MEMORANDUM & ORDER RE: DEFENDANTS MOTION FOR NEW TRIAL OR REMITTITUR
TABLE OF CONTENTS
I. INTRODUCTION..........................................................87
II. BACKGROUND............................................................90
III. TENENBAUM’S CHALLENGE TO THE DAMAGES AWARD ...... CO fr- 1
A Tenenbaum’s Constitutional Challenge to the Jury’s Award must Be Addressed................................................ CD H
B. Tenenbaum’s Due Process Challenge........................... CO CR
1. What standard should the Court employ in evaluating Tenenbaum’s constitutional challenge? CO Ü1
*87 a. Williams........................................................95
b. The Supreme Court’s Punitive Damages Jurisprudence...............95
c. Is the Supreme Court’s recent punitive damages jurisprudence relevant to this case?..........................................100
2. The BMW Guideposts...............................................103
a. The Third BMW Guidepost......................................103
b. The Second BMW Guidepost.....................................Ill
e. The First BMW Guidepost.......................................115
3. What is the maximum constitutionally permissible damages award in this case?........................................................116
IV. MISCELLANEOUS ITEMS................................................118
A. Fair Use..............................................................118
B. Tenenbaum’s Evidentiary Challenge......................................119
V. CONCLUSION............................................................121
I. INTRODUCTION
This copyright case raises the question of whether the Constitution’s Due Process Clause is violated by a jury’s award of $675,000 in statutory damages against an individual who reaped no pecuniary reward from his infringement and whose individual infringing acts caused the plaintiffs minimal harm. I hold that it is.
Joel Tenenbaum (“Tenenbaum”), the defendant in this action, was accused of using file-sharing software to download and distribute thirty copyrighted songs belonging to the plaintiffs. The plaintiffs are a group of the country’s biggest recording companies. 1 Their lawsuit against Tenenbaum is one of thousands that they have brought against file sharers throughout the country. Tenenbaum, like many of the defendants in these suits, was an undergraduate when his file-sharing was detected.
Although the plaintiffs presented evidence that Tenenbaum illegally downloaded and shared thousands of recordings, the trial focused on his infringement of the plaintiffs’ copyrights in thirty songs. As to these songs, Tenenbaum’s liability for infringement was not seriously in question. Since he admitted engaging in conduct that clearly constituted copyright infringement at trial, I directed judgment in the plaintiffs’ favor on this issue. The only questions for the jury were whether Tenenbaum’s infringements were willful and what amount of damages was appropriate.
In Tenenbaum’s case, the plaintiffs chose statutory damages over actual damages as the remedy. See 17 U.S.C. § 504(a), (c)(1). “Statutory damages” are damages specially authorized by Congress that may be obtained even in the absence of evidence of the harm suffered by the plaintiff or the profit reaped by the defendant. Under the relevant statute, the jury’s award could be no less than $750 for each work that Tenenbaum infringed and no more than $30,000 or $150,000, depending on whether the jury concluded that Tenenbaum’s conduct was willful. Id. § 504(c)(l)-(2). The jury did find that Tenenbaum willfully infringed the plaintiffs’ copyrights and imposed damages of $22,500 per song, yielding a total award of $675,000.
While that award fell within the broad range of damages set by Congress, Tenenbaum challenged it as far exceeding any plausible estimate of the harm suffered by *88 the plaintiffs and the benefits he reaped. He filed a motion for new trial or remittitur, raising both common law and constitutional grounds. 2 In addition to the plaintiffs opposing Tenenbaum’s motion, the United States government also intervened and filed a memorandum in support of the constitutionality of 17 U.S.C. § 504(c) as applied in this case. (Electronic Order Granting United States’ Mot. to Intervene, March 25, 2009, Case No. 03-cv-11661-NG); see also 28 U.S.C. § 2403(a) (providing that the Attorney General of the United States must be notified of, and may intervene in, any case in which the constitutionality of a federal statute is questioned); Fed.R.Civ.P. 5.1.
Significantly, the common-law doctrine of remittitur would have enabled this Court to entirely avoid the constitutional challenge, always the better choice. Remittitur permits a court to review a jury’s award to determine if it is “grossly excessive, inordinate, shocking to the conscience of the court, or so high that it would be a denial of justice to permit it to stand.”
Correa v. Hosp. San Francisco,
The plaintiffs in this case, however, made it abundantly clear that they were, to put it mildly, going for broke. They stated in open court that they likely would not accept a remitted award. And at a retrial on the issue of damages, I would again be presented with the very constitutional issues that the remittitur procedure was designed to avoid. I am thus obliged to deal with Tenenbaum’s constitutional challenge.
For many years, businesses complained that punitive damages imposed by juries were out of control, were unpredictable, and imposed crippling financial costs on companies. In a number of cases, the federal courts have sided with these businesses, ruling that excessive punitive damages awards violated the companies’ right to due process of law. These decisions have underscored the fact that the Constitution protects not only criminal defendants from the imposition of “cruel and unusual punishments,” U.S. Const, amend. VIII, but also civil defendants facing arbitrarily high punitive awards.
While this body of law is not entirely clear or consistent, it has both a procedural and substantive component. It prevents the awarding of damages without adequate procedural protections, but it also seeks to define the outer limits of what excessive punishment is. Thus, the Supreme Court has held that punitive damages awarded against BMW were grossly excessive, and therefore unconstitutional, in a lawsuit claiming that the manufacturer failed to disclose that the plaintiffs new luxury car had been repainted prior to sale.
BMW of N. Am., Inc. v. Gore,
*89 To be sure, Tenenbaum’s case is different in several respects from the Court’s punitive damages jurisprudence. Since the jury’s award fell within the range set by Congress, Tenenbaum was arguably on notice of the amount of damages that might be awarded to the plaintiffs. But that fact — notice—does not preclude constitutional review. While the parties disagree as to the content of the review of an award of statutory damages, they agree that some form of constitutional review is appropriate.
In reviewing the jury’s award, I must “accord ‘substantial deference’ to legislative judgments concerning appropriate sanctions for” copyright infringement.
BMW,
But since constitutional rights are at issue, deference must not be slavish and unthinking. This is especially so in this case since there is substantial evidence indicating that Congress did not contemplate that the Copyright Act’s broad statutory damages provision would be applied to college students like Tenenbaum who file-shared without any pecuniary gain.
I must also accord deference to the jury’s verdict. As a general matter, damages are uniquely in the jury’s competence. But unlike the Court, the jurors did not have access to data regarding the amount of statutory damages imposed in other copyright infringement actions. A comparison between the jury’s award in this case and the statutory damages awards in other copyright cases demonstrates that the jury’s award here was a serious outlier. The statutory provision under which the jurors imposed their award also did not offer any meaningful guidance on the question of what amount of damages was appropriate. It merely instructs the fact finder to select an amount within an extraordinarily broad range — which here went from $22,500 to $4,500,000 given Tenenbaum’s willful infringement of thirty works — that it “considers just.” 17 U.S.C. § 504(c)(l)-(2).
Weighing all of these considerations, I conclude that the jury’s award of $675,000 in statutory damages for Tenenbaum’s infringement of thirty copyrighted works is unconstitutionally excessive. This award is far greater than necessary to serve the government’s legitimate interests in compensating copyright owners and deterring infringement. In fact, it bears no meaningful relationship to these objectives. To borrow Chief Judge Michael J. Davis’ characterization of a smaller statutory damages award in an analogous file-sharing case, the award here is simply “unprecedented and oppressive.”
Capitol Records Inc. v. Thomas,
For the reasons I discuss below, I reduce the jury’s award to $2,250 per infringed work, three times the statutory minimum, for a total award of $67,500. Significantly, this amount is more than I might have awarded in my independent judgment. But the task of determining the appropriate damages award in this case fell to the jury, not the Court. I have merely reduced the award to the greatest *90 amount that the Constitution will permit given the facts of this case.
There is no question that this reduced award is still severe, even harsh. It not only adequately compensates the plaintiffs for the relatively minor harm that Tenenbaum caused them; it sends a strong message that those who exploit peer-to-peer networks to unlawfully download and distribute copyrighted works run the risk of incurring substantial damages awards. Tenenbaum’s behavior, after all, was hardly exemplary. The jury found that he not only violated the law, but did so willfully.
Reducing the jury’s $675,000 award, however, also sends another no less important message: The Due Process Clause does not merely protect large corporations, like BMW and State Farm, from grossly excessive punitive awards. It also protects ordinary people like Joel Tenenbaum. 3
II. BACKGROUND
Peer-to-peer networks allow users to share with others digital files stored on their computers.
