Lost Profits in a Multicomponent World
Lost Profit s in a Multicomponent World
Bernard Chao 0 1
0 This Article is brought to you for free and open access by the Law Journals at Digital Commons @ Boston College Law School. It has been accepted for inclusion in Boston College Law Review by an authorized editor of Digital Commons @ Boston College Law School. For more information , please contact
1 University of Denver Sturm College of Law
Follow this and additional works at: http://lawdigitalcommons.bc.edu/bclr Part of the Commercial Law Commons, Intellectual Property Law Commons, and the Science and Technology Law Commons Recommended Citation
Abstract: Given our adversarial system, it is not surprising that plaintiffs
davance creative damages theories that would help them maximize their
recvoeries. In patent law, one recurring tactic for patenteisesto seek remedies
based on the entire infringing product instead ofthe specific feature covered
by the patent. This distinction can significantly inflate remedies
becausemodern multicomponent products contain thousands, sometimes hundreds of
thousands, of different features. Thus, entire products are ordeorfs magnitude
larger, more complex, and more valuable than individual features.
In recent years, the Supreme Court hsa sensibly rejected attempts to base
patent remedies on entire products in the context of permanent injunctions and
design patents. The Federal Circuit nonetheless continues to allow patentees
to recover all the lost profits associated with an entire infrniging product even
when the patent at issue only covers one aspect of a multicomponent product.
In Mentor Graphics Corp. v. EVE-USA, Inc., the Federal Circuitrecently af
firmed a $36,417,661 award, giving the patentee all of the lost profits caused
by the sales of the defendant’s infringing semiconductor emulator syste.ms
The patent, however, covered only one feature of the defendant’s product. The
court explicitly rejected attempts to apportion profits between those
attributable to the patented feature and other significant factors.
This Article argues that the failure to consider apportionment is wrong on
both the law and policy. From a doctrinal perspective, the Federal Circuit has
misinterpreted Supreme Court precedent, dating back to the nineteenth
century, to arrive at an overly simplistic “but for” test to assess damages. From a
policy perspective, awarding lost profits based on the entire infringing
prdouct—rather than just the feature—compensates the patentee for value she did
not create and deters innovation in technologies that operate with or build
upon other technology (“complementary technology”). Accordingly, thisArticle
argues that it is time to realignthe lost profits doctrine to make it consistent
with other types of patent remedies. Patentees should only be compensated
based on the value of the patent they hold. That means focusing the remedy on
the infringing feature and not the infringing product.
© 2018, Bernard Chao. All rights reserved.
* Associate Professor, University of Denver Sturm College of Law. I would like to thank
Jorge Contreras, Thomas Cotter, Dmitry Karshtedt, Amy Landers, Mark Lemley, Brian Love,
Viva Moffat, Lucas Osborn, Justin Pidot and Melissa Wasserman for their helpful comments on
various drafts of this Article.
In an adversarial legal system such as ours, parties are incentivized to
seek the greatest possible recovery from a legal actionT.herefore, it is not
surprising that plaintiffs advance creative damages theories that would help
them maximize their recoveries. In patent law, one recurring tactic fora-p
tentees is to seek remedies based on the entire infringing product instead of
the specific feature covered by the patent. This distinction can significantly
inflate remedies because modern multicomponent products contain
thousands, sometimes hundreds of thousands, of different features. Entire
products are orders of magnitude larger, more complex,and more valuable than
their individual features. For example, a smartphone is far more than just
the beveled shell that holds its electronics or the software that allows a user
to unlock the phone with a single swipe (two separately patented features on
Apple’s iPhone).1 Numerous different innovations contribute to the success
of multicomponent products, and awarding all of a product’s lost profits
based on one invention would crowd out retunrs for other important
technologies that contribute to the success of these products.
In recent years, the United States Supreme Court has sensibly rejected
attempts to base patent remedies on entire products in the contextosf both
permanent injunctions and design patents.2 The United States Court of
Appeals for the Federal Circuintonetheless continues to allow patentees to
recover all of the lost profits associated with an entire infringing produ,ct
even when the patent at issue only covers one aspect of tphraotduct. In
2017, in Mentor Graphics Corp. v. EVE-USA, Inc., the Federal Circuit
affirmed a $36,417,661 award,giving the patentee all ofthe lost profits
caused by sales of the defendant’s infringing semiconductor emulator ss-y
tems.3 This decision explicitly rejected attempts to apportion profits
between those attributable to the patented feature and other significant factors
unrelated to the patented feature.
This Article argues that the failure to consider apportionment is wrong
on both the law and policy. From a doctrinal perspective, the Federal Circuit
has misinterpreted Supreme Court precedent,dating back to thenineteenth
century, to arrive at an overly simplistic “but for” test to assedssamages.
From a policy perspective, awarding lost profits based on the entire
infringing product compensates the patentee for value that he or she did not create
and deters innovation in complementary technologyi.e(. technology that
operates with other technology to create products). Accordingly, this Article
argues that it is time to realign the lost profits doctrine to make it consistent
with other types of patent remedies. Patentees should only be compensated
based on the value of the patentthat they hold. That means focusing the
remedy on the infringing feature and not the infringing product.
The issue in Mentor Graphics is just the latest battleground in one of
patent law’s most fundamental debates. Some commentators think oaf- p
tents as a kind of private property.4 The most straightforward implication of
this view is that a prevailing patentee should be able to obtain a permanent
injunction as a “matter of cours5e.”Most commentators, however, now
think of patent law as a type of policy lev,eerven seeing it as a kind of
regulation.6 In this vein, both courts and commentators rely on patent law’s
goal of promoting innovation as their guiding principle7. To the extent that
4 See Richard A. Epstein, The Disintegration of Intellectual Property? A Classical Liberal eR
sponse to a Premature Obituary, 62 STAN. L. REV. 455, 46
) (“The basic principles of
property law are alive and well, and they are capable of reasonable extension to all forms of intellectual
property.”); see also Bernard Chao, Causation and Harm in a Multicomponent World, 164U. PA. L.
REV. ONLINE 61, 67–74 (2016) [hereinafter Chao,Causation and Harm], http://scholarship.law.
cc/6RRM-RUCA] (discussing how the Federal Circuit’s view of patents as property contributed to
the weakening of the “causal nexus” test for assessing irreparable harm); F. Scott KieffP,roperty
Rights and Property Rules for Commercializing Invention, s85 MINN. L. REV. 697, 703 (2001)
(noting that “the treatment of patents as property rights is necessary to facilitate investment in the
complex, costly, and risky commercialization activities required to turn nascent inventions into
new goods and services”); Adam Mossoff,Patents as Constitutional Private Property: The
Hsitorical Protection of Patents Under the Takings Clause, 87 B.U. L. REV. 689, 701 (2007) (noting
that courts and Congress have historically classified “patents as property rights”).
5 See Epstein, supra note 4, at 489 (“In the patent context, this basic element of property
theory has long been held ‘to provide for an exclusive right to inventors to make, use and vend their
inventions,’ leading to the creation of a ‘complete monopoly’ for which injunctive reliefe-is r
quired as a matter of course.”(quoting Cont’l Paper Bag Co. v. E. Paper Bag Co., 210 U.S. 405,
6 See Shubha Ghosh, Patents and the Regulatory State: Rethinking the Patent Bargain Meat
phor After Eldred, 19 BERKELEY TECH. L.J. 1315, 1325 (2004) (asserting that patent law should
be viewed as “economic regulation that overlaps with administrative law”); Mark A. Lemley, The
Regulatory Turn in IP, 36 HARV. J.L. & PUB. POL’Y 109, 110 (2013) (discussing how patent law
could be viewed as both similar to property, as well as a “regulatory intervention” that limits
property rights); Ted Sichelman, Purging Patent Lawof “Private Law” Remedie,s 92 TEX. L.
REV. 517, 566 (2014) (proposing that “patent law remedies should be viewed . . . as part of a
larger regulatory mechanism grounded in public law concepts”). One need not, however, accept the
regulatory view of patent alw to believe that innovation is its goal. After all, the Constitution
speaks in terms of “promot[ing] the Progress of Science” when discussing patent rights. U.S.
CONST. art. I, § 8, cl. 8.
7 See, e.g., Bilski v. Kappos, 561 U.S. 593, 649 (2010) (Stevens, J., concurring) (expressing
concerns about “stifl[ing] . . . progress” of “scientific and technological work”when discussing
inventors receive financial rewards, it is simply a byproduct of encouraging
innovation.8 Advocates of the policy lever view worry that giving strong
remedies for patents on a single feature will discourage complementary
innovation; inventions that work with or build on the work of othe9rsF.or
example, two scholars criticized the prior rule of presumptively issuing
permanent injunctions to prevailing patentees based on concerns about
“patent holdup.”10 In 2006, in eBay Inc. v. MercExchange, L.L.C., the Supreme
Court changed that rule, holding that the courts must now weigh
fourequitable factors in determining whether to issue a permanent injunction1.1
Although the result in eBay was clearly at odds with the property view of
aptents, the reasoning in Justice Thomas’s majority opinion did not engage in
this dispute. Rather, it releid on principles from other substantive areas of
the law.12 This has allowed both property advocates and their opponents to
continue their debate in other doctrinal areas of patent l a,wincluding lost
Even though the Supreme Court has resolved theinjunction issue and
reduced the leverage that individual patented features have on entire
products, the Federal Circuit continues to take a fundamentally different view
when calculating lost profits.Given that the Federal Circuit has exclusive
patentable subject matter); Quanta Comput., Inc. v. LG Elecs., Inc., 553 U.S. 617, 626 (2008)
(discussing the role of patents in“promot[ing] the progress of science and useful arts” in the
context of patent exhaustion q(uoting Motion Picture Patents Co. v. Universal Film Mfg. Co., 243
U.S. 502, 511 (1917))); Graham v. John Deere Co., 383 U.S. 1, 9 (1966) (noting that patent law’s
purpose was “to bring forth new knowledge”); WILLIAM M. LANDES & RICHARD A. POSNER, THE
ECONOMIC STRUCTURE OF INTELLECTUAL PROPERTY LAW 294 (2003) (“The standard rationale
of patent law is that it is an efficient method of enabling the benefits of researchand development
to be internalized, thus promoting innovation and technological progress.”).
8 See Motion Picture, 243 U.S. at 511 (“[T]his court has consistently held that the primary
purpose of our patent laws is not the creation of private fortunes for the owners of patents but is
‘to promote the progress of science and useful arts’ . . . .”) quoting U.S. CONST. art. I, § 8)).
9 See, e.g., Mark A. Lemley, The Economics of Improvement in Intellectual Property Law, 75
TEX. L. REV. 989, 1067 (1997) (asserting that if those who wish to improve upon an invention are
forced to pay high prices to license the original invention, an “underproduction of improvements”
will result); Robert P. Merges & Richard R. Nelson,On the Complex Economics of Patent Scope,
90 COLUM. L. REV. 839, 843–44 (1990) (“Without extensively reducing the pioneer’s incentives,
the law should attempt at the margin to favor a competitive environment for improvements, rather
than an environment dominated by the pioneer f.”ir)m; Suzanne Scotchmer, Standing on the
Shoulders of Giants: Cumulative Research and the Patent Law, J. ECON. PERSP., Winter 1991, at
29, 32 (explaining how giving original inventors broad patent protection “can lead to deficient
incentives to develop second generation products”).
