An Ironic Model: Follow Exxon Because it’s NOT a Constitutional Ruling

Way back in October, I got half-way to proposing a constitutional model for statutory damage awards. More precisely, I got through the part where I reject the two leading models, which happened to have been the competing models in Capitol Records v. Thomas-Rasset—(1) St. Louis, Iron Mtn. & S. Ry. Co. v. Williams and (2) BMW of N. Am., Inc. v. Gore. My starting off point, which you may or may not agree with, is that we have a right to some degree of predictability of the legal consequences of our actions, which is necessary for us to go about our business, and that statutory damages have become out of control and unpredictable.

What I’d like to propose is, ironically, that we avoid the whole constitutional question because the Copyright Act is federal law, and federal law already has a way of dealing with a close cousin (in my opinion) of statutory damages: exemplary damages. It is, thus, the Supreme Court’s opinion in Exxon, not Gore and not Williams*, that governs here.

* Williams governs, too, in that it governs a legislative body’s power to enact and set statutory damages, but Williams doesn’t answer the question about a court’s power to make a specific award of statutory damages where there is a wide and, therefore, unpredictable range of awards.

Goring Thomas-Rasset with Williams

Neither model was really appropriate to the specific question posed by Thomas-Rasset: when is a specific award of statutory damages unconstitutionally unreasonable? Williams dealt with statutory damages, but it addressed the broader question of the enabling legislation is within the legislature’s power to enact, not the specific question of what a jury does within the legislatively-approved framework. In other words, following Williams , a court might reasonably hold that a statutory range of $750 to $150,000 is within Congress’ powers, but it says nothing about whether an award of $150,000 for infringement of my children’s finger paintings is permissible.

By contrast, Gore provided a framework for dealing with a specific award of non-compensatory damages, but it dealt with punitive damages, not statutory damages. It’s a bad fit because statutory damages combine the characteristics of compensatory and punitive damages. Perhaps as a result, Gore’s framework is almost nonsensical when applied to statutory damages.

In a perfect world, there would be legal authority on what limitations, if any, the U.S. Constitution places on a court’s specific award of statutory damages (beyond that already built into the statute). But no legal authority exists (that I, or anyone else, has been able to find).

Constitutional Avoidance

But let’s take a step back for a moment. There’s a more fundamental question. The Copyright Act is, after all, a federal statute, so we don’t have tricky and complex questions of federalism. That’s one reason why Gore isn’t applicable: Gore asked, “What does the U.S. Constitution have to say about the way states handle punitive damages?” But that’s getting ahead of ourselves. The initial, unasked unquestion is: “How do the federal courts handle punitive damages?”

Again, statutory damages aren’t quite the same thing as punitive damages, but they have a lot in common. Statutory damages have an exemplary component, but they aren’t exclusively exemplary in nature, unlike punitive damages. They give almost unfettered discretion to the factfinder, and in that respect, they’re unpredictable. To be sure, statutory damages have a floor and a ceiling, but with a ratio of 200:1 from maximum to minimum, having a range doesn’t make them much more predictable.

If you accept that statutory damages are a kind of exemplary damages, or are at least exemplary in character*, then shouldn’t we treat statutory damages in a manner similar to how exemplary damages are treated under federal law? All the more so when Congress itself provides no guidance, which pretty much invites the courts to come up with its own guidance.

* If we don’t, then how do we justify the really huge awards? Call me crazy, but it seems clear to me that most of the various judgments against Thomas-Rasset were attributable to a desire to make an example out of here: to deter others and to punish her (the kicker is that the awards only got bigger when she testify so unconvincingly).

It’s pretty rare for courts to apply federal law to punitive damages. They are almost entirely a creature of state law. As luck would have it, though, there is a U.S. Supreme Court that directly sets forth how federal law handles punitive damages.

Clarifying the Law with Crude Oil

That decision is Exxon Shipping Co. v. Baker. Exxon dealt with the infamous oil spill from the tanker Exxon Valdez in Prince William Sound, Alaska. One of the strange aspects of the case was that the accident didn’t take place in any state, but in open water, so the court had to follow the rarely-used admiralty law (called “maritime law” in the opinion). Admiralty law is federal law, and while it governs certain matters unique to shipping (such as salvage of wrecked ships), it also governs many matters that are ordinarily the province of state law, such as damage to property and persons. In the Exxon case, the claims had to do with the massive disruption of the livelihoods of the fishermen and others by the oil spill.

