Unites States of America v. Apple, Inc. et al

Filing 261

RESPONSE to Public Comments on the Proposed Macmillan Final Judgment. Document filed by Unites States of America. (Attachments: # 1 Exhibit Kohn Comment)(Fairchild, Stephen)

IN THE UNITED STATES DISTRICT COURT
FOR THE SOUTHERN DISTRICT OF NEW YORK

__________________________________________
UNITED STATES OF AMERICA
Plaintiff,
v.
APPLE, INC.,
HACHETTE BOOK GROUP, INC.,
HARPERCOLLINS PUBLISHERS, L.L.C.
VERLAGSGRUPPE GEORG VON
HOLTZBRINK PUBLISHERS, LLC
d/b/a MACMILLAN,
THE PENGUIN GROUP,
A DIVISION OF PEARSON PLC,
PENGUIN GROUP (USA), INC. and
SIMON & SCHUSTER, INC.,
Defendants.

)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)

Civil Action No.12-CV-2826 (DLC)

COMMENTS OF AMICUS CURIAE BOB KOHN
REGARDING PROPOSED FINAL JUDGMENT AS TO DEFENDANTS
THE PENGUIN GROUP, A DIVISION OF PEARSON PLC
AND PENGUIN GROUP (USA), INC.

TABLE OF CONTENTS
INTRODUCTION......................................................................................................................... 1
I.

STANDARD OF REVIEW FOR TUNNEY ACT CASES IN THIS CIRCUIT ............................................ 2

II.

THE SETTLEMENT’S RESTRICTIONS ON AGENCY PRICING IS NOT IN THE PUBLIC
INTEREST .......................................................................................................................................................... 5
A. The Settlement’s Restrictions on Agency Contracts Reverse the Benefits to the Public of Defendants’
Agency Contracts .............................................................................................................................................. 5
B. Restrictions on Agency Pricing Have a Harmful Impact Upon the Relevant Market ..................................... 13
C. Restrictions on Agency Pricing Have a Harmful Impact Upon the Public Generally ..................................... 16

III. THE GOVERNMENT HAS FAILED TO COMPLY WITH THE TUNNEY ACT BY NOT
DISCLOSING DOCUMENTS THAT WERE DETERMINATIVE............................................................ 21
IV. THE COURT SHOULD APPROVE A REVISED PENGUIN SETTLEMENT ........................................ 23

CONCLUSION ........................................................................................................................... 24

ii

INTRODUCTION
These comments are respectfully submitted in opposition to several provisions of the
proposed Final Judgment (hereinafter, “Macmillan Settlement”) 1 between the United States (the
“Government”) and defendants Verlagsgruppe Georg Von Holtzbrink GmbH and Holtzbrink
Publishers, LLC d/b/a/ Macmillan (collectively, “Macmillan”).
This public interest determination boils down to whether the Court should accept as
reasonable the Government’s conclusion that the higher prices that allegedly ensued after
the adoption of the agency contracts harmed consumers and the lower prices likely to ensue
after entry of the proposed Final Judgment will benefit consumers. If it can be shown, as a
matter law, that the lower prices sought by the proposed settlement harm consumers, the relevant
market and the public generally, then Government’s conclusions about the proposed settlement
are unreasonable and the settlement cannot be in the public interest.
If the Defendants' agency agreements, by definition, eliminated price competition at the
retail level, and if those agreements were intrinsically legal—as the Government has admitted
and as the District Court has now held—then the e-book price adjustments that followed the
adoption of the agency agreements were the result of the correction of a pre-existing market
failure brought about by the undisputed, systematic below marginal cost pricing of such e-books
by Amazon, which the Government has now admitted Amazon was doing. 2 By imposing
restrictions upon agency pricing, the settlements in this proceeding reverse the adjustment of
such e-book prices back up to their efficient equilibrium (i.e., Amazon's marginal cost) and
1

Macmillan Settlement (Dkt#174-1)
See Section II below. Thus, the comments herein are focused upon new facts and the applicability of
several key rulings of the District Court in its Opinion & Order dated September 5, 2012 (Dkt# 113),
which was issued after the previous submissions of amicus curiae in this matter regarding the Final
Judgment as to Hachette, HarperCollins, and Simon & Schuster (the “Initial Final Judgment”)—including
Comments of Bob Kohn (ATC-0143, April 30, 2012), Proposed Amicus Curiae Brief of Bob Kohn dated
August 13, 2012 (Dkt#97-1), and Amicus Curiae Brief of Bob Kohn dated September 4, 2012 (Dkt#110).
2

1

therefore, as a matter of law, such restrictions have a negative impact upon the relevant market
and the public generally.
Accordingly, it would be in the public interest for the Court to approve a revised
settlement in a way that eliminates the restrictions upon Defendants’ use of the agency model
while maintaining the decree’s remaining provisions to deter exchanges of competitive
information and monitor compliance during an appropriate timeframe. 3 The proposed settlement
between the United States and defendants The Penguin Group, A Division Of Pearson Plc. and
Penguin Group (USA), Inc. (“Penguin”)

4

and the Final Judgment 5 entered with respect to

Hachette, HarperCollins, and Simon & Schuster should similarly be revised.
I.