10 See Mark A. Lemley & Carl Shapiro, Patent Holdup and Royalty Stacking, 85 TEX. L. REV.
1991, 2008–10 (2007) (explaining how granting an injunction for infringement of single feature
can give the patentee disproportionate bargaining power).
11 eBay, 547 U.S. at 388.
12 See id. at 391–92 (citing the “well-established principles of equity” and the right to exclude
under property law).
appellate jurisdiction over patent law, this inconsistency is particularly
troublesome. In Mentor Graphics, the court relied solely on “but for” analysis
and awarded lost profits based on the entire infringing produ1c3t.If a
patentee cannot obtain an injunctionagainst a single feature because
thenijunction can be used to holdup an entire productr,ecovering all of the lost
profits caused by the infringing product should also not be allowed. In sum,
the Federal Circuit’s approach to lost profits is clearly at tnesion with the
Supreme Court’s decision on injunctions. Moreover, sound policy weighs in
favor of reducing the leverage that patents on individual features,
evenmiportant features, can exert. Otherwise, owners of these patents will be able
to effectively control products that were created by a community of
diffreent inventors. The solution is to apportion lost profits based on the relative
contribution of the patented feature.
Part I of thisArticle reviews the basics of modern patent lost profits
law.14 Relying on the facts ofMentor Graphics, it then explains why the
current test calculates lost profits based on an entire infringing product
without considering what portion of those profits are actually attributable to
the infringing feature.15 In other words, when calculating lost profits, patent
law does not apportion. Part II takes a step back into history and explains
why the modern view of lost profits is actually based on a misinterpretation
of two longstanding Supreme Court cases,Yale Lock Manufacturing Co. v.
Sargent and Aro Manufacturing Co. v. Convertible Top ReplacementCo.16
Part II then returns to the present to critically examine the Federal Circuit’s
futile attempts to reconcile the Supreme Court’s historical apportionment
requirement with theMentor Graphics decision’s refusal to consider the
value that other factors contributed to the infringing product. 17
Part III analyzes the issue of apportionment and lost profits from a
policy perspective.18 It describes both eBay and Samsung Electronics Co. v.
Apple Inc. and explains why the Supreme Court refused to allow injunctions
and design patent damages to be based on an entire product in these cases.
Part III then discusses further policy concernsi,ncluding issues of royalty
stacking, multiple “but for” causes, and basic fairness.19
13 Mentor Graphics, 851 F.3d at 1290.
14 See infra notes 20–59 and accompanying text.
15 See infra notes 42–59 and accompanying text.
16 Aro Mfg. Co. v. Convertible Top Replacement Co., 377 U.S. 476, 507 (1964)Y;ale Lock
Mfg. Co. v. Sargent, 117 U.S. 536, 549–51 (1886); see infra notes 60–125 and accompanying text.
17 Mentor Graphics, 851 F.3d at 1280; see infra notes 103–125 and accompanying text.
18 See infra notes 126–195 and accompanying text.
19 Samsung, 137 S. Ct. at 433;eBay, 547 U.S. at 388; see infra notes 161–195 and
I. MODERN LOST PROFITS DOCTRINE
A. Lost Profits Basics
The statute governing compensation for patent infringement, 35 U.S.C.
§ 284, states: “Upon finding for the claimant the court shall award the
claimant damages adequate to compensate for the infringement, but in no
event less than a reasonable royalty for the use made of the invention by the
infringer .. . .”20 Lost profits are generally sought when the infringer’s
products compete with the patentee’s product2s1. The lost profits remedy
seeks to restore the patent holder to the financial condition that it would
have occupied had there been no infringemen2t2. Reasonable royalties, on
the other hand, are the minimum damages for those patentees who do not
meet the requirements of lost profits.23 It is significantly more difficult to
establish lost profits than reasonable royalties24. Importantly, the recovery
of lost profits does not have a punitive aspect.25 Other types of damages can
be imposed for misconduct.26
Probably the most well-known modern decision on lost profits is
Panduit Corp. v. Stahlin Bros. Fibre Wor k,s decided by theSixth Circuit in
1978.27 Later in this Article, I will discuss this case in more dept2h8.For
now, it is sufficient to understand thta Panduit requires patentees to prove
four elements to recover lost profits. Thepatentee must show: “(1) demand
for the patented product, (2) absence of acceptable noninfringing siu-bst
20 35 U.S.C. § 284 (2012).
21 See, e.g., Mentor Graphics, 851 F.3d at 1280.
22 Panduit Corp. v. Stahlin Bros. Fibre Works, 575 F.2d 1152, 1156 (6th Cir. 1978).
23 Zelin Yang, Damaging Royalties: An Overview of Reasonable Royalty Damages, 29
BERKELEY TECH. L.J. 647, 649 (2014).
24 See Mark A. Lemley, Distinguishing Lost Profits from Reasonable Royaltie,s 51 WM. &
MARY L. REV. 655, 657–58 (2009) (noting the difficulty of establishing lost profits); Yang,supra
note 23 (noting that those who cannot meet the standard for lost profits can get reasonable roly-a
ties). To establish lost profits, patentees must show that they would have been capable of profiting
from the patent under the Panduit factors discussed below. See infra notes 27–31 and
accompanying text. Patentees seeking reasonable royalties are those unable to establish lost profits. Yang,
supra note 23, at 650. The amount of reasonable royalties that such patentees receive is the market
price for the licensing fee that the infringer would have paid the patentee to use the patenItd.. at
25 See Pall Corp. v. Micron Separations, Inc., 66 F.3d 1211, 1223 (Fed. Cir. 199(5st)ating
that “the purpose of compensatory damages is not to punish the infringer, but to make the patentee
26 See Halo Elecs., Inc. v. Pulse Elesc., Inc., 136 S. Ct. 1923, 193
) (describing the
standard for recovering enhanced damages for “egregious” infringement behavior); Octanet-Fi
ness, LLC v. Icon Health & Fitness, Inc., 134 S. Ct. 1749, 1756–57 (2014) (discussing the
recovery of attorney’s fees in exceptional cases).
27 Panduit, 575 F.2d at 1156.
28 See infra notes 97–102 and accompanying text.
tutes, (3) his manufacturing and marketing capability to exploit the demand,
and (4) the amount of the profit he would have made.”29 Each of these
factors is intended to ensure that the patentee only receives damages for profits
that it would have earned “but for” the infringement3.0 For example, under
the first factor, the patentee should not beable to recover lost profits unless
the infringer’s customers would have demanded the patentee’s products in
the absence of infringement. Similarly, under the second factor, lost profits
are unavailable if the infringer’s customers would have chosen a legitimate,
or non-infringing, third party supplier’s products instead of the patentee’s
products. The third factor ensures that the patentee had the ability to supply
any customers that would have purchased its products in the absence of
nifringement. The fourth factor insists that profits be calculated based on the
financial circumstances that would have existed but for the infringement. In
other words, the lost profits calculation must account for how the market
would have looked without the infringer. Forexample, the patentee might
have incurred increased costs to supply more products. O,r it may have
enjoyed supra-competitive profits because the infringer was kept out of the
market. In short, the four so-called Panduit factors use the same kind of
“but for” causation analysis used in tort law.31
B. Mentor Graphics
The Federal Circuit issued its opinion inMentor Graphics in 2017.32
In one respect, it looks like an easy application of thPeanduit factors
because “but for” the defendant Synopsys’ infringement,the plaintiff Mentor
Graphics would have captured Synopsys’ sales. Mentor Graphics and
Snyopsys (the company that acquired VEE-USA), both made emulation and
verification systems. These systems are incredibly complicated and
expensive. They are used totest semiconductor chip designs that contain up to
two billion circuits.33 The profits on a single unit can exceeodne million
Mentor Graphics and Synopsys were embroiled in a large patent war,
with each side alleging that the other party infrindgeits patents3.5 In the
end, only one patent made its way to the jury, a Mentor Graphics pate n36t.
That patent, U.S. Patent No6.,240,376 (“the ‘376 patent”), was entitled
“Method and Apparatus for Gate-Level Simulation of Synthesized Register
Transfer Level Designs with Source-Level Debugging.”37 It covered
technology that allowed an emulator to obtain more information about the rc-i
cuit by inserting “test probes” at certain points in the soft w38arTeh.ese
probes in turn gave the emulator the ability to measur“eintermediate
values.”39 This additional information allowed chip designers to debug chip
designs faster and more accurately4.0 Mentor Graphics alleged that
Synopsys’ ZeBu emulators used this technology and infringed the ‘376 patent.41
In the damages portionof the trial, Mentor Graphics argued that “but
for” the sale of forty-five of Synopsys’ infringing ZeBu emulators,
customers would have purchased Mentor GraphicsV’eloce emulators instead4.2
Accordingly, Mentor Graphics asked for allof the profits that these
potential sales would have generated.43 In response, Synopsys attempted to argue
that any lost profits should be apportioned based on the relative contribution
of the infringing feature4.4 Indeed, focusing only on debugging, the Zebu
emulators had many other important featur e,sincluding: dynamic probes,
static probes, flexible probes, memory access, and triggers.45 The trial court,
however, did not permit Synopsys to offer evidence of the valuethat any of
these other features contributed to the infringing ZeBu emulators.46 At trial,
Mentor Graphics prevailed. The jury found that Synopsys infringed the ‘376
patent and awarded Mentor Graphics $36,417,661 in lost profits.47
35 Mentor Graphics, 851 F.3d at 1280–81.
36 Id. at 1280. The trial court granted summary judgments on the other patents finding that
they were either invalid for various different reasons or that the infringement claims were barred.
As the Federal Circuit noted, the facts in this case alloewd for a
“remarkably simple” application of thePanduit factors.48 The lost profits at
issue concerned sales that Synopsys made to Intel4.9 Mentor Graphics
successfully argued that it would have sold its emulators to Intel if Synopsys’
emulators did not include two features covered by the ‘376 patent.50 For the
purposes of appeal, Synopsys accepted those arguments as true.51 Synopsys
also did not challenge a finding that there were no noin-fringing
alternatives.52 Thus, a straightforward application of thePanduit factors led to an
award encompassing allof the profits that Mentor Graphics would have
made had it sold its emulators to Intel instead of Synopsys.
Rather than challenging the factual findings, Synopsys argued that the
trial court should have instructed the jury to discount any lost profits by
only taking into account the relative contributionthat the patented invention
made to the entire produc5t3. This principle, known as apportionment, is
already part of the reasonable royalty calculation.54 Although there is a
historical basis for also apportioning lost profits, that has not beeanpart of
modern patent law.55 Mentor Graphics did acknowledge this precedent in
name, but it ignored the substance of what apportionmenrtequires.56
Instead, Mentor Graphics said that the patentee was entitled to be “made
whole.”57 Under this view, the fact finder does not need to considewrhich
lost profits should be attributed to the infringing feature andwhich should
be attributed to other factors.
In an intuitive sense, the Federal Circuit is absolutely correct. The only
way to restore Mentor Graphics to the position that it would have held had
Synopsys not sold any infringing productswas to award Mentor Graphics
all of its lost profits without apportionment.
The problem with this kind of “but for” analysis, howeveisr, that it
improperly confuses harm caused by iannfringing product with harm
48 See Mentor Graphics, 851 F.3d at 1286.
51 Id. at 1286–87.
52 Id. at 1287.
54 See Ericsson, Inc. v. D-Link Sys., Inc., 773 F.3d 1201, 1226 (Fed. Cir. 2014)[ W(“]here
multi-component products are involved, the governing rule is that the ultimate combination of
royalty base and royalty rate must reflect the value attributable to the infringing features of the
product, and no more.”); VirnetX, Inc. v. Cisco Sys., Inc., 767 F.3d 1308, 1326 (Fed. Cir. 2014)
(“No matter what the form of the royalty, a patentee must take care to seek only those damages
attributable to the infringing features.”).