Thus, when the jury returned a verdict including punitive damages against Exxon, the question was not whether—or, at least, not initially whether—there was some Constitutional limitation on punitive damages (the question posed in Gore). The federal courts were not being asked to review whether the way a state handles punitive damages violates the Constitution. Instead, the initial question is more basic: how, exactly, does the federal system handle the rare instance of punitive damages?

This is not a constiutional question, but a question we must ask before we get to the constitutional question and, indeed, may obviate the need to ask any constitutional questions (always a point in favor*).

* Courts are admonished to avoid constitutional issues if there is another way to decide the matter.

The Court was clearly most concerned about the inconsistency and unpredictability of punitive awards, and so it spent the bulk of its opinion (on this issue) on (1) the historical basis for punitive damages, the strength of which prevented the Court from simply chucking them; and (2) consideration of various methods for keeping punitive damages under control.

It considered three methods, one of which it described as “verbal” and the other two as “quantitative.” The first method is to come up with really good jury instructions, but the Court rejects it as just window-dressing. The second method is to come up with judicial maxima and minima, rather like the sentencing guidelines for federal crimes. But the Court rejects that as too complex because no guidelines could possibly take into account the wide variety of civil (as opposed to criminal) wrongs. In the end, it endorsed the third method, which was a fairly strict 1:1 ratio of punitive damages to compensatory damages. Without proof of really egregious behavior, punitive damages shouldn’t exceed compensatory damages.

Alas, we cannot simply apply the Supreme Court’s preferred method to statutory damages because the point of statutory damages is to avoid the need to prove actual damages. It is simply a society judgment embodied in the Copyright Act that any work of original creativity, even my children’s finger painting, deserves a certain minimum level of protection—$750 worth of protection, to be precise. Without, it would be far too easy to get away with pretty egregious copying because most acts of creativity have little or no economic value—or the damages that flow from the copying are too hard to nail down. This is, in fact, one of the problems that Gore ran into.

The other “quantative” method, placing a cap on the award, just restates the problem that we’re trying to solve. As the Supreme Court noted, that problem is pretty serious: “The trouble is, though, that there is no ‘standard’ tort or contract injury, making it difficult to settle upon a particular dollar figure as appropriate across the board.” Exactly. There a countless types of creative works, and countless ways to infringe them. A cap of $35,000 or $150,000 is a blunt instrument, especially where multiple works are at issue.

If you are the unhappy recipient of demand letter, accusing you of downloading pornographic content over the Internet using a BitTorrent protocol, is it any comfort to you that the maximum you can be found liable for is $150,000, even if you have no idea what BitTorrent is? More to the point, if you are building a business model that traipses into some grey area of copyright law—and there are numberless grey areas—and risks mulitple infringements, is it sufficient information to know that, if the law goes against you, you and your company will be liable for damages whose spread is represented by a ratio of 200:1? If 2:1 is too much for the Supreme Court, how is 200:1 OK?

Three General Principles for Exemplary Damages

So that leaves us with the “verbal” method, so maligned by the Supreme Court. Having surveyed some states’ jury instructions for punitive damages, the Court wrote:

These examples leave us skeptical that verbal formulations, superimposed on general jury instructions, are the best insurance against unpredictable outliers. Instructions can go just so far in promoting systemic consistency when awards are not tied to specifically proven items of damage (the cost of medical treatment, say), and although judges in the States that take this approach may well produce just results by dint of valiant effort, our experience with attempts to produce consistency in the analogous business of criminal sentencing leaves us doubtful that anything but a quantified approach will work. A glance at the experience there will explain our skepticism.

One of the jury instructions the Court rejected was this one from Maryland:

An award for punitive damages should be:

(1) In an amount that will deter the defendant and others from similar conduct.

(2) Proportionate to the wrongfulness of the defendant’s conduct and the defendant’s ability to pay.

(3) But not designed to bankrupt or financially destroy a defendant.