STANDARD OF REVIEW FOR TUNNEY ACT CASES IN THIS CIRCUIT
In its response to amicus curiae’s comments to the proposed Penguin Settlement, the

Government did not deny that the standard of review it proposed has not been followed by
the Court of Appeals for the Second Circuit. The proper standard to be followed by the
District Court is set forth in the Second Circuit decisions cited in this section.
Prior to entry of a proposed final judgment, the Tunney Act requires a court to determine
that such entry is “in the public interest.” 15 U.S.C. § 16(e)(1). Though the statute does not
define that phrase, the Second Circuit has held that such words “take meaning from the purposes
of the regulatory legislation.” United States v. American Cyanamid Co., 719 F.2d 558, 565 (2d
Cir. 1984). More specifically, “the Court should look to the elements of that species of antitrust
violation.” United States v. Int’l Bus. Machines Corp., 163 F.3d 737, 742 (2d Cir. 1988).
The Government touched upon the alleged species of antitrust violation when it stated in
3

The District Court has the power to do so under Second Circuit precedent cited in Section IV below.
Penguin Settlement (Dkt#162-1)
5
Initial Final Judgment (Dkt#119).
4

2

this action that, “Low prices, of course, are one of the principal goals of the antitrust law.”
Dkt#81 (Gov’t Response) at 22 (citing, Atlantic Richfield Co. v. USA Petroleum Co., 495 U.S.
328, 340 (1990) (opinion by Justice Brennan). But, in citing Atlantic Richfield, the Government
omitted Justice Brennan’s crucial caveat, which Justice O’Conner specifically included when she
quoted Justice Brennan: “Low prices benefit consumers regardless of how those prices are set,
and so long as they are above predatory levels, they do not threaten competition.” Brook Group
v. Brown and Williamson Tobacco Corp., 509 U.S. 209, 223 (1993) (opinion by Justice
O’Conner quoting, Atlantic Richfield at 340) (emphasis added).
As noted, this case is not about limiting e-book price competition at the retail level
through the adoption of the agency model. It’s about the rise in e-book prices after the switch to
the agency model. The operative terms of each of the settlements restricts the agency model such
that e-book prices will be lowered for a period of time, at not only cost to the Publisher
Defendants, but to competitors of Amazon, independent booksellers, authors of copyrighted
works, and the public at large who will suffer from the misallocation of resources. It is in that
context the Penguin Settlement is to be considered, and the Tunney Act directs courts to consider
several factors, including “the impact of entry of such judgment upon competition in the relevant
market or markets” and its impact “upon the public generally.” 15 U.S.C. § 16(e)(1).
Though normally a court makes only the minimal determination of whether a settlement
agreement is to be accorded the status of a judicially enforceable decree, the court has a “larger
role” where the consent judgment resolves “antitrust suits brought by the United States.” Int’l
Bus. Machines, 163 F.3d 737, 740 (quoting Janus Films v. Miller, 801 F.2d 578, 582 (2d Cir.
1986)).
Some courts, including the District Court, have held that “the relevant inquiry is whether

3

the Government has established an ample ‘factual foundation for the Government’s decisions
such that its conclusions regarding the proposed settlement are reasonable’.” Dkt#113 at 14
(quoting United States v. Keyspan, 783 F.Supp.2 633, 637-38 (S.D.N.Y. 2011). Research by this
writer has been unable to uncover a single decision by the Second Circuit Court of Appeals
which adopts the Keyspan standard of review. The standard in this Circuit appears to be that
which was established in American Cyanamid and IBM, cited above.
In any event, the argument below remains unaffected should the District Court again
embrace the Keyspan formulation. Even under the Keyspan standard, the courts may “give due
respect to the government's perception of its case” (Keyspan at 638) and may be “deferential to
the government's predictions as to the effect of the proposed remedies” (Microsoft, 56 F.3d at
1461), but such respect is not due, and such deference is not earned, where the Government’s
conclusions regarding the settlement are unreasonable. Keyspan at 637-38.
In inquiring whether the Final Judgment is in the public interest, or whether
Government’s conclusions are reasonable, the courts need not engage, as the Government has
suggested, in a “full-blown, lengthy and expensive trial.” Dkt#81 (Gov’t Response to Comments)
at 44. On the contrary, the courts may inquire into the reasonableness of the Government’s
conclusions assuming all of the factual allegations regarding defendants’ conduct in the
Complaint, stripped of their bare conclusions (i.e., the subject of this inquiry under Keyspan), are
true. To defer to both the Government’s alleged facts as well as their conclusions would indeed
make “a mockery of judicial power.”
Amicus curiae is not suggesting that the Court “construct [its] own hypothetical case and
then evaluate the decree against the case,” or “redraft the complaint to inquire into other matters
that the United States did not pursue” or “engage in an unrestricted evaluation of what relief

4

would best serve the public.” Dkt#5 (Initial CIS) §VII (quoting United States v. Microsoft Corp.,
56 F.3d 1448, 1459-62 (D.C. Cir. 1995); Keyspan at 638; United States v. BNS, Inc., 858 F.2d
456, 462 (9th Cir. 1988)). Nor is the Court being asked to “reach beyond the complaint to
evaluate claims the government did not make.” Microsoft, 56 F.3d at 1459.
On the contrary, in evaluating the reasonableness of the Government’s conclusions, the
District Court need simply assume all of the Government’s factual allegations as true, and add to
those allegations the Court’s own factual findings. It is respectfully submitted that, when Reason
is applied to the factual foundation, the District Court’s only conclusion can be that elements of
the Penguin Settlement that restrict the operation of the agency model, for however short a
period, are not in the public interest.
II.

THE SETTLEMENT’S RESTRICTIONS ON AGENCY PRICING IS NOT IN
THE PUBLIC INTEREST
The Macmillan Settlement is not in the public interest because the restrictions it imposes

on agency pricing unwinds the pro-competitive impacts of Defendants’ conduct upon the
relevant market and the favorable impacts of such conduct upon the public generally.
A.