55 See infra notes 60–72 and accompanying text.
56 Mentor Graphics, 851 F.3d at 1283 n.3.
57 Id. at 1290.
caused by an infringing feature. Patent remedies should focus on then-i
fringing feature, and the only way to properly determine the damages
asosciated with that harm is to apply apportionment prinMciepnletosr.
Graphics, nonetheless, does not stand alone. Modern lower court decisions
have similarly refused to apply apportionment to lost profits.58 Part II of this
Article explains how this view stems from a mistaken understanding of
Supreme Court precedent dating back to the nineteenth century.59
Even though apportioning lost profits has deep historical roots, modern
cases like Mentor Graphics Corp. v. EVE-USA, Inc., Inc. have generally
rejected that concept.60 One scholar has provided a thorough account of the
history of apportioning lost profit6s1. He explained that “[b]etween 1854
and 1915, the Supreme Court decided more than two dozen significant
patent damages cases and in each, adhered to the rule that a patentee must
satisfy the apportionrequirement to recover for infringeme.n”t62 Many of
these cases dealt with the remedy of restitution6.3 Restitution allowed a
patent holder to capture theinfringer’s profits as opposed to awardingto the
patentee profits it would have made absent any infringement (lost profits).
Restitution was essentially eliminated by a 1946 amendment to the patent
laws.64 This development should not have affected the applicability of
apportionment to lost profits calculations. After all, a few of the earlier die-c
sions concerned lost profits6.5 Moreover, the concept of apportionment is
equally suitable for recovering lost profits as it is to restitution. No matter
how the patent holder is compensated, apportionment acknowledges that
other factors beyond the patent at issue drive value.
Modern courts have nonetheless paid little attention to this precedent.
The explanations for rejecting it differ. One district court wrongly said that
apportionment only applied to restitution.66 Several recent district court
decisions have relied on the absence of modern authority supporting
apportionment.67 In Mentor Graphics, the Federal Circuit temporarily relied on
yet another rationale before revising its view. The original panel’s decision
in Mentor Graphics conceded that apportionment applied to all types of
patent damage recoveries, but argued that thePanduit factors somehow
incorporated apportionment principles6.8 Perhaps understanding that there
was no basis for this conclusion, an en bancopinion (denying rehearing)
relied on a wholly different basis6.9 Rather than endorsing the idea that the
Panduit analysis applied apportionment principles, the en banc opinion said
that meeting the Panduit factors satisfied the entire market value rule,
making apportionment unnecessary.70 Although some commentators have
advocated for apportionment, others have agreed with how courts have recently
viewed the lost profits doctrine.71
Regardless of how they arrived at the test,modern decisions now say
that the sole test for determining lost profits is a simple “but for” analys,is
and they use that framework to reject apportionmen7t.2 “But for” analysis
and apportionment, however, are really two steps in a more comprehensive
approach to lost profits. Ironically, the foundation for the current si-ngle
minded focus on “but for” analysis isthe 1886 Supreme Court decision of
Yale Lock Manufacturing Co. v. Sargent that illustrates the tw-ostep ap
Although the Supreme Court applied “but for” analysis in Yale Lock, it
also recognized that many factors beyond the patented invention causead
patent holder to lose profits, and it apportioned lost profits according7l4y.
That message has been lost, as courts today have focused exclusively on
Yale Lock’s “but for” causation language7.5 The result is that modern lost
profits doctrine fails to recognize the difference between aninfringing
feature and an infringing product, and thus overcompensates patent holders. A
proper appreciation of this Supreme Court decision, however,can help
restore apportionment to its proper role.
A. Yale Lock’s Two-Step Test
The patent at issue in Yale Lock covered the use of a particular kind of
“turning-bolt” for use in a combination lock7.6 The defendant’s sale of
infringing locks forced the patent holder to sell its own products at a lower
price—one dollar less for one lock and two dollarsless for another lock.77
Modern terminology would classify this as a patent holder inlogs profits
due to price erosion.78 The Supreme Court explained that lost profitsmust
be calculated by examining the “difference between [the patent holder’s]
pecuniary condition after the infringement, and what his condition would
have been if the infringement had not occurred.7”9 This is the well-known
“but for” analysis language that later cases would repeatedly quotem.-I
portantly, the “but for” analysis is only the first step inYale Lock’s damages
determination. This step calculates the profits lost due to the sale ofhte
infringing products— here, the infringing combination locks.It does no,t
however, attempt to determine what profits were lost because of the
infringing feature—here, the turning bolt. That determination occurs in the second
step of Yale Lock’s analysis.
Yale Lock did not simply award the patent holder alolf the profits it
lost because of the sale of the infringing products. Instead, the Supreme
Court affirmed the underlying decision and reduced the patent holder’s lost
profits by fifty percent because of “other causes which gave to the
defendant an advantage in selling its locks.8”0 Those other causes included third
party patented technology found in the infringing lock, other
“superiorxeternal attractions” found in the infringing lock, the “shape” ofthe infringing
locks’ cases, and “commercial success” that the defendant was having in
sales.81 Thus, the second step in Yale Lock’s analysis is apportioning the lost
profits between the profits lost due to the patented invention,and those
attributed to other factors.82 Importantly, this second step did not just apply to
minor infringing features. The infringing turning bolt Yianle Lock was
found to be an “essential feature of the infringing lock,s” and the Supreme
Court characterized the infringement as causing the patent holder’s “entire
In sum, in Yale Lock, the Supreme Court applied a twos-tep analysis.
In the first step, it used “but for” analysis to determine allof the lost profits
attributable to the infringing product. Recognizing that theinfringing
feature was only part of theinfringing product, the Court then proceeded to a
second step. The Court apportioned lost profits between the patented feature
and other contributions made by the infringer. Importantly, the Supreme
Court did not justlook at the product’s patented features when applying
apportionment. It also considered other factors that may have contributed to
the infringer’s profits.Yale Lock even used the modern term
“apportionment” when describing the concept.84 Of course, because of the second step,
81 Id. at 549–50. In modern terms, the latter factor could consist ofdefendant’s brand
recognition or superior marketing or manufacturing abilities.
82 In Yale Lock, the split between the invention and other factors was 50/50,and there was no
indication that the numbers were derived from anything other than a crude estimate.
83 Yale Lock, 117 U.S. at 551–52.
84 Id. at 550.
the patent holder only recovered the portion of lost profits attributable to the
infringing feature, differentiating Yale Lock from modern patent law.85
B. The Supreme Court Drops Step Two
The reason that the modern day understanding of Yale Lock stops with
step one can be traced to the cursory analysis of calculating damages that
the Supreme Court provided in 1964, in Aro Manufacturing Co. v.
Convertible Top Replacement Co.86 This mistake is somewhat
understandable,given that the Court was focused on a different issue. Therelevant part in Aro
focused on explaining what damages one responsible party (the
contribuotry infringer) must pay when money has already been paid to the patent
holder on behalf of another responsible party (the direct infringer).87
A more detailed examination ofAro helps explain the
misunderstanding. The plaintiff Convertible Top’s patent covered a“top-structure for
automobile convertibles.”88 The defendant Aro “produce[d] fabric components
designed as replacements for w-oorunt fabric portions of convertible
tops.”89 The primary issue on appeal was the level of knowledge that
Cnovertible Top had to prove to make out a case of contributory infringement
under 35 U.S.C. § 271(c).90 The Supreme Court held that there must be a
showing that Aro not only knew of the patent, but also that the use of the
component would infringe the patent.91
For our purposes, the important discussionin Aro dealt with damages.
Even though neither party briefed or argued damages, the Supreme Court
felt compelled to address what it believed to be a faulty assumptiomnade
by the parties.92 Specifically, the Court went out of its way to point out that
if earlier payments made on behalf of the direct infringers(customers who
purchased Ford automobiles) fully compensated the patentee for tnh-e i
fringement, then Aro would owe nothing for contributing to the customers’
infringement.93 Before reaching that conclusion,however, the Court
dsicussed what full compensation for patent infringement meant. Relying on
the oft quoted passage fromYale Lock, the Supreme Court explained that
patent damages restore a patent holderto the position that they would have
occupied had there been no infringement.94
Of course, this is only the first step inYale Lock’s analysis. Aro does not
mention the second apportionment step. This omission is not surprising given
the context. Aro was explaining that the patent holder could not obtain a
double recovery from different joint tortfeasors.95 There was no reason to go into
the details of how to calculate lost profits. Indeed, the term “lost profits” does
not even appear in the decision. Moreover,Aro certainly did not grapple with
the additional complexity that arises when considering apportionment. Courts
and commentators alike have nonetheless relied on both Aro and Yale Lock to
suggest that calculating lost profits boils down to applying a “but for”
cauastion analysis and nothing more9.6 As a result, the concept of apportionment
has disappeared from modern lost profits doctrine.
C. The Growth of “But For” Analysis
Even though Aro’s discussion of “but for” causation wasdicta, courts
have clearly accepted Aro’s abridged version of Yale Lock as both accurate
and complete. Consider Panduit Corp. v. Stahlin Bros. Fibre Work,s the
Sixth Circuit’s 1978 decision that is widely cited for its four-factor test used
to determine lost profits.97 Panduit relied on both Aro and Yale Lock for the
proposition that damages for patent infringement under 35 U.S.C2. 84§
should be calculated using “but for” analysi9s8. As stated earlier, each of
Panduit’s four factors expresses a different aspect of “but for” analys99is.
Despite quoting Yale Lock, Panduit does not suggest that apportionment is
part of the lost profits calculation. In other wordPs,anduit relies on Yale
Lock’s step one “but for”analysis yet omits its step two apportionment
94 Id. at 507 (“[Patent damages] have been said to constitute ‘the difference between [the
patentee’s] pecuniary condition after the infringement, and what his condition would have been if
the infringement had not occurred.’” (quoting Yale Lock, 117 U.S. at 552)).
95 Id. at 512 (noting that the discussion “is but an application of the rule that full satisfaction
received from one tortfeasor prevents further recovery against another”).
96 See infra notes 97–114.
97 Panduit Corp. v. Stahlin Bros. Fibre Works, 575 F.2d 1152, 1156 (6th Cir. 1978).
98 See id. (“[Patent damages] have been said to constitute ‘the difference between [the
patentee’s] pecuniary condition after the infringement, and what his condition would have been if the
infringement had not occurred’.” (quoting Yale Lock, 117 U.S. at 552) (citing Aro, 377 U.S. at
99 See supra notes 29–31 and accompanying text.
To be fair, the question of apportionment was not at issue iPnanduit.
