"The Apple iPod iTunes Anti-Trust Litigation"
Filing
757
RESPONSE (re 751 Administrative Motion to File Under Seal Plaintiffs' Memorandum of Law in Opposition to Defendant's Motion for Summary Judgment and to Exclude Expert Testimony of Roger G. Noll; Plaintiffs' Responsive Separate Statement in Support of ) filed byApple Inc.. (Attachments: # 1 Declaration of Amir Q. Amiri in Support of Apple's Response to Plaintiffs' Admin. Motion to File Under Seal, # 2 Proposed Order by Apple Granting Plaintiffs' Admin. Motion to File Under Seal, # 3 Exhibit - Apple's (Proposed) Redactions to Plaintiffs' Exhibit Nos. 1-3, and 54, # 4 Exhibit - Apple's (Proposed) Excerpt to Plaintiffs' Exhibit No. 22, # 5 Exhibit - Apple's (Proposed) Redactions to Plaintiffs' Exhibit Nos. 9-11; 14; 48; 50-53; and 62, # 6 Apple's (Proposed) Redactions to Plaintiffs' Memorandum of Law in Support of its Opposition to Motion for Summary Judgment, etc., # 7 Apple's (Proposed) Redactions to Plaintiffs' Responsive Separate Statement in Support of its Opposition to Motion for Summary Judgment, etc.)(Amiri, Amir) (Filed on 1/21/2014)
*APPLE'S (PROPOSED) REDACTIONS*
EXHIBIT 1
[Filed Under Seal]
UNITED STATES DISTRICT COURT
NORTHERN DISTRICT OF CALIFORNIA
OAKLAND DIVISION
____________________________________
|
APPLE IPOD ITUNES ANTI-TRUST
|
|
LITIGATION
|
____________________________________|
Lead Case No. C-05-00037-YGR
DECLARATION OF ROGER G. NOLL
ON LIABILITY AND DAMAGES
My name is Roger G. Noll, and I have submitted six declarations in this
proceeding. 1 My declaration of January 18, 2011, contains a description of my
qualifications and a list of the antitrust cases in which I submitted a declaration, was
deposed, and/or testified at trial. Since that date I have published a book and several
articles, and have received two awards: the Alfred E. Kahn Distinguished Career Award
from the American Antitrust Institute and the Distinguished Member Award of the
Transportation and Public Utilities Group of the American Economic Association. My
updated curriculum vita is attached as Appendix A to this declaration.
Since January 2011, I have testified in person in the following proceedings.
SmithKlein Beecham d/b/a GlaxoSmithKline vs. Abbott Laboratories (U.S.
District Court, Oakland);
Novell vs. Microsoft (U. S. District Court, Salt Lake City);
DVD CCA vs. Kaleidescape (Superior Court, San Jose); and
1. Declaration of Roger G. Noll (July 15, 2008), Reply Declaration of Roger G. Noll
(October 19, 2009), Declaration of Roger G. Noll (January 18, 2011), Reply Declaration
of Roger G. Noll (March 28, 2011), Supplemental Declaration of Roger G. Noll (July 18,
2011), and Second Supplemental Declaration of Roger G. Noll (September 23, 2011).
CONFIDENTIAL – ATTORNEYS EYES ONLY
In the Matter of Adjustment of Rates and Terms for Pre-existing Subscription and
Satellite Digital Audio Radio Service (Copyright Royalty Board, Washington, D. C.).
I have submitted declarations and have been deposed in the following additional cases.
Sarah Perez, et al., vs. State Farm Mutual Automobile Insurance Co., et al. (U.S.
District Court, San Jose);
Federal Trade Commission vs. Cephalon (U.S. District Court, Philadelphia);
In re Text Messaging Antitrust Litigation (U.S. District Court, Chicago); and
In re NCAA Student Athlete Name and Likeness Licensing Litigation (U.S.
District Court, Oakland).
I am the co-author of an amicus submission to the Federal Communications Commission.
Petition to Reconsider Sports Blackout Rules (Federal Communications
Commission).
ASSIGNMENT
Attorneys for the class plaintiffs in this litigation have asked me to undertake an
antitrust economics analysis of the liability and damages issues in this litigation. I have
been asked to determine whether the update to the iTunes digital media player software,
known as iTunes 7.0, caused harm to competition in a relevant market for portable digital
media players and, if so, to calculate the damages to members of the class of purchasers
of iPods from the date at which the update was issued on September 12, 2006, until the
end of the class period, March 31, 2009.
In undertaking this assignment I have read the legal submissions by the parties
and the decisions by the court in this case, the defendant’s answers to interrogatories, the
CONFIDENTIAL – ATTORNEYS EYES ONLY
2
expert reports submitted on behalf of the defendant in earlier phases of this litigation,
numerous discovery documents and depositions, and many publications about the sound
recording, consumer electronics and wireless communication industries. The discovery
material that I reviewed, including the material that I reviewed for my prior reports, is
listed in Appendix B. Appendix B also includes the publications on which I relied,
including publications from my prior reports. I also have relied on my 45 years of
experience in analyzing the economics of the communications industry. In undertaking
my analysis, including the statistical analysis of the data that have been produced by the
defendant, I have been assisted by the professional staff at Economists, Inc.
SUMMARY
The plaintiffs in this litigation allege that Apple maintained and enhanced its
monopoly power in the market for portable digital media players by releasing iTunes 7.0,
an update of the software that is used to store and catalog digital audio files on a personal
computer and an iPod, and making other changes to the internal electronics of new iPods
models. For convenience, I refer to all of these changes as the iTunes 7.0 update.
The class contains all entities except government agencies and Apple employees
that purchased iPods directly from Apple from September 12, 2006 (the release date of
iTunes 7.0) until March 31, 2009 (the day before Apple began selling downloads of audio
recordings from all of the major record companies that were not protected by a digital
rights management (DRM) system). Apple classifies its customers into two groups.
Apple defines “resellers” as entities that purchase iPods for resale, including retail stores
and wholesale distributors that sell to retail stores. Apple defines “direct purchasers” as
CONFIDENTIAL – ATTORNEYS EYES ONLY
3
customers that do not buy for resale, including individual consumers, corporations,
elementary and secondary schools, and government agencies. Apple also classifies
university bookstores as direct purchasers.
The iTunes 7.0 update disabled Harmony, a software product from RealNetworks
that allowed owners of iPods to assemble a library of audio files from the RealPlayer
Music Store (RMS) that could be played on both iPods and portable digital media players
that competed with iPods. The plaintiffs allege that disabling Harmony increased the cost
of switching from iPods to other brands of portable digital media players and thereby
harmed consumers by limiting choice and causing higher prices for iPods.
I have undertaken an antitrust economic analysis of these allegations. This
section summarizes my conclusions.
First, the conduct that is alleged in the plaintiffs’ complaint is an example of
“lock-in,” which is a form of foreclosure that arises from actions that increase the cost to
consumers of switching to a product that has better quality and/or a lower price. In this
case the iTunes 7.0 update raised the cost of switching from iPods to competing portable
digital media players by eliminating the ability of consumers to collect a library of
downloads that could be played on all players. Thus, the iTunes 7.0 update had the effect
of increasing the extent of lock-in for iPod owners. This effect is important because the
average replacement rate for iPods during the class period is short, about two years.
Second, although the presence of market power is not necessary for lock-in to
reduce efficiency and to cause harm to competition, in this case Apple enjoyed market
power during the class period in two relevant antitrust markets: the market for portable
digital media players, which includes iPods, and the market for permanent downloads of
CONFIDENTIAL – ATTORNEYS EYES ONLY
4
digital audio files, which includes Apple’s download service, originally the iTunes Music
Store but subsequently the iTunes Store (iTS). Apple enjoyed monopoly power in both
markets from the launch of iTS in April 2003 until at least the end of the class period.
Third, Apple enhanced and maintained its monopoly power in portable digital
media players by making iPods incompatible with audio files that were downloaded from
sites other than iTS. Periodic replacement of a portable digital media player is attractive
to consumers because rapid technological change allows greater memory, faster and more
powerful internal electronics, longer battery life, sleeker design from miniaturization, and
better sound reproduction. After the iTunes 7.0 update, users of new iPods could not
acquire downloads that could be played on both iPods and competing portable digital
media players until iTS and its competitors began to sell downloads that were not
encrypted. As a result, consumers who bought iPods with the iTunes 7.0 update and who
purchased downloads could not avoid buying from iTS, and if they did, they would face a
switching cost if they chose to replace their iPod with a competing player. This higher
switching cost increased Apple’s monopoly power in the market for portable digital
media players.
Fourth, Apple’s actions to make downloads from RMS incompatible with new
iPods enabled Apple to charge higher prices for iPods than otherwise would have been
the case. The damages in this case are the overcharge on iPods during the class period
due to the incompatibility that was created by iTunes 7.0. Several hedonic regression
models were estimated using Apple’s transactions records. The preferred model is a
logarithmic price equation that excludes observations with missing data or “outlier”
prices.
CONFIDENTIAL – ATTORNEYS EYES ONLY
5
ECONOMIC BACKGROUND
I understand that the issues in this case have been narrowed to the competitive
effects of Apple’s update of iTunes 7.0. To analyze this issue requires information about
how the evolution of digital distribution of audio files led to technical incompatibility
between Apple’s products and the products of competitors, and about how technical
incompatibility caused lock-in and thereby reduced the intensity of competition. The
goal of this section is to explain the causes and effects of lock-in as it applies to portable
digital media players. Because lock-in can arise for reasons other than anticompetitive
conduct, the discussion covers events other than the iTunes 7.0 update that affected the
extent to which Apple’s customers were locked in to iPods.
The History of Audio Downloads and Portable Digital Media Players
Until the 1990s sound recordings were offered to consumers only in physical
formats, such as audio tapes, vinyl records, or compact discs (CDs). By the 1980s digital
files could be delivered to personal computers over the telecommunications network, but
the speed of network transmission was too slow to make delivery of high-quality musical
CONFIDENTIAL – ATTORNEYS EYES ONLY
6
recordings practical. Eventually technological progress enabled telecommunications
carriers to support the delivery of high-quality sound recordings.
By the mid 1990s the quality of telecommunications had advanced sufficiently to
permit the digital distribution of music over the Internet. Beginning about 1997 several
firms introduced technology to deliver digital audio files to a computer, 2 but the record
industry was slow to embrace the technology. The first e-commerce Internet sites that
sold audio files from major record labels operated as mail-order retailers, offering the
opportunity to buy physical copies of recordings via the mail or other delivery services.
By 2000 nearly 85 percent of retailers had launched web sites for this purpose, and
Internet sites accounted for about three percent of retail sales. 3
An important advantage of Internet distribution is that it eliminates the need to
manufacture, store, ship and display physical products. Despite the prospect for a large
cost reduction for suppliers and much greater convenience for customers, the record
companies were reluctant to offer most of their catalogues for sale as downloads over the
Internet. As a result Internet sites such as Napster and Grokster had an opportunity to
offer illegal “file sharing” services without serious competition from legal Internet
sources. By 1999, when Napster was launched, “retailers and wholesalers have been
2. “Statement by Mike Farrace,” Hearings before the Committee of the Judiciary, U.S.
Senate, April 3, 2001, 2001 WL 323720 (F.D.C.H.). This testimony also is available on
the web site of the National Association of Recording merchandisers, www.narm.com.
3. “2000 Annual Survey Results,” National Association of Recording Merchandisers, pp.
1, 5, and Music and the Internet: Celestial Jukebox, Lehman Brothers (Europe),
November 9, 2000, pp. 21-24, 28, as cited in Roger G. Noll, “Napster’s Copyright
Misuse Defense and the Future of Internet Distribution of Music,” an essay derived from
the public version of my expert report in the Napster case.
CONFIDENTIAL – ATTORNEYS EYES ONLY
7
ready, willing and able to deliver secure online entertainment...” 4 During the period
when Napster operated, legal retail sales of digital music were minuscule. In 2000 digital
distribution revenues “were almost too small to measure.” 5 Exhibit 1 shows the amount
of revenue accounted for by physical copies, downloads, and digital streaming services.
In 2004 downloads accounted for only about 1.3 percent of total sales.
Meanwhile, the first portable digital media player, called the MPMan F10, was
brought to market in 1998, but the first player to be favorably received was the Rio
PMP300 a few months later. 6 In 2001, Apple entered the market with its first iPod, and
became one of the handful of players to obtain significant sales. These early products
were accompanied by software, called somewhat confusingly a digital media player, that
could transfer, store, and play digital audio files from a CD on a personal computer.
During the period when digital distribution of music was dominated by illegal
file-sharing sites, record companies jointly developed a business plan for downloads with
two key elements. First was to protect audio files by using DRM systems, one purpose of
which was to create an impediment to file sharing among consumers. Second, DRM also
would allow record companies to control the distribution and uses of digital music, such
as by limiting the number of times that a file could be played without further payment.
To carry out this plan the then-five major record distribution companies 7 formed two
4. “Statement of National Association of Recording Merchandisers,” Hearings before the
Committee of the Judiciary, U. S. Senate, April 3, 2001, p. 1 (www.narm.com).
5. NARM, “2000 Annual Survey Results,” op. cit., p. 6.
6. Alex Cosper, “The History of the Portable MP3 Player,” eHow, no date, at
http://www.ehow.com/about_5409458_history-portable-mp-players.html.
7. The five companies have become three: Universal/EMI, Sony/BMG, and Warner.
CONFIDENTIAL – ATTORNEYS EYES ONLY
8
joint ventures, MusicNet and PressPlay, to sell rights to distribute recordings to retail
download sites. BMG, EMI and Warner were partners in MusicNet, while Sony and
Universal were partners in PressPlay.
The five major record companies gave MusicNet and PressPlay identical nonexclusive licenses to the same recordings. The only major difference between these sites
was that they used different DRM formats (RealNetworks and Microsoft). The services
shared two undesirable features: a limited selection of music and extensive restrictions
on the use of downloads. Because of these limitations, an article in a leading computer
trade magazine ranked them 9th on the list of the worst tech products of all time. 8
Before consumers had time to make their own judgments about these services, the
record companies were handed a major setback in their infringement litigation against
Napster. The judge in that case refused to issue a permanent injunction against Napster
until the court decided whether the record companies’ involvement in MusicNet and
PressPlay was anticompetitive and so constituted copyright misuse. 9 Within a few
months the record companies divested these joint ventures.
Approximately one year after the Napster decision, Apple launched iTS in April
2003. The important breakthroughs for iTS were a much larger inventory of audio files
than were available on previous download sites and some relief from limitations on the
use of downloads. Other sites with these characteristics were not authorized until six
months later. One limitation on iTS and its competitors was the requirement from record
8. Dan Tyson, “The 25 Worst Tech Products of All Time,” PC World, May 26, 2006, at
www.pcworld.com/article/125772-3/the_25_worst_tech_products_of_all_time.html.
9. “Memorandum and Order,” In re Napster, Inc., Copyright Litigation, U.S. District
Court (Northern California), Case No. MDL 00-1369 MHP, February 2002.
CONFIDENTIAL – ATTORNEYS EYES ONLY
9
companies to use DRM protection. From its launch until the end of the class period the
downloads for sale on iTS from four of the five major record companies were protected
by Apple’s proprietary DRM system, FairPlay. 10
FairPlay and other DRM systems for downloads can be circumvented legally, but
the process is cumbersome and costly. Media player software allows computers to burn
CDs from downloaded audio files, although there are limits to the number of times each
sound recording can be burned. Files that are burned to a CD are not DRM-protected, so
a consumer can convert a DRM-protected file to an unprotected file by burning the CD
and then reloading it on the computer. Because converting protected files to unprotected
files is costly and time consuming, customers who purchased downloads became locked
in to the particular DRM system that was compatible with their portable digital media
player. Because Apple chose not to license FairPlay 11 and not to permit downloads from
other Internet sites to play on an iPod, Apple’s customers also were locked in to both iTS
for downloads and iPods for players. 12
10. EMI abandoned DRM protection for recordings sold by iTS in April 2007. The other
labels did not allow iTS to sell recordings in an unprotected format until 2009.
11. Before Apple introduced the iPhone, Motorola was licensed to manufacture a feature
phone that included a portable digital media player and that could access iTS, but this
license was not renewed after Apple introduced the iPhone. Matthew Hicks, “Motorola
Previews iTunes Phone,” January 7, 2005, eWeek.com. See also AIIA00328028-29
(Apple internal email regarding Motorola’s official press announcement). HewlettPackard apparently is the only equipment manufacture to have an agreement with Apple
to market an iPod with the HP brand. These devices also were capable of playing audio
files in the FairPlay format.
12. For analysis that reaches similar conclusions about the effects of FairPlay and other
DRM systems on consumer welfare, see Neil Weinstock Netanel, “Temptations of the
Walled Garden: Digital Rights Management and Mobile Phone Carriers,” Journal on
Telecommunications and High-Technology Law Vol 6 (2007-08), pp. 77-100, and
Thierry Reyna and Ludmila Striukova, “White Knight or Trojan Horse? The
CONFIDENTIAL – ATTORNEYS EYES ONLY
10
From the launch of iTS through the end of the class period, the most important
competing proprietary DRM formats were Microsoft’s Windows Media Audio (WMA)
and RealAudio from RealNetworks. Both were licensed to others and were compatible
with many portable digital media players, although Microsoft also had a proprietary
version of WMA that was used only in connection with its portable digital media player,
Zune (discontinued in October 2011 13). Thus, neither WMA (except the Zune version)
nor RealAudio caused their customers to be locked in to one brand of portable media
player, although both created switching costs for consumers who use a portable media
player that supports one DRM system (say, RealAudio) but who would like to switch to a
player that supports the other DRM system (say, WMA). Because many players support
each DRM system, the lock-in effect of these systems reduces competition only among
download sites, not among portable media players.
The software that enables consumers to load and play digital audio files is called,
somewhat confusingly, a digital media player (not to be confused with portable digital
media player, which is a physical device for playing digital audio files). A digital media
player allows a consumer to transfer, store, catalog and play audio and video files on a
personal computer, to burn those files on a CD if permitted by the DRM system, and to
transfer and catalog these files to a portable digital media player. Apple’s iTunes is the
only digital media player that can download and play a recording in the FairPlay format.
Recordings in the FairPlay format can not be transferred to and played on any portable
digital media player other than an iPod without using another program and, in almost all
Consequences of Digital Rights Management for Consumers, Firms and Society,”
Communications & Strategies No. 69 (1st Quarter, 2008), pp. 109-26.
13. Owen Williams, “Zune Player Officially Discontinued,” Simcity, October 4, 2011, at
http://www.neowin.net/news/zune-player-officially-discontinued.
CONFIDENTIAL – ATTORNEYS EYES ONLY
11
cases, a CD burner or other electronic equipment. 14 Likewise, iTunes can not convert
WMA or RealAudio DRM-protected files into a format that can be played by an iPod.
During the class period audio files in each of the latter two formats could be played on
several brands of portable digital media players. Thus, iTunes played an essential role in
maintaining technical incompatibility between the defendant’s products and competing
products in the relevant markets.
Several other formats that have no DRM protection have been available since
before the launch of iTS and can be played on iPods. The most important DRM-free
formats are AAC and MP3. While iPods are not compatible with RealAudio and WMA,
they can play audio files in unprotected MP3 and AAC formats. 15 By January 2008, the
major record companies all had agreed to let Internet vendors other than iTS sell
downloads in unprotected formats. After that date iPod owners could buy downloads
from sites other than iTS and use iTunes to load these files on an iPod. In January 2009
Apple announced that iTS would sell audio files in the unprotected AAC format. By
April 1, 2009, iTS’s transition to DRM-free files was complete. Since that time nearly all
new downloads could be played on any portable digital media player so that all first time
buyers of portable digital media players were free of lock-in arising from a proprietary
audio file format; however, owners of older iPods who had substantially libraries of
DRM-protected audio files remained locked in when they replaced their old player.
The events that culminated in the iTunes 7.0 update began in July 2004 when
14. Some hackers have offered programs for breaking FairPlay’s encryption and
converting a protected file from iTMS to an unprotected AAC or MP3 file, but these
programs are not wholly successful and in at least some cases may be illegal.
15. For more details about which audio formats are compatible with an iPod, see Apple’s
web site: http://support.apple.com/kb/HT1334.
CONFIDENTIAL – ATTORNEYS EYES ONLY
12
RealNetworks released RealPlayer 10.5, an upgrade of RealNetwork’s counterpart to
Apple’s iTunes media player software. This upgrade included Harmony, a software
product that allowed users to play a download from RMS on an iPod. In August,
RealNetworks announced its “Freedom of Choice” campaign to promote Harmony,
offering records at half price (49 cents per song and $4.99 per CD) on RMS. 16
As discussed in the report by plaintiffs’ technical expert, Dr. David Martin, that
was submitted with plaintiffs’ opposition to defendant’s motion for summary judgment,
in October 2004 Apple issued iTunes 4.7, an iTunes update that restored incompatibility
between new iPods and files that were downloaded from RMS. This update prevented
consumers from playing RMS downloads on new iPod models and made old downloads
from RMS that were compatible with a customer’s old iPod incompatible with a new
iPod. An iPod owner who had taken advantage of the RealNetworks 50 percent off sale
to buy downloads from RMS could not play these files on a new iPod.
In April 2005 RealNetworks responded to iTunes 4.7 by releasing an upgrade of
Harmony that restored compatibility between iPods and audio files from RMS. This
upgrade worked for about a year and a half. According to Dr. Martin, the iTunes 7.0
update, including firmware on new iPod models that were released after September 12,
2006, created a new form of incompatibility between iPods and audio files that had been
downloaded from RMS. RealNetworks never overcame this new incompatibility. As a
result, new iPod owners who wanted to download audio files that were protected by a
digital rights management system were forced to acquire these recordings from iTS, and
16. The RealNetworks press release, dated August 17, 2004, is available on PRNewswire
at http://www.prnewswire.com/news-releases/realnetworks-kicks-off-freedom-of-choicecampaign-with-biggest-music-sale-in-history-71643667.html.
CONFIDENTIAL – ATTORNEYS EYES ONLY
13
in so doing continue to build a library that locked them into iPods.
The Economics of Lock-In
The incompatibility between some iPods and downloads from RMS had two
effects on owners of these iPods. First, these iPods could not play DRM-protected audio
files from RMS, forcing customers to use iTS to obtain downloads. Second, owners of
iPods who had used RMS and thereby had lower switching costs, by being forced to use
iTS for downloads, began building a library of files that increased the cost of switching to
a competing player. An iPod owner who purchased downloads from iTS and who
contemplated switching to a competing player faced a choice of three costly alternatives:
(1) abandon playing files obtained from iTS on the new player; (2) burn iTS audio files
onto CDs and upload them to a computer for the purpose of loading them onto the new
player; or (3) repurchase the audio files obtained from iTS in a DRM format that was
compatible with the new player. Each option imposes a switching cost.
Another consequence of the incompatibility between iPods and downloads from
RMS involves a network effect. A network effect occurs when the value of a product is
greater if other people buy the same product. Because a download can be played on
several portable digital media players, members of the same family can share downloads
if the audio files are compatible with the iPods. Suppose a music customer, Person A,
has not bought any audio files from iTS, does not want to play any recordings that family
members have acquired from iTS, and so does not face a cost to switch to a competing
player. But if other family members would like to play audio files that are downloaded
by Person A, they will suffer a loss if Person A starts downloading files from RMS.
CONFIDENTIAL – ATTORNEYS EYES ONLY
14
Switching costs and network effects cause “lock-in,” 17 which occurs when a
customer incurs a cost by changing brands. Lock-in does not imply that customers
cannot change brands. Instead it refers to a circumstance in which a customer has an
incentive not to change brands. Lock-in allows a brand profitably to set prices above the
competitive level. For simplicity, call the brand that a customer currently owns the
incumbent and call the alternative brand the challenger. Assume that a consumer’s old
portable media player is worn out or obsolete. Also assume that the incumbent’s price is
Pi, the challenger’s price is the competitive market price Pc, the switching cost is S, and
the value of the network effect benefit of the incumbent is N. If the incumbent and the
challenger are otherwise identical in function and quality (i.e., except for switching costs
and network effects the products are perfect substitutes), the incumbent can retain the
customer if Pi < Pc + S + M. Thus, the incumbent has market power in that it profitably
can set price above the competitive level. If customers of all brands are locked in, the
same argument holds for each brand, and all prices will be above the competitive level.
One inference to be drawn from the preceding analysis is that lock-in is not an
either/or condition. Specifically, a change in switching costs, S, causes an equal change
in the maximum price, Pi, that the incumbent can charge. Thus, small changes in S have
small effects on Pi. If a product has numerous sources of lock-in, including a strong
brand-name reputation, the elimination of another source of lock-in, such as technical
17. On the economics of lock-in, see Carl Shapiro and Hal Varian, Information Rules: A
Strategic Guide to the Information Economy, Harvard University Press, 1999; Pei-yu
Chen and Lorin M. Hitt, “Information Technology and Switching Costs,” in Terrence
Hendershott, Handbook on Economics and Information Systems, Elsevier, 2006, pp. 43770; Joseph Farrell and Paul Klemperer, “Coordination and Lock-in: Competition with
Switching Costs and Network Effects,” in Mark Armstrong and Robert Porter, Handbook
of Industrial Organization Vol. 3, Elsevier, 2007, pp. 1967-2072.
CONFIDENTIAL – ATTORNEYS EYES ONLY
15
incompatibility, will not cause Pi to fall to Pc.
An important characteristic of lock-in is that even if markets otherwise are
competitive, lock-in still can cause higher prices. 18 Suppose that in the beginning no
customers have made a purchase so no customers are locked in to any brand. If suppliers
know that new customers will be locked in after purchase, and if they cannot negotiate in
the first period the prices that will be charged in future periods because important factors
that influence price, such as technological change and input prices, are not predictable,
then both customers and suppliers know that repurchase prices will exceed the
competitive level. In this circumstance, suppliers will compete for customers by offering
initial prices that are below the competitive level. The magnitude of the initial price cuts
and the subsequent overcharges will be determined by the fraction of future purchases
that are accounted for by new customers.
The preceding example of lock-in in competitive markets may seem to leave both
consumers and suppliers unharmed because the monopoly profits in later period are offset
by price cuts in the initial period. While suppliers are not harmed by this circumstance,
consumers usually are, although there are exceptions. In most cases the product in
question does not operate perfectly up to a date at which it ceases to function and so must
be replaced. Instead, the performance gap between new and old products tends gradually
to increase over time. In this case a consumer will purchase a new product when the
value of the performance differential becomes greater than the net price (the gross price
18. This paragraph summarizes the analysis in Severin Borenstein, Jeffrey K. MackieMason, and Janice S. Netz, “Antitrust Policy in Aftermarkets,” Antitrust Law Journal
Vol. 63 (1994-95), pp. 455-82. This analysis was applied in Eastman Kodak v. Image
Technical Services, 112 S. Ct. 2072 (1992). The competitive first purchase was a highspeed photocopier and the monopolized subsequent purchases were repair services.
CONFIDENTIAL – ATTORNEYS EYES ONLY
16
minus any trade-in or resale value) of buying a new product. If products are sold at the
competitive price in all periods, consumers will replace old products more frequently
than if the price of new products in later period exceeds the competitive price. The delay
in replacement from the date at which it is efficient (determined by the competitive price)
to a later date (determined by the monopoly price) harms consumers.
Lock-in can lead to higher prices even if not all customers are locked in. Lock-in
causes demand in later periods to be less elastic (i.e., less responsive to changes in price).
A seller that could distinguish between buyers according to whether they are locked in
would be able profitably to increase price to the locked in customer while retaining the
same price and profitability on sales to customers who were not locked in. But a seller
that cannot make this distinction will perceive that the overall demand for the product has
become less elastic. The new elasticity will be a weighted average of the elasticities of
the locked in and not locked in customers. 19
Whether lock-in harms consumers is an empirical issue. The previous examples
are based on a market in which all consumers buy a product in the first period, with no
19. For example, suppose the market initially has 100 customers, each of whom has a
demand, Q, for the product given by Q = 10 – P, where P is the price. For ease of
exposition, assume that the product is costless to produce, so that the supplier maximizes
total revenue R = 100(10 – P)P = 1000P -100P2. In this case the profit-maximizing price
is 5 and the supplier sells 500 units, five to each customer, for a total profit of 2500.
Suppose that 20 of these customers become locked in, meaning that their replacement
demand is less elastic. Thus, in the period in which the product is replaced, 80 customers
have the old demand but 20 customers have a new, less elastic demand, given by Q = 7 .4P. Note that at the old price, 5, each of these customers would still buy 5 units. Thus,
the demand faced by the supplier in the replacement period is 80(10 – P) + 20(7 - .4P) =
940 – 88P. Revenue is now (940 – 88P)P, the profit-maximizing price is 5.34, the
quantity sold is 475, and profit has increased to 2536.50. The harm to consumers has two
components: the 34 cent price increase (a total cost of 36.50) and a loss of 25 units of the
product, which had a net value to consumers of ½(.34)(25) = 4.25 (the familiar deadweight loss of monopoly).
CONFIDENTIAL – ATTORNEYS EYES ONLY
17
new customers arriving in the next period when the first customers are locked in, and in
which there are no economies of scale that cause the incremental cost of production to
fall as production is increased. In a market that is growing rapidly and in which costs fall
as production increases, a supplier is especially interested in attracting new customers, so
that prices in later periods may be determined primarily by the desire to attract them.
Thus, lock-in normally has a greater effect on price as the fraction of sales that are
accounted for by replacement purchases grows.
The economics of lock-in applies to digital audio files and portable digital media
players. A download is a highly durable asset that can be used without degradation in
quality for an indefinitely long period. A digital audio file loses economic value to a
consumer only if the consumer no longer wants to listen to it.
A portable media player has a shorter economic life, in part because the products
can be lost, stolen or broken, and in part due to rapid technological progress in
microelectronics. The electronics in portable media players are semiconductor products
that follow Moore’s Law, first enunciated by Intel co-founder Gordon Moore. According
to Moore’s Law, the amount of functionality that can be placed on a semiconductor of a
given size doubles every 18 months. For portable media players, Moore’s Law is the
source of larger memory, better sound quality, miniaturization, and the addition of video.
These factors cause the ownership life cycle (average replacement rate) of portable digital
media players and cell phones to be between 18 and 24 months. 20
20. Jemima Kiss, “How Big Is the iPod Installed Base?” Guardian, September 9, 2009
(reporting discussion with executive at Forrester Research); Larry Dignan, “Tablet
Replacement Rates: More Like an MP3 Player than PC,” ZDNet January 4, 2011
(reporting a Forrester Research study); “Mobile Phone Lifecycles,” GSM Association,
2006 (reporting that about half of phone sales are replacements and that the replacement
CONFIDENTIAL – ATTORNEYS EYES ONLY
18
The economics of lock-in in the information technology sector of the economy
has been extensively studied. Several studies examine the effects of changes in
telecommunications policies that reduce switching costs and network effects.
One example is the introduction of “number portability,” which allows phone
customers who change carriers to keep their old telephone numbers. In the absence of
number portability customers who switch carriers must go to the trouble and expense of
notifying those with whom they communicate that their numbers have changed. Number
portability eliminates this switching cost. Scholars have quantified switching costs by
studying the effect of adopting number portability on prices, churn (the rate at which
customers switch carriers), and the market share of the dominant carrier. These studies
find that number portability increases churn, lowers the market share of the dominant
carrier, and reduces prices, from which they conclude that this reduction in switching
costs intensified competition among carriers. 21
rate is about 18 months); “The Life Cycle of a Cell Phone,” U.S. Environmental
Protection Agency, 2005 (reporting cell phone replacement rate of 18 months); John
Paczkowski, “I Got a Fever, and the Only Prescription is… More iPhone!” All Things
Digital, June 25, 1010 (reporting that the replacement cycle for iPhones is 14.7 months);
Victor H., “Americans Replace Their Cell Phones Every 2 Years, Finns – Every Six, a
Study Claims,” Phonearena.com, July 11, 2011 (reporting a study by Recon Analytics
finding that the replacement rate for mobile phones was 18.7 months in 2007, 19.6 in
2008, and 21.1 in 2009).
21. See V. Brian Viard, “Do Switching Costs Make Markets More or Less Competitive?
The Case of 800-Number Portability,” Rand Journal of Economics Vol. 18, No. 1 (Spring
2007), pp. 146-63 (finding U.S. toll-free service prices fell after number portability was
adopted); Jung Eun Ku, Sang Woo Lee, and Tschanghee Hyun, “Value of Number
Portability on Internet Phones,” ETRI Journal Vol. 32, No. 1 (February 2010), pp. 469-71
(finding than number portability for mobile phones that can access the Internet increased
consumer welfare $1.50 per month per user and doubled use in two months); Tokio
Otsuka and Hitoshi Mitomo, “User Benefits and Operator Costs of Mobile Number
Portability in Japan and Impact on Market Competitiveness,” Telecommunications Policy
(in press), at http://www.sciencedirect.com/science/journal/aip/03085961 (finding that
mobile number portability reduced prices and the market share of the leading carrier).
CONFIDENTIAL – ATTORNEYS EYES ONLY
19
Another example is the effect of unlocking mobile devices. The two principal
technologies for mobile telephones are CDMA and GSM. CDMA telephones are
“locked” in the sense that a phone can access the telecommunications network through
only one carrier. Thus, a subscriber to a CDMA network (such as Verizon or Sprint in
the U.S.) must buy a new telephone to switch to another carrier, thereby creating a lockin to the customer’s original carrier. GSM telephones are not necessarily tied to one
carrier. GSM telephones include a Subscriber Identification Module on a removable card
(SIM card) that enables a customer to use the phone to connect through a specific carrier.
A customer can switch carriers by changing the SIM card unless the carrier has placed a
lock on the mobile device. In the U.S. a GSM customer must obtain the consent of the
carrier (AT&T or T-Mobile) to unlock a phone and replace the SIM card. GSM is more
commonly used than CDMA in the rest of the world and is used exclusively in the
European Union. In many countries, including the European Union, GSM carriers are
prohibited from locking phones, so customers can access the network through multiple
carriers by simply replacing the SIM card. Research on the effect of unlocking mobile
telephones concludes that prices are lower in nations in which phones are not locked. 22
These studies reference other research on the same topic.
22. Akohiro Nakamura, “Estimating Switching Costs Involved in Changing Mobile
Phone Carriers in Japan: Evaluation of Lock-in Factors Related to Japan’s SIM Card
Locks,” Telecommunications Policy 34 (2020), pp. 736-46 (finding from surveys that
eliminating locked mobile phones would benefit at least 20 percent of customers and that
unlocking combined with eliminating incompatibilities in content that is available on
each carrier would cause a substantial increase in consumer welfare); Lucio Fuentelsaz,
Juan Pablo Maicas, and Yolando Polo, “Switching Costs, Network Effects, and
Competition in the European Mobile Telecommunications Industry,” Information
Systems Research Vol. 23, No. 1 (March 2012), pp. 93-108 (finding that the nations in
Europe that have the lowest prices have the lowest switching costs from the combined
effects of number portability and phone unlocking, and that this effect is intensified in
nations in which carrier network effects are greater). These studies also reference other
CONFIDENTIAL – ATTORNEYS EYES ONLY
20
Lock-in also arises from network effects in communications when prices (whether
voice or data) depend on whether customers connect through the same carrier or different
carriers. Communication requires terminating the connection to the called party. Most
nations use “calling party pays,” in which the carrier of the customer who initiates
communication pays a termination fee to the terminating carrier. A carrier then passes on
the termination fee to the customer who originates the communication. Carriers compete
for customers over the price of originating a connection, but a carrier has a monopoly on
terminations to its own customers, which leads to a monopoly termination price if other
policies do not prevent it. Carriers also compete by offering lower termination charges
for communications that originate and terminate on the carrier’s network. This practice
creates a network effect: customers who frequently communicate pay less if they all buy
service from the same carrier. This network effect causes lock-in among customers who
buy service from the same carrier. The policies that can be used to overcome this lock-in
effect are: (1) regulate the price of termination, or (2) adopt “bill and keep,” in which
carriers (and customers) are not charged for termination on other networks. Research has
shown that “bill and keep,” by eliminating the monopoly in termination and the lock-in
effect of affinity groups to a particular carrier, causes the lowest prices. 23
As a theoretical matter, lock-in does not necessarily harm the competitive process.
Nevertheless, theoretical and empirical research concludes that the common result is that
lock-in makes competition less intense and harms consumers. Professors Farrell and
research on the same issues.
23. See Fuentelsaz, et al., op. cit. and Stephen C. Littlechild, “Mobile Termination
Charges: Calling Party Pays Versus Receiving Party Pays,” Telecommunications Policy
Vol. 30 (2006), pp. 242-77; David Harbord and Marco Pagnozzi, “Network-Based Price
Discrimination and ‘Bill-and-Keep’ vs. ‘Cost-Based’ Regulation of Mobile Termination
Rates,” Journal of Network Economics Vol. 9, No. 1 (March 2010), pp. 1-46.
CONFIDENTIAL – ATTORNEYS EYES ONLY
21
Klemperer conclude that “switching costs seem more likely to lower than to raise
efficiency, so when firms favor switching costs the reason often is because they enhance
monopoly or oligopoly power by directly raising prices or by inhibiting new entry.” 24
MARKET DEFINITION AND MARKET POWER
The plaintiffs allege that Apple enjoys monopoly power in the markets for digital
audio files and portable digital media players. From the perspective of the economics of
switching costs, the plaintiffs have alleged more than is required for lock-in to have an
anticompetitive effect. In lock-in markets in which firms compete intensely for the first
purchase, firms can enjoy ex post monopoly profits even if the market appears to be
structurally competitive. The best outcome for consumers in this type of market is that
competition in initial purchases is so intense that firms compete away their ex post
monopoly profits by setting initial prices far below costs – a circumstance that is called a
“bargain-then-ripoff pattern of prices.” 25 While this outcome leaves firms no better off
than had switching costs not been present, it harms consumers through its effects on
delaying or reducing purchases at ripoff prices.
Notwithstanding this caveat, the facts about the markets for digital audio files and
potable digital media players support the plaintiffs’ allegations. This section identifies
the markets in which iTS audio files and iPod portable digital media players are sold, and
then examines whether Apple enjoys monopoly power in these markets.
24. Farrell and Klemperer, op. cit., p. 2006.
25. Ibid.
CONFIDENTIAL – ATTORNEYS EYES ONLY
22
Market Definition
The purpose of relevant market analysis is to identify products that are close
substitutes. In antitrust economics, a relevant market consists of a reference product (the
product that is the subject of the complaint) and any close substitutes for that product that
could be profitably monopolized if all products were offered by the same seller. Products
are substitutes on the demand side if buyers would switch from one product to another in
response to a small reduction in the relative price of the latter. Products are substitutes on
the supply side if sellers would switch production from another product to a product that
is a close substitute in response to a small increase in the relative price of the latter.
The task of market definition is to identify the closest substitutes for a reference
product. To be close substitutes, products must be sufficiently similar that consumers
regard them as substitutes for performing the same functions, and must be conveniently
available to purchase them in the same geographic area. The relevant market consists of
the smallest number of products that, if sold by a single supplier, would be able to impose
a small but significant non-transitory increase in price (SSNIP) in comparison with the
prices that are charged when each product is sold separately. 26
Economists use several methods to identify a relevant market. In some cases,
economists estimate the cross-elasticity of demand (that is, how the sales of one product
are affected by the price of another product) between the reference product and each
other product that might be regarded as close substitutes. In most cases data limitations
preclude econometric estimation of cross-elasticity of demand. Econometric estimation
of cross-elasticities of demand is usually impossible for products that have extensive
26. Horizontal Merger Guidelines, U. S. Department of Justice and Federal Trade
Commission, 2010 (henceforth Merger Guidelines).
CONFIDENTIAL – ATTORNEYS EYES ONLY
23
product differentiation and that are rapidly evolving, as was the case of portable digital
media players during the class period.
If reliable estimation of cross-elasticity of demand is not feasible, economists look
for indirect evidence that products are close substitutes: 27 similarity of components and
functional uses, statements outside the context of litigation by executives and industry
analysts about their beliefs about which products are close competitors, and surveys of
buyers about which products they considered before buying a product that is a candidate
to be included in a relevant market.
The plaintiffs allege two relevant markets: downloads of digital audio files and
portable digital media players. The geographic area for these markets is the United
States. The reference products are iPods and iTS as a seller of downloads of digital audio
files. To define the relevant markets that include iTS or iPods involves collecting
information about prices, product characteristics, and informed beliefs among buyers,
sellers and industry observers about the closest substitutes for each reference product and
plausible close substitutes.
Two other products enter the analysis of competition in these product markets.
These are digital media players and the formats for DRM systems for digital audio files.
The analysis of competitive conditions in the markets for digital audio files and portable
digital media players is affected by these products, but does not hinge on the formal
definition of the product market for either of them.
27. The sources of evidence that are used in market definition, including internal records
of firms and their buyers as well as reports by industry analysts, are discussed more
completely in the Merger Guidelines, pp. 4-6, 10-11.
CONFIDENTIAL – ATTORNEYS EYES ONLY
24
Portable Digital Media Players
The alleged anticompetitive conduct in this litigation involves technical features
of iPods, iTS, and iTunes that reduce the substitutability between iPods other portable
digital media players. In the retail market that includes iPods, final consumers can
choose among portable digital media players. The possibility for substitution by final
consumers provides an opportunity for retailers other than Apple’s own retail outlets to
engage in substitution in the reseller market. Hence, the appropriate focus for defining
the relevant market for portable digital media players for both types of class members is
to identify close substitutes for iPods among final consumers, regardless of whether
Apple sold the product to a direct purchaser or a reseller.
The key characteristic of portable digital media players is the ability to play a
large number of digital audio files on a compact mobile device. The technology of
portable digital players has evolved since Apple introduced its first iPod in 2001. Rapid
technological progress in microprocessors, memory devices, batteries, and wireless
communications has been used to make players smaller and lighter, to increase the
storage capacity of players, and to expand the functionality of players. In 2005 Apple
introduced an iPod that could play digital video files and began to sell video downloads
through iTS. 28 In 2007 Apple introduced the iPod touch, which can access iTS over the
Internet and can be used for other applications, including video games. 29
The most obvious close substitutes for an iPod are other portable digital media
28. “Apple Unveils the New iPod” and “Apple Announces iTunes 6 with 2,000 Music
Videos, Pixar Short Films & Hit Shows,” Apple Press Releases, October 12, 2005. See
also http://www.apple-history.com/?page=gallery&model=ipod_video.
29. “Apple Unveils the iTunes WiFi Music Store” and “Apple Unveils iPod Touch,”
Apple Press Releases, September 5, 2007.
CONFIDENTIAL – ATTORNEYS EYES ONLY
25
players. 30 As of January 2011, Amazon.com offered the following brands of portable
digital media players: Archos, Coby, Cowon, Creative, Ematic, Ibiza, iPods, iRiver,
Latte, Meizu, Philips, Pyrus, SanDisk, Samsung, Sony, Toshiba and Zune. 31 In 2007,
CNet, a leading on-line source for reviews of consumer electronics, reviewed the
following portable digital media players: Altec Lansing, Apple, Archos, Coby, Creative
Zen, Cowon, iRiver, Microsoft, Philips, Samsung, San Disk, Shure and Sony. 32
. 33
30. For an example of detailed comparisons of the iPod and its leading substitutes, see
“Portable Digital Players: iPods Rule but Consider Other Brands,” Consumer Reports,
November 2006, at http://www.consumerreports.org/cro/electronics-computers/audiovideo/audio/ipods-mp3-players/mp3-players-11-06/overview/1106_mp3_ov_1.htm.
31. In January 2011 Amazon posted lists of portable digital media players that it sold at
http://www.amazon.com/MP3-Players-Audio-Video/b/ref=amb_link_86347991_ 3?ie=
UTF8&node=172630&pf_rd_m=ATVPDKIKX0DER&pf_rd_s=top-1&pf_rd_ r=
0TT5Q01MYZKSP88YNK53&pf_rd_t=301&pf_rd_p=157251702&pf_rd_i=mp3;
http://www.amazon.com/gp/search/ref=sr_tc_2_1?rh=n%3A1264866011%2Ck%3Amp3
&keywords=mp3&ie=UTF8&qid=1294860099&sr=1-2-tc; and http://www.amazon.com/
MP3-Players-Portable-Audio-Video/b/ref=amb_link_157669822_24?ie=UTF8&node=
1264866011&pf_rd_m=ATVPDKIKX0DER&pf_rd_s=center-5&pf_rd_r=
1T562SPGE1105HXYHVFZ&pf_rd_t=101&pf_rd_p=945911022&pf_rd_i=172630.
32. Jasmine France, “MP3 Players That Shaped 2007,” CNet, February 8, 2008, at
http://reviews.cnet.com/4321-6490_7-6606044.html lists the five best MP3 players for
2007, from Sony, Apple, Archos, SanDisk and Creative Zen. The others can be found by
searching the product manufacturers on the CNet site.
33. Defendant Apple Inc.’s First Amended Objections and Answers to Plaintiff’s Second
Set of Interrogatories 9-13, p. 5. Rio exited the market in 2005 and so was not available
during the class period.
CONFIDENTIAL – ATTORNEYS EYES ONLY
26
In 2005 cell phones “transitioned from relatively simple voice and text messaging
devices to gizmos capable of nearly everything a PDA can do, including instant
messaging (typically on AOL or Yahoo's IM services), playing music (primarily MP3
and AAC files), displaying snippets of TV shows, capturing 1-megapixel photos, and
running complex games.” 34 Cell phones that are bundled with portable digital media
player plausibly could be substitutes for a stand-alone player. 35 The first music phones
suffered from relatively low sound quality, slow download speeds from a PC to the
player, limited storage capacity, and limited battery life when used to download and then
to play audio files. 36 While performance improved over the next two years, the
34. Grace Acquino, “Dialed in: Best Cell Phones of 2005,” PC World, December 29,
2005, at http://www.pcworld.com/article/123742/article.html.
35 Because the price of a feature phone and, later, a smart phone is roughly equal to the
sum of the prices of a digital media player and a mobile telephone, these products are an
alternative to separate purchases of each device. Today mobile telephone penetration in
the United States is over 275 million, so that a very large fraction of consumers who want
a portable digital audio player also are likely to want a mobile telephone.
36. Thomas J. Fitzgerald, “Music to Your Cell Phone,” New York Times, July 7, 2005, at
http://www.nytimes.com/2005/07/07/technology/circuits/07basics.html?_r=0. The
Motorola Rokr, introduced in September 2005, was the first smart phone that could
access iTS, but Motorola’s iTS compatible cell phone was sold for only a few months.
See “The Music Cell Phone,” in History of Mobile Phones, Sutliffian Press, 2007,
http://www.xtimeline.com/evt/view.aspx?id=26731. A synopsis of a review of the LG
Fusic in 2006, while favorable concerning style and ease of use, stated that “the Fusic is
geared more toward the occasional listener than the audiophile.” Stan Horaczek, “LG
Fusic Music Phone Reviewed,” July 8, 2006, at http://www.engadget.com/2006/07/08/lgfusic-music-phone-reviewed/.
CONFIDENTIAL – ATTORNEYS EYES ONLY
27
performance shortfalls remained and the response was not enthusiastic. 37
In 2007 Apple introduced the iPhone, a “smart phone” that could access the
Internet and also included the features of an iPod. 38 Since the summer of 2007 the
closest functional substitutes for a portable digital media player is a smart phone that
includes the functionality of a portable digital media player. 39 In 2008 some financial
analysts concluded that iPhones and iPods were substitutes. 40 But during the class period
the functionality of smart phones fell short of the functionality of iPods. The main
37. A review of the LG Chocolate, offered by Verizon Wireless, observed that “U.S.
wireless operators have recently started shipping devices that download and play music to
increase revenue from their data services, which are still primarily used for text
messaging… Analysts, however, say it'll be awhile before carriers can offer a device that
would pose a strong challenge to the iPod.” Antone Gonsolves, “Verizon Launches iPod
Like Music Phone,” Information Week, July 31, 2006, at http://www.informationweek.
com/verizon-wireless-launches-ipod-like-musi/191600770. An article about the
anticipated release of the iPhone stated that “more than half of Americans with musiccapable phones also carry MP3 players” and that consumers “complain that existing
music phones make it difficult to synchronize their music collections and download
music…” Olga Kharif, “Another Music Phone? Yawn…” Bloomberg-Business Week,
October 18, 2006, at http://www.businessweek.com/stories/ 2006-10-18/another-musicphone-yawn-businessweek-business-news-stock-market-and-financial-advice. For a
similar assessment, see Yuki Noguchi, “Another Shot at a Music Phone,” Washington
Post, November 7, 2006, at http://voices.washingtonpost.com/posttech/2006/11/xm_
satellite_on_your_phone.html.
38. See “iPhone Premiers This Friday Night at Apple Retail Stores,” Apple press release,
June 28, 2007.
39. An example of an analyst’s report that reached this conclusion is Charlie Wolf,
“AAPL: Its MacWorld and We’re Just Living in It, Upgrading Apple from Buy to Strong
Buy,” Needham and Company, January 23, 2008.
40. “AAPL: The Reason for the iPhone’s Reported Woes Is Closer than You Think: It’s
the iPod touch,” Needham & Company, January 28, 2008, and “Apple’s Negative
Guidance Tone at the FQ1 Call Means a Lower Valuation of Multiple: Hold,” Kintisheff
Research, January 23, 2008. The former focuses on iPods cannibalizing the sale of
iPhones, and the latter focuses on iPhones cannibalizing iPod sales.
CONFIDENTIAL – ATTORNEYS EYES ONLY
28
shortfalls were storage capacity, size, and battery life. 41
Although the Apple iPhone received great attention as a breakthrough in smart
phones, the big jump in smart phone sales and use occurred after the end of the class
period. The ability of cell phones to substitute for portable digital media players is an
element of the usability of cell phones for data services. The annual reports on the
wireless industry by the Federal Communications Commission (FCC) contain data about
the penetration of high-speed data services over wireless devices.
The 2011 FCC wireless report covers the period during and immediately after the
class period. 42 The number of mobile telephone users who subscribed to Internet access
service was 26.5 million (out of 261 million mobile subscribers) in December 2008,
compared to 86 million who had devices that were capable of receiving communications
at 200 kilobits per second (the FCC’s threshold for defining high-speed access, which
also is the necessary speed for receiving high quality streaming music services). Data
were not collected on smart phones in use until after the end of the class period, but as of
June 2009 the number was 40.7 million. The fraction of adults who report ever having
used a mobile device for Internet access was 19 percent in December 2007 and 25
percent in April 2009. By comparison, the fraction of adults who reported owning a
portable digital media player was 47 percent, and the fraction of young adults (19-34)
41. Michael Kwan, “Feature: Why Standalone MP3 Players Still Exist,” Mobile
Magazine September 17, 2008, at http://www.mobilemag.com/2008/09/17/feature-whystandalone-mp3-players-still-exist/.
42. Federal Communications Commission, Annual Report and Analysis of Competitive
Market Conditions With Respect to Mobile Wireless, Including Commercial Mobile
Services, June 24, 2011, pp. 8-10, 95-102. The data series in this report end sometime
between December 2009 and June 2010.
CONFIDENTIAL – ATTORNEYS EYES ONLY
29
was 74 percent, in 2010. 43
Both sales of smart phones and the use of smart phones for data services grew
spectacularly in the two years after the class period came to an end. The next FCC
“annual report” on wireless was released in 2013, and it documents the explosive growth
in the use of wireless devices for data services. 44 While total mobile telephone
subscribers grew modestly to 285 million by December 2010 and 298 million in
December 2011, the number of subscribers who had devices that were capable of
receiving high-speed data services rose to 115.7 in December 2009, 151.6 million in
December 2010, and 183.7 million in December 2011. Of these, the number who
obtained high-speed Internet access rose to 56.3 million in December 2009, 97.5 million
in December 2010 and 142.1 million by the end of 2011. Thus, the number of mobile
phone subscribers who even had the capability of downloading audio files to their cell
phone increased nearly fivefold since the end of the class period. Between December
2007 and December 2011, data use on mobile devices increased 73 fold. In 2010, most
mobile users expected that their mobile phone would replace their portable digital audio
player by 2015, but by 2011 34 percent reported that this replacement had occurred. 45
The preceding data indicate that, indeed, smart phones are now competitive
substitutes for portable digital media players; however, the extent to which they are close
43. Kathryn Zucker, Generations and Their Gadgets, Pew Internet and American Life
Project, February 3, 2011, at http://www.pewinternet.org/Reports/2011/Generations-andgadgets/Overview.aspx.
44. Federal Communications Commission, Annual Report and Analysis of Competitive
Market Conditions With Respect to Mobile Wireless, Including Commercial Mobile
Services, March 21, 2013, pp. 9-11, 155-75.
45. Opportunity Calling: The Future of Mobile Communications – Take Two, Oracle
Communications, October 31, 2011, p. 6.
CONFIDENTIAL – ATTORNEYS EYES ONLY
30
substitutes was only just beginning at the end of the class period. Exhibit 8 shows the
relationship between sales of iPods and other portable digital media players and sales of
iPhones. The sale of iPhones in 2007 and 2008 were tiny compared to the sales of iPods.
In 2009, sales of portable digital media players started to decline, and in late 2010
iPhones finally began to outsell iPods. The quarterly sales of iPods, shown in Exhibit 9,
make the pattern much clearer. The historical peak in iPod sales occurred in the last
quarter (Christmas season) of 2008. The second quarter of 2009 was the first time that
iPod sales were below their sales one year earlier. Bearing in mind that in this period
most smart phones were not used for Internet access, these data indicate that smart
phones did not begin to have a competitively significant effect on the market for portable
digital media players until after the end of the class period.
The other product that is a candidate to be in the relevant market for portable
digital media players is portable CD players. My search of publications and documents
from Apple leads me to conclude that portable CD players had become obsolete and
unimportant before the beginning of the class period. According to one review, portable
CD players “have almost completely lost market share to MP3 players. However, for
those who have a collection of CDs and no desire to spend hours converting them to
digital format on a computer, portable CD players are still available.” 46 Consumers
Reports, the publication of Consumers Union, has not reviewed portable CD players
since 2002, and its current web site shows no hits for a search on “portable CD
46. “Portable CD Players Reviews,” ConsumerSearch, January 2008, at http://www.
consumersearch.com/portable-cd-players/review.
CONFIDENTIAL – ATTORNEYS EYES ONLY
31
players.” 47 CNet has not reviewed a portable CD since before the beginning of the class
period. 48 No products other than portable digital media players were considered by
Apple in the documents from which Exhibits 2-7 were created. Thus, there is no
evidence that CD players imposed a competitive constraint on portable digital media
players in the relevant time period for this litigation.
Digital Audio Files
The closest substitutes for iTS are the web sites that offer permanent downloads
of audio recordings that are distributed by the major record companies. The labels that
are distributed by the then five, now three, major record distribution companies account
for 85 to 90 percent of all sales of sound recordings. Internet sites that sell downloads of
recordings from the four major record distribution firms include Amazon.com,
BuyMusic, Napster, Puretracks, WalMart and Zune. 49 During the class period Rhapsody
also sold downloads, but they abandoned this business on April 1, 2013. Several sites
entered the download business before iTS, but iTS was the first site that offered a full
catalog of digital audio recordings from all of the major record distribution companies.
Some other online sellers of permanent downloads, such as eMusic and Ruckus, were not
close substitutes for iTS during the class period because they offered downloads only
from independent distributors or artists who had no distributor. Most download sites,
47. See http://www.consumerreports.org/cro/search.htm?query=portable+cd+players.
48. A search for “editors review portable CD players” on CNet obtained no relevant hits.
See http://reviews.cnet.com/1770-5_7-0.html?query=editors+review+portable+cd+
player&tag=srch&searchtype=products.
49. For descriptions and reviews of audio download sites, see http://music-downloadreview.toptenreviews.com/. I have not included the sites that may not be legal.
CONFIDENTIAL – ATTORNEYS EYES ONLY
32
list is akin to a CD except that the content of the list is selected by the user. A customer
can then listen to a play list on a personal computer or an Internet-enabled wireless
communications device (a smart phone or a tablet computer). Among Internet streaming
options, on-demand services that are available on a mobile wireless device are like
playing audio files on an iPod. And there is evidence that on-demand services have
substituted for downloads. A recent joint study by NPD Group and the National
Association of Recording Merchandisers (NARM) concluded that on-demand music
services detract from sales of sound recordings. 53 In response to this study, ST Holdings,
which owns about 200 record labels, notified Spotify as well as Napster, Rdio and Simfy
that it no longer will allow its recordings to be included in their services due to their
detrimental effect on sales. 54 Hence, the available evidence supports the conclusion that
these services are part of the relevant market that includes iTS.
The problem is that on-demand services that were supported by mobile devices
had no significant number of customers during the class period: only 1.8 million
subscribers in 2011 and 3.4 million subscribers in 2012. 55 The oldest Internet streaming
service is Rhapsody (originally Listen.fm), which acquired rights to recordings from the
How On-Demand Interactive Streaming Services Navigate the Digital Music Rights
Licensing Landscape,” November 21, 2012, at http://papers.ssrn.com/sol3/papers.
cfm?abstract_id=2179111.
53. “Study: Spotify is Detrimental to Music Purchasing,” Digital Music News,
November 15, 2011, at http://www.digitalmusicnews.com/permalink/2011/111115
cannibal#VIZ3-3IxRZUcRMwuQcs_9g.
54. Corey Tate, “Rdio, Spotify and Napster Lose 200 Record Labels Due to NARM
Study,” Spacelab, November 19, 2011, at http://www.thespacelab.tv/spaceLAB/2011/
11November/MusicNews-064-Rdio-Spotify-Napster-NARM-NPD.htm.
55. Joshua P. Friedlander, “News and Notes on 2012 RIAA Music Industry Shipment
and Revenue Statistics,” RIAA, at http://www.riaa.com/keystatistics.php?content_
selector=2008-2009-U.S-Shipment-Numbers.
CONFIDENTIAL – ATTORNEYS EYES ONLY
34
major record companies in 2002. But Rhapsody could not support access by mobile
telephones until 2011 with the release of Rhapsody 5.0, the company’s version of a
digital media player. Likewise, Spotify, currently the most popular on-demand service,
was not launched in the U.S. until the summer of 2011. MOG (“Music On the Go”) was
not available on mobile phones until December 2009.
The reason that on-demand services only recently have been supported on mobile
wireless devices is that the quality of the wireless network only recently was good
enough to allow high-quality live audio streaming. Whereas a download does not need to
be heard as the recording is being received, an on-demand streaming service requires that
the transmission rate of the audio file be fast enough to support high-quality sound
reproduction. At the beginning of the class period, U.S. wireless carriers had rolled out
3G digital wireless service. The original 3G wireless provided data transmission at a
peak rate of 200 kilobits per second (kbs), 56 which was a crucial step in developing audio
programming for mobile wireless devices because it allowed audio services to equal or
surpass the quality of FM radio. Today, U.S. wireless carriers employ 3G wireless
technologies that are capable of sustained bit rates of more than a megabit per second
(mbs) and are in the process of upgrading their networks to 4G technology. 57 4G service
can sustain substantially higher data speeds, thereby making mobile wireless devices
capable of receiving video transmissions that are comparable to DVDs and high-
56 See James Martin, "Mobile Computing: The Newest Wireless Technology," PC
World, November 14, 2002, at http://www.pcworld.com/article/106149/mobile_
computing_the_newest_wireless_technology.html.
57. Federal Communications Commission, Annual Report and Analysis of Competitive
Market Conditions With Respect to Mobile Wireless, Including Commercial Mobile
Services, June 24, 2011, pp. 72-78 (henceforth Mobile Wireless Report).
CONFIDENTIAL – ATTORNEYS EYES ONLY
35
definition cable television.
Reflecting the lack of impact of these sites during the class period, a list of the
best web sites of 2009 includes only one on-demand site – Spotify – and notes that it is
available only in Europe. 58 Billboard, the leading trade publication for the music
industry, did not modify its method for measuring hit records to take into account ondemand streaming services until March 14, 2012. 59 To explain why this new feature was
adopted, Billboard quoted the Vice President of the NARM, who stated” “The last year
has seen an explosion of both subscribers and traffic to music subscription services, and
the business is now contributing meaningfully to the music industry's growing digital
music revenues.” Thus, on-demand services cannot have had any competitive effect on
iTS until long after the end of the class period.
A customized service (examples are Last.fm and Pandora) allows consumers to
list artists and songs that they like and then customizes the play list to suit the consumer’s
preferences. Pandora, the most popular customized streaming service, sponsors the
Music Genome Project, a computer algorithm for classifying music and determining a
consumer’s music preferences. 60 Last.fm’s algorithm for customizing play lists is called
58. Adam Fisher, “The 50 Best Websites of 2009,” Time, August 24, 2009, at
http://www.time.com/time/specials/packages/completelist/0,29569,1918031,00.html.
The list also includes two customized streaming services, Pandora and Last.fm, which are
discussed elsewhere in this report.
59. Billboard Staff, “Hot 100 Impacted by New On-Demand Songs Chart,” Billboard,
March 14, 2012, at http://www.billboard.com/articles/news/502020/hot-100-impactedby-new-on-demand-songs-chart.
60. See http://www.pandora.com/about/mgp.
CONFIDENTIAL – ATTORNEYS EYES ONLY
36
“scrobbling.” 61 On these sites consumers indicate whether they like or dislike a
recording as it is played. The computer algorithm uses this information to construct an
ever-evolving play list.
Most likely customized sites are not close enough substitutes for play lists that are
constructed by the consumer to be in the same relevant market as download and ondemand sites. The play list includes recordings that are new to the user and that the user
may wish to buy. Thus, customized music services are more like preprogrammed radio
stations (terrestrial and satellite) and webcasting than like on-demand services that allow
a consumer to control the play list.
Preprogrammed stations are well-understood to promote record sales, not to
substitute for them. For decades record companies have encouraged terrestrial stations to
play their sound recordings by giving stations free copies of recordings, providing
promotional materials for new releases, and making artists available for interviews.
Record companies also have used “payola” – cash payments and other gifts to disk
jockeys and program directors – to induce radio stations to play their recordings. 62 FCC
rules prohibit broadcasters from accepting fees to promote a particular sound recording or
artist without disclosing that the promotion is an advertisement, regardless of whether the
fee goes to the station owners or to an employee who can influence program content. 63
The FCC’s rules are not mere window dressing. Enforcement actions for these
61. See http://www.last.fm/about.
62. See Ronald Coase, “Payola in Radio and Television Broadcasting,” Journal of Law
and Economics Vol. 22, No. 2 (October 1979), pp. 269-328.
63. The FCC’s rules are described at http://www.fcc.gov/guides/payola-rules.
CONFIDENTIAL – ATTORNEYS EYES ONLY
37
rules occur regularly. 64 In 2007 the FCC settled complaints against four large groups of
radio stations (CBS, Citadel, Clear Channel and Entercom) for accepting cash and other
considerations from record companies in return for playing their sound recordings. 65 The
four groups agreed to pay a total of $12.5 million to the FCC. In 2011, the FCC settled a
complaint against Emmis Austin Radio Broadcasting for accepting payola from a record
store, a concert venue and a booking agent to play recordings by a heavy metal rock band
that was appearing locally. 66 Bribing radio station employees and risking FCC sanctions
would make no sense if record companies did not believe that radio play time induced
greater sales of sound recordings. Additional evidence that non-interactive streaming
services are not regarded as substitutes for permanent sales is that the NARM study did
not identify any customized service as detracting from record sales, and no record
company or artists have refused to allow their recordings to be played on these services.
In any event, even if non-interactive streaming services are substitutes for digital
downloads today, they were not effective competitors during the class period for the same
reason that on-demand services were not. These services only recently became available
on mobile devices. Pandora released its first application that enabled consumers to
64. Since 2007, the FCC has undertaken 17 enforcement actions with respect to this rule.
See http://transition.fcc.gov/eb/broadcast/sponsid.html.
65. Federal Communications Commission, “Broadcasters Pay $12.5 Million to Resolve
Possible ‘Payola’ Violations,” April 13, 2007, at http://transition.fcc.gov/eb/News_
Releases/DOC-272304A1.html.
66. Federal Communications Commission, Order: In the Matter of Emmis Austin Radio
Broadcasting Company, L.P., File No. EB-06-IH-2944, July 22, 2011, at
http://transition.fcc.gov/eb/Orders/2011/DA-11-888A1.html.