The Settlement’s Restrictions on Agency Contracts Reverse the Benefits to
the Public of Defendants’ Agency Contracts

The Government has concluded that the restrictions on the agency contracts in the
proposed settlement will benefit consumers. This public interest determination boils down to
whether the Court should accept as reasonable the Government’s conclusion that the higher
prices that ensued after the adoption of the agency contracts were harmful and the lower prices
the Government says are “likely” to ensue after entry of the proposed Final Judgment are
beneficial to the relevant market and the public generally. If it can be shown, as a matter law,
that such lower prices are harmful, the relevant market and the public generally, then
5

Government’s conclusions about the proposed settlement are unreasonable and the settlement
cannot be in the public interest.
It is self-evident that the purpose and effect of the agency agreements, by definition, was
to eliminate e-book price competition among e-retailers. This is because, under the agency
model, price competition for e-books occurs among the publishers, not the retailers. Indeed, the
Government’s admission that the elimination of price competition at the retail level is not
unlawful is virtually codified in each of the settlements. 6 It should be clear, then, that this action
was never about eliminating e-book price competition at the retail level. It was merely about the
increase in e-book prices that ensued after the adoption of the agency contracts. But if those
increase in prices were beneficial, then any decease in prices caused by the proposed settlement
is harmful. Even if the publishers had not colluded as the Government has alleged, the
adoption of agency pricing would still have resulted in higher e-book prices, because the
agency model forced Amazon to cease selling e-books at below their marginal cost.
Until recently, the Government carefully avoided ever admitting that Amazon was selling
trade e-books at below marginal cost. Until the Court’s Opinion & Order regarding entry of the

6

See, for example, Macmillan Settlement (Dkt#174-1) at §V.D. (“After the expiration of prohibitions in
Sections V.A. and V.B of this Final Judgment, this Section V.D. shall not prohibit Penguin from
unilaterally entering into or enforcing any agreement with an E-book Retailer that restricts, limits, or
impedes the E-book Retailer from setting, altering, or reducing the Retail Price of any of Penguin’s Ebooks or from offering price discounts or any other form of promotions to encourage consumers to
Purchase any of Penguin’s E-books”). Oddly, in support of its ruling that the Government met the
“minimal” standard of review for the initial Final Judgment, the District Court found that it was
“undisputed that the Agency Agreements disallowed retail price discounting” and that after the adoption
of such agreements “a consumer could not find Publisher Defendants’ newly-released and bestselling
books for $9.99 at any retailer.” Dkt#113 (Order) at 32-33. If the District Court was suggesting here that
the agency agreements, by disallowing discounting, were somehow unlawful, then the District Court will
need to explain how the agency agreements, which intrinsically eliminate price competition at the retail
level, can be considered unlawful and “not intrinsically unlawful” at the same time. Cf. Dkt#113 (Order)
at 17. If all the District Court was suggesting was that e-book prices rose above the $9.99 price, it has not
explained how such price rises could have harmed consumers or were otherwise unlawful. If the Court is
suggesting here that all price increases harm consumers or are necessarily unlawful, then it clearly
adopted the Government’s mistaken view of pricing under the antitrust law as its own.
6

Initial Final Judgment, it was thought to be undisputed that, under the retail model, Amazon was
selling e-books made available by the defendant publishers and by most, if not all, independent
book publishers, at below its marginal cost, consistently, “[f]rom the time of its Kindle launch”
until Amazon’s acceptance of the agency model. See, Dkt#5 CIS §II.A. Even the Government
directly alleged facts in its Complaint and Competitive Impact Statement that Amazon was
selling e-books at below their marginal cost, something which the Government has never
denied. 7 Amazon’s marginal cost, under the retail model, was at least equal to the fixed
wholesale price (e.g., $13.00, which is 50% of a typical list price of $26.00) it pays to the
publishers for each incremental e-book it sells. 8 Dkt#5 (CIS) §II.A. These books were re-sold by
Amazon for “$9.99-or-less.” Id.; Dkt#1 (Complaint) at ¶30. Thus, for example, at a marginal cost
per e-book of $13.00, Amazon was selling such e-books at $3.01 (more than 23%) below its
marginal cost.
The Government stated it “reviewed data from Amazon and others” to investigate
Amazon’s e-book distribution business. Dkt#81 (Gov’t Response) at 21. By the end of its
investigation, the Government concluded that Amazon was not engaging in predatory pricing. Id.
Dkt#1 (Complaint) ¶30. It is noteworthy, however, that in reaching such conclusion, the
Government never alleged or asserted that Amazon was not selling below marginal cost.
The Government merely asserted that no objector supplied evidence that “Amazon threatens to
drive out competition and obtain monopoly power.” Dkt#81 at 22. Needless to say, that is not the
same as saying Amazon did not re-sell e-books below its marginal cost—a necessary element of

7

It may well be that the District Court was mislead by certain carefully crafted statements about
Amazon’s pricing policies set forth in the Government’s Complaint and Response to Public Comments
dated August 3, 2012, discussed below.
8
Amazon has additional marginal costs, such as the cost of wirelessly delivering each incremental e-book
to a Kindle e-reader, which is estimated to cost nearly 15 cents per e-book delivery. See, Kindle Direct
Publishing Agreement, Pricing Page (as of July 28, 2012).
7

predatory pricing.
Being unable to deny facts which it provided in its Complaint and CIS, and then without
ever categorically denying that Amazon was selling e-books at below marginal cost, the
Government jumped to its peculiar conclusion: based on its investigation and review of data
from Amazon and others, “Amazon’s e-book distribution business has been consistently
profitable.” Dkt#1 ¶30; repeated in Dkt#81 (Gov’t Response) at 4, 21-22).
Implicit in that conclusion was the legal standard which the Government applied to its
review of data. The Government provides no citation for the proposition that the standard for
predatory pricing under Section 2 of the Sherman Act (15 U.S.C. §2) is whether the predator’s
business “has been consistently profitable.”
In Brook Group v. Brown and Williamson Tobacco Corp., 509 U.S. 209, 222-24 (1993),
the Supreme Court set forth two elements for an action for predatory pricing. The first is to
establish that prices were set at “below an appropriate measure of its rival’s costs.” Interpreting
its prior holdings, the Court stated, “below-cost prices should suffice.” Id at 223. This was
consistent with the standard previously adopted by the Second Circuit. Northeastern Telephone
Co. v. American Tel. & Tel. Co., 651 F.2d 76, 86-89 (2d. Cir. 1981), cert. denied, 455 U.S. 943
(1982) (“The relationship between a firm’s prices and its marginal costs provides the best single
determinate of predatory pricing”). Marginal cost is traditionally defined as “the increment to
total cost that results from producing an additional increment of output.” Northeastern at 87
(quoting Areeda & Turner).
This prerequisite is clearly satisfied by the Government’s own factual allegations. By
selling e-books for “$9.99-or-less” that it purchased at higher amounts Amazon was selling such
e-books at below their marginal cost. Unless the price cutter can show that its low price is purely