The patent holder was unable to recover any lost profits because the Sixth
Circuit held that it had failed to satisfy Panduit’s fourth factor.100 There was
insufficient evidence to show the amount of profits the patent holder would
have made but for the defendant’s infringement.101 Because the patent
holder had failed to show that it deserved any lost profits, the court did not need
to discuss whether to apportion those profits or how to do so. The problem,
however, is that most courts accept the Panduit factors as complete
regarding lost sales. Courts regularly instruct juries about the fourPanduit factors
without mentioning apportionment.102
Panduit’s interpretation of Aro and Yale Lock hardly stands alone. In
1995, in Rite-Hite Corp. v. Kelley Co., an en banc panel of the Federal
Circuit applied a “but for” analysis to extend the reach of lost profits
recoveries.103 Quoting Aro, which was itself quoting Yale Lock, the Federal Circuit
held that the patent holder’s lost profits could encompass additional ssale
that it would have made for products that did not use the patented in-ve
tion.104 Interestingly, four of the judges took a contrary view and interpreted
Aro in much the way thisArticle does. Judge Nies’ dissent argued thatAro
was “simply addressing preclude[ing] double recovery” and that it did not
hold that “but for” analysis was the only limitation on damages.105 The
majority opinion avoided addressing this particular point, and subsequent
decisions continue to rely onAro and Yale Lock to focus exclusively on “but
100 Panduit, 575 F.2d at 1156.
101 Id. at 1156–57.
102 See, e.g., FEDERAL CIRCUIT BAR ASSOCIATION, MODEL PATENT JURY INSTRUCTIONS
§§ 6.2–.4 (2016); NORTHERN DISTRICT OF CALIFORNIA, MODEL PATENT JURY INSTRUCTIONS,
§§ 5.3–.5 (2015).
103 Rite-Hite Corp. v. Kelley Co., 56 F.3d 1538, 1546 (Fed. Cir. 1995).
104 Id. at 1545 (“The question to be asked in determining damages is ‘how much had the
aPtent Holder and Licensee suffered by the infringement. And that question [is] primarily: had the
Infringer not infringed, what would the Patentee Holder–Licensee have made?’” (quoting Aro, 377
U.S. at 507)). The court then went on to say:
Recovery for lost sales of a device not covered by the patent in suit is not of course
expressly provided for by thepatent statute. Express language is not required,
however . . . . [D]amages should be awarded “where necessary to afford the plaintiff full
compensation for the infringement.” Thus, to refuse to award reasonably foreseeable
damages necessary to make Rite-Hite whole would be inconsistent with the meaning
of § 284.
Id. at 1546–47 (quoting Gen. Motors Corp. v. Devex Corp., 461 U.S. 648, 654 (1983)).
105 Id. at 1558 (Nies, J., dissentingin part). Judge Nies’ opinion also relied on historical
Supreme Court precedent apportioning lost profits to rebut the view that damages should be
calculated using “but for” analysis. Id. at 1565–66.
The “but for” analysis in patent law is not boundless. TRhiete-Hite
majority noted that courts have used the concepts of “proximate cause” or
“foreseeability” to limit damages10.6 Applying those principles to patent
law, the court said that, “[i]f a particular injury was or should have been
reasonably foreseeable by an infringing competitor .. . that injury is
generally compensable absent a persuasive reason to the contrary1.0”7 Although
the foreseeability limitation is concerned with remote injuries, that
limitation does apply to the lost profits cases at issue here. Losing sales to ann-i
fringing competitor is not a remote injury; it is entirely foreseeable. Thus,
the issue is not whether to compensate the patentee for theseinjuries. The
issue is simply determining the proper amount of compensation. Limitations
based on proximate cause do not help determine that amount.108
Notably, the current critique ofRite-Hite’s “but for” analysis has
miplications for other areas of damages law beyond apportionment.109 It
suggests that several different damages theories are built upon shaky precedent,
including Rite-Hite’s holding that patentees can recover lost profits for sales
of unpatented products.110 It may also cause us to reconsider whether a
patentee should be able to recover lost profits for so called “convoyed sales”
or unpatented items sold with the patented item1.11 Moreover, my critique
made elsewhere buttresses claims that “but for” analysis should take a back
106 Id. at 1546 (majority opinion).
107 Id. Although Rite-Hite specifically addressed foreseeability, it also noted that other toosl
like “logic, common sense, justice, policy and precedent” have been used to limit legal respoin-s
bility. Id. (quoting W. PAGE KEETON ET AL., PROSSER AND KEETON ON THE LAW OF TORTS § 42,
at 279 (5th ed. 1984)).
108 Indeed, tort law has confronted similar issues. Prosser suggests that once causation is
established, and when feasible, apportionment should be used to limit a defendant’s liability to that
part of the harm of which defendant’s conduct has been a cause fainct. WILLIAM L. PROSSER,
HANDBOOK OF THE LAW OF TORTS § 52, at 313–14 (4th ed. 1971).
109 See Sichelman, supra note 6, at 538 (providing a policy-based critique of “make whole”
patent remedies and asserting that “makew-hole remedies may impede efficient levels of
110 See Rite-Hite, 56 F.3d at 1547. Others have criticized this holding on policy grounds.See,
e.g., CHRISTINA BOHANNAN & HERBERT HOVENKAMP, CREATION WITHOUT RESTRAINT:
PROMOTING LIBERTY AND RIVALRY IN INNOVATION 52 (2012) (noting that “when innovations are
cumulative, interconnected, or exist in complementary relationships, a rule giving the entire
surplus to the inventor isnot even conceptually possible”);see also Timothy Holbrook, Boundaries,
Extraterritoriality, and Patent Infringement Damages, 92 NOTRE DAME L. REV. 1745, 179
) (noting that “[i]t may be time to revisit the foreseeability/proximate cause aspect
111 See e.g., Rite-Hite, 56 F.3d at 1550 (stating that lost profits could apply to unpatented
items that form a “functional unit” with the patented item); Paper Converting Mach. Co. v.
Magna-Graphics Corp., 745 F.2d 11, 23 (Fed. Cir. 1984) (allowing lost profits damages for
different pieces of equipment in a rewinder assembly even though the patent only covered a single
machine in that assembly).
seat to comity concerns when thinking about extraterritorial dama1g1e2s.
These topics, however, are beyond the scope of thisArticle and will be left
for another day.
Bringing the discussion back toMentor Graphics, we can see that the
opinion carried forward Rite-Hite’s misunderstanding of precedent. Mentor
Graphics justified calculating lost profits relying solely on “but for”
anaylsis by citing Aro and Yale Lock and saying that “Section 284 damages ‘have
been said to constitute the difference between [the pateeen’st] pecuniary
condition after the infringement, and what his condition would have been if
the infringement had not occurred.’” 113
Just as in Rite-Hite, the Federal Circuit in Mentor Graphics made two
mistakes. It failed to appreciate thatYale Lock applied a two-step approach
to calculating lost profits that includde apportionment. Also, the court did
not understand that the comments regarding “but for” causation in Aro were
an afterthought unrelated to the issuethat the Supreme Court was addres
ing in that caes. This mistake is easy to make toda,y and many modern
commentators have perpetuated this misunderstanding1.14 It is not too late,
however, to correct the error. Either the Federal Circuit or the Supreme
Court can take a closer look at these Supreme Court decisions n,oawnd
restore apportionment principles to lost profits calculations.
D. “But For” ≠ Apportionment
Unlike other modern decisions refusing to use apportionment in lost
profits calculations, the original panel decision in Mentor Graphics
conceded that apportionment applied to all types of damages
calculation,sincluding lost profits1.15 The opinion took the odd positio,nhowever, that “but
for” analysis somehow incorporates apportionment principles.116
Specifically, the court said that, “Panduit’s requirement that patentees prove demand
for the product as a whole and the absence of n-ionnfringing alternatives
ties lost profit damages to specific claim limitations and ensures that
dmaages are commensurate with the value of the patented featusr.e”117 As
explained earlier, however, the Panduit factors simply reflect different specific
aspects of “but for” causation11.8 They do not reduce lost profit awards
based on other factors that may contribute to a product’s success.
Perhaps recognizing that the original apnel’s analysis was flawed, an
en banc session of the Federal Circuit denied rehearing on a wholly
diffreent basis.119 Rather than suggesting that thePanduit factors somehow
incorporated apportionment, the en banc opinion said that meeting
thePanduit factors satisfied the entire market value rulem, aking apportionment
unnecessary.120 The entire market rule is neither apportionment nor a
substitute for it. Rather, the entire market rule says that if the patentee can show
that “the patented feature drives the demand for an entire multi-component
product, a patentee may be awarded damages as a percentage of revenues or
profits attributable to the entire product.”121 Notably, the entire market rule
only says when the “entire product” can be used, but not the appiraotper
“percentage.”122 That latter issue is a distinct question that apportionment
Yet the en banc opinion reasoned that the first twoPanduit factors,
“‘demand for the patented product’ (factor one) and an ‘absence of
acceptable noninfringing substitutes’ (factor two1)2”3 together contemplate
“demand for the patented product as a whole” as well as the particular features
of the claimed invention in particula1r2.4 Relying on these statements, the
opinion concludes that “further apportionment is unnceessary.”125 The text
of these factors belies the Federal Circuit’s reasoning. The two factors only
discuss the infringing product but they fail to mention the specific
118 See supra notes 7–8 and accompanying text.
119 Mentor Graphics II, 870 F.3d at 1300.
120 Id. The entire market value rule allows a patentee to obtain compensation based on the
market price for a multicomponent product when their infringed-upon patent is only for one of the
components. Yang, supra note 23, at 654. This rule puts the patentee in the position it would have
been in without the infringement because without the infringement, the patentee could have made
and sold its own multicomponent product. Id.; see also Mentor Graphics II, 870 F.3d at 1300.
121 See, e.g., LaserDynamics, Inc. v. Quanta Comput., Inc., 694 F.3d 51, 67 (Fed. Cir. 2012)
123 Mentor Graphics II, 870 F.3d at 1300
(quoting Panduit, 575 F.2d at 1156)
ing features. Thus, there is no way for thePanduit test to consider the
particular patented features.
Instead, the first two Panduit factors serve the same function as the
entire market value rule. They serve to identify the universe of profits at issue
—in other words, the correct base to consider.Like the entire market value
rule, however, they do not say what portion of those profits should be
apportioned to the infringing feature. That is supposed to occur when the fact
finder applies Yale Lock’s second step—apportionment. Unfortunately, that
step has disappeared from the Federal Circuit’s lost profits jurisprudence.
III. CALIBRATING DAMAGES
This Article does not endorse the use of apportionment based on
precedent alone. There are several policy reasons that weigh in favor of using
apportionment to calculate lost profits. Part III explains the reasons in favor
of apportionment and criticizes the current method for calculating lost
profits.126 First, Part III explains how the Supreme Court has recently refused to
allow patent holders to recover remedies on entire products in the context of
both permanent injunctions and design patents.127 It then argues that by
failing to focus on theinfringing feature, lost profits law is inconsistent with
how these other remedies value patent right1s2.8 Second, Part III explains
why awarding lost profits on entire infringing products overcompensates
patentees and disincentivizes complementary innovatio1n29. This can be
particularly problematic because the current test allows for multiple “but
for” causes. As a result, an infringer may ultimately be liable for multiple
awards that sweep up all of a products’ profits. Third, Part III addresses the
critics of apportionment by arguing that concerns about making patentees
whole, although understandable, should not be prioritized overproperly
allocating innovation incentives1.30 As part of this argument, Part IIIx-e
plains why the failure to apportion profits results ina very similar problem
to that found in discussion of patent holdup.131 Fourth, Part III explains how
basing lost profits on entire products leads to uneven treatment of similarly
situated companies.132 Finally, Part III rejectsMentor Graphics Corp. v.