CONFIDENTIAL – ATTORNEYS EYES ONLY
38
access this music service on a mobile phone in 2008. 67 Most use of Pandora today is on
mobile devices, but that is a recent event. Pandora estimated that mobile wireless users
accounted for 50.5 percent of its listener hours in 2011, but only 4.6 percent in 2009. 68
The insignificance of on-demand and non-interactive digital streaming services is
documented in Exhibit 1, which shows the breakdown of revenues to the record industry
from various sources. Most revenue from digital sales is from downloads. In 2012,
digital downloads accounted for $2.9 billion in industry sales, compared to $1.0 billion
from all other digital sources, including satellite broadcasting and preprogrammed
webcasting. 69 The latter accounted for 15 percent of industry revenue in 2012, but only 4
percent in 2008 and 5 percent in 2009. Moreover, as shown in Exhibit 1, revenue from
digital sources other than downloads did not change substantially from 2006 through
2008, which includes all but the last quarter of the class period. Thus, digital streaming
services were too small during the class period to be a competitive restraint on audio
download services.
The last candidate for inclusion in the relevant market for downloads of audio
files is the sale of physical copies of sound recordings. Exhibit 1 also shows that sales of
physical copies have declined substantially for the past decade. During the period in
which revenues from downloads grew substantially (2004-2007), physical copy sales fell
by $4 billion. During the period that revenues from downloads grew less rapidly and
then stabilized (2007-2010) sales of physical copies dropped another $4 billion. Clearly
67. Ibid., p. 44.
68. Pandora Media Inc., Form 10-Q for Quarter Ended July 31, 2011, September 2,
2011, p. 33.
69. Friedlander, op. cit.
CONFIDENTIAL – ATTORNEYS EYES ONLY
39
CDs are in decline, but the issue is whether the decline indicates that CDs are a sufficient
competitive constraint on downloads of digital audio files to be included in the relevant
market. The continued decline after download sales stopped growing indicates that CDs
are not constraining sales of downloads. If they were, then download sales would have
captured a large share of the drop in CD sales after 2007, which they have not.
The main attraction of physical copies of audio files is that they generally have
higher sound quality, although as download speeds over the Internet increase, this
advantage is likely to disappear. For several other reasons physical copies are not close
substitutes for downloads. One reason is that the process of buying downloads is more
convenient. Downloads can be obtained immediately, rather than requiring a trip to a
brick-and-mortar store or a wait for delivery if purchased from an Internet vendor.
Another reason is that most recordings are not available as physical copies except as part
of a CD. Consumers of downloads can create their own personalized albums from the
millions of songs that are available on a download web site. A final advantage of
downloads is that storage requires much less space, which is of greater importance to
consumers who have extensive collections of recordings.
A great deal of academic research has focused on whether music on the Internet,
especially illegal file-sharing, is responsible for the decline in sales of CDs. A
comprehensive survey examines nearly 80 studies on the relationship between filesharing and CD sales. 70 This issue is relevant to whether digital downloads compete with
CDs because illegal file-sharing is a form of permanent download at the very attractive
70. Volker Grassmuch, “Academic Studies on the Effect of File-Sharing on the
Recorded Music Industry: A Literature Review,” available at http://papers.ssrn.com/sol3/
papers.cfm?abstract_id=1749579.
CONFIDENTIAL – ATTORNEYS EYES ONLY
40
price of zero – but the very unattractive prospect of being detected and sued by the record
industry. The survey finds that the research on whether file-sharing is responsible for the
decline in CD sales is inconclusive, but “ one lone author – against all basic intuition and
common sense – attributes the entire decline in recorded music sales and even more to
file-sharing” (p. 30).
So what explains the decline in CD sales? The research survey identifies several
causes: (1) a switch in retail sales of CDs from large inventory music stores to low
inventory, big box retailers; (2) a shift in record company strategy to release fewer
records and to focus more on releases by star performers; (3) a temporary sales boom in
the 1990s when consumers replaced libraries of vinyl records and audio tapes with CDs;
(4) a mistaken decision by the record companies to adopt DRM technology and to try to
use it to move to a form of metered use (limiting plays, limiting devices), capped by the
decision by Sony-BMG to include a “root kit” on CDs that enabled the company to
control a customer’s computer; (5) the abandonment of the single recording release in
favor of albums; and (6) the rise of a robust, Internet-based market for used CDs.
A subsequent study since the survey was completed shed a little more light on the
issue. 71 This study uses a survey among French consumers to inquire whether illegal
(free) file-sharing competes with each of the different types of legal Internet music
services as well as CD sales. The study has several interesting findings. First, free
(legal), advertising-supported streaming services have caused a substantial reduction in
illegal file-sharing. Second, Internet streaming services have no effect on CD sales but
71. Godefroy DangNguyen, Sylvain Dejean, and Francois Moreau, “Are Streaming and
Other Music Consumption Modes Substitutes or Complements?” March 16, 2012, at
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2025071.
CONFIDENTIAL – ATTORNEYS EYES ONLY
41
actually promote on-line download sales. In brief, the positive effect of sampling new
material outweighs the negative substitution effect.
The robust conclusion from the research literature is that Internet music is not a
primary cause of the decline in sales of physical recordings. The finding that streaming
has no effect on CD sales but promotes downloads implies that downloads are not
competitive substitutes for either CDs or streaming services.
Market Power
Market power is the ability to control prices or exclude competitors. Economists
use both direct and indirect measures of market power.
Direct measures of market power include the profits and mark-ups of price over
average variable cost and incidents in which a competitor was driven from the market or
abandoned an attempt to enter the market as a direct result of the defendant’s actions. 72
A firm with market power is able to set price-cost margins and to earn profits that exceed
an appropriate competitive benchmark and also is able successfully to defend its sales
and excess profits against attempts by competitors to capture a larger market share. A
substantial, sustained increase in price-cost margins for a profitable product is a reliable
indicator of increased market power because, in a competitive market, prices are driven
towards the long-run average cost of production. Hence, if competitive conditions in a
market do not change, price changes through time should reflect only changes in costs.
72. Another direct indicator of market power is the own-price elasticity of demand (the
responsiveness of sales to price) for the reference product; however, a reliable estimate of
own-price elasticity is not feasible here.
CONFIDENTIAL – ATTORNEYS EYES ONLY
42
Margin Analysis
Profit margins are regularly used by economists and financial analysts to ascertain
the market power of a firm. One standard technique is to measure “pass-through” of a
change in variable costs. 73 Another standard technique is to examine whether a firm’s
percentage mark-up changes in response to an event that potentially could affect the
firm’s market power without affecting its costs or product quality.
For iTS, Apple has produced only highly aggregated data about revenues and
costs. These are not sufficient to undertake a margin analysis for downloads of audio
files, which are the relevant product in this case. Hence, other methods of ascertaining
Apple’s market power in this product market are necessary.
For iPods, two sources of information are available for analyzing profit margins.
One is publications that study Apple’s financial performance in electronic devices, and
the other is data produced by Apple in discovery about revenues and costs by model.
Financial analysts regularly provide interpretations of data about Apple’s publicly
reported sales, costs and profits. For example, one financial analyst reports that the price
difference between two models of iPods that differ only in memory capacity is more than
double the difference in cost. 74 This price difference could not be sustained in a
competitive market, and therefore must be the result of market power. An academic
analysis of the relationship between price and component costs finds that in 2006 Apple
73. Variable costs are the component of cost that depends on output and sales. A profitmaximizing firm will set the price of a product based on its marginal or incremental cost
and its firm-specific elasticity (price responsiveness) of demand. The accounting cost
data that most closely corresponds to marginal cost is average variable cost.
74. “The Mix Get Richer: New iPod touch & iPhone Capacities,” Credit Suisse,
February 5, 2008.
CONFIDENTIAL – ATTORNEYS EYES ONLY
43
Market Concentration
An indirect indicator of market power is seller concentration in the presence of
barriers to entry. A measure of market concentration that economists commonly use is
the Herfindahl-Hirschman Index (HHI), which equals the sum of the squares of the
market shares of the firms in the market. In the presence of barriers to entry, an HHI
exceeding roughly 2500 80 is regarded as sufficient to infer that large firms in the market
possess market power, and an HHI between 1500 and 2500 is sufficient to “warrant
concern” about the intensity of competition. 81 In the presence of barriers to entry, the
2500 benchmark implies that a firm is likely to enjoy unilateral market (monopoly)
power if its market share exceeds 50 percent.
Several private companies regularly collect data about market shares for audio
downloads and portable digital media players. 82
80. A market with four firms of equal size has an HHI of 2500.
81. Merger Guidelines, op. cit., p. 19.
82.
CONFIDENTIAL – ATTORNEYS EYES ONLY
47
,
As discussed above, some portion of cell phone sales should be included in the
market for portable digital media players since smart phones began to be used as an MP3
player. A large survey of mobile subscribers in late 2011, long after the end of the class
period, found that between 21 and 24 percent of mobile subscribers “listened to music”
on their cell phone. Because webcasting and streaming services accounted for some of
this activity, 86 these numbers substantially exceed the fraction of cell phone users who
use their phone as an MP3 player. The proper indicator for Apple’s market power in
portable digital media players includes only the share of smart phone purchases for which
the buyer actually wants to use the device as a portable digital media player.
86. “comScore Reports December 2011 U.S. Mobile Subscriber Market Share,”
comScore, February 2, 2012, at http://www.comscore.com/Insights/Press_Releases/
2012/2/comScore_Reports_December_2011_U.S._Mobile_Subscriber_Market_Share.
CONFIDENTIAL – ATTORNEYS EYES ONLY
49
unilateral market (monopoly) power in the presence of entry barriers.
Barriers to Entry
In addition to market share information, the market structure approach to
ascertaining market power also requires showing that conditions in the market are
conducive to the exercise of market power. The most important of these conditions is the
presence of barriers to entry. A barrier to entry is any condition that would prevent a firm
from either entering a market or expanding its output in a market in which it is already
present. Examples of barriers to entry are high fixed costs that require an entrant to sell a
large amount of output at existing market prices in order to operate profitably and
intellectual property rights that protect an incumbent from competition. Anticompetitive
acts also can create a barrier to entry. An example is tying or bundling. In the presence
of tying or bundling, an entrant must succeed in successfully producing both products,
rather than only one, in order to compete in either market.
One form of entry barrier in the information technology sector is the high fixed
cost of R&D that is necessary to create new products. High fixed costs are a barrier to
entry because they require that a firm be able to set prices above average variable cost
and achieve significant market share in order to find entry into a market attractive.
Portable digital media players were a rapidly evolving technology throughout the class
period, and the R&D effort necessary to produce each succeeding generation of players
was a barrier to entry.
Intellectual property also can constitute a barrier to entry. For example, a major
issue in this litigation has been whether the “crippleware” that is part of Apple’s digital
CONFIDENTIAL – ATTORNEYS EYES ONLY
51
rights management system (for example, the code that prevented Harmony from playing
FairPlay files that was introduced by iTunes 4.7 and iTunes 7.0) was a legitimate exercise
of its intellectual property rights or an anticompetitive act to exclude competitors.
Regardless of the resolution of that issue, crippleware is an example of a barrier to entry.
Likewise, lock-in that is created by technical incompatibility also creates a barrier
to entry. The effect of Apple’s “walled garden” involving iTS, iPods and iPhones, and
iTunes digital media player creates switching costs for users. The lock-in from switching
costs reduces the intensity of competition partly because it creates a barrier to entry.
During the class period, in order to play digital recordings acquired from iTS on a
portable digital media player, a consumer was forced to buy an iPod because only an iPod
could play recordings in the DRM-protected FairPlay format that is used by iTS. Even
after Apple stopped selling recordings encrypted with DRM-protected FairPlay in 2009,
consumers who had a pre-existing library of digital recordings purchased from iTS were
precluded from playing those files directly on any portable digital player other than an
iPod unless they paid Apple to upgrade their files to the iTunes Plus format or burned
their old files to a CD and reloaded them on their personal computer. 92
As discussed in the expert report of Dr. David Martin, RealNetworks attempted to
compete against iTS by inventing a digital media player that could load and play the
digital audio recordings offered by its RMS download service on an iPod. RealNetworks
was thwarted in this effort when the defendant changed its encryption code to defeat the
compatibility between Harmony and iPods. This conduct is an example of how
92. See “iTunes Store: iTunes Plus Frequently Asked Questions (FAQ)” on Apple’s
website, http://support.apple.com/kb/ht1711. iTunes users can upgrade a previouslypurchased song to a DRM-free version for $0.30 or an album for 30% of its price.
CONFIDENTIAL – ATTORNEYS EYES ONLY
52
. 96
The extent to which an iPod user is locked in is determined by the number of
audio recordings in the user’s library that are protected by FairPlay. By mid 2008, near
the end of the class period, iTS sales of audio recordings topped five billion, 97 or an
average of about 70 recordings for every iPod that has been registered by iTS. Most of
these files were in the DRM-protected FairPlay format. Thus, many iPod owners are
likely to own many recordings that cannot be played on any competing brand of portable
digital media player.
Harmony reduced switching costs by giving iPod owners the opportunity to buy
audio recordings from RMS that could be played on other portable digital media players
when the user decided to buy a new one. Had Harmony not been blocked in October
2004, and again in September 2006, iPod users would have had access to another source
of audio files that were compatible with portable digital media players other than iPods
for over three years before audio downloads without DRM protection were available
“This means that the iTunes Music Store, with its catalog of over 1 million songs, works
with 65% of all MP3players and 92% of all hard drive based music players being sold
today. How can anyone say that this is a disadvantage? The iTunes Music Store is the
world’s number one online music store …” (See Apple_AIIA_01384979.)
96. Robert Sample, Stephanie Sun and Thompson Wu, “Happy Holidays!” Credit Suisse,
October 3, 2007.
97. “iTunes Store Tops Over Five Billion Songs Sold,” Apple Press Release, June 19,
2008.
CONFIDENTIAL – ATTORNEYS EYES ONLY
54
from other Internet vendors and over four years before iTS switched to DRM-free audio
recordings. As a result, the proportion of audio downloads that were incompatible with
competing portable digital media players would have begun to decline years earlier.
If iPod users who purchased audio recordings from iTS and from RMS using
Harmony decided to replace an old iPod, the cost of switching to a competing product
would have been lower had Harmony survived, thereby intensifying competition between
iPod and other brands. Given the rapid technological progress in portable digital media
players during the class period, Harmony plausibly would have increased competition
against iPods substantially before DRM-free audio recordings became available. By
updating iTunes to block interoperability with Harmony, Apple preserved the lock-in of
iPod owners to iTS and iPods.
The switch to DRM-free audio recordings is likely to have reduced the lock-in of
iPod users. The transition to DRM-free content was slow, proceeding for nearly two
years. In April 2007, EMI announced a “premium” version of downloads of audio files
(singles, but not albums, selling for a higher price), and iTS was the first to offer these
downloads, with others expected to follow in a few weeks. 98 In May 2007, iTS
introduced iTunes Plus, which initially offered recordings from EMI, some independent
record labels, and some unaffiliated artists in an unprotected format. iTunes Plus
recordings could be loaded onto some portable digital media players other than iPods,
although doing so required manipulation of the files using both iTunes and another digital
media player. Other audio download sites also made deals with EMI, but other sites
98. “EMI Launches DRM-Free Superior Sound Quality Downloads Across Its Entire
Digital Repertoire,” Webwire, April 3, 2007, at http://www.webwire.com/ViewPressRel.
asp?aId=31281.
CONFIDENTIAL – ATTORNEYS EYES ONLY
55
began to offer DRM-free EMI files in August.
Also in August 2007, Universal announced that it would conduct an experiment
from August 2007 to January 2008, in which it would release some DRM-free audio
recordings to several audio download services, including Amazon.com, Best Buy,
Google, RMS and WalMart. 99 Several download sites were launched in August and
September that offered some Universal DRM-free recordings.
Universal and the other two major distribution companies, Sony-BMG and
Warner, committed to sell a large number of audio recordings without DRM protection
between late December 2007 and early January 2008, and the transition to offering a
large inventory of DRM-free audio recordings by audio download services was complete
in March 2008. In January 2009, Apple announced that it would sell audio recordings
from all of the major distribution companies without DRM protection, and by April 1,
2009, a large repertoire of audio recordings could be purchased on iTS without DRM
protection. As a result, audio files from many sites can be loaded, catalogued and played
on an iPod, and DRM-free audio recordings from iTS can be played on other portable
digital media players, although doing so requires using two digital media players.
The widespread availability of DRM-free audio downloads should have reduced
the lock-in of iPod owners to both iTS and replacement iPods. DRM-free audio
recordings enable iPod users to buy audio recordings from competitors of iTS. As time
progressed, a larger proportion of a consumer’s library, and all of the more recent
acquisitions, were in a DRM-free format. For some users, the value of older DRM-
99. Colleen Bowen, “Universal Launches DRM-Free Music Test,” TWICE, August 10,
2007, at http://www.twice.com/article/258659-Universal_Music_Launches_DRM_Free_
Music_Test.php.
CONFIDENTIAL – ATTORNEYS EYES ONLY
56
protected recordings probably will decline, in which case some users may decide that the
loss of the ability to store and play old audio recordings on a new portable digital media
player is no longer an important reason not to replace an old iPod with another brand.
Conclusions on Market Power
Based on the evidence about the nature of the products at issue in this litigation,
the performance of the products in the market, and Apple’s internal documents and data, I
conclude that Apple enjoyed market power in the relevant markets for downloads of
digital audio files and portable digital media players. The crucial evidence here is the
presence of persistent monopoly power in iPods. The design of the iTunes digital media
player and iTS created a lock-in to iPods that made the latter products largely immune to
intense competition on both price and quality from other electronics firms. Whereas the
initial attainment of monopoly power in iPods arose because of the first-in advantage in
downloads of digital audio files that was given to Apple by the record companies,
Apple’s DRM technology played a critical role in maintaining and enhancing that
advantage. The iTunes 7.0 update re-established the lock-in that had been eroded by
Harmony, thereby enhancing the barrier to entry in portable digital media players due to
lock-in. As a result Apple enjoyed both a high market share and a record high profit
margin iPods after Harmony was disabled through the end of the class period. The effect
of this lock-in is quantified in the damages analysis elsewhere in this report.
ANTICOMPETITIVE SOURCES AND EFFECTS OF MONOPOLY POWER
The issue in this litigation focuses on conduct by Apple in 2006 and 2007 to raise
CONFIDENTIAL – ATTORNEYS EYES ONLY
57
switching costs from iPods to competing portable digital media players. The plaintiffs’
allegation pertaining to this conduct has two components. The first is a test from antitrust
economics as to whether conduct that creates lock-in is anticompetitive. The second is
the empirical question of the quantitative significance of the lock-in effect as measured
by its effect on market prices. In this case the quantifiable harm to competition is the
overcharge on iPods that was the result of the conduct that created lock-in.
Tests for Anticompetitive Conduct
If a firm enjoys market power, economic analysis can be used to determine
whether its market power is due all or in part to anticompetitive acts. Firms may enjoy
market power due to “superior foresight and efficiency,” i.e., their products are cheaper
and/or better because they have superior technology and/or management. For example,
innovations that are protected by valid intellectual property rights or that otherwise are
difficult for competitors to copy can be a source of market power. Likewise, if the
production technology in an industry exhibits economies of scale that are sufficiently
strong that only a small number of firms can achieve the minimum efficient scale of
production, firms are likely to enjoy market power and to earn excess profits. In antitrust
economics, obtaining market power from these sources is not anticompetitive.
Firms also may acquire or maintain market power by anticompetitive means. The
ultimate test for whether conduct is anticompetitive is whether it harms consumers. In
antitrust economics, conduct is regarded as unambiguously anticompetitive if it increases
or maintains market power, does not improve the quality or diversity of products
available, is unrelated to the legitimate protection of intellectual property rights, and
CONFIDENTIAL – ATTORNEYS EYES ONLY
58
requires costly action by the firm that undertakes it. An act may be anticompetitive even
if it provides benefits to consumers and is not costly to the firm with market power, but
only if no reasonable alternative means can obtain the same benefits to consumers.
Technical Incompatibility as an Anticompetitive Source of Market Power
Apple’s conduct to re-establish the lock-in of iPod owners was anticompetitive if
it was costly to implement, provided no benefit to consumers, but increased profits only
because it increased Apple’s market power. The expert report of Dr. David Martin shows
that the changes in the DRM system that were associated with iTunes 7.0 and iTunes 7.4
produced no benefits to consumers. But the creation of technical incompatibility without
a consumer benefit has a direct cost and an opportunity cost.
The direct cost is the incremental cost of creating incompatibility. The issue here is not
that the defendant had to incur costs to implement its proprietary file format, but that the
defendant was forced to incur additional costs for actions that had no purpose other than
to create or maintain incompatibility. Lines of code are also a rough indicator of the cost
of software.
Opportunity cost in this context refers to the sacrifice of sales in one product in
order to create and maintain incompatibility between its competitors and the other
product. The issue here is whether a vendor that owned only a download site (e.g., iTS)
and the software that was needed to access that site (e.g. iTunes) has a profit incentive to
make that site compatible with only one brand of portable digital media player. The
CONFIDENTIAL – ATTORNEYS EYES ONLY
59
answer, of course, is no – an independent owner of iTS would want to make the site
accessible to enough portable digital media players to ensure that the player market was
competitive. Indeed, RealNetworks pursued this path by working with several
manufacturers of portable digital media players to produce products that could access
RMS and its streaming service, Rhapsody. Harmony was a component of this strategy as
it sought to make iPods compatible with RMS.
The same argument applies to an independent supplier of iPods. To maximize
sales, a stand-alone supplier of iPods would make the product compatible with download
sites other than iTS. An independent iPod supplier would like to lock-in iTS customers
by making iTS compatible only with iPods, but it would not want to prevent customers of
competing download sites from buying iPods. For this reason, an independent iPods
supplier would not take actions to prevent RMS customers from using iPods.
The benefits of lock-in to Apple consisted of increased profits from the lock-in.
As discussed elsewhere, the market share of iTS hit an historical low after Harmony was
released, but iTS then experienced a record market share and profit margin after
Harmony was disabled and the incompatibility between RealPlayer and iPods was reestablished. These additional sales and profits arose despite the fact that, as discussed by
Dr. Martin, the changes in iTunes 7.0 did nothing to benefit consumers by improving the
performance of iTS or iPods.
Harm to Competition
In antitrust economics, “harm to competition” by a seller with market power
refers to reductions in the welfare of consumers. In this litigation one alleged harm to
CONFIDENTIAL – ATTORNEYS EYES ONLY
60
consumers is higher prices for iPods. The Amended Complaint alleges that because it
excluded actual and potential competition by updating its software to preclude
interoperability, Apple was able to maintain and enhance its market power in the relevant
market for portable digital media players. As a result, plaintiffs allege, the prices of
iPods were higher than they would have been in the absence of the defendant’s
anticompetitive acts. 100 This effect is the source of damages, so the discussion of the
quantification of this harm is discussed in the section about damages. Suffice to say here
that damages are calculated from an econometric model of iPod pricing that quantifies
the effect of the lock-in that was created by disabling Harmony. The econometric model
also quantifies the pro-competitive benefits arising from the introduction of Harmony and
the end of DRM-protected audio files. This analysis shows that the magnitude of the procompetitive benefit of DRM-free audio not only was substantial, but was greater than the
anticompetitive harm due to iTunes 7.0.
Consumers also can suffer financial harm in ways that normally are not included
in the calculation of damages. One example is the “dead-weight loss” arising from
higher prices. 101 Dead-weight loss is the loss of welfare arising from the reduction in
output that occurs when prices exceed the incremental cost of production. An
approximation of dead-weight loss is ½(Pm – Pc)(Qc – Qm), where Pm and Pc are the
prices under monopoly and competition, and Qm and Qc are the quantities sold under
monopoly and competition.
Because the conduct at issue in this case caused iPod prices to be higher, sales of
100. Amended Complaint, pp. 16-17.
101. Christopher R. Leslie, “Antitrust Damages and Deadweight Loss,” Antitrust
Bulletin, Vol. 51, No. 3 (Sept. 2006).
CONFIDENTIAL – ATTORNEYS EYES ONLY
61
iPods thereby were lower. Replacement customers, who account for about half of iPod
purchases, are likely to be especially sensitive to changes in price. Because they already
enjoy the services of an older model, they have less to gain from purchasing a new model
than a customer who does not have a portable digital media player. The dead-weight loss
here arises because customers respond to higher prices by increasing the duration of the
replacement period.
Once DRM-free audio files became available, consumers could overcome lock-in
by repurchasing audio files that had been protected by FairPlay. These consumers would
not be included in the damage calculation in this case if they purchased DRM-free files
and then switched from an iPod to another player, but they still would have suffered
harm.
Defendant’s anticompetitive acts also can harm consumers by reducing the
intensity of competition among other firms in the market. These firms may charge higher
prices for other products in the relevant market, either because they cannot take away a
significant amount of business from dominant incumbents by lowering their prices or
because the market power of dominant firms prevents them from achieving scale
economies that would lead to lower prices if the market were more competitive.
One way that lock-in can cause harm to competition is by slowing technological
progress. The incentive to innovate is provided by the sales that a firm expects to make if
CONFIDENTIAL – ATTORNEYS EYES ONLY
62
it produces a new product with lower cost, higher quality and new features. Lock-in
reduces expected sales from innovation because switching costs increase the incremental
value of a new product that is necessary to induce locked-in customers to switch. Hence,
lock-in reduces the incentive to innovate. A 2012 review of portable digital media
players states: “If you can save money by buying last year's model, go for it. The
innovation in MP3 players has been flat for years and a 2011 MP3 player is going to
work just as well as one released today.” 102
Business Justifications
A business justification is a benefit to consumers arising from an act that reduced
competition. Higher profits and greater sales are not business justifications. Instead, an
act that causes anticompetitive harm is reasonable if it provides benefits to consumers
that cannot be obtained by any reasonable alternative, less anticompetitive means.
Although the business justifications that Apple will offer at trial have not yet been
submitted, the defendant’s submissions earlier in this litigation assert the following
justifications. First, maintaining incompatibility between Apple’s products and its
competitor’s products benefited consumers because “Apple’s products worked better
together than with competitors’ products.” 103
102. Donald Bell, “MP3 Player Buying Guide,” CNET Reviews, November 5, 2012, at
http://reviews.cnet.com/mp3-player-buying-guide/.
103. Apple’s Reply in Support of Motion to Dismiss or, Alternatively, for Summary
Judgment, p. 1.
CONFIDENTIAL – ATTORNEYS EYES ONLY
63
participate in these markets, only Apple and Microsoft sold products that were
technically incompatible with the products of competitors. 111
The analysis of business justifications must address whether consumers are
harmed by the opportunity to deal with multiple vendors for each product (rather than
only for the group of products), and whether the opportunity for choice among vendors of
products in each relevant market offsets the benefit, if any, of integration from a single
vendor. Consumers normally do not need vendors to protect them against considering
and then rejecting a marketing message that seamless integration is valuable.
In the past decade, after the Microsoft case ended, two new wireless-enabled
computers – smart phones and eReaders/tablet computers (initially separate products but
now converged) – were enthusiastically received by consumers. Apple is successful in
both products, but both markets are much more competitive than the market for portable
digital media players, and both markets seem destined to be dominated by Google’s
Android operating system, which supports many brands of products.
In smart phones, many brands of smart phones use Android, the most successful
of which is the Samsung Galaxy. The share of smart phones that use Android rose from
51 percent in the fourth quarter of 2011 to 71 percent in the fourth quarter of 2012, while
Apple iOS (the operating system for the iPhone), fell from 24 percent to 21 percent. 112
111. Initially, Microsoft’s Windows media player software, online music store, and Zune
portable digital media player were compatible with the products of all vendors in the
relevant markets other than Apple. In 2006, Microsoft developed new versions of all
three products, all named Zune, that, like Apple’s products, were incompatible with the
products of other vendors, and closed its MSN Music Store, which had sold downloads in
DRM-protected recordings in the WMA and WMV formats for use on multiple players.
112. Chuck Jones, “Android Solidifies Smartphone Market Share,” Forbes, February 13,
2013, at http://www.forbes.com/sites/chuckjones/2013/02/13/android-solidifiesCONFIDENTIAL – ATTORNEYS EYES ONLY
66
Among Android smart phones, 47 percent are accounted for by Samsung and 37 percent
are accounted for by Kindle, the eReader from Amazon. 113
In tablet computers, the iPad, being the first product of its kind, had a first-in
advantage that was similar to the iPod after iTS was introduced, but Apple’s market share
is now below half and falling as others have entered with quality products. 114 A recent
study by an industry analyst concluded: “Initially, Apple with its iOS dominated the U.S.
tablet PCs market; however, Android based tablet PCs are expected to take over Apple's
share in the coming years.” 115 Except for portable digital media players, the experience
in consumer electronics during the new millennium is that consumers prefer choice – to
mix complementary products from different vendors.
In the relevant markets in this case, the only other firm that tried to construct a
“walled garden” of products that were incompatible with all other products was Microsoft
with Zune. The first Zune, released in 2006, received mixed reviews, but the second
smartphone-market-share/ (reporting sales data from Gartner).
113. Chuck Jones, “Samsung Dominates Android Devices,” Forbes, February 7, 2013, at
http://www.forbes.com/sites/chuckjones/2013/02/07/samsung-dominates-androiddevices/ (reporting data from Localytics).
114. Shane Richmond, “Apple iPad Market Share Falls as Tablet Market Booms,” The
Telegraph, January 31, 2012, at http://www.telegraph.co.uk/technology/apple/9839631/
Apple-iPad-market-share-falls-as-tablet-market-booms.html (reporting NPD data for the
fourth quarter of 2012). See also Nikhil Subramanium, “Apple’s Tablet Market Share for
Q4 Falls in Face of Samsung Surge,” Tech2, February 2, 2013, at http://tech2.in.com/
news/tablets/apples-tablet-market-share-for-q4-falls-in-face-of-samsung-surge/733692
(reporting sales data from IDC for 2012).
115. Press release from Research and Markets, reported as “Research and Markets:
Tablet PC Market – U.S. Industry Analysis, Size, Share, Growth and Forecast, 20122018,” Yahoo Finance, February 5, 2013, at http://finance.yahoo.com/news/researchmarkets-tablet-pc-market-161100545.html.
CONFIDENTIAL – ATTORNEYS EYES ONLY
67
release, in 2007, was received very favorably. The CNet review stated: 116
“Having survived its freshman hazing, the Zune is back for
its sophomore revenge, and the iPod has every reason to be
frightened. With a new design, higher capacity, wireless
sync capability, larger screen, and integrated support for
audio and video podcasts, the new 80GB Zune ($249) is
finally giving everyone a true alternative to the iPod.”
The CNet editors gave the Zune the same rating (four stars, excellent) as the iPod Touch.
But consumers rejected Microsoft’s product. The Zune never achieved significant market
share, and eventually was withdrawn from the market.
The experiences with consumer electronics over the last decade do not support the
conclusion that integrated systems are better. When consumers are given the choice,
most prefer products that do not lock them in to a single vendor for multiple products.
DAMAGES
In antitrust economics, damages are calculated by comparing actual prices for the
reference product with prices that would have been charged in the “but-for” world in
which the alleged anticompetitive acts had not occurred. A damage analysis estimates
prices in a hypothetical more competitive market – the “competitive benchmark” – that
would have been present had there been no anticompetitive conduct. 117
In analyzing damages in this matter, the task is to calculate the extent to which the
alleged anticompetitive conduct enabled the defendant to set higher prices for iPods than
116. “Microsoft Zune,” CNet, November 7, 2007, at http://reviews.cnet.com/mp3players/microsoft-zune-second-generation/4505-6490_7-32638989.html.
117. In their text Professors Kip Viscusi, Joseph Harrington and John Vernon state the
general principle: “Standard antitrust practice is to calculate damages... as the additional
revenue on the units sold.” W. Kip Viscusi, Joseph E. Harrington, Jr., and John M.
Vernon, Economics of Regulation and Antitrust, 4th Edition, MIT Press, 2005, p. 145.
CONFIDENTIAL – ATTORNEYS EYES ONLY
68
otherwise would have been charged if the market for portable digital media players had
been more competitive. The competitive benchmark is not necessarily an intensely
competitive market. Instead, it represents the degree of competition that would have
been present had the anticompetitive acts not occurred, which in some circumstances is
an oligopoly. Thus, a valid damage analysis must take into account that in the absence of
anticompetitive conduct, Apple would have enjoyed some market power in iPods.
General Considerations in Estimating Damages
The class in this matter includes both end-users and intermediaries who bought
iPods from the defendant. These two types of customers paid different prices.
Consumers paid retail prices from the online or traditional retail outlets of the Apple
Store, while wholesale distributors and large retail competitors of the Apple Store paid
reseller prices from Apple’s wholesale distribution operation. Consequently, the method
for calculating damages should take into account whether the product was sold at retail or
wholesale. If markets are competitive, the wholesale and retail prices of a firm that
operates in both markets differ according to the firm’s sales costs in the two distribution
channels; however, a firm that enjoys market power in manufacturing may have the
power to engage in effective price discrimination among categories of buyers. If so, the
amount of damages per unit sold will differ between retail and wholesale buyers.
Methods of Damage Estimation
Economists use three basic approaches to establishing competitive benchmark
CONFIDENTIAL – ATTORNEYS EYES ONLY
69
prices: “before-after,” “yardstick,” and “mark-up.” 118 My experience in other class
action litigation provides further evidence that these methods are standard within the
economics profession and in antitrust litigation. I have successfully proposed one or
more of these methods for calculating damages in several class action antitrust cases
dealing with many different types of products: luxury tableware, 119 dynamic random
access memory (DRAM), 120 static random access memory (SRAM), 121 compact discs, 122
118. These approaches are described and widely accepted in scholarly writings in
antitrust economics. Professors Roger Blair and David Kaserman devote a section of
their text to damages calculation. They summarize the standard approaches in antitrust
economics by stating that “the measure of damage is roughly equal to the wealth
transferred to the monopolist from the buyers.” Roger D. Blair and David L. Kaserman,
Antitrust Economics, Richard D. Irwin, 1985, p. 78. They list “three basic theories or
how one goes about measuring... overcharges” as the “before and after theory,” the
“yardstick theory,” and the “market share theory.” Ibid., pp. 78-79. (The last is only
relevant to estimating the lost profits of a competitor that is harmed by anticompetitive
conduct, so is not pertinent here.) The “yardstick approach to damage estimation is
based upon a comparison of the plaintiff’s experience with that of a firm or market that
was unaffected by the illegal activity... A plaintiff that is claiming damage due to
overcharges may attempt to compare the prices it paid with those charged in similar
markets where there was no antitrust violation.” The before-after and yardstick methods
are described in John Johnson, “Economic Approaches to Antitrust Damage Estimation,”
National Economic Research Associates, January 2005. All three methods (with the
mark-up approach separated into three ways that it can be implemented) are discussed in
John M. Connor, “Forensic Economics: An Introduction with Special Emphasis on Price
Fixing,” Journal of Competition Law and Economics, Vol. 4, No. 1 (March 2008), pp.
31-59. According to Professor Connor: “The principal challenge for forensic economists
is to calculate the relative competitive benchmark price...” Ibid., p. 45. He then goes on
to describe the principal methods of calculating damages as the “before and after
method” (which he dates to the 1920s), the “yardstick method” (which he notes has been
used in cases involving bread, milk and construction services), the “cost-based
approach,” the “constant-margin approach” (which was used in the Vitamin E
conspiracy), and the game theory method. Ibid., pp. 46-53. His yardstick approach is the
same method that I call the yardstick method, and his constant-margin and game-theory
approaches are the methods that I call the mark-up test.
119. In Re Tableware Antitrust Litigation, U.S. District Court, San Francisco, California.
120. In Re: DRAM Antitrust Litigation, U.S. District Court, San Francisco, California.
CONFIDENTIAL – ATTORNEYS EYES ONLY
70
and repair and maintenance of high-speed photocopiers 123 and body imaging devices. 124
After having examined the data and the other evidence in this litigation, I have
decided that the best method for calculating damages as well as proving anticompetitive
impact is the “before-after” method. The “before-after” method compares prices of the
reference products (here, iPods) before and/or after the occurrence of the anticompetitive
acts with prices during the damage period. I have implemented this method by
estimating an econometric model of price formulation for iPods.
The basic approach that I have adopted starts with a hedonic model of iPod prices
that takes into account the fact that Apple sells many models of iPods with different
features. In a hedonic equation, price is expressed as a function of the qualitative
attributes of the product. 125 The qualitative attributes that are included in the regressions
are indicator variables for each product class (classic, mini, nano, shuffle, touch), an
indicator variable for the U2 special editions, the logarithm of time measured as the
number of months between the current month and the month in which the iPod was
introduced (2001) (to represent technical change not captured in other variables),
121. In Re: Static Random Access Memory (SRAM) Litigation, U.S. District Court, San
Francisco.
122. Consolidated Compact Disc Antitrust Litigation, U.S. District Court, Los Angeles,
California.
123. R&D Business Systems v. Xerox, U.S. District Court, Marshall, Texas.
124. Southeast Georgia Regional Medical Center vs. General Electric Corporation, U.S.
District Court, Brunswick, Georgia.
125. The hedonic model that I have estimated is similar to a model that was used to
estimate a price equation for mobile telephones. See Ralf Dewenter, Justus Haucap,
Ricardo Luther, and Peter Rotzel, “Hedonic Prices in the German Market for Mobile
Phones,” Information Economics and Policy 31 (2007), pp. 4-13.
CONFIDENTIAL – ATTORNEYS EYES ONLY
71
126
Finally, indicator variables are used to take the value of one during periods when
some factor that is likely to affect competition in the market is present and zero
otherwise. The structure of the model is similar to an “event study,” a commonly used
procedure in financial economics to detect the effect of an unanticipated event on stock
prices. Among the events for which an indicator variable was created are the period
when Harmony was operational, the periods after it was disabled (by iTunes 4.7 and by
iTunes 7.0), and the periods when the competitors and then iTS sold DRM-free audio
126. Multicollinearity occurs when two or more independent variables are so highly
correlated that the coefficients in the model are imprecisely estimated. If the data set is
large, the only solution is to eliminate some of the correlated variables.
CONFIDENTIAL – ATTORNEYS EYES ONLY
72
files. The coefficients on these indicator variables quantify the competitive effects of
changes in the extent of compatibility between iPods and competing download sites.
Apple introduced the first iPod in 2001. While the initial versions of the iPod
were successful, the iPod was not dominant in the market for portable digital media
players. The initial event that caused Apple to obtain monopoly power in both audio
downloads and portable digital media players was the launch of iTS along with an iPod
that was the only portable digital media players that could play audio recordings in the
FairPlay DRM system. These products were launched in April 2003. The initial period
before iTS was created and the period after the launch of iTS establish the “before”
period for measuring Apple’s market power prior to the events surrounding the release of
Harmony, the attempts to disable Harmony, and the movement to DRM-free files. This
procedure provides a separate estimate of the price effect of each event that changed
competitive conditions in the market for iPods.
Two important events occurred in October 2003. First, iTS became accessible on
Windows-based personal computers, and second, the major record labels licensed
competing web sites to sell audio downloads with DRM protection. During a transition
period between October 2003 and February 2004, several Internet sites – including
Napster, RMS and WalMart – became fully operational as competitors to iTS. By March
2004, iTS had several competitors in audio downloads. The effect of iTS on iPod prices
is captured by an indicator variable that takes the value of one after iTS is launched and is
zero for the period before that date.
Harmony was introduced in July 2004. Harmony enabled many brands of
portable digital media players, including iPods, to be compatible with the RealNetworks
CONFIDENTIAL – ATTORNEYS EYES ONLY
73
audio download service. Harmony converted files that used the RealNetworks Helix
DRM protection into either the protected WMA or FairPlay format for use on an
otherwise incompatible portable digital media player. The important aspect of Harmony
for purposes of damage estimation is that it enabled iPod users to download audio files
from a site other than iTS. Harmony was operational until Apple’s software upgrade of
October 2004, iTunes 4.7. RealNetworks apparently made another attempt in 2005 to
restore Harmony’s compatibility with iPods, but Harmony again was made incompatible
with iPods with the iTunes 7.0 update in September 2006. RealNetworks never
overcame the incompatibility created by iTunes 7.0.
The indicator variable for Harmony takes the value of one for all periods after
Harmony was launched. This procedure is followed for two reasons. First, the iTunes
7.0 update applied only to new iPods. Harmony continued to work on old iPods, which
meant that consumers who had used Harmony on old iPods continued to be able to do so,
which would have affected the extent to which they were locked in to iPods. Second, this
procedure permits separating the positive competitive effect of Harmony from the
negative competitive effect of disabling it after iTunes 7.0 was released. If the Harmony
variable were zero during the periods that it was disabled by iTunes 4.7 and iTunes 7.0,
the coefficients on the latter variables would be the net effect of those events under the
assumption that Harmony did not exist. To separate the effect of these updates from the
effect of Harmony requires setting the Harmony variable to one for the entire data period
after its launch.
Between October 2004 and April 2005, and then again from September 2006 on,
updates to the Apple DRM system made Harmony incompatible with new iPods.
CONFIDENTIAL – ATTORNEYS EYES ONLY
74
Indicator variables for the period in which iTunes 4.7 and iTunes 7.0 made Harmony
incompatible with iPods also are included in the model. Between May 2007 and March
2009 legal audio download sites made a gradual transition to DRM-free audio recordings.
EMI allowed DRM-free downloads in April 2007, and iTS began selling EMI audio
downloads without DRM protection in May 2007. Competitors to iTS began selling
some DRM-free downloads from EMI and Universal in August 2007. All four other
major record companies allowed Apple’s competitors to sell a complete digital repertoire
of DRM-free downloads by January 2008, and by March 2008 this change in policy had
been fully implemented. By January 2009, all major record companies had agreed to let
iTS sell DRM-free downloads, which was fully implemented on April 1, 2009.
Competing audio download sites had an advantage compared to iTS until iTS began to
provide complete DRM-free downloads. Separate indicator variables are used for the full
implementation of DRM-free audio files by competitors and by iTS.
The periods for the various indicator variables that measure potential competitive
effects are: pre-iTS (November 3, 2001 to April 27, 2003); post-iTS (from April 28,
2003 to the end of the period for which Apple has produced transactions data); Harmony
available (beginning July 26, 2004 to the end of the data period); Harmony disabled with
update 4.7 (beginning October 26, 2004 and ending September 11, 2006 when update 7.0
was released); update 7.0 (beginning September 12, 2006 and running to the end of the
data period); iTS competitors fully DRM-free (beginning January 1, 2008 and running to
the end of the data period); and iTS fully DRM-free (beginning March 31, 2009 and
running to the end of the data period). The pre-iTS period is the omitted category.
In addition to events affecting competition in the market, other factors also could
CONFIDENTIAL – ATTORNEYS EYES ONLY
75
affect the price of iPods. The demand for iPods is likely to depend on the availability of
digital downloads, so that one factor affecting price changes for iPods over time may be
the number of permanent downloads that are available on iTS. Moreover, digital media
players, like all electronics products, experience rapid technological progress. To take
technology into account, economists frequently use a time trend as well as a measure of
unit cost.
The specification that I have concluded produces the most reliable estimate of
damages is to regress the logarithm of price on indicator variables plus the logarithm of
variables that are continuous; however, I also have estimated the linear specification. I
have estimated separate equations for the two types of class members, resellers
(wholesale customers) and direct purchasers (retail customers).
In the preferred
regression I have eliminated the outlier observations; however, I have also estimated the
regressions with the outlier observations included. Finally, to avoid bias in coefficient
estimates arising from the fact that the data set contains a very large number of very small
purchases, I have weighted each observation by the quantity sold.
The two preferred regressions (logarithmic equations with outliers removed) are
reported in Exhibit 13; the other regressions (logarithmic equations with outliers
included, and linear equations with and without outliers) are contained in Appendix C.
The linear specifications produce higher estimated damages, but I believe that they are
CONFIDENTIAL – ATTORNEYS EYES ONLY
76
less reliable. The other logarithmic equations produce results that are close to the results
of the preferred equations.
The calculations of damages from these regressions are reported in Exhibits 14
through 16. The method for calculating damages for each type of buyer is to multiply the
percentage mark-up during the iTunes 7.0 period times the average transaction price for
each model, defined by a class, generation and family as they appear in Apple’s data
base. The disaggregated damages are then aggregated to a class and generation, then to
just a class, and then to a total for each type of buyer.
Exhibit 15 disaggregates damages to the level of class
and generation, while Exhibit 16 further disaggregates damages to the level of class,
generation and family.
Reseller Price Regression
This section reports the results of the price regression on the current version of
Apple’s reseller transactions data. The reseller transactions data cover the period from
November 3, 2001, through March 26, 2011. Each record refers to a shipment to a
reseller.
Because of the remaining problems
and unanswered questions about the transactions data, many records were eliminated
CONFIDENTIAL – ATTORNEYS EYES ONLY
77
from the data base for purposes of econometric estimation. The transactions that were
deleted are as follows.
1. Records with missing shipment dates.
2. Transactions involving refurbished products.
3. Transactions reporting non-positive quantities or revenues.
4. Records with a missing entry for a data field that is required in the regression.
5. Sales to individuals and other non-resellers (per May 3, 2011 email from
Apple counsel).
For each customer, a variable was created measuring the total purchases by that
customer in the same quarter as a particular transaction occurred. Because so many
CONFIDENTIAL – ATTORNEYS EYES ONLY
78
variation in the dependent variable, and an adjusted R2 of 0.0 indicates that the equation
explains nothing. The values of the adjusted R2 for this regression is 0.9866, which
means that virtually all of the variation in prices across models of iPods and among time
periods is explained by these equations.
Another test of the quality of a regression is whether the estimated coefficients are
consistent with expectations derived from economic theory. For example, prices for
consumer electronics generally fall through time due to the presence of ubiquitous
learning-by-doing and Moore’s Law. The positive coefficient on the logarithm of time
bears out this expectation.
One result of interest is the coefficient on unit cost, which is positive and highly
significant in all specifications. While this result shows substantial cost pass-through, the
pass-through rate also is substantially less than one. This result is consistent with the
research publications cited elsewhere that Apple engages in price discrimination among
products, which is another indicator of the presence of market power.
relationship between the dependent variable and the additional explanatory variables.
The version of R2 that is reported here takes this phenomenon into account, and so is a
more accurate measure of the explanatory power of the regression than unadjusted R2.
CONFIDENTIAL – ATTORNEYS EYES ONLY
80
Damages are calculated by multiplying the percentage increase in price due to the
iTunes 7.0 update, which is calculated from the coefficient on the iTunes 7.0 variable, by
the price of each class/family/generation of iPod. Exhibit 14 shows the average damage
per unit sold and the total amount of damages to resellers for the entire class period.
Exhibit 15.1 disaggregates the reseller damages to a class and generation of iPod, and
Exhibit 16.1 further disaggregates the reseller damages to a class/generation/family.
Direct Sales Regressions
The direct sales transaction data cover the period from November 2001 through
December 2011. Each record has a line item on an invoice of a direct purchase of an
iPod.
The price is calculated by dividing the “amount”
variable by the “quantity” variable of that particular line item on the invoice.
CONFIDENTIAL – ATTORNEYS EYES ONLY
81
Major Data Issues
Apple has not provided sufficient information to link all returns and exchanges to
the original sales. As a result the data that were used in the regressions may have some
double counting. For example, a customer might order an iPod online, then return it to an
Apple store to exchange it for another color or to trade in the initial model for another
model at a different price.
Returns data are important because they imply that a prior transaction should be
eliminated or, if the return is a trade-in, adjusted. The transactions data contain some
information about returns, but Apple has not provided sufficient information to identify
all returns and exchanges. Ideally, if Apple had provided complete information, cleaning
the data would have involved identifying price-related adjustments to all sales. For
returns for credit, the entire transaction could be removed from the data. For adjustments
to original sales, the true transaction price could be calculated from the set of records
involving the same sale.
CONFIDENTIAL – ATTORNEYS EYES ONLY
83
the original sale and subsequent returns. Apple provided no dictionary for the variables
in the data. The text files contain an invoice identifier and original invoice identifier, but
these files provide neither an invoice item number that can be used to match exactly with
a record in the transactions data nor the original invoice item number that would also
allow a match with the original sales data. This was the first time Apple produced data
that included the original invoice identifier, making clear that documents Apple had
provided previously lacked a key linkage variable.
In response to Plaintiff’s request for clarification, Apple provided vague and
inaccurate answers and sent another file, “PD2_Invoice_Xref_CreditsDebitsReturns.zip,”
which contains a similar linkage for online sales. This file contains a different set of
variables, but does include an invoice identifier and item number as well as the
corresponding original invoice identifier and original item number information.
The newly produced data files were used to attempt to match returns and
adjustments with their corresponding original sales. The “returns.zip” file (also known as
PD3 for retail sales) contains 2,880,937 records and 11 variables, of which six were not
provided previously. The five variables that may be used to match with the original file
are invoice identifier, product identification, billing quantity, billing amount, and original
invoice identifier. When comparing specific records, the billing amount in this new file
did not match the billing amount contained in the original direct sales file. The likely
difference between PD3’s billing amount and the billing amount in the transactions data
is a restocking fee. Without further explanation from Apple, a proper investigation of
this possibility was not feasible in the few days between the receipt of the data and the
deadline for the report.
CONFIDENTIAL – ATTORNEYS EYES ONLY
85
credit or any adjustments using Apple’s cross reference files were deleted from the
regression analysis to minimize biases that may be introduced by including inaccurate
prices. This procedure also is likely to cause damages to be underestimated because
potentially legitimate sales have been excluded from the calculations.
If more information were
available, some of these records are likely to have been identified as legitimate and
included in the analysis.
The most important variables in the regression are the indicator variables for
competitive conditions, especially for iTunes 7.0. For an indicator variable to include the
correct transactions, the date of purchase must be accurately recorded. The direct sales
data contain two dates, the shipment date and the billed date. Because only online sales
have a shipment date, billed date is used for purposes of determining the date of the sale.
Another key variable in the econometric model is price, which is calculated by
dividing the amount variable by the quantity variable if the value of the quantity variable
is not equal to zero. The goal is to calculate a price that Apple charged for an iPod in a
given transaction. Prices should exclude other costs, such as taxes and delivery charges.
If a price adjustment is made for customer dissatisfaction due to late delivery, the
adjustment is not related to the iPod purchase price. But some post-sale promotions and
price adjustments change the actual price paid. To calculate accurate prices, the
transaction amount must be accurate, and so must include returns, credits, and
adjustments that were applied to the original transaction.
CONFIDENTIAL – ATTORNEYS EYES ONLY
87
Direct Sales Regression
CONFIDENTIAL – ATTORNEYS EYES ONLY
88
This section reports the results of the price regression on the current version of
Apple’s direct sales transactions data. For each transaction, a set of quantity indicator
variables were created based on the total quantity being purchased. The categories are:
Greater than 5, greater than 10, greater than 20, greater than 50 and greater than 100.
Thus, the price effect of a particular size of transaction is measured by the sum of the
coefficients of all quantity ranges up to that amount.
Exhibit 13.2 shows the results of the preferred regression. 133 The values of the
adjusted R2 for the preferred regression is 0.9813 which means that virtually all of the
variation in prices across models of iPods and among time periods is explained by these
equations. Exhibit 14 shows the average damage per unit sold and the total amount of
damages to direct purchasers. Exhibit 15.2 disaggregates the damages to direct
purchasers to a class and generation of iPod, and Exhibit 16.2 further disaggregates the
damages to direct purchasers to class/generation/family.
The coefficient on unit cost is positive and highly significant, indicating
substantial cost pass-through. But the results also indicate that much of the price
differences among iPod models are explained by model features, indicating substantial
price discrimination.
133. The equation also has been estimated with adjustments for heteroskedasticity and
the results are the same using heteroskedasticity-consistent standard errors.
CONFIDENTIAL – ATTORNEYS EYES ONLY
89
CONFIDENTIAL – ATTORNEYS EYES ONLY
90
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX A
CURRICULUM VITAE
ROGER G. NOLL
PERSONAL
Date and Place of Birth: March 13, 1940; Monterey Park, California
EDUCATION
East High School, Salt Lake City, Utah, 1958
B.S. (Math, Honors), California Institute of Technology, 1962
A.M., Ph.D. (Economics), Harvard University, 1965, 1967
SCHOLARSHIPS, FELLOWSHIPS AND AWARDS
National Merit Scholarship 1958-62
National Defense Education Act Fellowship 1962-66 (declined)
Harvard Prize Fellowship 1962-63
National Science Foundation Fellowship 1963-64
Guggenheim Fellow 1983-84
Rhodes Prize for Undergraduate Teaching, Stanford University, 1994
Distinguished Service Award, Public Utilities Research Center, University of Florida, 2001
Distinguished Lecture Award, Brookings-AEI Joint Center on Regulation and Markets, 2006
Alfred E. Kahn Distinguished Career Award, American Antitrust Institute, 2012
Distinguished Member Award, Transportation and Public Utilities Group, American Economic
Association, 2013
POSITIONS HELD
Instructor, California Institute of Technology, 1965-67
Assistant Professor, California Institute of Technology, 1967-69
Senior Staff Economist, Council of Economic Advisers, 1967-68
Associate Professor, California Institute of Technology, 1969-71
Senior Fellow and Co-director, Studies in the Regulation of Economic Activity, Brookings Institution,
1970-73
Professor, California Institute of Technology, 1973-82
Visiting Professor, Graduate School of Business, Stanford University, 1976-77
Chair, Division of the Humanities and Social Sciences, California Institute of Technology, 1978-82
Reuben Gustavson Lecturer, University of Chicago, April 1981
Institute Professor of Social Sciences, California Institute of Technology, 1982-84
Donald Gilbert Memorial Lecturer, University of Rochester, December 1982
Fellow, Center for Advanced Study in the Behavioral Sciences, 1983-84
Professor of Economics, Stanford University, 1984-2006 (Emeritus 2006-)
Visiting Scholar, Hoover Institution, 1984-85
Professor by Courtesy, Department of Political Science, Stanford University, 1985-2006
Professor by Courtesy, Graduate School of Business, Stanford University, 1986-2006
Veblen-Clark Lecturer, Carleton College, May 1986
Director, Public Policy Program, Stanford University, 1986-2002
David Kinley Lecturer, University of Illinois, May 1987
Sunderland Fellow, Law School, University of Michigan, Fall 1988
Morris M. Doyle Centennial Professor in Public Policy, Stanford University, 1990-2002
Jean Monnet Professor, European University Institute, Spring 1991
Associate Dean, Humanities and Sciences, Stanford University, l991-92
Visiting Professor, University of California, San Diego, 1993
2
Visiting Fellow, Brookings Institution, 1995-96
Nonresident Senior Fellow, Brookings Institution, 1996-99
Director, American Studies Program, Stanford University, 2001-02
Visiting Scholar, London School of Economics, Spring 2001 and Spring 2002
Senior Fellow, American Antitrust Institute, 2002Director, Stanford Center for International Development, 2002-06
Kim Thomas Lecturer, Whittier College, 2010
TEACHING EXPERIENCE
Undergraduate: Introductory Economics, Intermediate Microeconomic Theory, Introduction to
Econometrics, Antitrust and Regulation, Economic History of Medieval Europe, History of Economic
Thought, Economic Policy Analysis, Economics of Sports, Political Economy of the West
Graduate: Antitrust and Regulation, Economic Policy Analysis, Applied Microeconomic Theory,
Experimental Economics
RESEARCH INTERESTS
Antitrust and Regulation, Technology Policy, Political Economics, Political Economy of Law
MEMBERSHIP ON BOARDS AND COMMITTEES
President's Task Force on Communications Policy (CEA Staff Representative), 1967-68
President's Task Force on Suburban Problems, 1968
President's Committee on Urban Housing, 1968
President’s Task Force on Public Broadcasting, 1968
Department of Commerce Technical Advisory Board Panel on Venture Capital, 1968-69
Committee on the Multiple Uses of the Coastal Zone, National Council on Marine Resources and
Engineering, 1968
Secretary, President's Interagency Task Force on Income Maintenance, 1968
Task Force on Application of Economic Analysis to Transportation Problems, National Research Council,
1970-73
Committee on Technological Forecasting on Behalf of the Environment, Office of Science and
Technology, 1970-71
Board of Economic Advisers, Public Interest Economics Foundation, 1974-84
Executive Committee, Caltech Environmental Quality Laboratory, 1970-71
Faculty Board, Caltech, 1974-76
Advisory Commission on Regulatory Reform, Senate Committee on Government Operations, 1975-77
Chair, Fourth Annual Telecommunications Policy Research Conference, 1975-76
Committee on Satellite Communications, National Academy of Sciences, 1975-76
Advisory Council, Jet Propulsion Laboratory, 1976-82
Chair, Committee to Monitor the Desegregation Plan of the Los Angeles Unified School District, Los
Angeles Superior Court, 1978-79
Advisory Council, National Aeronautics and Space Administration, 1978-81
Advisory Council, National Science Foundation, 1978-89
Board of Advisers, National Institute of Economics and Law, 1978-84
Research Advisory Board, Committee for Economic Development, 1979-82
President's Commission for a National Agenda for the Eighties, 1980
Board of Directors, Economists, Inc., 1981Review Panel, NSF Regulation and Public Policy Program, 1981-84
Board of Editors, Journal of Economic Literature, 1981-90
Advisory Board, Solar Energy Research Institute, 1982-91
Board of Directors, Cornell Pelcovits and Brenner, Inc., 1982-1988
3
Chair, Advisory Panel on Information Technology R&D, Office of Technology Assessment, 1983-84
Supervisory Board of Editors, Information Economics and Policy, 1982-88
Advisory Committee on Integrated Environmental Management Program, Environmental Protection
Agency, 1983-85
Commission on Behavioral and Social Sciences and Education, National Research Council, 1984-90
Advisory Panel, NSF Policy Research and Analysis Division, 1984
Director, Program on Regulatory Policy, Stanford Institute for Economic Policy Research, 1984Panel on Clean Air, Science Advisory Board, Environmental Protection Agency, 1985-86
Board of Editors, Review of Economics and Statistics, 1985-2002
Contributing Editor, Regulation, 1986-93
Energy Research Advisory Board, Department of Energy, 1986-89
President & Chairman of the Board, Telecommunications Policy Research Foundation, 1986-87
Coordinating Editor, Information Economics and Policy, 1988-92
Board of Directors, International Telecommunications Society, 1988-92
Advisory Board of Editors, Journal of Risk and Uncertainty, 1988Acid Rain Advisory Committee, Environmental Protection Agency, 1990-91
Secretary of Energy Advisory Board, 1990-95
International Board of Editors, International Journal of the Economics of Business, 1993Faculty Senate, Stanford University, 1993-95, 98-02, 04-06
California Council on Science and Technology, 1995-2001
Panel on Universities, President's Committee of Advisors on Science and Technology, 1996
Committee on Intellectual Property and the Information Infrastructure, National Research Council, 1997-9
Board of Editors, Journal of Sports Economics, 1999Board of Associate Editors, Economics of Governance, 1999Board of Advisors, American Antitrust Institute, 2000Board on Science, Technology and Economic Policy, National Research Council, 2000-2006
Committee on Universal Postal Service, National Research Council, 2008
SPONSORED RESEARCH
"Opinions of Policemen." International Association of Chiefs of Police, 1969
"Studies in the Regulation of Economic Activity." Brookings Institution and Ford Foundation, 1970- 3
"Government Policies and Technological Innovation." National Science Foundation National R&D
Assessment Program, 1973-4
"The Social Consequences of Earthquake Prediction," National Aeronautics and Space Administration,
1974-6
"Nuclear Safety Regulation." National Science Foundation RANN Program, 1975-7
"The Public Television Station Program Cooperative." National Science Foundation RANN Program,
1975-7
"The Station Allocation Game." Federal Communications Commission, 1977
"Energy Policy Studies." Various donors, 1978-84
"Economics of Oil Leasing" and "Issues in Utility Pricing." Department of Energy, 1978-9
"The Economics of Boxing, Wrestling and Karate." California Athletic Commission, 1978
"Implementing Tradable Emissions Permits." California Air Resources Board, 1979-82
"Social Science and Regulatory Policy." National Science Foundation, 1980-2
"The Political Economy of Public Policy." National Science Foundation and Center for Economic Policy
Research, Stanford University, 1983-4
"SIEPR Program on Regulatory Policy." various donors, 1987"The Economics of Research Universities and Scholarly Communication." Brown Center for Education
Policy, Brookings Institution, 1995-6
"Coordination of Regulatory Reform," Organization for Economic Cooperation and Development, 1996
"The Future of the Research University," Carnegie Foundation, 1996
“SCID Program in Economic Policy Reform,” Various donors, 2002-06
4
Membership on Committees and Boards, cont’d
CONSULTING
Special Assistant to the President, Ford Foundation, 1969
Space Technology Applications, Jet Propulsion Laboratory, 1969
Panel on the Abatement of Particulate Emissions, National Research Council, 1971
Sloan Commission on Cable Communications, 1971
President's Commission on Government Procurement, 1971
Senate Subcommittee on Antitrust and Monopoly, 1971-72
MCI, Inc., 1972-73, 1983, 1986
National Science Foundation, 1973, 1975
Department of Justice, Antitrust Division, 1974-77, 1979-81, 1993-97
Internal Revenue Service, 1976-77
RAND Corporation, 1974-82
Los Angeles Lakers, 1974-75
National Football League Players Association, 1974-76, 1987-93, 2008, 2010Office of Telecommunications Policy, 1975-77
National Basketball Association Players Association, 1975-76, 1987-88, 1994
Naval Ordnance Test Station, 1975
Commission on Law and the Economy, American Bar Association, 1977-78
Aspen Institute Program on Communications and Society, 1977
National Commission on Electronic Funds Transfer, 1977
Business Round Table, 1978
Federal Communications Commission, 1977-81
Food and Drug Administration, 1978
Carnegie Commission on the Future of Public Broadcasting, 1978
Department of Energy, 1979
Office of Technology Assessment, 1980
Kerr-McGee Corporation, 1980
CBS, Inc. 1982-83
Environmental Protection Agency, 1982-83
Showtime/The Movie Channel, 1983, 1985
Harlequin Books, 1984
Lake Huron Broadcasting, 1984
National Collegiate Athletics Association, 1984
National Medical Enterprises, 1985, 1987-88
Camellia City Telecasters, 1985-86
Brown and Root, Inc., 1985-86
McDermott, Inc., 1985-86
Major League Baseball Players Association, 1985, 1994
United Cable Television and American Television and Communications, 1985
United States Football League, 1985-86
City of Anaheim, 1986
Technicolor, 1986
Metro-Mobile, 1986-89
Hewlett-Packard, 1986-1990, 1991
Echostar, 1987, 1994-95, 2002-03, 2004-05
Continental Airlines, 1987-88
Home Box Office, 1988-89
Bell South Cellular, 1989
Western Union, 1989
Minnesota Twins, 1989
Northwest Airlines, 1989
Pepsico, 1989
Yellow Phone, 1989-91
Dialog, 1990-91
5
Consulting, cont’d
California Public Utilities Commission, 1989-90
American Newspaper Publishers Association, 1990
Humana, 1990-91
Powell, Goldstein, Frazer and Murphy, 1990-93
South Coast Air Quality Management District, 1990-91
Federal Trade Commission, 1990-91, 2010Delta Airlines, 1991
California Cable Television Association, 1991
Bureau of Competition Policy, Government of Canada, 1991
R&D Business Systems, et al. 1991-95
International Entertainment Group, 1992-93
Nike, Inc., 1992
World Bank, 1992Gemini, Inc. 1992-94
Servicetrends, Inc., 1993-94
William Sullivan, 1993-95
Sure Safe Industries, 1993
U. S. Department of Justice, Civil Division 1994-95
Kopies, Inc., et al. 1995-1999
Telecom Technical Services, et al., 1995-1999
Digital Distribution, Inc.. 1996-1999
Silvey, et al., 1996-2000
Aguillar, et al. 1996-2000
Wadley Medical Center, 1997-2001
Oakland Raiders, 1997-2000
Major League Soccer Players Association, 1997-2000
Class Plaintiffs, Brand Name Prescription Drugs Litigation, 1998-9
Class Plaintiffs, Compact Disc Litigation, 1999-2003
Class Plaintiffs, State Microsoft Antitrust Litigation (California, Iowa, Minnesota, New York), 2000-2007
Kingray, 2000
Napster, 2000-2
Metropolitan Intercollegiate Basketball Association, 2002-5
Congressional Budget Office, 2002
Pioneer and Scientific Atlanta, 2002-3
Lenscrafters, 2003-4, 2009-12
Seven Network, 2003-7
Sports Car Clubs of America, 2003-05
Intertainer, 2003-05
Class Plaintiffs, DRAM Antitrust Litigation 2005-7
Class Plaintiffs, Honeywell Antitrust Litigation, 2005Class Plaintiffs, Tableware Antitrust Litigation, 2005-7
Class Plaintiffs, White, et al., v. NCAA, 2006-8
Sirius Satellite Radio and XM Satellite Radio, 2006-7
Class Plaintiffs, Cartier Antitrust Litigation, 2006-7
Monte Carlo Country Club and Socit Mongasque pour l’Exploitation du Tournai de Tennis, 2007
Pearle Vision, Inc., 2007-8
Class Plaintiffs, Apple iTunes/iPod Antitrust Litigation, 2007Class Plaintiffs, SRAM Antitrust Litigation, 2007-9
Fair Isaac, 2007-9
Houston Baptist University, 2008.
