Nemet et al v. Volkswagen Group of America, Inc. et al

Filing 97

Amended ORDER by Judge Charles R. Breyer denying (7697) Plaintiffs' Motion to Exclude; denying (7699) Plaintiffs' Motion to Exclude; granting (7706) Defendants' Motion to Exclude in case 3:15-md-02672-CRB; denying (81) Plaintiffs� 39; Motion to Exclude; denying (83) Plaintiffs' Motion to Exclude; dismissing case for lack of jurisdiction in case 3:17-cv-04372-CRB (crblc2, COURT STAFF) (Filed on 11/12/2020)Any non-CM/ECF Participants have been served by First Class Mail to the addresses of record listed on the Notice of Electronic Filing (NEF)

1
2
3
4

UNITED STATES DISTRICT COURT

5

NORTHERN DISTRICT OF CALIFORNIA

6
7

IN RE: VOLKSWAGEN “CLEAN DIESEL”
MARKETING, SALES PRACTICES, AND
PRODUCTS LIABILITY LITIATION
_____________________________________/

8
9
10

This Order Relates To:
Dkt. Nos. 7673, 7697, 7699, 7706;
Nemet, No. 3:17-cv-4372-CRB
_____________________________________

United States District Court
Northern District of California

11
12

MDL No. 2672 CRB (JSC)

ORDER EXCLUDING EVIDENCE AND
DISMISSING CASE FOR LACK OF
JURISDICTION

This case involves a putative class of automobile consumers who bought or leased

13
14

cars without receiving a feature for which they paid, but who then unwittingly sold or

15

returned the same cars without giving that feature for which they were compensated.
These consumers purchased or leased then parted with cars that they thought

16
17

produced low emissions. In actuality, Volkswagen had misled them, and everyone else,

18

about the emissions. As the Court has explained, it is possible that Volkswagen’s fraud

19

injured these consumers if the amount they paid for low emissions they did not get was

20

greater than the amount they later received for low emissions they did not give. See Order

21

re First MTD (dkt. 5374) at 21.
Plaintiffs have now submitted expert evidence purporting to prove their injuries.

22
23

But the evidence is either irrelevant, unreliable, or both. The Court thus grants

24

Defendants’ motion to exclude the evidence and denies Plaintiffs’ motions to exclude

25

Defendants’ expert evidence. Without any injury Plaintiffs lack standing, which means the

26

Court lacks jurisdiction and must dismiss this case.1

27
28

1

This order amends the Court’s previously filed order (dkt. 7769) to correct two typographical
errors. The orders are substantially the same.

1
2
3

I.

BACKGROUND
The Court has previously described Volkswagen’s fraud:

11

Over the course of six years, Volkswagen sold nearly 500,000
Volkswagen– and Audi-branded TDI “clean diesel” vehicles,
which they marketed as being environmentally friendly, fuel
efficient, and high performing. Consumers were unaware,
however, that Volkswagen had secretly equipped these
vehicles with a defeat device that allowed Volkswagen to
evade United States Environmental Protection Agency
(“EPA”) and California Air Resources Board (“CARB”)
emissions test procedures. Specifically, the defeat device
produces regulation-compliant results when it senses the
vehicle is undergoing testing, but operates a less effective
emissions control system when the vehicle is driven under
normal circumstances. It was only by using the defeat device
that Volkswagen was able to obtain Certificates of Conformity
from EPA and Executive Orders from CARB for its TDI diesel
engine vehicles. In reality, these vehicles emit nitrogen oxides
(“NOx”) at a factor of up to 40 times over the permitted limit.

12

In re: Volkswagen “Clean Diesel” Mktg., Sales Practices, & Prod. Liab. Litig., No. 15-md-

13

02672-CRB (JSC), 2016 WL 6248426, at *1 (N.D. Cal. Oct. 25, 2016).

4
5
6
7
8
9

United States District Court
Northern District of California

10

14

This lawsuit involves a putative class of consumers with whom VW has not settled:

15

those who owned or leased an affected car but who resold the car or exited the lease before

16

September 18, 2015, when the public learned that Volkswagen had been selling diesel cars

17

in the United States since 2009 that were equipped with defeat devices.