8

“‘promotional,’ e.g., a ‘free sample’,” or “show that it expects costs to fall when sales increase,”
then “the firm cannot rationally plan to maintain the low price.” Barry Wright Corp. v. ITT
Grinnell Corp., 724 F.2d 227, 230-34 (1st Cir. 1983) (opinion by Circuit Judge Breyer).
Amazon’s practice of below marginal cost pricing was no Macy’s “promotion.” These
were not “free samples” or “loss leaders.” Amazon’s below-cost pricing was consistently applied
to newly-released and bestselling trade e-book made available by the Defendant Publishers and
by most, if not all, independent book publishers, from the time of its Kindle launch until
Amazon’s acceptance of the agency model. Dkt#5 §II.A. “Without special circumstances there is
little to be said in economic or competitive terms for such a price cut.” Barry Wright, 724 F.2d
227, 231. The Government has supplied neither the public nor the District Court with any
evidence of special circumstances to justify Amazon’s below marginal cost pricing. 9
Nor can Amazon expect its marginal cost for e-books to decrease with volume. Amazon’s
marginal cost for selling each incremental e-book having a list price of $26.00 will always
remain $13.00, whether Amazon sells one copy or one million copies. “At a minimum, one
would wonder why this firm would cut prices on ‘incremental production’ below its ‘avoidable’
costs unless it later expected to raise its prices and recoup its losses.” Barry Wright, 724 F.2d
227, 232.
At first, the District Court seemed to embrace the Government’s odd conclusion that
“Amazon’s e-book business was ‘consistently profitable’.” Dkt#113 (Order) at 39. But then, it
went on to elucidate the law of predatory pricing correctly, only to overlook the Government’s
own factual allegations supporting Amazon’s below cost pricing and the apparent care the
Government took never to deny it.
9

If the Government has materials or documents that would shed light on this key issue—containing a
“smoking gun” or its “exculpatory opposite”—it is bound by the Tunney Act to share it with the public
and the District Court. See discussion below in Section II.E.
9

For example, this is how the Government defended Amazon against the charges by
hundreds of commentators of below cost pricing:
No objector to the proposed Final Judgment has supplied evidence that, in the dynamic
and evolving e-book industry, Amazon threatens to drive out competition and obtain the
monopoly pricing power which is the ultimate concern of predatory pricing law. [Dkt#81
(Gov’t Response) at 22 (emphasis added)].
While addressing the ultimate concern of predatory pricing law, the Government left out
the law’s immediate concern, the first prong of the test under Brook Group and Northeastern:
whether Amazon was selling at below marginal cost. The Government never denied this, because
the “evidence” of Amazon’s selling below marginal cost was not only obvious and abundant, it
was factually supported in the Government’s own Complaint and CIS.
Yet, when the District Court addressed the issue of below marginal cost pricing, it stated,
[W]hile comments complain that Amazon’s $9.99 price for newly-re-leased and
bestselling e-books was ‘predatory,’ none of them attempts to show that Amazon’s ebook prices as a whole were below marginal cost. [Dkt#113 at 39]. 10
Of course, none of public commentators attempted to “show” it, because none of them
had to. It was undisputed. As the Government observed, Amazon’s below-cost pricing was the
“overarching theme” of the public comments. Dkt#81 at 2. Facts supporting it were directly
alleged in the Government’s own Complaint and Competitive Impact Statement and the
Government has never since denied that Amazon was selling e-books at below marginal cost.
10

At a procedural level, what the District Court has done here is suggest that it was not the Government’s, but the
public’s responsibility somehow “to show” that Amazon was selling below marginal cost. Where in the Tunney Act
does the public have the right to conduct discovery on Amazon’s marginal costs? (How else could they have learned
this information?) Needless to say, nowhere. On the contrary, the Tunney Act specifically puts the burden of
producing such facts upon the Government, requiring that “materials and documents that were considered
determinative” in formulating the settlement shall “be made available to the public at the district court.” 15 U.S.C. §
16(b). The District Court has simply turned the Government’s statutory responsibility to the public on its head:
instead of the Government, the public is now somehow responsible for filing with the District Court materials and
documents that could be the “smoking gun” (or exculpatory opposite) of Amazon’s below marginal cost pricing.
See, discussion in Section III below regarding United States v. Bleznak, 153 F.3d 16 (2d Cir. 1998). The District
Court should either require the Government to produce its documents relating to its investigation of Amazon’s trade
e-book pricing or appoint a special master under the Tunney Act, Section 16(f), to conduct discovery directly on
Amazon to achieve that end.
10

Indeed, the express terms of each of the proposed settlements—which allow each Publisher
Defendant to prevent Amazon from selling the publisher’s “entire catalogue at a sustained
loss”—is an express admission by the Government that Amazon was selling below marginal
cost. See, for example, Competitive Impact Statement as to Macmillan (Dkt#175) at 8;
Macmillan Settlement (Dkt#174-1) at Section VI.B.
Finally, to cap it off, the Government has now finally conceded that Amazon’s e-book
prices as a whole were below marginal cost. Specifically, in its Response to Public Comments
on the Penguin Settlement, the Government asserted that the degree to which Amazon may
discount under the settlement is limited to the “the level of [the e-retailer’s] aggregate
commission.” Government Response to Penguin Settlement at 12. Thus, according to the
Government, the settlement permits Amazon to resume selling e-books at below marginal
cost, albeit at prices “closer to their marginal cost” than before. Id. at 12-13. 11
In its criticism of the public’s failure to come forth with sufficient facts regarding
Amazon’s below marginal cost pricing, the District Court seemed careful to use the phrase “as a
whole,” somehow suggesting that it has to be shown that Amazon was offering their e-books “as
a whole” below marginal cost. Even if the District Court was meaning to suggest that such a
showing is an essential element of an action against for predatory pricing, it is beside the point:
to judge the reasonableness of the Government’s conclusions regarding the proposed settlements,
it need not be proven that that Amazon would actually be liable in an action for predatory pricing
in order to show that the rise in e-book prices that occurred after the switch to the agency model
could not have harmed to consumers. That is, it cannot logically be denied that such increase
in e-book prices constituted a correction of the market failure that Amazon created with
respect to the huge number of e-books it systematically, and undisputedly, sold below its
11