U. S. Department of Justice, U. S. Attorney’s Office, San Francisco, 2008-9
Novell, 2008-11
GlaxoSmithKline, 2008-11
Class Plaintiffs, Flash Memory Antitrust Litigation, 2008-10
MobiTV, 2009-10
6
Consulting, cont’d
AT&T, 2009-10
Verizon, 2009-10
Ericsson, 2009-10
Kaleidescape, 2011-12
Class Plaintiffs, Text Messaging Antitrust Litigation, 2011Class Plaintiffs, California Automobile Insurance Antitrust Litigation, 2011Sirius XM, 2011Class Plaintiffs, NCAA Student-Athlete Name and Likeness Licensing Litigation, 2011-
BOOKS AND MONOGRAPHS
Reforming Regulation: An Evaluation of the Ash Council Report. Brookings Institution, 1971.
Economic Aspects of Television Regulation, co-authors Merton J. Peck and John J. McGowan. Brookings
Institution, 1973. Winner, National Association of Educational Broadcasters Annual Book Award, 1974.
Government and the Sports Business, editor. Brookings Institution, 1974.
The Political Economy of Deregulation, co-author Bruce Owen. American Enterprise Institute, 1983.
Regulatory Policy and the Social Sciences, editor. University of California Press, 1985.
The Technology Pork Barrel, co-author Linda R. Cohen. Brookings Institution, 1991.
The Economics and Politics of Deregulation. European University Institute, 1991.
Constitutional Reform in California: Making State Government More Effective and Responsive,
co-editor Bruce E. Cain. University of California Institute of Governmental Studies, 1995.
Sports, Jobs, and Taxes, co-editor Andrew Zimbalist. Brookings Institution, 1997.
Challenges to Research Universities, editor. Brookings Institution, 1998
A Communications Cornucopia, co-editor Monroe E. Price. Brookings Institution, 1998.
The Economics and Politics of the Slowdown in Regulatory Reform. AEI Press, 1999.
The Digital Dilemma, 17 co-authors (Committee on Intellectual Property Rights and the Emerging
Information Infrastructure). National Academy Press, 2000.
Bridging the Digital Divide, editor. California Council on Science and Technology, 2001.
Economic Reform in India, co-editors Nicholas C. Hope, Anjini Kochar, and T.N. Srinivasan. Cambridge
University Press, 2013
ARTICLES IN SCHOLARLY PUBLICATIONS
"Urban Concentration: Prospects and Implications." In Increasing Understanding of Public Problems and
Policies. Farm Foundation, 1969.
"Metropolitan Employment and Population Distribution and the Conditions of the Urban Poor.” In
Financing the Metropolis: Public Policy in Urban Economics: The Urban Affairs Annual Reviews, IV,
John P. Crecine, ed. Sage Publications, 1970. Brookings Reprint No. 184.
7
"National Communications Policy: Discussion--Spectrum Allocation Without Markets." American
Economic Review Papers and Proceedings 60(2) (May 1970).
"The Behavior of Regulatory Agencies." Review of Social Economics 24(1) (March 1971): 15-19.
Brookings Reprint No. 219 (November 1971).
"Summary and Conclusions," co-author William Capron. In Technological Change in Regulated
Industries, William Capron, ed. Brookings Institution, 1971.
"The Nature and Causes of Regulatory Failure." Administrative Law Review 23(4) (June 1971): 424-437.
Revised version published as "The Economics and Politics of Regulation." Virginia Law Review 57(6)
(September 1971): 1016-1032.
"Mass Balance, General Equilibrium and Environmental Externalities," co-author, John Trijonis.
American Economic Review 61(4) (September 1971): 730-735.
"Selling Research to Regulatory Agencies." In The Role of Analysis in Regulatory Decisionmaking: The
Case of Cable Television, Rolla Edward Park, ed. Heath-Lexington, 1973.
"Relative Prices on Regulated Transactions of the Natural Gas Pipelines," co-author Paul W. MacAvoy.
Bell Journal of Economics and Management Science 4(1) (Spring 1973): 212-234.
"Regulating Prices in Competitive Markets," co-author Lewis A. Rivlin. Yale Law Journal 82(7) (June
1973): 1426-1434.
"Attendance and Price Setting." In Government and the Sports Business, Roger G. Noll, ed. Brookings
Institution, 1974. Reprinted in The Economics of Sport, Andrew Zimbalist, ed. Edward Elgar, 2001.
"The U.S. Team Sports Industry." In Government and the Sports Business, Roger G. Noll, editor.
Brookings Institution, 1974. Abridged version reprinted in Public Policies Toward Business:
Reading and Cases, William G. Shephard, ed. Irwin, 1975.
"Alternatives in Sports Policy." In Government and the Sports Business, Roger G. Noll, ed. Brookings
Institution, 1974. Abridged version in Public Policies Toward Business: Readings and Cases, William G.
Shephard, ed. Irwin 1975. Revised version in Handbook of Social Science of Sport, Gunther R. R.
Luschen and George H. Sage, eds. Stripes Publishing Co., 1980.
"The Social Costs of Government Intervention." In The Business-Government Relationship in American
Society: Reassessment, Neil H. Jacoby, editor. University of California Press, 1975.
"The Consequence of Public Utility Regulation of Hospitals." In Controls on Health Care. Washington,
D.C.: National Academy of Sciences, 1975.
"Information, Decision-Making Procedures and Energy Policy." American Behavioral Scientist, Vol. 19,
No. 3 (January/February 1976): 267-278. Reprinted in Current Issues in Social Policy, W. B. Littrell and
G. Sjoberg, eds. Sage, 1976.
"Breaking Out of the Regulatory Dilemma: Alternatives to the Sterile Choice." Indiana Law Journal 51(3)
(Spring 1976): 686-699. Reprinted in Corporate Practice Commentator 19(1) (Spring 1977): 99-114.
"Safety Regulation," co-authors Nina Cornell and Barry Weingast. In Setting National Priorities: The
Next Ten Years, Henry Owen and Charles L. Schultze, eds. Brookings Institution, 1976.
"Major League Team Sports." In The Structure of American Industry, Walter Adams, ed. 5th ed.
Macmillan, 1977. 6th ed. Macmillan, 1981.
8
"An Experimental Market for Public Goods: The PBS Program Cooperative," co-author John A. Ferejohn.
American Economic Review Papers and Proceedings 66(2) (May 1976): 267-273.
"Government Policy and Technological Innovation: Where Do We Stand and Where Do We Go?" In
Innovation, Economic Change and Technology Policies, K.A. Stroetmann, ed. Birkauser-Verlag, 1977.
"Economic Policy Research on Cable Television: Assessing the Costs and Benefits of Cable
Deregulation," co-authors S. M. Besen, B. M. Mitchell, B. M. Owen, R. E. Park, and J. N. Rosse. In
Deregulation of Cable Television, Paul W. MacAvoy, ed. American Enterprise Institute, 1977.
"The Economic Implications of Regulation by Expertise: The Case of Recombinant DNA Research," coauthor Paul A. Thomas. In Research with Recombinant DNA. National Academy of Sciences, 1977.
"The Dilemma of Consumer Advocacy." In Regulatory Reform, W.S. Moore, ed. American Enterprise
Institute, 1978.
"Voters, Bureaucrats and Legislators: A Rational Choice Perspective on the Growth of Bureaucracy," coauthor Morris P. Fiorina. Journal of Public Economics 9(3) (May 1978): 239-254.
"Public Utilities and Solar Energy Development--Institutional Economic Considerations." In The Solar
Market: Proceedings of the Symposium on Competition in the Solar Industry, Federal Trade Commission,
1978. Excerpted version in On the Economics of Solar Energy, Stephen L. Feldman and Robert M.
Wirtshafter, eds. D.C. Heath, Lexington Books, 1980.
"Uncertainty and the Formal Theory of Political Campaigns," co-author John A. Ferejohn. American
Political Science Review 72(2) (June 1978): 492-505.
"The Rationale for Mandated Cost Increases." In Economic Effects of Government-Mandated Costs,
Robert F. Lanzillotti, ed. University Presses of Florida, 1978.
"Regulation and Computer Services." In The Computer Age, Michael L. Dertouzos and Joel Moses, eds.
MIT Press, 1979: 254-284.
"An Experimental Analysis of Decisionmaking Procedures for Discrete Public Goods: A Case Study of a
Problem in Institutional Design," co-authors John A. Ferejohn and Robert E. Forsythe. In Research in
Experimental Economics, Vol. I, Vernon L. Smith, ed. JAI Press, 1979.
"Voters, Legislators and Bureaucracy: Institutional Design in the Public Sector," co-author Morris P.
Fiorina. American Economic Review Papers and Proceedings 68(2) (May 1978): 256-260.
Translated into Italian in Problemi Di Amministrazione Pubblica 4(2) (1979): 69-89.
"Practical Aspects of the Construction of Decentralized Decisionmaking Systems for Public Goods," coauthors John A. Ferejohn and Robert Forsythe. In Collective Decisionmaking, Clifford S. Russell, ed.
Resources for the Future, 1979.
"Majority Rule Models and Legislative Elections," co-author Morris P. Fiorina. Journal of Politics, Vol.
41, No. 4 (November 1979): 1081-1104.
"Antitrust Exemptions: An Economic Overview." In National Commission for the Review of Antitrust
Laws and Procedures, Report to the President and the Attorney General, Vol. II. U.S. Department of
Justice, 1979.
"Regulatory and Nonregulatory Strategies for Controlling Health Care Costs," co-author Alain Enthoven.
In Medical Technology: The Culprit Behind Health Care Costs, Stuart H. Altman and Robert Blendon, eds.
Sun Valley Forum on National Health. U.S. Department of Health, Education and Welfare Publication No.
(PHS) 79-3216, 1979.
9
"The Game of Health Care Regulation: Comments on Feldman/Roberts." In Issues in Health Care
Regulation, Richard S. Gordon, ed. McGraw-Hill Book Co., 1980.
“Discussion: Regulatory Institutions in Historical Perspective.” American Economic Review Papers and
Proceedings, 70(2) (May 1980): 316-317.
"The Economics of Disaster Defense: The Case of Building Codes to Resist Seismic Shock," co-author
Linda Cohen. Public Policy 29(1) (Winter 1981): 1-29. Reprinted in The Economics of Natural
Disasters, Vol. I, Howard Kunruether and Adam Rose, eds. Edard Elgar, 2004.
"Regulation in Theory and Practice: An Overview," co-author Paul L. Joskow. In Studies in Public
Regulation, Gary Fromm, ed. MIT Press, 1981.
"Designing a Market for Tradable Emissions Permits," co-author Robert W. Hahn. In Reform of
Environmental Regulation, Wesley Magat, ed. Lexington Books, 1982.
"Implementing Marketable Emissions Permits." American Economic Review Papers and Proceedings
72(2) (May 1982): 120-124.
"An Experimental Examination of Auction Mechanisms for Discrete Public Goods," co-authors John A.
Ferejohn, Robert Forsythe and Thomas R. Palfrey. In Research in Experimental Economics, Vol. II,
Vernon L. Smith, ed. JAI Press, 1982. Reprinted in Experimental Foundations of Political Science,
Donald R. Kinder and Thomas R. Palfrey, eds. University of Michigan Press, 1993.
"Implementing Tradable Emissions Permits," co-author Robert W. Hahn. In Reforming Social Regulation,
Leroy Graymer and Frederick Thompson, eds. Sage Publications, 1982.
"The Case Against the Balanced Budget Amendment: Comments on Aranson and Rabushka." In The
Economic Consequences of Government Deficits, Laurence H. Meyer, ed. Kluwer-Nyhoff Publishing,
1983.
"The Feasibility of Marketable Emissions Permits in the U.S." In Public Sector Economics, Jorg Finsinger,
ed. Macmillan Press, Ltd., 1983.
"Barriers to Implementing Tradable Air Pollution Permits: Problems of Regulatory Interactions," co-author
Robert W. Hahn. Yale Journal on Regulation 1(1) (1983): 63-91.
"The Future of Telecommunication Regulation." In Telecommunications Today and Tomorrow, Eli Noam,
ed. Harcourt Brace Jovanovich, 1983.
"The Political Foundations of Regulatory Policy." Zeitschrift fur die gesamte Staatswissenschaft (Journal
of Institutional and Theoretical Economics) 139(3) (1983): 377-404. Reprinted in Congress: Structure
and Policy, Mathew McCubbins and Terry Sullivan, eds. Cambridge University Press, 1987.
"The Regulation of Surface Freight Transportation: The Welfare Effects Revisited," co-author Ronald R.
Braeutigam. The Review of Economics and Statistics 66(1) (1984): 80-87.
"Prospective Payment: Will It Solve Medicare's Financial Problem," co-author Alain C. Enthoven. Issues
in Science and Technology 1(1) (1984): 111-116. Reprinted in Health Industry Today 48(3) (1985): 16-24.
"The Preferences of Policy Makers for Alternative Allocations of the Broadcast Spectrum," co-author
Forrest Nelson. In Antitrust and Regulation: Essays in Memory of John J. McGowan, Franklin M. Fisher,
ed. MIT Press, 1985.
10
"Government Regulatory Behavior: A Multidisciplinary Survey and Synthesis." In Regulatory Policy and
the Social Sciences, Roger G. Noll, ed. University of California Press, 1985.
"'Let Them Make Toll Calls': A State Regulator's Lament." American Economic Review Papers and
Proceedings 75(2) (1985): 52-56.
"State Regulatory Responses to Competition and Divestiture in the Telecommunications Industry." In
Antitrust and Regulation, Ronald E. Grieson, ed. Lexington Books, 1986.
"Funding and Knowledge Growth: Comments." Social Studies of Science 16(1) (1986): 135-42.
"The Political and Institutional Context of Communications Policy." In Marketplace for
Telecommunications, Marcellus S. Snow, ed. Longman, Inc., 1986.
"Government R&D Programs for Commercializing Space," co-author Linda R. Cohen. American
Economic Review Papers and Proceedings 76(2) (1986): 269-73.
"Administrative Procedures as Instruments of Political Control," co-authors Mathew D. McCubbins and
Barry R. Weingast. Journal of Law, Economics and Organization 3(2) (1987): 243-77. Abridged version
in State and Federal Administrative Law, Arthur Earl Bonfield and Michael Asimow, eds. West
Publishing, 1989. Reprinted in Economic Regulation, Paul Joskow, ed. Edward Elgar, 2000; Regulation
and Regulatory Processes, Cary Coglianese and Robert Kagan, eds. Ashgate Publishing, 2007; The
Economics of Administrative Law, Susan Rose-Ackerman, ed., Edward Elgar, 2007; Rational Choice
Politics, Torun Dewan, Keith Dowling and Kenneth Shepsle, eds., Sage, 2009.
"Communications." In The New Palgrave, John Eatwell, Murray Milgat, and Peter Newman, eds.
MacMillan, 1987.
"Comment: Settlement Incentives and Follow-on Litigation." In Private Antitrust Litigation, Lawrence J.
White, ed. MIT Press, 1988.
"The Political Economy of NASA's Applications Technology Satellite Program," co-author Linda R.
Cohen. In Space Applications Board, Proceedings of a Symposium on Space Communications Research
and Development. National Research Council, 1988.
"Economics, Politics and Government Research and Development," co-author Linda Cohen. In
Technology and Politics, Michael E. Kraft and Norman J. Vig, eds. Duke University Press, 1988.
"The Anticompetitive Uses of Regulation: United States v. AT&T (1982)," co-author Bruce M. Owen. In
The Antitrust Revolution, John E. Kwoka, Jr., and Lawrence J. White, eds. Scott, Foresman, 1988.
"The Economics of Sports Leagues." In Law of Professional and Amateur Sports, Gary A. Uberstine, ed.
Clark Boardman, 1988.
"Preface: Symposium on Telecommunications Demand." Information Economics and Policy 3(4)
(1988): 275.
"U.S. v. AT & T: An Interim Assessment," co-author Bruce M. Owen. In Future Competition in
Telecommunications, Stephen P. Bradley and Jerry A. Hausman, eds. Harvard Business School Press,
1989.
"Structure and Process, Politics and Policy: Administrative Arrangements and the Political Control of
Agencies," co-authors Mathew D. McCubbins and Barry R. Weingast. Virginia Law Review 75(2)
(March 1989): 431-482. Reprinted in The Political Economy of Regulation, Thomas Lyon, ed. Edward
Elgar, 2007.
11
"Telecommunications Regulation in the 1990s." In New Directions in Telecommunications Policy, Vol. I,
Paula R. Newberg, ed. Duke University Press, 1989.
"Economic Perspectives on the Politics of Regulation." In Handbook of Industrial Organization, Vol. II.
Richard Schmalensee and Robert Willig, eds. North Holland Publishing Co., 1989. Reprinted in
Regulation and the Law, Anthony I. Ogus, ed. Edward Elgar Publishing, 2001.
"Comment: Peltzman on Deregulation." Brookings Papers on Economic Activity: Microeconomics
(1989): 48-58.
"Pricing of Telephone Services," co-author Susan Smart. In After the Breakup, Barry G. Cole, ed.
Columbia University Press, 1990.
"Positive and Normative Models of Procedural Rights: An Integrative Approach to Administrative
Procedures," co-authors Mathew D. McCubbins and Barry R. Weingast. Journal of Law, Economics and
Organization 6 (1990): 307-332.
"Some Implications of Cognitive Psychology for Risk Regulation," co-author James Krier. Journal of
Legal Studies 19 (June 1990): 747-779. Excerpts reprinted in Foundations of the Economic Approach to
Law, Avery Weiner Katz, ed. Oxford University Press, 1998. Reprinted in Behavioral Law and
Economics, Cass R. Sunstein, ed. Cambridge University Press, 2000.
"Environmental Markets in the Year 2000," co-author Robert Hahn. Journal of Risk and Uncertainty 3
(December, 1990): 347-363.
"Commentary: The Prospects for Using Market Incentives for Conservation of Biological Diversity." In
The Preservation and Valuation of Biological Resources, Gordon H. Orians, Gardner M. Brown, Jr.,
William E. Kunin, and Joseph E. Swierzbinski, eds. University of Washington Press, 1990.
"Slack, Public Interest, and Structure-Induced Policy," co-authors Mathew D. McCubbins and Barry R.
Weingast. Journal of Law, Economics and Organization 6 (1990): 203-212.
"How to Vote, Whether to Vote: Strategies for Voting and Abstaining on Congressional Roll Calls," coauthor Linda R. Cohen. Political Behavior 13(2) (1991): 97-127.
"Rational Actor Theory, Social Norms, and Policy Implementation: Applications to Administrative
Processes and Bureaucratic Culture," co-author Barry R. Weingast. In The Economic Approach to Politics,
Kristen Renwick Monroe, ed. Harper Collins, 1991.
"The National Aerospace Plane: An American Technological Long Shot, Japanese Style," co-authors Linda
R. Cohen and Susan A. Edelman. American Economic Review Papers and Proceedings, 81(2) (May
1991): 50-53.
"The Economics of Intercollegiate Sports." In Rethinking College Athletics, Judith Andre and David N.
James, eds. Temple University Press, 1991.
"Comparative Structural Policies," co-author Haruo Shimada. In Parallel Politics, Samuel Kernell, ed.
Brookings Institution, 1991.
"Structural Policies in the United States." In Parallel Politics, Samuel Kernell, ed. Brookings Institution,
1991.
"On Regulating Prices for Physicians." In Regulating Doctors' Fees, H.E. Frech III, ed. AEI Press, 1991.
"ISDN and the Small User: Regulatory Policy Issues," co-author William Lehr. In Integrated Broadband
Networks, Martin C.J. Elton, ed. North-Holland, 1991.
12
"Computer Reservation Systems and Their Network Linkages to the Airline Industry," co-author
Margaret E. Guerin-Calvert. In Electronic Services Networks, Margaret E. Guerin-Calvert and Steven S.
Wildman, eds. Praeger, 1991.
"Professional Basketball: Economic and Business Perspectives." In The Business of Professional Sports,
Paul D. Staudohar and James A. Mangan, eds. University of Illinois Press, 1991.
"An Economic Analysis of Scientific Journal Prices: Preliminary Results," co-author W. Edward
Steinmueller. Serials Review 18 (1992): 32-37.
"The Theory of Interpretive Canon and Legislative Behavior," co-authors Mathew McCubbins and Barry
Weingast. International Review of Law and Economics 12 (1992): 235-238.
"Positive Canons: The Role of Legislative Bargains in Statutory Interpretation," co-authors Mathew
McCubbins and Barry Weingast. Georgetown Law Journal 80 (February 1992): 705-742. Reprinted
in Sutherland Statutory Construction, 5th edition, Norton J. Singer, ed., Clark Boardman Callaghan, 1992,
and Sutherland Statutes and Statutory Interpretation, 6th edition, Norman J. Singer and J. D. Shambie
Singer, eds., West, 2000.
"Comment: Judicial Review and the Power of the Purse." International Review of Law and Economics 12
(June 1992): 211-213.
"Comment: Standard Setting in High-Definition Television." Brookings Papers on Economic Activity:
Microeconomics, 1992: 80-89.
"Research and Development," co-author Linda R. Cohen. In Setting Domestic Priorities: What Can
Government Do? Henry J. Aaron and Charles L. Schultze, eds. Brookings Institution, 1992.
"The Economics of Information: A User's Guide." In The Knowledge Economy: Annual Review of the
Institute for Information Studies. Aspen Institute, 1993.
"Downsian Thresholds and the Theory of Political Advertising." In Information, Participation, and
Choice, Bernard Grofman, ed. University of Michigan Press, 1993.
"Economic Regulation," co-author Paul L. Joskow. In American Economic Policy in the 1980s, Martin
Feldstein, ed. University of Chicago Press, 1994.
"Legislative Intent: The Use of Positive Political Theory in Statutory Interpretation," co-authors Mathew
McCubbins and Barry Weingast. Law and Contemporary Problems 57 (Winter/Spring 1994): 3-37.
Reprinted in Public Choice and Public Law, Daniel A. Farber, ed. Edward Elgar, 2006, and Legal
Institutions and Economic Development, Robert A. Cooter and Francesco Parisi, eds., Edward Elgar, 2010.
"The Origins of State Railroad Regulation: The Illinois State Constitution of 1870," co-author Mark T.
Kanazawa. In The Regulated Economy, Claudia Goldin and Gary D. Libecap, eds. University of Chicago
Press, 1994.
"Privatizing Public Research," co-author Linda R. Cohen. Scientific American, September 1994, pp. 7277. Reprinted in Leading Economic Controversies of 1996, Edwin Mansfield, ed. New York: Norton,
1996.
"Japanese and American Telecommunications Policy," co-author Frances M. Rosenbluth. Communications
and Strategy 15 (3eme trimestre 1994): 13-46.
"Public Policy and the Admission of the Western States," co-author David W. Brady. In The Political
Economy of the American West, Terry L. Anderson and Peter J. Hill, eds. Rowman and Littlefield, 1994.
13
“Introduction: Regulation and the New Telecommunications Infrastructure.” The Changing Nature of
Telecommunications/Information Infrastructure. National Academy Press, 1995.
"Telecommunications Policy: Structure, Process, Outcomes," co-author Frances M. Rosenbluth. In
Structure and Policy in Japan and the United States, Mathew D. McCubbins and Peter Cowhey, eds.
Cambridge University Press, 1995.
"Politics and the Courts: A Positive Theory of Judicial Doctrine and the Rule of Law," co-authors
Mathew D. McCubbins and Barry R. Weingast. Southern California Law Review 68 (September 1995):
1631-83.
“Feasibility of Effective Public-Private R&D Collaboration: The Case of Cooperative R&D Agreements,”
co-author Linda R. Cohen. International Journal of the Economics of Business 2 (1995): 223-240.
"Principles of State Constitutional Design," co-author Bruce E. Cain. In Constitutional Reform in
California, Bruce E. Cain and Roger G. Noll, eds. University of California Institute for Governmental
Studies, 1995. Excerpt reprinted in Madison Review 3 (Fall 1997): 7-11.
"The Role of Antitrust in Telecommunications." Antitrust Bulletin (Fall 1995): 501-528.
"Executive Organization: Responsiveness vs. Expertise and Flexibility." In Constitutional Reform in
California, Bruce E. Cain and Roger G. Noll, eds. University of California Institute for Governmental
Studies, 1995.
"Privatizing Public Research: The New Competitiveness Strategy," co-author Linda R. Cohen. In
The Mosaic of Economic Growth, Ralph Landau, Timothy Taylor, and Gavin Wright, eds. Stanford
University Press, 1996.
"Comment: Preferences, Promises, and the Politics of Entitlement." In Individual and Social
Responsibility, Victor R, Fuchs, ed. University of Chicago Press, 1996.
"Is There a Role for Benefit-Cost Analysis in Environmental, Health and Safety Regulation?," 10 coauthors. Science 272 (April 12, 1996): 221-222. Reprinted in Environmental and Development
Economics, Vol. 2, No. 2 (May 1997), pp. 196-201, and Economics of the Environment, Robert N. Stavins,
ed. Norton, 2000, and subsequent editions.
“Benefit-Cost Analysis in Environmental, Health, and Safety Regulation: A Statement of Principles,” ten
co-authors. AEI, 1996.
"Reforming Risk Regulation." The Annals of the AAPSS 545 (May 1996): 165-75.
"The Complex Politics of Catastrophe Economics." Journal of Risk and Uncertainty 12 (May 1996):
141-46.
"The Economics of Information." In The Economics of Information in the Networked Environment,
Meredith A. Butler and Bruce R. Kingma, eds. Association of Research Libraries, 1996. Reissued by
Haworth Press, 1999.
"The Economics of Scholarly Publications and the Information Superhighway." In The Internet and
Telecommunications Policy, Gerald W. Brock and Gregory L. Rosston, eds. Lawrence Erlbaum, 1996.
"Regulatory Reform as International Policy." In Regulatory Reform and International Market Openness.
Organisation for Economic Cooperation and Development, 1996.
14
"The Future of the National Laboratories," co-author Linda R. Cohen. Proceedings of the National
Academy of Sciences 93 (November 1996): 12678-85.
"Research and Development after the Cold War," co-author Linda R. Cohen. In Commercializing High
Technology: East and West, Judith B. Sedaitis, ed. Roman and Littlefield, 1997.
“Internationalizing Regulatory Reform.” In Comparative Disadvantage? Social Regulations and the
Global Economy, Pietro S. Nivola, ed. Brookings Institution, 1997.
“The International Dimension of Regulatory Reform: With Applications to Egypt.” Distinguished Lecture
Series #8. Egyptian Center for Economic Studies, 1997.
“Build the Stadium–Create the Jobs!” co-author Andrew Zimbalist. In Sports, Jobs and Taxes, Roger G.
Noll and Andrew Zimbalist, eds. Brookings Institution, 1997.
“The Economic Impact of Sports Teams and Facilities,” co-author Andrew Zimbalist. In Sports, Jobs and
Taxes, Roger G. Noll and Andrew Zimbalist, eds. Brookings Institution, 1997.
“Sports, Jobs and Taxes: The Real Connection,” co-author Andrew Zimbalist. In Sports, Jobs and Taxes,
Roger G. Noll and Andrew Zimbalist, eds. Brookings Institution, 1997.
“Legislative Control of Bureaucratic Policy Making,” co-authors Mathew D. McCubbins and Barry R.
Weingast. New Palgrave Dictionary of Economics and the Law. London: New Palgrave, 1997.
“The American Research University: An Introduction.” In Challenges to Research Universities, Roger G.
Noll, ed. Brookings Institution, 1998.
“Universities, Constituencies, and the Role of the States,” co-author Linda R. Cohen. In Challenges to
Research Universities, Roger G. Noll, ed. Brookings Institution, 1998.
“Students and Research Universities,” co-author Gary Burtless. In Challenges to Research Universities,
Roger G. Noll, ed. Brookings Institution, 1998.
“The Economics of University Indirect Cost Reimbursement in Federal Research Grants,” co-author
William P. Rogerson. In Challenges to Research Universities, Roger G. Noll, ed. Brookings Institution,
1998.
“The Future of Research Universities.” In Challenges to Research Univesities, Roger G. Noll, ed.
Brookings Institution, 1998.
“Communications Policy: Convergence, Choice, and the Markle Foundation,” co-author Monroe E. Price.
In A Communications Cornucopia, Roger G. Noll and Monroe E. price, eds. Brookings Institution, 1998.
“Economic Perspectives on the Athlete’s Body.” Stanford Humanities Review 6(2) (1998): 69-73.
“The Bell Doctrine: Applications in Telecommunications, Electricity, and Other Network Industries,” coauthor Paul L. Joskow. Stanford Law Review 51(5) (1999): 1249-1315.
“The Political Origins of the Administrative Procedure Act,” co-authors Mathew D. McCubbins and Barry
R. Weingast. Journal of Law, Economics, and Organization 15(1) (1999): 180-217.
“Competition Policy in European Sports after the Bosman Case.” In Competition Policy in Professional
Sports: Europe after the Bosman Case, Claude Jeanrenaud and Stefan Ksenne, eds. Standaard Editions
Ltd., 1999.
15
“The Economics of Information,” Journal of Library Administration 26(1) (1999): 47-55.
“The Business of College Sports and the High Costs of Winning.” Milken Institute Review 1(3)
(3rd Quarter 1999): 24-37. Reprinted in The Business of Sports, Scott Rosner and Kenneth Shropshire,
eds. Jones and Bartlett, 2004.
“Antitrust and Labor Markets in Professional Sports.” In The Role of the Academic Economist in
Litigation Support, Daniel J. Slottje, ed. North-Holland, 1999.
“Reforming Urban Water Systems in Developing Countries,” co-authors Mary M. Shirley and Simon
Cowan. In Economic Policy Reform: The Second Stage, Anne O. Krueger, ed. University of Chicago,
2000.
“Telecommunications Reform in Developing Countries.” In Economic Policy Reform: The Second Stage,
Anne O. Krueger, ed. University of Chicago: 2000.
“Regulatory Reform and International Trade Policy.” In Deregulation and Interdependence in the AsiaPacific Region, Takatoshi Ito and Anne O. Krueger, eds. University of Chicago/NBER, 2000.
“Comment: Hong Kong’s Business Regulation in Transition.” In Deregulation and Interdependence in
the Asia-Pacific Region, Tahatoshi Ito and Anne O. Krueger, eds. University of Chicago/NBER, 2000.
“The Digital Divide: Definitions, Measurement, and Policy Issues,” co-authors Dina Older-Aguilar,
Richard R. Ross and Gregory L. Rosston. In Bridging the Digital Divide, Roger G. Noll, ed. California
Council on Science and Technology, 2001.
“The Digital Divide: Diagnosis and Policy Options,” co-author Dina Older-Aguilar. In Bridging the
Digital Divide, Roger G. Noll, ed. California Council on Science and Technology, 2001.
“Intellectual Property, Antitrust and the New Economy,” co-author Linda R. Cohen. University of
Pittsburgh Law Review 62(3) (Spring 2001): 453-73.
“The Economics of Promotion and Relegation in Sports Leagues: The Case of English Football.” Journal
of Sports Economics 3(2) (May 2002): 169-203. Reprinted in The Economics of Association Football, Bill
Girard, ed., Edward Elgar, 2006.
“The Economics of Urban Water Systems.” In Thirsting for Efficiency, Mary M. Shirley, ed. Pergamon,
2002.
“Comment: Small-Scale Industry Policy in India.” In Economic Policy Reforms and the Indian
Economy, Anne O. Krueger, ed. University of Chicago, 2002.
“The Economics of the Supreme Court’s Decision on Forward Looking Costs,” co-author Gregory L.
Rosston. Review of Network Economics 1(2) (September 2002): 81-9.
“Federal R&D in the Antiterrorist Era.” In Innovation Policy and the Economy, Vol. 3, Adam B. Jaffe,
Josh Lerner and Scott Stern, eds. MIT Press, 2003.
“The International Dimension of Regulatory Reform with Applications to Egypt.” In Challenges and
Reforms of Economic Regulations in MENA Countries, Imed Limam, ed. American University in Cairo
Press, 2003.
“The Economics of Contraction in Baseball.” Journal of Sports Economics 4(4) (November 2003): 367-88.
“The Organization of Sports Leagues.” Oxford Review of Economic Policy 19(4) (Winter 2004): 530-51.
16
“The Conflict over Vertical Foreclosure in Competition Policy and Intellectual Property Law.” Journal of
Institutional and Theoretical Economics 160(1) (March 2004): 79-96.
“New Tool for Studying Network Industry Reforms in Developing Countries,” co-authors Scott J.
Wallsten, George Clarke, Luke Haggarty, Rosario Kaneshiro, Mary Shirley and Lixin Colin Xu. Review of
Network Economics 3(3) (September 2004): 248-82.
“Comment: Who Benefits Whom in Local Television Markets?” In Brookings-Wharton Papers on Urban
Affairs 2004, William G. Gale and Janet Rothenberg Pack, eds. Brookings Institution, 2004.
“‘Buyer Power’ and Economic Policy.” Antitrust Law Journal 72(2) (2005): 311-40. Spanish version
“‘Poder de Compra’ y Poltica Econmica.” In Libra Competencia y Retail: Un Anlisis Crtico, Paulo
Montt Rettig and Nicole Behme Zalaquett, eds. Abeledo Perrot Legal Publishing, Santiago, Chile, 2010.
“Einstein’s Interoffice Memo.” Science 309: 5740 (September 2, 2005): 1490-1. Reprinted in CAUTACPPU Bulletin (December 2005): 2.
“The Politics and Economics of Implementing State-Sponsored Embryonic Stem Cell Research.” In States
and Stem Cells, Aaron D. Levine, ed. Policy Research Institute for the Region, Princeton University, 2006.
“Designing an Effective Program of State-Sponsored Human Embryonic Stem-Cell Research.” Berkeley
Technology Law Journal 21(3) (Summer 2006): 1-33.
“Universal Telecommunications Service in India,” co-author Scott J. Wallsten. In India Policy Forum
2005-06, Suman Bery, Barry Bosworth and Arvind Pamagariya, eds. Brookings Institution, 2006.
Reprinted in Spectrum Law and Governance, Ramakistaiah Jilla, ed. Icfai University Press, 2008.
“Conditions for Judicial Independence,” co-authors Mathew D. McCubbins and Barry R. Weingast.
Journal of Contemporary Legal Issues 15(1) (September/October 2006): 107-29.
“Sports Economics after Fifty Years.” In Sports Economics after Fifty Years, Placido Rodriquez, Stefan
Ksenne and Jaume Garcia, eds. University of Oveido Press, 2006.
“Broadcasting and Team Sports.” Scottish Journal of Political Economy 54(3) (July 2007): 400-21.
“Comment: Housing Subsidies and Homeowners.” Brookings/Wharton Papers on Urban Affairs 2007,
Gary Burtless and Janet Rothenberg Pack, eds. Brookings Institution, 2007.
“The Political Economy of Law,” co-authors Mathew D. McCubbins and Barry R. Weingast. In
Handbook of Law and Economics, A. Mitchell Polinsky and Steven Shavell, eds. North-Holland
Publishers, 2007.
“The Economic Significance of Executive Order 13,422.” Yale Journal on Regulation 25(1) (Winter
2008), pp. 113-24.
“The Wines of West Africa: History, Technology and Tasting Notes.” Journal of Wine Economics Vol. 3,
No. 1(May 2008), pp. 85-94.
“Administrative Law Agonistes,” co-authors Mathew D. McCubbins, Daniel B. Rodriguez and Barry R.
Weingast. Columbia Law Review Sidebar Vol. 108 (April 29, 2008), pp. 15-22.
“Comment: London Congestion Charges.” Brookings-Wharton Papers on Urban Affairs 2008, Gary
Burtless and Janet Rothenberg Pack, eds. Brookings Institution, 2008.
17
“Priorities for Telecommunications Reform in Mexico.” In No Growth without Equity? Inequality,
Interests, and Competition in Mexico, Santiago Levy and Michael Walton, eds. Palgrave MacMillan,
2009.
“Malleable Constitutions: Reflections on State Constitutional Reform,” co-author Bruce E. Cain. Texas
Law Review Vol. 87, No. 7 (June 2009), pp. 1517-44.
“Regionalising Infrastructure Reform in Developing Countries,” co-authors Nancy C. Benjamin and
Ioannis N. Kessides. World Economics Vol. 11, No. 3 (July-September 2010), pp. 79-108.
“Institutional Causes of California’s Budget Problem,” co-author Bruce E. Cain. California Journal on
Politics and Policy Vol. 2, No. 2 (September 2010), pp. 1-35.
“Comment” (on articles by William Nordhaus and John Weyant). Energy Economics Vol. 33, No. 4 (July
2011), pp. 683-6.
“Impediments to Innovation in Legal Infrastructure.” I/S: A Journal of Law and Policy for the
Information SocietyVol. 8, No. 1 (Summer 2012), pp. 61-71.
“Introduction,” co-authors Nicholas C. Hope, Anjini Kochar, and T N. Srinivasan. In Economic Reform in
India, Nicholas C. Hope, Anjini Kochar, Roger Noll and T.N. Srinivasan, eds. Cambridge University
Press, 2013.
“An Assessment of Indian Telecommunications Reform,” co-author Scott J. Wallsten. In Economic
Reform in India, Nicholas C. Hope, Anjini Kochar, Roger Noll and T.N. Srinivasan, eds. Cambridge
University Press, 2013.
“Alfred E. Kahn, 1917-2010,” co-author Paul L. Joskow. Review of Industrial Organization Vol. 42, No.
2 (March 2013), pp. 107-26.
OTHER PROFESSIONAL PUBLICATIONS
"Central City–Suburban Employment Distribution." In Final Report: Statistical Papers. President's Task
Force on Suburban Problems, Department of Housing and Urban Development, December 1968.
"Current Mortgage Finance Problems." In Final Report: Policy and Program Papers. President's Task
Force on Suburban Problems, Department of Housing and Urban Development, December 1968.
“A Statistical Analysis of the International Association of Chiefs of Police Poll of the Opinions of the
Opinions of Policemen,” Jet Propulsion Laboratory, Fall 1969.
“The Economics of Professional Basketball,” co-author Benjamin A. Okner. Brookings Reprint No. 258,
(1972). From “Statements,” Senate Antitrust Subcommittee, Hearings on S. 2373, September 21-23, 1971,
Professional Basketball (Part I) and May 3, 1972, Professional Basketball (Part II).
“Prospects and Policies for CATV,” co-authors John J. McGowan and Merton J. Peck. In On the Cable:
Report of the Sloan Commission on Cable Communications. New York: McGraw-Hill, 1971.
"The Price Commission and Regulated Public Utilities." Compendium on Price and Wage Controls: Now
and the Outlook for 1973. Joint Economic Committee, U.S. Congress, 1972.
"The Administration Bill to Deregulate Surface Transportation." In Transportation Policy: The EconomicPolitical Interface, Anthony M. Woodward, ed. Business Research Center, Syracuse University, 1972.
18
"The Case for Viewer Sovereignty," co-authors Merton J. Peck and John J. McGowan. Research Report
No. 135. Brookings Institution, June 1973.
"Regulating the Medical Services Industry." In The Changing Role of the Public and Private Sectors in
Health Care. National Health Council, 1973.
"Decentralization of Public Television." In Conference on Communications Policy Research: Papers and
Proceedings. Office of Telecommunications Policy, Washington, D.C., 1973.
"The Product Market in Sports." In Conference on the Economics of Professional Sports: Proceedings,
George Burman, ed. National Football League Players Association, Washington, D.C., May 7, 1974.
"Government Administrative Behavior and Technological Change." In Government Policies and
Technological Innovation, Vol. II. National Technical Information Service, 1974.
"Public Policy and Innovation: Two Cases," co-author W. David Montgomery. In Government Policies
and Technological Innovation, Vol. II. National Technical Information Service, 1974.
"Subsidization through Regulation: The Case of Broadcasting," co-authors John J. McGowan and Merton
J. Peck. In The Economics of Federal Subsidy Programs: Part 8 - Selected Subsidies. U.S. Government
Printing Office, 1974.
"Comments Regarding Limitations on Programming Available for Broadcast on Pay-TV Channels."
Submitted to Federal Communications Commission. Published in Communications--the Pay-Cable
Industry, Subcommittee on Antitrust and Monopoly, Senate Committee on the Judiciary, June 1975.
"Empirical Studies of Utility Regulation." In Rate of Return Regulation: Proceedings of the Future
Planning Conference. Federal Communications Commission, 1976.
"The Role of Competition in the Electronic Media." In Proceedings of the Symposium on Media
Concentration, Vol. 1, December 14-15, 1978. Bureau of Competition, Federal Trade Commission.
"Adaptive Approaches to the CO2 Problem." In Carbon Dioxide, Climate and Society: A Research
Agenda, Vol. II. U.S. Department of Energy, 1980.
"The Rationale for Social Regulation." In Government Regulation: New Perspective, Andrew R. Blair, ed.
University of Pittsburgh, Graduate School of Business, October 1981.
"Institutional Aspects of Geothermal Development," co-authors Tom K. Lee, Venkatraman Sadanand and
Louis L. Wilde. In Geothermal Probabilistic Cost Study, Vol. II (JPL 5030-491). Jet Propulsion
Laboratory, 1981.
"Looking for Villains in the Energy Crisis." In Energy Independence for the United States: Alternative
Policy Proposals, Nake M. Kamrany, ed. Fundamental Books, 1981; and in U.S. Options for Energy
Independence, Nake M. Kamrany, ed. Heath Lexington Books, 1982.
"What Makes Reform Happen?," co-author Bruce Owen. Regulation 7(2) (March/April 1983): 19-24.
"Recent Social Policy: Comment." In Telecommunications Access and Public Policy, Alan
Baughcum and Gerald Faulhaber, eds. Ablex Publishing Corp., 1984.
"The Economic Viability of Professional Baseball: Report to the Major League Baseball Players
Association." In Representing Professional Athletes and Teams, Philip R. Hochberg and Martin E.
Blackman, eds. Practicing Law Institute, 1986.
19
"The Twisted Pair." Regulation 11(3/4) (1987): 15-22.
"Regulation After Reagan." Regulation 12(3) (1988): 13-20.
"Statement of Goals and Strategies for State Telecommunications Regulation," numerous co-authors.
Aspen Institute, 1989.
"Wirtschaftswachstum, High-Tech-Produkte und internationale Handelspolitik" (in German). In
Deutsch-amerikanische Beziehungen: Politische Freundschaft und wirtschaftliche Kondurrenx?
Haus der Evangelischen Publizistik, 1989.
"The Contemporary Development of International Trade: Approaches, Issues and Problems." In A
Developing World Economy: The Ethics of International Cooperation. Vesper International, 1990.
"Competitive Issues: Enforcement Priorities and Economic Principles." In Antitrust Issues in Regulated
Industries. Charles River Associates, 1991.
"Responding to Referees and Editors." CSWEP Newsletter (February 1993): 15-17. Reprinted in CSWEP
Newsletter Special Reprint Issue No. 2 (1996).
"Biographical Sketches of CSWEP Board Members." CSWEP Newsletter (October 1994).
“Privatization Won’t Foster R&D,” co-author Linda R. Cohen. Public Affairs Report 36(3) (May 1995).
"Constitutional Reform in California," co-author Bruce E. Cain. CPS Brief 7(14) (December 1995).
“Sports, Jobs, and Taxes: Are New Stadiums Worth the Coast?,” co-author Andrew Zimbalist. Brookings
Review 15(3) (Summer 1997): 35-39. Reprinted in Readings in Urban Economics: Issues and Public
Policy, Robert W. Wassmer, ed. Blackwell Publishing, 2000.
“Taxpayers Foot Bill for Sports Boom.” Brookings Newsletter 7(2) (Autumn 1997): 7.
“Unleashing Telecommunications: The Case for True Competition,” co-author Robert Litan. Brookings
Policy Brief # 39 (November 1998). Available at www.brookings.org/comm/policybriefs/pb39 htm.
“Telecommunications Reform in Romania.” In Romania: Regulatory and Structural Assesment in the
Network Utilities, Ioannis Kessides, ed. World Bank Report 20546 RO, 2000.
“Is U.S. Science Policy at Risk?”, co-author Linda R. Cohen. Brookings Review 19(1) (Winter
2001): 10-15.
“Broadband Telecommunications Policy: Ending the Chaos.” SIEPR Policy Brief, December 2001.
“The Supreme Court’s Decision on FCC Pricing Rules,” co-author Gregory Rosston. SIEPR Policy Brief,
May 2002.
“The FCC’s New Television Ownership Rules.” SIEPR Policy Brief, June 2003.
“The Uncertain Future of the Telecommunications Industry,” co-author Robert E. Litan. Brookings Policy
Brief #129, January 2004.
“The Painful Implementation of California’s Stem Cell Research Program,” SIEPR Policy Brief, October
2005.
“The Foreign Aid Paradox.” SIEPR Policy Brief, October 2006.
20
“For More Efficient Subsidy Schemes,” co-author T. N. Srinivasan. Hindu Business Line, April 27, 2006.
(Originally “More Efficient Subsidy Scheme Benefits Consumers, Government, and Economy,” SIEPR
Policy Brief, May 2006.)
“The Regulatory Component of Health Care Reform.” Fresh-Thinking Project, November 2007.
“Designing It Right,” co-author Joe Nation. Environmental Finance vol. 10, no. 2 (December
2008/January 2008), p. 49.
“”Toward a 21st Century Health Care System: Recommendations for Health Care Reform,” 49 co-authors.
Annals of Internal Medicine Vol. 150, No. 7 (April 7, 2009), pp. 493-5.
“Competitive Implications of the Proposed Acquisition of T-Mobile by AT&T Mobility,” co-author
Gregory Rosston. SIEPRPolicy Brief, March 2011.
“Would California Be Better Off Without Ballot Initiatives? No – but Make a Lot of Fixes,” Up For
Discussion, October 2011, Zocalo Public Square, at http://zocalopublicsquare.org/thepublicsquare/2011
/10/04/this-doggone-direct-democracy/read/up-for-discussion/.
OTHER PROFESSIONAL PAPERS
"An Economic Analysis of Network Operations Research Techniques." Ph.D. dissertation, Harvard
University, 1967.
"Perspectives on Rural-Urban Migration." Council of Economic Advisors, November 1968.
"The Economics of Pollution Abatement." Presented at Annual Meeting of the American Association for
the Advancement of Science, December 1970.
"Comments Regarding the Public Interest in Commission Rules and Regulations Relating to Cable
Television, Signal Importation and the Development of UHF Independent Commercial Stations," coauthors John J. McGowan and Merton J. Peck. Submitted to FCC Docket 18397-A, February 10, 1971.
"A Dynamic Theory of Political Campaigning," co-author John A. Ferejohn. Annual Meeting of the
American Political Science Association, September 1972, Washington, D.C.
“Comments Regarding Limitations on Programming Available for Broadcast on Pay-TV Channels.”
Submitted to FCC Docket 19554. Social Science Working Paper No. 65, California Institute of
Technology, September 20, 1974.
"Statement on Regulatory Reform." In Regulatory Reform - 1975: Hearings before the Committee on
Government Operations, U.S. Senate, 94th Cong., 1st Sess., 1975.
"The Causes of Regulatory Failures." In Oversight of Civil Aeronautics Board Practices and Procedures
Hearings before the Subcommittee on Administrative Practices and Procedures, Senate Committee on the
Judiciary, 94th Cong., 1st Sess., 1975.
"Responses to Disaster: Planning for a Great Earthquake in California," co-authors Linda Cohen and Barry
Weingast. Social Science Working Paper No. 131, California Institute of Technology, April 1977.
"Statement on Public Policy Toward Sports." Hearings: Select Committee on Professional Sports, U.S.
House of Representatives, September 1976.
21
"Statement on H.R. 11611." Drug Regulation Reform Act of 1978: Hearings before the Subcommittee on
Health and the Environment, U.S. House of Representatives, June 1978, p. 2156ff.
“Television and Competition.” Symposium on Media Concentration, Federal Trade Commission,
December 1978.
"The Economics of Boxing Regulation in California," co-authors Joel A. Balbien and James P. Quirk.
Social Science Working Paper No. 366, California Institute of Technology, January 1981. Report to
California Athletic Commission.
Implementing Tradable Emissions Permits for Sulfur Oxides Emissions in the South Coast Basin (three
volumes), co-authors Glen Cass and Robert Hahn. Report to California Air Resources Board. Caltech
Environmental Quality Laboratory, June 1983.
"Economic Effects of the Financial Interest and Syndication Rule," co-authors Robert Crandall and Bruce
Owen. Economists, Inc., April 1983. Submitted to Federal Communications Commission, Inquiry on
Television Networks.
"Pay and Performance in Baseball: Modeling Regulars, Reserves and Expansion," co-author Rodney D.
Fort. Social Science Working Paper No. 527, California Institute of Technology, Pasadena, CA 1984.
"Promises, Promises: Campaign Contributions and the Reputation for Services," co-author John Ferejohn.
Social Science Working Paper No. 545, California Institute of Technology, Pasadena, CA, 1984.
“The Economic Viability of Professional Baseball: A Report to the Major League Baseball Players’
Association.” Major League Baseball Players’ Association, 1985.
"Local Telephone Prices and the Subsidy Question," co-author Nina W. Cornell. Presented at
Annual Meetings of the American Economic Association, December 1984, and at the Conference
on Telecommunications Demand Modeling, Bell Communications Research, October 1985.
"The Economics of Bell Operating Company Diversification in the Post-Divestiture Telecommunications
Industry," co-authors Kenneth Baseman and Stephen Silberman. ICF, Incorporated, September 1986.
Submitted in First Triennial Review, Modification of Final Judgment, U.S. vs. AT&T.
"Two's Company, Three's a Crowd: Duverger's Law Reconsidered," co-author John A. Ferejohn.
Presented at Annual Meeting of the American Political Science Association, September 1987.
"Telecommunications Reform in Brazil." Report to the International Bank for Reconstruction and
Development, September 1990.
"Marketable Emissions Permits in Los Angeles." Report to the South Coast Air Quality Management
District. Center for Economic Policy Publication No. 285, Stanford University, January 1991.
"Statement on the Baseball Antitrust Exemption." Hearings before the Subcommittee on Antitrust,
Monopolies, and Business Rights, Committee on the Judiciary, United States Senate, December 1992.
“Regulatory Transparency in Telecommunications: Public Interest Standards for BOC Entry into Long
Distance,” co-author Robert Litan, August 1998. Submitted to FCC Docket CCBPol 98-4.
“Remedies Brief of Amicus Curie,” co-authors Robert Litan, William Nordhaus and Frederic
Scherer. U.S. v. Microsoft, U. S. District Court, District of Columbia, April 2000. Available at:
www.aei.brookings.org/publications/index.php?tab=author&authorid=14.
“ Promoting Efficient Use of Spectrum Through Elimination of Barriers to the Development of Secondary
Markets,” 36 co-authors. Submitted to Federal Trade Commission, February 2001.
22
“Notes on Privatizing Infrastructure Industries.” World Bank Development Report Planning Conference,
July 2001.
“Comments on Revised Proposed Final Settlement,” co-authors Robert Litan and William Nordhaus. U.S.
v. Microsoft, U. S. District Court, District of Columbia, January 2002.
“The Copyright Term Extension Act of 1998: An Economic Analysis,” Amicus Brief in Support of
Petititoners, Eldred, et al., vs. Aschroft, U. S. Supreme Court, May 2002, sixteen co-authors.
“Comments on the Federal Trade Commission’s Strategic Plan for 2003,” co-authors Robert W. Hahn and
Robert E. Litan. AEI-Brookings Joint Center for Regulatory Studies, July 2003.
“Brief of Amici Curiae,” 32 co-authors PSEG Fossil, et al., vs. Riverkeeper Inc., U. S. Supreme Court, July
21, 2008.
“A Statement on the Appropriate Role for Research and Development in Climate Policy,” 9 co-authors,
Berkeley Electronic Press, December 9, 2008.
“Comments of 71 Concerned Economists: Using Procurement Auctions to Allocate Broadband Stimulus
Grants,” National Telecommunications Information Agency and Rural Utilities Service, April 13, 2009.
“Amicus Curiae Brief in Support of Petitioner,” 19 co-authors, American Needle vs. National Football
League, U. S. Supreme Court, September 25, 2009.
“Regionalizing Telecommunications Reform in West Africa,” co-authors Ioannis N. Kessides and Nancy
C. Benjamin. Policy Research Working Paper No. 5126, World Bank, November 2009.
“Comment of Sports Economists on the FCC’s Sports Blackout Rules,” eight co-authors, submitted in
Docket MB 12-3, Federal Communications Commission, February 2012.
POPULAR PUBLICATIONS
"School Bonds and the Future of Pasadena." Pasadena Star-News (April 20, 1969): C-1.
"After Vietnam, Another Recession?" Caltech News (June 1969).
"People Is a Dirty Word," (with others). Pest Control Operators News 30 (February/March 1970).
(Transcript of panel discussion, radio station KMPC, Los Angeles.)
"Defending Against Disasters." Engineering and Science 39 (May-June 1976).
"Quake Prediction: For Public Officials the Problems Are Mind-Bending." Los Angeles Times (May 2,
1976), Part VIII: 5.
"Professional Sports: Should Government Intervene?" San Francisco Chronicle (June 7, 1977).
"Fact and Fancy About the Energy Crisis." Pasadena Star-News (July 27, 1980): 17.
"Looking for Villains in the Energy Crisis." Town Hall Reporter 13(8) (August 1980): 12.
"Using the Marketplace to Reform Regulation." Washington University Center for the Study of American
Business, Whittemore House Series 5 (1982).
23
"Leasing the Air: An Alternative Approach to Regulation?" Engineering and Science 46(1) (September
1982): 12-17.
"Television Ownership and the FCC: Let a Free Market Set the Pace." New York Times (August 26,
1984), Business Section: 2.
"Trends in California's Economy: Implications for the Future." Engineering and Science LVI (1) (Fall
1992): 9-13
"Help Business, but Beware of High-tech Pork," co-author Linda R. Cohen. USA Today, March 18, 1993,
p. 11A.
"The Decline of Public Support for R&D." Framtider International, Vol. 5 (1995): 23-27.
"Wild Pitch." New York Times, April 11, 1996. Reprinted in New York Times Op-Classic, 2008.
“Revisiting Telecom Reform,” co-author T. N. Srinivasan. Business Standard of India, August 21-22,
1999.
“32 Voices on the State of the Game,” 31 co-authors. The Biz of Baseball, December 15, 2006.
“Sharing a Stadium Makes Perfect Sense.” Oakland Tribune, February 9, 2007, Sports Section: 3.
“Six Views on Former Commissioner Bowie Kuhn,” five co-authors. The Biz of Baseball, March 26,
2007.
“Why Analysis of 49ers Move is Too Rosy.” San Jose Mercury News, April 15, 2007: 1P.
“Baseball Economics Roundtable,” six co-authors. The Biz of Baseball, May 3, 2007.
“Are Stadiums Worth the Price?” San Francisco Chronicle, July 8, 2007: E1, E3.
“Voices on Barry Bonds,” ten co-authors. The Biz of Baseball, November 27, 2007.
“Is Tim Gaithner’s Toxic Asset Plan Toxic?” seven co-authors. Foreign Policy, March 23, 2009.
“A Few Thoughts about Measure J,” San Francisco Chronicle, May 16, 2010.
REVIEW ARTICLES
"Assessing the Energy Situation." Science, 208(4445) (May 16, 1980): 701-702.
"Energy Data and Political Polarization." Science, 214(4524) (November 27, 1981): 1019.
"Handbook for Reform: Breyer on Regulation." Columbia Law Review, 83(4) (May 1983): 1109-1119.
"Rules in the Making, by Magat, Krupnick, and Harrington." Rand Journal of Economics, 18(3) (Autumn
1987): 461-464.
"Fields of Greed." New York Times Book Review (April 4, 1993): 24-25.
BOOK REVIEWS
"Cure for Chaos, by Simon Ramo." Engineering and Science 33 (November 1969).
24
"Technology and Market Structure, by Almarin Phillips." Journal of Economic Literature 10 (December
1972): 1253-1255.
"The Telecommunications Industry, by Gerald W. Brock." Journal of Economic Literature 20 (September
1982): 1096-97.
"Presidential Management of Science and Technology, by W. Henry Lambright." Science, 231(4739)
(February 14, 1986): 749.
"Telecommunications Economics and International Regulatory Policy, by Snow and Jusawalla."
Information Economics and Policy, 2(4) (1986): 318-319.
"The Economist's View of the World, by Steven E. Rhoads." American Political Science Review 81(1)
(March 1987): 339-341.
"Air Pollution Regulation, by Richard A. Liroff." Environmental Professional 9(4) (1987): 359-360.
"The Sports Industry and Collective Bargaining, by Paul D. Staudohar." Industrial and Labor Relations
Review 41(2) (January 1988): 314-315.
"The Social Context of New Information and Communication Technologies, by Elia Zureik and Dianne
Hartling." Information Economics and Policy 3(2) (1988): 204.
"The United States and the Direct Broadcast Satellite, by Sara Fletcher Luther." Information Economics
and Policy 4(1) (1989/90): 83-84.
"Regulation and Markets, by Daniel F. Spulber." Journal of Economic Literature 28(4) (December 1990):
1757-1759.
"A Legislative History of the Communications Act of 1934, edited by Max Paglin." Information Economics
and Policy 4(2) (1990): 190-94.
"International Telecommunications in Hong Kong: The Case for Liberalization, by Milton Mueller."
Information Economics and Policy 4(3) (1990): 273-274.
"Strategy and Choice, edited by Richard J. Zeckhauser." Journal of Policy Analysis and Management,
12(2) (Spring 1993): 386-388.
"Risky Business: Breaking the Vicious Circle, by Stephen Breyer." Regulation, 16(3) (1993): 82-83.
"Government's Role in Innovation, by D.P. Leyden and A.N. Link." Journal of Economics (Zeitschrift für
National Okönomie) 54(3) (1994): 333-335.
"Regulation, Organizations, and Politics, by Lawrence S. Rothenberg." American Political Science
Review 89(3) (September 1995): 768-769.
“The Political Economy of Public Administration: Institutional Choice in the Public Sector, by Murray J.
Horn.” Economic Record 73 (June 1997): 187-188.
“Making and Breaking Governments, by Michael Laver and Kenneth A. Shepsle.” Zeitschrift fr
Nationalkonomie (Journal of Economics) 66(3) (December 1997): 324-326.
“The Economics of Sports Broadcasting, by Chris Gratton and Harry Arne Solberg.” Journal of Media
Economics Vol. 23, No. 1 (2010): 42-5
2/13
25
APPENDIX B
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
Document Title
1.
Order Granting in Part Defendant’s Motion for Judgment on the Pleadings; Ordering Supplemental
Briefing
2.
Order Granting Defendant’s Motion for Judgment on the Pleadings As to the First Cause of Action for
Violations of Section 1 of the Sherman Act and The Fifth Cause of Action for Violations of the Cartwright
Act
3.
Order Decertifying Classes Without Prejudice to Being Renewed; Inviting Further Motions
4.
Defendant’s Motion for Decertification of Rule 23(B)(3) Class
5.
Expert Report of Dr. Michelle M. Burtis
6.
Declaration of Michael Scott in Support of Defendant’s Motion for Decertification of Rule 23(B)(3) Class
7.
Plaintiffs’ Memorandum in Opposition to Defendant’s Motion for Decertification of Rule 23(b)(3) Class
8.
Reply Declaration of Roger G. Noll
9.
Declaration of Thomas R. Merrick in Support of Plaintiffs’ Memorandum in Opposition to Defendant’s
Motion for Decertification of Rule 23(b)(3) Class
10.
Defendant’s Reply In Support of Motion for Decertification of Rule 23(B)(3) Class
11.
Reply Expert Report of Dr. Michelle M. Burtis
12.
Declaration of David C. Kiernan in Support of Reply on Motion for Decertification of Rule 23(B)(3) Class
13.
Amended Consolidated Complaint for Violations of Sherman Antitrust Act, Clayton Act, Cartwright Act,
California Unfair Competition Law, Consumers Legal Remedies Act, And California Common Law of
Monopolization
14.
Defendant Apple Inc.’s First Amended Answer and Defenses to Plaintiffs’ Amended Consolidated
Complaint
15.
Order Granting in Part and Denying in Part Defendant’s Motion to Dismiss; Denying as Premature
Defendant’s Motion for Summary Judgment; Granting Indirect Purchaser Plaintiff Leave to File an
Amended Complaint
16.
Apple’s Motion to Dismiss or, Alternatively, For Summary Judgment
17.
Declaration of Michael Scott in Support of Apple’s Motion to Dismiss or, Alternatively, For Summary
Judgment
18.
Declaration of Jeffrey Robbin in Support of Defendant’s Motion to Dismiss or in the Alternative Motion
For Summary Judgment
19.
Plaintiffs’ Memorandum in Opposition to Apple’s Motion to Dismiss or, Alternatively, For Summary
Judgment
20.
Declaration of Thomas R. Merrick in Support of Plaintiffs’ Memorandum in Opposition to Apple’s Motion
to Dismiss or, Alternatively, For Summary Judgment
21.
Declaration of Paula M. Roach Pursuant to Rule 56(f) of the Federal Rules of Civil Procedure in Support
of Plaintiffs’ Opposition to Apple’s Motion to Dismiss or, Alternatively, For Summary Judgment
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
22.
Apple’s Reply in Support of Motion to Dismiss or, Alternatively, For Summary Judgment
23.
Declaration of David C. Kiernan in Support of Apple’s Reply in Support of Motion to Dismiss or,
Alternatively, For Summary Judgment
24.
Plaintiffs’ Reply in Support of Their Motion for Additional Discovery Pursuant to Rule 56(f)
25.
Declaration of Thomas R. Merrick in Support of Motion for Additional Discovery Pursuant to Rule 56(f)
26.
Transcript of Proceedings for May 10, 2010
27.
Defendant Apple Inc.’s Objections and Responses to Plaintiffs’ Amended First Set of Requests for
Admission
28.
Defendant Apple Inc.’s Response to Plaintiffs’ Amended First Set of Requests for Production of
Documents
29.
Defendant’s Response to Plaintiffs’ Second Set of Requests for Production of Documents
30.
Defendant’s Response to Plaintiffs’ Third Set of Requests for Production of Documents
31.
Defendant’s Responses to Plaintiffs’ First Set of Interrogatories Directed at Defendant Apple Computer,
Inc.
32.
Defendant Apple Inc.’s First Amended Objections and Answers to Plaintiffs’ Amended First Set of
Interrogatories
33.
Defendant Apple Inc.’s Supplemental Objections and Answers to Plaintiffs’ Amended First Set of
Interrogatories
34.
Defendant Apple Inc.’s First Amended Objections and Answers to Plaintiffs’ Second Set of Interrogatories
14-16, 18-19
35.
Defendant Apple Inc.’s First Amended Objections and Answers to Plaintiffs’ Second Set of Interrogatories
9-13
36.
Defendant Apple Inc.’s Objections and Answers to Plaintiffs’ Third Set of Interrogatories
37.
Defendant Apple Inc.’s Responses to Plaintiff’s First Set of Requests for Production of Documents
[Indirect Purchaser Plaintiff Action]
38.
Defendant Apple Inc.’s Responses to Plaintiff’s First Set of Interrogatories [Indirect Purchaser Plaintiff
Action]
39.
Declaration of Roger G. Noll (July 15, 2008)
40.
Reply Declaration of Roger G. Noll (October 19, 2009)
41.
Declaration of Roger G. Noll (January 18, 2011)
42.
Reply Declaration of Roger G. Noll (March 28, 2011)
43.
Supplemental Declaration of Roger G. Noll (July 18, 2011)
44.
Second Supplemental Declaration of Roger G. Noll (September 23, 2011).
45.
Consolidated Complaint for Violations of Sherman Antitrust Act, Clayton Act, Cartwright Act, California
Unfair Competition Law, Consumer Legal Remedies Act, and California Common Law of Monopolization,
filed April 19, 2007.
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
46.
Defendant Apple Inc.’s Responses to Plaintiff’s First Set of Interrogatories, dated August 28, 2008 (Somers
v. Apple, Inc., Case No. C 07-6507 JW), Exhibits A & B.
47.
Defendant Apple Inc.'s Responses to Plaintiff's First Set of Interrogatories with Attachments, Somers v.
Apple Inc. , August 28, 2008.
48.
Affidavit of Gary L. French, Ph.D. with Exhibits and Appendices, The Apple iPod iTunes Antitrust
Litigation , February 13, 2009.