See A & M Records, Inc. v. Napster, Inc.,
In 1999, Tenenbaum began using the peer-to-peer network Napster to download copyrighted sound recordings from other users. He also made copyrighted songs saved on his computer available to other users through his “shared folder.”
(See
Tr. Tenenbaum Trial Testimony 41:13 to 42:3, 91:16-20, July 30, 2009, Case No. 07-cv-11446-NG, document # 20.) After Napster was forced to shut down for contributing to copyright infringement on a massive scale,
see A & M Records,
Tenenbaum was aware that his conduct was illegal. Before he began using Kazaa, he understood that Napster had closed because it was facilitating copyright infringement. (Tr. Tenenbaum Trial Testimony 42:9 to 43:11.) In addition, a student handbook published by Tenenbaum’s undergraduate institution clearly warned that the sharing of copyrighted works over peer-to-peer networks could subject a student to civil liability, criminal penalties, and academic disciplinary action. (Trial Ex. 26 at 11-12, Ex. G to Pis.’ Opp’n to Def.’s Mot. for New Trial or Remittitur.) He even continued to file-share after the *91 plaintiffs sent him a letter demanding that he cease his infringing activities. (See Tr. Tenenbaum Trial Testimony 10:18 to 11:12, 49:5-7, 72:10-23.)
On August 7, 2007, the plaintiffs in this case — five major recording companies— brought suit against Tenenbaum for infringing their registered copyrights through his online downloading and distribution. Instead of accepting responsibility for his actions, Tenenbaum sought to shift blame to his family members and other visitors of his family’s home by suggesting that they could have used the file-sharing software installed on his computer. (Id. at 17:18 to 21:19.) He admittedly lied in sworn responses to discovery requests. (Id. at 89:7-13, 98:12-15.) He also made several misleading or untruthful statements in his deposition testimony. For example, he suggested that a computer he used to download and distribute songs through Kazaa had been destroyed when in fact it had not. (Id. at 48:2 to 49:18, 73:12-24, 99:18 to 101:9.)
As explained above, Tenenbaum’s liability to the plaintiffs for copyright infringement was never seriously in dispute at trial. In fact, I granted the plaintiffs’ motion for judgment as a matter of law on the issue of infringement after Tenenbaum admitted to downloading and distributing the thirty sound recordings at issue in this case. (Electronic Order, July 31, 2009, Case No. 03-cv-11661-NG.) The only issues for the jury, then, were whether Tenenbaum’s infringing conduct was willful and how much the plaintiffs should be awarded in damages.
The jury’s damages award was governed by 17 U.S.C. § 504. Section 504 provides a copyright owner a choice as to the damages that she may recover from an infringer. The owner may select to recover her actual damages and the infringer’s profits, or she may instead elect to recover statutory damages. 17 U.S.C. § 504(a), (c)(1). For an ordinary case of non-willful infringement, permissible statutory damages range from $750 to $30,000 per infringed work. Id. § 504(c)(1). For a case of willful infringement, the statutory damages range is $750 to $150,000. Id. § 504(c)(2). If the infringer can prove that she “was not aware and had no reason to believe that his or her acts constituted an infringement of copyright,” statutory damages of not less than $200 may be awarded. Id.
The plaintiffs in this case elected to receive statutory damages. As explained above, the jury found that Tenenbaum’s infringements were willful and imposed damages of $22,500 per song, for a total award of $675,000.
III. TENENBAUM’S CHALLENGE TO THE DAMAGES AWARD
A. Tenenbaum’s Constitutional Challenge to the Jury’s Award must Be Addressed
Tenenbaum contends that the jury’s award of $675,000 in statutory damages was grossly excessive and thus violated the Due Process Clause. He suggests, however, that I can avoid reaching the question of the award’s constitutionality in a number of ways. First, I could hold that section 504 does not permit the plaintiffs to receive statutory damages because they have not offered evidence that they suffered more than nominal actual damages. Second, I could order a new trial based on alleged errors in my jury instructions. Third, I could reduce the award under the common law doctrine of remittitur. 4
*92
Generally, courts prefer to avoid confronting constitutional questions when they can reasonably rest their holdings on other grounds.
See, e.g., Edward J. DeBartolo Corp. v. Fla. Gulf Coast Bldg. & Constr. Trades Council,
First, his proffered interpretation of section 504 is implausible. Section 504(c)(1) clearly provides that a copyright owner suing for infringement “may elect, at any time before final judgment is rendered,” to recover statutory damages instead of actual damages and the infringer’s profits. 17 U.S.C. § 504(c)(1). The statute does not contain any provision requiring the copyright owner to prove that she suffered more than nominal damages before she may make this election. Tenenbaum does not cite any evidence from section 504’s legislative history or any case law that supports his interpretation of the statute. Indeed, every authority confirms what the language of section 504 clearly indicates— statutory damages may be elected even if the plaintiff cannot, or chooses not to, prove that she incurred more than nominal damages.
See, e.g., L.A. News Serv. v. Reuters Television Int’l, Ltd.,
Tenenbaum’s challenge to my jury instructions also fails. He argues that I should not have instructed the jury in the language of the statute, specifically that its damages award had to fall within the range of $750 to $150,000 per infringed work. Instead, he contends that I should merely have instructed the jury to return whatever award it considered “just” without mentioning the statutory minimum and maximum. If the jury’s award then fell outside of the permissible statutory range, I could have adjusted the wayward award to bring it within the bounds set by Congress. My instructions, however, correctly articulated the statutory damages ranges authorized by Congress and did so in a way that was neither confusing nor misleading.
See Davet v. Maccarone,
Finally, I cannot easily avoid Tenenbaum’s constitutional challenge through the remittitur procedure. Remittitur is a common law doctrine that permits a court to reduce an award by a jury that is “grossly excessive, inordinate, shocking to the conscience of the court, or so high that it would be a denial of justice to permit it to stand.”
Correa,
Remittitur, however, requires the plaintiffs’ cooperation. In deference to a civil litigant’s Seventh Amendment right to trial by jury, a court employing the remittitur procedure must offer the plaintiff the option of rejecting the reduced award and instead proceeding to a new trial on the issue of damages.
See
11 Charles Alan Wright, Arthur R. Miller
&
Mary Kay Kane,
Federal Practice and Procedure
§§ 2807, 2815 (2d ed. 1995 & Supp.2010). In contrast, when a court concludes that a jury’s award is
unconstitutionally
excessive, it can simply reduce the excessive award without giving the plaintiff the option of a new trial.
See Bisbal-Ramos v. City of Mayaguez,
The plaintiffs in this case have made it clear that they almost certainly would not accept a remitted award and would instead opt for a new trial. In an analogous file-sharing case in the District of Minnesota, Capitol Records Inc. v. Thomas-Rasset, the recording-company plaintiffs — four of whom are also plaintiffs in this case— rejected a remitted damages award of $2,250 per infringed work. 7 Notice of Pis.’ Decision Re: Remittitur, Capitol Records, Inc. v. Thomas-Rasset, No. 06-cv-1497-MJD-RLE (D.Minn. Feb. 8, 2010). At the hearing on Tenenbaum’s motion for new trial or remittitur, I specifically asked the plaintiffs’ counsel whether they would also reject remittitur in this case. Their attorney answered that “in all likelihood” they would. (Hearing Tr. 4-5, Feb. 23, 2010, Case No. 07-cv-11446-NG, document #42.)
Thus, it appears that I cannot avoid a new trial on the issue of damages through the remittitur procedure. And at the retrial of damages, I would be forced to confront the very constitutional question that the remittitur procedure was intended to avoid. In particular, I would have to decide whether to limit the range within which the jury could award damages in order to ensure that the jury’s award was not constitutionally out-of-bounds. I would also have to consider Tenenbaum’s objections to the constitutionality of any award that the new jury returned.
Since Tenenbaum’s constitutional challenge appears unavoidable in light of the plaintiffs’ stated reluctance to accept a reduced damages award, I will not enter an order of remittitur. Instead, I will proceed to consider whether the jury’s award violated the Fifth Amendment’s Due Process Clause. 8
*95 B. Tenenbaum’s Due Process Challenge
1. What standard should the Court employ in evaluating Tenenbaum’s constitutional challenge?
a. Williams
Tenenbaum, the plaintiffs, and the U.S. government all agree that the jury’s statutory damages award is subject to some form of review under the Due Process Clause. They simply disagree as to the standard that I should use in evaluating whether the jury’s award is unconstitutionally excessive. The Supreme Court case most directly on point — and the only one that the plaintiffs and the government concede applies to this case — is
St. Louis, I.M. & S. Ry. Co. v. Williams,
In
Williams,
the Supreme Court squarely considered the issue of whether a jury’s award within a statutorily prescribed range violated the Due Process Clause. The plaintiffs in the case, two sisters, sued a railroad that charged them 66 cents more than the statutorily prescribed fare.