How the Government could construe this as “a goal Mr. Kohn applauds” is baffling.
11

marginal cost. To the extent the Court believes it must be shown that Amazon’s below-cost
pricing of e-books had to be offered “as a whole” below marginal cost, the Court can now turn to
the Government’s admission of that fact—i.e., that Amazon is now selling e-books below,
though “closer to” their marginal cost.
In a footnote in its Response to the Penguin Settlement (note 6 at 13), the Government
notes Northeastern found only that prices below marginal costs will be “presumed predatory”
(citing 651 F.2d 76, 78 (2d Cir. 1981). And added, “To succeed on a predatory pricing claim, an
antitrust plaintiff must also establish that there is a ‘dangerous probability’ that the defendant
will later ‘recoup[] its investment in below-cost prices.’ Brook Group Ltd. v. Brown &
Williamson Tobacco Corp., 509 U.S. 209, 224 (1993).” The Government should have kept
reading; Kohn’s comments to the Penguin Settlement goes on to say:
The second prerequisite for predatory pricing is a demonstration that the predator has a
“dangerous probability[] of recouping its investment in below-cost pricing.” Brook Group, 509
U.S. 209, 224. In order to recoup their losses, the predator “must obtain enough market power to
set higher than competitive prices.” Id. at 225. This prerequisite was satisfied by the District
Court’s factual finding that Amazon achieved a “90 percent monopoly” of the e-book
market. Dkt#113 34-35. The existence of the requisite market power ordinarily may be inferred
from the predominant share of the market. See, United States v. Grinnell Corp., 384 U.S. 563,
571 (1966) (in which 87% of the market constituted “monopoly power”).
Moreover, in the Second Circuit, once the first prerequisite is demonstrated, the second is
presumed. With such a presumption in hand, the Government should not have been expecting
members of the public to produce “evidence” of Amazon’s threat to drive out competition
(Dkt#81 at 22), but disclosing to the District Court exculpatory facts that would overcome that

12

presumption. Without the presence of facts to overcome the presumption, conduct that
resulted in raising illegally-low (consumer welfare-harming) prices back up to the
consumer welfare-optimizing level of marginal cost could not have, as a matter of law,
harmed consumers.
The Government now having conceded that Amazon’s e-book prices as a whole were
below marginal cost, the settlements cannot be in the public interest, because any increase in ebook prices back up to the level of Amazon’s marginal cost, could not have harmed consumers.
On the contrary, the alleged increase in prices could only have benefited consumers, because
selling below marginal cost, as recognized by the Second Circuit in Northeastern discussed
below, causes a misallocation of resources. Thus, with lower prices not only being the stated
intent of the proposed settlements, but the only logical result of them, the Government’s
conclusions about the settlements, even under the Keyspan standard, could not have been
reasonable—in which case, the settlements cannot be held to be in the public interest. Moreover,
to this point, it doesn’t matter whether the alleged collusive conduct was unjustifiable. Higher
prices resulting from the settlements, because they have a harmful impact upon the relevant
market and the public generally, cannot be in the public interest.
B.

Restrictions on Agency Pricing Have a Harmful Impact Upon the Relevant
Market

The Macmillan Settlement reverses at least two undisputed pro-competitive impacts upon
the relevant market: reducing Amazon’s monopoly power and monopsony power. Meanwhile,
the settlement’s only alleged pro-competitive impact has been found by the District Court not to
exist.
1.

Impact Upon Amazon’s Monopoly Power

As noted, the District Court found that it is “undisputed” that Amazon had achieved a “90
13

percent monopoly” in the e-book market and that, following the course of defendants’ conduct as
alleged in the Complaint, Amazon’s market share in trade e-books dropped from an undisputed
90% to 60%. Dkt#113 at 34-34. The restrictions placed on the Defendants’ use of agency pricing
set forth in the settlements unwind the pro-competitive effects which the District Court found to
be undisputed.
2.

Impact Upon Amazon’s Monopsony Power

The Complaint itself alleges ample facts to demonstrate that defendant publishers faced
in Amazon a single buyer generating 90% of their e-book revenues who both wielded and
exercised demonstrable monopsony power. Dkt#1 (Complaint) at ¶80. 12 It has been said that a
monopsony is the “mirror image” of monopoly. Todd v. Exxon Corp., 275 F.3d191, 202 (2d Cir.
2001). Under the textbook economic definition, the monopsonist, in depressing the price of the
goods purchased, transfers wealth to itself from the supplier of the goods. According to the
Government’s own Merger Guidelines, the monopsonist will not pass along the lower price input
to its downstream consumers. Horizontal Merger Guidelines of 2010 at §12. 13 And, as discussed
infra, the use of monopsony power to reduce the price paid to publishers and authors for e-book
distribution rights is antithetical to the exercise of the rights of copyright owners, whose pricing
is already significantly constrained by the free-rider/piracy problem. 14
When Amazon’s 90% monopoly of the e-book market dropped to 60%, Amazon’s

12

Amazon’s exercise of this power against Macmillan was not the first time. Doreen Carvajal, “Small
Publishers Feel Power of Amazon’s Buy Button,” New York Times (June 16, 2008). See, Comments of
Author’s Guild (ATC-0214 at 3, 5-6. http://www.justice.gov/atr/cases/apple/comments/atc-0214.pdf.
13
U.S. Dept. of Justice and Federal Trade Commission, Horizontal Merger Guidelines (August, 19, 2010)
http://www.justice.gov/atr/public/guidelines/hmg-2010.pdf
14
Moor-Law 527 F.Supp. 758, 763, discussed infra. See also, John Cirace, CBS v. ASCAP: An Economic
Analysis of a Political Problem, 47 Fordham L. Rev. 277, 293-94 (1978-79) (the collusive conduct by
music publishers in Broadcast Music was justified, not by reduced transaction costs, but by the need to
balance the monopsony power of the three TV network buyers of musical performances), discussed in
Proposed Amicus Curiae Brief of Bob Kohn (Dkt#97-1) at 12-14.
14

monopsony power dropped correspondingly. Neither the Government nor the District Court
disputed the fact that this pro-competitive drop in Amazon’s monopoly and monopsony power
were directly attributed to Defendants’ switch to the agency model.

