Rewarding Socially Beneficial Innovation: A Further Thought on the Aereo Case and High-Frequency Trading
By Michael Dorf
In Professor Hockett's preview of the Aereo oral argument last week, he argued that it would be wonderful if, in considering how to resolve the particular dispute under the Copyright Act, the Justices paid some considerable attention to the broader question of the implications of public ownership of the airwaves. In a related spirit, I want to use the oral argument (transcript now available here) as the launching point for a discussion about how Congress and the courts ought to think about innovation--both in the current context and more broadly.
First, a reminder of the dispute's basics: Aereo has facilities in various cities around the country at which it rents tiny antennae and disk space on an individualized basis, thereby allowing its customers to receive via the internet recorded broadcast programming. As Prof. Hockett explained, the key question in the case is whether Aereo's service amounts to a "public performance", and thus infringes the copyrights of the broadcasters, or amounts to mere time-shifting by the end users, and thus is permissible as not a public performance (pursuant to the doctrine of the 1984 Sony case).
Prompted by the briefing, during the oral argument, Justice Sotomayor raised concerns about the implications of plaintiffs' argument. In particular, she worried that if Aereo's transmission of the recorded signals of the broadcasters from their antenna/dvr farms in Brooklyn etc counts as an unlicensed "public performance" in violation of the broadcasters' copyright, then cloud computing might be in trouble.
Suppose I pay for and download a licensed copy of the Beatles song Across the Universe. If I keep the file on my home computer, I can play it over and over again, without the need for a public performance license, because this is a private performance. However, under the plaintiffs' view in Aereo, if my home computer syncs my files to the cloud (via Dropbox, iCloud, Box, OneDrive, GoogleDrive, or whatever), then if I happen to play Across the Universe from my cloud-based storage rather than from the hard drive of my computer, the cloud server has engaged in a public performance for which an additional license is required. And even if an additional license is not required where the file that is stored in the cloud and then played is licensed, the cloud service provider would be responsible for public performance copyright infringement in the circumstances in which its customers upload and then download unlicensed copies of material--which the cloud service provider cannot possibly police.
Paul Clement, arguing for the broadcasters, had to find a way to say that Aereo is engaged in a public performance but that Dropbox is not. He analogized Aereo to a car dealer--providing new cars--and analogized Dropbox to a valet parking service--merely providing space--and argued further that this distinction was somehow captured by the text of the Copyright Act. He also tried to assure the Court that it could find against Aereo without deciding the cloud question. That assurance was met with skepticism from every Justice who spoke, and Dave Frederick, arguing for Aereo, stoked the Justices' skepticism by offering various arguments for the indistinguishability of Aereo from Dropbox et al.
To my mind, the crucial policy question was posed by Justice Ginsburg. Dissenting from the Second Circuit's ruling for Aereo, Judge Chin wrote the following:
Pursuant to the substance-over-form doctrine, a transaction that serves no economic purpose other than avoidance of some tax will be regarded as taxable. The doctrine doesn't apply everywhere. In particular, sometimes Congress specifically creates unprincipled exceptions that serve more or less as safe harbors; in effect, Congress writes into the Internal Revenue Code particular circumstances where form matters. But in general, the substance-over-form doctrine acts as a background principle of construction.
I would model a parallel principle for intellectual property on tax law's substance-over-form doctrine: Unless Congress has specifically provided otherwise, firms should not be rewarded for "innovating", where the innovation does not serve a technological purpose independent of fitting into some relatively advantageous legal category. As applied to Aereo, if the Court agrees with Judge Chin, the principle would mean that Aereo could be held liable for performance copyright infringement even though cloud service providers would not be. Why? Because (by hypothesis), Aereo's service is indistinguishable from a technologically superior system that would infringe the broadcasters' copyright, which Aereo is only avoiding in order to circumvent copyright; by contrast, cloud computing is socially useful quite independent of the law, as it provides storage backup and convenience for sharing and syncing files across devices.