Id.
at 64,
In rejecting this claim and upholding the constitutionality of the Arkansas court’s awards, the Supreme. Court noted that the awards’ validity should not be tested merely by comparing the small amount of the overcharges with the magnitude of the judgments obtained by the sisters.
Id.
at 67,
b. The Supreme Court’s Punitive Damages Jurisprudence
Although Williams upheld the constitutionality of the Arkansas jury’s awards, it recognized the possibility that civil damages may in some instances be so excessive as to violate the Constitution. Over the past two decades, the Supreme Court has built on this insight by constructing a *96 rather elaborate doctrinal framework for testing the constitutionality of punitive damages awards.
The Court’s recent punitive damages jurisprudence, which I survey in detail below, is animated by the basic premise that “[t]he touchstone of due process is protection of the individual against arbitrary action of government.”
Browning-Ferris Indus. of Vt., Inc. v. Kelco Disposal, Inc.,
There is no question that these standards have both substantive and procedural components. In other words, while the Supreme Court requires courts imposing punitive damages to afford defendants certain procedural protections, procedural regularity is not alone sufficient for a punitive damages award to survive scrutiny under the Due Process Clause. Instead, the amount of the award produced by proper procedures must also not be “ ‘grossly excessive’ in relation to [the] legitimate punitive damages objectives” of deterring and punishing misconduct.
BMW,
In
Browning-Ferris Industries of Vermont, Inc. v. Kelco Disposal, Inc.,
In
Pacific Mutual Life Insurance Co. v. Haslip,
TXO Production Corp. v. Alliance Resources Corp.,
Honda Motor Co. v. Oberg,
In
BMW of North America, Inc. v. Gore,
The Court began its inquiry into the constitutionality of the jury’s award using the language of substantive due process review. The Court noted that “[pjunitive damages may properly be imposed to further a State’s legitimate interests in pun
*98
ishing unlawful conduct and deterring its repetition.”
Id.
at 568,
The Court’s opinion, however, then took a turn for the procedural. In the introduction to the majority’s discussion of the three famous BMW guideposts, the Court stated:
Elementary notions of fairness enshrined in our constitutional jurisprudence dictate that a person receive fair notice not only of the conduct that will subject him to punishment, but also of the severity of the penalty that a State may impose. Three guideposts, each of which indicates that BMW did not receive adequate notice of the magnitude of the sanction that Alabama might impose ..., lead us to the conclusion that the $2 million award against BMW is grossly excessive....
Id.
at 574-75,
The guideposts, however, seem to contemplate a highly substantive review of a jury’s punitive damages award. They require a court reviewing the constitutionality of a jury’s punitive damages award to consider “(1) the degree of reprehensibility of the defendant’s misconduct; (2) the disparity between the actual or potential harm suffered by the plaintiff and the punitive damages award; and (3) the difference between the punitive damages awarded by the jury and the civil penalties authorized or imposed in comparable cases.”
State Farm,
In reviewing the reprehensibility of the defendant’s conduct, a court should consider whether:
the harm caused was physical as opposed to economic; the tortious conduct evinced an indifference to or a reckless disregard of the health or safety of others; the target of the conduct had financial vulnerability; the conduct involved repeated actions or was an isolated incident; and the harm was the result of intentional malice, trickery, or deceit, or mere accident.
State Farm,
The second guidepost’s ratio analysis requires a court to “consider whether punitive damages bear a reasonable relationship to the harm that the defendant’s conduct caused or is likely to have caused.”
Mendez-Matos v. Municipality of Guaynabo,
The third guidepost instructs a court to compare the punitive damages award to civil penalties authorized or imposed for similar misconduct.
State Farm,
As noted above, these guideposts — although introduced with rhetoric regarding the Court’s procedural concern about “fair notice” — have a significant substantive bite to them. This tension in the language used by the Court in its punitive damages case law is of more than mere academic interest. The distinction between substantive and procedural due process is an important component of the plaintiffs’ and the U.S. government’s argument that the BMW guideposts do not apply to Tenenbaum’s case. If the Court’s major concern in BMW was ensuring that defendants have notice of the civil penalties that may be imposed upon them, BMW’s relevance to the case at bar may be minimal. Unlike in BMW, where the jury’s discretion to award punitive damages was not capped by any statutory maximum, the jury’s award in this case had to fall within the range of $750 to $150,000 per infringed work. Although I have doubts whether this extraordinarily broad statutory range afforded Tenenbaum fair notice of the liability he might face for file-sharing, see infra note 13, it is indisputable that section 504(c) clearly set forth the minimum and maximum statutory damages available for each of his acts of infringement.
Cases decided after
BMW,
however, have reaffirmed that a court’s review of a jury’s punitive award under the Due Process Clause has a significant substantive component.
Cooper Industries, Inc. v. Leatherman Tool Group, Inc.,
State Farm Mutual Automobile Insurance Co. v. Campbell,
The Campbells sued State Farm for its bad faith failure to settle for an amount within the policy limit, and during the damages phase of the trial, they introduced evidence that State Farm’s conduct was part of a broader, nationwide policy to maximize profits by capping payouts on claims.
State Farm,
Significantly, the Supreme Court began its review of the constitutionality of the $145 million punitive damages award by noting that “there are procedural and substantive constitutional limitations” on such awards.
Id.
at 416,
Finally, in
Philip Morris USA v. Williams,
The U.S. Supreme Court vacated the Oregon Supreme Court’s judgment and remanded for reconsideration of the propriety of a jury instruction that Philip Morris offered at trial.
Id.
at 357-58,
c. Is the Supreme Court’s recent punitive damages jurisprudence relevant to this case?
The plaintiffs and the government argue that the Supreme Court’s recent punitive *101 damages jurisprudence does not apply to statutory damages. Instead, they contend that the only standard applicable to this case is the one articulated in Williams. There is a split of authority on this issue, 10 but as described below, the damages award in this case fails under either test.
While I conclude that the due process principles articulated in the Supreme Court’s recent punitive damages case law are relevant to Tenenbaum’s case, the differences between the two approaches are, in practice, minimal. At their root, the standards articulated in
Williams, BMW,
and
State Farm
all aim at providing defendants with some protection against arbitrary government action in the form of damages awards that are grossly excessive
*102
in relation to the objectives that the awards are designed to achieve. Indeed, early twentieth century cases such as
Williams
were the seedlings from which the Supreme Court’s recent punitive damages jurisprudence sprouted.
Browning-Ferris,
the case that rejected a challenge to a punitive damages award under the Excessive Fines Clause, cited
Williams
as an example of a prior opinion in which the Court had expressed “the view that the Due Process Clause places outer limits on the size of a civil damages award pursuant to a statutory scheme.”
Browning-Ferris,
Furthermore,
BMW
and
State Farm
are not irrelevant in a case involving statutory damages merely because the defendant arguably has “fair notice” of the amount of damages that might be imposed on him. As noted above, the Supreme Court has recognized that its punitive damages jurisprudence has both procedural and substantive components.
State Farm,
Lower courts have recognized as much by applying the
BMW
guideposts to punitive damages awards subject to statutory caps. For example, the First Circuit in
Romano v. U-Haul International,
Even the rigorous
BMW
guideposts, however, suggest that a district court judge should afford “substantial deference” to a jury’s award of statutory damages within the range set by Congress.
BMW,
In addition, when applying
BMW’s
second guidepost, which looks at the ratio of punitive to compensatory damages, I must remain mindful of the fact that statutory damages in copyright infringement cases are not only, or even primarily, intended to punish copyright infringers. They are also intended to compensate copyright owners in instances where the harm imposed by the infringer’s conduct is difficult to calculate.