18

Volkswagen contended that these consumers lacked standing because they were not

19

injured by Volkswagen’s fraud. See Order re First MTD at 1. The Court rejected many of

20

Plaintiffs’ injury theories, but held that Plaintiffs had adequately pleaded several theories:

21

[1] For all Plaintiffs, it is plausible that they overpaid for the class vehicles when
they paid a premium for a low-emission vehicle and (in the case of former owners)
only received a portion of that premium back through resale because of
depreciation, or (in the case of former lessees) did not recover that premium.

22
23
24
25
26
27
28

[2] Former owners were also plausibly injured when they paid inflated financing
fees for the class vehicles [tied to their overpayment for low emissions], but former
lessees were not plausibly injured by paying lease acquisition and lease termination
fees [with no relationship to any overpayment for low emissions].
[3] If there was no “clean diesel” premium, former owners were not plausibly
injured just by purchasing the class vehicles [because they were made whole when
2

they resold their cars], but former lessees plausibly were [because they did not resell
their cars and were plausibly injured when they paid money to lease vehicles that
they plausibly would not have leased but-for the emissions fraud]—although the
amount of their injury may be difficult to determine.

1
2
3
4

Id. at 21. The Court acknowledged that it “may be challenging for Plaintiffs to prove the
amount that they paid specifically for low emissions. And it is doubtful that Plaintiffs’

5

overpayment would have been equivalent to the entire cost of the package if the package
6
7
8
9

also included features that Plaintiffs received” like strong fuel economy and better driving
performance. Id. at 10–11.2 Although only the “low emissions” premium’s decline in
value “would be a concrete injury,” and proving that injury may be difficult or impossible,
the Court held that it would be “premature” to dismiss Plaintiffs’ Complaint at the pleading

10

stage. Id. at 11.
11
United States District Court
Northern District of California

To streamline adjudication of the question whether Plaintiffs have standing, the
12
13

Court directed the parties to select specific plaintiffs for whom Plaintiffs would “be
required to offer proof of damages before the case proceeds further.” Order re Second

14

MTD (dkt. 6864) at 3. The Court explained that after evidence of damages had been
15

submitted, the Court would “consider whether the evidence is admissible, whether the

16

methodologies offered to prove damages can be used to calculate damages on a class-wide
17

basis, and whether the trier of fact could reasonably conclude that the identified damages
18

are recoverable under the relevant causes of action.” Id. The parties selected four

19

plaintiffs: Angela Matt Architect, Inc. (Angela Matt), a former Audi lessee; Bradley
20

Conner, a former Volkswagen lessee; Ingrid Salgado, a former Volkswagen owner; and
21

Michael Skena, a former Volkswagen owner. See Submission re Proof of Damages (dkt.
22

7673 at 9-11).
23

a.

Plaintiffs’ Submission on Proof of Damages

24

On August 28, 2020, Plaintiffs submitted their Proof of Damages, see Submission
25

re Proof of Damages, and supporting reports, see Gaskin Report (dkt. 7673-4); Stockton
26
27
28

2

Plaintiffs do not dispute that they received certain features in exchange for the overall premium
they paid for diesel vehicles relative to gasoline vehicles.
3

1

Report (dkt. 7673-5); Weir Report (dkt. 7673-6).3
Plaintiffs’ Submission begins with several legal contentions aimed at broadening

2
3

the damages question. They argue that the correct measure of damages is the amount

4

Plaintiffs paid for their cars less the fair market value of those cars at the time of sale, and

5

“need not account for benefits received after the purchase,” that is, the vehicles’ inflated

6

resale or residual value. See Submission re proof of Damages at 12, 14–16. Plaintiffs then

7

argue that if the Court considers any “offset or reduction” to the initial overcharge damage,

8

“it is VW’s burden to prove the amount of the reduction.” Id. at 16–17.
To prove their injuries, Plaintiffs present evidence from their experts, who calculate

9

the “fair market value” of the class vehicles “at the time of sale.” Id. at 19. Plaintiffs’

11
United States District Court
Northern District of California

10

experts employ varying analyses to calculate these damages for the selected plaintiffs.
First, Plaintiffs present a “conjoint analysis” conducted by experts Steve Gaskin and

12
13

Colin Weir. This analysis (unlike others presented by Plaintiffs) attempts to quantify the

14

low emissions premium that Plaintiffs paid. Id. at 23. Mr. Gaskin constructed a survey by

15

identifying eight vehicle attributes, then asking respondents to choose between different

16

configurations of attributes at different prices. Id. at 25; see also Gaskin Report ¶ 14. To