49.
Reply Affidavit of Gary L. French, Ph.D., Regarding Class Certification with Exhibits, The Apple iPod
iTunes Antitrust Litigation , May 14, 2009.
50.
Expert Report of Dr. Michelle M. Burtis with Exhibits, Somers v. Apple, Inc. , June 17, 2009.
51.
Apple's Motion for Summary Judgment, The Apple iPod iTunes Antitrust Litigation , January 18, 2011.
52.
Apple's Opposition to Renewed Motion for Class Certification, The Apple iPod iTunes Antitrust Litigation ,
February 28, 2011.
53.
Expert Report of Dr. Michelle M. Burtis with Exhibit A, The Apple iPod iTunes Antitrust Litigation ,
February 28, 2011.
54.
Plaintiffs’ Memorandum in Opposition to Apple’s Motion for Summary Judgment, The Apple iPod iTunes
Antitrust Litigation , February 28, 2011.
55.
Apple's Reply in Support of its Motion for Summary Judgment, The Apple iPod iTunes Antitrust
Litigation , March 18, 2011.
56.
Apple’s Opposition to Motion to Exclude the Opinions of Defendant’s Expert, Dr. Michelle M. Burtis,
Ph.D., The Apple iPod iTunes Antitrust Litigation , April 11, 2011.
57.
Apple’s Supplemental Objections to Reply Declaration of Roger C. Noll and Supplemental Opposition to
Class Certification Motion, The Apple iPod iTunes Antitrust Litigation , April 11, 2011.
58.
Expert Report of Dr. Michelle M. Burtis in Support of Apple Inc.'s Opposition to Plaintiffs' Motion to
Exclude with Exhibits, The Apple iPod iTunes Antitrust Litigation , April 11, 2011.
59.
Apple’s Opposition to Motion to Compel, The Apple iPod iTunes Antitrust Litigation , April 12, 2011.
60.
Apple’s Reply In Support of Its Motion For Summary Judgment, April 18, 2011.
61.
Apple's Motion for Summary Judgment, April 18, 2011
62.
Plaintiffs’ Memorandum In Opposition To Apple’s Motion For Summary Judgment, April 18, 2011.
63.
Order Granting in Part and Denying in Part Defendant’s Motion for Summary Judgment; Denying as
Premature Plaintiffs’ Motion for Class Certification, The Apple iPod iTunes Antitrust Litigation , May 19,
2011.
64.
Apple’s Supplemental Brief Re Class Certification, The Apple iPod iTunes Antitrust Litigation , June 6,
2011.
65.
Supplemental Brief In Support of Plaintiffs’ Renewed Motion for Class Certification and Response to
Court’s May 19, 2011 Order, The Apple iPod iTunes Antitrust Litigation , June 6, 2011.
66.
Apple Exhibit A, Class Definition, June 23, 2011.
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
67.
Apple’s Further Supplemental Brief Re Class Certification, The Apple iPod iTunes Antitrust Litigation ,
June 23, 2011.
68.
Plaintiff's Response to the Court's June 22, 2011 Order Requiring Further Supplemental Briefing, The
Apple iPod iTunes Antitrust Litigation , June 23, 2011.
69.
Apple’s Response to Professor Noll’s July 18 Declaration, The Apple iPod iTunes Antitrust Litigation ,
July 22, 2011.
70.
Supplemental Report of Dr. Michelle M. Burtis with Exhibits, The Apple iPod iTunes Antitrust Litigation ,
July 22, 2011.
71.
Transcript of Proceedings (Motions Hearing) for June 27, 2011.
72.
Second Supplemental Report of Dr. Michelle M. Burtis with Exhibits, The Apple iPod iTunes Antitrust
Litigation , November 14, 2011.
73.
Order Granting Plaintiffs’ Motion for Class Certification, The Apple iPod iTunes Antitrust Litigation ,
November 22, 2011.
74.
Defendant Apple Inc.’s Answer and Defenses to Plaintiffs’ Consolidated Complaint
75.
Defendant’s Notice of Motion and Motion to Dismiss Antitrust Claims
76.
Defendant’s Responses to Plaintiffs’ First Set of Interrogatories Directed at Apple Computer, Inc.
77.
Apple’s Supplementary Response to Interrogatories, Interrogatory No. 6.
78.
Consolidated Compact Disc Antitrust Litigation , U. S. District Court, Los Angeles, California.
79.
Decision, U. S. v. Microsoft , U. S. Court of Appeals for District of Columbia.
80.
Findings of Fact, U. S. v. Microsoft , U. S. District Court for the District of Columbia.
81.
In Re: DRAM Antitrust Litigation , U. S. District Court, San Francisco, California.
82.
In Re: Static Random Access Memory (SRAM) Litigation , U. S. District Court, San Francisco.
83.
In Re: Tableware Antitrust Litigation , U. S. District Court, San Francisco, California.
84.
“Memorandum and Order,” In re Napster, Inc., Copyright Litigation , U.S. District Court (Northern
California), Case No. MDL 00-1369 MHP, February 2002.
85.
R&D Business Systems v. Xerox , U. S. District Court, Marshall, Texas.
86.
Southeast Georgia Regional Medical Center v. General Electric Corporation , U. S. District Court,
Brunswick, Georgia.
87.
Declaration of Augustin Farrugia In Support Of Defendant's Renewed Motion For Summary Judgment,
January 2011.
88.
Declaration of David C. Kiernan in Support of Apple Opposition to Plaintiffs' Motion to Exclude the
Opinions of Defendant's Expert, Dr. Michelle M. Burtis, Ph.D. with Exhibits, The Apple iPod iTunes
Antitrust Litigation , April 11, 2011.
89.
Declaration of David C. Kiernan in Support of Apple’s Opposition to Plaintiffs’ Motion to Compel, The
Apple iPod iTunes Antitrust Litigation , April 12, 2011.
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
90.
Declaration of David F. Martin In Support Of Plaintiffs' Opposition To Apple's Motion For Summary
Judgment, with Exhibit A, February 28, 2011.
91.
Declaration of Jeffrey Robbin In Support Of Defendant's Renewed Motion For Summary Judgment, April
18, 2011.
92.
Declaration of Amanda Marks
93.
Declaration of Howie Singer
94.
Declaration of Lawrence Kanusher
95.
Declaration of Mark Piibe
96.
Deposition Transcript of Roger G. Noll, taken September 19, 2008
97.
Deposition Transcript of Michelle M. Burtis, Ph.D, taken June 23, 2009 and Exhibits 1-3
98.
Deposition Transcript of Michelle M. Burtis, Ph.D, taken September 30, 2009 and Exhibits 1-2
99.
Deposition Transcript of Roger G. Noll, taken October 27, 2009 and Exhibit 1
100.
Deposition Transcript of Jeffrey L Robbin, taken December 3, 2010 and Exhibits 1-20
101.
Deposition Transcript of Augustin J. Farrugia, taken December 8, 2010 and Exhibits 21-31
102.
Deposition Transcript of David K. Heller, taken December 15, 2010 and Exhibits 32- 51
103.
Deposition Transcript of Arthur Rangel, taken December 17, 2010 and Exhibits 80-94
104.
Deposition Transcript of Eddy Cue, taken December 17, 2010 and Exhibits 52-74
105.
Deposition Transcript of Mark Donnelly, taken December 20, 2010 and Exhibits 95- 118
106.
Deposition Transcript of Michelle M. Burtis, Ph.D, taken March 14, 2011 and Exhibits 1-2
107.
Deposition Transcript of David M. Martin, JR., Ph.D, taken March 18, 2011
108.
Deposition Transcript of Roger G. Noll, taken April 7, 2011
109.
120610_ResellerLocations_SG
110.
AIIA iPod Gross Margin Reports
111.
AIIA_iTunes iTS Fcsts Q4FY05-Q4FY10
112.
AIIA_US iPod Sales_Family Level_Direct-Indirect-OEM.xlsx
113.
AIIA_US Sales_PPN Level_FY02-FY10
114.
AMR iPod Forecasts_12-13-10 WW Music Advertising FY06-FY10
115.
Funds iPod Data FY2007
116.
Funds iPod Data FY2008
117.
Funds iPod Data FY2009
118.
Funds iPod Data FY2010
119.
Funds iPod Data Q3_Q4FY2006
120.
FY 2010 iPhone Summary
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
121.
iPod Costed BOM Data
122.
iPod LCPM – Launch-Q2FY09
123.
iPod LCPM as of 11-30-10
124.
iTunes Client FY2002 to FY2010_cc Monthly Expenses
125.
NPD Techworld’s MP3 Player Market Tracking Service Data 2002-2010
126.
NPD_DigitalOnlyJuly06_Dec11.xlsx
127.
NPD_PhysicalandDigital_July06_Dec11.xlsx
128.
NPDMP3Aug07Update.xls
129.
NPDMP3October2005Update.xls
130.
NPDUSMP3April2009.xls
131.
NPDUSMP3July2010.xls
132.
NPDUSMP3June08Update.xls
133.
NPDUSMP3PriceBandJuly2010.xls
134.
US iTS P&Ls_Q3FY03-Q4FY10
135.
US Reseller Soft Dollar Programs (BDF)
136.
US Sales_iPods_Direct_Indirect_OEM_FY02-FY10
137.
US TV Advertising Spend by Campaign_FY04-FY1
138.
US_Sales_PPN Level_FY02-FY10
139.
Worldwide Sales _iPods_FY02-FY10
140.
WW Music Advertising FY06-FY10
141.
WW_iPod Exp. Level View Data
142.
Direct Sales Transaction Data
143.
Price Override Data
144.
Reseller Transaction Data
145.
iPod BDF Accruals 2001-2011
146.
BDF Retailer Names
147.
Apple’s answers to the “Additional Questions for Apple Regarding Matching of Direct Sales and Returns,”
SFI-819714v.1.
148.
SFI_808637_2_Responses to Price Override Questions.pdf
149.
88493_7 Responses to Costed BOMs and Gross Margins
150.
752597_1.xlsx - channel code descriptions node 5600
151.
792310_1.xlsx - reason codes for restocking fees
152.
ChannelHIERARCHY (full) 02.05.13.xls
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
153.
MPN Attributes.xlsx
154.
ORDER_TYPE 09.25.12.xlsx
155.
SFI_744267_1 09.25.12 Answers to 8.22.2012 Data questions - direct sales transaction data.DOCX
156.
SFI_791971_1_PriceOverrideReasonDesc.XLSX
157.
Price_group_1.xlsx
158.
Price_group_7.xlsx
159.
Price_group_MISC.xlsx
160.
Price_overirde_12.xlsx
161.
Price_overirde_code14.xlsx
162.
POS_PD3_Invoice_Xref2002.txt
163.
POS_PD3_Invoice_Xref2003.txt
164.
POS_PD3_Invoice_Xref2004.txt
165.
POS_PD3_Invoice_Xref2005.txt
166.
POS_PD3_Invoice_Xref2006.txt
167.
POS_PD3_Invoice_Xref2007.txt
168.
POS_PD3_Invoice_Xref2008.txt
169.
POS_PD3_Invoice_Xref2009.txt
170.
POS_PD3_Invoice_Xref2010.txt
171.
POS_PD3_Invoice_Xref2011.txt
172.
POS_PD3_Invoice_Xref2012.txt
173.
PD2_Invoice_Xref_CreditsDebitsReturns.txt
174.
2011 Year-End Shipment Statistics, RIAA, http://www.riaa.com/keystatistics.php?content_selector=riaashipment-database-log-in
175.
Piper Jaffrey, "Apple Inc. - NPD Macs Continue to Track up as Apple Meets iMac Demand", Thomson
One Report #21611871, March 18, 2013, at 4.
176.
RIAA Year End Industry Shipment and Revenue Statistics,
http://www.riaa.com/keystatistics.php?content_selector=research-shipment-database-overview.
177.
AIIA00187783-823
178.
Apple AIIA00093729
179.
Apple_AIIA_ 000904429-31
180.
Apple_AIIA_00019916
181.
Apple_AIIA_00090405-07
182.
Apple_AIIA_00090447-79
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
183.
Apple_AIIA_00090471
184.
Apple_AIIA_00099408
185.
Apple_AIIA_00142141-43
186.
Apple_AIIA_00182671
187.
Apple_AIIA_00799692
188.
Apple_AIIA_00975685
189.
Apple_AIIA_00979727
190.
Apple_AIIA_01278697
191.
Apple_AIIA_01384979
192.
Apple_AIIA_0712612
193.
Apple_AIIA_B_000001
194.
Apple_AIIA_B_000104 – 199
195.
Apple_AIIA_B_000106 - Apple_AIIA_B_000365
196.
Apple_AIIA_B_000115 -Apple_AIIA_B_000149
197.
Apple_AIIA_B_000150 -Apple_AIIA_B_000176
198.
Apple_AIIA_B_000177 -Apple_AIIA_B_000189
199.
Apple_AIIA_B_000190 -Apple_AIIA_B_000192
200.
Apple_AIIA_B_000200 – 299
201.
Apple_AIIA_B_000204 -Apple_AIIA_B_000229
202.
Apple_AIIA_B_000249 -Apple_AIIA_B_000257
203.
Apple_AIIA_B_000258
204.
Apple_AIIA_B_000259 -Apple_AIIA_B_000264
205.
Apple_AIIA_B_000265 -Apple_AIIA_B_000312
206.
Apple_AIIA_B_000300 – 393
207.
Apple_AIIA_B_000313 -Apple_AIIA_B_000328
208.
Apple_AIIA_B_000340 -Apple_AIIA_B_000347
209.
Apple_AIIA_B_000348 -Apple_AIIA_B_000376
210.
Apple_AIIA_B_000394
211.
Apple_AIIA_B_000796
212.
Apple_AIIA_B_001540 – 571
213.
Apple_AIIA_B_002813 – 835
214.
Apple_AIIA_B_003134 – 170
215.
Apple_AIIA_B_003359 – 372
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
216.
Apple_AIIA_B_012059 – 094
217.
Apple_AIIA_B_012668 – 679
218.
Apple_AIIA_B_012951 – 952
219.
Apple_AIIA_B_012964
220.
Apple_AIIA_B_012974 – 996
221.
Apple_AIIA_C_00000411
222.
Apple_AIIA_C_00003450
223.
Apple_AIIA00000001 – 005
224.
Apple_AIIA00000674
225.
Apple_AIIA00002406
226.
Apple_AIIA00002879 – 894
227.
Apple_AIIA00002895
228.
Apple_AIIA00002896
229.
Apple_AIIA00002897
230.
Apple_AIIA00002905 – 912
231.
Apple_AIIA00019916
232.
Apple_AIIA00031960
233.
Apple_AIIA00042247
234.
Apple_AIIA00042401
235.
Apple_AIIA00042418
236.
Apple_AIIA00042838
237.
Apple_AIIA00042905
238.
Apple_AIIA00049006
239.
Apple_AIIA00049400
240.
Apple_AIIA00057978
241.
Apple_AIIA00062113
242.
Apple_AIIA00089936
243.
Apple_AIIA00089937
244.
Apple_AIIA00089938
245.
Apple_AIIA00089939
246.
Apple_AIIA00090359 – 360
247.
Apple_AIIA00090361 – 364
248.
Apple_AIIA00090365 – 369
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
249.
Apple_AIIA00090370 – 372
250.
Apple_AIIA00090400
251.
Apple_AIIA00090405 – 407
252.
Apple_AIIA00090412 – 414
253.
Apple_AIIA00090415 – 417
254.
Apple_AIIA00090418 – 421
255.
Apple_AIIA00090429 – 431
256.
Apple_AIIA00090435 – 437
257.
Apple_AIIA00090441 – 443
258.
Apple_AIIA00090447 – 479
259.
Apple_AIIA00090453
260.
Apple_AIIA00090456 – 458
261.
Apple_AIIA00090467 – 468
262.
Apple_AIIA00090471
263.
Apple_AIIA00090472
264.
Apple_AIIA00090472 – 473
265.
Apple_AIIA00090474 – 478
266.
Apple_AIIA00090479 – 480
267.
Apple_AIIA00090481
268.
Apple_AIIA00090482 – 483
269.
Apple_AIIA00090495 – 497
270.
Apple_AIIA00090498
271.
Apple_AIIA00090510 – 511
272.
Apple_AIIA00090519
273.
Apple_AIIA00090536
274.
Apple_AIIA00090539
275.
Apple_AIIA00090600 – 602
276.
Apple_AIIA00090607
277.
Apple_AIIA00090608
278.
Apple_AIIA00090611 – 613
279.
Apple_AIIA00090615
280.
Apple_AIIA00090616 – 617
281.
Apple_AIIA00090667
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
282.
Apple_AIIA00090703
283.
Apple_AIIA00090850 – 851
284.
Apple_AIIA00090853 – 854
285.
Apple_AIIA00090862 – 866
286.
Apple_AIIA00090867 – 874
287.
Apple_AIIA00090920 – 922
288.
Apple_AIIA00090940
289.
Apple_AIIA00090953 – 957
290.
Apple_AIIA00090976
291.
Apple_AIIA00091049 – 051
292.
Apple_AIIA00091511
293.
Apple_AIIA00091512 – 514
294.
Apple_AIIA00091678
295.
Apple_AIIA00091714 – 715
296.
Apple_AIIA00091723 – 724
297.
Apple_AIIA00091742
298.
Apple_AIIA00091761 – 762
299.
Apple_AIIA00091766
300.
Apple_AIIA00091780
301.
Apple_AIIA00091783
302.
Apple_AIIA00091826 – 837
303.
Apple_AIIA00091931
304.
Apple_AIIA00091932 – 933
305.
Apple_AIIA00091967
306.
Apple_AIIA00092158
307.
Apple_AIIA00092162
308.
Apple_AIIA00092166
309.
Apple_AIIA00092243
310.
Apple_AIIA00092245
311.
Apple_AIIA00092246 – 247
312.
Apple_AIIA00092287
313.
Apple_AIIA00092348 – 378
314.
Apple_AIIA00092415 – 416
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
315.
Apple_AIIA00092454
316.
Apple_AIIA00093325 – 331
317.
Apple_AIIA00093477 – 480
318.
Apple_AIIA00093504
319.
Apple_AIIA00093729
320.
Apple_AIIA00093729 – 736
321.
Apple_AIIA00093858
322.
Apple_AIIA00093859
323.
Apple_AIIA00093861
324.
Apple_AIIA00093875 – 876
325.
Apple_AIIA00093895
326.
Apple_AIIA00094066 – 068
327.
Apple_AIIA00094079 – 084
328.
Apple_AIIA00094086
329.
Apple_AIIA00094118
330.
Apple_AIIA00094123
331.
Apple_AIIA00094124 – 126
332.
Apple_AIIA00094362 – 369
333.
Apple_AIIA00094370 – 382
334.
Apple_AIIA00094564 – 569
335.
Apple_AIIA00095065 – 070
336.
Apple_AIIA00095090 – 091
337.
Apple_AIIA00095101
338.
Apple_AIIA00095111 – 113
339.
Apple_AIIA00095142
340.
Apple_AIIA00095169
341.
Apple_AIIA00095170
342.
Apple_AIIA00095170 – 172
343.
Apple_AIIA00095208
344.
Apple_AIIA00095213 – 217
345.
Apple_AIIA00095842 – 847
346.
Apple_AIIA00095849 – 852
347.
Apple_AIIA00096615 – 629
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
348.
Apple_AIIA00096929
349.
Apple_AIIA00096956
350.
Apple_AIIA00097184
351.
Apple_AIIA00097185
352.
Apple_AIIA00097188
353.
Apple_AIIA00097194
354.
Apple_AIIA00097211 – 217
355.
Apple_AIIA00097678
356.
Apple_AIIA00098220 – 222
357.
Apple_AIIA00098367 – 369
358.
Apple_AIIA00098373 – 375
359.
Apple_AIIA00098387 – 389
360.
Apple_AIIA00098416
361.
Apple_AIIA00098417
362.
Apple_AIIA00098435
363.
Apple_AIIA00098486 – 488
364.
Apple_AIIA00098491 – 493
365.
Apple_AIIA00098511
366.
Apple_AIIA00098533
367.
Apple_AIIA00098686
368.
Apple_AIIA00098798
369.
Apple_AIIA00098815 – 819
370.
Apple_AIIA00099064
371.
Apple_AIIA00099065 – 066
372.
Apple_AIIA00099084
373.
Apple_AIIA00099100 – 104
374.
Apple_AIIA00099339 – 341
375.
Apple_AIIA00099351
376.
Apple_AIIA00099373
377.
Apple_AIIA00099408 – 409
378.
Apple_AIIA00099770 – 771
379.
Apple_AIIA00099814 – 818
380.
Apple_AIIA00099928 – 929
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
381.
Apple_AIIA00100369 – Apple_AIIA00100371
382.
Apple_AIIA00100515
383.
Apple_AIIA00100796
384.
Apple_AIIA00100868
385.
Apple_AIIA00101375
386.
Apple_AIIA00101436 – 437
387.
Apple_AIIA00101459 – 464
388.
Apple_AIIA00101873 – 875
389.
Apple_AIIA00102397 – 405
390.
Apple_AIIA00102445
391.
Apple_AIIA00102487
392.
Apple_AIIA00103629 – 675
393.
Apple_AIIA00104005 – 016
394.
Apple_AIIA00104297 – 312
395.
Apple_AIIA00104338 – 353
396.
Apple_AIIA00104429 – 449
397.
Apple_AIIA00105655
398.
Apple_AIIA00105765 – 767
399.
Apple_AIIA00105851 – 861
400.
Apple_AIIA00105859
401.
Apple_AIIA00105860
402.
Apple_AIIA00105861
403.
Apple_AIIA00105896 – 898
404.
Apple_AIIA00105931
405.
Apple_AIIA00105959
406.
Apple_AIIA00106003
407.
Apple_AIIA00106024
408.
Apple_AIIA00106128 – 130
409.
Apple_AIIA00106519 – 871
410.
Apple_AIIA00113574
411.
Apple_AIIA00116092
412.
Apple_AIIA00116382
413.
Apple_AIIA00116422
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
414.
Apple_AIIA00122467
415.
Apple_AIIA00142141
416.
Apple_AIIA00155931 – 922
417.
Apple_AIIA00164871
418.
Apple_AIIA00165562
419.
Apple_AIIA00165759
420.
Apple_AIIA00174684 – 707
421.
Apple_AIIA00182671
422.
Apple_AIIA00185670
423.
Apple_AIIA00187610
424.
Apple_AIIA00187793
425.
Apple_AIIA00230354
426.
Apple_AIIA00231292
427.
Apple_AIIA00231314
428.
Apple_AIIA00231322
429.
Apple_AIIA00232273
430.
Apple_AIIA00233512
431.
Apple_AIIA00234674
432.
Apple_AIIA00256728
433.
Apple_AIIA00289709
434.
Apple_AIIA00290624
435.
Apple_AIIA00291320
436.
Apple_AIIA00291806
437.
Apple_AIIA00322844
438.
Apple_AIIA00322919 – 920
439.
Apple_AIIA00323137 – 138
440.
Apple_AIIA00323212 – 219
441.
Apple_AIIA00323333 – 334
442.
Apple_AIIA00323409 – 412
443.
Apple_AIIA00323457 – 458
444.
Apple_AIIA00325638 – 642
445.
Apple_AIIA00325883 – 886
446.
Apple_AIIA00325894
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
447.
Apple_AIIA00326226 – 229
448.
Apple_AIIA00327951
449.
Apple_AIIA00328028 – 029
450.
Apple_AIIA00328030 – 031
451.
Apple_AIIA00329373
452.
Apple_AIIA00329663 – 665
453.
Apple_AIIA00330298
454.
Apple_AIIA00797947
455.
Apple_AIIA00798033
456.
Apple_AIIA00798119
457.
Apple_AIIA00798211
458.
Apple_AIIA00799009 – 010
459.
Apple_AIIA00799335 – 336
460.
Apple_AIIA00799485 – 486
461.
Apple_AIIA00799672
462.
Apple_AIIA00799706
463.
Apple_AIIA00799707 – 709
464.
Apple_AIIA00799710 – 713
465.
Apple_AIIA00799768 – 769
466.
Apple_AIIA00802966
467.
Apple_AIIA00807080
468.
Apple_AIIA00807080-81
469.
Apple_AIIA00807085
470.
Apple_AIIA00809613 – 616
471.
Apple_AIIA00812423
472.
Apple_AIIA00815239 – 241
473.
Apple_AIIA00817408
474.
Apple_AIIA00818342
475.
Apple_AIIA00818701 – 703
476.
Apple_AIIA00819421 – 423
477.
Apple_AIIA00924813
478.
Apple_AIIA00924823
479.
Apple_AIIA00924825
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
480.
Apple_AIIA00925323
481.
Apple_AIIA00928375
482.
Apple_AIIA00928486
483.
Apple_AIIA00928878
484.
Apple_AIIA00940041 – 067
485.
Apple_AIIA00940807
486.
Apple_AIIA00942002
487.
Apple_AIIA00960350
488.
Apple_AIIA00974519
489.
Apple_AIIA00974838
490.
Apple_AIIA00974904
491.
Apple_AIIA00976211
492.
Apple_AIIA00979462 – Apple_AIIA00979469
493.
Apple_AIIA00980609 – Apple_AIIA00980620
494.
Apple_AIIA00983892
495.
Apple_AIIA01044664
496.
Apple_AIIA01044830
497.
Apple_AIIA01045450 – 455
498.
Apple_AIIA01045458 – 466
499.
Apple_AIIA01045475
500.
Apple_AIIA01055551
501.
Apple_AIIA01056176
502.
Apple_AIIA01071045
503.
Apple_AIIA01071222
504.
Apple_AIIA01072060
505.
Apple_AIIA01072123
506.
Apple_AIIA01072452
507.
Apple_AIIA01115498
508.
Apple_AIIA01203889
509.
Apple_AIIA01205270
510.
Apple_AIIA01209939
511.
Apple_AIIA01213489
512.
Apple_AIIA01230360
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
513.
Apple_AIIA01232836
514.
Apple_AIIA01245226
515.
Apple_AIIA01275937 – Apple_AIIA01275942
516.
Apple_AIIA01275943 – Apple_AIIA01275952
517.
Apple_AIIA01275953 – Apple_AIIA01275962
518.
Apple_AIIA01275963 – Apple_AIIA01275974
519.
Apple_AIIA01275975 – Apple_AIIA01275980
520.
Apple_AIIA01275981 – Apple_AIIA01275987
521.
Apple_AIIA01275988 – Apple_AIIA01275996
522.
Apple_AIIA01275997 – Apple_AIIA01276005
523.
Apple_AIIA01276006 – Apple_AIIA01276015
524.
Apple_AIIA01276016 – Apple_AIIA01276023
525.
Apple_AIIA01276024 – Apple_AIIA01276031
526.
Apple_AIIA01276032 – Apple_AIIA01276039
527.
Apple_AIIA01276040 – Apple_AIIA01276047
528.
Apple_AIIA01276048 – Apple_AIIA01276057
529.
Apple_AIIA01276058 – Apple_AIIA01276068
530.
Apple_AIIA01276069 – Apple_AIIA01276076
531.
Apple_AIIA01276077 – Apple_AIIA01276085
532.
Apple_AIIA01276086 – Apple_AIIA01276093
533.
Apple_AIIA01276094 – Apple_AIIA01276102
534.
Apple_AIIA01276103 – Apple_AIIA01276109
535.
Apple_AIIA01276110 – Apple_AIIA01276116
536.
Apple_AIIA01276117 – Apple_AIIA01276132
537.
Apple_AIIA01276133 – Apple_AIIA01276148
538.
Apple_AIIA01276149 – Apple_AIIA01276156
539.
Apple_AIIA01276157 – Apple_AIIA01276164
540.
Apple_AIIA01276165 – Apple_AIIA01276172
541.
Apple_AIIA01276173 – Apple_AIIA01276185
542.
Apple_AIIA01276186 – Apple_AIIA01276192
543.
Apple_AIIA01276193 – Apple_AIIA01276200
544.
Apple_AIIA01276201 – Apple_AIIA01276208
545.
Apple_AIIA01276209 – Apple_AIIA01276216
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
546.
Apple_AIIA01276217 – Apple_AIIA01276225
547.
Apple_AIIA01276230 – Apple_AIIA01276232
548.
Apple_AIIA01276248 – Apple_AIIA01276258
549.
Apple_AIIA01276259 – Apple_AIIA01276271
550.
Apple_AIIA01276272 – Apple_AIIA01276282
551.
Apple_AIIA01276284 – Apple_AIIA01276293
552.
Apple_AIIA01276296 – Apple_AIIA01276305
553.
Apple_AIIA01276307 – Apple_AIIA01276317
554.
Apple_AIIA01276319 – Apple_AIIA01276328
555.
Apple_AIIA01276330 – Apple_AIIA01276339
556.
Apple_AIIA01276342 – Apple_AIIA01276350
557.
Apple_AIIA01276353 – Apple_AIIA01276361
558.
Apple_AIIA01276363 – Apple_AIIA01276371
559.
Apple_AIIA01276373 – Apple_AIIA01276380
560.
Apple_AIIA01276383 – Apple_AIIA01276390
561.
Apple_AIIA01276393 – Apple_AIIA01276400
562.
Apple_AIIA01276402 – Apple_AIIA01276409
563.
Apple_AIIA01276411 – Apple_AIIA01276418
564.
Apple_AIIA01276421 – Apple_AIIA01276428
565.
Apple_AIIA01276429
566.
Apple_AIIA01276430 – Apple_AIIA01276437
567.
Apple_AIIA01276439 – Apple_AIIA01276445
568.
Apple_AIIA01276447 – Apple_AIIA01276454
569.
Apple_AIIA01276456 – Apple_AIIA01276463
570.
Apple_AIIA01276465 – Apple_AIIA01276471
571.
Apple_AIIA01276474 – Apple_AIIA01276480
572.
Apple_AIIA01276483 – Apple_AIIA01276489
573.
Apple_AIIA01276493 – Apple_AIIA01276500
574.
Apple_AIIA01276504 – Apple_AIIA01276511
575.
Apple_AIIA01276512 – Apple_AIIA01276520
576.
Apple_AIIA01276525 – Apple_AIIA01276532
577.
Apple_AIIA01276538 – Apple_AIIA01276545
578.
Apple_AIIA01276551 – Apple_AIIA01276558
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
579.
Apple_AIIA01276560 – Apple_AIIA01276568
580.
Apple_AIIA01276570 – Apple_AIIA01276582
581.
Apple_AIIA01276576 - Apple_AIIA01384647
582.
Apple_AIIA01276584 – Apple_AIIA01276595
583.
Apple_AIIA01276596 – Apple_AIIA01276609
584.
Apple_AIIA01276610 – Apple_AIIA01276623
585.
Apple_AIIA01276624 – Apple_AIIA01276637
586.
Apple_AIIA01276638 – Apple_AIIA01276652
587.
Apple_AIIA01276653 – Apple_AIIA01276668
588.
Apple_AIIA01276670 – Apple_AIIA01276682
589.
Apple_AIIA01276684 – Apple_AIIA01276696
590.
Apple_AIIA01276699 – Apple_AIIA01276711
591.
Apple_AIIA01276714 – Apple_AIIA01276729
592.
Apple_AIIA01276735 – Apple_AIIA01276750
593.
Apple_AIIA01276756 – Apple_AIIA01276771
594.
Apple_AIIA01276774 – Apple_AIIA01276789
595.
Apple_AIIA01276790 – Apple_AIIA01276798
596.
Apple_AIIA01276799 – Apple_AIIA01276807
597.
Apple_AIIA01276808 – Apple_AIIA01276817
598.
Apple_AIIA01276818
599.
Apple_AIIA01276821 – Apple_AIIA01276830
600.
Apple_AIIA01276831 – Apple_AIIA01276841
601.
Apple_AIIA01276842 – Apple_AIIA01276852
602.
Apple_AIIA01276853 – Apple_AIIA01276863
603.
Apple_AIIA01276864 – Apple_AIIA01276875
604.
Apple_AIIA01276876 – Apple_AIIA01276884
605.
Apple_AIIA01276885 – Apple_AIIA01276894
606.
Apple_AIIA01276895 – Apple_AIIA01276911
607.
Apple_AIIA01276915 – Apple_AIIA01276923
608.
Apple_AIIA01276924 – Apple_AIIA01276933
609.
Apple_AIIA01276934 – Apple_AIIA01276941
610.
Apple_AIIA01276942 – Apple_AIIA01276950
611.
Apple_AIIA01276951 – Apple_AIIA01276959
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
612.
Apple_AIIA01276960 – Apple_AIIA01276968
613.
Apple_AIIA01276969 – Apple_AIIA01276976
614.
Apple_AIIA01276977 – Apple_AIIA01276984
615.
Apple_AIIA01276985 – Apple_AIIA01276992
616.
Apple_AIIA01276993 – Apple_AIIA01277001
617.
Apple_AIIA01277002 – Apple_AIIA01277011
618.
Apple_AIIA01277012 – Apple_AIIA01277036
619.
Apple_AIIA01277037 – Apple_AIIA01277046
620.
Apple_AIIA01277047 – Apple_AIIA01277061
621.
Apple_AIIA01277062 – Apple_AIIA01277077
622.
Apple_AIIA01277078 – Apple_AIIA01277096
623.
Apple_AIIA01277099 – Apple_AIIA01277110
624.
Apple_AIIA01277113 – Apple_AIIA01277124
625.
Apple_AIIA01277127 – Apple_AIIA01277138
626.
Apple_AIIA01277142 – Apple_AIIA01277153
627.
Apple_AIIA01277154 – Apple_AIIA01277161
628.
Apple_AIIA01277162 – Apple_AIIA01277170
629.
Apple_AIIA01277171 – Apple_AIIA01277182
630.
Apple_AIIA01277183 – Apple_AIIA01277194
631.
Apple_AIIA01277198 – Apple_AIIA01277209
632.
Apple_AIIA01277210 – Apple_AIIA01277219
633.
Apple_AIIA01277220 – Apple_AIIA01277229
634.
Apple_AIIA01277230 – Apple_AIIA01277242
635.
Apple_AIIA01277243 – Apple_AIIA01277260
636.
Apple_AIIA01277261 – Apple_AIIA01277275
637.
Apple_AIIA01277277 – Apple_AIIA01277292
638.
Apple_AIIA01277294 – Apple_AIIA01277309
639.
Apple_AIIA01277312 – Apple_AIIA01277327
640.
Apple_AIIA01277328
641.
Apple_AIIA01277344 – Apple_AIIA01277352
642.
Apple_AIIA01277355 – Apple_AIIA01277362
643.
Apple_AIIA01277363 – Apple_AIIA01277371
644.
Apple_AIIA01277376 – Apple_AIIA01277383
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
645.
Apple_AIIA01277386 – Apple_AIIA01277401
646.
Apple_AIIA01277405 – Apple_AIIA01277413
647.
Apple_AIIA01277414 – Apple_AIIA01277421
648.
Apple_AIIA01277422 – Apple_AIIA01277429
649.
Apple_AIIA01277430 – Apple_AIIA01277437
650.
Apple_AIIA01277438 – Apple_AIIA01277448
651.
Apple_AIIA01277465 – Apple_AIIA01277475
652.
Apple_AIIA01277476 – Apple_AIIA01277486
653.
Apple_AIIA01277541 – Apple_AIIA01277550
654.
Apple_AIIA01277553 – Apple_AIIA01277568
655.
Apple_AIIA01277569 – Apple_AIIA01277577
656.
Apple_AIIA01277578 – Apple_AIIA01277587
657.
Apple_AIIA01277588 – Apple_AIIA01277595
658.
Apple_AIIA01277596 – Apple_AIIA01277603
659.
Apple_AIIA01277604 – Apple_AIIA01277611
660.
Apple_AIIA01277612 – Apple_AIIA01277621
661.
Apple_AIIA01277622 – Apple_AIIA01277631
662.
Apple_AIIA01277632 – Apple_AIIA01277644
663.
Apple_AIIA01277645 – Apple_AIIA01277657
664.
Apple_AIIA01277658 – Apple_AIIA01277670
665.
Apple_AIIA01277671 – Apple_AIIA01277684
666.
Apple_AIIA01277685 – Apple_AIIA01277695
667.
Apple_AIIA01277750 – Apple_AIIA01277759
668.
Apple_AIIA01277760 – Apple_AIIA01277774
669.
Apple_AIIA01277775 – Apple_AIIA01277790
670.
Apple_AIIA01277791 – Apple_AIIA01277809
671.
Apple_AIIA01277812 – Apple_AIIA01277823
672.
Apple_AIIA01277826 – Apple_AIIA01277837
673.
Apple_AIIA01277840 – Apple_AIIA01277851
674.
Apple_AIIA01277855 – Apple_AIIA01277866
675.
Apple_AIIA01277870 – Apple_AIIA01277881
676.
Apple_AIIA01277882 – Apple_AIIA01277891
677.
Apple_AIIA01277892 – Apple_AIIA01277901
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
678.
Apple_AIIA01277902 – Apple_AIIA01277914
679.
Apple_AIIA01277915 – Apple_AIIA01277929
680.
Apple_AIIA01277931 – Apple_AIIA01277946
681.
Apple_AIIA01277949 – Apple_AIIA01277964
682.
Apple_AIIA01277967 – Apple_AIIA01277982
683.
Apple_AIIA01277983 – Apple_AIIA01277991
684.
Apple_AIIA01277992 – Apple_AIIA01278001
685.
Apple_AIIA01278002 – Apple_AIIA01278011
686.
Apple_AIIA01278014 – Apple_AIIA01278021
687.
Apple_AIIA01278032 – Apple_AIIA01278044
688.
Apple_AIIA01278046 – Apple_AIIA01278057
689.
Apple_AIIA01278059 – Apple_AIIA01278070
690.
Apple_AIIA01278073 – Apple_AIIA01278084
691.
Apple_AIIA01278087 – Apple_AIIA01278102
692.
Apple_AIIA01278108 – Apple_AIIA01278123
693.
Apple_AIIA01278129 – Apple_AIIA01278144
694.
Apple_AIIA01278147 – Apple_AIIA01278162
695.
Apple_AIIA01278163 – Apple_AIIA01278172
696.
Apple_AIIA01278173 – Apple_AIIA01278182
697.
Apple_AIIA01278183 – Apple_AIIA01278195
698.
Apple_AIIA01278196 – Apple_AIIA01278208
699.
Apple_AIIA01278209 – Apple_AIIA01278221
700.
Apple_AIIA01278222 – Apple_AIIA01278235
701.
Apple_AIIA01278236 – Apple_AIIA01278246
702.
Apple_AIIA01278301 – Apple_AIIA01278310
703.
Apple_AIIA01278311 – Apple_AIIA01278320
704.
Apple_AIIA01278321 – Apple_AIIA01278330
705.
Apple_AIIA01278331 – Apple_AIIA01278340
706.
Apple_AIIA01278345 – Apple_AIIA01278392
707.
Apple_AIIA01278393 – Apple_AIIA01278419
708.
Apple_AIIA01278420 – Apple_AIIA01278430
709.
Apple_AIIA01278485 – Apple_AIIA01278494
710.
Apple_AIIA01278496 – Apple_AIIA01278505
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
711.
Apple_AIIA01278506 – Apple_AIIA01278518
712.
Apple_AIIA01278519 – Apple_AIIA01278531
713.
Apple_AIIA01278532 – Apple_AIIA01278545
714.
Apple_AIIA01278600 – Apple_AIIA01278609
715.
Apple_AIIA01278610 – Apple_AIIA01278619
716.
Apple_AIIA01278623 – Apple_AIIA01278634
717.
Apple_AIIA01278638 – Apple_AIIA01278649
718.
Apple_AIIA01278652 – Apple_AIIA01278663
719.
Apple_AIIA01278671 – 674
720.
Apple_AIIA01278680 – 683
721.
Apple_AIIA01278697
722.
Apple_AIIA01278810 – 811
723.
Apple_AIIA01288063 – 064
724.
Apple_AIIA01288139 – 143
725.
Apple_AIIA01289173
726.
Apple_AIIA01289403
727.
Apple_AIIA01290132 – 135
728.
Apple_AIIA01290799
729.
Apple_AIIA01290800 – 801
730.
Apple_AIIA01333622
731.
Apple_AIIA01333622-28
732.
Apple_AIIA01335531 – 532
733.
Apple_AIIA01341443
734.
Apple_AIIA01344648 – 649
735.
Apple_AIIA01346841 – 843
736.
Apple_AIIA01373827
737.
Apple_AIIA01373945
738.
Apple_AIIA01373953
739.
Apple_AIIA01384143 – Apple_AIIA01384160
740.
Apple_AIIA01384161 – Apple_AIIA01384171
741.
Apple_AIIA01384172 – Apple_AIIA01384182
742.
Apple_AIIA01384183 – Apple_AIIA01384190
743.
Apple_AIIA01384191 – Apple_AIIA01384198
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
744.
Apple_AIIA01384199 – Apple_AIIA01384205
745.
Apple_AIIA01384206 – Apple_AIIA01384258
746.
Apple_AIIA01384259 – Apple_AIIA01384265
747.
Apple_AIIA01384266 – Apple_AIIA01384275
748.
Apple_AIIA01384276 – Apple_AIIA01384286
749.
Apple_AIIA01384287 – Apple_AIIA01384294
750.
Apple_AIIA01384295 – Apple_AIIA01384302
751.
Apple_AIIA01384303 – Apple_AIIA01384364
752.
Apple_AIIA01384365 – Apple_AIIA01384375
753.
Apple_AIIA01384376 – Apple_AIIA01384400
754.
Apple_AIIA01384401 – Apple_AIIA01384426
755.
Apple_AIIA01384427
756.
Apple_AIIA01384442
757.
Apple_AIIA01384455
758.
Apple_AIIA01384456
759.
Apple_AIIA01384469 – Apple_AIIA01384470
760.
Apple_AIIA01384483 – Apple_AIIA01384491
761.
Apple_AIIA01384496 – Apple_AIIA01384506
762.
Apple_AIIA01384507 – Apple_AIIA01384517
763.
Apple_AIIA01384518 – Apple_AIIA01384528
764.
Apple_AIIA01384529 – Apple_AIIA01384539
765.
Apple_AIIA01384540 – Apple_AIIA01384567
766.
Apple_AIIA01384569 – Apple_AIIA01384596
767.
Apple_AIIA01384597 – Apple_AIIA01384622
768.
Apple_AIIA01384623 – Apple_AIIA01384630
769.
Apple_AIIA01384631 – Apple_AIIA01384640
770.
Apple_AIIA01384641 – Apple_AIIA01384650
771.
Apple_AIIA01384651 – Apple_AIIA01384659
772.
Apple_AIIA01384664 – Apple_AIIA01384687
773.
Apple_AIIA01384965
774.
Apple_AIIA01384966 – 967
775.
Apple_AIIA01384975 – 976
776.
Apple_AIIA01384977 – 978
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
777.
Apple_AIIA01384979
778.
Apple_AIIA01385106
779.
Apple_AIIA01385354 – 356
780.
Apple_AIIA01385373 – 374
781.
Apple_AIIA01385433 – 436
782.
Apple_AIIA01385470 – 471
783.
Apple_AIIA01385473
784.
Apple_AIIA01385473 – 474
785.
Apple_AIIA01385473-74
786.
Apple_AIIA01385548
787.
Apple_AIIA01385569 – 570
788.
Apple_AIIA01385578 – 581
789.
Apple_AIIA01385602 – 603
790.
Apple_AIIA01385617 – 621
791.
Apple_AIIA01385670 – 671
792.
Apple_AIIA01385684 – 685
793.
Apple_SOM00000001 – 002
794.
Apple_SOM00000539 – 547
795.
Apple_SOM00000596
796.
Apple_SOM00007134 – 184
797.
Apple_SOM00007206 – 214
798.
IT Hardware Apple Computer , Deutsch Bank (September 21, 2005)
799.
Apple Computer Inc. – Apple Rings the Register, Credit Suisse (September 8, 2005)
800.
Apple Computer: They Did It Again – Innov New Products, Lehman Brothers (September 8, 2005)
801.
Equity Research Note: September 14, 2005, Apple Computer, Inc., Needham
802.
Apple Computer Inc. Investment Case, Morgan Stanley (September 8, 2005)
803.
Apple Computer: Cingular to Offer iTunes Motorola Phone, Lehman Brothers (August 31, 2005)
804.
Apple Computer: More iPod Cycling up Ahead, Cowen & Co. (September 6, 2005)
805.
AAPL: Monster iPods and Guidance, Plus We Still Think Core Macs Accelerate in FY05; Raising
Estimates and Target to $53 – BUY , Fulcrum Global Partners (October 14, 2004)
806.
Apple Computer Inc. Not the First Bite, But Plenty of Juice in the Apple, JPMorgan Securities Inc.
(October 14, 2004)
807.
Q4 Preview: iPod Invasion, Credit Suisse First Boston (2004)
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
808.
AAPL: Initiating Coverage with a Neutral Rating, Credit Suisse First Boston (August 18, 2004)
809.
Apple Computer: Nano Estimates and Raising Px Tx, Lehman Brothers (September 22, 2005)
810.
AAPL Morning Meeting Notes, Fulcrum Global Partners LLC (October 14, 2004)
811.
Apple Computer Inc., Credit Suisse First Boston (August 17, 2004)
812.
Apple Computer, Inc., Needham (June 21, 2004)
813.
Apple Computer Inc. iPod for Windows or Bust, Needham (July 17, 2002)
814.
Apple Inc. Initiating Coverage with An Outperform Rating and 12-month Price Target of $200, Credit
Suisse (August 7, 2008)
815.
Apple Inc. Upgrading rating to Overweight based on increased confidence in long- term growth outlook;
Raising estimates and price target, Thomas Weisel Partners (April 7, 2008)
816.
Apple Inc. Macworld Recap: An Uneventful Finale, Credit Suisse (January 7, 2009)
817.
Apple Inc. The DRM-Free Movement Begins – ALERT, JPMorgan (May 30, 2007)
818.
Apple Inc. iPod, iMac, iPhone, I Hold at This Price, Scotia Capital (March 30, 2009)
819.
Apple Inc. A Strategic Nugget in the Accounting Change, Credit Suisse (November 3, 2009)
820.
AAPL: Macworld Product Introductions as Expected, Smith Barney Citigroup (January 7, 2004)
821.
AAPL: iPod Mini Revolution, Credit Suisse First Boston (January 7, 2004)
822.
Apple Computer Inc. iPod Steals Show at Macworld & Fiscal 1Q04 Preview, UBS Investment Research
(January 7, 2004)
823.
Apple Computer Inc. MacWorld: Mac Turns 20, iPod Goes “mini” & More, JPMorgan (January 7, 2004)
824.
Apple Computer Inc. Macworld Expo San Fran Preview, Morgan Stanley (January 5, 2004)
825.
AAPL: Apple iPods Could Drive Upside, Credit Suisse First Boston (December 19, 2003)
826.
AAPL: Hewlett-Packard to Resell iPod, Distribute iTunes, Smith Barney Citigroup (January 9, 2004)
827.
Apple Can Hear the Music With iTunes and iPod, UBS (December 2, 2003)
828.
Apple Computer, Inc. Going for Broke: Apple’s Lock in, Lock Out Music Strategy, Needham (November
10, 2003)
829.
AAPL: Analyst Meeting Sets Positive Tone, Credit Suisse First Boston (November 6, 2003)
830.
Apple Computer, Inc. Investment Recommendation, Needham (October 23, 2003)
831.
AAPL: Apple Unveils iTunes for Windows – Reiterating 3H Rating, Smith Barney Citigroup (October 16,
2003)
832.
AAPL: iTunes Launched in Europe, Should Fuel iPod Momentum, UBS Investment Research (June 15,
2004)
833.
Apple Computer Huge iPod Quarter, As Advertised, SG Cowen (January 15, 2004)
834.
AAPL: Apple Beats Q1 Despite Weaker PowerMac, Suisse First Boston (January 15, 2004)
835.
AAPL: A Closer Look at the iPod Opportunity, UBS Investment Research (June 3, 2004)
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
836.
Apple Computer Inc. iPods in Focus in Meeting with iPod Guru, UBS Investment Research (February 6,
2004)
837.
First Read: Apple Computer iPod Mini Looks like a Hit, UBS Investment Research (March 25, 2004)
838.
“The Mix Get Richer: New iPod touch & iPhone Capacities,” Credit Suisse, February 5, 2008.
839.
Robert Sample, Stephanie Sun and Thompson Wu, “Happy Holidays!” Credit Suisse, October 3, 2007.
840.
Music and the Internet: Celestial Jukebox , Lehman Brothers (Europe), November 9, 2000, pp. 21-24, 28
841.
Roger G. Noll, “Napster’s Copyright Misuse Defense and the Future of Internet Distribution of Music," In
re Napster, Inc., Copyright Litigation .
842.
"3rd UPDATE: Apple Unveils TV Adapter, New Movie Service," Dow Jones Factiva via LexisNexis,
September 13, 2006.
843.
"A Report on iTunes Bugs," NY Times , September 18, 2006,
http://pogue.blogs.nytimes.com/2006/09/18/18pogues-posts-2/
844.
“AAPL: The Reason for the iPhone’s Reported Woes Is Closer than You Think: It’s the iPod touch,”
Needham & Company, January 28, 2008
845.
“AAPL: Weak Macro Environment; Downgrading on Poor Risk/Reward,” Morgan Keegan Company,
April 7, 2008.
846.
Ahrens, Frank, "Apple Nears Deal for Feature Films on IPod; Plan Has Major Studios Divided,"
Washington Post via LexisNexis, September 7, 2006.
847.
Ahrens, Frank, "Movie Downloads, Coming Soon to An IPod Near You," Washington Post via LexisNexis,
September 12, 2006.
848.
Alan S. Manne, “A Linear Programming Model of the U. S. Petroleum Refining Industry,” Econometrica
Vol. 26, No. 1 (January 1958), pp. 67-106.
849.
Alexander E. Reppel, Isabelle Szmigin, and Thorsten Gruber, “The iPod Phenomenon: Identifying a
Market Leader’s Secrets through Qualitative Market Research,” Journal of Product and Brand
Management Vol. 15, No. 4 (2006), pp. 239-49.
850.
"Analysts: iPod sales expected to decline 7.2% year over year," CNN Money , July 15, 2011,
http://tech.fortune.cnn.com/2011/07/15/analysts-ipods-sales-expected-to-decline-7-2-year-over-year/
851.
“Apple Adds Video Camera to iPod Nano, Cuts Prices for iPod Touch,” Huffington Post, September 9,
2009, at http://www.huffingtonpost.com/2009/09/09/apple-cuts-ipod-price-tag_n_280582.html.
852.
“Apple Slashes iPod Prices,” Techtree.com, September 10, 2009, available at
http://www.techtree.com/India/News/Apple_Slashes_iPod_Prices_Adds_More_Storage/551-106219-8933.html.
853.
“Apple’s Negative Guidance Tone at the FQ1 Call Means a Lower Valuation of Multiple: Hold,”Kintisheff
Research, January 23, 2008.
854.
"Apple looks to squash bugs with iTunes 7.0," AppleInsider , September 27, 2006,
http://appleinsider.com/articles/06/09/27/apple_looks_to_squash_bugs_with_itunes_701.
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
855.
"Apple Releases iTunes 7.0.1 to Address Numerous Issues," DailyTech , September 27, 2006,
http://www.dailytech.com/Apple+Releases+iTunes+701+to+Address+Numerous+Issues/article4362.htm
856.
"Apple releases iTunes 7.0.1 update," Gigaom , September 27, 2006, http://gigaom.com/2006/09/27/applereleases-itunes-701-update/
857.
"Apple stomps competitors in flash-based MP3 player market," Mac Daily News , March 28, 2013,
http://macdailynews.com/2005/09/02/apple_stomps_competitors_in_flash_based_mp3_player_market/
858.
"Apple Unveils the New iPod Shuffle," PR Newswire Association via LexisNexis, September 12, 2006.
859.
"Apple's Movie Launch Leaves Some Wanting More," Dow Jones Factiva via LexisNexis, September 14,
2006.
860.
"Apple's streaming music service could debut in 2013 -- analyst", Cnet News, November 30, 2012,
http://news.cnet.com/8301-13579_3-57556472-37/apples-streaming-music-service-could-debut-in-2013analyst/
861.
Apple 2012 Annual Report, Form 10-K
862.
Apple Inc. financial statements.
863.
Ariel Pakes, “A Reconsideration of Hedonic Price Indices with an Application to PCs,” American
Economic Review Vol. 93, No. 5 (December 2003), pp. 1578-96.
864.
"As Music Streaming Grows, Royalties Slow to a Trickle," NY Times, January 28, 2013,
http://www.nytimes.com/2013/01/29/business/media/streaming-shakes-up-music-industrys-model-forroyalties.html?_r=0
865.
Barker, Garry, "Bugs to bear; MACMAN," The Age Company Limited via LexisNexis, September 28,
2006.
866.
Barker, Garry, "That's Life; MACMAN," The Age Company Limited via LexisNexis, September 21, 2006.
867.
Bond, Paul, "Disney chief 'bullish' on iTunes movie delivery," Reuters via LexisNexis, September 20,
2006.
868.
"Business Brief -- Walt Disney Co.: Deal With Apple Sparks Sale Of 125,000 Movies in Week," WSJ via
LexisNexis, September 20, 2006.
869.
C. E. Buddington and T. E. Baker, “A History of Mathematical Programming in the Petroleum Industry,”
Interfaces Vol. 20, No. 4 (July-August 1990), pp. 117-27.
870.
Carl Shapiro and Hal R. Varian, Information Rules: A Strategic Guide to the Network Economy, Harvard
Business School Press, 1999.
871.
Charles Haddad, “Sweet Music from iPod,” Business Week, October 31, 2001, available at
http://www.businessweek.com/bwdaily/dnflash/oct2001/nf20011031_4266.htm.
872.
Charlie Wolf, “AAPL: Its MacWorld and We’re Just Living in It, Upgrading Apple from Buy to Strong
Buy,” Needham and Company, January 23, 2008.
873.
Chmielewski, Dawn C., "The Online Box Office Is Growing," LA Times via LexisNexis, September 6,
2006.
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
874.
“Chris Brandrick, “Apple Cuts iPod Prices Ahead of Today’s Event,” PC World, September 9, 2009, at
http://www.pcworld.com/article/171636/apple_cuts_ipod_prices_ahead_of_todays_event.html.
875.
Crothers, Brooke, "How many devices can a smartphone, tablet replace?," CNET News, July 10, 2011,
http://news.cnet.com/8301-13924_3-20078244-64/how-many-devices-can-a-smartphone-tablet-replace/
876.
Darwin Klingman, Nancy Phillips, David Steiger, Ross Wirth, Rema Padman, and R. Krishnan, “An
Optimization Based Integrated Short-Term Refined Petroleum Planning System,” Management Science
Vol. 33, No. 7 (July 1987), pp. 813-30.
877.
David A. Freedman and Stephen C. Peters, “Bootstrapping an Econometric Model: Some Empirical
Results,” Journal of Business and Economic Statistics Vol. 2, No. 2 (April 1984), pp. 150-58.
878.
David Smith, “Why the iPod is Losing Its Cool,” The Observer, September 10, 2006, available at
http://www.guardian.co.uk/technology/2006/sep/10/news.theobserver1.
879.
Delahunty, James, "iPod market share at 73.8 percent, 225 million iPods sold, more games for Touch than
PSP & NDS: Apple," News by Afterdawn, September 9, 2009,
http://www.afterdawn.com/news/article.cfm/2009/09/09/ipod_market_share_at_73_8_percent_225_million
_ipods_sold_more_games_for_touch_than_psp_nds_apple
880.
"Digital music maze; Competing file formats leave low-techies longing for days of CD," Grand Rapids
Press (Michigan) via LexisNexis , March 21, 2007.
881.
"Digital music maze; Competing file formats leave low-techies longing for days of CD," Grand Rapids
Press via LexisNexis, November 6, 2006.
882.
"Disappearing songs and other iTunes 7 woes," CNet, September 26, 2006, http://reviews.cnet.com/830110921_7-6645665-4.html.
883.
Donald O. Parsons and Edward J. Ray, “The United States Steel Consolidation: The Creation of Market
Control,” Journal of Law and Economics Vol. 18, No. 1 (April 1975), pp. 181-219.
884.
Douglas Staiger and James H. Stock, “Instrumental Variable Regressions with Weak Instruments,”
Econometrica Vol. 65, No. 3 (May 1997), pp. 557-86.
885.
Emi Nakamura, “Pass-through in Retail and Wholesale,” American Economic Review Papers and
Proceedings Vol. 98, No. 2 (May 2008), pp. 430-37.
886.
Eric M. Olson, Andrew J. Czaplewski, and Stanley F. Slater, “Stay Cool,” Marketing Management Vol.
14, No. 5 (September/October 2005), pp. 14-17.
887.
Ernst R. Berndt, “The Measurement of Quality Change,” The Practice of Econometrics: Classic and
Contemporary , Ch. 4, Addison-Wesley, 1991.
888.
Fairplay, Wikipedia , http://en.wikipedia.org/wiki/FairPlay.
889.
Farivar, Cyrus, "iTunes 7 turning out to have major glitches," Engadget, September 14, 2006,
http://www.engadget.com/2006/09/14/itunes-7-turning-out-to-have-major-glitches/.
890.
George J. Stigler and Robert A. Sherwin, “The Extent of the Market,” Journal of Law and Economics Vol.
28, No. 3 (October 1985), pp. 555-85.
891.
George Symeonides, “Comparing Cournot and Bertrand Equilibrium in Differentiated Duopoly with
Product R&D,” International Journal of Industrial Organization Vol. 21, No. 1 (January 2003).
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
892.
Global Cool Hunt 2003/04 , Hill and Knowlton, undated, accessed October 9, 2009, at
http://www.signsofthetime.nl/image/globalcoolhuntfinal.pdf.
893.
Gollner, Phillipp, "Apple may launch movie downloads, analysts say," Reuters via LexisNexis, September
5, 2006.
894.
Gonsalves, Antone,"Apple Addresses iTunes 7 Problems," InformationWeek , September 21, 2006,
http://www.informationweek.com/apple-addresses-itunes-7-problems/193004674
895.
Graeme Reynolds and Chris Walters, “The Use of Customer Surveys for Market Definition and
the Competitive Assessment of Horizontal Mergers,” Journal of Competition Law and Economics Vol. 4,
No. 2 (June 2008), pp. 411-31.
896.
"Growth of Paid Downloads vs. Streaming, 2012 vs. 2011…," Digital Music News , January 3, 2013,
http://www.digitalmusicnews.com/permalink/2013/20130103cannibalism2012
897.
Hall, James, "Smartphones wipe out sales of MP3 players," December 26, 2012,
http://www.telegraph.co.uk/technology/news/9741910/Smartphones-wipe-out-sales-of-MP3-players.html
898.
Henry Norr, “Apple iPod Has Its Charms,” San Francisco Chronicle, October 29, 2001, available at
http://www.sfgate.com/cgibin/article.cgi?file=/chronicle/archive/2001/10/29/BU215879.DTL&type=business.
899.
Humberto Barretto and Frank M. Howland, Introductory Econometrics Using Monte Carlo Simulation
with Microsoft Excel , Cambridge University Press, 2006, p. 492.
900.
“IFPI Digital Music Report 2013: Engine of a digital world,” International Federation of the Phonographic
Industry (2013), available at www.ifpi.org.
901.
"Importing/converting Real Player files into iTunes," Apple Support Communities,
https://discussions.apple.com/thread/710076?start=0&tstart=0
902.
"iPod getting away from the basics," Chicago Tribune via LexisNexis , September 25, 2006.
903.
"iPod still has 70% of MP3 player market," MacTech , http://www.mactech.com/2012/07/24/ipod-still-has70-mp3-player-market
904.
iPod Sales Chart, Wikipedia, http://en.wikipedia.org/wiki/File:Ipod_sales_per_quarter.svg#file
905.
“iTunes Player Hits a High Note, Passes RealPlayer – U. S. Broadband Penetration Increases to 86.79%
among Active Internet Users – January 2008 Bandwidth Report,” WebSiteOptimization.com, January
2008.
906.
"iTunes 7 - Breaks Airtunes !," Apple Support Communities, October 2, 2011,
https://discussions.apple.com/thread/637159?start=30&tstart=0
907.
"iTunes 7 bugs: doesn't Apple test software before releasing it?," The Techsploder, September 14, 2006,
http://juha.saarinen.org/1308.
908.
"iTunes 7 DRM Cracked," SlashGear, http://www.slashgear.com/itunes-7-drm-cracked-131624/
909.
"iTunes 7 might delete your Files! Silently!," Maba , September 16, 2006,
http://maba.wordpress.com/2006/09/19/itunes-7-might-delete-your-files-silently/
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
910.
"iTunes 7 problem," MacRumors.com , September 20, 2006,
http://forums.macrumors.com/archive/index.php/t-235988.html
911.
"iTunes 7 Problems Arise," SoftPedia, September 15, 2006, http://news.softpedia.com/news/iTunes-7Problems-Arise-35666.shtml
912.
"iTunes 7.0 & iPod: Frozen/stuck playback at end of file," Apple Support Communities, September 21,
2006,
https://discussions.apple.com/message/3176969?messageID=3176969#3176969?messageID=3176969
913.
"iTunes 7.0 (#4): Persistent crashes: Delete .plist files, turn off plug-ins; fix for broken podcasts; selecting
libraries; more," Cnet News, September 16, 2006, http://reviews.cnet.com/8301-13727_7-10327208263.html%202%20of
914.
"iTunes 7.0 broke my ProTools," Avid Pro Audio Community , September 27, 2006,
http://duc.avid.com/archive/index.php/t-178082.html
915.
"iTunes 7.0 crashes at movies page," iLounge , September 2006, http://forums.ilounge.com/itunes-relatedmac-pc-applications/173675-itunes-7-0-crashes-movies-page.html
916.
"iTunes 7.0 crashes on exit," Apple Support Communities, October 4, 2006,
https://discussions.apple.com/thread/648839?start=0&tstart=0
917.
"iTunes 7.0 deletes my files," YouTube , http://www.youtube.com/watch?v=-zfEMmgcuNc
918.
"iTunes 7.0 distorted songs bug," Apple Support Communities, October 11, 2006,
https://discussions.apple.com/message/3109621?messageID=3109621
919.
"iTunes 7.0 has killed my iPod," Apple Support Communities, January 19, 2007,
https://discussions.apple.com/message/3320606?messageID=3320606
920.
"iTunes 7.0 persistent crashes," Apple Support Communities, September 28, 2006,
https://discussions.apple.com/thread/648807?start=0&tstart=0
921.
"iTunes 7.0 Problem - DON'T UPGRADE!," Apple Support Communities, September 23, 2006,
https://discussions.apple.com/message/3192547?messageID=3192547
922.
"iTunes 7.0 PROBLEM with song recognition--HELP," Apple Support Communities, September 20, 2006,
https://discussions.apple.com/message/3133578?messageID=3133578
923.
"iTunes 7.0 Special Report: iPod synchronization issues: Problems transferring music, crashes, making
sure software is up to date," Cnet News, September 26, 2006, http://reviews.cnet.com/8301-13727_710327372-263.html
924.
"iTunes 7.0.1 = crashes!," Apple Support Communities, October 17, 2006,
https://discussions.apple.com/thread/667815?start=0&tstart=0
925.
"iTunes 7.0.1 Repeatedly Crashes," Apple Support Communities, October 3, 2006,
https://discussions.apple.com/thread/668140?start=0&tstart=0
926.
"itunes 7.0.1 still broken," Apple Support Communities, October 3, 2006,
https://discussions.apple.com/thread/666709?start=0&tstart=0
927.
"iTunes Dominates Download Market & Streaming Audio Grows," CE Pro, October 10, 2012,
http://www.cepro.com/article/itunes_dominates_download_market_streaming_audio_grows/
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
928.
"iTunes Compatibility Kit Allows iPod Users To Switch To Prism DuroSport", March 31, 2006, available
at http://www.prweb.com/releases/2006/03/prweb365364.htm.
929.
J. D. Sargan, “The Estimation of Economic Relationships Using Instrumental Variables” Econometrica
Vol. 28, No. 3 (July 1958), pp. 393-415.
930.
James H. Gary and Glenn E. Handwerk, Petroleum Refining: Technology and Economics , Marcel Dekker,
2001, p. 5.
931.
James M. Griffen, :The Econometrics of Joint Production: Another Approach,” Review of Economics and
Statistics Vol. 59, No. 4 (November 1977), pp. 389-97.
932.
Janusz A. Ordover, Garth Saloner, and Steven C. Salop, “Equilibrium Vertical Foreclosure,” American
Economic Review Vol. 80, No. 1 (March 1990), pp. 127-42.
933.