See F.W. Woolworth Co. v. Contemporary Arts, Inc.,
Nevertheless, even in a copyright infringement action, there should be some nexus between the jury’s statutory damages award and the actual damages suffered by the plaintiff and the profits, if any, obtained by the defendant. 4 Nimmer & Nimmer,
supra,
§ 14.04[E][l][a], at 14-95;
id.
at 14-96 (“[Statutory damages ... should be woven out of the same bolt of cloth as actual damages.”);
see also Thomas-Rasset,
In summary, I conclude that it is appropriate to apply the three
BMW
guideposts to the jury’s award in this case. However, in applying these guideposts, I will remain cognizant of two factors that distinguish this case from a typical case in which punitive damages are awarded: (1) the jury’s award fell within a range authorized by Congress, and (2) the maximum and minimum amount of statutory damages that could be imposed for each of Tenenbaum’s acts of infringement was clearly set forth in section 504(c). While the
BMW
guideposts are helpful aids, my ultimate task is to determine whether the jury’s statutory damages award is “grossly excessive” in relation to the government’s legitimate interests in prescribing such awards — namely, compensating copyright owners and deterring infringement.
BMW,
2. The BMW Guideposts
a. The Third BMW Guidepost
Since the third
BMW
guidepost is arguably the most troublesome for Tenenb
*104
aum’s argument that the jury’s award violated the Due Process Clause, I begin with it. On its face, this guidepost, which counsels courts to consider “the difference between [the jury’s punitive award] and the civil penalties authorized or imposed in comparable cases,” weighs heavily in the plaintiffs’ favor.
BMW,
The most recent act of Congress addressing section 504(c)’s statutory damages provisions is the Digital Theft Deterrence and Copyright Damages Improvement Act of 1999 (hereinafter “Digital Theft Deterrence Act”), Pub.L. No. 106-160, 113 Stat. 1774, which increased the section’s statutory damages ranges to their current levels. 11 The timing of the Act suggests that legislators did not have in mind the problem of consumers sharing music through peer-to-peer networks when the Act was drafted. While the predecessor to the bill that eventually became the Digital Theft Deterrence Act was first introduced on May 11, 1999, see 145 Cong. Rec. 9233 (1999), Napster — the peer-to-peer network that brought file-sharing into the mainstream — was not released until June 1, 1999. Matt Hartley, The Phenorn That Launched a Billion Downloads, Globe & Mail (Can.), May 11, 2009, at A7.
To be sure, the legislation’s timing does not unambiguously militate in Tenenbaum’s favor. As the plaintiffs note, the Digital Theft Deterrence Act was not signed into law until December 1999, at which point Napster had been up and running for six months. Furthermore, the House Judiciary Committee’s report on the No Electronic Theft (NET) Act of 1997, Pub.L. No. 105-147, 111 Stat. 2678, which amended various statutory provisions governing the availability of criminal penalties for copyright infringement, noted that “an audio-compression technique, commonly referred to as MP-3, now permits infringers to transmit large volumes of CD-quality music over the Internet.” H.R.Rep. No. 105-339, at 4 (1997). And *105 well before 1999, recording companies had begun suing the operators of websites that provided users with unauthorized access to copyrighted sound recordings. (See Pis.’ Opp’n to Def.’s Mot. for New Trial or Remittitur 31, Case No. 07-cv-11446-NG, document # 36 (listing cases).)
The plaintiffs also emphasize the following language from the House Judiciary Committee’s report on an early version of the Digital Theft Deterrence Act:
By the turn of the century the Internet is projected to have more than 200 million users, and the development of new technology will create additional incentive for copyright thieves to steal protected works.... Many computer users are either ignorant that copyright laws apply to Internet activity, or they simply believe that they will not be caught or prosecuted for their conduct. Also, many infringers do not consider the current copyright infringement penalties a real threat and continue infringing, even after a copyright owner puts them on notice that their actions constitute infringement and that they should stop the activity or face legal action. In light of this disturbing trend, it is manifest that Congress respond appropriately with updated penalties to dissuade such conduct.
H.R.Rep. No. 106-216, at 3 (1999). According to the plaintiffs, this paragraph clearly indicates that Congress intended section 504(c)’s increased statutory damages ranges to deter individuals such as Tenenbaum from exploiting the Internet to engage in copyright violations.
Tenenbaum rejoins that this language from the committee report does not indicate that Congress intended for
file sharers
to face massive statutory damages awards. Much of the paragraph quoted by the plaintiffs was taken verbatim from the House Judiciary Committee’s report on the 1997 NET Act.
Compare
H.R.Rep. No. 106-216, at 3,
with
H.R.Rep. No. 105-339, at 4. The NET Act was intended to “reverse the practical consequences of
United States v. LaMacchia,
Since the Digital Theft Deterrence Act of 1999 was passed only two years after the NET Act and explicitly renewed its call for the Sentencing Commission to reevaluate the guidelines provisions for criminal copyright infringement, see Digital Theft Deterrence Act, sec. 3, 28 U.S.C.A. § 994 note; H.R.Rep. No. 106-216, at 4 (indicating that the low sentences meted out to criminal infringers discouraged the Department of Justice from bringing such prosecutions), Tenenbaum argues that Congress passed the 1999 Act primarily to target “malicious large scale operations like LaMacchia’s,” not individual file shar *106 ers such as Tenenbaum. (Def.’s Mot. & Mem. for New Trial or Remittitur 21, Case No. 07-cv-11446-NG, document #26.) While Tenenbaum’s account of the Act’s legislative history is interesting, I am skeptical whether there is as big a difference between Tenenbaum and LaMaechia as Tenenbaum claims. True, Tenenbaum did not create a software program that would allow users to share copyrighted materials. In this sense, he was more like a user of LaMacehia’s electronic bulletin board than he was like LaMaechia himself. In any event, Tenenbaum, like LaMaechia, not only downloaded copyrighted materials without authorization, he also distributed them by putting them in his shared folder. Furthermore, like LaMaechia, Tenenbaum’s conduct was willful, even though it was not carried out for commercial gain.
However, later statements by Senators Orrin Hatch and Patrick Leahy, two sponsors of the Digital Theft Deterrence Act, strongly suggest that Tenenbaum is correct; they did not anticipate that individuals such as Tenenbaum who engaged in noncommercial file-sharing would be subjected to liability for statutory damages under section 504(c). Hatch and Leahy presided over a Senate Judiciary Committee hearing titled “Music on the Internet: Is There an Upside to Downloading?” on July 11, 2000. Music on the Internet: Is There an Upside to Downloading?: Hearing Before the S. Comm, on the Judiciary, 106th Cong. (2000). During the hearing, the committee members demonstrated how the peer-to-peer system Gnutella is used by downloading and then playing a song by the band Creed. Id. at 7. As the committee was downloading the Creed song, Senator Leahy proudly proclaimed that he was doing some of his own downloading on his laptop. Id. at 7, 61. When one of the developers of Gnutella pointed out to the committee members that they might be engaging in copyright infringement, Senator Hatch responded that their downloading and public performance of the Creed song qualified as “fair use” since it was carried out for “educational and governmental purposes.” Id. at 40. Nevertheless, the senators’ willingness to download copyrighted sound recordings through a peer-to-peer network during a committee hearing suggests, at the very least, that they did not view such downloading as particularly reprehensible.
And this inference from the senators’ conduct is largely confirmed by their words. Although Senator Hatch noted that peer-to-peer technology had the capacity, “if misused, to rob [artists] of their livelihood,” id. at 3, he also praised the development of Gnutella as “quite an accomplishment,” id. at 8. And Senator Leahy added:
[W]hen I go on college campuses, as many of us do, to talk and everybody is talking about what they have downloaded, how they share, and so on, and when my kids pick up a “Black Muddy River,” which happens to be one of my favorites of the Dead, and send it to me — they have heard a new version — and I log on in the morning while I am having my breakfast and there it is, I mean this is a whole different world, and I think we have to recognize that on where we go.
Id. at 62.
Senator Hatch’s tolerance of, if not admiration for, peer-to-peer networks was even more on display at a special Judiciary Committee hearing held on October 9, 2000, at Brigham Young University (“BYU”). See Utah’s Digital Economy and the Future: Peer-to-Peer and Other Emerging Technologies: Hearing Before the S. Comm, on the Judiciary, 106th Cong. (2000). Shawn Fanning, the founder of Napster, was the star witness at this hearing, and Senator Hatch repeatedly *107 praised Fanning, expressing how “proud” he was of Fanning and even suggesting that Fanning should become a professor at BYU or run for political office. See id. at 2-3, 29, 34. Obviously, Senator Hatch’s comments should be taken with a large grain of salt. They are not authoritative statements of Congress and certainly do not control how the copyright statutes should be interpreted. (Also, Senator Hatch’s effusive praise of Fanning may well have stemmed from his awareness that he was appearing before an audience of college students, a sizable portion of whom likely used Napster.) But his comments nevertheless suggest that he did not anticipate that the statutory damages scheme over which his committee had jurisdiction would be applied to users of Napster and other peer-to-peer networks.