17

reference the low emissions premium, the survey compared option (1): engines with

18

software designed to detect if the vehicle is undergoing emissions tests, that would meet

19

emissions standards during testing but emit levels of NOx up to 35 times the federal legal

20

standard during normal driving, for which the manufacturer “may eventually offer a free

21

update that reduces the level of NOx emissions during normal driving, but the update may

22

somewhat degrade the fuel economy and performance of the vehicle”;4 with option (2):

23

engines that would meet emissions standards during both testing and normal driving. See

24

id. at 26–27. Based on consumers’ responses to these prompts and a regression analysis,

25
26
27
28

Plaintiffs’ experts have also reponded to Volkswagen’s experts’ reports. See Gaskin Reply (dkt.
7673-7); Stockton Reply (dkt. 7673-9); Weir Reply (dkt. 7673-8).
4
This option noted that there would be no physical indications of the software and that the survey
respondents would be “unaware” that their vehicles have it. Submission re Proof of Damages at
26.
4
3

1

Mr. Gaskin estimated the difference in value “due to a change from ‘Diesel engine that

2

meets NOx emission standards both during testing and normal driving’ to ‘Diesel engine

3

that only meets NOx emission standards during testing.’” Id. at 28; see also Gaskin Report

4

¶ 60. Depending on the price of the vehicle, Mr. Gaskin found overpayment percentages

5

varying from 8.5% to 60.5%. See Gaskin Rpt. Exhibit K (dkt. 7707-13). Mr. Gaskin

6

picked the lowest number (i.e., the “most conservative”) “just to be safe.” Gaskin Tr. (dkt.

7

7707-2) 176:11–21. Accordingly, Mr. Gaskin found that value declined by 8.5% in

8

vehicles with the defeat device. Gaskin Report ¶ 63. Using this figure, Mr. Weir then

9

calculated damages for each selected Plaintiff. Id. at 28-29. At this stage, he did not

10

United States District Court
Northern District of California

11
12
13
14

consider the effects of resale or depreciation. See id.
Second, Plaintiffs present various additional analyses from their expert Edward
Stockton, including:
-a “diminished value proxy analysis” that examines the decline in market
prices for class vehicles after the emissions fraud was revealed to calculate
the magnitude of overpayment at the time of initial purchase. Id. at 29–32;

15
16
17
18
19
20
21
22
23
24

-a “TDI Premium” analysis that attempts to measure the premium paid for
the “bundle of attributes” offered in all TDI vehicles by (1) comparing TDI
vehicle prices to gasoline vehicle prices, and, alternatively, (2) using a
regression analysis to “isolate the value that the market places on the TDIs
compared with similar gasoline vehicles.” Id. at 32–35;
-a “retail replacement” analysis that calculates damages based on “the value
of a remedy that would have been available had VW disclosed the defeat
device immediately upon the sale of the Subject Vehicles.” Id. at 35; and
-an “additional indirect overpayment effects” analysis that estimates
additional indirect costs associated with the initial overpayment, including
interest expense and increased taxes. Id. at 37–38.
Third, near the end of their submission, Plaintiffs present a damages theory that

25

accounts for depreciation. Mr. Weir and Mr. Stockton calculated the selected plaintiffs’

26

vehicles’ overall depreciation rates between the time of acquisition and the time of

27

disposal. Id. at 38. Mr. Weir measured depreciation by the percentage of purchase price

28

recovered at resale and multiplied this rate by the initial overpayment as determined by the
5

1

conjoint analysis. Id. at 39-40. Mr. Stockton performed a substantially similar analysis,

2

but also applied it to lessees, and used both an “individual method” (examining individual

3

transaction information to determine vehicle depreciation) and a “market-based method”

4

(using industry information and the vehicles’ mileage). Id. at 41-42.

5

b. Defendants’ Response and Evidence

6

Defendants have responded to Plaintiffs’ submission, see Response to Submission

7

(dkt. 7706), and submitted accompanying expert declarations and deposition transcripts,

8

see Exhibits (dkt. 7707-1–7707-20).

9

Defendants argue that Plaintiffs have the burden of showing they were injured
despite the inflated resale and residual value of their cars, and that Plaintiffs failed to carry

11
United States District Court
Northern District of California

10

this burden because their experts “did not measure any emissions premium or calculate

12

whether and by how much it depreciated.” Response to Submission at 4.