126 See infra notes 134–195 and accompanying text.
127 See infra notes 136–152 and accompanying text.
128 See infra notes 153–160 and accompanying text.
129 See infra notes 161–182 and accompanying text.
130 See infra notes 183–193 and accompanying text.
131 See infra notes 187–188 and accompanying text.
132 See infra note 194 and accompanying text.
EVE-USA, Inc.’s conclusion that “but for” analysis somehow incorporates
A. Infringing Feature << Infringing Product
The primary policy reason for using apportionment is to properly
cailbrate lost profits awards to the value of the patented invention. It may seem
self-evident that a patent on part of a product is entitled to a smaller reward
than a patent on the entire productP.atentees have nonetheless repeatedly
sought to obtain remedies based on the entire product when the patent
covered just an aspect of the product.
Allowing patentees a disproportionately large remedy harms
innovation by disincentivizing others from developing complementary technology
that either builds on or works with other technology1.34 To illustrate,
imagine giving the inventor of a fundamental television technology the right to
capture all of the profits from any kind of television, even those that
incorporate later-developed improvements. In other words, assume that any kind
of television would infringe that patent. Other inventors would have
noincentive to develop better televisions because the original patentee would get
all of the profits. As a result, neither high definition television nor playback
features would be invented until the original patent expired. These concerns
should not be overemphasized because lost profits law does not affect
innovators nearly as muchas the more common remedy of reasonable royl-a
ties.135 The relatively infrequent use of lost profits law, however, should not
prevent us from adopting rules that identify precisely what the patentee
Recognizing this problem, the Supreme Court recentlyoverturned two
attempts to obtain remedies based on the entire product in the context of
injunctions and design patent damages. The first such decision weaBs ay
Inc. v. MercExchange, L.L.C.in 2006.136 For many years, courts
automaitcally awarded prevailing patentees a permanent injunction against further
infringement.137 That, however, gave patentees owning patents covering
part of a product the ability to enjoin a company from making and selling
the product as a whole. In some cases, infringers could redesign their
products to exclude the infringing feature, but those costs might actually be
higher than the value of the patented technology1.38 In other cases, the
patent may cover technology essential to making the larger product. Examples
include patents on interfaces or patents essential to making products comply
with different technical standards, or “standards essential patents.”139 These
kinds of patents gave their owners the ability to “houldp” manufacturers
and demand settlements that were far larger than the value the patented
invention actually contributed to the product.140
Although the primary rationale underlyingeBay was restoring
traditional equitable principles to patent law injunctions, several Justices were
also concerned aubt o
holdup.141 eBay requires
courts to apply the same
test they use in other
areas of the law when
determining if ann- i
junction should be
sisued.142 Lower courts
have interpreted eBay’s
four-factor test as ensuring a focus on the infringing feature.143 Specifically,
when assessing the first factor—whether continued infringement will cause
the patentee irreparable harm—the Federal Circuit has held that there must
be a “causal nexus” between theinfringing feature and demand for the
infringing product.144 This rule limits the problem of patent holdup.145
In a suit between smartphone titans in the Northern District of
California which began in 2011,Samsung Electronics Co.v. Apple Inc., a jury
found that Samsung infringed various Apple dgensi patents.146 The jury
awarded Apple $399 million in damag,ersepresenting “the entire profit
Samsung made from its sales of the infringing smartphone1s4.7” Consider
one of the design patents involved (depicteadbove).148 The patent claim
only covers the black, flat front face of the phone. The idea that the owner
of a design patent covering a phone case can recover allof the profits of a
product as complex as a smartphone makes no economic sense. oAnes
scholar put it, “As applied to a modern, multicomponent product it drais-t
cally overcompensates the owners of design patents, and correspondingly
undervalues technical innovation and manufacturing know-how.”149 Indeed,
the huge Samsung jury verdict suggests that the tens of thousands of other
patents that smartphones use have no value.150
The Supreme Court wisely rejected the lower court’s ruling in
Samsung Electronics in 2016, holding that the term “article of manufacture” in
35 U.S.C. §289 could apply to “both a product sold to a consumer and a
component of that product, whether sold separately or no1t5.1” The Court
relied on a statutory construction unique to design patents, nont policy,
and thus its decision does not have a broad impact on other areas of the law.
Samsung Electronics, however, does advance an important policy goal. By
explaining that a design patent may only cover a part of a product, the new
rule effectively requires courts to at least consider applying a form of
apportionment to the recovery of design patent damages.152
By solely using “butfor” analysis, Mentor Graphics accomplished
precisely what botheBay and Samsung Electronics rejected. Mentor
Graphics provided a remedy based on the entire infringing product and not
only the patented feature. In contrast, one of the waythsat eBay reduced
patent holdup was to make it much harder for patent holders to prevent
someone from obtaining an injunction against an entire product when the
patent only covered a component of it.153 Likewise, Samsung Electronics
reduced the likelihood that a design patentowner could recover all of the
profits earned from an infringing product when the patent only covers one
aspect of that product. Courts must now consider whether the article of
manufacture covered by a design patent is simply a component. If it is, the
patent holder will only recover profits attributable to that componenatn,d
not those attributable to the entire product.
In contrast, relying exclusively on “but for” analysis fails to
disntiguish between lost profits attributable to a feature and lost profits
attributable to the entire produ;ctunder “but for” analysi,s they are one and the
same.154 The facts of Mentor Graphics clearly illustrate this point. Mentor
Graphics obtained $36,417,661,which represents allof the profits that
Mentor Graphics would have made but for the sale of Synpoyss’ infringing
products.155 Presumably, the result would have been the same if the
infringing product incorporated technology covered by ten, or evenone hundred,
other patents.156 Indeed, Mentor Graphics did not even lower its damages
demand after it lost four of the five patents it had been assertin1g57. In the
“but for” damages world, prevailing on any one sufficiently important
patent means that a patentee recovers all of the lost profits associated with the
Moreover, Synopsys was not allowed to introduce evidence that there
were “multiple drivers of [customer] demand.”158 That information is
irrelevant if the specific value of theinfringing feature is not used to determine
damages. This is not supposition; it is what the district court judge said:
I am just saying you cannot apportion. When you start saying,
“Here is how much of the purchase price for each sale was really
driven by other features,” then other than on royalty, it doestn’
come up. That’s apportionment.159
Of course, the end result is that the patentee recovers lost profits on the
entire product. This result clearly conflicts with the policy underlying both
eBay and Samsung.160 If left standing, the rule will overcompensate patent
holders and unnecessarily burden companies that make and sell
B. Royalty Stacking
A related critique of relying solely on “but for” lost profits analysis is
the concern about overlapping patent rights. Here, the concern is not just
about the outsized remedy that a patent on a single feature receives, but how
that remedy interacts with all of the other patents involved in
multicomponent products. Modern technology products often use hundreds, if not
hundreds of thousands, of patented inventions16.1 Those patents are often
owned by many different parties. The “royalty stacking” problem occurs
when these different patent owners demand higher aggregate royalties than
they would have had they acted as a group.162 If the combination of patent
licenses and damages awards become excessive, this will likely deter later
innovation.163 That risk is only heightened when a single award amounts to
all of the profits that a patent holder would have made. In Mentor Graphics,
Synopsys raised one variation of this concern when it pointed out that
products may contain many features patented by different companies, and each
of those patented features could be a “but for” cause of a company’s
The Federal Circuit inMentor Graphics concluded that this situation
was unrealistic because “[u]nder Panduit . . . there can only be one recovery
of lost profits for any particular sale1.6”5 The opinion illustrated its point
using a laptop example offered by Synopsys. In the example, customers
demand a number of separate features in a lapto,pincluding a high
resolution screen, a responsive keyboard, a fast wireless network receiver, and an
extended-life battery.166 The opinion reasoned that “[i]f each are patented by
separate companies, and no manufacturer has the right to sell them all, then
no manufacturer could obtain lost profits on such a laptop (none could
satisfy the demand for everything).”167 The patentee can only recover a
reasonable royalty.168 On the other hand, if the only patented component is the
extended life battery, “one patentee can obtain lost profits on the overall
product.”169 Thus, the court in Mentor Graphics reasoned that there can only be
one patent that ever gives rise to lost profits recovery.
The Federal Circuit’s analysis, howeveri,s flawed on several levels.
First, it fails to even discuss the more fundamental concern of royalty
stacking. Second, with respect to the narrower question of multiple “but for”
causes, the opinion fails to appreciate how thePanduit test creates a hypothetical
world with no non-infringing alternatives. This hypothetical world looks
dfiferent when different patents are at issue in a particular lawsuit. The result is
that there may be different “but for” causes depending on what patent is being
asserted. Alternatively, the opinion may actually be adding a new element to
the Panduit test for lost profits that does not exist and makes little sense
be162 Contreras, supra note 161, at 18–19.
163 See Lemley & Shapiro, supra note 138, at 2015–16 (“[T]he combined royalty burden
associated with royalty stacking may make it unprofitable for the downstream firm to conduct the
R&D and incur the other costs necessary to develop the product in question.”).
164 Mentor Graphics, 851 F.3d at 1289.
169 Id. at 1289–90.
cause it assumes knowledgethat real-world companies do not possess. The
following two subsections explain these arguments in further depth.
1. Excessive Payments
This subsection first addresses the larger concern: whether a single
award based on all of a patent holder’s lost profits will force an infringer to
pay excessive fees for allof the patent rights that it must acquire. Keep in
mind that this problem exists even if we acceptMentor Graphics’
proposition that there is only one patented feature thatdrives consumers’
purchasing decisions.170 Consider the situation in which the Federal Circuit
decision is left standing and Synopsys pays Mentor Graphics the $36 million
award in lost profits. If we use Mentor Graphics’ profits per emulator as a
rough estimate of the profits that Synopsys made, the result is that Synopsys
just paid all of its profits for a license to a single patent.
This outcome makes little economic sense because a product’s entire
value is not attributable to one patent. What happens when a third-party
patent holder later approaches Synopsy,s demanding that it license the thir
dparty patent because of past infringement? If Synopsys has already paid all
of its profits to Mentor Graphics, then the payment of any additional
royalties will mean that Synopsys sold its emulators at a loss.171 Even if the other
patented feature is relatively minor, that third-party patentee is still entitled
to a royalty. In other words, its quite possible that a manufacturer must
obtain a license to a critical patenti.(e. if infringed would result in a lost
profits award) in addition to less important patentsi.e(. where lost profits
are unavailable but a payment of a reasonable royalty is still required). If
the infringer must pay all of the patentee’s lost profits for the use of the
critical patent, that willeliminate crowd out the infringe’rs ability to pay for
other licenses and earn any profits. It will also prevent the infringer from
earning a return on the investments it made in technologies that likely laso
contributed to the infringing product’s success.172
171 The profits that defendant Synopsys earned are not necessarily the same as Mentor
Graphics’ lost profits. In multicomponent contexts, however, there is no reason to believe that one
party’s (the infringer or the patentee’s) profits will be systematically higher than the other’s. That
is especially true when there is no evidence that the infringer avoided investing in research and
development by copying. Thus, for simplicity’s sake, we can assume that their per product profit
was roughly the same.
172 See Brian J. Love,Patentee Overcompensation and the Entire Market Value Rule, 60
STAN. L. REV. 263, 264–65, 280 (2007) (noting that the entire market value rule causes a similar
problem for reasonable royalty calculations and asserting that “if a producer of a complex product
has already been compelled to pay a reasonable royalty to a patentee based on the entire value of
This analysis does not even consider all of the non-patent reasons why a
product is profitable, which may include the company’s reputat1i7o3n.For
example, Apple can charge a premium that other lesser known companies
cannot. Larger profits are also obtained by taking greater risks. Companies
can have trade secrets, better employees,and general know-how that all
contribute to the bottom line. A damages award that captures all of the infringer’s
profits assumes that none of these factors contribute to a product’s success.174
2. Multiple “But For” Causes
The Federal Circuit has concluded that there can only be one patent
that is the “but for” cause fo lost profits.This issue, however, is less
important than the royalty stacking problem. That is because the multiple “but
for” problem undoubtedly occurs far less frequently than situations where a
multicomponent product is covered by many less criticalpatents. For these
cases, the problem is that a company will have to make both a hefty lost
profits payment and many presumably smaller reasonable royalty payments.