3.

Alleged Pro-Competitive Impact Upon the Relevant Market Never
Existed

Initially, in its CIS, the Government concluded that the agency price restrictions in the
Final Judgment would have pro-competitive impact upon the relevant market, because the
termination of defendants’ agency contracts would reverse the elimination of e-book price
competition among e-retailers. Dkt#5 §III.A.1. But the District Court has since ruled that this
alleged pro-competitive impact never existed. Since the District Court found (Dkt#113 at 33), as
the Government recognized (Dkt#81 at vi), that Defendants’ agency contracts were not
“intrinsically unlawful,” the effect of the agency contracts—the replacement of e-book price
competition at the retail level with e-book price competition at the publisher level—was not
“intrinsically” anticompetitive.
The worst that could be said about the shift to the agency model was that it had a neutral
impact upon the e-book market. Thus, since the objective of the Penguin Settlement is to return
e-book price competition to e-retailers (Dkt#5 III.A.1), such objective is not pro-competitive at
all. 15

15

Subsequent to entry of the initial Final Judgment, two of the world largest trade book publishers
announced their intention to merge. Eric Pfannner & Any Chozick, “Random House and Penguin Merger
Creates Global Giant,” New York Times (October 29, 2012) (“The deal, analysts said, would give the new
company, to be called Penguin Random House, greater scale to deal with the challenges arising from the
growth of electronic books and the power of Internet retailers…companies [who have] the size to let them
negotiate better terms on prices.” Given the impact of the Final Judgment upon the relevant market, the
proposed merger would appear, not unlike the alleged collusive switch to the agency model, to be a procompetitive response to Amazon’s monopsony power. See, Cirace, supra. Apparently, the Justice
Department would not disagree, as it has announced that it will not raise an anti-trust objection to the
15

C.

Restrictions on Agency Pricing Have a Harmful Impact Upon the Public
Generally

Defendants’ conduct, as alleged, had undisputed positive impacts upon the public
generally that would be unraveled by the Penguin Settlement.
1.

Harms the Public Interest in Promoting the Creation of Works of
Authorship

One of the essential the considerations in promoting the public interest underlying
copyright and antitrust is the distinctive nature of the products at issue—copyrightable works of
authorship delivered and consumed in digital form.
Recognizing the “tax” 16 that the copyright imposes upon consumers required to achieve
its important public purpose, the Department of Justice and the Federal Trade Commission
jointly devised a specific policy to be applied when enforcing the antitrust laws in cases
involving copyrights and other forms of intellectual property. The antitrust enforcement policy of
the United States with respect to copyrighted works, such as e-books, is stated in the Antitrust
Licensing Guidelines for the Licensing of Intellectual Property (April 6, 1995) (“IP
Guidelines”). 17 The IP Guidelines recognize that intellectual property has important
characteristics that distinguish it from other forms of property. IP Guidelines §2.1. Antitrust
analysis must take these differences into account in evaluating specific market circumstances in
which transactions regarding them occur. Id.
E-books, like music performances and downloads, have the classic characteristics of what

merger. Leslie Kaufman, “Justice Department Approves Random House-Penguin Merger,” New York
Times (February 14, 2013).
16
Though it results in “a tax on readers for the purpose of giving a bounty to writers” (Golan v. Holder,
132 S.Ct. 873, 899 (2012) (opinion by Justice Ginsburg, quoting T. Macaulay, Speeches on Copyright 25
(E. Miller ed. 1913), copyright remains an essential means by which “an important public purpose may be
achieved.” Sony Corp. of America v. Universal Studios, Inc., 464 U.S. 417, 429 (1984).
17
http://www.justice.gov/atr/public/guidelines/0558.pdf.
16

are known as, “public goods.” See, Broadcast Music, Inc. v. Moor-Law, Inc., 527 F.Supp. 758
(D. Del. 1981), aff’d without published opinion, 691 F.2d 490 (3d Cir. 1982). First, unlike
private goods (e.g., apples or printed books), which can be withheld from the market and
released only in exchange for payment, an e-book can be consumed without leaving any less for
others to consume. See, Moor-Law at 763. Second, the digital nature of e-books facilitates the
reproduction of perfect copies at virtually no cost, making it difficult for the copyright owner to
exclude persons who do not pay for consuming the e-book—a problem known as the free rider
problem, 18 misappropriation, 19 infringement, 20 illegal file sharing, 21 or piracy.
For these reasons, courts have recognized that “the natural market forces of supply and
demand do not operate normally on pricing in this market.” Moor-Law at 763. This is because
the cost of acquiring a copy of and consuming an additional printed book costs the consumer
something. By contrast, because of the free rider or piracy problem, the marginal cost of
acquiring and consuming an additional e-book can be as low as zero. Moor-Law at 763.
The public good characteristics of e-books make normal cost-based pricing infeasible. Id.
For example, the price that a publisher can charge for an e-book is subject to a natural constraint:
illegal competitors charging zero for the same e-book. In Moor-Law, the District Court found, as
an observed fact, that “the free rider problem does provide a significant constraint on the price
[the copyright owner] charges.” Id. The free rider problem tends to make the copyright owner’s
enforcement costs high. Because higher enforcement costs can more than consume increased
revenue from a higher price, the copyright owner considers this problem when setting a price. Id.
Thus, there is an inherent limit to how much “tax” may be exacted from consumers by