13

Defendants also move to exclude Plaintiffs’ expert evidence, arguing that Plaintiffs’

14

experts’ methodologies are unreliable. Id. at 5. In support, Defendants have submitted

15

expert reports from Professors Peter Rossi and Lorin Hitt. See Rossi Rpt. (dkt. 7707-10);

16

Hitt Rpt. (dkt. 7707-11).

17
18
19
20
21
22
23
24
25
26
27
28

As most relevant here, Rossi opines that Mr. Gaskin and Mr. Weir’s reports are
unreliable for the following paraphrased reasons:
-Plaintiffs’ experts have not shown that low emissions decline in value,
unlike other aspects of a vehicle that are subject to wear.
-Mr. Gaskin’s conjoint analysis is not applicable to used vehicles, so it does
not calculate what Plaintiffs received for low emissions when they resold
their vehicles (or when their lease payments were lower due to increased
residual values).
-A conjoint survey measures only consumer demand and cannot calculate an
“overpayment” because it does not measure supply-side factors that are
necessary to calculate market price. “Willingness to Pay” (WTP) is not
equal to market price.
-The conjoint survey does not isolate a low emissions premium because it
required respondents to account for the vague chance of a manufacturer
6

1

“update” and an unspecified resulting harm to fuel economy and
performance.

2
3
4

-The survey is flawed because (1) the sample includes owners of both
gasoline and diesel-powered vehicles, (2) the sample size is inadequate, and
(3) more than half of respondents were female but the vast majority of diesel
owners are male.

5
6
7
8
9
10

United States District Court
Northern District of California

11
12
13
14
15
16
17
18
19
20

-Neither Mr. Gaskin nor Mr. Weir measured the margin of statistical error in
their damages estimates.
-Mr. Gaskin’s WTP-based overpayment calculation resulted in wildly
varying overpayment percentages, showing unrealistic sensitivity to base
price.
-Mr. Gaskin’s analysis produced irrational results that show its unreliability.
For example, it indicates that certain consumers would be willing to pay over
$9,000 for a navigation system in a $16,000 car and that VW is a pricepremium brand compared to Audi.
-Mr. Gaskin did not include the right vehicle attributes in the survey or
properly determine how those features should be described to consumers.
-The assumption that an overpayment percentage (e.g., 8.5%) can be applied
in a uniform fashion to all years, makes, and models of vehicles is
unsupported by any data or economic argument.
Rossi Rpt. at 4–7, 27–31, 44.
As most relevant here, Professor Hitt opines that Mr. Stockton and Mr. Weir’s
analyses are unreliable for the following paraphrased reasons:
-Mr. Stockton does not attempt to measure any low emissions premium.

21
22
23
24
25
26
27
28

-Plaintiffs’ depreciation analysis assumes without support that a vehicle’s
overall rate of depreciation can be applied to measure the depreciation of the
alleged emissions premium.
-Because empirical evidence shows that diesel vehicles generally depreciate
more slowly than comparable gasoline vehicles, diesel-specific attributes
(such as the level of NOx emissions) depreciate differently, and may not
depreciate at all.
Hitt Rpt. at 6–7, 43, 92–93.
c. Plaintiffs’ Motions to Exclude
7

Plaintiffs have moved to exclude portions of these reports. See Mot. to Exclude

1
2

Rossi (dkt. 7699); Mot. to Exclude Hitt (dkt. 7697). Plaintiffs argue that one of Professor

3

Rossi’s main points—that Mr. Gaskin’s WTP analysis did not calculate a projected market

4

price—is irrelevant. Mot. to Exclude Rossi at 1–2. Plaintiffs argue that because “the only

5

thing that should change in the but-for world is the “harmful act,” keeping the “quantity

6

supplied of the Vehicles” the same, such that WTP reflects market price, is proper. Id. at

7

6. Plaintiffs also argue that Professor Hitt’s depreciation arguments assume that a

8

“premium” can depreciate, even though only assets can depreciate. Mot. to Exclude Hitt at

9

3–9.

10

United States District Court
Northern District of California

11

II.

LEGAL STANDARD
The Court has a duty to examine its own jurisdiction, see, e.g., B.C. v. Plumas

12

Unified Sch. Dist., 192 F.3d 1260, 1264 (9th Cir. 1999), and thus whether Plaintiffs have

13

standing to sue in federal court, see Spokeo, Inc. v. Robins, 136 S. Ct. 1540, 1547 (2016)

14

(explaining that standing is jurisdictional because without it, there is no Article III case or

15

controversy).