Jason Dedrick, Kenneth L. Kraemer and Greg Linden, “Who Profits from Innovation in Global Value
Chains?: A Study of the iPod and Notebook PCs,” Industrial and Corporate Change, June 22, 2009,
available at http://icc.oxfordjournals.org/cgi/reprint/dtp032r1.
934.
Jeffrey M. Wooldridge, “Cluster Sample Analysis in Applied Econometrics” American Economic Review
Vol. 93, No. 2 (May 2003), pp. 133-38.
935.
Jeffrey M. Wooldridge, Econometric Analysis of Cross Section Data , MIT Press (2002), especially
Chapter 11, Section 11.5, pp. 328-332.
936.
Jerry Hausman, Gregory Leonard and J. Douglas Zone, “Competitive Analysis with Differentiated
Products,” Annales d'Économie et de Statistique No. 34 (April-June 1994), pp. 159-80.
937.
Joe Wilcox, “Media2Go Team Gets Creative,” C/Net News.com, March 13, 2003.
938.
Joel L. Horowitz, “The Bootstrap,” in James Heckman and Edward Leamer, Handbook of Econometrics
Vol. 5, North Holland (2001), pp. 3159-228.
939.
John Johnson, “Economic Approaches to Antitrust Damage Estimation,” National Economic Research
Associates, January 2005.
940.
John Johnson, “Economic Approaches to Antitrust Damage Estimation,” NERA Economic Consulting,
January 2005.
941.
John M. Connor, “Forensic Economics: An Introduction with Special Emphasis on Price Fixing,” Journal
of Competition Law and Economics Vol 4, No. 1 (March 2008), pp. 31-59.
942.
Jonathan Baker and Timothy F. Bresnahan, “The Gains from Merger or Collusion in Product Differentiated
Industries,” Journal of Industrial Economics Vol. 33, No. 4 (June 1985), pp. 427-44.
943.
Joshua D. Angrist and Alan B. Krueger, “Instrumental Variables and the Search for Identification: From
Supply and Demand to Natural Experiments,” Journal of Economic Perspectives Vol. 15, No. 4 (Fall
2001), pp. 69-85.
944.
Joshua D. Angrist and Jörn-Steffen Pischke, Mostly Harmless Econometric: An Empiricists Companion ,
Chapter 8, pp. 293-325, Princeton University Press, 2009.
945.
Joshua D. Angrist, Guido W. Imbens, Donald B. Rubin, “Identification of Causal Effects Using
Instrumental Variables,” Journal of the American Statistical Association Vol. 91, No. 434
(June 1996), pp. 444-55.
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
946.
Keating, Gina, "Disney CEO touts new iTunes movie downloads," Reuters via LexisNexis, September 19,
2006.
947.
Ken Belson, “Wireless: What Can Apple Do to Keep the iPod Cool?” New York Times, May 9, 2005,
available at http://www.nytimes.com/2005/05/08/technology/08ihtwireless09.html.
948.
Kivanc Kirgiz, Michelle Burtis, and David A. Lunin, “Petroleum-Refining Industry Business Interruption
Losses Due to Hurricane Katrina,” Journal of Business Valuation and Economic Loss Analysis Vol. 4, No.
2 (2009), no pages (available online only from Berkeley Electronic Press).
949.
Krazit, Tom, "Apple's Jobs calls for DRM-free music," Cnet News via LexisNexis, February 6, 2007.
950.
Kwan, Michael, "Feature: Why Standalone MP3 Players Still Exist," Mobile Magazine, September 17,
2008, http://www.mobilemag.com/2008/09/17/feature-why-standalone-mp3-players-still-exist/
951.
"Laptop crashed using iTunes 7.0 with Windows Vista," Tech Support Forum, February 27, 2007,
http://www.techsupportforum.com/forums/f108/laptop-crashed-using-itunes-7-0-with-windows-vista.html
952.
Larry V. Hedges and Christopher H. Rhoads, “Correcting an Analysis of Variance for Clustering,” British
Journal of Mathematical and Statistical Psychology Vol. 64, No. 1 (February 2011), pp. 20-37.
953.
Mark Heflinger, “Zune MP3 Market Share up to 4%, Creative Drops to 2%,” Media Wire, May 12, 2008.
954.
Markoff, John, "Apple Plans To Inhabit Living Room," NY Times via LexisNexis, September 13, 2006.
955.
Marsal, Katie, "iTunes usage overtakes RealPlayer for the first time," AppleInsider, January 31, 2008,
http://appleinsider.com/articles/08/01/31/itunes_usage_overtakes_realplayer_for_the_first_time.
956.
Martell, Duncan, "UPDATE 1-Apple to offer movies on iTunes music store," Reuters via LexisNexis,
September 12, 2006.
957.
Matthew Hicks, “Motorola Previews iTunes Phone,” January 7, 2005, eWeek.com.
958.
Menn, Joseph, "Apple, Disney in a Bind on Movie Downloads?," LA Times via LexisNexis, September 26,
2006.
959.
Menn, Joseph, "Companies try to threaten iTunes, but Apple still wins," Chicago Tribune via LexisNexis,
August 27, 2007.
960.
Menn, Joseph, "ITunes' rivals may lift Apple," LA Times via LexisNexis, August 22, 2007.
961.
Merger Guidelines, U. S. Department of Justice and Federal Trade Commission, 1994.
962.
Michael H. Riordan, “Anticompetitive Vertical Integration by a Dominant Firm,” American Economic
Review Vol. 88, No. 5 (December 1998), pp. 1232-48.
963.
Michael H. Riordan, “Competitive Effects of Vertical Integration,” in Paolo Buccirossi (editor), Handbook
of Antitrust Economics , MIT Press, 2008, pp. 145-82.
964.
"Midi conversion broken?!," Apple Support Communities, March 21, 2006,
https://discussions.apple.com/thread/324488?start=0&tstart=0
965.
Moulds, Josephine, "technology Hacker unpeels Apple's iTunes DVD Jon says he has cracked playback
restrictions.," The Daily Telegraph via LexisNexis, October 26, 2006.
966.
"MP3 Players - 2010," Pew Internet & American Life Project , Accessed March 13, 2013,
http://www.pewinternet.org/Reports/2010/Gadgets/Report/Mp3-players.aspx
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
967.
"MP3 Players - 2011," Pew Internet & American Life Project , February 3, 2011,
http://www.pewinternet.org/Reports/2011/Generations-and-gadgets/Report/Mp3-players.aspx
968.
Musgrove, Mike, "A Messy Age for Music; Confusion Reigns In the Expanding Digital World,"
Washington Post via LexisNexis, October 22, 2006.
969.
"My Problem With iTunes 7," MediaLoper, September 19, 2006, http://medialoper.com/my-problem-withitunes-7/
970.
Naoki Watanabe, Ryo Nakajima, and Takanori Ida, “Quality-Adjusted Prices of Mobile Phone Handsets
and Carriers’ Product Strategies: The Japanese Case,” Discussion Paper No. 1224, Department of Social
Systems Management, University of Tsukuba, January 2009.
971.
"New Pew data: There's a good (and growing) chance you're reading this on your phone," Nieman
Journalism Lab , June 26, 2012, http://www.niemanlab.org/2012/06/new-pew-data-theres-a-good-andgrowing-chance-youre-reading-this-on-your-phone/
972.
"Nielsen: No evidence of on-demand music streams hurting download sales," LA Times, March 14, 2012,
http://latimesblogs.latimes.com/entertainmentnewsbuzz/2012/03/nielsen-on-demand-streams-downloadsales.html.
973.
Nirvikar Singh and Xavier Vives, “Price and Quantity Competition in a Differentiated Duopoly,” Rand
Journal of Economics Vol. 14, No. 4 (Winter 1984), pp. 546-54.
974.
"No one cares about the MP3 player market anymore," CNN Money Tech, August 22, 2011,
http://cnnmoneytech.tumblr.com/post/9253928206/no-one-cares-about-the-mp3-player-market-anymore
975.
O'Brien, Kevin J., "European Verdict on Appeal by Microsoft Expected Today," NY Times via LexisNexis,
September 17, 2007.
976.
"Opportunity Calling: The Future of Mobile Communications –Take Two," Oracle , October 31, 2011,
http://www.oracle.com/us/industries/communications/oracle-communications-future-mobile-521589.pdf
977.
"Opportunity Calling: The Future of Mobile Communications," Oracle , September 22, 2010,
http://www.oracle.com/us/industries/communications/oracle-communications-mobile-report-170802.pdf
978.
P. D. Chwelos, E. R. Berndt and I. M. Cockburn, “Faster, Smaller, Cheaper: An Hedonic Price Analysis of
PDAs,” Applied Economics Vol. 40, No. 22-24 (November-December 2008), pp. 2839-56.
979.
Patrick Bajari and Lanier Benkard, “Demand Estimation with Heterogeneous Consumers and Unobserved
Product Characteristics: A Hedonic Approach,” Journal of Political Economy Vol. 113, No. 6 (December
2005), pp. 1239-76.
980.
Paul D. Chwelos, Ernst R. Berndt, and Iain M. Cockburn, “Faster, Smaller, Cheaper: A Hedonic Price
Analysis of PDAs,” Applied Economics Vol. 40, No. 22 (November 2008), pp. 2839-56.
981.
Pegoraro, Rob, "Movie Downloads Remain a Production Worth Skipping," Washington Post via
LexisNexis, September 17, 2006.
982.
Philip Elmer-DeWitt, “How to Grow the iPod as the MP3 Market Shrinks,” Fortune, January 20, 2008,
available at http://tech.fortune.cnn.com/2008/01/29/beyond-the-incredible-shrinking-ipod-market/.
983.
Pio Baake and Anette Boom, “Vertical Product Differentiation, Network Externalities, and Compatibility,”
International Journal of Industrial Organization Vol. 19, Nos. 1-2 (January 2001), pp. 267-84.
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
984.
Pogue, David, "New at Apple: Smaller iPods, Bigger Ideas," NY Times via LexisNexis, September 14,
2006.
985.
“Portable Digital Players: iPods Rule but Consider Other Brands,” Consumer Reports,
November 2006, available at http://www.consumerreports.org/cro/electronics-computers/audiovideo/
audio/ipods-mp3-players/mp3-players-11-06/overview/1106_mp3_ov_1.htm.
986.
“Portable Digital Players: iPods Rule, but Consider Other Brands,” Consumer Reports, November 1, 2006.
987.
Portable Devices Instructions, RealPlayer Resource, 2006, http://www.realplayerresource.com/.
988.
"Report Brochure: Mobile Phones - US - January 2013," Mintel ,
http://academic.mintel.com/display/637571/
989.
"RIAA data shows streaming music up 19%, downloads up 22%," Electronista, March 27, 2012,
http://www.electronista.com/articles/12/03/27/riaa.shows.streaming.on.the.rise/
990.
Rob Walker, “The Guts of a New Machine,” New York Times, November 30, 2003.
991.
Robert H. Porter, “A Study of Cartel Stability: The Joint Executive Committee, 1880-1886,” Bell Journal
of Economics Vol. 14, No. 2 (Autumn 1983), pp. 301-14.
992.
Roger D. Blair and David L. Kaserman, Antitrust Economics , Richard D. Irwin, 1985.
993.
Rolf Dewenter, Justus Haucup, Ricardo Luther and Peter Rotzel, “Hedonic Prices in the German Market
for Mobile Phones,” Telecommunications Policy Vol. 31, No. 1 (2007), pp. 4-13.
994.
Sanders, Tom, "Apple users report serious iTunes 7 problems," V3, September 15, 2006,
http://www.v3.co.uk/v3-uk/news/1960766/apple-users-report-itunes.
995.
Sayer, Peter, "Sony BMG to sell DRM-free music downloads through stores," InfoWorld, January 7, 2008,
http://www.infoworld.com/t/applications/sony-bmg-sell-drm-free-music-downloads-through-stores-341
996.
Selburn, Jordan, "Rising Media Tablet and Smartphone Sales Cut Demand for Single-Task Consumer
Products," iSuppli , July 28, 2011, http://www.isuppli.com/home-and-consumerelectronics/news/pages/rising-media-tablet-and-smartphone-sales-cut-demand-for-single-task-consumerproducts.aspx
997.
Shabtai Donnenfeld and Shlomo Weber, “Vertical Product Differentiation with Entry,” International
Journal of Industrial Organization Vol. 10, No. 3 (September 1992), pp. 449-72.
998.
"Smartphones Heavily Decrease Sales of iPod, MP3 Players," Tom's Hardware , December 31, 2012,
http://www.tomshardware.com/news/Smartphones-iPod-MP3-Players-Sales,20062.html.
999.
"Songs Disappearing from iTunes," Apple Support Communities, June 12, 2011,
https://discussions.apple.com/thread/1902641?start=345&tstart=0
1000.
Sorrel, Charlie, “Portable CD Player Sales Up 50%,” Wired , December 7, 2007, available at
http://www.wired.com/gadgetlab/2008/12/portable-cd-pla/.
1001.
"Spotlight on music download promos," EPM Communications via LexisNexis, December 15, 2007.
1002.
Steven Levy, The Perfect Thing: How the iPod Shuffles Commerce, Culture, and Coolness , Simon &
Schuster, 2006, pp. 75-106.
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
1003.
"Streaming and Micropennies: The Footnotes," NY Times, January 29, 2013,
http://mediadecoder.blogs.nytimes.com/2013/01/29/streaming-and-micropennies-the-footnotes/
1004.
"Streaming Media, IPTV, and Broadband Transport: Telecommunications Carriers and Entertainment
Services 2008-2013," The Insight Research Corporation, http://www.insight-corp.com/reports/IPTV08.asp
1005.
"Streaming music, not cannibalising download sales," U Music, December 3, 2012,
http://www.umusic.co.uk/umusic-blog/mythbusting/237-streaming-music-not-cannibalising-download-sales
1006.
"Thanks to digital downloads, music industry finally on road to recovery," GMA News , January 26, 2013,
http://www.gmanetwork.com/news/story/292104/scitech/technology/thanks-to-digital-downloads-musicindustry-finally-on-road-to-recovery
1007.
“The Mix Get Richer: New iPod touch & iPhone Capacities,” Credit Suisse, February
5, 2008.
1008.
“The NPD Group: Amazon MP3 Music Download Store Offers a New Hope for Digital Music Growth,”
NPD Group Press Release, April 15, 2008.
1009.
"The download on movies; Apple's iTV prepares to compete with DVD movie rentals, sales," Chicago
Tribune, LexisNexis , September 15, 2006.
1010.
"The Nielsen Company & Billboard’s 2012 Music Industry Report," The Business Wire , January 4, 2013,
http://www.businesswire.com/news/home/20130104005149/en/Nielsen-Company-Billboard%E2%80%99s2012-Music-Industry-Report
1011.
"The Ten Businesses the Smartphone Has Destroyed," Yahoo! Finance , November 12, 2010,
http://finance.yahoo.com/news/pf_article_111299.html
1012.
Tode, Chantal, "Smartphones are replacing MP3 players, digital cameras and GPS devices: Oracle,"
Mobile Marketer, November 1, 2011, http://www.mobilemarketer.com/cms/news/research/11371.html
1013.
"Topic: Why nothing from Apple wrt iTunes 7 problems?," iLounge Forums ,
http://forums.ilounge.com/itunes-related-mac-pc-applications/174562-why-nothing-apple-wrt-itunes-7problems.html.
1014.
"Unable To Play Radio In iTunes 7.0.1," Apple Support Communities, November 2, 2006,
https://discussions.apple.com/message/3305091?messageID=3305091
1015.
"UPDATE: Apple Launches Movie Downloads, New iPods," Dow Jones Factiva via LexisNexis,
September 13, 2006.
1016.
"upgrade to 7.0 problems," Apple Support Communities, December 28, 2006,
https://discussions.apple.com/message/3769565?messageID=3769565
1017.
"Upgrading To iTunes 7:0? You Might Want To Wait," Newstex Web Blogs via LexisNexis, September 19,
2006.
1018.
W. Kip Viscusi, Joseph E. Harrington, Jr., and John M. Vernon, Economics of Regulation and Antitrust ,
4th Edition, MIT Press, 2005.
1019.
Wakabayashi, Daisuke, "Apple seen linking movie downloads to TV," Reuters via LexisNexis, September
11, 2006.
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
1020.
"Warner Offers DRM-Free MP3s Via Amazon," Spin , September 29, 2007,
http://www.spin.com/articles/warner-offers-drm-free-mp3s-amazon
1021.
Warner Music Group F1Q07 (Qtr End 12/31/06) Earnings Call Transcript
1022.
"When will apple fix iTunes 7.0?," Apple Support Communities, October 8, 2006,
https://discussions.apple.com/thread/678309?start=0&tstart=0
1023.
"Why pay if it’s free? Streaming, downloading, and digital music consumption in the 'iTunes era'," LSE ,
2011,
http://www2.lse.ac.uk/media@lse/research/mediaWorkingPapers/MScDissertationSeries/2011/71.pdf
1024.
"Why Spotify Will Kill iTunes," Harvard Business Review, July 22, 2011,
http://blogs.hbr.org/cs/2011/07/why_spotify_will_kill_itunes.html
1025.
Wingfield, Nick, "Apple Computer Aims to Take Over Your Living-Room TV," WSJ via LexisNexis,
September 13, 2006.
1026.
Wright, Dale, "Is the Smartphone Killing the iPod?," Snugg Buzz , December 28, 2012,
http://www.thesnugg.com/news/2012/12/is-the-smartphone-killing-the-ipod/
1027.
Xavier Vives, “On the Efficiency of Bertrand and Cournot Equilibria with Product Differentiation,”
Journal of Economic Theory Vol. 36, No. 1 (June 1985), pp. 166-75.
1028.
Yongmin Chen, “On Vertical Mergers and Their Competitive Effects,” Rand Journal of Economics Vol.
12, No. 4 (Winter 2001), pp. 667-85.
1029.
"Young listeners opting to stream, not own music," CNN , June 16, 2012,
http://www.cnn.com/2012/06/15/tech/web/music-streaming
1030.
“Statement by Mike Farrace,” Hearings before the Committee of the Judiciary, U.S. Senate, April 3, 2001,
2001 WL 323720 (F.D.C.H.).
1031.
“2000 Annual Survey Results,” National Association of Recording Merchandisers.
1032.
“Statement of National Association of Recording Merchandisers,” Hearings before the Committee of the
Judiciary, U. S. Senate, April 3, 2001, (www.narm.com).
1033.
Dan Tyson, “The 25 Worst Tech Products of All Time,” PC World , May 26, 2006, at
www.pcworld.com/article/125772-3/the_25_worst_tech_products_of_all_time.html.
1034.
Matthew Hicks, “Motorola Previews iTunes Phone,” January 7, 2005, eWeek.com .
1035.
Neil Weinstock Netanel, “Temptations of the Walled Garden: Digital Rights Management and Mobile
Phone Carriers,” Journal on Telecommunications and High-Technology Law Vol 6 (2007-08).
1036.
Thierry Reyna and Ludmila Striukova, “White Knight or Trojan Horse? The Consequences of Digital
Rights Management for Consumers, Firms and Society,” Communications & Strategies No. 69 (1st
Quarter, 2008).
1037.
Owen Williams, “Zune Player Officially Discontinued,” Simcity , October 4, 2011, at
http://www.neowin.net/news/zune-player-officially-discontinued.
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
1038.
The RealNetworks press release, dated August 17, 2004, available on PRNewswire at
http://www.prnewswire.com/news-releases/realnetworks-kicks-off-freedom-of-choice-campaign-withbiggest-music-sale-in-history-71643667.html.
1039.
Carl Shapiro and Hal Varian, Information Rules: A Strategic Guide to the Information Economy , Harvard
University Press, 1999
1040.
Pei-yu Chen and Lorin M. Hitt, “Information Technology and Switching Costs,” in Terrence Hendershott,
Handbook on Economics and Information Systems , Elsevier, 2006.
1041.
Joseph Farrell and Paul Klemperer, “Coordination and Lock-in: Competition with Switching Costs and
Network Effects,” in Mark Armstrong and Robert Porter, Handbook of Industrial Organization Vol. 3,
Elsevier, 2007.
1042.
Severin Borenstein, Jeffrey K. Mackie-Mason, and Janice S. Netz, “Antitrust Policy in Aftermarkets,”
Antitrust Law Journal Vol. 63 (1994-95).
1043.
Jemima Kiss, “How Big Is the iPod Installed Base?” Guardian , September 9, 2009
1044.
Larry Dignan, “Tablet Replacement Rates: More Like an MP3 Player than PC,” ZDNet January 4, 2011
1045.
“Mobile Phone Lifecycles,” GSM Association, 2006
1046.
“The Life Cycle of a Cell Phone,” U.S. Environmental Protection Agency, 2005
1047.
John Paczkowski, “I Got a Fever, and the Only Prescription is… More iPhone!” All Things Digital , June
25, 1010 (reporting that the replacement cycle for iPhones is 14.7 months)
1048.
Victor H., “Americans Replace Their Cell Phones Every 2 Years, Finns – Every Six, a Study Claims,”
Phonearena.com , July 11, 2011
1049.
V. Brian Viard, “Do Switching Costs Make Markets More or Less Competitive? The Case of 800-Number
Portability,” Rand Journal of Economics Vol. 18, No. 1 (Spring 2007)
1050.
Jung Eun Ku, Sang Woo Lee, and Tschanghee Hyun, “Value of Number Portability on Internet Phones,”
ETRI Journal Vol. 32, No. 1 (February 2010)
1051.
Tokio Otsuka and Hitoshi Mitomo, “User Benefits and Operator Costs of Mobile Number Portability in
Japan and Impact on Market Competitiveness,” Telecommunications Policy (in press), at
http://www.sciencedirect.com/science/journal/aip/03085961
1052.
Akohiro Nakamura, “Estimating Switching Costs Involved in Changing Mobile Phone Carriers in Japan:
Evaluation of Lock-in Factors Related to Japan’s SIM Card Locks,” Telecommunications Policy 34 (2020)
1053.
Lucio Fuentelsaz, Juan Pablo Maicas, and Yolando Polo, “Switching Costs, Network Effects, and
Competition in the European Mobile Telecommunications Industry,” Information Systems Research Vol.
23, No. 1 (March 2012)
1054.
Fuentelsaz, et al., op. cit. and Stephen C. Littlechild, “Mobile Termination Charges: Calling Party Pays
Versus Receiving Party Pays,” Telecommunications Policy Vol. 30 (2006)
1055.
David Harbord and Marco Pagnozzi, “Network-Based Price Discrimination and ‘Bill-and-Keep’ vs. ‘CostBased’ Regulation of Mobile Termination Rates,” Journal of Network Economics Vol. 9, No. 1 (March
2010)
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
1056.
Horizontal Merger Guidelines , U. S. Department of Justice and Federal Trade Commission, 2010
1057.
“iTunes Competitive Landscape,” October 2007
1058.
“Study: Spotify is Detrimental to Music Purchasing,” Digital Music News, November 15, 2011, at
http://www.digitalmusicnews.com/permalink/2011/111115 cannibal#VIZ3-3IxRZUcRMwuQcs_9g.
1059.
Corey Tate, “Rdio, Spotify and Napster Lose 200 Record Labels Due to NARM Study,” Spacelab,
November 19, 2011, at http://www.thespacelab.tv/spaceLAB/2011/ 11November/MusicNews-064-RdioSpotify-Napster-NARM-NPD.htm.
1060.
James Martin, "Mobile Computing: The Newest Wireless Technology," PC World, November 14, 2002, at
http://www.pcworld.com/article/106149/mobile_computing_the_newest_wireless_technology.html.
1061.
Federal Communications Commission, Annual Report and Analysis of Competitive Market Conditions
With Respect to Mobile Wireless, Including Commercial Mobile Services , June 24, 2011.
1062.
Adam Fisher, “The 50 Best Websites of 2009,” Time , August 24, 2009, at
http://www.time.com/time/specials/packages/completelist/0,29569,1918031,00.html.
1063.
Ronald Coase, “Payola in Radio and Television Broadcasting,” Journal of Law and Economics Vol. 22,
No. 2 (October 1979).
1064.
Federal Communications Commission, “Broadcasters Pay $12.5 Million to Resolve Possible ‘Payola’
Violations,” April 13, 2007, at http://transition.fcc.gov/eb/News_Releases/DOC-272304A1.html.
1065.
Federal Communications Commission, Order: In the Matter of Emmis Austin Radio Broadcasting
Company, L.P., File No. EB-06-IH-2944, July 22, 2011, at http://transition.fcc.gov/eb/Orders/2011/DA-11888A1.html.
1066.
Pandora Media Inc., Form 10-Q for Quarter Ended July 31, 2011, September 2, 2011, p. 33.
1067.
“Apple Unveils the New iPod” and “Apple Announces iTunes 6 with 2,000 Music Videos, Pixar Short
Films & Hit Shows,” Apple Press Releases, October 12, 2005. See also http://www.applehistory.com/?page=gallery&model=ipod_video.
1068.
“Apple Unveils the iTunes WiFi Music Store” and “Apple Unveils iPod Touch,” Apple Press Releases,
September 5, 2007.
1069.
“Portable Digital Players: iPods Rule but Consider Other Brands,” Consumer Reports , November 2006,
at http://www.consumerreports.org/cro/electronics-computers/audio-video/audio/ipods-mp3-players/mp3players-11-06/overview/1106_mp3_ov_1.htm.
1070.
Jasmine France, “MP3 Players That Shaped 2007,” CNet, February 8, 2008, at
http://reviews.cnet.com/4321-6490_7-6606044.html
1071.
Grace Acquino, “Dialed in: Best Cell Phones of 2005,” PC World , December 29, 2005, at
http://www.pcworld.com/article/123742/article.html.
1072.
“The Music Cell Phone,” in History of Mobile Phones , Sutliffian Press, 2007,
http://www.xtimeline.com/evt/view.aspx?id=26731.
1073.
“AAPL: The Reason for the iPhone’s Reported Woes Is Closer than You Think: It’s the iPod touch,”
Needham & Company, January 28, 2008
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
1074.
“Apple’s Negative Guidance Tone at the FQ1 Call Means a Lower Valuation of Multiple: Hold,”
Kintisheff Research, January 23, 2008.
1075.
“AAPL: Its MacWorld and We’re Just Living in It, Upgrading Apple from Buy to Strong Buy,” Needham
and Company, January 23, 2008.
1076.
“Portable CD Players Reviews,” ConsumerSearch, January 2008, at
http://www.consumersearch.com/portable-cd-players/review.
1077.
“The NPD Group: Amazon MP2 Music Download Store Offers a New Hope for Digital Music Growth,”
NPD Press Release, April 15, 2008.
1078.
“comScore Reports December 2011 U.S. Mobile Subscriber Market Share,” comScore, February 2, 2012,
at http://www.comscore.com/Insights/Press_Releases/
2012/2/comScore_Reports_December_2011_U.S._Mobile_Subscriber_Market_Share.
1079.
“iTunes Store: iTunes Plus Frequently Asked Questions (FAQ)” on Apple’s website,
http://support.apple.com/kb/ht1711.
1080.
“The Music Cell Phone,” in History of Mobile Phones , Sutliffian Press, 2007,
http://www.xtimeline.com/evt/view.aspx?id=26731.
1081.
Alex Cosper, “The History of the Portable MP3 Player,” eHow , no date, at
http://www.ehow.com/about_5409458_history-portable-mp-players.html.
1082.
Antone Gonsolves, “Verizon Launches iPod Like Music Phone,” Information Week , July 31, 2006, at
http://www.informationweek.com/verizon-wireless-launches-ipod-like-musi/191600770.
1083.
Billboard Staff, “Hot 100 Impacted by New On-Demand Songs Chart,” Billboard , March 14, 2012, at
http://www.billboard.com/articles/news/502020/hot-100-impacted-by-new-on-demand-songs-chart.
1084.
Christopher R. Leslie, “Antitrust Damages and Deadweight Loss,” Antitrust Bulletin , Vol. 51, No. 3 (Sept.
2006).
1085.
Colleen Bowen, “Universal Launches DRM-Free Music Test,” TWICE , August 10, 2007, at
http://www.twice.com/article/258659-Universal_Music_Launches_DRM_Free_ Music_Test.php.
1086.
Donald Bell, “MP3 Player Buying Guide,” CNET Reviews , November 5, 2012, at
http://reviews.cnet.com/mp3-player-buying-guide/.
1087.
EMI Launches DRM-Free Superior Sound Quality Downloads Across Its Entire Digital Repertoire,”
Webwire , April 3, 2007, at http://www.webwire.com/ViewPressRel. asp?aId=31281.
1088.
Federal Communications Commission, Annual Report and Analysis of Competitive Market Conditions
With Respect to Mobile Wireless, Including Commercial Mobile Services , March 21, 2013.
1089.
Godefroy DangNguyen, Sylvain Dejean, and Francois Moreau, “Are Streaming and Other Music
Consumption Modes Substitutes or Complements?” March 16, 2012, at
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2025071.
1090.
Grace Acquino, “Dialed in: Best Cell Phones of 2005,” PC World , December 29, 2005, at
http://www.pcworld.com/article/123742/article.html.
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
1091.
Jason Dedrick, Kenneth L. Kraemer and Greg Linden, “Who Profits from Innovation in Global Value
Chains? A Study of the iPod and Notebook PCs,” Industrial and Corporate Change , June 22, 2009, at
http://icc.oxfordjournals.org/cgi/reprint/dtp032r1.
1092.
Joshua P. Friedlander, “News and Notes on 2012 RIAA Music Industry Shipment and Revenue Statistics,”
RIAA, at http://www.riaa.com/keystatistics.php?content_ selector=2008-2009-U.S-Shipment-Numbers.
1093.
Kathryn Zucker, Generations and Their Gadgets , Pew Internet and American Life Project, February 3,
2011, at http://www.pewinternet.org/Reports/2011/Generations-and-gadgets/Overview.aspx.
1094.
Olga Kharif, “Another Music Phone? Yawn…” Bloomberg-Business Week , October 18, 2006, at
http://www.businessweek.com/stories/ 2006-10-18/another-music-phone-yawn-businessweek-businessnews-stock-market-and-financial-advice.
1095.
Philip Elmer-DeWitt, “How to Grow the iPod as the MP3 Market Shrinks,” Fortune , January 20, 2008,
accessed on http://apple20.blogs.fortune.cnn.com/2008/01/29/beyond-the-incredible-shrinking-ipodmarket/.
1096.
Rick Marshall, “Oh Mercy : How On-Demand Interactive Streaming Services Navigate the Digital Music
Rights Licensing Landscape,” November 21, 2012, at http://papers.ssrn.com/sol3/papers.
cfm?abstract_id=2179111.
1097.
See Ralf Dewenter, Justus Haucap, Ricardo Luther, and Peter Rotzel, “Hedonic Prices in the German
Market for Mobile Phones,” Information Economics and Policy 31 (2007)
1098.
Stan Horaczek, “LG Fusic Music Phone Reviewed,” July 8, 2006, at
http://www.engadget.com/2006/07/08/lg-fusic-music-phone-reviewed/.
1099.
Thomas J. Fitzgerald, “Music to Your Cell Phone,” New York Times , July 7, 2005, at
http://www.nytimes.com/2005/07/07/technology/circuits/07basics.html?_r=0.
1100.
Volker Grassmuch, “Academic Studies on the Effect of File-Sharing on the Recorded Music Industry: A
Literature Review,” available at http://papers.ssrn.com/sol3/ papers.cfm?abstract_id=1749579.
1101.
Yuki Noguchi, “Another Shot at a Music Phone,” Washington Post , November 7, 2006, at
http://voices.washingtonpost.com/posttech/2006/11/xm_satellite_on_your_phone.html.
1102.
Apple Press Releases 2005-2011, available at http://www.apple.com/pr/library.
1103.
“Apple Announces iTunes 6 with 2,000 Music Videos, Pixar Short Films & Hit Shows,” Apple Press
Release, October 12, 2005.
1104.
“Apple Announces iTunes 7 with Amazing New Features," Apple Press Release, September 12, 2006.
1105.
“Apple Introduces New iPod Nano with Built-in Video Camera,” Apple Press Release, September 9, 2009.
1106.
“Apple Introduces New iPod Touch Lineup,” Apple Press Release, September 9, 2009.
1107.
“Apple Introduces the New iPod," Apple Press Release, September 12, 2006.
1108.
“Apple Unveils iPod Touch,” Apple Press Release, September 5, 2007.
1109.
“Apple Unveils the iTunes WiFi Music Store,” Apple Press Release, September 5, 2007.
1110.
“Apple Unveils the New iPod,” Apple Press Release, October 12, 2005.
1111.
“Apple’s iPod Shuffle Now Starts at Just $59,” Apple Press Release, September 9, 2009.
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
1112.
“Changes Coming to iTunes Store,” Apple Press Release, January 6, 2009.
1113.
Christoper R. Leslie, “Antitrust Damages and Deadweight Loss,” Antitrust Bulletin Vol. 51, No. 3 (Sept.
2006).
1114.
“iPhone Premiers This Friday Night at Apple Retail Stores,” Apple Press Release, June 28, 2007.
1115.
“iTunes Store Tops Over Five Billion Songs Sold,” Apple Press Release, June 19, 2008.
1116.
“Linkin Park Catalog, Exclusive Bonus Tracks & Digital Booklets Now Available on the iTunes Music
Store,” Apple Press Release, August 29, 2006.
1117.
“The NPD Group: Amazon MP2 Music Download Store Offers a New Hope for Digital Music Growth,”
NPD Press Release, April 15, 2008.
1118.
http://support.sas.com/
1119.
http://www.amazon.com/gp/bestsellers/electronics/16009311/ref=pd_ts_pg_1?ie=UTF8&pg=1.
1120.
http://www.apple.com
1121.
http://www.apple-history.com
1122.
http://www.cnet.com/ipod/.
1123.
Apple’s web site: http://support.apple.com/kb/HT1334.
1124.
http://www.consumerreports.org/cro/search.htm?query=portable+cd+players.
1125.
http://reviews.cnet.com/1770-5_70.html?query=editors+review+portable+cd+player&tag=srch&searchtype=products
1126.
http://www.amazon.com/MP3-Players-AudioVideo/b/ref=amb_link_86347991_%203?ie=UTF8&node=172630&pf_rd_m=ATVPDKIKX0DER&pf_rd
_s=top-1&pf_rd_%20r=0TT5Q01MYZKSP88YNK53&pf_rd_t=301&pf_rd_p=157251702&pf_rd_i=mp3
1127.
http://www.amazon.com/gp/search/ref=sr_tc_2_1?rh=n%3A1264866011%2Ck%3Amp3&keywords=mp3
&ie=UTF8&qid=1294860099&sr=1-2-tc
1128.
http://www.amazon.com/ MP3-Players-Portable-AudioVideo/b/ref=amb_link_157669822_24?ie=UTF8&node=
1264866011&pf_rd_m=ATVPDKIKX0DER&pf_rd_s=center-5&pf_rd_r=
1T562SPGE1105HXYHVFZ&pf_rd_t=101&pf_rd_p=945911022&pf_rd_i=172630.
1129.
Pandora. http://www.pandora.com/about/mgp.
1130.
Last FM. http://www.last.fm/about.
1131.
The FCC's Payola Rules. http://www.fcc.gov/guides/payola-rules.
1132.
http://transition.fcc.gov/eb/broadcast/sponsid.html.
1133.
Descriptions and reviews of audio download sites, see http://music-download-review.toptenreviews.com/.
1134.
http://support.apple.com/kb/ht1353
1135.
http://support.apple.com/specs/#iphone
1136.
http://support.apple.com/specs/#ipod
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX B
Documents/Data Considered By Professor Roger G. Noll and/or Economists Inc.
1137.
http://www.apple.com/pr/products/ipodhistory/
1138.
http://www.everymac.com/systems/apple/consumer_electronics/index-ipod.html
1139.
All materials cited in any of my 6 reports on this matter.
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPENDIX C
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
CONFIDENTIAL - ATTORNEYS EYES ONLY
APPLE TYING
Service List - 4/3/2013
Page 1 of 1
(06-0171)
Counsel for Defendant(s)
Robert A. Mittelstaedt
Craig E. Stewart
David Craig Kiernan
Jones Day
555 California Street, 26th Floor
San Francisco, CA 94104
415/626-3939
415/875-5700 (Fax)
Counsel for Plaintiff(s)
Andrew S. Friedman
Elaine A. Ryan
Todd D. Carpenter
Bonnett, Fairbourn, Friedman & Balint, P.C.
2325 E. Camelback Road, Suite 300
Phoenix, AZ 85016
602/274-1100
602/274-1199 (Fax)
Michael D. Braun
Marc L. Godino
Braun Law Group, P.C.
10680 West Pico Blvd., Suite 280
Los Angeles, CA 90064
310/836-6000
310/836-6010 (Fax)
Eric J. Belfi
Labaton Sucharow LLP
140 Broadway, 34th Floor
New York, NY 10005
212/907-0700
212/818-0477 (Fax)
Jacqueline Sailer
Murray Frank LLP
275 Madison Avenue, Suite 801
New York, NY 10016
212/682-1818
212/682-1892 (Fax)
Bonny E. Sweeney
Alexandra S. Bernay
Thomas R. Merrick
Robbins Geller Rudman & Dowd LLP
655 West Broadway, Suite 1900
San Diego, CA 92101
619/231-1058
619/231-7423 (Fax)
Roy A. Katriel
The Katriel Law Firm
1101 30th Street, N.W., Suite 500
Washington, DC 20007
202/625-4342
866/373-4023 (Fax)
*APPLE'S (PROPOSED) REDACTIONS*
EXHIBIT 2
[Filed Under Seal]
UNITED STATES DISTRICT COURT
NORTHERN DISTRICT OF CALIFORNIA
OAKLAND DIVISION
____________________________________
|
|
|
LITIGATION
|
____________________________________|
APPLE IPOD ITUNES ANTI-TRUST
Lead Case No. C-05-00037-YGR
REBUTTAL DECLARATION OF ROGER G. NOLL
ON LIABILITY AND DAMAGES
My name is Roger G. Noll, and previously I have submitted eight declarations in
this proceeding.1 My previous declarations contain my career history and qualifications.
My current curriculum vita, which has an up-to-date list of my publications, is attached to
this declaration as Appendix A.
Since submitting my report on liability and damages, I have submitted expert
reports and been deposed in the following new matters.
In re Application of Pandora Media, Inc. Related to U.S. vs. ASCAP (U.S. District
Court, New York City);
City of San Jose, et al., vs. Office of the Commissioner of Baseball, et al. (U.S.
District Court, San Jose); and
In re Electronic Books Antitrust Litigation (U.S. District Court, New York City).
1
Declaration of Roger G. Noll (July 15, 2008), Reply Declaration of Roger G. Noll
(October 19, 2009), Declaration of Roger G. Noll (January 18, 2011), Reply Declaration
of Roger G. Noll (March 28, 2011), Supplemental Declaration of Roger G. Noll (July 18,
2011), Second Supplemental Declaration of Roger G. Noll (September 23, 2011),
Declaration of Roger G. Noll on Liability and Damages (April 13, 2013; henceforth Noll
Merits Report), and Corrections to the Declaration of Roger G. Noll on Liability and
Damages (May 13, 2013; henceforth, Noll Corrections Report).
ASSIGNMENT
Attorneys for the class plaintiffs in this litigation have asked me to review the
Expert Report of Kevin M. Murphy (henceforth Murphy Report) and the Expert Report of
Robert H. Topel (henceforth Topel Report) that were submitted on behalf of defendant
Apple, Inc., to determine whether any of the evidence and analysis in these reports causes
me to change the conclusions in the Noll Merits Report, as amended in the Noll
Corrections Report. The attorneys for the class plaintiffs also have asked me to reestimate the damages regressions in the Noll Corrections Report to take into account new
information on the iPod models in which iTunes 7.0 was used. In undertaking this
assignment, I have read the expert reports and depositions of Professors Murphy and
Topel and other documents and publications that are referenced in the footnotes of this
declaration. In preparing this report I have been assisted by the staff of Economists, Inc.
SUMMARY AND CONCLUSIONS
This declaration reports new regression equations for demonstrating class-wide
harm and calculating damages. The most important source of differences between the
new regressions and the regressions that were reported in the Noll Merits Report is that
the new regressions limit the effect of iTunes 7.0 to iPod models for which the features of
iTunes 7.0 that disabled Harmony were enabled, according to information that Apple
produced after the Noll Merits Report was submitted. (Throughout this report, I use the
term “iPod model” to refer to a class/generation/family of iPod.)
The Murphy Report and the Topel Report make numerous criticisms of the
analysis of both liability and damages in my previous reports. Their most important
2
complaint is that the regressions that I used in the Noll Merits Report and the Noll
Corrections Report contain numerous errors in specifying and estimating the equations
that cause the results of those regressions to be unreliable, and that correcting these errors
causes the estimate of anticompetitive injury and damages to class members to be zero.
After reviewing these reports and undertaking further econometric analysis, I
have concluded that the criticisms by defendant’s experts fall into two categories. The
first consists of criticisms that have some basis in economics but that overstate the
importance of their criticisms. In these cases, regardless of whether I agree that the
criticisms warrant changing the procedures that I used, I have adopted the changes that
Professors Murphy and Topel propose, and find that these changes have only a small
effect on the amount of damages that is calculated from the regression equations. The
second category consists of criticisms that have no basis in economics and so do not
warrant changes in the analysis and methods in my previous reports.
Liability Issues
The Murphy Report contains several criticisms about my analysis of market
definition and market power. These criticisms are not based on a valid economic analysis
of either issue, and so do not cause me to change any of the arguments and analysis in the
Noll Merits Report.
Professor Murphy sets forth reasons that he believes that closed systems (“walled
gardens”) like the Apple combinations of iPods, iTunes media player software, and the
iTunes Store (iTS) are superior to a system composed of complementary products that are
purchased from different vendors. While Professor Murphy does not provide any
3
economic analysis to support this assessment, even if it were true Professor Murphy does
not explain why disabling the use of competing products is necessary or beneficial to
consumers if technologies that create lock-in are also more expensive because lock-in
diminishes competition.
Professor Murphy also argues that a seller of all complementary products will
charge lower, not higher, prices. This claim is incorrect because it assumes that the seller
of each component enjoys unilateral market power and overlooks the effect of lock-in. If
a technology locks customers in to buying all components from a single seller,
competition in each component is reduced, and reduced competition can cause the price
of each component to be higher, not lower.
Professor Murphy argues that digital downloads are not a properly defined market
because consumers can obtain recordings in other ways, and that in any case this issue is
irrelevant because the case is about lock-in. Regarding competition among types of
recordings (CDs, illegal file sharing, on-demand streaming), Professor Murphy does not
apply the methods of antitrust economics and offers no economic evidence to support the
conclusion that other forms of recordings are as close competitive substitutes for iTS as
are other download sites. In addition, the claim that lock-in effects render market
definition irrelevant is incorrect – one must still address which vendors of which products
would be the most intense competitors for the reference product in the absence of lock-in.
Professor Murphy also opines that my analysis of the sources of Apple’s market
power in iPods – and the attribution of some of this market power to iTunes 7.0 – is
incorrect because Apple also introduced iPods with new features when iTunes 7.0 was
released. This criticism ignores the fact that the regression equation that explains prices
4
includes indicator variables for class, other product attributes, and time (to represent
improved technology), which measure the effect of the characteristics of an iPod.
Criticisms Leading to Changes in Damages Calculations
Professors Murphy and Topel make four criticisms that lead me to make changes
in the regression equations in the Noll Merits Report.
First, defendant’s experts propose that quantity weights, rather than frequency
weights, should be used to estimate the regression equations. While I agree that quantity
weights produce better estimates of the standard errors of coefficients, the choice of a
weighting procedure does not affect the regression coefficients and, in a large data set
such as the Apple transactions records, leads to only minor changes in the results of the
tests of the statistical significance of the coefficients on the independent variables in the
regression equation. I show that this choice is inconsequential, but I use quantity weights
in the revised damages model.
Second, Professors Murphy and Topel criticize my use of the logarithm of time
rather than the scalar variable time in my regressions that use the logarithm of price as the
dependent variable. I agree that there is no theoretical reason to prefer the logarithm of
time to the scalar measure, so in the re-estimated damages equations I eliminate this
difference between us by adopting the specification that is preferred by Professors
Murphy and Topel.
Third, Professor Topel proposes that the appropriate date on which Apple’s
iTunes Store (iTS) adopted a DRM-free format for music was January 6, 2009, instead of
March 31, 2009, the date that was used in my previous reports. The facts are that Apple
5
made a transition to DRM-free music that began on January 6, but the significance of the
initial change on January 6 and the speed of the complete transition to making DRM-free
music available for all titles is unknown. Hence, a perfect measure of the availability of
DRM-free music from January 6 to March 31, 2009, cannot be constructed. Because the
issue of the best way to measure the importance of the introduction of DRM-free
recordings on the iTunes Store (iTS) during the period between January 6 and March 31
is not of major importance, in the revised regressions I adopt Professor Topel’s proposal
and start the variable that indicates the availability of DRM-free music on January 6,
rather than March 31.
Fourth, Professors Murphy and Topel argue that the indicator variable for
Harmony, software that allowed iPod users to download recordings from Internet sites
that used the RealNetworks DRM format, should distinguish between the first version,
which was available from July 2004 until April 2005 and was disabled by iTunes 4.7, and
the second version, which was introduced in April 2005 and worked around the disabling
features of previous versions of the iTunes digital media player software starting with
iTunes 4.7. In my new damages model, I have adopted this procedure.
Incorrect Criticisms that Require No Change
Professors Topel and Murphy make several criticisms that are not valid and that
would reduce, not increase, the reliability of the regression analysis if they were adopted.
First, Professor Murphy and Topel argue that the indicator variable for iTunes 4.7
and subsequent versions of iTunes prior to the introduction of iTunes 7.0 should continue
to take the value of one after this software was replaced by iTunes 7.0. This criticism
6
was directed at my original damages model, which was based on the assumption that
iTunes 7.0 was used in all iPods that were sold after September 12, 2006. After I
submitted the Noll Merits Report, Apple produced information that only some iPods that
were sold after September 12, 2006, used iTunes 7.0. In the damages equations that are
reported in this declaration, the indicator variable for iTunes 7.0 is set equal to one only
for iPod models that used iTunes 7.0 or its successors. The variable for prior versions of
iTunes that blocked Harmony, starting with iTunes 4.7, is now set to one after September
12, 2006, for iPods that did not use iTunes 7.0. I disagree with defendant’s economic
experts that the variable for software that disabled the first version of Harmony should
remain set to one for iPod models that used iTunes 7.0 and its successors. These iPods
did not contain both software programs. Apple chose to replace software that the court
previously found to be legal with software that plaintiffs now contend was illegal. The
regression equations that I use measure the effect of that change.
Second, Professor Murphy and Topel argue that, as a matter of economic theory,
the effect of iTunes 7.0 on the prices of iPods would not occur immediately when iTunes
7.0 was introduced, but would occur gradually as customers who purchased these models
became increasingly locked-in to Apple’s technology. While the extent of lock-in does
increase as a consumer buys more recordings that can be played only on iPods, this fact
does not imply that, as a theoretical matter, the price differential would increase through
time. The reason is that once iTunes 7.0 is launched, Apple no longer can compete for
customers who have used a portable digital media player that plays downloads in the
RealNetworks format, and so no longer has an incentive to try to attract those customers
by cutting the prices of iPods that use software that effectively blocks Harmony.
7
Third, Professors Murphy and Topel claim that the regression is unreliable
because it does not include every technical characteristic that differentiates models of
iPods. This argument is incorrect. The goal of a hedonic price regression is to explain as
much of the variance in prices of differentiated products as possible on the basis of the
characteristics of each product. To achieve this goal requires adding variables that
measure characteristics that differ among products up to the point that adding more such
variables does not contribute significantly to explaining price differences. The reason
that adding more characteristic variables to the price equations for iPods does not cause a
meaningful increase in the explanatory power of the regression is that the variables for
iPod class, capacity and cost that already are included in the regression already measure
the effects of technical characteristics that are peculiar to a particular model. If the
additional technical characteristics that Professors Murphy and Topel discuss are added
to the equation, the effect is to cause extreme multicollinearity (i.e., a high correlation
between separate independent variables in the equation). High multicollinearity causes a
reduction in the efficiency of the estimates of the coefficients in the regression equation,
making both the point estimates and the standard errors of the regression coefficients
unreliable. Because no issue in this litigation hinges on monetizing the value of any
particular technical feature of an iPod, nothing is gained by adding more characteristics
that do not increase the explanatory power of the regression but reduce the precision of
the estimates of the coefficients.
Fourth, Professors Murphy and Topel argue that my regression analysis suffers
from non-independence of the observations that arise from clustering effects.
8
The solution to this alleged problem that
Professors Murphy and Topel propose is to collapse the transactions data by using only
one observation for each specific price that Apple charges for a given iPod model during
a calendar quarter and then weighting that observation by the quantity of that iPod model
that was sold in that quarter at that price. At every step of the way, this argument by
defendant’s experts is incorrect as matters of fact and econometric theory.
Defendant’s experts are incorrect to claim that each transaction is not an
independent observation because Apple sets all prices. The statistical concept of
independence is not about whether different people set the price for different purchases,
but about the distribution of the error term in the price equation. For example, all of the
laws of physics are set by nature, but in an appropriately designed experiment separate
observations of natural phenomena that are collected for the purpose of quantifying these
laws are independent observations. Notwithstanding that Apple sets the price of each
iPod, each Apple customer makes a separate, independent decision whether to buy that
iPod, some other iPod, some other brand of portable digital media player, or no portable
digital media player at all, and, if the decision is to buy a particular iPod, how many to
purchase. Thus, each transaction record is analogous to a single experimental trial to
measure a law of nature, and so is properly regarded as an independent event.
9
Clustering problems arise
because of the way that a sample of observations is drawn from a larger population. A
cluster sample consists of a data-gathering technique that divides all observations into
groups, selects a sample of these groups from which observations will be drawn, and
then, for each sampled group, takes a sample of observations within the sampled groups.
The issues that give rise to a cluster problem are: (1) observations from a sample of
groups may not be representative of observations in all groups, and (2) membership in a
group may be correlated with other factors that could influence the dependent variable in
the regression but that are not included as independent variables in the regression.
Clustering does not apply to Apple’s transactions data because these data are not a
sample, let alone a clustered sample. Apple’s transactions data include the entire
population of all sales by Apple, not a sample of Apple’s sales for some models of iPods.
Moreover, even if the data were a sample of sales from a sample of iPod models, as the
number of clusters increases (here, the number of iPod models and calendar quarters), the
sampling error from clustering converges to zero. In Apple’s transactions data the
number of iPod models per calendar quarter is large enough to eliminate a concern about
problems from clustering.
A standard procedure for reducing cluster effects is to include indicator variables
10
for clusters. This procedure does not work if the number of clusters is small (i.e., a
handful), but does work if the number of clusters is large. The regression equations that I
use include a large number of indicator variables for iPod classes and capacity, so even if
the data were a cluster sample the clustering problem would not be important. Finally,
even if the number of clusters is small and observations are from a cluster sample, the
proposal to divide sales of iPod models into calendar quarters does not solve the problem.
Dividing the observations in each cluster among calendar quarters does not make the
sample more representative and does not correct for correlations between the identity of a
cluster and excluded independent variables.
For these reasons the criticisms by Professors Murphy and Topel regarding the
alleged lack of independence of price observations and the presence of cluster effects are
without merit. The sole effect of their proposal is vastly to reduce the number of
transactions observations that are used to estimate the regression equation, thereby
reducing the explanatory power of the regression analysis by eliminating most of the true
variance in price among transactions and, thereby, destroying the reliability and precision
of the regression results.
Recalculated Damages
Apple produced new information after the submission of the Noll Corrections
Report about the models of iPods that made use of iTunes 7.0 and its successors. I have
re-estimated the damages model making the following changes in the specification and
estimation procedures. First, the indicator variable for iTunes 7.0 takes the value one
only for models of iPods that used this version of iTunes, and not for other iPods that
11
were sold after September 12, 2006, when iTunes 7.0 was launched. Second, because
older versions of iTunes, beginning with iTunes 4.7, that used previous methods for
blocking Harmony continued to be used on other iPod models, the “Harmony Blocked”
indicator variable continues to take the value one for these models until the end of the
data period. Third, to take into account the fact that RealNetworks introduced a second
version of Harmony that worked around the disablement of the first version of Harmony
by iTunes 4.7, I added a second indicator variable that takes the value of one for all
transactions after the second version of Harmony was released. Fourth, I have adopted
January 6, 2009, instead of March 31, 2009, as the date on which the “DRM-free”
variable takes the value of one and for which damages end. Fifth, I have replaced the
logarithm of time in the regression by the scalar measure of time. Sixth, I have estimated
the equation using quantity-weighting, rather than frequency-weighting.
12
LIABILITY ISSUES
Professor Murphy argues that the analysis of liability in the Noll Merits Report is
implausible because it does not deal with several issues that Professor Murphy believes
are important. On all counts Professor Murphy’s arguments are not based on economic
analysis and are incorrect.
Superiority of Closed Systems
According to Professor Murphy, my analysis ignores “pro-competitive reasons”
(Murphy Report, p. 4) for creating closed systems. I do not ignore the fact that, in
principle, a closed system may perform better than a system in which the components are
acquired from separate vendors. Regardless of whether an iPod actually performs
downloads from iTS better than downloads from the RealPlayer Music Store (RMS) or
Amazon is irrelevant.2 The relevant issue is not whether Apple is correct in its beliefs
about the superiority of all-things Apple, but whether forcing consumers to buy either all
or no complementary products from Apple is a reasonable restriction on competition.
In the absence of actions to disable the use of competing products, consumers can
judge for themselves whether Apple’s “walled garden” is superior, rather than delegate
this evaluation to Apple. To prove that this restriction is reasonable requires showing
2
The Murphy Report (pp. 19-31) contains a discussion of Apple’s closed system, but
does not contain any information that is relevant to the issue of whether a download from
one of the competitors of iTS is in any way inferior to a download from iTS. Moreover,
this discussion ignores two facts: before the iTunes Music Store was launched in April
2003, iPods loaded music from CDs, which are not in a protected format and can be
loaded to a portable digital media player, and since January 2009 downloads have been
sold without DRM protection from all Internet vendors, without doing any harm to iPods
or iPod users. Thus, there is no evidence for the premise of Professor Murphy’s
argument that closed systems perform better than systems with components from
different vendors.
13
that the performance of an iPod is degraded if it tries to play a download from a vendor
other than iTS and that consumers are harmed if they are allowed to make their own
decisions about where to acquire downloads. Professor Murphy offers no economic
evidence to support either of these arguments.
Extent of Use of Harmony
Professor Murphy asserts that proof of liability must involve showing that iPod
owners used Harmony and purchased downloads from the RMS (Murphy Report, pp. 4,
33). This claim also is incorrect. The economic argument about the effects of changes in
the market conditions for portable digital media players is that they affected the relative
prices of iPods. If Apple lowered iPod prices due to Harmony and raised these prices
after Harmony was blocked, the purpose would have been to minimize the extent to
which consumers changed their pattern of consumption among both portable digital
media players and sources of downloads of sound recordings.
Nothing in the economic theory of demand indicates, as Professor Murphy
implies (Murphy Report, p. 6), that some threshold level of users of Harmony among
iPod owners is necessary for Harmony to have had an effect on iPod prices. Instead, the
effect of a technology that allows consumers to buy complementary products from
different vendors is to make the price of these components more elastic, and the effect of
conduct that eliminates this possibility is to make demand for components less elastic. If
Apple managed to avoid most or even all of a switch in sales arising from a technology
such as Harmony by lowering prices and improving its products, one could observe little
or no actual use of the technology because it was discouraged by the price effect.
14
Double Marginalization
Third, the Murphy Report (p. 21) argues that the sale of complementary products
by a single seller necessarily lowers the prices of components because the single seller
can take into account the effect of the price of one component on the sale of another.
This argument is based on the premise that each component is sold by a firm with
monopoly power. Elsewhere Professor Murphy notes that RMS accounts for a few
percent of download sales, which is inconsistent with the premise that RealNetworks
possesses significant market power. Professor Murphy’s argument is incorrect if the
markets for complementary products are competitive. Moreover, if Professor Murphy’s
argument were correct, a firm that sold all of the complementary products would not need
to make components incompatible with complementary products from other suppliers
because it would obtain all sales without creating lock-in by charging lower prices.
Market Definition and Market Power
The Murphy Report (pp. 6, 59-63) makes several arguments challenging the
analysis of market definition and market power in the Noll Merits Report.
According to Professor Murphy, a digital download of sound recordings “is a
‘market’ without meaning” because the “central question” is whether iPod users are
locked in to iPods and RMS cannot possibly be important in determining whether iPod
users are locked in because it has “a very small share of any music on an iPod” (Murphy
Report, p. 6). Here Professor Murphy confuses the issue of defining a market with the
issue of whether Apple’s conduct reduced competition in that market. The low market
15
share of RMS, which entered after the iTunes Music Store (iTMS) was launched, is the
expected consequence of lock-in by the dominant provider of portable digital media
players at the time that RMS entered the market. The Murphy Report (p. 61) also states
that the definition of the market in which iTS operates is “irrelevant” because “Plaintiffs
have not alleged that Apple attempted to increase prices for digital downloads above the
competitive level…” Whether Apple’s pricing of digital downloads is monopolistic or
competitive is not a relevant issue in this case. Here Professor Murphy ignores that
market power is the ability of a firm to cause prices to exceed the competitive level or to
exclude competitors. The latter is the relevant issue in this case, and the definition of the
relevant market in which iTS competes is relevant to determining whether Apple has
excluded competitors.
Professor Murphy claims that all forms of obtaining music, including physical
copies (CDs), on-demand Internet streaming, and illegal file sharing should be in the
relevant market that includes iTS. Market definition is about identifying the close
competitive substitutes for a reference product (here, iPods and digital downloads).
Professor Murphy does not show that CDs, illegal file sharing, and on-demand streaming
services are equally close substitutes for iTS as a competing source of permanent
downloads. The Noll Merits Report (pp. 31-42) explains why these other sources of
recordings are not as close substitutes for iTS as other Internet sites that sell permanent
downloads of recordings. Briefly, the success of sites like iTS, RMS, Amazon and others
demonstrates that many consumers do not want to obtain recordings in a way that
infringes copyrights. Notwithstanding Professor Murphy’s dismissal of the point, ondemand streaming services for play on portable devices were not important until after the
16
end of the class period, so whether they were in the relevant market is irrelevant to the
issue of whether Apple enjoyed market power in digital downloads. Finally, CDs are not
the closest substitutes for digital downloads because they are bundles of many recordings
in a pre-programmed format.
The Murphy Report (pp. 61-62) also disputes my analysis of the market power
that Apple enjoys in the market for portable digital media players. According to
Professor Murphy, lock-in played no role in the success of iPods, based on the
observation that “Apple achieved great success and was able to charge a price premium
even before it was established as the market leader.” Here Professor Murphy seems to
agree that Apple enjoyed market power in portable digital media players, and instead is
disagreeing with the conclusion that Apple’s market power increased with the launch of
iTMS and its resulting lock-in effect. The “great success” with iPods before iTMS
involved obtaining a far lower share of a far smaller market for portable digital media
players than existed after iTMS and other digital download sites were launched.
In summary, Professor Murphy’s opinions on market definition are not based on
an application of the methods that economists use to define a relevant market, and his
analysis of market power does not deal systematically with the relevant facts that an
economist would use to determine whether a seller has market power and, if so, how its
17
market power has changed over time.
Simultaneous Improvements
The Murphy Report (p. 6) argues that the iTunes 7.0 variable measures more than
just the effect of disabling Harmony, but also other technical features that occurred
simultaneously.
ECONOMETRIC ISSUES
Professors Murphy and Topel assert that the econometric model that I use to
demonstrate anticompetitive injury and to calculate damages on a class-wide basis is
unreliable for three reasons: (1) the use of frequency weights on each transaction record;
(2) the details of the specification of the regression equation; and (3) the failure to take
into account an alleged clustering of the data that causes transaction observations not to
satisfy the assumption that observations in the data set are independent. Professors
Murphy and Topel further claim that if the model is adjusted to take their criticisms into
account, the estimated effect of iTunes 7.0 on iPod prices is zero, implying that
18
consumers were not damaged by disabling Harmony.
Weighting
The Murphy Report (p. 47) and the Topel Report (pp. 25-27) criticize the use of
frequency weighting of observations in the regression model on the grounds that it causes
the precision of the regression coefficients to be over-estimated. As Professors Murphy
and Topel state, the price of a specific iPod (model/generation/family) in each transaction
is calculated as the ratio of the total charge for that specific iPod divided by the number
of units of that iPod that were purchased. Thus, the unit of observation in the regression
analysis is a transaction that includes this calculated price and values for the other
independent variables, including the indicator variables for class and capacity, variables
representing the sales quantity in the transaction, and variables indicating technical
features of iPods. The sales quantity was included as an independent variable to take into
account the possibility of volume discounts.
The core of the weighting issue is as follows. The procedure for estimating a
19
regression equation calculates the values of the coefficients on the independent variables
to minimize the sum of the squares of the prediction errors from the estimated equation.
That is, if the true price in a transaction is P and the estimated price from the regression
equation is P*, the method for estimating the regression calculates coefficients that
minimize the sum of (P* - P)2 over all transaction observations.
The most commonly used procedure for estimating an econometric model is an
unweighted regression, which means that each observation is given equal weight in
calculating the coefficients of the variables in the regression equation. Thus, a
transaction in which 20,000 units of a specific iPod are sold is given the same weight as a
transaction in which one unit of the same iPod is sold. If PH is the price of the transaction
involving 20,000 units and PL is the price of the transaction involving one unit, a
regression in which there were N transactions of one unit and one transaction for 20,000
units would minimize (PH* - PH) + N(PL* - PL). For values of N that are greater than one,
an unweighted regression produces a more precise estimate of PL* than of PH*.
The problem with using this procedure is that it yields a biased estimate of total
damages because it does not take into account the fact that far more damages are at stake
in obtaining an accurate estimate of damages for the transaction that involves 20,000
units than in accurately estimating damages for the one-unit transactions. In the
preceding example, if there are 100 transactions for one unit, an error of ten cents in the
estimate of PL produces an error of $10 in total damages, but the same error in the
estimate of PH causes an error in damages of $2,000.
The standard solution to this problem is to weight each observation by the amount
transacted. That is, in the preceding example, instead of minimizing the unweighted sum
20
(PH* - PH) + 100(PL* - PL), the procedure minimizes (20,000)(PH* - PH) + 100(PL* - PL).3
Notice that an unweighted regression assigns most of the weight to the small transactions,
while a weighted regression assigns most of the weight to large transactions.
The Murphy Report and the Topel Report do not criticize my decision to estimate
a weighted regression. Instead, their criticism is about the use of frequency weighting
rather than quantity weighting. Frequency weighting causes the “degrees of freedom”4 in
a regression to be overstated, which in turn causes an under-estimate of the standard
errors of the regression coefficients.
I agree that, as a general proposition, quantity weights are preferred to frequency
weights. But the choice between these weighting methods is not important in this
circumstance for two reasons. First, the choice between frequency weights and quantity
weights has no effect on the estimated coefficients of the independent variables and, as a
result, the amount of damages. The only statistics that can be affected by the choice of
weights are the standard errors of the coefficients and the measure of fit of the equation,
which is adjusted R2. Second, when the number of observations is extremely large (here,
in the millions), the effect of the choice of a weighting method on the outcome of tests of
statistical significance of the regression coefficients and of adjusted R2 is small. Once the
degrees of freedom in a regression are in the millions, a further increase is unlikely to
3
Typically each weight is divided by a constant, such as the total quantity sold in all
transactions. While this procedure makes the coefficients in the regression easier to
interpret, it does not affect the fit of the equation, the statistical significance of the
coefficients, or the amount of damages.
4
Usually the statistic “degrees of freedom” equals the number of observations minus the
number of independent variables in the equation. In both data sets (direct and reseller)
the number of observations is in the millions and the number of independent variables is
less than 75, so the “degrees of freedom” statistic is in the millions. But if observations
in the data are not statistically independent, the degrees of freedom statistic is adjusted
downwards to reflect the extent of correlation among the errors in the observations.
21
have much of an effect on the outcome of tests of statistical significance for the
regression coefficients or the equation as a whole.
The effect of the weighting method also is greater if the
data include “outlier” observations that are likely to be incorrect and, therefore, to create
more unexplained variation in prices.
To illustrate that the choice of the weighting method has a minimal effect on the
regression results, I estimated two identical versions of the original logarithmic
regression equation in the Noll Merits Report, one with frequency weights and the other
with quantity weights, using reseller data that includes outlier observations. These
specifications do not make any of the changes in the regression equations that are
discussed elsewhere in this declaration and that I have adopted in the new damages model
and so directly address the claim by defendant’s experts that the use of frequency weights
dramatically affected the results of my regression analysis.