18

Id.
IP Guidelines §1.0.
20
Copyright Act, 17 U.S.C. §501(a).
21
A&M Records, Inc. v. Napster, Inc., 239 F.3d 1004, 1011 (9th Cir. 2001).
19

17

book publishers for e-books, because they must compete with free riders who can readily take
advantage of the e-books’ digital form to cheaply trade in pirated copies without payment.
Amazon has repeatedly used its monopsony power in attempts to lower its cost for books.
In January, 2008, Amazon removed the buy buttons from more than 1,000 self-published printed
books and threatened that these books would be permanently removed from the Amazon website
unless the books were published via Amazon’s own BookSurge print-on-demand service. See,
Comments of Authors Guild (ATC-0214) at 4. That 2008 incident was a precursor to what
Amazon did to Macmillan on Friday, January 29, 2010. Dkt#1 ¶80. Had Amazon succeeded in
putting Macmillan out of business, there would have been one less major publisher competing to
acquire publishing rights from authors—resulting in authors being paid less for publication rights
to their e-book manuscripts.
The use of monopsony power to reduce the price paid to publishers and authors for ebooks is antithetical to the exercise of the rights of copyright owners, whose pricing is already
significantly constrained by the free-rider/piracy problem. Moor-Law at 763. At the same time,
there is no corresponding benefit to consumers. According to the Government’s own Merger
Guidelines, the monopsonist will not pass along the lower price input to its downstream
consumers. Horizontal Merger Guidelines of 2010 at §12.
These critical distinctions between e-books and printed books should have been part of
the factual foundation underlying the Government’s decisions in this case. Yet, no mention of
these distinctions, nor mention of the IP Guidelines, can be found in any of the Government’s
filings in this action or the District Court’s Entry Order. Dkt#113. By failing to take them into
account, the Government’s conclusions regarding the Penguin Settlement were not reasonable.
By restricting the freedom of authors and publishers to control the price of their works to the

18

public, in way that actually harms competition in the relevant market and the public generally,
the proposed settlement unjustifiably interferes with the essential means by which the U.S.
Constitution provides them for the purpose of promoting the important public interest in the
creation of Writings. 22

2.

Harms the Public Interest in Efficient Prices for E-Books

In its response to amicus curiae’s comments to the proposed Penguin Settlement (at 12),
the Government made a startling admission: the discounting the settlements require publishers to
allow is limited to “the level of [the e-retailer’s] aggregate commission” and thus, an e-retailer
“will be selling e-books closer to their marginal cost.”
Hence, the agency model, which is intrinsically lawful, which put an end to below
marginal cost pricing, returned e-book prices to their economically efficient level for
consumers; the settlement, in allowing e-retailers to resume selling e-books at prices that
are below (though “closer to”) their marginal cost, partially reverses the market-correcting
adjustment in prices following the agency contracts. Reversing such price adjustments,
even partially (by virtue of the aggregate limitation), causes an economically inefficient
result that harms consumers and the public generally.
While the public has a Constitutionally-recognized interest in the promotion of works of
authorship, it does not have any interest in low prices. See, Dkt#81 at 21-22 (where the
Government expresses its mistaken belief that “low prices” are a “core ambition” of free
markets). Low prices are not the core ambition of either the law or the markets. Efficient prices
are. As the Second Circuit has held, consumer welfare is not maximized by “low” pricing, but by
22

U.S. Constitution, Article I, §8; Copyright Act, 17 U.S.C. §101 et.seq.; Sony Corp., 464 U.S. 417, 429
(1984).
19

“marginal cost” pricing. Northeastern 651 F.2d at 87-88.
By seemingly borrowing the Government’s mistaken primacy of “low prices,” the
District Court was lead, for example, to reject the legitimate concerns expressed by “bricks &
mortar” retailers. Retailers of printed books are not seeking protection from “the working of the
market” or the “vicissitudes” of competition. Dkt#113 at 23-24. The digital disintermediation of
printed book retailers may well be inevitable, but Amazon’s below marginal cost pricing does
not constitute a “working of the market” or “competition on the merits.” As Areeda & Turner (at
712) explain, selling below marginal cost leads to an improper allocation of resources and
“greatly increases the probability that rivalry will be extinguished or prevented for reasons
unrelated to the efficiency of the monopolist.” See, Northeastern, 651 F.2d at 87-88.
By allowing a predatory-induced market failure to resume for another two years, the
Penguin Settlement will cause the natural disintermediation of physical bookstores to occur
faster than they would in a competitive market, putting at risk the businesses of physical
retailers, large and small, perhaps years before the digital evolution of the book industry would
otherwise disintermediate them. The immediate beneficiary of this accelerated digital
disintermediation is the predator itself, Amazon. This is precisely the kind of misallocation of
resources that predatory pricing causes and why the Second Circuit makes it presumptively
illegal. It invokes the very essence of the Second Circuit’s declaration that the purpose of
antitrust law is “not to protect businesses from the working of the market,” but “to protect the
public from the failure of the market.” United States v. Int’l Bus. Machines, 163 F.3d 737, 74142 (2d Cir. 1998) (emphasis added).

20

III.