16

To have standing, Plaintiffs must establish (1) that they have suffered an injury in

17

fact, (2) that their injury is fairly traceable to Defendants’ conduct, and (3) that their injury

18

will likely be redressed by a favorable decision. See Lujan v. Defenders of Wildlife, 504

19

U.S. 555, 560–61 (1992). To establish the first of these elements, Plaintiffs must

20

demonstrate that they “suffered ‘an invasion of a legally protected interest’ that is

21

‘concrete and particularized’ and ‘actual or imminent, not conjectural or hypothetical.’”

22

Spokeo, 136 S. Ct. at 1548 (quoting Lujan, 504 U.S. at 560). Plaintiffs bear “the burden of

23

proof” to establish that they suffered an injury in fact “with the manner and degree of

24

evidence required at the successive stages of the litigation.” Lujan, 504 U.S. at 561. At

25

the pleading stage, “general factual allegations . . . may suffice.” Maya v. Centex Corp.,

26

658 F.3d 1060, 1068 (9th Cir. 2011). But at summary judgment, Plaintiffs must provide

27

some evidence of “specific facts” to establish injury. Id. (quoting Lucas v. S.C. Coastal

28

Council, 505 U.S. 1003, 1014 n.3 (1992)).
8

Determining whether Plaintiffs have provided some evidence of their injury, may

1
2

require the Court to determine whether Plaintiffs’ evidence is admissible. Under Rule 402

3

of the Federal Rules of Evidence, “[i]rrelevant evidence is not admissible.” Fed. R. Evid.

4

402.5 For potentially relevant expert evidence, trial judges act as “gatekeepers” and

5

exclude unreliable evidence offered by experts. See Daubert v. Merrell Dow Pharm., Inc.,

6

509 U.S. 579, 597 (1993). Under Rule 702 of the Federal Rules of Evidence, expert

7

evidence is admissible only if

8

(a) the expert’s scientific, technical, or other specialized knowledge will help
the trier of fact to understand the evidence or to determine a fact in issue;
(b) the testimony is based on sufficient facts or data;
(c) the testimony is the product of reliable principles and methods; and
(d) the expert has reliably applied the principles and methods to the facts of
the case.

9
10

United States District Court
Northern District of California

11
12

Fed. R. Evid. 702.

13

III.

14

DISCUSSION
Here, the Court must determine whether Plaintiffs have provided any admissible

15

evidence of an injury based on Plaintiffs’ submission, Defendants’ expert reports and

16

motion to exclude, and Plaintiffs’ motions to exclude. In some respects, Plaintiffs

17

disregard (or seek to relitigate) issues decided in the Court’s prior orders—that is, the

18

Court’s rejection of damages theories unrelated to a low emissions premium and that

19

premium’s depreciation. Where Plaintiffs do attempt to prove damages under the Court’s

20

previously approved theories, their analyses are unreliable and thus inadmissible.6

21

a. Which Party Must Prove Injury?

22

To begin, the Court rejects Plaintiffs’ attempt to resurrect previously rejected

23
24
25
26
27
28

Under Rule 401 of the Federal Rules of Evidence, evidence is relevant if “it has any tendency to
make a fact more or less probable than it would be without the evidence . . . and the fact is of
consequence in determining the action.” Fed. R. Evid. 401(a)–(b).
6
One theory did not depend on a low emissions premium: Lessees were plausibly injured even in
the absence of such a premium because they did not resell their cars and paid money to lease
vehicles that they plausibly would not have leased but-for the emissions fraud. But Plaintiffs have
not attempted to quantify this injury, which the Court previously noted would be exceedingly
difficult to prove. See Order re First MTD at 21.
9
5

1

damages theories by shifting the burden to Defendants to prove any “offset” in damages.

2

The burden is on Plaintiffs to show evidence of their injury. See Lujan, 504 U.S. at 561.

3

The Court’s prior order requiring Plaintiffs “to offer proof of damages”—and not requiring

4

Defendants to prove the absence of damages—is consistent with this principle. Order re

5

Second MTD at 3. Plaintiffs argue that when a party raises a mitigation of damages

6

defense, that party has the burden of proof. See Submission re Proof of Damages at 16.