Just as in tax law, so too in intellectual property law, an anti-circumvention principle can be difficult to apply at the margins. Does some transaction really serve no tax-independent economic purpose? Are there really no advantages of lots of tiny antennae? As the exchange between CJ Roberts and Dave Frederick illustrated, a good lawyer will be able to create doubt about such matters. But that's a reason why anti-circumvention principles are simply one factor in what will sometimes be a complex calculus.
We might also want to think even more broadly about how to structure the law to incentivize useful innovation, rather than rent-seeking innovation. That task will often be pursued best by legislative action. For example, an attentive Congress would write rules of law that distinguish between particular forms of regulation-avoiding innovation and socially useful innovation. And doing so would extend beyond intellectual property.
In that connection, consider one of the leading sub-plots discussed in the Michael Lewis book Flash Boys (about which I blogged at length here). Lewis describes the multi-million dollar construction of a path for a fiber-optic cable between Chicago and northern New Jersey, the sole purpose of which was to shave off tiny fractions of a second in the time it takes to send a signal between the markets at each end, and thus to give an advantage to high-frequency traders (HFTs) using the new line rather than those using the slightly longer, and thus slightly slower, pre-existing line. The new line did not create anything of general value, even for the HFTs themselves, but once it existed, the HFTs needed to pay to use it so that they wouldn't lose out to other HFTs.
Now note that the construction of the straighter line between Chicago and New Jersey did not engineer around legal categories, so in that sense, it doesn't fit within the same anti-circumvention story. But it's part of a still-larger story about how our society goes about encouraging socially useful innovation. It should serve as a counterweight to the standard response of those on Wall Street who reflexively resist regulation by claiming that regulation would stifle innovation. The answer to that response should be: Stifling innovation is only a problem where the stifled innovation would have served to enhance social welfare.
In Professor Hockett's preview of the Aereo oral argument last week, he argued that it would be wonderful if, in considering how to resolve the particular dispute under the Copyright Act, the Justices paid some considerable attention to the broader question of the implications of public ownership of the airwaves. In a related spirit, I want to use the oral argument (transcript now available here) as the launching point for a discussion about how Congress and the courts ought to think about innovation--both in the current context and more broadly.
First, a reminder of the dispute's basics: Aereo has facilities in various cities around the country at which it rents tiny antennae and disk space on an individualized basis, thereby allowing its customers to receive via the internet recorded broadcast programming. As Prof. Hockett explained, the key question in the case is whether Aereo's service amounts to a "public performance", and thus infringes the copyrights of the broadcasters, or amounts to mere time-shifting by the end users, and thus is permissible as not a public performance (pursuant to the doctrine of the 1984 Sony case).
Prompted by the briefing, during the oral argument, Justice Sotomayor raised concerns about the implications of plaintiffs' argument. In particular, she worried that if Aereo's transmission of the recorded signals of the broadcasters from their antenna/dvr farms in Brooklyn etc counts as an unlicensed "public performance" in violation of the broadcasters' copyright, then cloud computing might be in trouble.
Suppose I pay for and download a licensed copy of the Beatles song Across the Universe. If I keep the file on my home computer, I can play it over and over again, without the need for a public performance license, because this is a private performance. However, under the plaintiffs' view in Aereo, if my home computer syncs my files to the cloud (via Dropbox, iCloud, Box, OneDrive, GoogleDrive, or whatever), then if I happen to play Across the Universe from my cloud-based storage rather than from the hard drive of my computer, the cloud server has engaged in a public performance for which an additional license is required. And even if an additional license is not required where the file that is stored in the cloud and then played is licensed, the cloud service provider would be responsible for public performance copyright infringement in the circumstances in which its customers upload and then download unlicensed copies of material--which the cloud service provider cannot possibly police.
Paul Clement, arguing for the broadcasters, had to find a way to say that Aereo is engaged in a public performance but that Dropbox is not. He analogized Aereo to a car dealer--providing new cars--and analogized Dropbox to a valet parking service--merely providing space--and argued further that this distinction was somehow captured by the text of the Copyright Act. He also tried to assure the Court that it could find against Aereo without deciding the cloud question. That assurance was met with skepticism from every Justice who spoke, and Dave Frederick, arguing for Aereo, stoked the Justices' skepticism by offering various arguments for the indistinguishability of Aereo from Dropbox et al.