The effect of the choice of a weighting procedure in the reseller regression is
reported in Exhibit 1. As expected from statistical theory, the regression coefficients are
identically the same in the frequency weighted and quantity-weighted equations, as is
adjusted R2. The standard errors of the regression coefficients are larger in the quantityweighted regression, but these differences are generally small compared to the values of
the coefficients. The coefficients on the variables that are relevant for establishing
22
liability and calculating damages5 all remain highly significant.6 Exhibit 2 shows the
results of switching from frequency weights to quantity weights for the direct sales
regression. As expected, the differences in these results are limited to small differences
in the standard errors of the coefficients, with all other results unchanged, including the
coefficient on the iTunes 7.0 variable that is used to calculate the amount of damages.
These results show that, despite the advantages of quantity weights compared to
frequency weights, the choice of a weighting method does not affect any conclusion in
my previous reports. Nevertheless, in the new damages regressions reported elsewhere, I
use quantity weighting rather than frequency weighting.
Specification Issues
Economists use the term “specification” to refer to the choice of a functional form
of a regression equation and the choice of the independent variables that are included in
the equation. Professors Murphy and Topel make five criticisms of the specification of
the regression equations that are estimated in the Noll Merits Report.
The first criticism is that time should be included in a logarithmic equation as a
linear variable, rather than a logarithmic variable (Murphy Report, pp. 52-53, and Topel
Report, pp. 43-46). The second criticism (Murphy Report, pp. 51-52, and Topel Report,
pp. 37-43) is that the indicator variable for the iTunes 4.7 and subsequent versions of
iTunes that blocked the use of Harmony but that the Court has found to be legal should
5
These are the indicator variables for iTS Launched, Harmony, Harmony Blocked
(iTunes 4.7), iTunes 7.0, Competitors DRM-free, and iTS DRM-free.
6
Two independent variables, the smaller quantity indicator and the indicator for 6GB of
memory, are statistically significant in the frequency-weighted regression and statistically
insignificant in the quantity weighted regression, but no conclusion in my prior reports
hinges on the significance of these variables.
23
remain “on” to the end of the data period, rather than be replaced by iTunes 7.0 when the
latter is released. The third criticism (Topel Report, pp. 20-21) is that the regression
equation should include separate indicator variables for the two versions of Harmony that
were released by RealNetworks. The fourth criticism (Topel Report, pp. 53-54) is that
my original specification used the wrong date for the start of the availability of DRM-free
music on iTS. The final criticism is that the regression equation does not include
measures of all possible technical characteristics of an iPod and other variables that might
affect its price (Murphy Report, pp. 53-56, and Topel Report, pp. 47-49).
This section explains the issues that lie behind these criticisms. In three cases (the
specification of time, the beginning date for DRM-free music, and the two versions of
Harmony), I use the procedures that Professors Murphy and Topel recommend and find
that they do not cause the coefficient on the iTunes 7.0 variable to become statistically
insignificant and the amount of damages to become zero. In the other cases, I conclude
that the criticisms by Professors Murphy and Topel are not valid.
Time
According to the Murphy Report (p. 53) and the Topel Report (p. 46), time in a
logarithmic model should enter as a scalar, not the logarithm of time. In the revised
damages regressions reported elsewhere, I use time, rather than the logarithm of time, in
the logarithmic price regression.
24
iTunes 4.7 and Harmony
The damages equations that were estimated in the Noll Merits Report included a
variable, “Harmony,” to indicate the availability of the Harmony software to enable iPod
users to download music from Internet retailers that sold music in RealNetworks’ DRM
format. These equations included two variables to measure versions of the iTunes digital
media player software that disabled Harmony. The first, labeled “Harmony Blocked,”
took the value of one after iTunes 4.7 was introduced and was switched to zero when
iTunes 7.0 was introduced. The second, labeled “iTunes 7.0,” took the value of one from
the time that iTunes 7.0 was introduced until the end of the data period. The Murphy
Report (pp. 51-52) and the Topel Report (pp. 37-43) argue that this specification is
incorrect and should be re-specified in two ways.
First, Professors Murphy and Topel propose that the equation should include two
indicators for Harmony – one for the initial version, and another for a later version that to
some degree got around the software in iTunes 4.7 that blocked the use of the first
version of Harmony. I accept this proposal and have separated the Harmony indicator
into “Harmony1” that takes the value of one after July 2004 and “Harmony2” that takes
the value of one after April 2005.
The second proposal by defendant’s experts is that “Harmony Blocked” should be
set equal to one until the end of the data period, thereby causing the iTunes 7.0 variable
to measure the incremental harm arising from the replacement of earlier versions of
iTunes that the Court had ruled to be legal. The context of this criticism is that in the
regressions in the Noll Merits Report, Harmony was assumed to have been disabled on all
models of iPods that were sold after September 12, 2006. The relevance of this criticism
25
has been affected by the fact that after the Noll Corrections Report was submitted, Apple
provided new information about
in iTunes 7.0 and its successors were enabled.
continued to use the methods for blocking Harmony
that were included in iTunes 4.7 and that had been worked around in the second version
of Harmony. In the revised regressions the “Harmony Blocked” indicator remains set
equal to one until the end of the data period for the models that were not affected by
iTunes 7.0 and is set equal to zero for the iPod models for which the
were enabled.
I expect that Professors Murphy and Topel continue to think that “Harmony
Blocked” should be set to one for iPod models on which the
were enabled, thereby
blocking Harmony. I believe that this recommendation is unwarranted. For these models
of iPods, Apple replaced technology that was ruled by the Court to be legal with new
technology that plaintiffs allege was not legal. The proper way to measure the impact of
enabling the
and using iTunes 7.0 is not to assume that these models contained the
old blocking technology from iTunes 4.7, which is the implicit assumption in the
proposal by Professors Murphy and Topel.
The Impact of iTunes 7.0
The Murphy Report and the Topel Report (p. 55) claim that the specification of
the iTunes 7.0 variable is inappropriate for measuring the anticompetitive effect of
blocking Harmony on the prices of iPods. Both argue that the lock-in effect arising from
preventing iPod owners from downloading recordings from websites that used the
RealNetworks encryption format would have become quantitatively more significant as
26
customers bought more recordings from Apple, and conclude from this argument that the
effect of iTunes 7.0 on iPod prices should have been gradually to increase them. The
problem with this argument is that it focuses on only one way in which the creation of a
mandatory closed system of products reduces competition and raises prices.
This argument ignores the fact that iTunes 7.0 immediately locked out a customer
who has been using a portable digital media player that used the RealNetworks DRM
format and who had purchased downloads from RMS. The immediate effect of iTunes
7.0 was to reduce the expected benefit to Apple in increased iPod sales from engaging in
price competition with these other portable digital media players, thereby causing the
demand for iPods to be less price elastic. The effect of lock-out is immediate, and the
principal anticompetitive effect of blocking Harmony on iPod prices arises from the
general reduction in competition among iPods and other portable digital media players
arising from the lock-out effect.7
The argument by defendant’s experts also is incorrect for another reason.
Whereas new iPod owners in late 2006 became more locked in to iPods over time, the
effect of this lock-in would not have affected the demand for subsequent iPods for a long
period because these purchasers would not soon make repurchase decisions. Thus, for
most of the damages period, the lock-in effect on new iPod purchasers would not be an
important factor affecting iPod prices.
7
The Murphy Report (p. 37) argues that lock-in makes an iPod less valuable and so, as a
matter of economic theory, reduce its prices. This statement is not true and reveals a lack
of understanding of the economic of lock-in effects. Lock-in makes otherwise identical
products less perfect substitutes, and thereby reduces the intensity of price competition
among these products.
27
DRM-Free Music
The Topel Report (pp. 53-54) criticizes my use of the date of March 31, 2009, as
the starting point for the availability of DRM-free music on the iTunes Store (iTS). The
basis for this criticism is that Apple announced in January 2009 that as of that date 80
percent of the recordings on iTS were available in a DRM-free format. Professor Topel
reports that he reset the start date of the DRM-free music variable “three months earlier,
as of January 6, 2009” (Topel Report, p. 54) and that this change reduced the effect of
iTunes 7.0.
The relevant variable for measuring the effect of the availability of DRM-free
downloads on iTS is not the fraction of titles that were available in a DRM-free format,
but the popularity of those titles. The impact of the catalog of DRM-free music could be
approximated by the relative sales of DRM-free music versus other music on, say, a
weekly basis. But these data were not produced by Apple in this litigation and so cannot
be put into the regression equation. Without such data, any specification of the effect of
DRM is going to be partially incorrect. For example, the implicit assumption that the full
effect of DRM-free music occurred immediately on January 6, 2009, is clearly incorrect.
The importance of the error in specifying the DRM-free start date as January 6 depends
on the popularity of the titles that were not available in a DRM-free format on that date.
In the revised damages regressions reported elsewhere, I have used January 6,
2009, rather than March 31, 2009, as the beginning of the DRM-free period on iTS. The
effect of this change is to cause true damages to be underestimated because not all
recordings were available in a DRM-free format on that date. The likely effect of this
change is a small reduction in the amount of damages. This effect is likely to be small
28
because the duration of the period being reclassified is short.
Excluded Variables
Professors Murphy and Topel argue that the damages equations that I estimated in
my prior reports have low quality. One reason they give is that some coefficients appear
to have the “wrong” sign, and the other is that some variables are excluded. For example,
the Topel Report (p. 47) observes that photo capability and photo and video capability
both have negative regression coefficients, which Professor Topel interprets as causing
the price of an iPod with these features to be lower than if these features had been left
out. According to the Topel Report (p. 47), “there can be little doubt that these features
are valuable to users.” Professor Topel then adds several new variables that measure
technical characteristics of iPods to the regressions (Topel Report, pp. 48-49, Exhibits
13A and 13B). A related criticism (Murphy Report, p. 55) is that the equations fail to
take into account the profit margin of products, which Apple’s committee for setting
prices states is an important element of decisions about prices.
These criticisms are incorrect for the same reason. The average incremental cost
of an iPod model is included in the regression.
Professor Murphy’s proposal to add profit margin to the regression equation is
entirely inappropriate. The cost of an iPod model is the basis for setting its profit margin.
The cost variable that already is included in the regression incorporates the consideration
of margins by Apple in setting prices. Adding margin as another independent variable,
far from improving matters, would constitute a fatal econometric error because it would
cause the dependent variable, price, to appear on both sides of the equation. If P is price,
29
C is incremental cost, and A, b and c are coefficients to be estimated in the regression,
then the regression equation is P = A + b(P – C) + cC = A + bP + (c-b)C. In this
meaningless regression, the estimate for b would be one, while A and (c-b) would be
zero, and the equation would prove precisely nothing.
Professors Murphy and Topel’s argument that many technical characteristics of an
iPod plausibly affect its value and so should be included in the hedonic price equation
also is not useful. The problem with this claim is that it loses sight of the reason for
estimating the hedonic price equation. The goal of the hedonic price equation is to
explain as much of the variation in price as possible, using measures of product features.
One important variable that captures a bundle of iPod characteristics is the iPod class.
Thus, if variables for all of the specific technical characteristics that distinguish an iPod
touch from an iPod nano were added to the equation, these variables would not add
explanatory power because they already are captured in the class indicator variables.
Professor Topel finds that adding additional technical characteristics has an
extremely small effect on the explanatory power of the regressions that I reported,8 and
8
When Professor Topel adds more variables that measure technical characteristics to the
regression equations, R2 increases by .0063 in the reseller equation and by .0053 in the
direct sales regression.
30
claims that this justifies including them.9 But adding more variables that have a minimal
effect on the explanatory power of a regression is worse than simply being unnecessary.
If some variables that are sources of value in an iPod add little explanatory power when
they are included in the regression equation, the reason is that they already are highly
correlated – if not perfectly correlated – with variables that already are in the equation.
For example, the indicator variable for an iPod nano is perfectly correlated with any
indicator variable for a feature that is found only in an iPod nano. The term that
econometricians use for including highly correlated variables in the same regression is
multicollinearity, and perfect multicollinearity of this form makes estimating a regression
equation impossible.
The problem that multicollinearity creates is to reduce the precision of the
estimated coefficients in the regression, making them less reliable as estimates of the true
effect of an independent variable on the dependent variable. Thus, the effect of piling on
more iPod characteristics that add no explanatory power to the hedonic regression is to
cause all of the estimated coefficients, including the coefficient on iTunes 7.0, to become
less reliably estimated.
If the goal of a regression were to estimate the contribution of a particular
technical characteristic to the price of an iPod, then adding highly correlated variables
might make sense if the loss of precision in estimating the coefficients were not so great
that many, if not all, coefficients became statistically insignificantly different from zero.
But the goal here is not to determine the implicit value of the attributes of an iPod, but to
produce a formula for calculating damages. The latter does not depend on the values of
9
Deposition of Robert Topel, pp. 149-50.
31
the coefficients on technical characteristics, but the reliability and precision of the
coefficient on the iTunes 7.0 indicator.
Independence and “Cluster Effects”
Both the Murphy Report (pp. 45-50) and the Topel Report (pp. 33-37) argue that
the methods that I used to analyze Apple’s transactions records are based on an erroneous
assumption that the transactions observations are independent when, in fact, these
observations occur in “clusters” of correlated observations. According to Professor
Murphy (Murphy Report, p. 46), I “made a fundamental error” in constructing my data
set and interpreting the regression results because I assumed that “literally millions of
identical price ‘observations’ are statistically independent.” The Murphy Report (p. 48)
also states that the “prices paid by purchasers (either individual consumers or resellers)…
were all set by Apple…” and “the vast majority of the prices for the same model in the
same time period were identical.” Professor Murphy then asserts that these facts cause
the observations to violate the independence assumption. Professors Murphy and Topel
claim (Murphy Report, p. 49, and Topel Report, pp. 35-36) that the residual errors in the
prices of each iPod model in each calendar quarter are correlated, by which they mean
that the average residual error in each quarter is not zero. From this result, Professors
Murphy and Topel conclude (Murphy Report, pp. 49-50, and Topel Report, pp. 36-37)
that the data are “clustered” according to the family of the iPod and that the proper
method for analyzing these data is to collapse all transactions at a given price for each
class/generation/family of iPod in a calendar quarter into a single observation.
At every step of the way this series of arguments is incorrect.
32
“Millions of Identical Prices”
Apple and its experts repeatedly have claimed that all Apple customers pay the
same price. This claim is untrue and is easy to disprove using Apple’s transactions data.
Appendix B contains histograms of the price and quantity data for every iPod family for
every calendar quarter during the entire data period. The reseller and direct purchaser
data are shown separately. Each figure shows the distribution of the prices charged and
the number of transactions at that price for each calendar quarter. The reseller
transactions data show substantial within-quarter variation in prices. The direct sales data
show that most transactions are at list price, but many are not. Thus, a major premise of
the arguments put forth by Professors Murphy and Topel – that all or virtually all prices
of a given iPod model are identical – is wildly incorrect. Moreover, these data understate
the true variation in prices because Apple has not provided the information that is
necessary to match all discounts to a specific transaction.
The Independence of Residual Errors
The Murphy Report and the Topel Report argue that the transactions observations
that were used to estimate the regression equations contain “clusters” of observations,
defined as sales of a single iPod class/generation/family during a specific calendar
quarter, in which the residual errors of the regression equation are correlated within each
cluster, and that as a result the regressions that I report are unreliable because they do not
include a correction for this cluster effect. These claims are incorrect, partly because the
statistical analysis that Professors Murphy and Topel use to demonstrate within-cluster
33
correlation is incorrect, and partly because applying the concept of cluster analysis to the
Apple transactions data is inappropriate as a matter of econometric theory.
In the empirical analysis that Professors Murphy and Topel present to prove that
errors are correlated within a cluster, Professors Murphy and Topel calculate the residual
error of the estimated price for each sale of an iPod within a cluster, divide the sample of
residual errors into two equal size groups, and show that the mean error in the two groups
is correlated. This analysis does not show that residual errors are correlated. Instead, this
analysis shows that the two samples generate similar estimates of the mean residual error,
which is to be expected if the samples are drawn randomly. Professors Murphy and
Topel do not test whether the mean residual errors from this procedure are statistically
significantly different from zero, which would have to be the case if the errors within a
cluster are correlated. Thus, the analysis of residual errors by Professors Murphy and
Topel does not actually test whether residual errors are correlated within a cluster.
Another problem with the procedure that Professor Murphy and Topel adopt is
that it is based on the erroneous assumption that the proper unit of analysis is the sale of
one iPod, not a transaction that may involve the sale of many iPods. The procedure that
is used by Professors Murphy and Topel to test for correlation of residual errors
calculates the residual error for each unit sold, which means that in the Best Buy
examples cited above, Professors Topel and Murphy will count the same residual error in
each of these transactions about 20,000 times. Of course, all of these 20,000 residual
errors will be the same, so if 10,000 are assigned to each of the two groups of residual
errors for each cluster, the mean values of the residual errors for each group will be
correlated simply because the same observation has been counted so many times in both
34
groups within the cluster.
My regression analysis is based on the assumption that the proper unit of analysis
is a transaction and that the error in the price equation is independently distributed across
transactions, not across each unit sold in a multi-unit transaction. Although I weight a
price observation in a transaction by the number of units of a given iPod that are sold in
order to better fit the prices for multi-unit transactions, the appropriate residual error for a
transaction is a single observation on the difference between estimated and actual price
for the transaction, not for each unit in the transaction. Thus, the appropriate procedure
to test for within-cluster correlation of errors is to examine the residual errors of
transactions, not the residual errors of the price of each unit sold.
As a practical matter, a particular model of iPod is represented by class and
capacity variables in the regression. One consequence of this procedure is that the mean
residual error for a specific class or capacity of iPod over the entire period in which it
was sold must be zero. This does not mean that the mean residual error is necessarily
zero in any specific calendar quarter – indeed, the probability that every quarter will have
exactly the same mean residual error is zero, even without any specification problem in
the equation. But Professors Murphy and Topel do not even attempt to determine
whether mean residual errors for transaction prices actually vary from quarter to quarter
for a given model of iPod and, if so, whether the differences are statistically significant or
quantitatively important.
35
The Relation between Independent Observations and Clustering
Professors Murphy and Topel discuss the independence of observations as if the
concept of independence applies to whether the price differs among observations. In fact
independence refers to the error term in the regression equation. Professors Murphy and
Topel are incorrect to state that two transactions involving the same product and the same
seller at exactly the same price necessarily are not independent. Even for sales at the
same posted price, the observations are independent because each transaction involves a
different buyer making an independent decision about whether to make a purchase and, if
so, how many products to buy. The price equations that are used to calculate damages
include variables that can affect supply (Apple) and demand (customer) decisions. Thus,
counting each transaction as a separate, independent observation is appropriate.10
Even using the same observation more than once does not cause the independence
assumption to be violated and can be statistically useful. A technique for dealing with
small data sets is “bootstrapping,” which is used in correcting some clustering
problems.11 The simple bootstrap method involves taking a random sample with
replacement from the original data (i.e., randomly drawing an observation, recording it,
10
Economists frequently use data sets with many transactions at the same price to study
supply and demand conditions in a market. For example, a study of grocery transactions
used all recorded retail sales to calculate the extent to which prices were correlated
among stores and the extent to which retail price changes were in response to wholesale
price changes. On a given day, the price of any specific grocery item was likely to be
identical in all sales, but differences in the identity of the purchaser and the quantity
purchased made each transaction a valid observation. See Emi Nakamura, “Pass-through
in Retail and Wholesale,” American Economic Review Papers and Proceedings, Vol. 98,
No. 2 (May 2008), pp. 430-37.
11
A. Colin Cameron and Douglas L. Miller, “A Practitioner’s Guide to Cluster-Robust
Inferences,” Journal of Human Resources (forthcoming), pp. 14-15.
36
and returning the observation to the data set before the next random draw).12 Sampling
with replacement implies that the same observation may be drawn more than once,
causing duplication of observations. A regression is estimated from this sample. This
procedure is repeated many times to produce many estimates of the regression
coefficients. The distribution of the regression coefficients is then used to construct an
estimate of the true coefficient.
If a valid statistical procedure that does not violate the independence assumption
can involve using the same observation more than once, the fact that the prices in two
distinct transactions are the same or were set by the same person cannot, by itself, violate
the independence assumption. Consider the problem of measuring the true acceleration
of a falling body due to gravity. Just as Apple determines every price and all of the
technical features of every iPod, so does nature set the true values for V, g and t in the
equation V = gt, where V is velocity, t is time, and g is the acceleration due to gravity of
a freely falling body. One can perform an experiment in which many observations are
made of V and t, and from these measurements one can estimate g. The fact that every
measurement is determined by exactly the same law of physics does not cause these
observations to violate the independence assumption.
In estimating the equation for V, problems can arise with the data. First, V and t
may not be measured accurately. Second, other factors may influence the speed of the
12
For a comprehensive discussion of bootstrapping, including methods that do not use
sampling with replacement, see Joel L. Horowitz, “The Bootstrap,” in James Heckman
and Edward Leamer, Handbook of Econometrics, Vol. 5, North Holland (2001), pp.
3159-228. For a simpler discussion focused on regression analysis, see David A.
Freedman and Stephen C. Peters, “Bootstrapping an Econometric Model: Some
Empirical Results,” Journal of Business and Economic Statistics Col. 2, No. 2 (April
1984), pp. 150-58.
37
falling object, such as imperfections in the release of the object and friction due to air
resistance, in which case the simple equation for V has a specification error in that
variables that influence V are excluded. Third, some sources of error may be random
variables with an expected value at or near zero, in which case a larger number of
measurements will produce a smaller error in the estimate of g. The experiment in a
physics lab to try to measure g does not include observations of every object that has ever
fallen anywhere on earth, but is a sample of all such events, and the larger the fraction of
all possible events that is included in the sample, the more accurate will be the estimate
of g. The problem arising from this fact is called sampling error.
A standard linear regression to estimate g from a sample of observations on V and
t assumes that these errors are independent – that is, that the magnitude of the error in one
observation does not reveal any information about the magnitude of the error in another
observation. This assumption would not be true, for example, if the measurement errors
were due to a design flaw in the clock that measured t so that all such measures were, say,
10 percent too high for all values of t that were observed. The presence of this problem
has nothing to do with whether the true underlying relationship is identically the same for
all observations – that is, whether nature sets physics formulas or Apple sets pricing
formulas – but whether the errors in trying to estimate either are independent.
Within this framework, clustering refers to one reason that the independence
assumption might not be true. Specifically, clustering refers to how observations are
collected. In the physics experiment, the observations on V and t are taken in the same
lab using the same equipment. If the acceleration due to gravity depends on location and
the accuracy of measurements depends on the equipment that is used, some of the errors
38
in the equation for V will be correlated among all observations from the same laboratory
using the same equipment. Consequently, the observations of V and t will not be
representative of objects that have fallen in other locations and could have been measured
by other types of equipment. Here the simple equation has a specification error in that
the variables for the measurement equipment and the location of the lab are not included
in the equation.
Clustering arises when a sample of observations suffers from the type of problem
that is represented by a dependence of the observations on the location of the experiment
and the equipment that is used. If the sample consists of observations taken randomly at
every location on earth using equipment that is drawn randomly from all possible types of
measuring devices, the independence assumption is satisfied; however, if observations
are taken at one laboratory using one clock and one radar gun for measuring velocity, the
errors in estimating the equation for estimating g are likely to depend on location and
equipment, and so are not independent.
The significance of the sampling aspect of cluster effects is that cluster analysis is
irrelevant if the data set is either representative of the entire population of observations or
is not a sample at all, but in fact is the entire population. In fact, Apple’s transactions
data are a population, not a sample, of all transactions of iPods. Moreover, Apple’s
customers were not randomly assigned to a model of iPod, but freely chose which iPods
they purchased.
For still another reason cluster analysis is irrelevant to the issue of undertaking a
regression analysis of Apple’s transaction data. In fact, cluster analysis pertains to a form
of regression equation that differs from the regression equations that I estimated. To
39
understand why cluster analysis is irrelevant for this reason it is useful to examine the
type of problem that cluster analysis is designed to solve.
The problem of clustering has been extensively analyzed in the study of the
effects of curriculum changes on the educational outcomes of children. The problem is to
use regression analysis to determine the effect on the outcome (e.g., test score), Yig, for
person i in group g of a treatment variable (e.g., a new curriculum) Xg, that takes the
value one if group g, to which person i belongs, received the treatment and zero
otherwise.13 The regression equation is then:
(1)
Yig = A + bXg + eig,
in which A is a constant, b is the coefficient on the treatment, and eig is the error in the
estimate of Y for person i in group g. The standard assumption about the error term is
that it is independent and identically distributed with a mean of zero and a finite variance,
which means that the value of eig does not depend on any of the other variables in the
equation and that the variance of the error term is the same for each observation.
The more general regression framework assumes that the regression contains two
types of independent variables.14 The first type, Xg, applies commonly to all members of
group g. For example, if a treatment is applied to some groups but not others, Xg is one
for groups with the treatment and zero otherwise. The second type, Zig, is any variable
13
The standard notation in cluster analysis is based on the convention of separately
numbering the members of each group, so that Y14 refers to the first person in the fourth
group and Y16 refers to the first person in the sixth group.
14
The analysis presented here is based on discussions of cluster effects in Jeffrey M.
Wooldridge, Econometric Analysis of Cross Section Data, MIT Press (2002), especially
Chapter 11, Section 11.5, pp. 328-332; Jeffrey M. Wooldridge, “Cluster Sample
Analysis in Applied Econometrics” American Economic Review Vol. 93, No. 2 (May
2003), pp. 133-38; Joshua D. Angrist and Jörn-Steffen Pischke, Mostly Harmless
Econometric: An Empiricists Companion, Chapter 8, pp. 293-325, Princeton University
Press, 2009; and Cameron and Miller, op. cit.
40
that takes different values for individuals within a group. A variable of this type may or
may not apply to more than a single group. The outcome variable is Yig for person i in
group g, and the estimated regression equation is then:
(2)
Yig = A + bXg + cZig + eig,
in which c is the regression coefficient on Zig. Of course, there can be several different
treatment variables (Xg) and individual variables (Zig), in which case all the variables are
interpreted as vectors, but I ignore this complexity here.
The econometric problems that can arise in the more general specification
involving clusters are as follows. First, the outcome for a member of a group may be
affected by factors other than the treatment variable that are common to all group
members. Second, the outcome of an individual may be influenced by an individuallevel variable that is not included in the regression but that is correlated with the group to
which an individual belongs. Third, the assumption of independent and identical
distributions for the error terms may be violated in that the mean and variance of the
errors differ among groups.
The first two problems are examples of an omitted variable, which is a form of
specification error. The first problem arises because the treatment and other common
causal factors are all rolled into the same variable, the group indicator Xg. For example,
if a group is students in a particular class at a particular school, the quality of the teacher
can affect outcomes as well as the treatment effect. The second problem arises because
not all individual-level variables are represented in Zig. For example, if a group is
students in a school, the education of the parents may differ between schools that are
given the new curriculum and schools that continue to use the old curriculum. In both
41
cases, the estimated coefficient on the treatment effect in (2) will confound the effect of
the treatment with the effects of other variables that are correlated with it.
The standard solution for omitted variables problems is not to omit variables.
One straight-forward way to add variables that take into account factors that are common
within a group is to add indicator (fixed effect) variables for each specific group and
additional variables that may be correlated with the identity of the group. “If the model
includes cluster-specific fixed effects, and we believe that within-cluster correlation of
errors is solely driven by a common shock process, then we may not be worried about
clustering. The fixed effects will absorb away the common shock, and the remaining
errors will have zero within-cluster correlation. More generally, control variables may
absorb systematic within-cluster correlation.”15
To illustrate the omitted variable problem in cluster samples, regressions using
equation (1) have generated results implying that smaller class sizes cause improvements
in test scores. Subsequent analysis shows that this simple model over-estimates the
magnitude and statistical significance of the effect of class size because smaller class
sizes tend to be found in schools in which students have wealthier, more educated parents
and better teachers. If these other factors are included in the regression, the effect of
class size falls substantially (sometimes to zero). Moreover, the new estimate of the
class-size effect is much smaller than the lower bound of the 95 percent confidence
interval on the coefficient for class size in the simple regression. This result shows that
the simple regression under-estimated the standard error of the regression coefficient as
well as over-estimated the value of the coefficient.
15
Cameron and Miller, op. cit., p. 21.
42
The third potential problem in equation (2) is that the error terms may not be
independent and identically distributed. Instead they may exhibit autocorrelation (e.g., if
observations are arranged by group, errors within a group will tend not to average zero)
and/or heteroskedasticity (e.g., the variance of the error term may differ among groups).
In an analysis in which individual observations are assigned to groups, the variance of the
standard errors can differ among groups if the number of observations in each group
differs or if excluded variables apply to some groups but not others.
The importance of problems related to the error term depends on the nature of the
data. One standard solution to these problems is to add more data by including more
clusters in the sample. As the number of groups (clusters) grows larger, the importance
of within-cluster error problems diminishes. The other standard solution is to test for and
correct autocorrelation and heteroskedasticity.
The next issue is how clustering problems relate to estimating the price effect of a
change in competitive conditions in the market for iPods and the equations that I have
estimated to measure these effects. Professors Murphy and Topel define a cluster as a
group of iPod transactions for a specific class/generation/family of iPod in a specific
quarter during the data period. These clusters bear no relationship to the clustering
problem in econometrics. The entering assumption of clustering models is that the
individuals that are represented in the data can be divided into groups that are affected by
common factors. “A cluster sample is typically a cross-section on individuals… where
each individual is part of a cluster.”16 The individuals in the Apple transaction data
include resellers – distributors and retailers who sell Apple iPods. Most of these Apple
16
Wooldridge, 2002, op. cit., p. 229.
43
customers appear in many (perhaps all) clusters. Likewise, even some direct purchasers
buy more than one model of iPod. In each case, the purchaser decides which iPods to
buy, rather than being assigned to buy a specific model by Apple. Thus, the transactions
data are not analogous to test scores of students from different classes that are assigned
different teaching materials. Likewise, the seller always is Apple – the clusters do not
represent different individual sellers that may not be representative of all sellers.
Professors Murphy and Topel cluster transaction observations by time periods.
The necessary condition for a correction for clustering to be useful is that strictly “withincluster correlation” is “substantial, which means the usual OLS standard errors can be
very misleading.”17 There is no plausible reason to believe that a distinct group-specific
effect influences transaction prices for a given class/generation/family of iPod differently
in two adjacent quarters. Neither the seller nor the product changes between quarters, nor
does the formula that Apple uses to set list prices and to give discounts to resellers. For
most resellers the identity of buyers (such as
) also does not change from quarter
to quarter. There is no basis for assuming that two clusters that are adjacent in time but
otherwise represent sales of the same iPod are substantially affected by different common
group-level effects, and Professors Murphy and Topel have not identified any plausible
candidates for differential effects across adjacent quarters.
Professors Murphy and Topel also do not consider the fact that clustering is a
problem of small samples in terms of both the number of clusters that are included in the
analysis and the number of observations per cluster. Assume that several clusters are
subjected to the same treatment effect, but, as is required for cluster analysis to be useful,
17
Wooldridge, 2003, op. cit., p. 134.
44
each cluster is affected by a distinct group-level common variable that is unobserved. In
this case, indicator variables for the cluster will account for the cluster-specific effects,
with the efficacy of the cluster variables in dealing with the cluster problem rising as the
number of clusters increases. The number of clusters identified by Professors Murphy
and Topel is in the hundreds, which is far more than the number that econometricians
normally would regard as creating a potential problem.18
In this case, one can also have statistical problems if the number of observations
in each cluster differs so that the sampling error differs among clusters. Substantial
disparity in sample sizes among groups can affect the efficiency of the estimate of a
cluster effect and cause heteroskedasticity across groups. But in this case, the number of
observations in each cluster also is large. As a result, the impact of heteroskedasticity is
small. With large sample sizes within clusters, the mean value of the dependent variable
within each cluster is a very precise estimate of the true mean, regardless of whether the
error terms in the observations in each cluster have the same variance.19
One underlying factor that gives rise to the clustering problem is unobserved
influences on outcomes at the group level. “Observations within a cluster are thought to
be correlated as a result of an unobserved cluster effect.”20 But, as explained above, this
within-cluster effect can be taken into account by introducing fixed effects for each
cluster and other variables that account for differences within a cluster. Equations (1)
18
Angrist and Pischke, op. cit, whimsically state that 42 is the magic number of clusters
that is sufficient to stop worrying about the distribution of errors. Of course, no magic
solution exists, but the effects of violating the standard assumptions about the distribution
of errors declines as the number of clusters increases.
19
Wooldridge, 2003, op. cit, p. 136.
20
Wooldridge, 2002, op. cit., p. 229.
45
and (2) are regressions in which the treatment is an independent variable, but neither
indicator variables for the clusters nor independent variables that might have a common
effect within a cluster enter the regression. By comparison, the regression equations that
I have estimated include indicator variables for iPod classes and capacities, other
characteristics of products (e.g., cost), and separate indicator variables for several
treatment effects (different competitive conditions in the market). Thus, the condition for
clustering to be a problem because of unobserved common effects is not present.
The important point to emphasize is that none of the problems giving rise to a
clustering adjustment is present in the iPod transactions data. First, there is no evidence
that the residual errors are correlated within a group of transactions for a given model of
iPod in a given calendar quarter. Second, the regressions on iPod transactions were not a
sample of types of iPods, but are based on all transactions. Thus, the iPod data require no
correction to adjust for sampling. Third, even if the transactions data were a sample, the
sample size within each group of transactions for each type of iPod is large, as is the
number of such groups. The magnitude of cluster effects diminishes with sample size,
and these sample sizes are large enough to cause the effect of clustering to become
unimportant. Fourth, variables to account for differences among these groups of
transactions are present in the data set and are included in the regressions.
The final question is what harm can arise from using an inappropriate technique
for adjusting the data when the data set does not fit the conditions that would make the
adjustment appropriate. The problem is that if there is no clustering effect, the standard
procedure for “improving” the estimates of the standard errors actually makes matters
worse by causing an upward bias in the estimates of the standard errors of the regression
46
coefficients.21 For this reason, the proposal to apply cluster adjustments to a data set that
does not actually contain clusters and that does not have a problem of confounding
unobserved effects of iPod models with the relevant treatment effect (the introduction of
iTunes 7.0) is not only unhelpful, but causes more harm than good in the quality of the
regression estimates.
The procedure that Professors Murphy and Topel adopt to deal with the perceived
problem of clustering collapses the residual errors for all transactions of a specific model
of an iPod during a given calendar quarter into a single observation. This procedure
vastly reduces the number of transactions observations, and thereby vastly reduces the
number of degrees of freedom in the regression. This procedure leaves Professors
Murphy and Topel with a regression that cannot reliably explain anything, all in pursuit
of correcting a non-existent clustering problem.
The procedure that Professors Murphy and Topel use is not even one that is most
recommended by econometricians when clustering is a problem. In dealing with true
cluster effects, econometricians recommend adding more variables and adjusting the
standard errors of the regression coefficient upward by a factor that depends on the
number of clusters and the number of observations per cluster.22 The former has been
done in the regression equations in the Noll Merits Report, and the latter adjustment
approaches zero as the number of clusters and observations grows large. Thus, although
the specifications of my regression equations were not adopted to deal with cluster
effects, these equations are more consistent with the standard methods for coping with
cluster effects than the methods advocated by Professors Murphy and Topel.
21
Angrist and Pischke, op. cit., pp. 293-308.
22
Cameron and Miller, op. cit., pp. 6-20.
47
RE-ESTIMATED DAMAGES MODEL
Separate damage equations were estimated for direct sale and reseller purchases.
These equations calculate damages by comparing actual prices for the reference product
with estimates of the prices that would have been charged in the “but-for” world in which
the alleged anticompetitive act – the introduction of iTunes 7.0 that disabled Harmony –
had not occurred. The regressions use the “before-after” method, in which the estimate
of damages is based on the differences in prices of models of iPods before and after the
occurrence of the anticompetitive act, as measured by a coefficient on a variable that
takes the value of one for iPod models that included iTunes 7.0 or its successors.
The regressions that are used to calculate damages are estimated from Apple’s
transactions data. Defendant’s economic experts, Professors Murphy and Topel, did not
add or subtract any transactions in their analysis and instead used the same data that I
used in the Noll Merits Report. The regressions use the logarithm of price from these
transactions records as the dependent variable.
23
Another feature that disabled Harmony,
. The iTunes 7.0 variable is set equal to one for these models.
48
The new regression equations differ from the regressions in the Noll Merits
Report in ways discussed elsewhere in this report. First, a variable for time has replaced
the logarithm of time. Second, the value of the “Harmony Blocked” variable has been set
to one from the date that iTunes 4.7 was introduced until the end of the data period for
models of iPods that did not use iTunes 7.0 or its successors. Third, the indicator
variable for Harmony has been divided into the two versions of Harmony, with the first
set equal to one from July 2004 until the end of the data period and the second set equal
to one for all periods after April 2005. Fourth, the date on which iTS became DRM-free
has been set as January 6, 2009, rather than March 31, 2009. Fifth, the regression has
been estimated using quantity weights, rather than frequency weights.
The reduced-form hedonic price equations that I have estimated are reported in
Exhibits 3A (resellers) and 3B (direct purchasers). The damages are applicable to all
iPod sales, including those excluded from the analysis. These price equations allow me
to calculate damages for the transactions excluded from the regression as long as I use net
sales revenues to account for returns.
To calculate damages from these equations, I start with the quantity and net sales
49
revenues for those models for which the iTunes 7.0 technology that disabled Harmony
had been enabled for every transaction that occurred between the date that iTunes 7.0 was
introduced and the date that iTS became DRM-free. For the latter date, I use January 6,
2009, although the use of the January date causes damages to be underestimated for two
reasons. First, iTS customers continued to buy some encrypted recordings from iTS
beyond that date. Second, the lock-in effect remained in force for all iPod users who had
bought encrypted downloads from iTS in the past. These customers would not stop being
locked in just because their most recent purchases were in a DRM-free format. The
effect of continuing lock-in is to make the demand for iPods less price elastic, and
thereby to cause the profit-maximizing price to be higher than otherwise would have been
the case had iTunes 7.0 not prevented iPod users from buying music from RMS.
For each class/generation/family, total net sales quantity and total net sales
amount are calculated. The estimated damage is calculated by applying the price
overcharge percentage (calculated from the regression coefficients on the iTunes 7.0
variable) to the units sold in each class/generation/family.
For an individual transaction, damages can be calculated from the regression
equations by multiplying the price overcharge percentage calculated from the iTunes 7.0
variable by the actual amount that was billed for the iPods that were purchased. If
several models were purchased in a single transaction, total damages associated with that
purchase are the sum of the damages for each iPod model, which is just the percentage
overcharge multiplied by the total amount billed. Damages among customers will vary
according to the price that they paid, with customers who received larger discounts
receiving smaller damages.
50
51
DECLARATION OF SERVICE BY MAIL AND E-MAIL
I, the undersigned, declare:
1.
That declarant is and was, at all times herein mentioned, a citizen of the United
States and a resident of the County of San Diego, over the age of 18 years, and not a party to or
interested party in the within action; that declarant’s business address is 655 West Broadway,
Suite 1900, San Diego, California 92101.
2.
That on November 25, 2013 declarant served the attached REBUTTAL
DECLARATION OF ROGER G. NOLL ON LIABILITY AND DAMAGES by depositing a
true copy thereof in a United States mailbox at San Diego, California in a sealed envelope with
postage thereon fully prepaid and addressed to the parties listed below.
3.
That there is a regular communication by mail between the place of mailing and
the place so addressed.
4.
Also on November 25, 2013, I served the attached REBUTTAL DECLARATION
OF ROGER G. NOLL ON LIABILITY AND DAMAGES on the parties in the within action by
e-mail addressed as follows:
COUNSEL FOR DEFENDANTS:
NAME
FIRM
EMAIL
Robert A. Mittelstaedt
Jones Day
ramittelstaedt@jonesday.com
Craig E. Stewart
Jones Day
cestewart@jonesday.com
Caroline Nason Mitchell
Jones Day
cnmitchell@jonesday.com
David C. Kiernan
Jones Day
dkiernan@jonesday.com
I declare under penalty of perjury that the foregoing is true and correct. Executed on
November 25, 2013, at San Diego, California.
s/ Shonda L. Landry
SHONDA L. LANDRY
CONFIDENTIAL - ATTORNEYS' EYES ONLY
CONFIDENTIAL - ATTORNEYS' EYES ONLY
CONFIDENTIAL - ATTORNEYS' EYES ONLY
CONFIDENTIAL - ATTORNEYS' EYES ONLY
CONFIDENTIAL - ATTORNEYS' EYES ONLY
CONFIDENTIAL - ATTORNEYS' EYES ONLY
CONFIDENTIAL - ATTORNEYS' EYES ONLY
CONFIDENTIAL - ATTORNEYS' EYES ONLY
CONFIDENTIAL - ATTORNEYS' EYES ONLY
CONFIDENTIAL - ATTORNEYS' EYES ONLY
CONFIDENTIAL - ATTORNEYS' EYES ONLY
CONFIDENTIAL - ATTORNEYS' EYES ONLY
CONFIDENTIAL - ATTORNEYS' EYES ONLY
CONFIDENTIAL - ATTORNEYS' EYES ONLY
CONFIDENTIAL - ATTORNEYS' EYES ONLY
CONFIDENTIAL - ATTORNEYS' EYES ONLY
CONFIDENTIAL - ATTORNEYS' EYES ONLY
CONFIDENTIAL - ATTORNEYS' EYES ONLY
CONFIDENTIAL - ATTORNEYS' EYES ONLY
APPENDIX A
CURRICULUM VITAE
ROGER G. NOLL
PERSONAL
Date and Place of Birth: March 13, 1940, Monterey Park, California
EDUCATION
East High School, Salt Lake City, Utah, 1958
B.S. (Math, Honors), California Institute of Technology, 1962
A.M., Ph.D. (Economics), Harvard University, 1965, 1967
SCHOLARSHIPS, FELLOWSHIPS AND AWARDS
National Merit Scholarship 1958-62
National Defense Education Act Fellowship 1962-66 (declined)
Harvard Prize Fellowship 1962-63
National Science Foundation Fellowship 1963-64
Guggenheim Fellow 1983-84
Rhodes Prize for Undergraduate Teaching, Stanford University, 1994
Distinguished Service Award, Public Utilities Research Center, University of Florida, 2001
Distinguished Lecture Award, Brookings-AEI Joint Center on Regulation and Markets, 2006
Alfred E. Kahn Distinguished Career Award, American Antitrust Institute, 2012
Distinguished Member Award, Transportation and Public Utilities Group, American Economic
Association, 2013
POSITIONS HELD
Instructor, California Institute of Technology, 1965-67
Assistant Professor, California Institute of Technology, 1967-69
Senior Staff Economist, Council of Economic Advisers, 1967-68
Associate Professor, California Institute of Technology, 1969-71
Senior Fellow and Co-director, Studies in the Regulation of Economic Activity, Brookings Institution,
1970-73
Professor, California Institute of Technology, 1973-82
Visiting Professor, Graduate School of Business, Stanford University, 1976-77
Chair, Division of the Humanities and Social Sciences, California Institute of Technology, 1978-82
Reuben Gustavson Lecturer, University of Chicago, April 1981
Institute Professor of Social Sciences, California Institute of Technology, 1982-84
Donald Gilbert Memorial Lecturer, University of Rochester, December 1982
Fellow, Center for Advanced Study in the Behavioral Sciences, 1983-84
Professor of Economics, Stanford University, 1984-2006 (Emeritus 2006-)
Visiting Scholar, Hoover Institution, 1984-85
Professor by Courtesy, Department of Political Science, Stanford University, 1985-2006
Professor by Courtesy, Graduate School of Business, Stanford University, 1986-2006
Veblen-Clark Lecturer, Carleton College, May 1986
Director, Public Policy Program, Stanford University, 1986-2002
David Kinley Lecturer, University of Illinois, May 1987
Sunderland Fellow, Law School, University of Michigan, Fall 1988
1
Positions, cont’d
Morris M. Doyle Centennial Professor in Public Policy, Stanford University, 1990-2002
Jean Monnet Professor, European University Institute, Spring 1991
Associate Dean, Humanities and Sciences, Stanford University, l991-92
Visiting Professor, University of California, San Diego, 1993
Visiting Fellow, Brookings Institution, 1995-96
Nonresident Senior Fellow, Brookings Institution, 1996-99
Director, American Studies Program, Stanford University, 2001-02
Visiting Scholar, London School of Economics, Spring 2001 and Spring 2002
Senior Fellow, American Antitrust Institute, 2002Director, Stanford Center for International Development, 2002-06
Kim Thomas Lecturer, Whittier College, 2010
TEACHING EXPERIENCE
Undergraduate: Introductory Economics, Intermediate Microeconomic Theory, Introduction to
Econometrics, Antitrust and Regulation, Economic History of Medieval Europe, History of Economic
Thought, Economic Policy Analysis, Economics of Sports, Political Economy of the West
Graduate: Antitrust and Regulation, Economic Policy Analysis, Applied Microeconomic Theory,
Experimental Economics
RESEARCH INTERESTS
Antitrust and Regulation, Technology Policy, Political Economics, Political Economy of Law
MEMBERSHIP ON BOARDS AND COMMITTEES
President's Task Force on Communications Policy (CEA Staff Representative), 1967-68
President's Task Force on Suburban Problems, 1968
President's Committee on Urban Housing, 1968
President’s Task Force on Public Broadcasting, 1968
Department of Commerce Technical Advisory Board Panel on Venture Capital, 1968-69
Committee on the Multiple Uses of the Coastal Zone, National Council on Marine Resources and
Engineering, 1968
Secretary, President's Interagency Task Force on Income Maintenance, 1968
Task Force on Application of Economic Analysis to Transportation Problems, National Research Council,
1970-73
Committee on Technological Forecasting on Behalf of the Environment, Office of Science and
Technology, 1970-71
Board of Economic Advisers, Public Interest Economics Foundation, 1974-84
Executive Committee, Caltech Environmental Quality Laboratory, 1970-71
Faculty Board, Caltech, 1974-76
Advisory Commission on Regulatory Reform, Senate Committee on Government Operations, 1975-77
Chair, Fourth Annual Telecommunications Policy Research Conference, 1975-76
Committee on Satellite Communications, National Academy of Sciences, 1975-76
Advisory Council, Jet Propulsion Laboratory, 1976-82
Chair, Committee to Monitor the Desegregation Plan of the Los Angeles Unified School District, Los
Angeles Superior Court, 1978-79
2
Boards, cont’d
Advisory Council, National Aeronautics and Space Administration, 1978-81
Advisory Council, National Science Foundation, 1978-89
Board of Advisers, National Institute of Economics and Law, 1978-84
Research Advisory Board, Committee for Economic Development, 1979-82
President's Commission for a National Agenda for the Eighties, 1980
Board of Directors, Economists, Inc., 1981Review Panel, NSF Regulation and Public Policy Program, 1981-84
Board of Editors, Journal of Economic Literature, 1981-90
Advisory Board, Solar Energy Research Institute, 1982-91
Board of Directors, Cornell Pelcovits and Brenner, Inc., 1982-1988
Chair, Advisory Panel on Information Technology R&D, Office of Technology Assessment, 1983-84
Supervisory Board of Editors, Information Economics and Policy, 1982-88
Advisory Committee on Integrated Environmental Management Program, Environmental Protection
Agency, 1983-85
Commission on Behavioral and Social Sciences and Education, National Research Council, 1984-90
Advisory Panel, NSF Policy Research and Analysis Division, 1984
Director, Program on Regulatory Policy, Stanford Institute for Economic Policy Research, 1984Panel on Clean Air, Science Advisory Board, Environmental Protection Agency, 1985-86
Board of Editors, Review of Economics and Statistics, 1985-2002
Contributing Editor, Regulation, 1986-93
Energy Research Advisory Board, Department of Energy, 1986-89
President & Chairman of the Board, Telecommunications Policy Research Foundation, 1986-87
Coordinating Editor, Information Economics and Policy, 1988-92
Board of Directors, International Telecommunications Society, 1988-92
Advisory Board of Editors, Journal of Risk and Uncertainty, 1988Acid Rain Advisory Committee, Environmental Protection Agency, 1990-91
Secretary of Energy Advisory Board, 1990-95
International Board of Editors, International Journal of the Economics of Business, 1993Faculty Senate, Stanford University, 1993-95, 98-02, 04-06
California Council on Science and Technology, 1995-2001
Panel on Universities, President's Committee of Advisors on Science and Technology, 1996
Committee on Intellectual Property and the Information Infrastructure, National Research Council, 1997-9
Board of Editors, Journal of Sports Economics, 1999Board of Associate Editors, Economics of Governance, 1999Board of Advisors, American Antitrust Institute, 2000Board on Science, Technology and Economic Policy, National Research Council, 2000-2006
Committee on Universal Postal Service, National Research Council, 2008
SPONSORED RESEARCH
"Opinions of Policemen." International Association of Chiefs of Police, 1969
"Studies in the Regulation of Economic Activity." Brookings Institution and Ford Foundation, 1970- 3
"Government Policies and Technological Innovation." National Science Foundation National R&D
Assessment Program, 1973-4
"The Social Consequences of Earthquake Prediction," National Aeronautics and Space Administration,
1974-6
"Nuclear Safety Regulation." National Science Foundation RANN Program, 1975-7
"The Public Television Station Program Cooperative." National Science Foundation RANN Program,
1975-7
"The Station Allocation Game." Federal Communications Commission, 1977
"Energy Policy Studies." Various donors, 1978-84
3
Sponsored Research, cont’d
"Economics of Oil Leasing" and "Issues in Utility Pricing." Department of Energy, 1978-9
"The Economics of Boxing, Wrestling and Karate." California Athletic Commission, 1978
"Implementing Tradable Emissions Permits." California Air Resources Board, 1979-82
"Social Science and Regulatory Policy." National Science Foundation, 1980-2
"The Political Economy of Public Policy." National Science Foundation and Center for Economic Policy
Research, Stanford University, 1983-4
"SIEPR Program on Regulatory Policy." various donors, 1987"The Economics of Research Universities and Scholarly Communication." Brown Center for Education
Policy, Brookings Institution, 1995-6
"Coordination of Regulatory Reform," Organization for Economic Cooperation and Development, 1996
"The Future of the Research University," Carnegie Foundation, 1996
“SCID Program in Economic Policy Reform,” Various donors, 2002-06
CONSULTING
Special Assistant to the President, Ford Foundation, 1969
Space Technology Applications, Jet Propulsion Laboratory, 1969
Panel on the Abatement of Particulate Emissions, National Research Council, 1971
Sloan Commission on Cable Communications, 1971
President's Commission on Government Procurement, 1971
Senate Subcommittee on Antitrust and Monopoly, 1971-72
MCI, Inc., 1972-73, 1983, 1986
National Science Foundation, 1973, 1975
Department of Justice, Antitrust Division, 1974-77, 1979-81, 1993-97
Internal Revenue Service, 1976-77
RAND Corporation, 1974-82
Los Angeles Lakers, 1974-75
National Football League Players Association, 1974-76, 1987-93, 2008, 2010Office of Telecommunications Policy, 1975-77
National Basketball Association Players Association, 1975-76, 1987-88, 1994
Naval Ordnance Test Station, 1975
Commission on Law and the Economy, American Bar Association, 1977-78
Aspen Institute Program on Communications and Society, 1977
National Commission on Electronic Funds Transfer, 1977
Business Round Table, 1978
Federal Communications Commission, 1977-81
Food and Drug Administration, 1978
Carnegie Commission on the Future of Public Broadcasting, 1978
Department of Energy, 1979
Office of Technology Assessment, 1980
Kerr-McGee Corporation, 1980
CBS, Inc. 1982-83
Environmental Protection Agency, 1982-83
Showtime/The Movie Channel, 1983, 1985
Harlequin Books, 1984
Lake Huron Broadcasting, 1984
National Collegiate Athletics Association, 1984
National Medical Enterprises, 1985, 1987-88
Camellia City Telecasters, 1985-86
4
Consulting, cont’d
Brown and Root, Inc., 1985-86
McDermott, Inc., 1985-86
Major League Baseball Players Association, 1985, 1994
United Cable Television and American Television and Communications, 1985
United States Football League, 1985-86
City of Anaheim, 1986
Technicolor, 1986
Metro-Mobile, 1986-89
Hewlett-Packard, 1986-1990, 1991
Echostar, 1987, 1994-95, 2002-03, 2004-05
Continental Airlines, 1987-88
Home Box Office, 1988-89
Bell South Cellular, 1989
Western Union, 1989
Minnesota Twins, 1989
Northwest Airlines, 1989
Pepsico, 1989
Yellow Phone, 1989-91
Dialog, 1990-91
California Public Utilities Commission, 1989-90
American Newspaper Publishers Association, 1990
Humana, 1990-91
Powell, Goldstein, Frazer and Murphy, 1990-93
South Coast Air Quality Management District, 1990-91
Federal Trade Commission, 1990-91, 2010Delta Airlines, 1991
California Cable Television Association, 1991
Bureau of Competition Policy, Government of Canada, 1991
R&D Business Systems, et al. 1991-95
International Entertainment Group, 1992-93
Nike, Inc., 1992
World Bank, 1992Gemini, Inc. 1992-94
Servicetrends, Inc., 1993-94
William Sullivan, 1993-95
Sure Safe Industries, 1993
U. S. Department of Justice, Civil Division 1994-95
Kopies, Inc., et al. 1995-1999
Telecom Technical Services, et al., 1995-1999
Digital Distribution, Inc.. 1996-1999
Silvey, et al., 1996-2000
Aguillar, et al. 1996-2000
Wadley Medical Center, 1997-2001
Oakland Raiders, 1997-2000
Major League Soccer Players Association, 1997-2000
Class Plaintiffs, Brand Name Prescription Drugs Litigation, 1998-9
Class Plaintiffs, Compact Disc Litigation, 1999-2003
Class Plaintiffs, State Microsoft Antitrust Litigation (California, Iowa, Minnesota, New York), 2000-2007
Kingray, 2000
Napster, 2000-2
5
Consulting, cont’d
Metropolitan Intercollegiate Basketball Association, 2002-5
Congressional Budget Office, 2002
Pioneer and Scientific Atlanta, 2002-3
Lenscrafters, 2003-4, 2009-12
Seven Network, 2003-7
Sports Car Clubs of America, 2003-05
Intertainer, 2003-05
Class Plaintiffs, DRAM Antitrust Litigation 2005-7
Class Plaintiffs, Honeywell Antitrust Litigation, 2005Class Plaintiffs, Tableware Antitrust Litigation, 2005-7
Class Plaintiffs, White, et al., v. NCAA, 2006-8
Sirius Satellite Radio and XM Satellite Radio, 2006-7, 2011-12
Class Plaintiffs, Cartier Antitrust Litigation, 2006-7
Monte Carlo Country Club and Societe Monegasque pour l’Exploitation du Tournai de Tennis, 2007
' '
'
Pearle Vision, Inc., 2007-8
Class Plaintiffs, Apple iTunes/iPod Antitrust Litigation, 2007Class Plaintiffs, SRAM Antitrust Litigation, 2007-9
Fair Isaac, 2007-9
Houston Baptist University, 2008.
U. S. Department of Justice, U. S. Attorney’s Office, San Francisco, 2008-9
Novell, 2008-11
GlaxoSmithKline, 2008-11
Class Plaintiffs, Flash Memory Antitrust Litigation, 2008-10
MobiTV, 2009-10
AT&T, 2009-10
Verizon, 2009-10
Ericsson, 2009-10
Kaleidescape, 2011-12
Class Plaintiffs, Text Messaging Antitrust Litigation, 2011Class Plaintiffs, California Automobile Insurance Antitrust Litigation, 2011Class Plaintiffs, NCAA Student-Athlete Name and Likeness Licensing Litigation, 2011-
BOOKS AND MONOGRAPHS
Reforming Regulation: An Evaluation of the Ash Council Report. Brookings Institution, 1971.
Economic Aspects of Television Regulation, co-authors Merton J. Peck and John J. McGowan. Brookings
Institution, 1973. Winner, National Association of Educational Broadcasters Annual Book Award, 1974.
Government and the Sports Business, editor. Brookings Institution, 1974.
The Political Economy of Deregulation, co-author Bruce Owen. American Enterprise Institute, 1983.
Regulatory Policy and the Social Sciences, editor. University of California Press, 1985.
The Technology Pork Barrel, co-author Linda R. Cohen. Brookings Institution, 1991.
The Economics and Politics of Deregulation. European University Institute, 1991.
6
Books, cont’d
Constitutional Reform in California: Making State Government More Effective and Responsive,
co-editor Bruce E. Cain. University of California Institute of Governmental Studies, 1995.
Sports, Jobs, and Taxes, co-editor Andrew Zimbalist. Brookings Institution, 1997.
Challenges to Research Universities, editor. Brookings Institution, 1998
A Communications Cornucopia, co-editor Monroe E. Price. Brookings Institution, 1998.
The Economics and Politics of the Slowdown in Regulatory Reform. AEI Press, 1999.
The Digital Dilemma, 17 co-authors (Committee on Intellectual Property Rights and the Emerging
Information Infrastructure). National Academy Press, 2000.
Bridging the Digital Divide, editor. California Council on Science and Technology, 2001.
Economic Reform in India, co-editors Nicholas C. Hope, Anjini Kochar, and T.N. Srinivasan. Cambridge
University Press, 2013
ARTICLES IN SCHOLARLY PUBLICATIONS
"Urban Concentration: Prospects and Implications." In Increasing Understanding of Public Problems and
Policies. Farm Foundation, 1969.
"Metropolitan Employment and Population Distribution and the Conditions of the Urban Poor.” In
Financing the Metropolis: Public Policy in Urban Economics: The Urban Affairs Annual Reviews, IV,
John P. Crecine, ed. Sage Publications, 1970. Brookings Reprint No. 184.
"National Communications Policy: Discussion--Spectrum Allocation Without Markets." American
Economic Review Papers and Proceedings 60(2) (May 1970).
"The Behavior of Regulatory Agencies." Review of Social Economics 24(1) (March 1971): 15-19.
Brookings Reprint No. 219 (November 1971).
"Summary and Conclusions," co-author William Capron. In Technological Change in Regulated
Industries, William Capron, ed. Brookings Institution, 1971.
"The Nature and Causes of Regulatory Failure." Administrative Law Review 23(4) (June 1971): 424-437.
Revised version published as "The Economics and Politics of Regulation." Virginia Law Review 57(6)
(September 1971): 1016-1032.
"Mass Balance, General Equilibrium and Environmental Externalities," co-author, John Trijonis.
American Economic Review 61(4) (September 1971): 730-735.
"Selling Research to Regulatory Agencies." In The Role of Analysis in Regulatory Decisionmaking: The
Case of Cable Television, Rolla Edward Park, ed. Heath-Lexington, 1973.
"Relative Prices on Regulated Transactions of the Natural Gas Pipelines," co-author Paul W. MacAvoy.
Bell Journal of Economics and Management Science 4(1) (Spring 1973): 212-234.
7
Scholarly Articles, cont’d
"Regulating Prices in Competitive Markets," co-author Lewis A. Rivlin. Yale Law Journal 82(7) (June
1973): 1426-1434.
"Attendance and Price Setting." In Government and the Sports Business, Roger G. Noll, ed. Brookings
Institution, 1974. Reprinted in The Economics of Sport, Andrew Zimbalist, ed. Edward Elgar, 2001.
"The U.S. Team Sports Industry." In Government and the Sports Business, Roger G. Noll, editor.
Brookings Institution, 1974. Abridged version reprinted in Public Policies Toward Business:
Reading and Cases, William G. Shephard, ed. Irwin, 1975.
"Alternatives in Sports Policy." In Government and the Sports Business, Roger G. Noll, ed. Brookings
Institution, 1974. Abridged version in Public Policies Toward Business: Readings and Cases, William G.
Shephard, ed. Irwin 1975. Revised version in Handbook of Social Science of Sport, Gunther R. R.
Luschen and George H. Sage, eds. Stripes Publishing Co., 1980.
"The Social Costs of Government Intervention." In The Business-Government Relationship in American
Society: Reassessment, Neil H. Jacoby, editor. University of California Press, 1975.
"The Consequence of Public Utility Regulation of Hospitals." In Controls on Health Care. Washington,
D.C.: National Academy of Sciences, 1975.
"Information, Decision-Making Procedures and Energy Policy." American Behavioral Scientist, Vol. 19,
No. 3 (January/February 1976): 267-278. Reprinted in Current Issues in Social Policy, W. B. Littrell and
G. Sjoberg, eds. Sage, 1976.
"Breaking Out of the Regulatory Dilemma: Alternatives to the Sterile Choice." Indiana Law Journal 51(3)
(Spring 1976): 686-699. Reprinted in Corporate Practice Commentator 19(1) (Spring 1977): 99-114.
"Safety Regulation," co-authors Nina Cornell and Barry Weingast. In Setting National Priorities: The
Next Ten Years, Henry Owen and Charles L. Schultze, eds. Brookings Institution, 1976.
"Major League Team Sports." In The Structure of American Industry, Walter Adams, ed. 5th ed.
Macmillan, 1977. 6th ed. Macmillan, 1981.
"An Experimental Market for Public Goods: The PBS Program Cooperative," co-author John A. Ferejohn.
American Economic Review Papers and Proceedings 66(2) (May 1976): 267-273.
"Government Policy and Technological Innovation: Where Do We Stand and Where Do We Go?" In
Innovation, Economic Change and Technology Policies, K.A. Stroetmann, ed. Birkauser-Verlag, 1977.
"Economic Policy Research on Cable Television: Assessing the Costs and Benefits of Cable
Deregulation," co-authors S. M. Besen, B. M. Mitchell, B. M. Owen, R. E. Park, and J. N. Rosse. In
Deregulation of Cable Television, Paul W. MacAvoy, ed. American Enterprise Institute, 1977.
"The Economic Implications of Regulation by Expertise: The Case of Recombinant DNA Research," coauthor Paul A. Thomas. In Research with Recombinant DNA. National Academy of Sciences, 1977.
"The Dilemma of Consumer Advocacy." In Regulatory Reform, W.S. Moore, ed. American Enterprise
Institute, 1978.
"Voters, Bureaucrats and Legislators: A Rational Choice Perspective on the Growth of Bureaucracy," coauthor Morris P. Fiorina. Journal of Public Economics 9(3) (May 1978): 239-254.
8
Scholarly Articles, cont’d
"Public Utilities and Solar Energy Development: Institutional Economic Considerations." In The Solar
Market: Proceedings of the Symposium on Competition in the Solar Industry, Federal Trade Commission,
1978. Excerpted version in On the Economics of Solar Energy, Stephen L. Feldman and Robert M.
Wirtshafter, eds. D.C. Heath, Lexington Books, 1980.
"Uncertainty and the Formal Theory of Political Campaigns," co-author John A. Ferejohn. American
Political Science Review 72(2) (June 1978): 492-505.
"The Rationale for Mandated Cost Increases." In Economic Effects of Government-Mandated Costs,
Robert F. Lanzillotti, ed. University Presses of Florida, 1978.
"Regulation and Computer Services." In The Computer Age, Michael L. Dertouzos and Joel Moses, eds.
MIT Press, 1979: 254-284.
"An Experimental Analysis of Decision-Making Procedures for Discrete Public Goods: A Case Study of a
Problem in Institutional Design," co-authors John A. Ferejohn and Robert E. Forsythe. In Research in
Experimental Economics, Vol. I, Vernon L. Smith, ed. JAI Press, 1979.
"Voters, Legislators and Bureaucracy: Institutional Design in the Public Sector," co-author Morris P.
Fiorina. American Economic Review Papers and Proceedings 68(2) (May 1978): 256-260.
Translated into Italian in Problemi Di Amministrazione Pubblica 4(2) (1979): 69-89.
"Practical Aspects of the Construction of Decentralized Decision-Making Systems for Public Goods," coauthors John A. Ferejohn and Robert Forsythe. In Collective Decisionmaking, Clifford S. Russell, ed.
Resources for the Future, 1979.
"Antitrust Exemptions: An Economic Overview." In National Commission for the Review of Antitrust
Laws and Procedures, Report to the President and the Attorney General, Vol. II. U.S. Department of
Justice, 1979.
"Regulatory and Nonregulatory Strategies for Controlling Health Care Costs," co-author Alain Enthoven.
In Medical Technology: The Culprit Behind Health Care Costs, Stuart H. Altman and Robert Blendon, eds.
Sun Valley Forum on National Health. U.S. Department of Health, Education and Welfare Publication No.
(PHS) 79-3216, 1979.
"Majority Rule Models and Legislative Elections," co-author Morris P. Fiorina. Journal of Politics, Vol.
41, No. 4 (November 1979): 1081-1104.