7

But Defendants do not raise a mitigation defense—they challenge whether, given that

8

Plaintiffs sold their vehicles or had their leases terminate before Volkswagen’s fraud

9

became public, Plaintiffs suffered any damages at all.
Thus, Plaintiffs’ argument that they need not calculate the difference between what

11
United States District Court
Northern District of California

10

they paid and what they later received for low emissions fails for the reasons explained in

12

the Court’s prior order. Consider a counterfeit Louis Vuitton purse, sold as authentic for

13

$1,000 and later resold as authentic for $800. Was the first purchaser harmed? Maybe. If

14

a known inauthentic purse would have cost $100 and resold for $80, then the first

15

purchaser would have recouped all but $20, instead of all but $200, upon resale. In that

16

sense, the first purchaser was injured to the tune of $180. If the first purchaser could

17

provide evidence of the authentic and counterfeit purses’ relative depreciation, the first

18

purchaser could prove an injury. Note that the first purchaser’s initial overpayment of

19

$900 is plainly larger than any injury—were the first purchaser awarded $900 after

20

reselling the purse for $800, the first purchaser would get a significant windfall.

21

The present controversy is more complicated than the Louis Vuitton example for

22

several reasons. Authenticity is a single variable that the first purse buyer did not receive

23

and later did not give. But here, Plaintiffs paid a premium for some features they did

24

receive (better fuel economy and driving performance) and another feature they did not

25

(low emissions). This makes the low emissions premium and its depreciation more

26

difficult to isolate and measure. Moreover, it is easier to measure relative depreciation

27

between authentic and counterfeit purses because there are established markets for both (in

28

New York City, for example, a person could buy an authentic purse on Fifth Avenue or a
10

1

knockoff purse on Canal Street). Here, the market for vehicles that are identical but-for

2

the low emissions feature is hypothetical and less easily observable.

3

The difficulty of proving an injury is no reason to reject it at the motion to dismiss

4

stage. See Order re First MTD at 11. But it is also no reason to accept a theory of injury

5

untethered to the harm that Plaintiffs experienced. Therefore, Plaintiffs’ expert evidence is

6

admissible only to the extent it could reliably show Plaintiffs’ damages given their

7

vehicles’ inflated resale or residual values.
b. Plaintiffs’ Evidence

9

The Court excludes Plaintiffs’ conjoint analysis and depreciation analyses as

10

unreliable under Rule 702 of the Federal Rules of Evidence; the Court excludes Plaintiffs’

11
United States District Court
Northern District of California

8

other analyses as irrelevant under Rule 402 of the Federal Rules of Evidence.

12
13

1. Conjoint Analysis
As discussed above, Mr. Gaskin’s Conjoint Analysis measured a demand-price

14

difference between vehicles with engines that meet emission standards only during

15

testing—and which allow the engine to emit levels of NOx up to 35 times the federal

16

standard—and engines that meet NOx standards before and during testing. See

17

Submission re Proof of Damages at 12, 14–16. Such an analysis could conceivably

18

account for the premium that Plaintiffs paid for low emissions.

19

But as conducted, Mr. Gaskin’s analysis could not reliably estimate the market price

20

premium “specifically for low emissions.” Order re First MTD at 10–11. Professor Rossi

21

persuasively critiques every aspect of Mr. Gaskin’s analysis. See Rossi Rpt. (dkt. 7707-

22

10). But several aspects of Mr. Gaskin’s analysis stand out as flawed even to the untrained

23

eye. Each provides an independent reason to exclude the evidence as unreliable.

24

First, Mr. Gaskin does not actually calculate a market price premium; he examines

25

only what consumers say they would be willing to pay for certain vehicles. As Defendants

26

point out, this ignores the “supply” part of the supply/demand curve. See Rossi Rpt. ¶ 18.

27

“[D]istrict courts across the country have excluded choice-based conjoint analyses that fail

28

to accurately account for supply-side considerations.” MacDougall v. Am. Honda Motor
11

1

Co., No. 17-1079 JBG, 2020 WL 5583534, at *6 (C.D. Cal. Sept. 11, 2020) (collecting

2

cases). Calculating a premium by comparing (1) the actual price of vehicles sold with

3

defeat devices to unwitting consumers, and (2) consumers’ willingness-to-pay for vehicles

4

with the defeat device (as opposed to the market price of vehicles with the defeat device) is

5

comparing apples and oranges. And presuming that Defendants would have sold the same

6

number of cars, at the exact price that consumers would have been willing to pay, is not a