If there are multiple “but for”causes, the problem is magnified because the
company will make multiple substantial lost profits payments. TheMentor
Graphics decision relied heavily onthe belief that there can be only one
“but for” cause to reject apportionment. As discussed below, that analysis is
Mentor Graphics said that there can only be one “but for” cause
ebcause “there can only be one recovery of lost profits for any particular
sale.”176 On its face, this conclusion seems quite plausible. Recall the laptop
example. If a laptop manufacturer is infringing two separate patents
covering a high-resolution screen and a fas-twireless network receiver, both
patent holders will not be able to prove lost profits at the same time.
Remmeber that, to recover lost profits, each patentee must show that if there had
been no infringement, customerswould have purchased the patentee’s
products instead of the infringer’s productIsf. the owner of the h-igh
resolution screen patent, however, shows that the infringing manufacturer’s
customer would have purchased its laptop, that meanthsat the customer
the complex product, the producer has effectively compensated the patentee for the value
contrbiuted by each and every component of that product”).
173 See Lemley, supra note 24, at 663 (“Most commonly, other patents also contribute to [a]
product .. . [and] know-how, materials, and marketing efforts almost always contribute some
value.”) (citation omitted).
174 This is also inconsistent with theYale Lock analysis discussed above. See supra notes 80–
82 and accompanying text.
175 See infra notes 176–182 and accompanying text.
176 Mentor Graphics, 851 F.3d at 1289.
would not have purchased the laptop from the owner of fa-swt ireless. The
reverse is also true. If the evidence shows that the customer would have
purchased a laptop from the owner of the fas-twireless patent, the evidence
clearly cannot show that the customer would have purchased a laptop from
the high-resolution screen patent holder.
This analysis, however, assumes that the Panduit factors are
attempting to determine what would have happened in the real world had there
been no infringement. That is only partially true. The second Panduit factor
modifies what is otherwise a re-awlorld determination by pretending that
“but for” the infringement, the infringer’s customers would not have chosen
to purchase products from an infringing third party.177 The rationale for this
modification makes sense. It allows the patentee to recover lost profits even
when the infringer’s customers would have chosen a third party’s infringing
products in the absence of thdeefendant’s infringement. An unforeseen
consequence of this rule, however, is that juries in separate lawsuits may
find different “but for” causes. In other words, one jury may find that an
infringer’s customers would have purchased products from patentee A “but
for” the infringement of patent A,while a second jury may find that the
same customer would have purchased products from patentee B, but for the
infringement of patent B.
Consider Mentor Graphics’ laptop example in the context of two
separate lawsuits. Assume that the owner of the hig-hresolution screen paetnt
sues the laptop manufacturer and shows that customers insist on purchasing
laptops with high-resolution screens. This evidence will allow the patentee
to prove that “but for” the infringement of the high-resolution screen patent,
the infringer’s customers would have purchased the patentee’s laptop. The
second Panduit factor prevents the jury from considering the possibility that
customers would have purchased laptops from the other manufacturers
(nicluding the fas-wtireless patent holder) because theyo dnot sell n o-n
infringing alternatives, or they do not have a license to the hig-hresolution
Now assume that the owner of the fa-swtireless patent sues the same
laptop manufacturer in a second and separate lawsuit. The owner of the
fastwireless patent will also be able to present evidence that customers insist on
purchasing laptops that have a fas-twireless connection. This evidence can
be used to prove that “but for” the infringer’s sales, customers would have
purchased laptops from the owner of the f-awstireless patent. Again, the
177 Panduit Corp. v. Stahlin Bros. Fibre Works, 575 F.2d 1152, 115(6th Cir. 1978). The
second factor asks whethernon-infringing alternative products were available to the infringer’s
second Panduit factor prevents the jury from considering evidence that
customers would have purchased laptops from the owner of the high-resolution
screen patent. That is because whenPanduit asks about non-infringing
alternatives, the only relevant patent is the patent being asserted in the
lawsuit. Thus, under the current lost profits test, it is quite possible to find that
when a manufacturer infringes two different patents, it is causing two fd-i
ferent patentees to lose sales. That meansthat a given manufacturer may
end up paying two lost profits awards. If these awards are not apportioned,
the payments would likely drive the manufacturer out of the market.
The Federal Circuit’s ruling in Mentor Graphics could also be
interpreted as trying to avoid the situation described above by creating a new
legal rule that prevents the recognition of more than one “but for” cause.
Mentor Graphics said that when a product consists ofmultiple important
patents owned by different parties, “no manufacturer could obtain lost
its on such a[product] (none could satisfy the demand for everything).”
This may be a new variation of the secondPanduit factor, which requires
patentees to prove the absence of acceptable n o-innfringing substitutes.179
In practice, this factor limits the infringer’s ability to avoid paying lost
profits by arguing that customers would have purcheads a competing product
from a third party. If the third party’s products also infringe the patent at
issue, they are not a legitimate alternative source, and they cannot prevent a
lost profits recovery. In other words, even if true as a factual matter, the alw
will not allow an infringer to say “but for” the infringement, customers
would have chosen a third party’s infringing products instead of the
Mentor Graphics takes this analysis and applies it to the patentee. If
the patentee’s products infringe some third-party patent, they are not a
legitimate source for satisfying consumer demand. In other woMrdesn,tor
Graphics suggests that even if itis factually true that “but for” then- i
fringement, customers would have purchased the patentese’ products, the
law should not permit a patentee tomake that argument if those products
also infringe another’s patent.
From a purely doctrinal perspective, this new requirement has no basis
in the law. It does not appear that any decision exists that requires a patentee
to show that its products do not infringe other patents before it can recover
lost profits. As a practical matter, this requirement would be impossible to
implement. Patent assertions occur sequentially and many manufacturers do
178 Mentor Graphics, 851 F.3d at 1289.
179 See Panduit, 575 F.2d at 1156.
not know about numerous important patents.180 In fact, patent holders have
every reason to lie low.181 Thus, when lawsuits like Mentor Graphics go to
trial, the parties may not be aware of the other important patents in the field.
Moreover, even if they were aware of smoe of them, injecting third party
patents into a patent lawsuit will make trials even more complicated. There
will be disputes over whether these third patents are valid and infringed too.
Adding this additional layer of complexity to what is already amotnhge
most expensive kinds of litigation would be highly undesirable.182
C. Lost Profits Without Apportionment ≈ Injunction
One scholar has argued that it is apportionment, and not “but for”
causation damages, that is unfair because apportionment does not make
patentees whole.183 He points out that if juries were to apportion lost profits, the
patentee would only recover a percentage “of the profit it would have made
on sales that it would have it made but for [the] infringem1e84ntT.”hat
would leave the patentee “worse off than it would have been but for
thenifringement.”185 Although this fairness critique is factually accurate, keep in
mind that the goal of the patent system is not to enrich individual inventors,
but to properly incentivize innovation.186 When products are made up of a
single patented feature, such as some pharmaceutical drugs, weighing
competing interests is fairly straightforwardT.he landscape of modern muil-t
component products, however, is far more complex. To simply resotre
pa180 See JAMES BESSEN & MICHAEL J. MEURER, PATENT FAILURE: HOW JUDGES,
BUREAUCRATS, AND LAWYERS PUT INNOVATORS AT RISK 71–72 (2008) (describing the difficulty of
searching for relevant patents).
181 See Oskar Liivak, When Nominal Is Reasonable: Damages for the Unpracticed Pate,nt56
B.C. L. REV. 1031, 1041 (2015) (noting that the structure of patent law and reasonable royaltey- r
wards incentivizes patenting and waiting to enforce the patent later, rather than innovating); Peter S.
Menell & Michael J. Meurer,Notice Failure and Notice Externalities, 5 J. LEGAL ANALYSIS 1, 8–9
(2013) (explaining that patent holders have an incentive to hide their patents until after a company
has invested heavily in its infringing technology because they are in a more powerfruglaibnaing
position); see also Tun-Jen Chiang, The Reciprocity of Search, 66 VAND. L. REV. 1, 13 (2013)
(noting that “examples of w-eklnlown producers being held up by previously unknown patentees
182 See Paul R. Gugliuzza,Quick Decisions in PatentCases, 106 GEO. L.J. (forthcoming
2018) (manuscript at 1) (describing the high cost of patent litigation).
183 THOMAS F. COTTER, COMPARATIVE PATENT REMEDIES: A LEGAL AND ECONOMIC
ANALYSIS 115 (2013).
186 Motion Picture Patents Co. v. Universal Film Mfg. Co., 243 U.S. 502, 511 (1917) (“[T]his
court has consistently held that the primary purpose of our patent laws is not the creation ofi-pr
vate fortunes for the owners of patents, but is ‘to promote the progress of science and useful art’s
. . . .” (quoting U.S. CONST. art. I, § 8)).
tentees to the placethat they would have occupied prior to infringement
would impede innovation in these industries.
Unlike in the pharmaceutical industry, modern multicomponent
products build on thousands and thousands of other innovations. Thus,t iis
important for patent law to consider how its rules will affect the kind of
complementary innovation that exists in the technology sector. Of course, when
inventors use, or infringe,someone else’s patent, they still need to pay for
the privilege. The remedies, however, should not be so powerful that they
would give the patentee control over products that buiuldpon its patent.
That is particularly true when that patented feature is simply one of many
thousands of different features. Outsized remedies for infringement of such
features is what made issuing automatic injunctions to prevailing patentees
so problematic prior to eBay. 187 It effectively allowed a patent on one
faeture to hold up a multicomponent product.
Although there seems to be a general consensus that patent holdup can
be a serious problem, failing to apportion lost profits presents substantially
the same issue. When the law grants a permanent injunction against further
sales of a product, it effectively gives the patentee control over that product.
The patentee can then use the threat of the injunction to insist that thne- i
fringer pay all, or nearly all, of the profits that it could make on that
product.188 Awarding a patentee all of the patentee’s lost profits associated with a
multicomponent product gives the patentee a remedy of the same magnitude
as an injunction. The difference is primarily one of timing. Injunctions
allow a patentee leverage to negotiate a disproportionate shareof the
infringer’s future profits. Failing to apportion lost profits awards the patentee a
disproportionate share of the infringer’s past profits. Indeed, in some cases,
failing to apportion lost profits results in a stronger remedy than awarding
an injunction. For example, when there is an injunction, the infringer can
choose to redesign the infringing feature if those costs are lower than the
licensing fee that the patentee demands. In effect,those design costs place
an upward bound on the money the patentee can demand with an injunction.