My analysis here has used legislative history not to divine the meaning of an ambiguous statutory provision. The plain language of 17 U.S.C. § 504(c) authorized the jury’s award in this case. I must give effect to this clear statutory language, at least to the extent that the jury’s award does not run afoul of the Due Process Clause.
See Caminetti v. United States,
Rather, I have examined section 504(c)’s legislative history to better understand the types of defendants members of Congress had in mind when they last increased the provision’s statutory damages ranges. If Congress did not foresee that section 504(c) would be used to mulct individual file sharers such as Tenenbaum in damages, it makes no sense to say that I must defer to Congress’ judgment that section 504(c)’s statutory damages ranges are appropriate in cases such as Tenenbaum’s; section 504(c) does not embody any such judgment.
Congress undoubtedly intended for the Copyright Act to be flexible enough to account for the rise of new technologies. However, the fact that peer-to-peer file-sharing was just emerging when Congress passed the Digital Theft Deterrence Act suggests that I should not simply defer to Congress’ statutory regime and assume that the jury’s award, because it is within the statutorily authorized range, is sufficiently related to the government’s legitimate interests in compensating copyright owners and deterring potential infringers to pass constitutional muster. Further inquiry is required.
Although
BMW’s
third guidepost focuses on the magnitude of civil penalties authorized by legislators, it is also helpful to compare the jury’s award in this case to the awards imposed in other copyright cases.
12
See Zimmerman v. Direct Fed.
*108
Credit Union,
The case most comparable to Tenenbaum’s is that of Jammie Thomas-Rasset, the only other file sharer to go to trial. The first jury to hear Thomas-Rasset’s case found her liable for willfully infringing twenty-four sound recordings and awarded the plaintiffs $9,250 per song, for a total award of $222,000.
Thomas,
Tenenbaum’s culpability seems roughly comparable to that of Thomas-Rasset. Both knew that file-sharing was illegal but engaged in it anyway. Both refused to *109 accept responsibility for their actions, trying to shift blame to others and even lying under oath. And both engaged in multiple acts of infringement. See id. at 1053. Thus, it seems that the awards in both cases should be about the same, suggesting that the jury’s award of $675,000 in this case should be significantly reduced.
The jury’s $675,000 award appears especially excessive when it is compared to the damages imposed on other file sharers whose cases have not made it to trial. Most individuals sued in the recording industry’s campaign against file-sharing have either settled with the recording companies or have allowed default judgments to be entered against them. When defendants have defaulted, the recording companies have generally asked courts to impose the statutory minimum damage amount of $750 per infringed work, and courts have routinely granted these requests.
See, e.g., Elektra Entm’t Group, Inc. v. Carter,
The plaintiffs cannot reasonably rely on the argument that higher damages are appropriate in this case because Tenenbaum insisted on taking his case to a jury. Under 17 U.S.C. § 505, the plaintiffs may move for the court to award them the costs of their action and any attorneys’ fees that they have reasonably incurred. The threat of bearing the opposing party’s court costs and attorneys’ fees should generally deter a defendant in a copyright infringement action from unduly prolonging the proceedings when his liability is clear. While the damages award should be sufficient to cover the plaintiffs’ costs of detecting Tenenbaum’s infringement, this cost is incurred by copyright owners even in cases where the defendant defaults. Recording companies’ willingness to accept damages of only $750 per infringed work in such cases suggests that section 504(c)’s minimum damages provision is roughly sufficient to encourage the recording industry to ferret out copyright infringement. Since section 505 provides for the awarding of attorneys’ fees and court costs, there is no reason to further inflate awards under section 504(c) to allow plaintiffs to recover their litigation expenses.
The jury’s award in this case also appears egregious in light of the damages typically imposed on restaurants, bars, and other businesses that play copyrighted songs in their establishments without first acquiring the appropriate licenses. These defendants are arguably more culpable than Tenenbaum. Unlike Tenenbaum, who did not receive any direct pecuniary gain from his file-sharing, defendants in these cases play copyrighted music to create a more pleasurable atmosphere for their customers, thus generating more business and, consequently, more revenue.
See EMI Mills Music, Inc. v. Empress Hotel, Inc.,
*111 I cannot conceive of any plausible rationale for the discrepancy between the level of damages imposed in public-performance cases and the damages awarded in this case. The disparity strongly suggests that the jury’s $675,000 award is arbitrary and grossly excessive.
b. The Second BMW Guidepost
The second
BMW
guidepost requires a court to consider the ratio between the actual or potential harm to the plaintiff and the punitive award assessed by the jury.
My analysis under the second
BMW
guidepost must focus squarely on Tenenbaum’s individual conduct, the benefits that he derived from that conduct, and the harm that he caused. While the plaintiffs argue that they have lost billions of dollars in revenue due to file-sharing, the jury was not permitted to punish Tenenbaum for harm caused by other infringers.
Cf. Philip Morris,
In assessing the plaintiffs’ actual damages, it is helpful to ask the following question: Assuming that Tenenbaum was entitled to file-share, how much would the plaintiffs have been willing to pay Tenenbaum not to engage in the activity? Presumably, the plaintiffs would have been willing to pay an amount equal to the profits they lost as a result of Tenenbaum’s conduct. Each of the songs that Tenenbaum illegally downloaded can now be purchased online from the iTunes Music Store and other retailers for approximately $0.99 or $1.29 a piece. And for each $0.99 song sold on the iTunes Music Store, it appears that the recording companies only receive about $0.70.
See Starr v. Sony BMG Music Entm’t,
If we use the $0.70 wholesale price for music sold on the iTunes Music Store as a rough proxy for the plaintiffs’ profits from each sale, then Tenenbaum’s illegal downloading of the thirty sound recordings for which he was found liable deprived the plaintiffs of approximately $21 in profit, for a ratio of statutory damages to actual damages of approximately 32,143:1. If we assume that the damages to plaintiffs equaled $1 per song, then the ratio is 22,-500:1, and if we assume damages of $15 per song (because before individual songs were widely available online through services such as iTunes, Tenenbaum would have needed to purchase an entire album to obtain a song he desired), the ratio is 1,500:1.
The plaintiffs rejoin that Tenenbaum did not merely download the thirty songs listed on the jury’s verdict form. He also downloaded thousands of other songs and distributed these songs to countless other file sharers through his shared folder. However, it is hard to believe that Tenenbaum’s conduct, when viewed in isolation, had a significant impact on the plaintiffs’ profits. He almost certainly would not have purchased all of the songs he downloaded if they were not available for free; thus, not all of his downloads represented lost sales for the plaintiffs. Also, it seems likely that the individuals who downloaded songs from Tenenbaum’s shared folder would simply have found another free source for the songs had Tenenbaum never engaged in file-sharing. While file-sharing may be very economically damaging to the plaintiffs in the aggregate, Tenenbaum’s individual contribution to this total harm was likely minimal.
The harm suffered by the plaintiffs, however, is not the only factor relevant to
*113
my analysis under
BMW’s
second guidepost. I must also consider the benefits that Tenenbaum reaped from file-sharing. We generally require individuals who wish to reproduce or distribute a copyrighted work to purchase a license from the copyright owner in a voluntary transaction. In this sense, copyrights are protected by what academies in the field of law and economics call a “property rule.”
See
Guido Calabresi & A. Douglas Melamed,
Property Rules, Liability Rules, and Inalienability: One View of the Cathedral,
85 Harv. L.Rev. 1089, 1092 (1972);
see also
Blair & Cotter,
supra,
at 1614 (“[Ijntellectual property rights are a paradigm example of entitlements protected by property rules.”). Property rules are distinguished from liability rules, which permit one party to deprive another party of something to which the law says he is entitled by paying an objectively determined value for it. Calabresi & Melamed,
supra,
at 1092. The quintessential example of a liability rule is a rule that permits a factory to pollute only if it compensates surrounding homeowners by paying them an amount of damages determined by a court.
See id.
at 1115-24;
see also Boomer v. Atl. Cement Co.,
Protecting copyrights with property rules rather than liability rules helps ensure that copyright owners reap the full benefit of their legally sanctioned monopoly power, which in turn provides an incentive for individuals to engage in creative endeavors. When copyrights are protected by property rules, the price of a license to use a copyright is determined through ordinary market interactions. If copyrights were protected by liability rules, however, the price of a license would effectively be determined by judges and juries, and their valuations might well be less than the price that would be established by
market forces. See Blair & Cotter, supra, at 1589 & n. 16,1615-16.