THE GOVERNMENT HAS FAILED TO COMPLY WITH THE TUNNEY ACT
BY NOT DISCLOSING DOCUMENTS THAT WERE DETERMINATIVE
In its response to amicus curiae’s comments to the proposed Penguin Settlement (at 13),

the Government responded that the Court has already determined the materials supplied by the
Government provided an “ample factual foundation for [its] decisions regarding the proposed
Final Judgment.” Opinion & Order, Dkt# 113 at 12-13. The question of whether determinative
materials and documents must be disclosed does not turn on whether the Government has
provided an “ample factual foundation,” but whether under 9th Circuit law the Government made
available “documents that are either ‘smoking guns’ or the exculpatory opposite.” United States
v. Bleznak, 153 F.3d 16, 20 (2d Cir. 1998). Particularly in light of the Government’s admission
that the settlement permits Amazon to sell e-books as a whole below their marginal costs, the
Court should require the Government to disclose to the public by filing with this Court its
materials and documents that were determinative on the issue of whether Amazon’s e-book
business was “consistently profitable.”
As noted above, this amicus believes that either the Court has been mislead about the
Government’s investigation regarding Amazon’s pricing practices or the Court was simply
wrong to have put the burden on the public to “show that Amazon’s e-book prices as a whole
were below its marginal costs.” Id at 40. Otherwise, members of the public deserve an
explanation from the Court as to how they were expected to accomplish such a task (1) without
the Court exercising the powers it has at its disposal under 15 U.S.C.16(f)(1-3) or (2) without
being able to review the Government’s documents and materials that could constitute a “smoking
gun” or its “exculpatory opposite” on the issue of whether “Amazon’s e-book prices as a whole
were below its marginal costs.”
Compliance with the procedural requirements of the Tunney Act ensures the district court
21

has before it the information it needs in order to make an informed determination whether the
decree is in the public interest. Massachusetts v. Microsoft, 373 F.3d 1199, 1246 (D.C. Cir.
2004). Accordingly, the Act requires that the Government make available to the public at the
district court “any other materials and documents which the United States considered
determinative in formulating” the proposed decree. 15 U.S.C. §16(b) [emphasis added]. Such
requirement is confined “to documents that are either ‘smoking guns’ or the exculpatory
opposite.” United States v. Bleznak, 153 F.3d 16, 20 (2d Cir. 1998).
In its CIS, the Government asserted that it considered “no determinative materials or
documents” in formulating the settlement. Dkt#163§VIII. Given, however, that selling below
marginal cost is presumed illegal in this Circuit, it is not up to objectors to supply “evidence” to
the Government of Amazon’s threat “to drive out competition.” That threat was presumed when
the Government alleged facts that Amazon was pricing below marginal cost. Dkt# ¶¶2, 30;
Dkt#5 §II.A; Dkt#163 §II.A. Thus, the burden was not on the public, but rather on the
Government to overcome that presumption with facts supporting their conclusion that Amazon’s
below marginal cost pricing was not exclusionary.
Accordingly, materials or documents supporting the Government’s conclusion that
Amazon’s e-book business was “consistently profitable” would clearly constitute “either
‘smoking guns’ or the exculpatory opposite.” Bleznak, 153 F.3d 16, 20. This is because, to the
extent Amazon was pricing e-books at below its marginal cost, any conduct by defendants to
raise prices back to the marginal level could not have harmed consumers. Since the alleged harm
to consumers constituted a “substantial inducement to the government to enter into the consent
decree” (Bleznak at 21) such materials and documents were not merely relevant, but
determinative. By failing to disclose them to the public, it is not in compliance with the Tunney

22

Act and, because they go “to the essence of” the Government’s conclusions regarding the
Penguin Settlement, entry of such settlement must be denied unless and until such determinative
documents are disclosed. See, United States v. Bechtel Corp., 648 F.2 660, 664 (9th Cir. 1981).
IV.

THE COURT SHOULD APPROVE A REVISED PENGUIN SETTLEMENT
By entering the Initial Final Judgment in this case, the District Court has made a

regrettable error that—by virtue of the restrictions on the agency pricing—has saddled
consumers with an unnecessary misallocation of resources in the trade e-book market—lasting
the next two years, perhaps longer. The Court should not compound its mistake by entering
either the proposed Macmillan Settlement or proposed Penguin Settlement—at least not without
modification. 23 Only by modifying all of the settlements in this case to remove their respective
restrictions on the Defendant’s agency agreements will the settlements align with the public
interest.
Accordingly, it is submitted that a settlement that contains only the following provisions
is in the public interest: Sections I, II, III, V (E and F only), VI (A only), VII, VIII, IX, X, XI,
and XII. The intention is to eliminate all of the restrictions on agency pricing, which would
otherwise have the effect of unwinding the pro-competitive and impacts upon the relevant
market and favorable impacts upon the public generally of Defendants’ conduct. At the same
time, the balance of the settlement would ensure that any future competitive information
exchanges between or among the Defendants will be enjoined and monitored for compliance
over an appropriate timeframe.
The Government cites no Second Circuit authority for the proposition that the District
23

The public interest does not require that the proposed Penguin Settlement be rejected in its entirety. The
District Court is urged to enter a revised Penguin Settlement comprised only of provisions that do not
unravel the pro-competitive effects that Defendants’ conduct had upon the relevant market and the public
generally.
23

Court’s authority under the Tunney Act is limited to merely approving or disapproving a consent
decree. Indeed, the Second Circuit, in de novo review of a District Court’s decision, has held that
itself may vacate a consent decree in part and “remand the case to the District Court for entry of
a revised decree consistent with [its] opinion.” Guardians Assoc. New York Police Dept., Inc. v.
Civil Service Comm’n of the City of New York, 630 F.2d 79, 113 (2d Cir. 1980). In its Response
to Comments on the Penguin Settlement, the Government did not deny that this Court,
under Guardians, has the authority to revise a consent decree.

CONCLUSION
In the Second Circuit, the courts have a “larger role” where a consent judgment resolves
“antitrust suits by the United States.” United States v. Int’l Bus. Machines Corp., 163 F.3d 737,
742 (2d Cir. 1988). The disinfectant of the Court’s public interest determination under the
Tunney Act need not be a placebo. Cf., Broadcast Music, Inc. v.DMX, Inc., 683 F.3d 32, 49 (2d
Cir. 2012). The Court is urged to set aside the “rubber stamp” which the Government submitted
in the form of its proposed standard of review and approve the proposed Penguin Settlement to
the extent it is revised to eliminate all restrictions upon the use of agency pricing.
DATED: April 29, 2013

Respectfully submitted,

_______________________
BOB KOHN
140 E. 28th St.
New York, NY 10016
Tel. [phone]
Fax. [phone]
eMail: [e-mail]

24