When the patentee is awarded lost profits without apportionmen,t however,
it is too late to try to design around the patent. Therefore, the same upward
bound does not exist.Regardless of the minor differences between
nijunctions and unapportioned lost profits,however, the point is the same. For
187 See supra notes 136–145 and accompanying text (discussing the eBay case).
188 Lemley & Shaprio, supra note 138, at 1994–2002 (detailing how the threat of injunction
provides negotiating leverage). Although the model looks at negotiating leverage prior to
litigation, modifying the chance of prevailing at litigation to 100% illustrates the leverage a patentee
would have with a permanent injunction.
multicomponent products, they are both outsized remedies that deter
In short, patent law needs to properly allocate incentives among the
many different innovators taht contribute to multicomponent product1s8.9
Importantly, most infringement does not result from copying, and we should
not assign moral blameor punish the infringerin these cases. Most n-i
fringement does not result from copying1.90 Moreover, infringers often add
their own contributions. That might take the form of creating unforeseen
variations or simply adding new technology. The patent system needs to
incentivize those that improve on existingtechnology.191 Apportioning lost
profits serves this goal and properly balances incentives for different
Critics of apportionment rely on the idea of making inventors whole—
restoring them to the position that they would have occupied had there been
no infringement. The law, however, should not concern itself with making
inventors whole at the cost of harming innovation19.2 Indeed, the Supreme
Court has repeatedly made it clear that patent law is focused on the public
interest, not individual enrichment1.93 With that framework in mind, iits
clear that patent law’s interest in apportioning lost profits is more
compelling than concerns about restoring individual inventors to the precise
financial status that they would have been in absent infringement.
189 Chao, The Infringement Continuum, supra note 114, at 1398–1400.
190 See Christopher A. Cotropia & Mark A. Lemley, Copying in Patent Law, 87 N.C. L. REV.
1421, 1424 (2009) (finding that copying is quite rare in patent litigation). To the extent that there
is intentional copying, § 284 of the patent lawsprovides courts the discretion to award enhanced
damages for this kind of intentional conductS.ee 35 U.S.C. § 284 (2012); Halo Elesc., Inc. v.
Pulse Elecs., Inc., 136 S. Ct. 1923, 19
191 See Scotchmer, supra note 9, at 30 (“The challenge is to reward early innovators fully for
the technological foundation they provide to later innovators, but to reward lateirnnovators
adequately for their improvements and new products as wel.”l); see also Landers, supra note 134, at
505 (“The question of incentives must encompass a concern for sequential invention in order to
fully conform with the central purpose of the patent system.”).
192 See Sichelman, supra note 6, at 538 (criticizing make-whole remedies as poorly suited for
193 See, e.g., Bilski v. Kappos, 561 U.S. 593, 649 (2010) (Stevens, J., concurring) (expressing
concerns about “stif[ling] the . . . progress” of “‘scientific and technological work’” when
discussing patentable subject matter (quoting Gottschalk v. Benson, 409 U.S. 63, 67 (197);2)Q)uanta
Comp., Inc. v. LG Elecs., Inc., 553 U.S. 617, 626 (2008) (discussing the role of patents in “op-r
mot[ing] the progress of scienceand useful arts” when ruling on the proper scope of patent
exhaustion (quoting U.S. CONST. art. I, § 8)); Graham v. John Deere Co., 383 U.S. 1, 9 (1966)
(asserting that patent law’s role was “to bring forth new knowledge” when deciding the standard for
D. Ordering Effects and Equal Treatment of Licensees
Yet another problem with refusing to apportion lost profits is that it
leads to uneven and inequitable resul,tswith some companies inevitably
paying damages that are effectively apportioned, while others do not. This
problem stems from the fact that companies learn about patents at different
times. If we think about what happens when a thir-dparty patentee licenses
a different patented feature to the original patentee and notto the infringer,
there are some surprisingly uneven consequences depending ownhen the
license is executed. Consider whatwould have happened if the third-party
patent holder in Mentor Graphics had licensed to Mentor Graphics before it
sued Synopsys.194 Mentor Graphics’ per unit profit wouldhave been
smaller, and any lost profits caused by Synopsys’ infringement wouldhave been
correspondingly smaller. In other words, Synopsys would have had to pay a
smaller damages award because Mentor Graphics had taken a license to a
third-party patent before the lawsuit. In effect, this is a kind of real world
apportionment of lost profits. The infringer, Synopsys, pays a smaller
amount in lost profits because some of the profits the patentee would have
made are instead channeled to pay another patentee.
The result is different if Mentor Graphics takes the licenasfeter its
lawsuit with Synopsys. Because Mentor Graphics profits shrink after the
lawsuit, those smaller profits cannot be used to reduce the damages
Synopsys owes. The information was simply unavailable. Tehresult is that the
amount of lost profits damages an infringer like Synopsys pays to a patentee
like Mentor Graphics depends on when the patentee takes its licenses to
other third-party patent holders. The damages are higher if the patentee can
somehow avoid obtaining the licenses it needs until its can recover lost
profits damages on its own patents firsTt.he time when a patent holder
takes a license to other patented features, however, should have no effect on
the damages the patent holder recovers.
This uneven result is caused by the fact thatMentor Graphics awards
the lost profits based on the entire product instead of just those attributable
with the patented feature.195 The same result would not occur if courts
apportioned lost profits. Because Mentor Graphics would only recover lost
profits attributable to the patented invention and not the entire product,
paying royalties on other features would have no effect on Mentor Graph’ics
recovery, regardless of when those payments took place.
This Article presents both doctrinal and policy cases for applying
apportionment to lost profits recovery. From a doctrinal perspective, Supreme
Court precedent does not support an exclusive “but for” test, but rather a
two-step test that includes both “but for” analysis and apportionment. From
a policy perspective, it makes no economic sense to award a patentee allof
the lost profits associated with an entire product when the patent onlyo-c
vers a single feature. Such a rule overcompensates patentees and wrongly
disincentivizes those that would make complementary technologies.
The task of apportionment is not an easy one. The perfect should not,
however, be the enemy of the good. As the Supreme Court said long ago
about apportioning damages:
1 See U.S Patent No. 8 , 046 ,721 (issued Oct. 25 , 2011 ) (swipe to unlock feature;)U .S. Patent No. D618 , 677 (issued June 29, 2010 ) (phone shell). The patent for the la-tdeerveloped slide to unlock feature was found invalid as obvious. See Apple Inc . v. Samsung Elecs. Co., 816 F.3d 788 , 806 (Fed. Cir. 2016 ), reinstated in part, vacated in part, 839 F.3d 1034 ( Fed. Cir . 2016 ).
2 See, e.g., Samsung Elecs . Co. v. Apple Inc., 137 S. Ct . 429 ( 2016 ) (design patents); eBay Inc . v. MercExchange, L.L.C. , 547 U.S. 388 ( 2006 ) (permanent injunctions).
3 851 F .3d 1275 , 1280 (Fed. Cir. 2017 ).
29 Panduit, 575 F.2d at 1156.
30 See id . (stating that a patent owner must prove the four factors in order to receive damages for “the profits on sales he would have made absent the infringement”).
31 See RESTATEMENT (THIRD) OF TORTS § 26 (AM . LAW. INST. 2000 ) (defining factual case of harm as occurring “when the harm would not have occurred absent the conduct”).
32 Mentor Graphics , 851 F.3d at 1275.
33 See MENTOR GRAPHICS , ANNUAL REPORT 4 ( 2016 ), http://s3.mentor.com/investor-relations/ FY16-annual-report .PDF [https://perma.cc/667W-5U55].
34 See Transcript of Record for October 9 , 2014 at 1918, Mentor Graphics Corp. v. EV-EUSA, Inc ., 2015 U.S. Dist. LEXIS 29720 (DO .r. Mar. 11 , 2015 ) ( Nos . 3 : 10 -cv-954 -MO , 3 : 12 -cv-1500 - MO , 3 : 13 -cv-579 -MO) [hereinafter October 9 , 2014 Transcript] (asserting, on behalf of Mentor Graphics, in its closing statement that the profit on the sale of a single unit was $1 , 121 ,000).
37 U.S. Patent No. 6 , 240 , 376 .
38 Mentor Graphics , 851 F.3d at 1281.
40 See id.
41 Id. at 1282.
42 Id. at 1283; October 9 , 2014 Transcript, supra note 34, at 1914 .
43 Mentor Graphics , 851 F.3d at 1283.
45 See Transcript of Record forOctober 6 , 2014 at 1303, Mentor Graphics , 2015 U.S. Dist. LEXIS 29720 ( Nos . 3 : 10 -cv-954 -MO , 3 : 12 -cv-1500 -MO , 3 : 13 -cv-579-MO).
46 Transcript of Record for October 8 , 2014 at 1649, Mentor Graphics , 2015 U.S. Dist. LEXIS 29720 ( Nos . 3 : 10 -cv-954 -MO , 3 : 12 -cv-1500 -MO , 3 : 13 -cv-579-MO); see infra notes 158-159 and accompanying text.
47 Mentor Graphics, 851 F.3d at 1283. Mentor Graphics asked for $47 , 191 , 819 . October 9, 2014 Transcript, supra note 34, at 1920 . II. MISUNDERSTANDING “BUT FOR” PRECEDENT
58 See infra notes 66-72 and accompanying text.
59 See infra notes 60-125 and accompanying text.
60 See supra notes 56-57 and accompanying text.
61 Eric Bensen , Apportioning of Lost Profits for Patent Infringement, LEXISNEXIS EMERGING ISSUES ANALYSIS (Apr . 2017 ), http://www.ericbensen. com/Resources/48.%20Apportionment% 20of%20Lost%20Profit%20for%20Patent%20Infringement .pdf [https://perma.cc/X4BM-HP2F] [hereinafter Bensen , Patent Infringement]; Eric E. Bensen, Apportionment of Lost Profits in Contemporary Patent Damages Cases , 10 VA. J.L. & TECH ., Summer 2005 , at 1, 23 - 45 , http://vjolt. org/wp-content/uploads/2017/Articles/vol10/issue3/v10i3_ a8 - Bensen .pdf [https://perma.cc/8CXJLZBJ] [hereinafter Bensen, Patent Damages Cases].
62 Bensen, Patent Infringement, supra note 61 , at 3 (citing Bensen, Patent Damages Cases, supra note 61, at 29- 42 ).
63 See, e.g., Dowagiac Mfg . Co. v. Minn. Moline Plow Co., 235 U.S. 641 , 648 ( 1915 ) (“As the exclusive right conferred by the patent was property and the infringement was a tortious taking of a part of that property, the normal measure of damages was the value of what was taken .”).
64 See Act of Aug. 1 , 1946 , Pub. L. No. 587 , 60 Stat . 778 , 778 ( 1946 ( )codified as 35 U .S.C. § 70 and subsumed into 35 U .S.C. § 285 ) (stating that “for [patent] infringement the complainant shall be entitled to recover general damages which shall be due compensation for making, using, or selling the invention, not less than a reasonable royalty” ); John M. Golden & Karen E. Sandrik , A Restitution Perspective on Reasonable Royaltie,s 36 REV . LITIG. 335 , 344 ( 2017 ) (explaining how the 1946 statute ended restitution as a remedy for utility patents ).
65 See Bensen , Patent Infringement, supra note 61 , at 3 (explaining how some of the leading apportionment cases involved the recovery of lost profits);see also Garretson v . Clark , 111 U.S. 120 , 121 ( 1884 ) (stating that patentees must apportion both “the defendant's profits and thae- p tentee's damages between the patented feature and the unpatented features”).