To avoid transforming the property rule protecting copyrights into a liability rule, the damages awarded in this case must be great enough to ensure that potential file sharers have an adequate incentive to purchase copyrighted songs from legitimate sources instead of downloading them from peer-to-peer networks.
See
Calabresi & Melamed,
supra,
at 1125 (“For us to charge the thief with a penalty equal to an objectively determined value of the property stolen would be to convert all property rule entitlements into liability rule entitlements.”). As a result, it would have been reasonable for the jury to return an award sufficiently high to disgorge the benefits Tenenbaum derived from file-sharing. The jury could also reasonably have increased its award to account for the fact that there was some probability — perhaps a significant one — that Tenenbaum’s file-sharing would go undetected. When there is some chance that infringing behavior will not result in civil liability, damages awards must be greater than the benefit that the potential infringer expects to derive from his illegal activity in order to achieve adequate deterrence.
See Taylor v. Merrick,
But even when these factors are considered, the jury’s award still appears grossly excessive. Tenenbaum did not derive any direct pecuniary gain from file-sharing. He did not, for example, sell the songs he illegally downloaded or charge for access to his shared folder. Instead, the “profit” that he reaped from his activities was more amorphous; he gained the ability to access an essentially unlimited variety of music on demand. In theory, we could quantify this benefit of file-sharing by asking how much Tenenbaum would have been willing to pay to engage in his infringing conduct. Although the record does not contain sufficient evidence to calculate this figure with precision, it is instructive to consider how much consumers are now willing to pay to have unlimited access to a large library of music. Today, a number of services allow users to access millions of songs from their computers and certain portable devices such as iPhones for a flat monthly fee of less than $15. See Brad Stone, Now Selling Music Files, Not Sharing, N.Y. Times, June 3, 2010, at Bl; see also Napster, http://www.napster.com (last visited July 7, 2010); Rhapsody, http://www.rhapsody.com/-discover (last visited July 7, 2010). Although these services are not the same as Napster or Kazaa, they do provide users with a wide assortment of music on demand. Tenenbaum began file-sharing in 1999 and was sued in 2007. Ignoring inflation and similar considerations, the cost of using a subscription service such as Rhapsody or “Napster to Go” for eight years would be approximately $1,440 or less. 16 Thus, it seems fair to say that the average consumer today would be willing to pay no more than $1,500 to engage in conduct roughly similar to Tenenbaum’s between 1999 and 2007.
I emphasize that I am not offering a definitive estimate of the benefit that Tenenbaum derived from file-sharing. Again, I recognize that the legal subscription services available today are not perfect substitutes for peer-to-peer networks such as Napster and Kazaa. Also, as the supply of a new technology increases, one would expect its price to correspondingly decline. Thus, the amount that an average consumer would be willing to pay for a subscription service today is almost certainly not the same as the amount that Tenenbaum would have been willing to pay to engage in file-sharing several years ago. Nevertheless, there can be no doubt that the benefit Tenenbaum derived from file-sharing is far less than the $675,000 in damages imposed in this case.
If we assume that Tenenbaum would have paid approximately $1,500 to engage in file-sharing from 1999 to 2007, the ratio between the statutory damages awarded in this case and the benefit he derived from his infringing conduct is 450:1, 17 far higher than the 114:1 ratio between the statutory damages awards approved in Williams and the profits reaped by the defendant railroad company from its overcharges. This shockingly high ratio between the *115 jury’s statutory damages award and Tenenbaum’s non-pecuniary “profits” cannot be justified by the fact that there was some probability that Tenenbaum’s file-sharing would not be detected. Since most individuals are risk averse, adequate deterrence can undoubtedly be obtained with an award that is much, much lower. 18 See Blair & Cotter, supra, at 1622; Polinsky & Shavell, supra, at 886-87, 913.
Two additional factors are relevant to my analysis under BMW’s second guidepost. First, section 504(e)’s statutory damages ranges likely include some amount to compensate copyright owners for the costs entailed in investigating and detecting infringing behavior. See Register of Copyrights Report, supra, at 103 (discussing how statutory damages help ensure a recovery sufficient “to warrant the expense of detecting infringements”). However, the Copyright Act does not contain a provision to correct for the fact that once a recording company has decided to devote the resources necessary to detect one act of infringement by a file sharer, the marginal cost of detecting additional acts of infringement is likely low since the investigation of the file sharer’s account on a peer-to-peer network will generally reveal a treasure trove of unlawfully downloaded works. As a result, the imposition of statutory damages in file-sharing cases where multiple copyrighted works have been infringed can produce awards much greater than necessary to provide copyright owners with an adequate incentive to investigate and detect infringement.
Similarly, section 504(c) is not attuned to what one author has called the issue of “proportionate reprehensibility.” Barker, supra, at 550, 552-53. Congress certainly had a legitimate interest in setting statutory damages ranges that would impose a measure of retribution on copyright infringers. However, the reprehensibility of a file sharer’s conduct does not increase linearly with the number of songs he downloads and shares. Someone who illegally downloads 1,000 songs may be more blameworthy than an individual who illegally downloads only one, but it seems odd to say that his conduct is 1,000 times more reprehensible. Section 504(c) ignores this issue entirely, providing the same statutory damages ranges for each infringed work no matter how many works are infringed. Consequently, the aggregation of statutory damages awarded under section 504(c) may result in unconscionably large awards.
In summary, the asymmetry between the relatively small harm suffered by plaintiffs and benefit reaped by Tenenbaum, on the one hand, and the jury’s extraordinarily high award, on the other, is so extreme as to “jar [the Court’s] constitutional sensibilities.”
Haslip,
c. The First BMW Guidepost
The “degree of reprehensibility of the defendant’s conduct” is “[p]erhaps the most important indicium of the reasonable
*116
ness” of a punitive award.
BMW,
Other factors, however, militate against Tenenbaum. He willfully engaged in thousands of acts of copyright infringement, knowing his conduct to be illegal but acting anyway.
See State Farm,
It seems fair to say that file-sharing, in general, is fairly low on the totem pole of reprehensible conduct. Although the activity is quite pernicious in the aggregate, it regrettably is quite common, and most file sharers do not receive any direct pecuniary gain from their activity (although they do save money by avoiding the need to pay for the music they download). But among this group of comparatively venial offenders, Tenenbaum is one of the most blameworthy since he engaged in the activity for a long period of time, knowing it to be illegal, and then lied in a futile attempt to cover his tracks.
3. What is the maximum constitutionally permissible damages award in this case?
Based on my review of the
BMW
factors and the standard articulated in
Williams,
I conclude that the jury’s award of $675,000 violates the Due Process Clause. The award bears no rational relationship to the government’s interests in compensating copyright owners and deterring infringement. Even under the
Williams
standard, the award cannot stand because it is “so severe and oppressive as to be wholly disproportioned to the offense and obviously unreasonable.”
And as Tenenbaum notes, a highly punitive award is likely less called for in his
*117
case than in
Williams.
Tenenbaum was an ordinary young adult engaging in noncommercial file-sharing, not a wealthy railroad bilking customers for its own profit. Furthermore, for the reasons set forth above, the egregiousness of the jury’s award becomes even more apparent when it is analyzed in light of the Supreme Court’s recent punitive damages jurisprudence. The amount of statutory damages imposed on Tenenbaum is simply “unprecedented and oppressive.”
Thomas,
I must reduce the jury’s unconstitutional $675,000 award to the maximum amount that is consistent with the dictates of the Due Process Clause.
See, e.g., Bach v. First Union Nat’l Bank,
Some will undoubtedly murmur that my decision to draw the constitutional line at $2,250 per infringed work is to some extent arbitrary. But this criticism applies to
any
line drawing process; it is always possible to argue that the line should have been drawn a bit differently.
Cf. Inter Med. Supplies, Ltd. v. FBI Med. Sys., Inc.,
I again emphasize that the total of the reduced award, $67,500, is significant and harsh. It adequately compensates the plaintiffs for the relatively minor harm that Tenenbaum caused them and, even more importantly, should serve as a strong deterrent against unlawful file-sharing. The award is higher than I might have awarded in my own independent judgment and is the maximum that the Constitution will permit given the facts of this case.
IV. MISCELLANEOUS ITEMS
The other issues raised in Tenenbaum’s motion for a new trial or remittitur warrant little comment. I address them briefly below.