7

way to reliably incorporate supply-side considerations. See In Re General Motors LLC

8

Ignition Switch Litig., 407 F. Supp. 3d 212, 239 (S.D.N.Y. Aug 6, 2019) (rejecting a

9

similar argument because it “would mean that a given product is supplied by the
manufacturer in the same quantity no matter what the price is” and because it would be

11
United States District Court
Northern District of California

10

“highly unlikely” that a vehicle manufacturer “would have wanted to sell the same amount

12

of cars at the price implied by [the] but-for analysis”) (citation omitted).
Second, Mr. Gaskin did not isolate a low emissions premium because the

13
14

information provided to respondents required them to consider certain vague effects going

15

beyond low emissions. Mr. Gaskin instructed respondents that, for engines that produced

16

low emissions only during testing, the vehicle manufacturer may “eventually offer a free

17

update that reduces the level of NOx emissions during normal driving, but the update may

18

somewhat degrade the fuel economy and performance of the vehicle.” Gaskin Rpt. ¶ 49.

19

Thus, Mr. Gaskin invited respondents to consider a potential reduction in other

20

performance measures under vague circumstances that might or might not arise. See Rossi

21

Rpt. ¶ 60. It is impossible to know how much this affected responses.7

22

Third, the results of Mr. Gaskin’s analysis indicate that his methodology was not

23

reliable. Mr. Gaskin’s results suggest that certain consumers value a $2,000 navigation

24

system in a $16,000 vehicle at $9,000, and that VW is a premium brand compared to

25
26
27
28

7

The Court is also skeptical that survey respondents could reliably determine how much they
would pay for a vehicle containing a defeat device given certain phrasing in the survey. See
Submission re Proof of Damages at 26. Consumers were told that they would “be unaware” that
their vehicle had the emissions cheating software. Id. This is a confusing and arguably
imponderable prompt, akin to asking a person how much they would pay for a car that displays an
offensive bumper sticker that the person will never notice or learn about.
12

1

Audi.8 Further, Mr. Gaskin calculated radically different “overcharge” percentages

2

depending on the price of the vehicle (ranging from 8.5% to 60.5%). An overpayment

3

ratio might sensibly vary depending on price, see Gaskin Reply ¶ 24, but a range that

4

resembles ‘somewhere between almost nothing and almost everything’ is facially

5

unrealistic. Merely picking the lowest number in that range does not remove doubt.

6

Finally, Mr. Gaskin’s conjoint analysis suffers from other apparent methodological

7

flaws. Conjoint analyses usually use “a pretest to ensure that questions on the final survey

8

are not confusing [or] misleading, and accurately measure respondent preferences.”

9

MacDougall, 2020 WL 5583534, at *7. Here, Mr. Gaskin’s pretest consisted of
“interviews . . . with 20 respondents to ensure that they ‘did not have difficulty with the

11
United States District Court
Northern District of California

10

questions and instructions; that they understood the choice exercise they were asked to

12

perform; that they looked at all or almost all of the features in making their choices; that

13

they did not think the questions were leading or biased; and that they felt the features were

14

evenly presented.’” Submission re Proof of Damages at 25 n.52 (quoting Gaskin Rpt.

15

¶ 35). But these interviews consisted of “free form” conversations and were not

16

standardized. Gaskin Tr. 232:9–22. Thus, any pretesting measured respondents’

17

impressions of the survey in a highly informal manner. And Plaintiffs have not offered a

18

persuasive response to the sampling/representativeness problems identified by Professor

19

Rossi.
Therefore, Plaintiffs’ conjoint analysis is unreliable and inadmissible under Rule

20
21

702 of the Federal Rules of Evidence.
2. Depreciation

22
23

Without a reliably calculated emissions premium, Plaintiffs’ depreciation

24

analysis is useless because it lacks a starting point from which to calculate a decline in

25

value for low emissions. But even accepting an initial 8.5% emissions premium,

26
27
28

Mr. Gaskin’s response is not persuasive. He accounts for such anomalies by taking the most
conservative estimate in his range. See Gaskin Reply ¶ 25. But the point is that Mr. Gaskin’s
starting point—the range—contains facially nonsensical data. Arbitrarily picking the most
conservative figure from an unreliably determined range does not make the result more reliable.
13
8

1

Plaintiffs’ approach to depreciation is not reliable, and this is equally fatal to their attempt

2

to prove injury.
Plaintiffs assume that the value of an emissions premium declines at the same rate

3

as the overall vehicle. That is plainly wrong. Car features and characteristics are likely to

5

affect depreciation to different degrees. Indeed, Mr. Gaskin’s report for Plaintiffs implies

6

that certain vehicle attributes may depreciate more than others. Mr. Gaskin acknowledged

7

that “consumers shopping for used vehicles consider a wide variety of factors that are

8

irrelevant to a new-car purchase (for example, mileage, accident history, and condition).”