66 See W.L. Gore & Assocs. v. Carlisle Corp., 198 U.S.P.Q 353 , 363 - 64 ( D. Del . 1978 ).
67 See Brocade Commc'ns Sys ., Inc. v. A10 Networks , Inc., No. C 1- 03428 PSG , 2013 WL 10601009, at *2 ( N.D. Cal . May 15, 2013 ) (asserting that although the concept of apportionment was “appealing,” the defendant's argument “has not cited any Supreme Court or Federal Circuit case requiring an apportionment when a patentee claims lost profits based onPanduit”). The court further stated, “Nor do any of the cases it does cite arise in the context oPfaanduit-based lost profits claim .” Id.; see also Universal Elecs., Inc. v. Universal Remote Control, Inc., No. SACV 12-00329 AG (JPRx) , 2014 WL 12586737, at *6 ( C.D. Cal . Apr. 21 , 2014 ).
68 Mentor Graphics Corp. v. EVE-USA, Inc ., 851 F.3d 1275 , 1287 - 88 (Fed. Cir. 2017 ).
69 See Mentor Graphics Corp . v. EVE-USA, Inc . (Mentor Graphics II) , 870 F.3d 1298 , 1300 (Fed. Cir. 2017 ).
71 See, e.g., Roger D. Blair & Thomas F. Cotter , Rethinking Patent Damages, TEX. INTELL. PROP. L.J. , Fall 2001 , at 1, 28 ( 2001 ) (suggesting that there is no need to apportion lost profits); Brent Rabowsky , Recovery of Lost Profits on Unpatented Products in Patent Infringement Cases , 70 S. CAL. L. REV . 281 , 295 ( 1996 ) (“[A]pplication of the entire market value rule and the generic 'but for' causation test eliminates the need for apportionment .”).
72 See Bensen , Patent Damages Cases, supra note 61, at 22-23 ( commenting that courts have “reduc[ed] a claim for lost profits to a simple 'but for' test, thereby allowing a patentee to recover its entire lost profit on a lost sale regardless of the contribution of the patent to those profits”).
73 Yale Lock Mfg . Co. v. Sargent, 117 U.S. 536 , 5495 - 1 ( 1886 ) ; see infra notes 86-96 and accompanying text (describing later courts' interpretations of Yale Lock) .
74 Yale Lock , 117 U.S. at 549-51.
75 See, e.g., Aro Mfg . Co. v. Convertible Top Replacement Co., 377 U.S. 476 , 507 ( 1964 ).
76 Yale Lock , 117 U.S. at 542.
77 Id. at 548. Interestingly, although the underlying master's decision found that the infringer's sales had caused the patent holder to lose sales too, no lost profits were awarded for these lost sales . Id. at 549.
78 See McSherry Mfg. Co. v. Dowagiac Mfg . Co., 163 F. 34 , 35 ( 6th Cir . 1908 ) (“The damages claimed were on account of reduction of prices, the right to recover which stands on exactly the same ground as damages on account of lost sale.”s). Although the typical lost profits case deals with lost sales and not price erosion, this distinction does not affect the analysis .
79 Yale Lock , 117 U.S. at 552.
112 Bernard Chao , Patent Imperialism, 109 NW. U. L. REV. ONLINE 77 , 83 - 86 ( 2014 ), https:// scholarlycommons.law.northwestern.edu/cgi/viewcontent.cgi?referer=https://www.google.com/ &httpsredir=1&article=1008&context=nulr_online [https/:/perma.cc/WSR5-XMZ7]; see also Carnegie Mellon Univ. v. Marvell Tech. Grp., Ltd., 807 F.3d 1283 , 1306 (Fed. Cir. 2015 ( r)ejecting “but for” analysis that would allow patentees to capture damages for sales made abroad).
113 Mentor Graphics, 851 F.3d at 1283-84 (quoting Yale Lock , 117 U.S. at 552) (citing Aro, 377 U .S. at 507). The decision went on to rely on Rite-Hite and Panduit . Id. at 1284-85.
114 See, e.g., Bernard Chao , The Infringement Continuum , 35 CARDOZO L. REV. 1359 , 1397 n. 208 ( 2014 ) ; Ned Conley, An Economic Approach to Patent Damages, 15 AIPLA Q .J. 354 , 368 ( 1987 ) (citing articles to either or bothAro and Yale Lock for applying straightforward “but for” analysis); Jerry A . Hausman , Gregory K. Leonard & J. Gregory Sidak , Patent Damages and Real Options: How Judicial Characterization of NoInn-fringing Alternatives Reduced Incentives to Innovate, 22 BERKELEY TECH . L.J. 825 , 833 n. 23 ( 2007 ).
115 Mentor Graphics , 851 F.3d at 1287.
116 See id . at 1288 ( asserting that apportionment “was properly incorporated into the lost profits analysis . . . through the Panduit factors”).
133 See infra note 195 and accompanying text.
134 See Scotchmer , supra note 9 , at 32-33 ( explaining that later innovators also need incentives to build on patented technology);see also Amy L . Landers, Patent Claim Apportionment, Patentee Injury, and Sequential Invention, 19 GEO. MASON L. REV. 471 , 504 - 09 ( 2012 ) (relying on the economics of cumulative invention to make the case for apportionment in calculating reasonable royalties).
135 PWC , 2015 PATENT LITIGATION STUDY: A CHANGE IN PATENTEE FORTUNES 8 ( 2015 ), https://www.pwc.com/us/en/forensic-services/publications/assets/2015-pwc -patent-litigation-study . pdf [https://perma.cc/NGR6-G4TL] (finding that 81% of damages awards between 2005 and 2014 contained a reasonable royalties component while only 31% contained a lost profits component).
136 eBay Inc . v. MercExchange, L.L.C. , 547 U.S. 388 , 392 - 94 ( 2006 ).
137 Id. at 391 (discussing the prior rule granting permanent injunctions on infringement b“-a sent exceptional circumstances”).
138 See Mark A. Lemley & Carl Shapiro , Patent Holdup and Royalty Stacking , 85TEX . L. REV . 1991 , 2002 ( 2007 ) (explaining when settlement values might be closer to redesign costs as opposed to the value of the contribution the patent makes to the overall product ).
139 See Bernard Chao , Horizontal Innovation and Interface Patents , 2016 WIS. L. REV. 287 , 316 (describing, for example, why other companies want access to patents covering things like Apple's interfaces that reach millions of Apple customers ).
140 See Lemley & Shapiro, supra note 138, at 2008- 10 ( describing how injunctions against an entire product can result in patent holdup).
141 eBay, 547 U.S. at 396- 97 (Kennedy, J., concurring) ( joined byStevens, Souter , and Breyer, JJ.) (suggesting that courts can avoid holdup by awarding damages instead of an injunction “[w]hen the patented invention is but a small component” of a larger product).
142 Id. at 391 (majority opinion) . Those factors are: (1) whether the plaintiff has suffered an irreparable harm, (2) whether there is an adequate remedy at law, (3) the balance of hardships on the respective parties, and (4) whether granting an injunction would disservice the public interest . See id.
143 See, e.g., Apple , Inc. v. Samsung Elecs. Co., 678 F.3d 1314 , 1324 (Fed. Cir. 2012 ).
144 Apple Inc . v. SamsungElecs . Co., 735 F.3d 1352 , 1364 (Fed.Cir. 2013 ). Unfortunately, the Federal Circuit subsequently watered down the causal nexus requirement in yet another smartphone decision. Apple Inc . v. Samsung Elecs. Co., 809 F.3d 633 , 641 (Fed. Cir. 2015 ). This new standard is misguided . See Chao, Causation and Harm, supra note 4 , at 74-76.
145 Chao, Causation and Harm, supra note 4 , at 75-76.
146 Samsung Elecs . Co. v. Apple Inc., 137 S. Ct . 429 , 433 ( 2016 ).
147 Id. Although lost profits for infringement of the more common utility patents are focused on what the patent owners lost, the remedy for infringing a design patent allows recovery of the infringer's profits . See 35 U.S.C. § 289 ( 2012 ) (stating the design patent infringer “shall be liable to the [patent] owner to the extent of his total profit, but not less than $250”) .
148 U.S. Patent No. D618 ,677 S (issued June 29, 2010 ).
149 Mark A. Lemley , A Rational System of Design Patent Remedies , 17 STAN. TECH. L. REV. 219 , 221 ( 2013 ) ; see also Peter Lee & Madhavi Sunder , The Law of Look and Feel , 90 S. CAL. L. REV . 529 , 590 ( 2017 ) (positing that awardinagll profits for design patent infringements “may vastly overcompensate [the patentee] and may not reflect the true economic contribution of the design to the product”).
150 See Steve Lohr , Apple-Samsung Case Shows Smartphone as Legal Magne,tN . Y. TIMES (Aug. 25 , 2012 ),https://www.nytimes.com/ 2012 /08/26/technology/apple -samsung-case-showssmartphone-as-lawsuit-magnet.html (noting that “as many as 250,000 patents can be used to claim ownership of some technical or design element in a smartphone”).
151 Samsung, 137 S. Ct . at 436.
152 See Sarah Burstein , The “Article of Manufacture” in 1887 , 32 BERKELEY TECH. L.J. 1 , 76 ( 2017 ) (relying on a historical perspective to suggest thatthe “article of manufacture” in Samsung Electronics was not the entire smartphone but the “bezel and the screen conglomeration”).
153 See Mark A. Lemley & Philip J. Weiser , Should Property or Liability Rules Govern nIformation? , 85 TEX. L. REV. 783 , 795 ( 2007 ) (noting that “when injunctive relief is not well it-a lored to the rights being protected, courts are left with the choice of giving no protection or giving too much protection”).
154 This was one of the primary reasons historical precedent used apportionmenSte .e, e.g., Philp v . Nock , 84 U.S. ( 17 Wall.) 460 , 462 ( 1873 ) (“Where the infringement is confined to a part of the thing sold, the recovery must be limited according .l”y); Seymour v . McCormick , 57 U.S. ( 16 How.) 480 , 489 ( 1853 ) (stating that “one who invents some improvement in the machinery of a mill could not claim that the profits of the whole mill should be the measure of damages for the use of his improvement”).
155 Mentor Graphics Corp. v. EVE-USA, Inc ., 851 F.3d 1275 , 1287 (Fed. Cir. 2017 ).
156 These numbers are quite realistic, as Mentor Graphics apparently had over a hunddreactive patents on emulation technology. Opening Brief and Addendum of Defend-aAntpspellants EVE-USA, Inc . at 48, Mentor Graphics , 851 F.3d 1275 ( Nos . 15 - 1470 , 15 - 1554 , 15 - 1556 ). The documents that supposedly illustrate this point were filed under seal and unavailable .
157 Id. at 49. The documents underlying these statements were also under seal .
158 Transcript of Record of October 8 , 2014 , supra note 46 at 1649.
159 Id. at 1650.
160 This concern is not new and is one of the primary reasons that the Supreme Court adopted apportionment years ago . See Seymour , 57 U.S. at 490-91 ( reversing a jury instruction that l-a lowed a patent holder to recover damages on the entire machine when the patent only covered an improvement) .
161 See Jorge L. Contreras , Technical Standards, Standards-Setting Organizations and Intellectual Property: A Survey of the Literature w(ith an Emphasis on Empirical Approaches), in 2 ANALYTICAL METHODS: RESEARCH HANDBOOKS ON THEECONOMICS OF INTELLECTUAL PROPERTY LAW (forthcoming) (manuscript at 8-9 ), https://dc.law.utah.edu/cgi/viewcontent.cgi?referer=https:// www.google.com/&httpsredir=1&article=1010&context=scholarship [https://perma.cc/MCL4F-T2T] (listing studies that identified thousands to hundreds of thousands of patents that potentially cover various technical standards); see also Chao, The Infringement Continuum , supra note 114 , at 1399; supra note 97 and accompanying text.