For the Supreme Court, this isn’t good enough to protect the interest of accused defendants, but it is considerably better than nothing, which is precisely what the jury had in Thomas-Rasset. Had the jury had these instructions, and if the judge could have reviewed the jury’s award based on these squishy, vague factors, they could not have come up with such shocking awards. The award would not have been permitted to “financially destroy” Thomas-Rasset. (And later juries, which returned even larger verdicts, might have put aside their anger at her apparent dishonesty and considered that, in the end, no one deserves to be financially ruined for allowing others to download 27 songs.)

A Difference Between a File-Sharer and Google

Rights holders need not run in terror from this. They might actually be better off with jury instructions like this for two reasons. First, the RIAA’s overkill victory over Thomas-Rasset did it no favors. It won far more than it could collect, but since copyright law treats the Thomas-Rassets of the world the same way as the Googles of the world, it felt compelled to pursue this P.R.-disaster of a judgment vigorously because conceding grounds for reduction to Thomas-Rasset would be tantamount to conceding the same to Google. That’s why it was so eager to settle with her, so it could announce a reduced amount without setting any precedent.* Under these jury instructions, Thomas-Rasset would have received a much lower amount anyway, but in a way that would have no effect on the Googles of the world, since the reduction would be tied to ability to pay.**

* Technically, the RIAA wanted Thomas-Rasset to agree to vacating a previous order reducing her award (which reduction was later overturned anyway), which just underlines the Hobson’s choice the RIAA was faced with: weaken your rights or look like an ass.

** As I’ve argued before, to a person of ordinary means, there really is no difference between the threat of $100,000 and $1,000,000: either award would financially destory the person. In fact, there might not be much of a difference between $25,000 and $100,000 since one suspects a $100,000 judgment creditor might be pretty lucky to collect 25% of that in bankruptcy.

Second, rights holders could concentrate on increasing the maximum of the statutory range. $150,000 is a lot to you and me, but it’s not very much for Google. During the Oracle v. Google case, there was a point at which Google thought it had mostly won, but might have to pay statutory damages for one copyrighte work (in the end, they got a complete victory), and Google would “only” have to pay no more than $150,000. That’s not enough to make big companies think twice. If statutory damages were tied to ability to pay, we could increase this upper limit enough to matter to the Googles of the world, without increasing the risk to ordinary folks.* We might even come up with a more sensible formula that takes into account the acts of infringement, rather than just the number of works infringed.

* One worry: a common form of copyright arbitrage is to threaten ordinary folks with judgments for copyright infringement “up to $150,000 per work infringed.” This threat—often completely bogus or at least highly exaggerated—might be more persuasive if it read “up to $1.5 million.” Then again, that might be so high that no one would take it seriously.

Refining the Crude Instructions

The test/instructions could be refined, but surely it’s the right place to start. They need to take into account the compensatory nature of the award, which necessarily means giving the jury an opportunity to consider evidence of actual harm (to the rights holder) and actual beneift (to the infringer) of the infringement. A single reproduction of my children’s finger paintings should be treated differently than a scheme to counterfeit thousands of CDs with millions of dollars in sales. They should probably also take into account intent—without duplicating the purpose of willfulness—since copyright infringement doesn’t require it (but obviously punitive damages does).

We might take a cue from fair use, which takes into account (1) the nature of the infringement (in terms of commerciality, “transformative” uses, etc.); (2) the nature of the work (in terms of creativity, usually); (3) the nature and amount taken; and (4) effect on the market for the underlying works.

But let’s take care not to get carried away with too many factors, which would only serve to confuse the jury and perhaps over-emphasize certain aspects.*

* In the Tenenbaum case, the court instructed the jury to take into account nine non-exhaustive factors, and the jury still came back with an award of $22,500 per song. Notably, none of the factors had anything to do with Tenenbaum’s ability to pay. Tenenbaum’s lawyers didn’t object.

A revised set of instructions might look like this, then:

An award of statutory damages should:

  1. Be proportionate to the actual harm done to the rights holder and the actual benefit accrued to the infringer;
  2. Account for infringer’s purpose for committing the infringing acts;
  3. Be sufficient to deter others from committing similar infringing acts;
  4. Be proportionate to infringer’s ability to pay and should not financially ruin the infringer.

You can probably do better. Let me know in the comments how would improve these factors.

Thanks for reading!