A. Fair Use
More than one and a half years after Tenenbaum filed his original answer to the plaintiffs’ complaint, and well after his attorney entered an appearance in this case, he amended his answer to include an affirmative defense that his file-sharing constituted a “fair use” under the Copyright Act.
Sony BMG Music Entm’t v. Tenenbaum,
In my December 2009 opinion, I hypothesized that “a defendant who used the new file-sharing networks in the technological interregnum before digital media could be purchased legally, but who later shifted to paid outlets,” might be able to rely on the defense of fair use. Id. I concluded, however, that Tenenbaum could not assert such a defense because the plaintiffs detected his file-sharing in August 2004, more than a year after the iTunes Music Store, which made authorized digital downloads widely available, debuted in April 2003. Id. at 222, 236.
Tenenbaum now argues that the “technological interregnum” recognized in dicta in my prior opinion actually extended to 2007, when the plaintiffs finally allowed consumers to purchase songs online that were not encrypted with digital rights management technologies (“DRM”). He also argues that I improperly rejected his argument that the plaintiffs created an “attractive nuisance” for young adults such as Tenenbaum by aggressively promoting their copyrighted works in a world in which file-sharing was prevalent. Finally, he claims that I erred by failing to afford sufficient weight to particular factors in my fair-use analysis, including the costs “borne by parents and schools charged with policing the online activities of chil *119 dren and students; costs on universities compelled to disclose the names of their own students using computers connected to their university network; and the intrusions upon the privacy of individuals entailed by forced inspections of their computers.” (Def.’s Mot. & Mem. for New Trial or Remittitur 7.)
I reaffirm my prior ruling on Tenenbaum’s fair-use defense. I considered both Tenenbaum’s “attractive nuisance” argument and the “policing costs” required to crackdown on file-sharing in my December 7 opinion.
See Tenenbaum,
Tenenbaum’s argument about the length of the technological “interregnum” is also unavailing. Tenenbaum effectively blames the plaintiffs for his conduct because they did not make their copyrighted works available in the format he preferred. Even if a copyrighted work’s commercial availability factors into the fair-use analysis, a consumer does not have a right to demand that a copyright owner make his work available in the exact format that the consumer desires.
Cf. Harper & Row, Publishers, Inc. v. Nation Enters.,
B. Tenenbaum’s Evidentiary Challenge
At trial, Tenenbaum attempted to introduce into evidence the full text of a letter that he wrote on November 21, 2005, offering to settle the plaintiffs’ claims against him for $500 and tendering a money order in that amount. Tenenbaum wrote the letter in response to a letter that the plaintiffs sent him in September 2005 demanding that he cease his file-sharing activities. (See Tr. Tenenbaum Trial Testimony 10:18 to 11:12, 49:5-7, 72:10-23.) In addition to the settlement offer, the letter contained the following promise from Tenenbaum:
While I do not have access to the computer [on which Tenenbaum had saved illegally downloaded files] at college, I will be home on November 22nd. If there are any files existing in violation of copyrights, I will destroy them at that time.
(Def.’s Mot. & Mem. for New Trial or Remittitur 9 (quoting the full text of the letter).) I allowed the plaintiffs to introduce the above-quoted language from the letter but required that Tenenbaum’s settlement offer be redacted.
Tenenbaum argues that this decision was erroneous in two respects: (1) his settlement offer need not have been excluded under Federal Rule of Evidence 408; and (2) shorn of its context in a letter offering settlement, the language quoted above appeared to the jury to be an unconditional promise to delete illegally downloaded music files, which provided fodder for the plaintiffs since evidence presented at trial showed that Tenenbaum did not in fact delete the files.
Rule 408 provides that evidence of “offering ... to furnish ... valuable consideration in compromising or attempting to compromise [a] claim” is “not admissible on behalf of any party ... to prove liability for, invalidity of, or amount of a claim that *120 was disputed as to validity or amount.” Fed.R.Evid. 408(a). Such evidence, however, may be admissible for other purposes, such as “proving a witness’s bias or prejudice; negating a contention of undue delay; and proving an effort to obstruct a criminal investigation or prosecution.” Id. 408(b).
Tenenbaum argues that his offer of settlement was admissible to show that as early as November 2005, he was willing to take responsibility for his file-sharing. 20 He contends that this evidence could have been used to combat the plaintiffs’ attempts to portray him as an unrepentant, “hardcore” file sharer. According to Tenenbaum, proof of his alleged acceptance of responsibility is not excludable under Rule 408 because he was not seeking to introduce it to “prove liability for, invalidity of, or amount of a claim.” Fed.R.Evid. 408(a); see also id. 408(b) (providing that Rule 408 “does not require exclusion if the evidence is offered for purposes not prohibited by subdivision (a)”).
Even if Tenenbaum was offering the redacted portion of the November 2005 letter for a purpose permitted by Rule 408, my ruling was nevertheless proper under Rule 403. See Fed.R.Evid. 403 (“Although relevant, evidence may be excluded if its probative value is substantially outweighed by the danger of unfair prejudice, confusion of the issues, or misleading the jury, or by considerations of undue delay, waste of time, or needless presentation of cumulative evidence.”). Admitting evidence of Tenenbaum’s settlement offer would have opened the door to further inquiry into the course of the parties’ settlement negotiations. Plaintiffs would have tried to show that Tenenbaum was intransigent in subsequent settlement discussions. This line of inquiry would have had little probative value, likely would have wasted the Court’s time, and potentially could have confused the jurors by inundating them with information that had little pertinence to the matters they were to decide. It also may have required the parties’ attorneys to take the stand as witnesses to testify about the course of the settlement discussions. This could have triggered a wave of disqualifications, a result that Rule 408 was intended to avert. See Fed.R.Evid. 408 advisory committee’s notes (2006 amendment) (“[Pjroof of statements and offers made in settlement would often have to be made through the testimony of attorneys, leading to the risks and costs of disqualification.”).
Tenenbaum’s argument that the plaintiffs improperly used the redacted letter as “damning evidence of perfidy” by showing that he failed to “destroy! ] the offending files as he had apparently and unequivocally promised to do” is meritless. (Def.’s Mot. & Mem. for New Trial or Remittitur 9.) To start, the plaintiffs primarily used the November 2005 letter to impeach Tenenbaum’s credibility by showing that he had waffled on the issue of whether a computer he used to download and distribute copyrighted works had been destroyed. *121 While his November 21, 2005, promise to delete any unlawfully downloaded files on the computer suggested that the computer was operational as of that date, Tenenbaum later appeared to take the position that the computer had been destroyed before his promise was made. (Tr. Tenenbaum Trial Testimony 52:1 to 53:6.) In fact, it was Tenenbaum’s own attorney, not the plaintiffs, who made explicit the fact that he had failed to fulfill his promise to delete the files. 21 (Id. at 73:12.)
But even if I ignore this serious hole in Tenenbaum’s argument, my ruling still stands. Despite Tenenbaum’s protestations to the contrary, a reasonable jury could not have construed his promise as contingent on the plaintiffs’ acceptance of his settlement offer. Tenenbaum did not use conditional language; instead, he directly asserted: “I will be home on November 22nd. If there are any files existing in violation of copyrights, I will destroy them at that time.” He also dated this letter November 21, 2005, and the date he set for the destruction of the files was November 22. It seems unlikely that the plaintiffs or their attorneys even received Tenenbaum’s letter by November 22. Thus, it is hard to see how Tenenbaum’s promise to destroy the files while home on a break from school could have been contingent on the plaintiffs’ acceptance of his settlement offer by that date.
In addition, any harm caused by my rulings with regard to the November 2005 letter was mitigated by the parties’ stipulation that they engaged in settlement negotiations that ultimately failed. (Joint Stipulation re Settlement Negotiations, Ex. J to Pis.’ Opp’n to Def.’s Mot. for New Trial or Remittitur.) Thus, even if I erred by excluding Tenenbaum’s settlement offer and admitting the redacted letter, the error was harmless because the jury was informed that the parties had attempted to resolve their dispute through a settlement agreement. See Fed.R.Civ.P. 61 (setting forth the harmless error standard).
V. CONCLUSION
The jury’s $675,000 award is wholly out of proportion with the government’s legitimate interests in compensating the plaintiffs and deterring unlawful file-sharing. No plausible rationale can be crafted to support the award. It cannot withstand scrutiny under the Due Process Clause.