9

Gaskin Rpt. ¶ 48 n.32. Unlike characteristics such as these, low emissions might hold a

10

relatively stable value over a vehicle’s lifetime compared to features that naturally wear

11
United States District Court
Northern District of California

4

down. That means the low emissions premium may play less of a role in vehicle

12

depreciation than, say, accumulated mileage. If so, then the consumer who purchased a car

13

for $25,000 while paying $4,000 for low emissions and then resold a car for $20,000 while

14

receiving $4,000 for low emissions was not harmed by Volkswagen’s fraud. The opposite

15

may also be true, but Plaintiffs’ analysis does not account for these possibilities at all.
Plaintiffs take issue with the word “depreciation” with respect to a non-physical

16
17

feature like low emissions. Mot. to Exclude Hitt at 3–9. But here, “depreciation” is

18

simply shorthand for the changing value that consumers might place on low emissions.

19

Plaintiffs must show how this value changes to prove damages, and they do not purport to

20

do so.

21
22

3. Mr. Stockton’s Analyses
Mr. Stockton’s other analyses are not relevant to any of the damages

23

theories approved by the Court. Indeed, Mr. Stockton testified at his deposition that he did

24

not attempt to calculate a low emissions premium. See Stockton Tr. (dkt. 7707-1) at 64:7–

25

15. Given that Mr. Stockton did not tailor his analyses to the pertinent damages theories, it

26

is no surprise that they do not tend to prove damages under those theories.

27
28

For example, Mr. Stockton’s “diminished value proxy analysis” examines the
decline in market prices for class vehicles after the emissions fraud was revealed to
14

1

calculate the magnitude of overpayment at the time of initial purchase. See Submission re

2

Proof of Damages at 29–32. It principally relies on the effects of disclosure, including

3

brand effects, even though Plaintiffs resold their cars before disclosure. The pertinent

4

market value does not rely on the public’s reaction to news that vehicles were fraudulent.

5

Customers still in possession of their fraudulent vehicles at the time of disclosure were

6

harmed by that reaction. Plaintiffs were not.

7

Similarly, Mr. Stockton’s “retail replacement” analysis calculates damages based on

8

“the value of a remedy that would have been available had VW disclosed the defeat device

9

immediately upon the sale of the Subject Vehicles.” Id. at 35. This analysis has nothing to
do with the emissions premium. Mr. Stockton’s “TDI Premium” analysis purports to

11
United States District Court
Northern District of California

10

measure the premium paid for the “bundle of attributes” offered in all TDI vehicles. Id. at

12

32–35. But Plaintiffs were required to submit evidence of “the amount of the premium

13

that was specifically for low emissions,” not the amount that covered the full bundle of

14

additional TDI vehicle features—some of which Plaintiffs received. Order re First MTD at

15

14. Finally, Mr. Stockton’s “additional indirect overpayment effects” analysis estimates

16

additional damages associated with any overcharge at the time of purchase, including

17

interest expenses and increased taxes. Submission re Proof of Damages at 37–38. An

18

analysis like this could be relevant (indeed, it resembles the financing costs theory of

19

liability previously approved by the Court); if Plaintiffs presented evidence of a low

20

emissions premium, there could be indirect but concrete financial harms associated with

21

that premium. But Mr. Stockton’s “indirect overpayment effects” analysis relies on Mr.

22

Stockton’s other overpayment analyses, which do not purport to measure a low emissions

23

premium.9
Thus, Mr. Stockton’s analyses are inadmissible as irrelevant under Rules 401 and

24
25
26
27
28

9

Although Plaintiffs have not raised this issue, the time value of money could give rise to a
concrete injury. Plaintiffs’ injury could comprise not only the difference between what they paid
and what they received for low emissions, but also the interest that they could have earned on the
money they overpaid between the time of purchase and the time of disposal. But this would still
require Plaintiffs to reliably calculate their initial overpayment for low emissions, and they have
not done so.
15