"The Game of Health Care Regulation: Comments on Feldman/Roberts." In Issues in Health Care
Regulation, Richard S. Gordon, ed. McGraw-Hill Book Co., 1980.
“Discussion: Regulatory Institutions in Historical Perspective.” American Economic Review Papers and
Proceedings, 70(2) (May 1980): 316-317.
"The Economics of Disaster Defense: The Case of Building Codes to Resist Seismic Shock," co-author
Linda Cohen. Public Policy 29(1) (Winter 1981): 1-29. Reprinted in The Economics of Natural Disasters,
Vol. I, Howard Kunruether and Adam Rose, eds. Edard Elgar, 2004.
"Regulation in Theory and Practice: An Overview," co-author Paul L. Joskow. In Studies in Public
Regulation, Gary Fromm, ed. MIT Press, 1981.
9
Scholarly Articles, cont’d
"Designing a Market for Tradable Emissions Permits," co-author Robert W. Hahn. In Reform of
Environmental Regulation, Wesley Magat, ed. Lexington Books, 1982.
"Implementing Marketable Emissions Permits." American Economic Review Papers and Proceedings 72(2)
(May 1982): 120-124.
"An Experimental Examination of Auction Mechanisms for Discrete Public Goods," co-authors John A.
Ferejohn, Robert Forsythe and Thomas R. Palfrey. In Research in Experimental Economics, Vol. II,
Vernon L. Smith, ed. JAI Press, 1982. Reprinted in Experimental Foundations of Political Science,
Donald R. Kinder and Thomas R. Palfrey, eds. University of Michigan Press, 1993.
"Implementing Tradable Emissions Permits," co-author Robert W. Hahn. In Reforming Social Regulation,
Leroy Graymer and Frederick Thompson, eds. Sage Publications, 1982.
"The Case Against the Balanced Budget Amendment: Comments on Aranson and Rabushka." In The
Economic Consequences of Government Deficits, Laurence Meyer, ed. Kluwer-Nyhoff Publishing, 1983.
"The Feasibility of Marketable Emissions Permits in the U.S." In Public Sector Economics, Jorg Finsinger,
ed. Macmillan Press, Ltd., 1983.
"Barriers to Implementing Tradable Air Pollution Permits: Problems of Regulatory Interactions," co-author
Robert W. Hahn. Yale Journal on Regulation 1(1) (1983): 63-91.
"The Future of Telecommunication Regulation." In Telecommunications Today and Tomorrow, Eli Noam,
ed. Harcourt Brace Jovanovich, 1983.
"The Political Foundations of Regulatory Policy." Zeitschrift fur die gesamte Staatswissenschaft (Journal
of Institutional and Theoretical Economics) 139(3) (1983): 377-404. Reprinted in Congress: Structure
and Policy, Mathew McCubbins and Terry Sullivan, eds. Cambridge University Press, 1987.
"The Regulation of Surface Freight Transportation: The Welfare Effects Revisited," co-author Ronald R.
Braeutigam. The Review of Economics and Statistics 66(1) (1984): 80-87.
"Prospective Payment: Will It Solve Medicare's Financial Problem," co-author Alain C. Enthoven. Issues
in Science and Technology 1(1) (1984): 111-116. Reprinted in Health Industry Today 48(3) (1985): 16-24.
"The Preferences of Policy Makers for Alternative Allocations of the Broadcast Spectrum," co-author
Forrest Nelson. In Antitrust and Regulation: Essays in Memory of John J. McGowan, Franklin M. Fisher,
ed. MIT Press, 1985.
"Government Regulatory Behavior: A Multidisciplinary Survey and Synthesis." In Regulatory Policy and
the Social Sciences, Roger G. Noll, ed. University of California Press, 1985.
"'Let Them Make Toll Calls': A State Regulator's Lament." American Economic Review Papers and
Proceedings 75(2) (1985): 52-56.
"State Regulatory Responses to Competition and Divestiture in the Telecommunications Industry." In
Antitrust and Regulation, Ronald E. Grieson, ed. Lexington Books, 1986.
"The Political and Institutional Context of Communications Policy." In Marketplace for
Telecommunications, Marcellus S. Snow, ed. Longman, Inc., 1986.
10
Scholarly Articles, cont’d
"Funding and Knowledge Growth: Comments." Social Studies of Science 16(1) (1986): 135-42.
"Government R&D Programs for Commercializing Space," co-author Linda R. Cohen. American
Economic Review Papers and Proceedings 76(2) (1986): 269-73.
"Administrative Procedures as Instruments of Political Control," co-authors Mathew D. McCubbins and
Barry R. Weingast. Journal of Law, Economics and Organization 3(2) (1987): 243-77. Abridged version
in State and Federal Administrative Law, Arthur Earl Bonfield and Michael Asimow, eds. West
Publishing, 1989. Reprinted in Economic Regulation, Paul Joskow, ed. Edward Elgar, 2000; Regulation
and Regulatory Processes, Cary Coglianese and Robert Kagan, eds. Ashgate Publishing, 2007; The
Economics of Administrative Law, Susan Rose-Ackerman, ed., Edward Elgar, 2007; Rational Choice
Politics, Torun Dewan, Keith Dowling and Kenneth Shepsle, eds., Sage, 2009.
"Communications." In The New Palgrave, John Eatwell, Murray Milgat, and Peter Newman, eds.
MacMillan, 1987.
"Comment: Settlement Incentives and Follow-on Litigation." In Private Antitrust Litigation, Lawrence J.
White, ed. MIT Press, 1988.
"The Political Economy of NASA's Applications Technology Satellite Program," co-author Linda R.
Cohen. In Space Applications Board, Proceedings of a Symposium on Space Communications Research
and Development. National Research Council, 1988.
"Economics, Politics and Government Research and Development," co-author Linda Cohen. In
Technology and Politics, Michael E. Kraft and Norman J. Vig, eds. Duke University Press, 1988.
"The Anticompetitive Uses of Regulation: United States v. AT&T (1982)," co-author Bruce M. Owen. In
The Antitrust Revolution, John E. Kwoka, Jr., and Lawrence J. White, eds. Scott, Foresman, 1988.
"The Economics of Sports Leagues." In Law of Professional and Amateur Sports, Gary A. Uberstine, ed.
Clark Boardman, 1988.
"Preface: Symposium on Telecommunications Demand." Information Economics and Policy 3(4)
(1988): 275.
"Telecommunications Regulation in the 1990s." In New Directions in Telecommunications Policy, Vol. I,
Paula R. Newberg, ed. Duke University Press, 1989.
"U.S. v. AT & T: An Interim Assessment," co-author Bruce M. Owen. In Future Competition in
Telecommunications, Stephen P. Bradley and Jerry A. Hausman, eds. Harvard Business School Press,
1989.
"Structure and Process, Politics and Policy: Administrative Arrangements and the Political Control of
Agencies," co-authors Mathew D. McCubbins and Barry R. Weingast. Virginia Law Review 75(2)
(March 1989): 431-482. Reprinted in The Political Economy of Regulation, Thomas Lyon, ed. Edward
Elgar, 2007.
"Economic Perspectives on the Politics of Regulation." In Handbook of Industrial Organization, Vol. II.
Richard Schmalensee and Robert Willig, eds. North Holland Publishing Co., 1989. Reprinted in
Regulation and the Law, Anthony I. Ogus, ed. Edward Elgar Publishing, 2001.
11
Scholarly Articles, cont’d
"Comment: Peltzman on Deregulation." Brookings Papers on Economic Activity: Microeconomics
(1989): 48-58.
"Pricing of Telephone Services," co-author Susan Smart. In After the Breakup, Barry G. Cole, ed.
Columbia University Press, 1990.
"Positive and Normative Models of Procedural Rights: An Integrative Approach to Administrative
Procedures," co-authors Mathew D. McCubbins and Barry R. Weingast. Journal of Law, Economics and
Organization 6 (1990): 307-332.
"Some Implications of Cognitive Psychology for Risk Regulation," co-author James Krier. Journal of
Legal Studies 19 (June 1990): 747-779. Excerpts reprinted in Foundations of the Economic Approach to
Law, Avery Weiner Katz, ed. Oxford University Press, 1998. Reprinted in Behavioral Law and
Economics, Cass R. Sunstein, ed. Cambridge University Press, 2000.
"Environmental Markets in the Year 2000," co-author Robert Hahn. Journal of Risk and Uncertainty 3
(December, 1990): 347-363.
"Commentary: The Prospects for Using Market Incentives for Conservation of Biological Diversity." In
The Preservation and Valuation of Biological Resources, Gordon H. Orians, Gardner M. Brown, Jr.,
William E. Kunin, and Joseph E. Swierzbinski, eds. University of Washington Press, 1990.
"Slack, Public Interest, and Structure-Induced Policy," co-authors Mathew D. McCubbins and Barry R.
Weingast. Journal of Law, Economics and Organization 6 (1990): 203-212.
"How to Vote, Whether to Vote: Strategies for Voting and Abstaining on Congressional Roll Calls," coauthor Linda R. Cohen. Political Behavior 13(2) (1991): 97-127.
"Rational Actor Theory, Social Norms, and Policy Implementation: Applications to Administrative
Processes and Bureaucratic Culture," co-author Barry R. Weingast. In The Economic Approach to Politics,
Kristen Renwick Monroe, ed. Harper Collins, 1991.
"The National Aerospace Plane: An American Technological Long Shot, Japanese Style," co-authors Linda
R. Cohen and Susan A. Edelman. American Economic Review Papers and Proceedings, 81(2) (May
1991): 50-53.
"The Economics of Intercollegiate Sports." In Rethinking College Athletics, Judith Andre and David N.
James, eds. Temple University Press, 1991.
"Comparative Structural Policies," co-author Haruo Shimada. In Parallel Politics, Samuel Kernell, ed.
Brookings Institution, 1991.
"Structural Policies in the United States." In Parallel Politics, Samuel Kernell, ed. Brookings Institution,
1991.
"On Regulating Prices for Physicians." In Regulating Doctors' Fees, H.E. Frech III, ed. AEI Press, 1991.
"ISDN and the Small User: Regulatory Policy Issues," co-author William Lehr. In Integrated Broadband
Networks, Martin C.J. Elton, ed. North-Holland, 1991.
"Professional Basketball: Economic and Business Perspectives." In The Business of Professional Sports,
Paul D. Staudohar and James A. Mangan, eds. University of Illinois Press, 1991.
12
Scholarly Articles, cont’d
"Computer Reservation Systems and Their Network Linkages to the Airline Industry," co-author
Margaret E. Guerin-Calvert. In Electronic Services Networks, Margaret E. Guerin-Calvert and Steven S.
Wildman, eds. Praeger, 1991.
"An Economic Analysis of Scientific Journal Prices: Preliminary Results," co-author W. Edward
Steinmueller. Serials Review 18 (1992): 32-37.
"The Theory of Interpretive Canon and Legislative Behavior," co-authors Mathew McCubbins and Barry
Weingast. International Review of Law and Economics 12 (1992): 235-238.
"Positive Canons: The Role of Legislative Bargains in Statutory Interpretation," co-authors Mathew
McCubbins and Barry Weingast. Georgetown Law Journal 80 (February 1992): 705-742. Reprinted
in Sutherland Statutory Construction, 5th edition, Norton J. Singer, ed., Clark Boardman Callaghan, 1992,
and Sutherland Statutes and Statutory Interpretation, 6th edition, Norman J. Singer and J. D. Shambie
Singer, eds., West, 2000.
"Comment: Judicial Review and the Power of the Purse." International Review of Law and Economics 12
(June 1992): 211-213.
"Comment: Standard Setting in High-Definition Television." Brookings Papers on Economic Activity:
Microeconomics, 1992: 80-89.
"Research and Development," co-author Linda R. Cohen. In Setting Domestic Priorities: What Can
Government Do? Henry J. Aaron and Charles L. Schultze, eds. Brookings Institution, 1992.
"The Economics of Information: A User's Guide." In The Knowledge Economy: Annual Review of the
Institute for Information Studies. Aspen Institute, 1993.
"Downsian Thresholds and the Theory of Political Advertising." In Information, Participation, and
Choice, Bernard Grofman, ed. University of Michigan Press, 1993.
"Economic Regulation," co-author Paul L. Joskow. In American Economic Policy in the 1980s, Martin
Feldstein, ed. University of Chicago Press, 1994.
"Legislative Intent: The Use of Positive Political Theory in Statutory Interpretation," co-authors Mathew
McCubbins and Barry Weingast. Law and Contemporary Problems 57 (Winter/Spring 1994): 3-37.
Reprinted in Public Choice and Public Law, Daniel A. Farber, ed. Edward Elgar, 2006, and Legal
Institutions and Economic Development, Robert A. Cooter and Francesco Parisi, eds., Edward Elgar, 2010.
"The Origins of State Railroad Regulation: The Illinois State Constitution of 1870," co-author Mark T.
Kanazawa. In The Regulated Economy, Claudia Goldin and Gary D. Libecap, eds. University of Chicago
Press, 1994.
"Privatizing Public Research," co-author Linda R. Cohen. Scientific American, September 1994, pp. 72-77.
Reprinted in Leading Economic Controversies of 1996, Edwin Mansfield, ed. New York: Norton, 1996.
"Japanese and American Telecommunications Policy," co-author Frances M. Rosenbluth. Communications
and Strategy 15 (3eme trimestre 1994): 13-46.
"Public Policy and the Admission of the Western States," co-author David W. Brady. In The Political
Economy of the American West, Terry L. Anderson and Peter J. Hill, eds. Rowman and Littlefield, 1994.
13
Scholarly Articles, cont’d
“Introduction: Regulation and the New Telecommunications Infrastructure.” The Changing Nature of
Telecommunications/Information Infrastructure. National Academy Press, 1995.
"Telecommunications Policy: Structure, Process, Outcomes," co-author Frances M. Rosenbluth. In
Structure and Policy in Japan and the United States, Mathew D. McCubbins and Peter Cowhey, eds.
Cambridge University Press, 1995.
"Politics and the Courts: A Positive Theory of Judicial Doctrine and the Rule of Law," co-authors
Mathew D. McCubbins and Barry R. Weingast. Southern California Law Review 68 (September 1995):
1631-83.
“Feasibility of Effective Public-Private R&D Collaboration: The Case of Cooperative R&D Agreements,”
co-author Linda R. Cohen. International Journal of the Economics of Business 2 (1995): 223-240.
"Principles of State Constitutional Design," co-author Bruce E. Cain. In Constitutional Reform in
California, Bruce E. Cain and Roger G. Noll, eds. University of California Institute for Governmental
Studies, 1995. Excerpt reprinted in Madison Review 3 (Fall 1997): 7-11.
"The Role of Antitrust in Telecommunications." Antitrust Bulletin (Fall 1995): 501-528.
"Executive Organization: Responsiveness vs. Expertise and Flexibility." In Constitutional Reform in
California, Bruce E. Cain and Roger G. Noll, eds. University of California Institute for Governmental
Studies, 1995.
"Comment: Preferences, Promises, and the Politics of Entitlement." In Individual and Social
Responsibility, Victor R, Fuchs, ed. University of Chicago Press, 1996.
"Privatizing Public Research: The New Competitiveness Strategy," co-author Linda R. Cohen. In
The Mosaic of Economic Growth, Ralph Landau, Timothy Taylor, and Gavin Wright, eds. Stanford
University Press, 1996.
"Is There a Role for Benefit-Cost Analysis in Environmental, Health and Safety Regulation?," 10 coauthors. Science 272 (April 12, 1996): 221-222. Reprinted in Environmental and Development
Economics, Vol. 2, No. 2 (May 1997), pp. 196-201, and Economics of the Environment, Robert N. Stavins,
ed. Norton, 2000, and subsequent editions.
“Benefit-Cost Analysis in Environmental, Health, and Safety Regulation: A Statement of Principles,” ten
co-authors. AEI, 1996.
"Reforming Risk Regulation." The Annals of the AAPSS 545 (May 1996): 165-75.
"The Complex Politics of Catastrophe Economics." Journal of Risk and Uncertainty 12 (May 1996):
141-46.
"The Economics of Information." In The Economics of Information in the Networked Environment,
Meredith A. Butler and Bruce R. Kingma, eds. Association of Research Libraries, 1996. Reissued by
Haworth Press, 1999.
"The Economics of Scholarly Publications and the Information Superhighway." In The Internet and
Telecommunications Policy, Gerald W. Brock and Gregory L. Rosston, eds. Lawrence Erlbaum, 1996.
14
Scholarly Articles, cont’d
"Regulatory Reform as International Policy." In Regulatory Reform and International Market Openness.
Organisation for Economic Cooperation and Development, 1996.
"The Future of the National Laboratories," co-author Linda R. Cohen. Proceedings of the National
Academy of Sciences 93 (November 1996): 12678-85.
"Research and Development after the Cold War," co-author Linda R. Cohen. In Commercializing High
Technology: East and West, Judith B. Sedaitis, ed. Roman and Littlefield, 1997.
“Internationalizing Regulatory Reform.” In Comparative Disadvantage? Social Regulations and the
Global Economy, Pietro S. Nivola, ed. Brookings Institution, 1997.
“The International Dimension of Regulatory Reform: With Applications to Egypt.” Distinguished Lecture
Series #8. Egyptian Center for Economic Studies, 1997.
“Build the Stadium–Create the Jobs!” co-author Andrew Zimbalist. In Sports, Jobs and Taxes, Roger G.
Noll and Andrew Zimbalist, eds. Brookings Institution, 1997.
“The Economic Impact of Sports Teams and Facilities,” co-author Andrew Zimbalist. In Sports, Jobs and
Taxes, Roger G. Noll and Andrew Zimbalist, eds. Brookings Institution, 1997.
“Sports, Jobs and Taxes: The Real Connection,” co-author Andrew Zimbalist. In Sports, Jobs and Taxes,
Roger G. Noll and Andrew Zimbalist, eds. Brookings Institution, 1997.
“Legislative Control of Bureaucratic Policy Making,” co-authors Mathew D. McCubbins and Barry R.
Weingast. New Palgrave Dictionary of Economics and the Law. London: New Palgrave, 1997.
“The American Research University: An Introduction.” In Challenges to Research Universities, Roger G.
Noll, ed. Brookings Institution, 1998.
“Universities, Constituencies, and the Role of the States,” co-author Linda R. Cohen. In Challenges to
Research Universities, Roger G. Noll, ed. Brookings Institution, 1998.
“Students and Research Universities,” co-author Gary Burtless. In Challenges to Research Universities,
Roger G. Noll, ed. Brookings Institution, 1998.
“The Economics of University Indirect Cost Reimbursement in Federal Research Grants,” co-author
William P. Rogerson. In Challenges to Research Universities, Roger G. Noll, ed. Brookings Institution,
1998.
“The Future of Research Universities.” In Challenges to Research Univesities, Roger G. Noll, ed.
Brookings Institution, 1998.
“Communications Policy: Convergence, Choice, and the Markle Foundation,” co-author Monroe E. Price.
In A Communications Cornucopia, Roger G. Noll and Monroe E. price, eds. Brookings Institution, 1998.
“Economic Perspectives on the Athlete’s Body.” Stanford Humanities Review 6(2) (1998): 69-73.
“The Bell Doctrine: Applications in Telecommunications, Electricity, and Other Network Industries,” coauthor Paul L. Joskow. Stanford Law Review 51(5) (1999): 1249-1315.
15
Scholarly Articles, cont’d
“The Political Origins of the Administrative Procedure Act,” co-authors Mathew D. McCubbins and Barry
R. Weingast. Journal of Law, Economics, and Organization 15(1) (1999): 180-217.
“The Economics of Information,” Journal of Library Administration 26(1) (1999): 47-55.
“Competition Policy in European Sports after the Bosman Case.” In Competition Policy in Professional
Sports: Europe after the Bosman Case, Claude Jeanrenaud and Stefan Kesenne, eds. Standaard Editions
'
Ltd., 1999.
“The Business of College Sports and the High Costs of Winning.” Milken Institute Review 1(3)
(3rd Quarter 1999): 24-37. Reprinted in The Business of Sports, Scott Rosner and Kenneth Shropshire,
eds. Jones and Bartlett, 2004.
“Antitrust and Labor Markets in Professional Sports.” In The Role of the Academic Economist in
Litigation Support, Daniel J. Slottje, ed. North-Holland, 1999.
“Reforming Urban Water Systems in Developing Countries,” co-authors Mary M. Shirley and Simon
Cowan. In Economic Policy Reform: The Second Stage, Anne O. Krueger, ed. University of Chicago,
2000.
“Telecommunications Reform in Developing Countries.” In Economic Policy Reform: The Second Stage,
Anne O. Krueger, ed. University of Chicago: 2000.
“Regulatory Reform and International Trade Policy.” In Deregulation and Interdependence in the AsiaPacific Region, Takatoshi Ito and Anne O. Krueger, eds. University of Chicago/NBER, 2000.
“Comment: Hong Kong’s Business Regulation in Transition.” In Deregulation and Interdependence in
the Asia-Pacific Region, Tahatoshi Ito and Anne O. Krueger, eds. University of Chicago/NBER, 2000.
“The Digital Divide: Definitions, Measurement, and Policy Issues,” co-authors Dina Older-Aguilar,
Richard R. Ross and Gregory L. Rosston. In Bridging the Digital Divide, Roger G. Noll, ed. California
Council on Science and Technology, 2001.
“The Digital Divide: Diagnosis and Policy Options,” co-author Dina Older-Aguilar. In Bridging the
Digital Divide, Roger G. Noll, ed. California Council on Science and Technology, 2001.
“Intellectual Property, Antitrust and the New Economy,” co-author Linda R. Cohen. University of
Pittsburgh Law Review 62(3) (Spring 2001): 453-73.
“The Economics of Promotion and Relegation in Sports Leagues: The Case of English Football.” Journal
of Sports Economics 3(2) (May 2002): 169-203. Reprinted in The Economics of Association Football, Bill
Girard, ed., Edward Elgar, 2006.
“The Economics of Urban Water Systems.” In Thirsting for Efficiency, Mary M. Shirley, ed. Pergamon,
2002.
“Comment: Small-Scale Industry Policy in India.” In Economic Policy Reforms and the Indian
Economy, Anne O. Krueger, ed. University of Chicago, 2002.
“The Economics of the Supreme Court’s Decision on Forward Looking Costs,” co-author Gregory L.
Rosston. Review of Network Economics 1(2) (September 2002): 81-9.
16
“Federal R&D in the Antiterrorist Era.” In Innovation Policy and the Economy, Vol. 3, Adam B. Jaffe,
Josh Lerner and Scott Stern, eds. MIT Press, 2003.
“The International Dimension of Regulatory Reform with Applications to Egypt.” In Challenges and
Reforms of Economic Regulations in MENA Countries, Imed Limam, ed. American University in Cairo
Press, 2003.
“The Economics of Contraction in Baseball.” Journal of Sports Economics 4(4) (November 2003): 367-88.
“The Organization of Sports Leagues.” Oxford Review of Economic Policy 19(4) (Winter 2004): 530-51.
“The Conflict over Vertical Foreclosure in Competition Policy and Intellectual Property Law.” Journal of
Institutional and Theoretical Economics 160(1) (March 2004): 79-96.
“New Tool for Studying Network Industry Reforms in Developing Countries,” co-authors Scott J.
Wallsten, George Clarke, Luke Haggarty, Rosario Kaneshiro, Mary Shirley and Lixin Colin Xu. Review of
Network Economics 3(3) (September 2004): 248-82.
“Comment: Who Benefits Whom in Local Television Markets?” In Brookings-Wharton Papers on Urban
Affairs 2004, William G. Gale and Janet Rothenberg Pack, eds. Brookings Institution, 2004.
“‘Buyer Power’ and Economic Policy.” Antitrust Law Journal 72(2) (2005): 311-40. Spanish version
“‘Poder de Compra’ y Politica Economica.” In Libra Competencia y Retail: Un Analisis Critico, Paulo
'
'
'
'
Montt Rettig and Nicole Behme Zalaquett, eds. Abeledo Perrot Legal Publishing, Santiago, Chile, 2010.
“Einstein’s Interoffice Memo.” Science 309: 5740 (September 2, 2005): 1490-1. Reprinted in CAUTACPPU Bulletin (December 2005): 2.
“The Politics and Economics of Implementing State-Sponsored Embryonic Stem Cell Research.” In States
and Stem Cells, Aaron D. Levine, ed. Policy Research Institute for the Region, Princeton University, 2006.
“Designing an Effective Program of State-Sponsored Human Embryonic Stem-Cell Research.” Berkeley
Technology Law Journal 21(3) (Summer 2006): 1-33.
“Universal Telecommunications Service in India,” co-author Scott J. Wallsten. In India Policy Forum
2005-06, Suman Bery, Barry Bosworth and Arvind Pamagariya, eds. Brookings Institution, 2006.
Reprinted in Spectrum Law and Governance, Ramakistaiah Jilla, ed. Icfai University Press, 2008.
“Conditions for Judicial Independence,” co-authors Mathew D. McCubbins and Barry R. Weingast.
Journal of Contemporary Legal Issues 15(1) (September/October 2006): 107-29.
“Sports Economics after Fifty Years.” In Sports Economics after Fifty Years, Placido Rodriquez, Stefan
Kesenne and Jaume Garcia, eds. University of Oveido Press, 2006.
'
“Broadcasting and Team Sports.” Scottish Journal of Political Economy 54(3) (July 2007): 400-21.
“Comment: Housing Subsidies and Homeowners.” Brookings/Wharton Papers on Urban Affairs 2007,
Gary Burtless and Janet Rothenberg Pack, eds. Brookings Institution, 2007.
“The Political Economy of Law,” co-authors Mathew D. McCubbins and Barry R. Weingast. In
Handbook of Law and Economics, A. Mitchell Polinsky and Steven Shavell, eds. North-Holland
Publishers, 2007.
17
Scholarly Articles, cont’d
“The Economic Significance of Executive Order 13,422.” Yale Journal on Regulation 25(1) (Winter
2008), pp. 113-24.
“The Wines of West Africa: History, Technology and Tasting Notes.” Journal of Wine Economics Vol. 3,
No. 1(May 2008), pp. 85-94.
“Administrative Law Agonistes,” co-authors Mathew D. McCubbins, Daniel B. Rodriguez and Barry R.
Weingast. Columbia Law Review Sidebar Vol. 108 (April 29, 2008), pp. 15-22.
“Comment: London Congestion Charges.” Brookings-Wharton Papers on Urban Affairs 2008, Gary
Burtless and Janet Rothenberg Pack, eds. Brookings Institution, 2008.
“Priorities for Telecommunications Reform in Mexico.” In No Growth without Equity? Inequality,
Interests, and Competition in Mexico, Santiago Levy and Michael Walton, eds. Palgrave MacMillan, 2009.
“Malleable Constitutions: Reflections on State Constitutional Reform,” co-author Bruce E. Cain. Texas
Law Review Vol. 87, No. 7 (June 2009), pp. 1517-44.
“Regionalising Infrastructure Reform in Developing Countries,” co-authors Nancy C. Benjamin and
Ioannis N. Kessides. World Economics Vol. 11, No. 3 (July-September 2010), pp. 79-108.
“Institutional Causes of California’s Budget Problem,” co-author Bruce E. Cain. California Journal on
Politics and Policy Vol. 2, No. 2 (September 2010), pp. 1-35.
“Comment.” Energy Economics Vol. 33, No. 4 (July 2011), pp. 683-6.
“Impediments to Innovation in Legal Infrastructure.” I/S: A Journal of Law and Policy for the Information
SocietyVol. 8, No. 1 (Summer 2012), pp. 61-71.
“Introduction,” co-authors Nicholas C. Hope, Anjini Kochar, and T N. Srinivasan. In Economic Reform in
India, Nicholas C. Hope, Anjini Kochar, Roger Noll and T.N. Srinivasan, eds. Cambridge University
Press, 2013.
“An Assessment of Indian Telecommunications Reform,” co-author Scott J. Wallsten. In Economic
Reform in India, Nicholas C. Hope, Anjini Kochar, Roger Noll and T.N. Srinivasan, eds. Cambridge
University Pess, 2013.
“Alfred E. Kahn, 1917-2010,” co-author Paul L. Joskow. Review of Industrial Organization Vol. 42, No. 2
(March 2013), pp. 107-26.
“Evaluación de las Políticas de Telecomunicaciones en México.” El Trimestre Económico Vol. 80, No. 3
(July-September 2013), pp. 603-50.
“The DRAM Antitrust Litigation (2008).” In The Antitrust Revolution, John E. Kwoka, Jr., and Lawrence
J. White, eds. Oxford University Press, 2013.
“Endogeneity in Attendance Demand Models.” In The Econometrics of Sport, Placido Rodriguez, Stefan
Kesenne and Jaume Garcia, eds. Edward Elgar, 2013
18
OTHER PROFESSIONAL PUBLICATIONS
"Central City–Suburban Employment Distribution." In Final Report: Statistical Papers. President's Task
Force on Suburban Problems, Department of Housing and Urban Development, December 1968.
"Current Mortgage Finance Problems." In Final Report: Policy and Program Papers. President's Task
Force on Suburban Problems, Department of Housing and Urban Development, December 1968.
“A Statistical Analysis of the International Association of Chiefs of Police Poll of the Opinions of the
Opinions of Policemen,” Jet Propulsion Laboratory, Fall 1969.
“The Economics of Professional Basketball,” co-author Benjamin A. Okner. Brookings Reprint No. 258,
(1972). From “Statements,” Senate Antitrust Subcommittee, Hearings on S. 2373, September 21-23, 1971,
Professional Basketball (Part I) and May 3, 1972, Professional Basketball (Part II).
“Prospects and Policies for CATV,” co-authors John J. McGowan and Merton J. Peck. In On the Cable:
Report of the Sloan Commission on Cable Communications. New York: McGraw-Hill, 1971.
"The Price Commission and Regulated Public Utilities." Compendium on Price and Wage Controls: Now
and the Outlook for 1973. Joint Economic Committee, U.S. Congress, 1972.
"The Administration Bill to Deregulate Surface Transportation." In Transportation Policy: The EconomicPolitical Interface, Anthony M. Woodward, ed. Business Research Center, Syracuse University, 1972.
"The Case for Viewer Sovereignty," co-authors Merton J. Peck and John J. McGowan. Research Report
No. 135. Brookings Institution, June 1973.
"Regulating the Medical Services Industry." In The Changing Role of the Public and Private Sectors in
Health Care. National Health Council, 1973.
"Decentralization of Public Television." In Conference on Communications Policy Research: Papers and
Proceedings. Office of Telecommunications Policy, Washington, D.C., 1973.
"The Product Market in Sports." In Conference on the Economics of Professional Sports: Proceedings,
George Burman, ed. National Football League Players Association, Washington, D.C., May 7, 1974.
"Government Administrative Behavior and Technological Change." In Government Policies and
Technological Innovation, Vol. II. National Technical Information Service, 1974.
"Public Policy and Innovation: Two Cases," co-author W. David Montgomery. In Government Policies
and Technological Innovation, Vol. II. National Technical Information Service, 1974.
"Subsidization through Regulation: The Case of Broadcasting," co-authors John J. McGowan and Merton J.
Peck. In The Economics of Federal Subsidy Programs: Part 8 - Selected Subsidies. U.S. Government
Printing Office, 1974.
"Comments Regarding Limitations on Programming Available for Broadcast on Pay-TV Channels."
Submitted to Federal Communications Commission. Published in Communications--the Pay-Cable
Industry, Subcommittee on Antitrust and Monopoly, Senate Committee on the Judiciary, June 1975.
"Empirical Studies of Utility Regulation." In Rate of Return Regulation: Proceedings of the Future
Planning Conference. Federal Communications Commission, 1976.
19
Other Publications, cont’d
"The Role of Competition in the Electronic Media." In Proceedings of the Symposium on Media
Concentration, Vol. 1, December 14-15, 1978. Bureau of Competition, Federal Trade Commission.
"Adaptive Approaches to the CO2 Problem." In Carbon Dioxide, Climate and Society: A Research
Agenda, Vol. II. U.S. Department of Energy, 1980.
"The Rationale for Social Regulation." In Government Regulation: New Perspective, Andrew R. Blair, ed.
University of Pittsburgh, Graduate School of Business, October 1981.
"Institutional Aspects of Geothermal Development," co-authors Tom K. Lee, Venkatraman Sadanand and
Louis L. Wilde. In Geothermal Probabilistic Cost Study, Vol. II (JPL 5030-491). Jet Propulsion
Laboratory, 1981.
"Looking for Villains in the Energy Crisis." In Energy Independence for the United States: Alternative
Policy Proposals, Nake M. Kamrany, ed. Fundamental Books, 1981; and in U.S. Options for Energy
Independence, Nake M. Kamrany, ed. Heath Lexington Books, 1982.
"What Makes Reform Happen?," co-author Bruce Owen. Regulation 7(2) (March/April 1983): 19-24.
"Recent Social Policy: Comment." In Telecommunications Access and Public Policy, Alan
Baughcum and Gerald Faulhaber, eds. Ablex Publishing Corp., 1984.
"The Economic Viability of Professional Baseball: Report to the Major League Baseball Players
Association." In Representing Professional Athletes and Teams, Philip R. Hochberg and Martin E.
Blackman, eds. Practicing Law Institute, 1986.
"The Twisted Pair." Regulation 11(3/4) (1987): 15-22.
"Regulation After Reagan." Regulation 12(3) (1988): 13-20.
"Statement of Goals and Strategies for State Telecommunications Regulation," numerous co-authors.
Aspen Institute, 1989.
"Wirtschaftswachstum, High-Tech-Produkte und internationale Handelspolitik" (in German). In
Deutsch-amerikanische Beziehungen: Politische Freundschaft und wirtschaftliche Kondurrenx?
Haus der Evangelischen Publizistik, 1989.
"The Contemporary Development of International Trade: Approaches, Issues and Problems." In A
Developing World Economy: The Ethics of International Cooperation. Vesper International, 1990.
"Competitive Issues: Enforcement Priorities and Economic Principles." In Antitrust Issues in Regulated
Industries. Charles River Associates, 1991.
"Responding to Referees and Editors." CSWEP Newsletter (February 1993): 15-17. Reprinted in CSWEP
Newsletter Special Reprint Issue No. 2 (1996).
"Biographical Sketches of CSWEP Board Members." CSWEP Newsletter (October 1994).
“Privatization Won’t Foster R&D,” co-author Linda R. Cohen. Public Affairs Report 36(3) (May 1995).
"Constitutional Reform in California," co-author Bruce E. Cain. CPS Brief 7(14) (December 1995).
20
Papular Publications, cont’d
“Taxpayers Foot Bill for Sports Boom.” Brookings Newsletter 7(2) (Autumn 1997): 7.
“Sports, Jobs, and Taxes: Are New Stadiums Worth the Coast?,” co-author Andrew Zimbalist. Brookings
Review 15(3) (Summer 1997): 35-39. Reprinted in Readings in Urban Economics: Issues and Public
Policy, Robert W. Wassmer, ed. Blackwell Publishing, 2000.
“Unleashing Telecommunications: The Case for True Competition,” co-author Robert Litan. Brookings
Policy Brief # 39 (November 1998). Available at www.brookings.org/comm/policybriefs/pb39.htm.
“Telecommunications Reform in Romania.” In Romania: Regulatory and Structural Assesment in the
Network Utilities, Ioannis Kessides, ed. World Bank Report 20546 RO, 2000.
“Is U.S. Science Policy at Risk?”, co-author Linda R. Cohen. Brookings Review 19(1) (Winter
2001): 10-15.
“Broadband Telecommunications Policy: Ending the Chaos.” SIEPR Policy Brief, December 2001.
“The Supreme Court’s Decision on FCC Pricing Rules,” co-author Gregory Rosston. SIEPR Policy Brief,
May 2002.
“The FCC’s New Television Ownership Rules.” SIEPR Policy Brief, June 2003.
“The Uncertain Future of the Telecommunications Industry,” co-author Robert E. Litan. Brookings Policy
Brief #129, January 2004.
“The Painful Implementation of California’s Stem Cell Research Program,” SIEPR Policy Brief, October
2005.
“The Foreign Aid Paradox.” SIEPR Policy Brief, October 2006.
“For More Efficient Subsidy Schemes,” co-author T. N. Srinivasan. Hindu Business Line, April 27, 2006.
(Originally “More Efficient Subsidy Scheme Benefits Consumers, Government, and Economy,” SIEPR
Policy Brief, May 2006.)
“The Regulatory Component of Health Care Reform.” Fresh-Thinking Project, November 2007.
“Designing It Right,” co-author Joe Nation. Environmental Finance vol. 10, no. 2 (December 2008/January
2008), p. 49.
“”Toward a 21st Century Health Care System: Recommendations for Health Care Reform,” 49 co-authors.
Annals of Internal Medicine Vol. 150, No. 7 (April 7, 2009), pp. 493-5.
“Competitive Implications of the Proposed Acquisition of T-Mobile by AT&T Mobility,” co-author
Gregory Rosston. SIEPRPolicy Brief, March 2011.
“Would California Be Better Off Without Ballot Initiatives? No – but Make a Lot of Fixes,” Up For
Discussion, October 2011, Zocalo Public Square, at http://zocalopublicsquare.org/thepublicsquare/2011
/10/04/this-doggone-direct-democracy/read/up-for-discussion/.
21
OTHER PROFESSIONAL PAPERS
"An Economic Analysis of Network Operations Research Techniques." Ph.D. dissertation, Harvard
University, 1967.
"Perspectives on Rural-Urban Migration." Council of Economic Advisors, November 1968.
"The Economics of Pollution Abatement." Presented at Annual Meeting of the American Association for
the Advancement of Science, December 1970.
"Comments Regarding the Public Interest in Commission Rules and Regulations Relating to Cable
Television, Signal Importation and the Development of UHF Independent Commercial Stations," coauthors John J. McGowan and Merton J. Peck. Submitted to FCC Docket 18397-A, February 10, 1971.
"A Dynamic Theory of Political Campaigning," co-author John A. Ferejohn. Annual Meeting of the
American Political Science Association, September 1972, Washington, D.C.
“Comments Regarding Limitations on Programming Available for Broadcast on Pay-TV Channels.”
Submitted to FCC Docket 19554. Social Science Working Paper No. 65, California Institute of
Technology, September 20, 1974.
"Statement on Regulatory Reform." In Regulatory Reform - 1975: Hearings before the Committee on
Government Operations, U.S. Senate, 94th Cong., 1st Sess., 1975.
"The Causes of Regulatory Failures." In Oversight of Civil Aeronautics Board Practices and Procedures
Hearings before the Subcommittee on Administrative Practices and Procedures, Senate Committee on the
Judiciary, 94th Cong., 1st Sess., 1975.
"Responses to Disaster: Planning for a Great Earthquake in California," co-authors Linda Cohen and Barry
Weingast. Social Science Working Paper No. 131, California Institute of Technology, April 1977.
"Statement on Public Policy Toward Sports." Hearings: Select Committee on Professional Sports, U.S.
House of Representatives, September 1976.
"Statement on H.R. 11611." Drug Regulation Reform Act of 1978: Hearings before the Subcommittee on
Health and the Environment, U.S. House of Representatives, June 1978, p. 2156ff.
“Television and Competition.” Symposium on Media Concentration, Federal Trade Commission,
December 1978.
"The Economics of Boxing Regulation in California," co-authors Joel A. Balbien and James P. Quirk.
Social Science Working Paper No. 366, California Institute of Technology, January 1981. Report to
California Athletic Commission.
Implementing Tradable Emissions Permits for Sulfur Oxides Emissions in the South Coast Basin (three
volumes), co-authors Glen Cass and Robert Hahn. Report to California Air Resources Board. Caltech
Environmental Quality Laboratory, June 1983.
"Economic Effects of the Financial Interest and Syndication Rule," co-authors Robert Crandall and Bruce
Owen. Economists, Inc., April 1983. Submitted to Federal Communications Commission, Inquiry on
Television Networks.
22
Papular Publications, cont’d
"Pay and Performance in Baseball: Modeling Regulars, Reserves and Expansion," co-author Rodney D.
Fort. Social Science Working Paper No. 527, California Institute of Technology, Pasadena, CA 1984.
"Promises, Promises: Campaign Contributions and the Reputation for Services," co-author John Ferejohn.
Social Science Working Paper No. 545, California Institute of Technology, Pasadena, CA, 1984.
“The Economic Viability of Professional Baseball: A Report to the Major League Baseball Players’
Association.” Major League Baseball Players’ Association, 1985.
"Local Telephone Prices and the Subsidy Question," co-author Nina W. Cornell. Presented at
Annual Meetings of the American Economic Association, December 1984, and at the Conference
on Telecommunications Demand Modeling, Bell Communications Research, October 1985.
"The Economics of Bell Operating Company Diversification in the Post-Divestiture Telecommunications
Industry," co-authors Kenneth Baseman and Stephen Silberman. ICF, Incorporated, September 1986.
Submitted in First Triennial Review, Modification of Final Judgment, U.S. vs. AT&T.
"Two's Company, Three's a Crowd: Duverger's Law Reconsidered," co-author John A. Ferejohn. Presented
at Annual Meeting of the American Political Science Association, September 1987.
"Telecommunications Reform in Brazil." Report to the International Bank for Reconstruction and
Development, September 1990.
"Marketable Emissions Permits in Los Angeles." Report to the South Coast Air Quality Management
District. Center for Economic Policy Publication No. 285, Stanford University, January 1991.
"Statement on the Baseball Antitrust Exemption." Hearings before the Subcommittee on Antitrust,
Monopolies, and Business Rights, Committee on the Judiciary, United States Senate, December 1992.
“Regulatory Transparency in Telecommunications: Public Interest Standards for BOC Entry into Long
Distance,” co-author Robert Litan, August 1998. Submitted to FCC Docket CCBPol 98-4.
“Remedies Brief of Amicus Curie,” co-authors Robert Litan, William Nordhaus and Frederic
Scherer. U.S. v. Microsoft, U. S. District Court, District of Columbia, April 2000. Available at:
www.aei.brookings.org/publications/index.php?tab=author&authorid=14.
“ Promoting Efficient Use of Spectrum Through Elimination of Barriers to the Development of Secondary
Markets,” 36 co-authors. Submitted to Federal Trade Commission, February 2001.
“Notes on Privatizing Infrastructure Industries.” World Bank Development Report Planning Conference,
July 2001.
“Comments on Revised Proposed Final Settlement,” co-authors Robert Litan and William Nordhaus. U.S.
v. Microsoft, U. S. District Court, District of Columbia, January 2002.
“The Copyright Term Extension Act of 1998: An Economic Analysis,” Amicus Brief in Support of
Petititoners, Eldred, et al., vs. Aschroft, U. S. Supreme Court, May 2002, sixteen co-authors.
“Comments on the Federal Trade Commission’s Strategic Plan for 2003,” co-authors Robert W. Hahn and
Robert E. Litan. AEI-Brookings Joint Center for Regulatory Studies, July 2003.
23
Other Publications, cont’d
“Brief of Amici Curiae,” 32 co-authors PSEG Fossil, et al., vs. Riverkeeper Inc., U. S. Supreme Court, July
21, 2008.
“Comments of 71 Concerned Economists: Using Procurement Auctions to Allocate Broadband Stimulus
Grants,” National Telecommunications Information Agency and Rural Utilities Service, April 13, 2009.
“A Statement on the Appropriate Role for Research and Development in Climate Policy,” 9 co-authors,
Berkeley Electronic Press, December 9, 2008.
“Amicus Curiae Brief in Support of Petitioner,” 19 co-authors, American Needle vs. National Football
League, U. S. Supreme Court, September 25, 2009.
“Regionalizing Telecommunications Reform in West Africa,” co-authors Ioannis N. Kessides and Nancy
C. Benjamin. Policy Research Working Paper No. 5126, World Bank, November 2009.
“Comment of Sports Economists on the FCC’s Sports Blackout Rules,” eight co-authors, submitted in
Docket MB 12-3, Federal Communications Commission, February 2012.
POPULAR PUBLICATIONS
"School Bonds and the Future of Pasadena." Pasadena Star-News (April 20, 1969): C-1.
"After Vietnam, Another Recession?" Caltech News (June 1969).
"People Is a Dirty Word," (with others). Pest Control Operators News 30 (February/March 1970).
(Transcript of panel discussion, radio station KMPC, Los Angeles.)
"Defending Against Disasters." Engineering and Science 39 (May-June 1976).
"Quake Prediction: For Public Officials the Problems Are Mind-Bending." Los Angeles Times (May 2,
1976), Part VIII: 5.
"Professional Sports: Should Government Intervene?" San Francisco Chronicle (June 7, 1977).
"Fact and Fancy About the Energy Crisis." Pasadena Star-News (July 27, 1980): 17.
"Looking for Villains in the Energy Crisis." Town Hall Reporter 13(8) (August 1980): 12.
"Using the Marketplace to Reform Regulation." Washington University Center for the Study of American
Business, Whittemore House Series 5.
"Leasing the Air: An Alternative Approach to Regulation?" Engineering and Science 46(1) (September
1982): 12-17.
"Television Ownership and the FCC: Let a Free Market Set the Pace." New York Times (August 26, 1984),
Business Section: 2.
"Trends in California's Economy: Implications for the Future." Engineering and Science LVI (1) (Fall
1992): 9-13
24
Papular Publications, cont’d
"Help Business, but Beware of High-tech Pork," co-author Linda R. Cohen. USA Today, March 18, 1993,
p. 11A.
"The Decline of Public Support for R&D." Framtider International, Vol. 5 (1995): 23-27.
"Wild Pitch." New York Times, April 11, 1996. Reprinted in New York Times Op-Classic, 2008.
“Revisiting Telecom Reform,” co-author T. N. Srinivasan. Business Standard of India, August 21-22,
1999.
“32 Voices on the State of the Game,” 31 co-authors. The Biz of Baseball, December 15, 2006.
“Sharing a Stadium Makes Perfect Sense.” Oakland Tribune, February 9, 2007, Sports Section: 3.
“Six Views on Former Commissioner Bowie Kuhn,” five co-authors. The Biz of Baseball, March 26, 2007.
“Why Analysis of 49ers Move is Too Rosy.” San Jose Mercury News, April 15, 2007: 1P.
“Baseball Economics Roundtable,” six co-authors. The Biz of Baseball, May 3, 2007.
“Are Stadiums Worth the Price?” San Francisco Chronicle, July 8, 2007: E1, E3.
“Voices on Barry Bonds,” ten co-authors. The Biz of Baseball, November 27, 2007.
“Is Tim Gaithner’s Toxic Asset Plan Toxic?” seven co-authors. Foreign Policy, March 23, 2009.
“A Few Thoughts about Measure J,” San Francisco Chronicle, May 16, 2010.
REVIEW ARTICLES
"Assessing the Energy Situation." Science, 208(4445) (May 16, 1980): 701-702.
"Energy Data and Political Polarization." Science, 214(4524) (November 27, 1981): 1019.
"Handbook for Reform: Breyer on Regulation." Columbia Law Review, 83(4) (May 1983): 1109-1119.
"Rules in the Making, by Magat, Krupnick, and Harrington." Rand Journal of Economics, 18(3) (Autumn
1987): 461-464.
"Fields of Greed." New York Times Book Review (April 4, 1993): 24-25.
BOOK REVIEWS
"Cure for Chaos, by Simon Ramo." Engineering and Science 33 (November 1969).
"Technology and Market Structure, by Almarin Phillips." Journal of Economic Literature 10 (December
1972): 1253-1255.
"The Telecommunications Industry, by Gerald W. Brock." Journal of Economic Literature 20 (September
1982): 1096-97.
25
Book Reviews, cont’d
"Presidential Management of Science and Technology, by W. Henry Lambright." Science, 231(4739)
(February 14, 1986): 749.
"Telecommunications Economics and International Regulatory Policy, by Snow and Jusawalla."
Information Economics and Policy, 2(4) (1986): 318-319.
"The Economist's View of the World, by Steven E. Rhoads." American Political Science Review 81(1)
(March 1987): 339-341.
"Air Pollution Regulation, by Richard A. Liroff." Environmental Professional 9(4) (1987): 359-360.
"The Sports Industry and Collective Bargaining, by Paul D. Staudohar." Industrial and Labor Relations
Review 41(2) (January 1988): 314-315.
"The Social Context of New Information and Communication Technologies, by Elia Zureik and Dianne
Hartling." Information Economics and Policy 3(2) (1988): 204.
"The United States and the Direct Broadcast Satellite, by Sara Fletcher Luther." Information Economics
and Policy 4(1) (1989/90): 83-84.
"Regulation and Markets, by Daniel F. Spulber." Journal of Economic Literature 28(4) (December 1990):
1757-1759.
"A Legislative History of the Communications Act of 1934, edited by Max Paglin." Information Economics
and Policy 4(2) (1990): 190-94.
"International Telecommunications in Hong Kong: The Case for Liberalization, by Milton Mueller."
Information Economics and Policy 4(3) (1990): 273-274.
"Strategy and Choice, edited by Richard J. Zeckhauser." Journal of Policy Analysis and Management,
12(2) (Spring 1993): 386-388.
"Risky Business: Breaking the Vicious Circle, by Stephen Breyer." Regulation, 16(3) (1993): 82-83.
"Government's Role in Innovation, by D.P. Leyden and A.N. Link." Journal of Economics (Zeitschrift für
National Okönomie) 54(3) (1994): 333-335.
"Regulation, Organizations, and Politics, by Lawrence S. Rothenberg." American Political Science
Review 89(3) (September 1995): 768-769.
“The Political Economy of Public Administration: Institutional Choice in the Public Sector, by Murray J.
Horn.” Economic Record 73 (June 1997): 187-188.
"
“Making and Breaking Governments, by Michael Laver and Kenneth A. Shepsle.” Zeitschrift fur
Nationalokonomie (Journal of Economics) 66(3) (December 1997): 324-326.
"
“The Economics of Sports Broadcasting, by Chris Gratton and Harry Arne Solberg.” Journal of Media
Economics Vol. 23, No. 1 (2010): 42-5.
“Handbook of International Sport Law, by James A. R. Nafziger and Stephen F. Ross.” Journal of Sports
Economics Vol. 14, No. 3 (June 2013), pp. 330-33.
9/13
26
APPENDIX B
11:35 Monday, November 25, 2013
1
11:35 Monday, November 25, 2013
2
11:35 Monday, November 25, 2013
3
11:35 Monday, November 25, 2013
4
11:35 Monday, November 25, 2013
5
11:35 Monday, November 25, 2013
6
11:35 Monday, November 25, 2013
7
11:35 Monday, November 25, 2013
8
11:35 Monday, November 25, 2013
9
11:35 Monday, November 25, 2013
10
11:35 Monday, November 25, 2013
11
11:35 Monday, November 25, 2013
12
11:35 Monday, November 25, 2013
13
11:35 Monday, November 25, 2013
14
11:35 Monday, November 25, 2013
15
11:35 Monday, November 25, 2013
16
11:35 Monday, November 25, 2013
17
11:35 Monday, November 25, 2013
18
11:35 Monday, November 25, 2013
19
11:35 Monday, November 25, 2013
20
11:35 Monday, November 25, 2013
21
11:35 Monday, November 25, 2013
22
11:35 Monday, November 25, 2013
23
11:35 Monday, November 25, 2013
24
11:35 Monday, November 25, 2013
25
11:35 Monday, November 25, 2013
26
11:35 Monday, November 25, 2013
27
11:35 Monday, November 25, 2013
28
11:35 Monday, November 25, 2013
29
11:35 Monday, November 25, 2013
30
11:35 Monday, November 25, 2013
31
11:35 Monday, November 25, 2013
32
11:35 Monday, November 25, 2013
33
11:35 Monday, November 25, 2013
34
11:35 Monday, November 25, 2013
35
11:35 Monday, November 25, 2013
36
11:35 Monday, November 25, 2013
37
11:35 Monday, November 25, 2013
38
11:35 Monday, November 25, 2013
39
11:35 Monday, November 25, 2013
40
11:35 Monday, November 25, 2013
41
11:35 Monday, November 25, 2013
42
11:35 Monday, November 25, 2013
43
11:35 Monday, November 25, 2013
44
11:35 Monday, November 25, 2013
45
11:35 Monday, November 25, 2013
46
11:35 Monday, November 25, 2013
47
11:35 Monday, November 25, 2013
48
11:35 Monday, November 25, 2013
49
11:35 Monday, November 25, 2013
50
11:35 Monday, November 25, 2013
51
11:35 Monday, November 25, 2013
52
11:35 Monday, November 25, 2013
53
11:35 Monday, November 25, 2013
54
11:35 Monday, November 25, 2013
55
11:35 Monday, November 25, 2013
56
11:35 Monday, November 25, 2013
57
11:35 Monday, November 25, 2013
58
11:35 Monday, November 25, 2013
59
11:35 Monday, November 25, 2013
60
11:35 Monday, November 25, 2013
61
11:35 Monday, November 25, 2013
62
11:35 Monday, November 25, 2013
63
11:35 Monday, November 25, 2013
64
11:35 Monday, November 25, 2013
65
11:35 Monday, November 25, 2013
66
11:35 Monday, November 25, 2013
67
11:35 Monday, November 25, 2013
68
11:35 Monday, November 25, 2013
69
11:35 Monday, November 25, 2013
70
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
*APPLE'S (PROPOSED) REDACTIONS*
EXHIBIT 3
[Filed Under Seal]
UNITED STATES DISTRICT COURT
NORTHERN DISTRICT OF CALIFORNIA
OAKLAND DIVISION
____________________________________
|
|
|
LITIGATION
|
____________________________________|
APPLE IPOD ITUNES ANTI-TRUST
Lead Case No. C-05-00037-YGR
SUPPLEMENTAL REBUTTAL DECLARATION OF
ROGER G. NOLL ON LIABILITY AND DAMAGES
My name is Roger G. Noll, and previously I have submitted nine declarations in
this proceeding.1 My previous declarations contain my career history and qualifications.
ASSIGNMENT AND SUMMARY
Attorneys for the class plaintiffs in this litigation have asked me to review the
Supplemental Report of Kevin M. Murphy and Robert H. Topel (December 20, 2013,
henceforth Joint Report) to determine if anything in this report causes me to alter any of
the opinions I expressed in either the Noll Merits Report or the Noll Merits Rebuttal.
The answer to this question is simple. The Joint Report contains no new
arguments and no valid new data analysis, so nothing in that report requires me to change
any of my analysis or conclusions.
1
Declaration of Roger G. Noll (July 15, 2008), Reply Declaration of Roger G. Noll
(October 19, 2009), Declaration of Roger G. Noll (January 18, 2011), Reply Declaration
of Roger G. Noll (March 28, 2011), Supplemental Declaration of Roger G. Noll (July 18,
2011), Second Supplemental Declaration of Roger G. Noll (September 23, 2011),
Declaration of Roger G. Noll on Liability and Damages (April 13, 2013, henceforth Noll
Merits Report), Corrections to the Declaration of Roger G. Noll on Liability and
Damages (May 13, 2013), and Rebuttal Declaration of Roger G. Noll on Liability and
Damages (November 25, 2013, henceforth Noll Merits Rebuttal).
The Joint Report perpetuates econometric mistakes that Professors Murphy and
Topel have made in prior reports. These mistakes are as follows. (1) The assertion that
the regression in the Noll Merits Rebuttal must be corrected to account for clustering
effects. (2) The claim that more independent variables should be added to a regression
equation if they add any explanatory power, even if these variables are highly collinear
with other variables in the equation. Repeated assertion does not make an incorrect
econometric argument correct, and Professors Murphy and Topel remain misinformed
about the appropriate econometric technique for estimating a reliable hedonic price
equation from Apple’s iPod transactions data.
The Joint Report claims that correlation among the residual errors when iPod
transaction data are grouped by quarters implies that the coefficients in the hedonic price
regressions in my prior reports suffer from a clustering problem that causes the results to
be unreliable. This claim remains false. The method that Professors Murphy and Topel
use to detect autocorrelation among the residuals remains just as incorrect as it was in
prior reports by defendant’s experts.
The method that defendant’s experts use to “correct” for this problem remains just
as incorrect as it was previously. Their method is equivalent to replacing observations on
transactions prices with quarterly average prices. The main effect of this procedure is
vastly to reduce the precision of the regression equation, thereby causing the coefficients
of the iTunes 7.0 variables to become insignificant. Because Professors Murphy and
Topel have no valid reason for picking calendar quarters as the basis for creating clusters,
these results are meaningless. Indeed, if one adopts a week as the time period for
creating clusters, which also shows non-zero mean residual errors by family/week, the
2
coefficient on the iTunes 7.0 variable is again significant for direct sales. Professors
Murphy and Topel have no basis for picking calendar quarters over weeks as the basis for
creating groups of observations, and so no reason to prefer their results to results from a
shorter time period.
The Joint Report continues to insist that the indicator variable for iTunes 4.7
should remain “on” for the iPod models that included iTunes 7.0, and adds the argument
that the new treatment of iTunes 7.0 is inconsistent with my former testimony.
Defendant’s experts give no new reason for leaving iTunes 4.7 turned “on” for models
that did not include iTunes 4.7 after the introduction of iTunes 7.0.
CLUSTERING
The Joint Report repeats the argument that the error terms in my regressions are
correlated within “clusters” (iPod product families in a quarter).2 As a baseline matter,
there is no clustering problem among iPod transactions. As explained by Professor
Wooldridge, clustering problems occur when data are drawn from a sample of a
population of groups, or clusters.3 Professors Murphy and Topel are guilty of ex post
clustering, which occurs when groups in either a sample or a population are identified as
belonging to a certain group, which is then after-the-fact labeled a cluster.4
2
Joint Report, pp. 4-5.
3
Declaration of Jeffrey M. Wooldridge in Support of Plaintiff’s Daubert Motion to
Exclude Certain Opinion Testimony of Robert H. Topel and Kevin M. Murphy
(henceforth, Wooldridge Declaration), December 20, 2013, p. 3.
4
Wooldridge Declaration p. 3. At his deposition, Professor Wooldridge highlighted the
issue of ex post clustering in the context of estimating the average hourly wage from a
random sample of data. He indicates that there are many ways to cluster such data (e.g.,
by occupation or by education level), but none is correct if the underlying data have been
gathered randomly so there is no cluster sample. Deposition of Jeffrey Wooldridge, p. 40.
3
In a footnote5 Professors Murphy and Topel assert that there is nothing to suggest
that clustering is limited to samples, and cite one of the references that I quoted in the
Noll Rebuttal. This statement is incorrect and misleading.
First, in the chapter of the book by Professors Angrist and Pischke that I cited, the
cluster example that the authors use is the STAR experiment, which was based on a
sample of classrooms (the clusters are classes). Footnote 10 of this chapter explains that
“most of the samples [emphasis added] that we work with are close enough to random
that we typically worry more about the dependence due to group structure [a reference to
excluded independent variables that are correlated with groups] than clustering due to
stratification.”6 These authors also state, in reference to their whimsical statement that
one can stop worrying about cluster effects when the number of clusters is 42 or more:
“If 42 is enough for the standard cluster adjustment to be reliable, and if less is too few,
then what should you do when the cluster count is low? First best is to get more clusters
by collecting more data (emphasis added).”7 Getting more clusters makes sense only if
the original number of clusters is less than the population of clusters. In short, the
chapter on cluster analysis in the book by Professors Angrist and Pischke is about
problems arising from cluster samples. To suggest otherwise is misleading.
Second, all of the other references that I cited in past reports refer to cluster
samples. The fact that Professors Angrist and Pischke did not make extensive use of the
word “sample” does not imply that cluster effects are not a sampling problem. Thus, the
5
Joint Report, footnote 21, p. 7.
6
Joshua D. Angrist and Jörn-Steffen Pischke, Mostly Harmless Econometric: An
Empiricists Companion, Chapter 8, Princeton University Press, 2009, p. 309.
7
Ibid, pp. 319-20.
4
claim in the Joint Report that clustering is not related to sampling errors is simply false.
Professors Murphy and Topel motivate their assertion that the iPod transactions
data exhibit a clustering problem by presenting a correlation analysis of residual errors.
Professors Murphy and Topel calculate the estimated residual for each transaction in each
data set (reseller or direct sales), and group these residual errors into families of iPods in
each quarter. For this procedure to make sense, some unobserved variable must be
present that affects prices in a family of iPods differently in different quarters. Professors
Murphy and Topel do not explain what this variable might be, or why the appropriate
time period is a calendar quarter, rather than a day, a week, or a year. In short, their
choice of calendar quarters as the basis for dividing the data into clusters is arbitrary.
After creating these unmotivated quarterly groups of residual errors by family,
Professors Murphy and Topel divide the residuals into two equal-size groups within each
product family/quarter group, calculate the average residual within each half, and plot the
mean residuals from each group against each other (Exhibits JT-4a and JT-4b of the Joint
Report). According to Professors Murphy and Topel, these graphs demonstrate that the
residual errors are correlated.
Professor Murphy’s and Professor Topel’s analysis is unreliable because the
groups into which they divide the observations are artificially created. The specification
of the price regressions guarantees that each combination of class and capacity of an iPod
will have a mean residual error of zero. If one then divides these residual errors into time
periods, the mean errors of these groups taken together still will be zero, but as a
statistical matter the mean residual errors for every such group is not expected to be zero.
Instead, the values of the mean residual errors among the groups will be randomly
5
distributed around zero, with some values above zero and some below.
To demonstrate that the results in the Joint Report have nothing to do with
whether a calendar quarter is somehow an appropriate basis for clustering the data, the
procedures that were used in the Joint Report were applied to two other ways to divide
the data by time period: by family/month and by family/week. The results of these
alternate specifications are shown in Exhibits 1a-b and 2a-b, and are qualitatively
identical to the family/quarter results in the Joint Report.8 As Exhibits 1a-b and 2a-b
show, using the same method that Professors Murphy and Topel use, residuals also are
correlated among these smaller groups. This is exactly the result one would expect. A
similar result would be observed if one plotted mean residual errors for each family/year
or family/day, or any other time period short of the full data period.
These results show that there is no basis in the correlation analysis for picking
quarterly data as the hypothetical clusters, as opposed to weekly or monthly data, or even
daily or yearly data. The correlations of residual errors are not due to the standard
problem of cluster samples – the presence of unobserved variables that affect outcomes
differentially among the groups – but are the expected result of dividing the residual
errors into a large number of arbitrarily selected groups.
The Joint Report “corrects” the non-existent clustering problem by grouping the
transactions data by quarter. This procedure is equivalent to vastly reducing the number
of observations in the regression, thereby vastly reducing the degrees of freedom and
vastly increasing the standard errors of the regression coefficients. Choosing ex-post
clusters by time period at any level of granularity is purely ad hoc. A family/quarter
8
Exhibits 1a (monthly) and 2a (weekly) show the results from the regression on the
reseller data, while 1b and 2b show the results from the direct sales regression.
6
include a large number of product characteristics that affect iPod prices, regressions with
any additional product attributes should be tested for multicollinearity. Professors
Murphy and Topel did not do so in either their original merits reports or the Joint Report.
The adjusted R2, “variance inflation factor” (VIF) and “condition number”, explained
below, are statistics that indicate when multicollinearity is likely to cause coefficients of a
regression model to be inaccurate and imprecisely measured.14 Professors Murphy and
Topel did not consider these statistics, and the statistics they do report are misleading.
Professors Murphy and Topel assert four justifications for including the additional
iPod attributes in the regressions (paragraph references are to the Joint Report):
1.
2.
3.
4.
With one or two exceptions (depending on the regression) the additional iPod
attributes are individually statistically significant (¶17).
The additional iPod attributes are jointly significant (¶17).
The additional iPod attributes increase the R2 of the reseller and direct sales
regressions, and not by a “small” amount (¶18).
Including the additional product attributes does not substantially increase the
estimated standard error of the iTunes 7.0 coefficient, as would occur if the
iTunes 7.0 variable were affected by multicollinearity. (¶21).
Professors Murphy and Topel do not explain how the first three statements justify
ignoring the problem multicollinearity should it exist. Even if variables are individually
or jointly significant and add to the explanatory power of a regression, that alone is
insufficient justification to include them in the regression. Defendant’s class certification
expert, Dr. Michelle Burtis, showed that events irrelevant to iPod pricing could be
statistically significant in a regression model.15 One needs a reason, derived from
economic theory or econometrics, to add variables to a regression, especially if these
Merits Report. See Deposition of Robert H. Topel, November 8, 2013, pp. 28-29.
14
See William H. Greene, Econometric Analysis (4th ed.), Prentice Hall 2000, pp. 257258 for a discussion of collinearity diagnostics.
15
Second Supplemental Report of Dr. Michelle M. Burtis, November 14, 2011, p. 20 and
Exhibit 7.
9
variables create a problem with reliably estimating the equation.