The fact that I reduce this award, however, obviously does not mean that Tenenbaum’s actions are condoned or that wholesale file-sharing in comparable circumstances is lawful. I have determined that Tenenbaum’s conduct was not “fair use” and that it infringed the plaintiffs’ copyrights. Furthermore, the jury’s award, even as reduced, is unquestionably severe and is more than adequate to satisfy the statutory purposes and the plaintiffs’ interests.
I GRANT Tenenbaum’s Motion for a New Trial or Remittitur (document # 26) insofar as it seeks a reduction in the jury’s award on the grounds that it is so grossly excessive as to violate the Constitution. I DENY the balance of Tenenbaum’s motion for the reasons stated in this opinion. I will amend the judgment in this case to reduce the jury’s award to $2,250 for each of the thirty infringed works.
SO ORDERED.
Notes
. In particular, the plaintiffs are Sony BMG Music Entertainment, Warner Bros. Records Inc., Atlantic Recording Corp., Arista Records LLC, and UMG Recordings, Inc.
. Tenenbaum raised a similar argument in a pretrial motion to dismiss under Federal Rule of Civil Procedure 12(b)(6). (Def.'s Mot. to Dismiss, Case No. 03-cv-11661-NG, document # 779.) I denied Tenenbaum’s motion without prejudice to his right to file a post-trial motion challenging the constitutionality of any award the jury might return. (Order re: Def.’s Mot. to Dismiss, Case No. 03-cv-11661-NG, document # 847.)
. Although I grant Tenenbaum’s motion for a new trial or remittitur insofar as it seeks a reduction of the jury’s statutory damages award, I deny the motion in all other respects. In particular, I reaffirm my prior rejection of Tenenbaum’s affirmative defense of fair use and deny his request for a new trial based on my admission of a redacted letter that Tenenbaum mailed to the plaintiffs soon after his file-sharing was detected.
. I could also avoid Tenenbaum’s constitutional challenge by holding that I erred in granting the plaintiffs summary judgment on Tenenbaum’s affirmative defense of fair use or admitting into evidence the redacted text of a letter that Tenenbaum sent to the plaintiffs *92 in November 2005. However, as I discuss in Part IV below, I reject each of these grounds for granting Tenenbaum a new trial.
. While Congress has instructed courts not to inform juries in Title VII cases that their awards are subject to a statutory ceiling,
see
42 U.S.C. § 1981a(c)(2);
Sasaki v. Class,
. I instructed the jurors that they could consider the following non-exhaustive list of factors in awarding statutory damages:
(a) The nature of the infringement;
(b) The defendant's purpose and intent;
(c) The profit that the defendant reaped, if any, and/or the expense that the defendant saved;
(d) The revenue lost by the plaintiff as a result of the infringement;
(e) The value of the copyright;
(f) The duration of the infringement;
(g) The defendant’s continuation of infringement after notice or knowledge of copyright claims; and
(h) The need to deter this defendant and other potential infringers.
(Jury Instructions 3, Case No. 03-cv-11661-NG, document # 909.) In addition, I informed them that if they found that Tenenbaum's infringements were willful, they could also consider this fact in arriving at a statutory damages award. (Id. at 4.)
. The jury had originally awarded $80,000 per song, for a total award of $1,920,000.
Capitol Records Inc. v. Thomas-Rasset,
. Although I do not employ the remittitur procedure, I reject the plaintiffs’ contention that it is unavailable in cases where a jury has returned a statutory damages award under the Copyright Act.
See Thomas-Rasset,
Furthermore, if judicial review were not available, section 504(c) would arguably be unconstitutional. In
Honda Motor Co. v. Oberg,
. Although the Supreme Court relied on its common-law authority in maritime cases, not on the Due Process Clause, in reducing the punitive damages award in
Exxon Shipping Co. v. Baker,
.
Compare Zomba Enters., Inc. v. Panorama Records, Inc.,
. Before this bill was passed, the statutory damages range for ordinary non-willful infringement was $500 to $20,000 per infringed work and the maximum award for willful infringement was $100,000. See Berne Convention Implementation Act of 1988, Pub.L. No. 100-568, § 10(b), 102 Stat. 2853, 2860.
Although Congress again revised section 504 in 2004, it merely added paragraph (3) of section 504(c), which creates a rebuttable presumption that an infringement was committed willfully if the infringer knowingly provided materially false contact information to a domain name registry "in registering, maintaining, or renewing a domain name used in connection with the infringement.” Intellectual Property Protection & Courts Amendments Act of 2004, Pub.L. No. 108-482, § 203, 118 Stat. 3912, 3916-17 (codified at 17 U.S.C. § 504(c)(3)). This provision is not implicated in this case.
. I note that the jurors in this case could not perform a similar analysis. Congress originally intended for judges, not juries, to determine the appropriate amount of statutory damages in copyright infringement actions.
See Feltner,
. I recognize that
Zimmerman
stated that "a reviewing court should search for comparisons solely to determine whether a particular defendant was given fair notice as to its potential liability for particular misconduct, not to determine an acceptable range into which an award might fall.”
But even under Zimmerman, my analysis of the statutory damages awards returned in other copyright cases is appropriate. Notice of section 504(c)'s extraordinarily broad statutory damages ranges, standing alone, does not in any meaningful sense constitute "fair notice” of the liability that an individual might face for file-sharing. In a case of willful infringement such as this one, the maximum damages per infringed work — $150,-000 — are 200 times greater than the statutory minimum of $750. Since the jury found that Tenenbaum willfully infringed thirty copyrights, its award could have ranged from a low of $22,500 to a high of $4,500,000. For anyone who is not a multi-millionaire, such "notice” is hardly more illuminating than the notice that BMW and State Farm had that their fraudulent conduct might lead to the imposition of a punitive damages award ranging from $0 to infinity.
Since section 504(c) failed to provide Tenenbaum with fair notice of the liability he could incur for file-sharing, it is imperative that I review other copyright cases to determine whether the jury’s $675,000 award here fell within a discernible pattern of awards of which Tenenbaum could have taken note, or was instead an unforeseeable outlier.
. These cases were not cherry picked to highlight the excessiveness of the jury’s *111 award. To get a better sense of the amount of statutory damages awarded in copyright actions, the Court ran a Lexis-Nexis search for all cases decided between January 1, 2008, and January 1, 2010, discussing statutory damages awards under section 504(c). The Court randomly selected approximately fifty of these cases to review; the cases included in the string cite above come from this list. Based on its independent review of the case law, the Court feels certain that these cases are not in any way abnormal.
. To be clear, I do not intend to suggest that the unauthorized sharing of copyrighted works over peer-to-peer networks is a socially beneficial activity. Aside from being illegal, such conduct may reduce demand for music from legitimate sources and thus dampen the monetary incentive for artists to create new works.
See Sony BMG Music Entm’t v. Tenenbaum,
. $15 per month for ninety-six months equals a total cost of $1,440.
. $675,000/$!,500 = 450/1.
. An individual is risk averse if she “preferís] a certain income to a risky income with the same expected value.” Robert S. Pindyck & Daniel L. Rubinfeld, Microeconomics 157 (5th ed. 2001). For example, a risk-averse individual would prefer a job with "a certain income of $20,000” to “a job yielding an income of $30,000 with probability .5 and an income of $10,000 with probability .5 (so that the expected income is $20,000).” Id. A risk-neutral individual, in contrast, would be indifferent between the two jobs because their expected incomes are equivalent. Id.
. Although Chief Judge Davis reduced the jury’s award in
Thomas-Rasset
under the common-law doctrine of remittitur and thus did not reach the question of whether the award violated the Due Process Clause, I nonetheless find Chief Judge Davis’ decision on the remittitur issue instructive in determining the maximum constitutionally permissible award in Tenenbaum’s case. While Chief Judge Davis relied on different grounds in reducing the jury’s award in
Thomas-Rasset,
the question he confronted was essentially identical to the one I face: What is “the maximum amount a juiy could reasonably award to both compensate Plaintiffs and address the deterrence aspect of the Copyright Act”?
Thomas-Rasset,
. He also argues that Rule 408 does not apply when a party seeks to admit his own settlement offer. This seems to have previously been the prevailing view,
see Innovative Eng’g & Consulting Corp. v. Hurley & Assocs., Inc.,
No. 1:05CV0764,
. The plaintiffs, however, did make a fleeting reference to Tenenbaum’s failure to delete the files in their closing argument. (Trial Tr. 77:6-9, July 31, 2005, Case No. 03-cv-11661-NG, document #916.)