When economic theory provides a reason to believe these variables might be
causally related to the dependent variable, if the additional attributes create a
multicollinearity problem, the problem must be addressed. The two methods for
addressing multicollinearity are to add more data (here not possible because the data are a
population, not a sample) and to eliminate one or more collinear variables.
Professors Murphy and Topel assert without proof that the regressions that they
ran do not suffer from multicollinearity. Tests for the presence of multicollinearity
applied to the models reported in the Joint Report (Exhibits JT-3a and JT-3b column 4)
show that, in fact, the variables that defendant’s experts add to the regression model
cause severe multicollinearity.
Three statistical tests are used to detect multicollinearity: adjusted R-squared,
condition number, and the variance inflation factor (VIF). The adjusted R-squared test,
when used in an “auxiliary” regression of a potentially multicollinear variable on the
other explanatory variables in a model, measures correlations among the variables. The
condition number measures how sensitive the output of a model is to small changes in
data. The VIF measures the severity of multicollinearity in a regression.
All but one of the iPod attributes that Professors Murphy and Topel add to the
regressions in the Noll Merits Rebuttal generate a multicollinearity problem.16 Each of
the iPod attributes that were added by Professors Murphy and Topel17 were regressed on
16
The exception is whether the iPod transaction was an original equipment manufacturer
(“OEM”) sale to Hewlett-Packard, an attribute applicable to reseller but not direct sale
transactions. Hewlett-Packard OEM transactions have a high condition number but low
adjusted R2 and VIF.
17
“Model 4” refers to the model reported in column 4 of the Joint Report Exhibits JT-3a
10
the other variables that were included in the regressions in the Noll Merits Rebuttal. The
adjusted R2, condition number, and VIF from each of these “auxiliary” regressions is
reported in Exhibit 3 and Exhibit 4. If the R2 from an auxiliary regression is high and
similar to the R2 of the main equation to be estimated, then multicollinearity is severe.18
The R2’s in the regressions in the Joint Report are 0.9930 (resellers) and 0.9866 (direct
sales). As shown in Exhibits 3 and 4, the R2’s from auxiliary regressions for all of the
additional product attributes are highly similar to the R2 of the main equation to be
estimated. The lowest R2 among auxiliary regressions in the reseller data (Exhibit 3) is
0.9767 (recharge hours). The lowest R2 among auxiliary regressions in the direct sales
data (Exhibit 4) is 0.9728 (recharge hours).
The condition number associated with each auxiliary regression indicates when
the coefficient estimates in the underlying regression are likely to be affected by
multicollinearity.19 A condition number exceeding 10 in a model with a large number of
independent variables is regarded as indicating that multicollinearity is likely to be a
serious problem.20 The smallest condition number in Exhibits 3 and 4 is 1548. These
and JT-3b. This model includes the additional iPod attributes. I previously performed a
similar analysis using the models in the initial Murphy Report and Topel Report with the
difference that each auxiliary regression included only the explanatory variables in the
Noll Merits Report and not the other product attributes added by Professors Murphy and
Topel. Results were similar.
18
Klein, L., An Introduction to Economics, Prentice-Hall (1962), as reported in Greene
op cit. p. 258.
19
Greene p. 258 and SAS v 9.3 documentation (http://support.sas.com/documentation/
cdl/en/statug/63962/HTML/default/viewer.htm#statug_reg_sect036.htm).
20
The condition number is not a reliable indicator of serious multicollinearity when the
number of independent variables is very small (two or three). See Alexis Lazaridis, “A
Note Regarding The Condition Number: The Case of Spurious and Latent
Multicollinearity,” Quality and Quantity Vol. 41, No. 1 (February 2007), pp. 123-35.
11
statistics indicate that each of the additional product attributes generates multicollinearity
problems in the underlying reseller or direct sales regression.
The VIF of each auxiliary regression is equal to 1/(1-Rk2), where Rk2 is the R2
from the auxiliary regression of the independent variable xk on the other variables
included in the underlying model. The VIF measures the inflation in the variances of the
parameter estimates in the regression due to multicollinearity among the explanatory
variables.21 Although the common “rule of thumb” is that a VIF that exceeds 10 is a
caution that multicollinearity presents a serious problem, this rule is not definitive
because the effect on the variance of the regression coefficient by itself is not sufficient to
know whether multicollinearity is sufficient to undermine the quality of the regression,
which in turn depends on the estimated t-statistic for the coefficient and whether the goal
of the regression is to test a hypothesis (e.g., is a coefficient positive?) or to estimate the
quantitative significance of a variable (as is the goal here).22 Because multicollinearity
expands the confidence range of plausible values for a regression coefficient, its effect
when the goal is a point estimate is more serious than when the goal is hypothesis testing.
Thus, the “rule of 10” is more appropriate when, as here, the goal is estimating a
regression coefficient precisely. And the values of the VIF for all of the variables that
have been added by Professors Murphy and Topel are far above 10. The smallest VIF in
Exhibits 3 and 4 is 37 (recharge hours in the direct sales data) and the highest VIF is 1429
(resolution in the reseller data). These values indicate that all of the additional variables
cause a multicollinearity problem in the main reseller or direct sales regression.
21
SAS v9.3 online documentation, http://support.sas.com/documentation/cdl/en/statug/
63962/HTML/default/viewer.htm#statug_reg_sect036.htm.
22
See Robert M. O’Brien, “A Caution Regarding Rules of Thumb for Variance Inflation
Factors,” Quality and Quantity Vol. 41, No. 5 (October 2007), pp. 673-90.
12
Finally, even adding these variables to the regression causes estimated damages to
fall, but not dramatically. The back-up Appendixes to the Joint Report contain
regressions in which nothing was changed from the regressions in the Noll Merits
Rebuttal other than to add the additional variables.
In conclusion, nothing in the Joint Report supports the conclusion that additional
attribute variables improve the quality of the regression. Adding these attributes to the
model worsens the precision of the estimates and renders the estimates sensitive to small
changes in the data and in the specification of the regression equation. The only question
is whether the quality of the model would be improved by substituting some variable in
the original equation with another variable.
iTUNES 7.0
The Joint Report again asserts that the indicator variable for iTunes 4.7 should
remain “on” for transactions of iPods that used iTunes 7.0. The essence of the argument
The Joint Report states that one reason the additional variables should be added is that
they increase R2. The additional attributes increase R2 by small amounts: 0.0062 in the
reseller regression and 0.0050 in the direct sales regression. The goal of a regression
analysis is not to maximize R2. The problem with using a gain in R2, however small, as
the sole criterion for evaluating whether variables should be added is that the goal of the
regression is to obtain reliable, precise estimates of the coefficients on iTunes 7.0. If the
additional variables create multicollinearity that undermines the precision of the estimates
of the coefficients, the goal of the regression is sacrificed for a small change in R2.
23
13
by Professor Murphy and Topel is that iTunes 4.7 was legal and could have been used in
these models, so that the proper estimation method is to turn on both variables so that the
coefficient on iTunes 7.0 measures the incremental harm due to iTunes 7.0.
14
*APPLE'S (PROPOSED) REDACTIONS*
EXHIBIT 54
[Filed Under Seal]
UNITED STATES DISTRICT COURT
NORTHERN DISTRICT OF CALIFORNIA
OAKLAND DIVISION
____________________________________
|
APPLE IPOD ITUNES
|
ANTITRUST LITIGATION
|
____________________________________|
Lead Case No. C-05-00037-YGR
DECLARATION OF JEFFREY M. WOOLDRIDGE IN SUPPORT OF PLAINTIFF’S
DAUBERT MOTION TO EXCLUDE CERTAIN OPINION TESTIMONY OF
ROBERT H. TOPEL AND KEVIN M. MURPHY
1. Introduction
My name is Jeffrey M. Wooldridge. I am a University Distinguished Professor of
Economics at Michigan State University (MSU), where I have taught since 1991. My first
academic appointment was as an assistant professor in the economics department at the
Massachusetts Institute of Technology, where I taught from 1986 to 1991.
2. Qualifications
I received bachelor’s degrees in computer science and economics from the University of
California, Berkeley, in 1982, and I received my doctorate in economics from the University of
California, San Diego, in 1986. I am trained as an econometrician and have taught and done
research on econometric methods for almost 30 years. One of my areas of expertise is problems
associated with nonrandom sampling, including missing data, stratified sampling, and cluster
sampling. I have written several articles about these subjects, and have given many lectures and
short courses on approaches to handling nonrandom samples. I have written two popular
textbooks in econometrics, an introductory book and a graduate-level text, Econometric Analysis
of Cross Section and Panel Data Methods (2010, MIT Press, 2e). I use my MIT Press book for a
1
CONFIDENTIAL – ATTORNEYS EYES ONLY
year-long course in microeconometric methods for second-year Ph.D. students, and I have based
several short summer school courses on the contents of the book.
I am a Fellow of the Econometric Society and of the Journal of Econometrics. I have
served on several editorial boards, including as editor of the Journal of Business and Economic
Statistics from 1998 to 2000, as co-editor of Econometric Theory from 2003 to 2005, and as
econometrics co-editor of Economics Letters from 1995 to 1998.
I have served as an econometrics consultant several times, including work for Charles
Rivers Associates when I taught at MIT, Arthur Andersen, Deloitte Consulting, and, currently,
for Industrial Economics, Inc, on a damage assessment project for the National Oceanic and
Atmospheric Administration.
Other evidence of my qualifications as an expert in econometrics generally and on
sampling issues specifically is given in my curriculum vita, which is included as Appendix A.
2. Assignment
I have been asked by the attorneys for the plaintiffs in this litigation to evaluate and
reconcile differing claims about the proper calculation of standard errors in separate expert
reports written by three economists, Roger G. Noll (for the plaintiffs), Kevin M. Murphy (for the
defendant), and Robert H. Topel (for the defendant). In separate reports, Professors Murphy and
Topel criticize the statistical inference in Professor Noll’s hedonic regression analysis for not
properly accounting for cluster correlation when computing standard errors for the regression
coefficients – and, in particular, the standard errors for the coefficients used to estimate damages.
In what follows I restrict my comments to issues associated with computing proper standard
errors and do not discuss model specification. In the remainder of this declaration I describe the
2
CONFIDENTIAL – ATTORNEYS EYES ONLY
clustering problem somewhat generally and determine whether and how the standard errors
should be adjusted for cluster correlation in Professor Noll’s regressions.
3. When is a Sample a Cluster Sample?
The simplest example of a cluster sample is when individual units are grouped, a priori,
into clusters, and those clusters are sampled from a population of clusters. After the clusters have
been chosen, sampling within the clusters may occur to obtain a subset of units, but the
substantive problem for computing standard errors is created when the clusters are originally
chosen. An example is when elementary schools in a state are sampled from the population of all
elementary schools in the state. After the schools have been drawn, data are collected for the
individual students – say, test scores and class sizes. If we wish to study the effects of class size
on test scores then we need to address the fact that student outcomes within a school are likely to
be influenced by common factors determined at the school level, such as (unmeasured) teacher
or principal quality. There are a couple of appropriate responses in such settings. One is to
include school “fixed effects” in the regression as a way of eliminating the cluster correlation
(and also allowing the policy variable, class size, to be systematically different across schools).
Or, one may compute standard errors that allow for correlation in the errors within a school. As I
argue in Wooldridge (2003), in some cases one may want to try both solutions at the same time.
A common misunderstanding with cluster sampling is that a clustering problem arises if
individual units that have been randomly sampled from the population are identified, ex post, as
belonging to certain groups, which are then turned into clusters. For example, suppose a random
sample of fourth-grade students is taken from the state of Michigan, and test scores and family
background variables are collected. A school identifier for each student is also included in the
data set. It would not be uncommon to sort the resulting data set so that students from the same
3
CONFIDENTIAL – ATTORNEYS EYES ONLY
school are in adjacent rows in the data set. Such a data set, which was obtained via random
sampling, is indistinguishable in appearance from a data set that was obtained by first sampling
schools and then obtaining a sample of students within each sampled school. But the
resemblance is superficial: When applying statistical methods, the latter requires the use of
methods for cluster samples while the former does not.
A simple logical exercise demonstrates that ex post grouping of a randomly drawn
sample cannot cause a clustering problem when it comes to statistical inference. After all, each
randomly drawn student is also associated with a school district. If we include a school district
identifier should we then treat the district as the cluster? In Michigan, groups of school districts
form what are known as intermediate school districts (ISDs). Is the ISD now the appropriate
level of clustering? We can keep going. Each ISD is either in the lower or upper peninsula of
Michigan. Does this mean we only have two independent clusters? Taking the argument to its
extreme, each student lives in the state of Michigan. Is it appropriate to think we have only a
single cluster, in which case no statistical inference is possible? The final perspective is clearly
preposterous, but by getting to it we see that viewing the data as a cluster sample at any level of
aggregation is incorrect. If we started with a random sample then we can and should analyze the
data using basic tools of statistics and econometrics based on random sampling. How we might
choose to group the students ex post is irrelevant provided the variables we use in our analysis
are collected as part of the random sampling scheme.
I will show in the next section that the clustering exercise undertaken by Professors
Murphy and Topel fits into the framework of ex post clustering, and is therefore an inappropriate
and unreliable application of clustering.
4
CONFIDENTIAL – ATTORNEYS EYES ONLY
It is important to know that computing so-called “cluster-robust” standard errors is not
harmless when the data have come from a random sample. (The same comment holds when the
sample is the population, a point that follows from the discussion in Section 5.) Clustering can
produce standard errors that are vastly inflated compared with the true precision of the estimates.
In other words, estimates that are properly deemed to be statistically significant when the correct
standard errors are used can be statistically insignificant when “cluster-robust” standard errors
are used. Empirical researchers sometimes fall into the trap of thinking conservative inference is
somehow preferred, but that is incorrect when appropriate methods of inference are available that
are not conservative. If we are interested in estimating the effect of an intervention we want our
confidence intervals to be as reliable as possible.
A separate problem with clustering is that, even when it is legitimate – so that the data
have actually been collected by sampling clusters – we need enough clusters that are relatively
small in order to justify the large-sample approximations. Hansen (2007) shows that when
applied to true cluster samples one should have a reasonably large number of clusters without
very large cluster sizes. See also the discussion in Donald and Lang (2007) and Wooldridge
(2003). Hansen’s simulations show that clustered standard errors perform reasonably well with
50 clusters and 20 observations per cluster. But clustered standard errors are not justified with,
say, 10 clusters and 200 observations per cluster. The higher is the ratio of observations per
cluster to number of clusters, the more poorly clustered standard errors behave. Generally, it is
difficult to know how well clustering approximates the true sampling error when the number of
clusters is small and cluster sizes are large. The resulting “cluster-robust” standard errors can be
very misleading as estimates of statistical precision. As I discuss in Section 4, Murphy and Topel
5
CONFIDENTIAL – ATTORNEYS EYES ONLY
use clustering in a setting where the cluster sizes are much larger than the number of clusters.
This is a second reason why their analysis is invalid.
There is a final, important point that needs to be emphasized before turning specifically
to Professor Noll’s analysis and the critiques by Professors Murphy and Topel. Namely, it is
inappropriate to use the amount of variation in either the explanatory variables or the response
variable across subgroups of the population as a guide for whether a data set constitutes a cluster
sample. Clustering is a property of how the data are collected and has nothing to do with how
much variation there is in the underlying population variable or variables. For example, suppose
we are interested in testing whether a chain of fast-food restaurants practices price discrimination
across two different kinds of neighborhoods – say “poor” and “not poor.” We randomly sample
transactions at fast food restaurants and obtain a sample of prices of purchased items. When we
sample a transaction, we observe the item bought (say, a medium soda), its price, and the kind of
neighborhood (poor or not poor). Particularly if the transactions data are collected over a narrow
time interval the price variation for a particular item in a particular geographic region may be
small. In fact, many of the transactions would have identical prices for identical menu items – for
example, small, medium, and large sodas. This does not mean that those three groups of sodas
should be treated as three clusters. In fact, this would be a bad idea for the two reasons discussed
above: (i) The ex post clustering of the data creates the illusion of a cluster sampling problem
where none exists, seriously understating the amount of independent information we have on
prices. (ii) The standard errors obtained using three clusters, probably with several hundred or
thousand observations within each cluster, have no statistical justification and are for all practical
purposes worthless. On the other hand, computing the average prices for each item and obtaining
6
CONFIDENTIAL – ATTORNEYS EYES ONLY
standard errors under random sampling is appropriate, and standard comparison-of-means tests
across poor and non-poor neighborhoods is entirely appropriate.
In this example, it would also be inappropriate to treat “poor” and “not poor” as two
clusters, in which case no statistical inference concerning the difference in means would be
possible because we could not even compute a standard error. [See, for example, Wooldridge
(2003).] It is also inappropriate to use, say, census blocks as clusters. If the transactions have
been randomly sampled from the population of all transactions then the data should be analyzed
using standard statistical methods for random sampling.
4. Clustering in Professor Noll’s Hedonic Regressions
I now turn to an evaluation of the claims made by Professors Noll, Murphy, and Topel
concerning the proper computation of standard errors using the transactions data for Apple iPods.
In determining whether clustering is needed, it is important to understand a couple of features of
the data. First, as explained by Professor Noll (rebuttal, 11/25/13), the data on transactions –
apart from dropping some records with missing data or outliers in prices – effectively represents
the entire population of transactions. Under a scenario that I describe in Section 5 – which views
the prices as counterfactual outcomes under different “treatment” regimes (before Harmony was
blocked and after it was blocked) – the standard errors computed as if the sample is random are
at least conservative. Therefore, if one is satisfied with conservative inference, one need not
worry about corrections that are sometimes used when the sample and population coincide.
7
CONFIDENTIAL – ATTORNEYS EYES ONLY
Based on my understanding of the
econometrics used by Professor Noll in his rebuttal report, Professor Murphy’s assertion is no
longer relevant. In the analysis that forms the basis for his rebuttal report, Professor Noll did not
expand each transaction into an observation for each iPod unit sold. Professor Noll does use
quantity weights that account for the number of units in each transaction and, strictly speaking,
these are the optimal weights for weighted least squares only when the unit-specific equations
have independent errors. However, as I have emphasized in my introductory textbook
(Wooldridge, 2013, pages 290-292), the weighted least squares standard errors can be made
robust to allow for the use of non-optimal weights. The same point is discussed, in the context of
sampling, in Solon, Haider, and Wooldridge (forthcoming). Professor Noll properly uses the
robust standard errors. But the more important point is that Professor Noll properly uses each
transaction as the unit of observation. There is no clustering problem introduced in Professor
Noll’s rebuttal report by expanding a transaction into several hundred or thousand observations
on individual iPod units.
If one properly views a transaction as a unit of observation, as is done by Professor Noll,
the data need not be treated as a cluster sample. In particular, the clustering procedure suggested
by Professors Murphy and Topel, which is to cluster the data at the product family/quarter level,
is inappropriate. With each shipment there is a new draw of characteristics of the units shipped,
quantity, and average price, and it is appropriate to treat these as independent pieces of
8
CONFIDENTIAL – ATTORNEYS EYES ONLY
information. By contrast, Professors Murphy and Topel make the mistake described in Section 3.
Namely, they use an ex post clustering based on characteristics of the shipments – product family
and time period. In his report, Murphy writes, “There are standard techniques that allow for
clustering in the calculation of regression standard errors. If it were the case that clustering is
unimportant, as assumed by Professor Noll, he could have employed these techniques to
demonstrate that accounting for clusters does not affect his results” (page 51). Professor Murphy
makes the mistake of assuming that clustering in the context of Professor Noll’s analysis at the
product family/quarter level is statistically valid. For the two reasons I explained in Section 3,
clustering is not valid, and likely to produce inference that is much too conservative: (i) The
clustering of the data by product family/quarter is done ex post, artificially creating cluster
correlation and unnecessarily producing conservative sampling variances. (ii) The number of
observations per cluster swamps the number of clusters, rendering the cluster standard errors
useless, even if we accepted the idea that clustering is needed in the first place (which I do not).
It may help to cover a simple case to further see the key difference between what is the
correct standard error calculation used by Professor Noll and the incorrect calculation by
Professors Murphy and Topel. Suppose there is only one iPod model with a fixed set of
characteristics. We obtain transaction records for this model both before and after Harmony was
blocked. For each transaction we obtain a price per iPod unit sold, and then compute the average
price within the two groups. A valid standard error and t statistic is obtained by treating the
transactions as independent draws – the approach used by Professor Noll. In this case, if we treat
the product class as a cluster, as suggested by Murphy and Topel, we cannot even compute a
standard error: the observations within the control and treatment groups are allowed to be
arbitrarily correlated. Wooldridge (2003) discusses essentially this same example, and points out
9
CONFIDENTIAL – ATTORNEYS EYES ONLY
that, if one follows the Murphy and Topel view, the staple of introductory statistics and basic
policy analysis with random interventions, a simple comparison of means, cannot be carried out.
With more product classes and multiple time periods then, technically, one can compute
clustered standard errors at the product family/quarter level – as is done in the Murphy Report
and the Topel Report. But it should not be done because it produces very unreliable standard
errors.
5. Inference when the Sample is the Population
Professor Noll comments in his rebuttal (page 10) that he is using the entire population of
transactions, and therefore there can be no cluster sampling problem because there is no
sampling. I agree with this assessment. As discussed by Professor Noll, his analysis of the
transactions data is essentially an event study, or an intervention analysis. Thus, it is important to
ask: At what level is the intervention? In my view, it is at the transaction level, and that is the
approach taken by Professor Noll.
If we observe the entire population of shipments, how can one meaningfully introduce
sampling uncertainty into the estimates? After all, if we can compute a price for every
transaction in the population, we can compute the differences in average prices among various
product family/time period groups without error.
One possibility, which fits well with intervention analysis, is to view the prices we
observe as only one possible set of prices from a set of counterfactual prices. Consider a simple
example, where there is only a single product (a particular iPod with a given set of features) and
a single intervention (Harmony is blocked). Then we can view each shipment in the population
as having two potential prices,
0 and
1 . The first of these is the price in the regime
where Harmony is not blocked, and the second is in the regime where Harmony is blocked. The
10
CONFIDENTIAL – ATTORNEYS EYES ONLY
goal is to estimate the so-called average treatment effect – see, for example, Imbens and
Wooldridge (2009) – here defined as
∑
where
1
0 ,
(1)
is the size of the population. The counterfactual nature of the thought experiment is
crucial, as we observe only one of
0 and
1 . Which price we observe depends on the
intervention status for transaction , which we denote using a binary (dummy) variable
. This
variable is equal to zero for transactions observed when Harmony is not blocked and equal to one
when Harmony is blocked. The observed price is then a random variable, and can be written as
1
is random because
Assume that
0
1
(2)
is random.
is assigned independent of the counterfactual prices – this is called the
“unconfoundedness” assumption in the treatment effects literature (see Imbens and Wooldridge,
2009). The standard estimator of is the difference in the sample averages between the “treated”
and “control” groups, which we denote
̂
.
(3)
This is an unbiased estimator of when the unconfoundedness assumption holds. For the
purposes of statistical inference, it can be shown that the variance of the estimator, conditional
on the number of treated units,
, is
,
̂ |
where
1
is the number of control units,
population variance of the
1 ,
,
(4)
∑
is the population variance of the
population variance of the treatment effects,
1
1
0 , and
,
is the
is the
0 . The first two terms in equation (4)
comprise the usual expression, which can be found in introductory texts on mathematical
11
CONFIDENTIAL – ATTORNEYS EYES ONLY
statistics, for the variance of the difference in means when obtaining random samples from a
large population. The last term is a correction that results from having access to the entire
population. When the treatment effect is constant, so that
1
0
for all , the last
term is zero, and the usual variance formula obtained under random sampling holds when we
observe the entire population. In general, the usual formula is actually conservative. An
important implication is that the reported standard errors actually overstate the sampling
variation in ̂ , resulting in conservative inference: If one finds statistical significance using the
usual standard errors then the estimates would be significant if one accounted for the final term
in (4).
The analysis for multiple regression is more complicated but the findings are similar:
using the usual heteroskedasticity-robust variance matrix estimator is at least conservative.
Professor Noll uses the heteroskedasticity-robust variance matrix estimator and so his inference
is appropriate.
The previous discussion also has relevance for the cluster sampling issue. Because,
hypothetically, each transaction could have been in multiple states of the world – different time
periods, different models shipped, before and after Harmony was blocked – it makes sense to
view the intervention for each transaction as an independent assignment. Therefore, there can be
no cluster sampling issue in the setting analyzed by Professor Noll. On the other hand, if we
sampled an entire population of clusters – such as all elementary schools in a particular state – in
order to evaluate a school-level intervention, then we should treat each school as its own cluster
of individual students. The conclusion would be that the usual cluster-sample inference obtained
for sampling clusters from a large population results in conservative inference. But this is a
different setting than the one studied by Professor Noll.
12
CONFIDENTIAL – ATTORNEYS EYES ONLY
6. Summary
(i) Clustering is inappropriate where, as here, the regressions use the entire population of
transactions. There can be no cluster sampling problem because there is no sampling.
(ii) Professor Noll’s use of heteroskedasticity-robust standard errors is appropriate. By
contrast, Professors Murphy and Topel, who suggest clustering at the family product/quarter
level, are guilty of ex post clustering. Their standard error calculations are much too imprecise.
(iii) Computing “cluster-robust” standard errors where, as here, clustering is not
appropriate, is not harmless. Clustering can produce standard errors that are vastly inflated
compared with the true precision of the estimates.
I declare that the foregoing is true to the best of my knowledge and belief.
_____________________________
Jeffrey M. Wooldridge
Executed at Mason, Michigan, December 20, 2013
13
CONFIDENTIAL – ATTORNEYS EYES ONLY
References
Donald, S.G. and K. Lang (2007), “Inference with Difference-in-Differences and Other Panel
Data,” Review of Economics and Statistics 89, 221-233.
Hansen, C.B. (2007), "Asymptotic Properties of a Robust Variance Matrix Estimator for Panel
Data when T is Large," Journal of Econometrics 141, 597-620.
Imbens, G.W. and J.M. Wooldridge (2009), “Recent Developments in the Econometrics of
Program Evaluation.” Journal of Economic Literature 47, 5–86.
Solon, G., S.J. Haider, and J.M. Wooldridge (forthcoming), “What Are We Weighting For?”
Journal of Human Resources.
Wooldridge, J.M. (2003), “Cluster-Sample Methods in Applied Econometrics,” American
Economic Review 93, 133-138.
Wooldridge, J.M. (2010), Econometric Analysis of Cross Section and Panel Data, second
edition. Cambridge, MA: MIT Press.
14
CONFIDENTIAL – ATTORNEYS EYES ONLY
Appendix A
CURRICULUM VITAE
JEFFREY M. WOOLDRIDGE
Office Address: Department of Economics
Michigan State University
110 Marshall-Adams Hall
East Lansing, MI 48824-1038
Phone: 517-353-5972
Fax: 517-432-1068
E-mail: wooldri1@msu.edu
1. ACADEMIC POSITIONS HELD
University Distinguished Professor of Economics, Michigan State University: July 1, 2001 to
present.
Co-Director, Economics of Education Specialization, Michigan State University: July 2009 to
present.
Professor of Economics, Michigan State University: July 1, 1993 to June 30, 2001.
Associate Professor of Economics, Michigan State University: September 1, 1991 to June 30,
1993.
Assistant Professor of Economics, Massachusetts Institute of Technology: June 1, 1986 to June
30, 1991.
2. ACADEMIC BACKGROUND
Ph.D., Economics, University of California, San Diego, 1986
B.A., Computer Science, B.A., Economics, University of California, Berkeley, 1982 (With High
Distinction in General Scholarship)
3. TEACHING
Graduate: Linear Models, Nonlinear Econometrics, Cross Section and Panel Data Econometrics,
Applied Econometrics, Mathematical Statistics, Mathematics for Economics
Undergraduate: Econometrics, Applied Econometrics, Statistics
15
CONFIDENTIAL – ATTORNEYS EYES ONLY
4. SOCIETY MEMBERSHIPS
American Economic Association
American Statistical Association
Econometric Society
Midwest Economics Association
5. EDITORIAL BOARDS
Co-Editor, Journal of Econometric Methods, 2010 to present.
Editorial Board, Journal of Economic Literature, 2004 to 2010.
Co-Editor, Econometric Theory, 2003 to 2005.
Advisory Editor, the Palgrave Handbook of Econometrics, 2004-2005.
Editor, Journal of Business and Economic Statistics, 1998 to 2000.
Co-Editor, Economics Letters, 1995 to 1998.
Associate Editor, Stata Journal, 2002 to present.
Associate Editor, Journal of Business and Economic Statistics, 1995 to 1997.
Associate Editor, Journal of Econometrics, 1995 to 1997; 2001 to 2005.
Associate Editor, Review of Economics and Statistics, 1993 to 1997.
6. AWARDS AND HONORS
President, Midwest Economics Association, 2010-2011.
Fellow, Institute for the Study of Labor (IZA), Bonn, Germany, elected 2009.
President-elect, Midwest Economics Association, 2009-2010.
First Vice President, Midwest Economics Association, 2005-2006.
Fellow of the Econometric Society, elected 2002.
Plura Scripsit Award, Econometric Theory, elected 2001.
Benjamin Meaker Visiting Professorship, University of Bristol, England, July 2000.
16
CONFIDENTIAL – ATTORNEYS EYES ONLY
Richard Stone Prize for “Econometric Methods for Fractional Response Variables with an
Application to 401(k) Plan Participation Rates” (with Leslie E. Papke), Journal of Applied
Econometrics, 1998. (Best paper in two volumes.)
Multa Scripsit Award, Econometric Theory, elected 1997.
Journal of Econometrics Fellow, elected November 1995.
Alfred P. Sloan Research Fellow, 1991-1994.
Teacher of the Year Award, Graduate Economics Association, MIT, 1991-1992.
Teacher of the Year Award, Graduate Economics Association, MIT, 1990-1991.
Teacher of the Year Award, Graduate Economics Association, MIT, 1988-1989.
Alfred P. Sloan Doctoral Dissertation Fellowship, 1985-1986.
Sea Grant Predoctoral Traineeship, 1983.
Regents Fellowship, University of California, 1982-1984.
7. PROFESSIONAL
Econometric Society Fellows Nominating Committee, 2006.
Scientific Committee, International Panel Data Conference, 2006, 2009, 2011, 2012.
Advisory Panel, National Science Foundation, Mathematical Social and Behavioral Sciences
Competition, Arlington, VA, May 2005.
Occasional Consultant, Industrial Economics, Inc.
Occasional Consultant, Stratus Consulting
Occasional Consultant, Deloitte Consulting
Occasional Consultant, Washington State Institute for Public Policy
Occasional Consultant, Arthur Andersen, Chicago, Illinois, 1995 to 2001
Occasional Consultant, Charles River Associates, Boston, Massachusetts, 1987 to 1996
17
CONFIDENTIAL – ATTORNEYS EYES ONLY
8. PRESENTATIONS AT MEETINGS
Invited Speaker, Mainz Econometrics Workshop, “Control Function Methods in Applied
Econometrics,” Mainz, Germany, August 2013.
Invited Speaker, Sydney Econometric Theory Workshop: “Quasi-Maximum Likelihood
Estimation and Testing for Nonlinear Models with Endogenous Explanatory Variables,” Sydney,
Australia, July 2012.
A.W. Phillips Lecture, Econometric Society Australasian Meeting, “Nonlinear Panel Data
Models with Heterogeneity and Endogeneity,” Melbourne, Australia, July 2012.
Invited Speaker, Stata Conference: “Fractional Response Models with Endogenous Explanatory
Variables and Heterogeneity,” Chicago, IL, July 2011.
Invited Speaker, Causality, Prediction, and Specification Analysis: Recent Advances and Future
Directions – A Conference in Honore of Halbert L. White, Jr.: “Quasi-Maximum Likelihood
Estimation and Testing for Nonlinear Models with Endogenous Explanatory Variables,” La
Jolla, CA, May 2011.
Presidential Address, Midwest Economics Association: “Thoughts on Heterogeneity in
Econometric Models,” St. Louis, MO, March 2011.
Keynote Speaker, Health Econometrics Workshop: “Correlated Random Effects Models with
Unbalanced Panels,” Ann Arbor, MI, October 2010.
Invited Speaker, 15th Conference on Panel Data: “Nonlinear Correlated Random Effects Models
with Unbalanced Panels,” Bonn, Germany, July 2009.
Invited Speaker, Canadian Econometrics Study Group: “Nonlinear Dynamic Panel Data Models
with Unobserved Effects,” Montréal, Quebec, September 2008.
“Nonparametric and Semiparametric Estimation of Partial Effects for Nonlinear Panel Data
Models,” Workshop on Nonparametric/Semiparametric Econometric Methods, Carleton
University, Ottawa, Ontario, September 2008.
Invited Speaker, Summer North American Stata Users Group: “Inference for Partial Effects in
Nonlinear Panel Data Models using Stata,” Chicago, IL, July 2008.
Featured Speaker, National Value-Added Modeling Conference: “Some Econometric
Considerations for Value-Added Modeling,” Madison WI, April 2008.
“Estimating Average Treatment Effects with Continuous and Discrete Covariates: The Case of
Swan-Ganz Catheterization,” Winter Meetings of the American Economic Association, New
Orleans, January 2008.
18
CONFIDENTIAL – ATTORNEYS EYES ONLY
Panel Member, “How to Mentor Junior Economists,” Winter Meetings of the American
Economic Association, New Orleans, January 2008.
“Panel Data Methods for Fractional Response Variables with an Application to Test Pass Rates”
(with L.E. Papke), Conference on The Use of Econometrics in Informing Public Policy Makers,
Rice University, April 2006.
Panel Member, Midwest Economics Association, Chicago, March 2006: “Teaching
Econometrics.”
Discussant, Midwest Economics Association, Chicago, March 2006: “Applied Time Series.”
Discussant, Winter Meetings of the Econometric Society, Philadelphia, January 2005: “Causal
Inference and Matching.”
“Inverse Probability Weighted M-Estimators for General Missing Data Problems,” Texas Camp
Econometrics, Fort Worth, TX, February 2004.
“On the Robustness of Fixed Effects and Related Estimators in Correlated Random Coefficient
Panel Data Models,” Midwest Econometrics Group, Columbia, MO, October 2003.
“Cluster-Sample Methods in Applied Econometrics,” Winter Meetings of the American
Economic Association,” Washington, DC, January 2003.
Discussant, Winter Meetings of the Econometric Society, Washington, DC, January 2003:
“Difference-in-Differences and Panel Data Estimation.”
“Inverse Probability Weighted M-Estimators for Sample Selection, Attrition, and Stratification,”
CeMMAP Microeconometrics Workshop, London, February 2002.
“Inverse Probability Weighted M-Estimators for Sample Selection, Attrition, and Stratification,”
Midwest Econometrics Group, Kansas City, October 2001.
“Unobserved Heterogeneity and Estimation of Average Partial Effects,” National Science
Foundation Symposium on Identification and Inference for Econometric Models, Berkeley,
August 2001.
“Applications of Generalized Method of Moments Estimation,” Winter Meetings of the
American Economic Association,” New Orleans, January 2001.
Discussant, Winter Meetings of the Econometric Society, New Orleans, January 2001: “Recent
Advances in Nonlinear Time Series.”
Invited Speaker, ESRC Econometric Study Group Conference, Bristol, England, July 2000:
“The Initial Conditions Problem in Dynamic, Nonlinear Panel Data Models with Unobserved
Heterogeneity.”
19
CONFIDENTIAL – ATTORNEYS EYES ONLY
Discussant, Joint Statistical Meetings, Indianapolis, August 2000: Journal of Business and
Economic Statistics Invited Session.
Plenary Speaker, Research Triangle Econometrics Conference, December 1998: “Estimating
Average Partial Effects Under Conditional Moment Independence Assumptions.”
Discussant, Winter Meetings of the Econometric Society, Chicago, January 1998: “Panel Time
Series” and “New Developments in Technical Econometrics.”
“Selection Corrections with a Censored Selection Variable,” Canadian Econometrics Study
Group, Windsor, September 1994.
Discussant, Joint Statistical Meetings, Toronto, August 1994: “Analysis of Data on Durations
and Counts.”
Discussant, Summer Meetings of the Econometric Society, Quebec City, June 1994: “Panel
Data Methods” and “Testing Using Nonparametric Methods.”
Discussant, Winter Meetings of the Econometric Society, Anaheim, January 1993: “Simulation
and Semiparametric Methods” and “Estimation and Testing in Systems of Equations.”
“Multiplicative Panel Data Models without the Strict Exogeneity Assumption,” Midwest
Econometrics Group, South Bend, IN, September 1991.
Discussant, Winter Meetings of the Econometric Society, Washington, DC, December 1990:
“Semiparametric Methods,” “Unit Roots and Cointegration: Methodology,” and “Specification
Testing.”
“Some Results on Specification Testing Against Nonparametric Alternatives,” Summer Meetings
of the Econometric Society, Ann Arbor, MI, July 1989.
Discussant, Summer Meetings of the Econometric Society, Ann Arbor, MI, July 1989:
“Specification Tests.”
“An Encompassing Approach to Conditional Mean Tests with Applications to Testing
Nonnested Hypotheses,” Winter Meetings of the Econometric Society, New York, December
1988.
Discussant, Summer Meetings of the Econometric Society, Minneapolis, July 1988: “Time
Series I.”
“A Capital Asset Pricing Model with Time-Varying Covariances,” World Congress of the
Econometric Society, Cambridge, MA, August 1985.
20
CONFIDENTIAL – ATTORNEYS EYES ONLY
9. SEMINARS
“A Control Function Approach to Estimating Switching Regression Models with Endogenous
Explanatory Variables and Endogenous Switching,” MSU, 2013.
“Evaluating Specification Tests in the Context of Value-Added Estimation,” Vanderbilt, 2012.
“Quasi-Maximum Likelihood Estimation and Testing for Nonlinear Models with Endogenous
Explanatory Variables,” EIEF, 2011. U.C. Riverside, University College London, 2012.
“Correlated Random Effects Models with Unbalanced Panels,” MSU, 2009; Bank of Italy, 2011.
“Minimum Distance Estimation Using Pseudo Panel Data,” MSU, Michigan, 2008.
“Panel Data Methods for Fractional Response Variables with an Application to Test Pass Rates”:
Harvard/MIT, British Columbia, 2007-2008.
“Inverse Probability Weighted Estimation for General Missing Data Problems”: Berkeley,
Harvard/MIT, Michigan, Montreal, Notre Dame, Ohio State, Penn State, Texas, 2002-2006.
“On the Robustness of Fixed Effects and Related Estimators in Correlated Random Coefficient
Panel Data Models”: Texas A&M, 2004.
“Unobserved Heterogeneity and Estimation of Average Partial Effects”: UCLA, UCSD, 2003.
“Simple Solutions to the Initial Conditions Problem for Dynamic, Nonlinear Panel Data Models
with Unobserved Effects”: Michigan, 2002.
“Instrumental Variables Estimation of the Average Treatment Effect in the Correlated Random
Coefficient Model”: NYU, 2000.
“Inverse Probability Weighted M-Estimators for Sample Selection, Attrition, and Stratification”:
University College London, 2000.
“The Initial Conditions Problem for Dynamic, Nonlinear Panel Data Models with Unobserved
Heterogeneity”: Penn State, 2000.
“Estimating Average Partial Effects Under Conditional Moment Independence Assumptions”:
Princeton, Northwestern, Harvard/MIT, Chicago, Bristol, 1997-2000.
“Asymptotic Properties of Weighted M-Estimators for Variable Probability Samples”: Michigan,
Research Triangle Institute, 1996-1997.
“Selection Corrections with a Censored Selection Variable”: Georgia State, Texas, Texas A&M,
Arizona State, Montreal, Florida, Virginia, 1994-1996.
21
CONFIDENTIAL – ATTORNEYS EYES ONLY
“Selection Corrections for Panel Data Models Under Conditional Mean Independence
Assumptions”: Research Triangle Institute, UCSD, Ohio State, Harvard/MIT, 1992-1993.
“Multiplicative Panel Data Models without the Strict Exogeneity Assumption”: Northwestern,
Purdue, Texas A&M, Rice, Windsor, Wisconsin, 1991-1992.
“Distribution-Free Estimation of Some Nonlinear Panel Data Models”: Yale, Harvard, Michigan,
Michigan State, Northeastern, 1990-1991.
“Some Alternatives to the Box-Cox Regression Model”: Boston College, Montreal, 1989.
“A Test for Functional Form Against Nonparametric Alternatives”: Harvard, Berkeley, UCSD,
UCSB, Pennsylvania, Rochester, 1989.
“On the Application of Robust, Regression-Based Diagnostics to Models of Conditional Means
and Conditional Variances”: Princeton, Johns Hopkins, Indiana, Illinois, Board of Governors of
the Federal Reserve, Brown, 1988-1989.
“A Unified Approach to Robust, Regression-Based Specification Tests”: Brandeis, CarnegieMellon, Queen’s, Toronto, Johns Hopkins, Chicago Graduate School of Business, Northwestern,
Cornell, UCSD, Penn State, Columbia, 1987-1988.
“Specification Testing and Quasi-Maximum Likelihood Estimation”: Harvard, Yale, Johns
Hopkins, 1986-1987.
10. LECTURES AND SHORT COURSES
University of Mainz Summer School, “New Developments in Panel Data Econometrics,” Mainz,
Germany, August 2013.
University of Crete Advanced Summer School in Economics and Econometrics, “Panel Data
Econometrics and Treatment Effect Estimation,” Crete, Greece, July/August 2013.
IZA European Summer School in Labor Economics, “Correlated Random Effects Panel Data
Models,” Buch/Ammersee, Germany May 2013.
American Accounting Association Workshop, “Linear and Nonlinear Panel Data Models,”
Washington, DC, August 2012.
National Centre for Econometric Research, Queensland University of Technology, “Panel Data
Econometrics,” Brisbane, Australia, July 2012.
Programme Evaluation for Policy Analysis, Institute for Fiscal Studies, “Microeconometric
Methods in Policy Evaluation,” London, June 2012.
22
CONFIDENTIAL – ATTORNEYS EYES ONLY
Distinguished Visitor, “Control Function Methods in Econometrics,” University of California,
Riverside, May 2012.
American Economic Association, “Cross-Section Econometrics” (with Guido Imbens), Chicago,
January 2012.
American Accounting Association (FARS Division), “Treatment Effect Estimation with
Unconfounded Assignment,” Chicago, January 2012.
LABOUR Lectures, EIEF, “Topics in Microeconometrics,” Rome, October 2011.
CIDE Summer School in Econometrics, “Topics in Panel Data Econometrics,” Bertinoro, Italy,
June 2011.
Czech National Bank, “Topics in Panel Data Econometrics,” Prague, May 2011.
Bonn Graduate School of Economics/IZA, “Treatment Effect Estimation and Selection Models,”
Bonn, Germany, July 2009.
Cemmap/University College London, “New Developments in Econometrics” (with Guido
Imbens), London, June 2009.
Canadian Labour Market and Skills Researcher Network/Canadian Economics Association, “A
Workshop in Applied Econometrics” (with Guido Imbens), Toronto, ON, May 2009.
International Monetary Fund, “Limited Dependent Variables,” Washington, DC, April 2009.
American Economic Association, “Cross-Section Econometrics” (with Guido Imbens), San
Francisco, CA, January 2009.
Federal Reserve Board, “A Course in Applied Econometrics,” Washington, DC, SeptemberOctober 2008.
Michigan State University Center for Statistical Training and Consulting, “Control Function and
Related Methods,” East Lansing, MI, October 2008.
Invited Speaker, Canadian Econometrics Study Group, “Nonlinear Dynamic Panel Data Models
with Unobserved Effects,” Montreal, QU, September 2008.
Carleton University Workshop on Robust Nonparametric Methods, “Semiparametric and
Nonparametric Estimaton of Partial Effects in Nonlinear Panel Data Models,” Ottawa, ON,
September 2008.
Institute for Research on Poverty, University of Wisconsin, Madison, “A Course in Applied
Microeconometrics” (with Guido Imbens), August 2008.
23
CONFIDENTIAL – ATTORNEYS EYES ONLY
Stata North American Users’ Group, “Inference for Partial Effects in Nonlinear Panel Data
Models using Stata,” Chicago, IL, July 2008.
National Conference on Value-Added Modeling, “Some Econometric Considerations for ValueAdded Modeling,” Madison, WI, April 2008.
Michigan State University Center for Statistical Training and Consulting, “Regression and
Propensity Score Methods for Treatment Effect Estimation and Policy Evaluation,” East
Lansing, MI, April 2008.
Michigan State University Department of Accounting, “Research Workshop on Selection
Models,” East Lansing, MI, April 2008.
Bureau of Economic Analysis/Federal Trade Commission, “A Course in Applied
Microeconometrics” (with Guido Imbens), Washington, DC, January/February 2008.
National Bureau of Economic Research Summer Institute (with Guido Imbens), Cambridge MA:
“What’s New in Econometrics?” July/August 2007.
Invited Lecturer, University of Helsinki/MTT, “Econometric Methods with Censored Data,”
Helsinki, Finland, June 2007.
Michigan State University Center for Statistical Training and Consulting, “Propensity Score
Methods for Treatment Effect Estimation and Policy Evaluation,” East Lansing, MI, April 2007.
Michigan State University Center for Statistical Training and Consulting, “Binary Response Models for
Longitudinal Data,” East Lansing, MI, August 2006.
Invited Lecture, NIPE Summer School, University of Minho, “Topics in Program Evaluation,”
Minho, Portugal, June 2006.
Invited Lecturer, CIDE Summer School in Econometrics, “Topics in Panel Data Econometrics,”
Bertinoro, Italy, June 2005.
Mathematical Economics Forum, Wake Forest University: “Econometric Issues in Estimating
Performance Effects of Spending in K-12 Schools,” October 2004.
Invited Lecturer, Banco de Portugal, Lisbon: “Topics in Microeconometrics,” June 2004.
Seminars on Analytical Methods, Statistics Canada, Ottawa, Ontario: “Cluster-Sample Methods
in Applied Econometrics,” April 2004.
XIVth Summer School of the European Economic Association, London: “Discrete Response
Models,” September 2003.
24
CONFIDENTIAL – ATTORNEYS EYES ONLY
Methods Workshop, Lister Hill Center for Health Policy, University of Alabama, Birmingham:
“Cluster-Sample Methods in Applied Econometrics,” March 2003.
Rodgers Clark Invited Lecturer, North Carolina State University: “Simple Solutions to the Initial
Conditions Problem in Nonlinear, Dynamic Panel Data Models with Unobserved Effects,” April
2002.
Invited Lecturer, Western Michigan University: “Estimation and Inference for Dependent
Processes” and “Selection Corrections for Cross Section and Panel Data,” February-March 1994.
Invited Lecturer, University of Pennsylvania: “Topics in Specification Testing,” January 1991.
Invited Lecturer, Universidad Complutense de Madrid: “The Econometric Treatment of
Nonstationary Time Series,” November 1990.
11. BIBLIOGRAPHY
Journal Articles
“What Are We Weighting For?” (with G. Solon and S.J. Haider), forthcoming, Journal of
Human Resources.
“Can Value-Added Measures of Teacher Performance be Trusted?” (with C.M. Guarino and
M.D. Reckase), forthcoming, Education Finance and Policy.
“Quasi-Maximum Likelihood Estimation and Testing for Nonlinear Models with Endogenous
Explanatory Variables,” forthcoming, Journal of Econometrics.
“Estimation of Dynamic Panel Data Models with Sample Selection” (with A. Semykina),
Journal of Applied Econometrics 28, 47-61, January/February 2013.
“Partial Maximum Likelihood Estimation of Spatial Probit Models” (with H. Wang and E.M.
Iglesias), Journal of Econometrics 172, 77-89, January 2013.
“A Simple Method for Estimating Unconditional Heterogeneity Distributions in Correlated
Random Effects Models,” Economics Letters 113, 12-15, October 2011.
“Estimating Panel Data Models in the Presence of Endogeneity and Selection” (with A.
Semykina), Journal of Econometrics 157, 375-380, August 2010.
“On Estimating Firm-Level Production Functions Using Proxy Variables to Control for
Unobservables,” Economics Letters 104, 112-114, September 2009.
25
CONFIDENTIAL – ATTORNEYS EYES ONLY
“Efficient Estimation of Average Treatment Effects with Mixed Categorical and Continuous
Data” (with Q. Li and J.S. Racine), Journal of Business and Economic Statistics 27, 206-223,
April 2009.
“Recent Developments in the Econometrics of Program Evaluation” (with G.W. Imbens),
Journal of Economic Literature 47, 5–86, March 2009.
“Difference-in-Differences Estimation,” Quantile 6, 25-46, March 2009. (In Russian.)
“Panel Data Methods for Fractional Response Variables with an Application to Test Pass Rates”
(with L.E. Papke), Journal of Econometrics 145, 121-133, July 2008.
“Estimating Average Treatment Effects with Continuous and Discrete Covariates: The Case of
Swan-Ganz Catheterization” (with Q. Li and J.S. Racine), American Economic Review 98, 357362, May 2008.
“Fixed Effects Instrumental Variables Estimation in Correlated Random Coefficient Panel Data
Models” (with I. Murtazashvili), Journal of Econometrics 142, 539-552, January 2008.
“Inverse Probability Weighted M-Estimation for General Missing Data Problems,” Journal of
Econometrics 141, 1281-1301, December 2007.
“Violating Ignorability of Treatment by Controlling for Too Many Factors,” Econometric Theory
21, 1026-1028, October 2005.
“Instrumental Variables Estimation with Panel Data,” Econometric Theory 21, 865-869, August
2005.
“Fixed Effects and Related Estimators for Correlated Random-Coefficient and Treatment Effect
Panel Data Models,” Review of Economics and Statistics 87, 385-390, May 2005.
“A Computational Trick for Delta-Method Standard Errors” (with L.E. Papke), Economics
Letters 86, 413-417, March 2005.
“Simple Solutions to the Initial Conditions Problem for Dynamic, Nonlinear Panel Data Models
with Unobserved Heterogeneity,” Journal of Applied Econometrics 20, 39-54, January 2005.
“Cluster-Sample Methods in Applied Econometrics,” American Economic Review 93, 133-138,
May 2003.
“Further Results on Instrumental Variables Estimation of Average Treatment Effects in the
Correlated Random Coefficient Model,” Economics Letters 79, 185-191, May 2003.
“ n -Consistent Estimation of a Partial Linear Model with Generated Regressors” (with Q. Li),
Econometric Theory 18, 625-645, June 2002.
26
CONFIDENTIAL – ATTORNEYS EYES ONLY
“Inverse Probability Weighted M-Estimators for Sample Selection, Attrition, and Stratification,”
Portuguese Economic Journal 1, 117-139, June 2002.
“Applications of Generalized Method of Moments Estimation,” Journal of Economic
Perspectives 15, 87-100, November 2001.
“Asymptotic Properties of Weighted M-Estimators for Standard Stratified Samples,”
Econometric Theory 17, 451-470, April 2001.
“A Framework for Estimating Dynamic, Unobserved Effects Panel Data Models with Possible
Feedback to Future Explanatory Variables,” Economics Letters 68, 245-250, September 2000.
“Asymptotic Properties of Weighted M-Estimators for Variable Probability Samples,”
Econometrica 67, 1385-1406, November 1999.
“Efficient Estimation of Panel Data Models with Strictly Exogenous Explanatory Variables”
(with K.S. Im, S.C. Ahn, and P. Schmidt), Journal of Econometrics 93, 177-201, November
1999.
“Distribution-Free Estimation of Some Nonlinear Panel Data Models,” Journal of Econometrics
90, 77-97, May 1999.
“On Two Stage Least Squares Estimation of the Average Treatment Effect in Random
Coefficient Models,” Economics Letters 56, 129-133, October 1997.
“Multiplicative Panel Data Models without the Strict Exogeneity Assumption,” Econometric
Theory 13, 667-678, October 1997.
“Econometric Methods for Fractional Response Variables with an Application to 401(k) Plan
Participation Rates” (with L.E. Papke), Journal of Applied Econometrics 11, 619-632,
November-December, 1996.
“Estimating Systems of Equations with Different Instruments for Different Equations,” Journal
of Econometrics 74, 387-405, October 1996.
“Selection Corrections for Panel Data Models Under Conditional Mean Independence
Assumptions,” Journal of Econometrics 68, 115-132, July 1995.
“A Simple Test for the Consistency of Dynamic Linear Regression in Rational Distributed Lag
Models” (with K.T. McClain), Economics Letters 48, 235-240, June 1995.
“On the Limits of GLM for Specification Testing: A Comment on Gurmu and Trivedi,”
Econometric Theory 10, 409-418, June 1994.
“A Simple Specification Test for the Predictive Ability of Transformation Models,” Review of
Economics and Statistics 76, 59-65, February 1994.
27
CONFIDENTIAL – ATTORNEYS EYES ONLY
“Contrastes de Especifiacion en Modelos Lineales con Variables Integrades” (Specification
Testing in Linear Models with Integrated Variables), Cuadernos Economicos de ICE 55, 243261, 1993.
“An Empirical Investigation of the Box-Cox Model and a Nonlinear Least Squares Alternative”
(with E. Berndt and M. Showalter), Econometric Reviews 12, 65-102, March 1993.
“A Test for Functional Form Against Nonparametric Alternatives,” Econometric Theory 8, 452475, December 1992.
“Some Alternatives to the Box-Cox Regression Model,” International Economic Review 33,
935-955, November 1992.
“Quasi-Maximum Likelihood Estimation and Inference in Dynamic Models with Time-Varying
Covariances” (with T. Bollerslev), Econometric Reviews 11, 143-172, September 1992.
“A Note on Computing R-Squared and Adjusted R-Squared for Trending and Seasonal Data,”
Economics Letters 36, 49-54, May 1991.
“Specification Testing and Quasi-Maximum Likelihood Estimation,” Journal of Econometrics
48, 29-55, April 1991.
“On the Application of Robust, Regression-Based Diagnostics to Models of Conditional Means
and Conditional Variances,” Journal of Econometrics 47, 5-46, January 1991.
“A Note on the Lagrange Multiplier and F-statistics for Two Stage Least Squares Regressions,”
Economics Letters 34, 151-155, November 1990.
“An Encompassing Approach to Conditional Mean Tests with Applications to Testing
Nonnested Hypotheses,” Journal of Econometrics 45, 331-350, September 1990.
“A Unified Approach to Robust, Regression-Based Specification Tests,” Econometric Theory 6,
17-43, March 1990.
“A Computationally Simple Heteroskedasticity and Serial Correlation Robust Standard Error for
the Linear Regression Model,” Economics Letters 31, 239-243, December 1989.
“Some Invariance Principles and Central Limit Theorems for Dependent Heterogeneous
Processes” (with H. White), Econometric Theory 4, 210-230, August 1988.
“A Capital Asset Pricing Model with Time-Varying Covariances” (with T. Bollerslev and R.F.
Engle), Journal of Political Economy 96, 116-131, February 1988.
28
CONFIDENTIAL – ATTORNEYS EYES ONLY
Books
Solutions Manual and Supplementary Materials for Econometric Analysis of Cross Section and
Panel Data, second edition. Cambridge, MA: MIT Press, 2011.
Econometric Analysis of Cross Section and Panel Data, second edition. Cambridge, MA: MIT
Press, 2010.
Introductory Econometrics: A Modern Approach, fourth edition. Cincinnati, OH: SouthWestern College Publishing, 2009.
Introductory Econometrics: A Modern Approach, International Edition, Thomson, 2006.
Introductory Econometrics: A Modern Approach (Greek Translation), Thomson, 2006.
Introductory Econometrics: A Modern Approach, third edition. Cincinnati, OH: South-Western
College Publishing, 2006.
Introductory Econometrics: A Modern Approach (Chinese Translation), Thomson, 2006.
Solutions Manual and Supplementary Materials for Econometric Analysis of Cross Section and
Panel Data. Cambridge, MA: MIT Press, 2003.
Introductory Econometrics: A Modern Approach, second edition. Cincinnati, OH: SouthWestern College Publishing, 2003.
Econometric Analysis of Cross Section and Panel Data. Cambridge, MA: MIT Press, 2002.
Introductory Econometrics: A Modern Approach (Spanish Translation), Thomson, 2001.
Introductory Econometrics: A Modern Approach. Cincinnati, OH: South-Western College
Publishing, 2000.
29
CONFIDENTIAL – ATTORNEYS EYES ONLY
Book Chapters
“Teaching Undergraduate Econometrics,” in International Handbook on Teaching and Learning
Economics, Gail Hoyt and KimMarie McGoldrick (eds.), 452-462. Cheltenham, United
Kingdom: Edward Elgar, 2011
“Econometrics: Panel Data Methods,” in Complex Systems in Finance and Econometrics.
Roberts Meyers (ed.), 215-237. Heidelberg, Germany: Springer, 2011.
“Stratified and Cluster Sampling,” in The New Palgrave Dictionary of Economics, second
edition. Stephen Durlauf and Lawrence E. Blume (eds.). London: Palgrave Macmillan, 2008.
“Probabilistic Sampling,” in the Handbook of Probability: Theory and Applications. Tamás
Rudas (ed.), 187-203. Los Angeles: Sage Publications, 2008.
“Instrumental Variables Estimation of the Average Treatment Effect in Correlated Random Coefficient
Models,” in Advances in Econometrics, Volume 21 (Modeling and Evaluating Treatment Effects in
Econometrics). Daniel Millimet, Jeffrey Smith, and Edward Vytlacil (eds.), 93‐117. Amsterdam: Elsevier,
2008.
“Unobserved Heterogeneity and Estimation of Average Partial Effects,” in Identification and
Inference for Econometric Models: Essays in Honor of Thomas Rothenberg. D.W.K. Andrews
and J.H. Stock (eds.), 27-55. Cambridge: Cambridge University Press, 2005.
“Central Limit Theorems for Dependent Heterogeneous Processes with Trending Moments”
(with H. White), in New Perspectives in Econometric Theory: The Selected Works of Halbert
White, Volume Two, 464-481. Cheltenham, UK: Edward Elgar, 2004.
“Diagnostic Testing,” in Companion to Theoretical Econometrics. B.H. Baltagi (ed.), 180-200.
Oxford: Blackwell, 2001.
“Asymptotic Properties of Some Specification Tests in Linear Models with Integrated
Processes,” in Cointegration, Causality, and Forecasting. R.F. Engle and H. White (eds.), 366384. Oxford: Oxford University Press, 1999.
“Quasi-Likelihood Methods for Count Data,” in Handbook of Applied Econometrics, Volume 2.
M.H. Pesaran and P. Schmidt (eds.), 352-406. Oxford: Blackwell, 1997.
“Score Diagnostics for Linear Models Estimated by Two Stage Least Squares,” in Advances in
Econometrics and Quantitative Economics. G.S. Maddala, P.C.B. Phillips, and T.N. Srinivasan
(eds.), 66-87. Oxford: Blackwell, 1995.
“Estimation and Inference for Dependent Processes,” in Handbook of Econometrics, Volume 4.
R.F. Engle and D.L. McFadden (eds.), 2639-2738. Amsterdam: North-Holland, 1994.
“Some Results on Sieve Estimation with Dependent Observations,” (with H. White), in
Semiparametric and Nonparametric Methods in Econometrics and Statistics. W.J. Barnett, J.
30
CONFIDENTIAL – ATTORNEYS EYES ONLY
Powell, and G. Tauchen (eds.), 459-493. Cambridge: Cambridge University Press, 1991.
Reprinted as Chapter 15 in H. White, Artificial Neural Networks: Approximation and Learning
Theory. Oxford: Blackwell, 1992.
Book Reviews
Book review of Dynamic Econometrics, by David F. Hendry, Economica 65, 296-298, May
1998.
Book review of Co-integration, Error-Correction, and the Econometric Analysis of Nonstationary Data, by A. Banerjee, J. Dolado, J.W. Galbraith, and D.F. Hendry, Journal of
Economic Literature 3, 820-821, June 1995.
Book review of Analog Estimation Methods in Econometrics by Charles F. Manski, Journal of
Economic Literature 28, 1738-1740, December 1990.
Miscellaneous
“Acknowledgement of Related Prior Work,” Econometric Theory 22, 1177-1178, December
2006.
“Statistical Significance is Okay, Too: Comment on ‘Size Matters’,” Journal of SocioEconomics 33, 577-579, November 2004.
“Fixed Effects Estimation of the Population-Averaged Slopes in a Panel Data Random
Coefficient Model,” Solution, Econometric Theory 20, 428-429, April 2004.
“Fixed Effects Estimation of the Population-Averaged Slopes in a Panel Data Random
Coefficient Model,” Problem, Econometric Theory 19, 411-412, April 2003.
“Robustness of Tests for Functional Form to Serial Correlation,” Solution, Econometric Theory
16, 795-796, October 2000.
“Robustness of Tests for Functional Form to Serial Correlation,” Problem, Econometric Theory
15, 778-780, October 1999.
“Consistency of OLS in the Presence of a Lagged Dependent Variable and Serially Correlated
Errors,” Solution, Econometric Theory 15, 260-261, April 1999.
“Consistency of OLS in the Presence of a Lagged Dependent Variable and Serially Correlated
Errors,” Problem, Econometric Theory 14, 285, April 1998.
“Asymptotic Properties of Tests for Heteroskedasticity Under Measurement Error,” Solution,
Econometric Theory 12, 402-403, June 1996.
31
CONFIDENTIAL – ATTORNEYS EYES ONLY
“The Asymptotic Power of RESET for Detecting Omitted Variables,” Solution, Econometric
Theory 12, 205, March 1996.
“Econometrics,” Encyclopedia Americana, page 614, 1996.
“Efficient Estimation Under Heteroskedasticity,” Solution, Econometric Theory 11, 797-798,
October 1995.
“Asymptotic Properties of Tests for Heteroskedasticity Under Measurement Error,” Problem,
Econometric Theory 11, 399-400, June 1995.
“The Asymptotic Power of RESET for Detecting Omitted Variables,” Problem, Econometric
Theory 10, 968-969, December 1994.
“Efficient Estimation Under Heteroskedasticity,” Problem, Econometric Theory 10, 223, March
1994.
“Comments on the Papers by McCall, Mullahy, and Sueyoshi,” American Statistical Association
1994 Proceedings of the Business and Economic Statistics Section, 17-19.
“Efficient Estimation with Orthogonal Regressors,” Problem, Econometric Theory 9, 687,
December 1993.
“Solution Set” in Asymptotic Theory for Econometricians by Halbert White, 191-223. Academic
Press: Orlando, 1984.
Papers under Review and Recent Unpublished Working Papers
“A Control Function Approach to Estimating Switching Regression Models with Endogenous
Explanatory Variables and Endogenous Switching” (with I. Murtazashvili), mimeo.
“Control Function Methods in Applied Econometrics.” Revised and resubmitted to Journal of
Human Resources.
“Correlated Random Effects Models with Unbalanced Panels,” mimeo, Michigan State
University Department of Economics, 2009. Under revision for Journal of Econometrics.
“An Evaluation of Empirical Bayes’ Estimation of Value-Added Teacher Performance
Measures” (with C.M. Guarino, M. Maxfield, M.D. Reckase, and P. Thompson), mimeo, MSU
Department of Economics, 2012. Submitted to Journal of Educational and Behavioral Statistics.
“Evaluating Specification Tests in the Context of Value-Added Estimation” (with C.M. Guarino,
M.D. Reckase, F. Smart, and B. Stacy), mimeo, Michigan State University Department of
Economics, 2012.
“Should Instrumental Variables be Used as Matching Variables?” mimeo, Michigan State
University Department of Economics, 2009.
32
CONFIDENTIAL – ATTORNEYS EYES ONLY
33
CONFIDENTIAL – ATTORNEYS EYES ONLY
Appendix B
Documents Considered by Professor Jeffrey M. Wooldridge
Declaration of Roger G. Noll on Liability and Damages, dated April 3, 2013 (including Exhibits
1-16 and Appendices A-C).
Corrections to Declaration of Roger G. Noll on Liability and Damages, dated May 31, 2013
(including Exhibits 14, 15.1, 15.2, 16.1, 16.2).
Rebuttal Declaration of Roger G. Noll on Liability and Damages, November 25, 2013 (including
Exhibits 1-6 and Appendices A-B).
Amended Expert Report of Robert H. Topel, dated August 19, 2013 2013 (including Exhibits 116 and Appendices A-D).
Corrections to Expert Report of Robert H. Topel Submitted July 19, 2013.
Amended Expert Report of Kevin M. Murphy, dated August 19, 2013 (including Exhibits 1-17
and Appendices A-D).
Corrections to Expert report of Kevin M Murphy Submitted July 19, 2013.
34
CONFIDENTIAL – ATTORNEYS EYES ONLY
Disclaimer: Justia Dockets & Filings provides public litigation records from the federal appellate and district courts. These filings and docket sheets should not be considered findings of fact or liability, nor do they necessarily reflect the view of Justia.
Why Is My Information Online?