To my mind, the crucial policy question was posed by Justice Ginsburg. Dissenting from the Second Circuit's ruling for Aereo, Judge Chin wrote the following:
Aereo's "technology platform" is . . . a sham. The system employs thousands of individual dime-sized antennas, but there is no technologically sound reason to use a multitude of tiny individual antennas rather than one central antenna; indeed, the system is a Rube Goldberg-like contrivance, over-engineered in an attempt to avoid the reach of the Copyright Act and to take advantage of a perceived loophole in the law.Justice Ginsburg asked Frederick whether Judge Chin was right. He said no, and offered arguably-plausible-sounding technological reasons for the many antennae rather than one big antenna, but various Justices were skeptical. CJ Roberts, in follow-up questioning, suggested that even if Aereo's technology were designed simply to take advantage of a loophole in the Copyright Act, that Aereo might still win the case, and that seems right to me: sometimes the law pretty plainly forbids X but not Y, even though Y is functionally equivalent to X; in such circumstances, persons and firms act legally when they do Y even though there is no pre-legal reason for doing Y rather than X. But insofar as the law is unclear, it does seem like the courts ought to be guided by an anti-circumvention presumption modeled on the tax law doctrine of substance over form.
Pursuant to the substance-over-form doctrine, a transaction that serves no economic purpose other than avoidance of some tax will be regarded as taxable. The doctrine doesn't apply everywhere. In particular, sometimes Congress specifically creates unprincipled exceptions that serve more or less as safe harbors; in effect, Congress writes into the Internal Revenue Code particular circumstances where form matters. But in general, the substance-over-form doctrine acts as a background principle of construction.
I would model a parallel principle for intellectual property on tax law's substance-over-form doctrine: Unless Congress has specifically provided otherwise, firms should not be rewarded for "innovating", where the innovation does not serve a technological purpose independent of fitting into some relatively advantageous legal category. As applied to Aereo, if the Court agrees with Judge Chin, the principle would mean that Aereo could be held liable for performance copyright infringement even though cloud service providers would not be. Why? Because (by hypothesis), Aereo's service is indistinguishable from a technologically superior system that would infringe the broadcasters' copyright, which Aereo is only avoiding in order to circumvent copyright; by contrast, cloud computing is socially useful quite independent of the law, as it provides storage backup and convenience for sharing and syncing files across devices.
Just as in tax law, so too in intellectual property law, an anti-circumvention principle can be difficult to apply at the margins. Does some transaction really serve no tax-independent economic purpose? Are there really no advantages of lots of tiny antennae? As the exchange between CJ Roberts and Dave Frederick illustrated, a good lawyer will be able to create doubt about such matters. But that's a reason why anti-circumvention principles are simply one factor in what will sometimes be a complex calculus.
We might also want to think even more broadly about how to structure the law to incentivize useful innovation, rather than rent-seeking innovation. That task will often be pursued best by legislative action. For example, an attentive Congress would write rules of law that distinguish between particular forms of regulation-avoiding innovation and socially useful innovation. And doing so would extend beyond intellectual property.
In that connection, consider one of the leading sub-plots discussed in the Michael Lewis book Flash Boys (about which I blogged at length here). Lewis describes the multi-million dollar construction of a path for a fiber-optic cable between Chicago and northern New Jersey, the sole purpose of which was to shave off tiny fractions of a second in the time it takes to send a signal between the markets at each end, and thus to give an advantage to high-frequency traders (HFTs) using the new line rather than those using the slightly longer, and thus slightly slower, pre-existing line. The new line did not create anything of general value, even for the HFTs themselves, but once it existed, the HFTs needed to pay to use it so that they wouldn't lose out to other HFTs.
Now note that the construction of the straighter line between Chicago and New Jersey did not engineer around legal categories, so in that sense, it doesn't fit within the same anti-circumvention story. But it's part of a still-larger story about how our society goes about encouraging socially useful innovation. It should serve as a counterweight to the standard response of those on Wall Street who reflexively resist regulation by claiming that regulation would stifle innovation. The answer to that response should be: Stifling innovation is only a problem where the stifled innovation would have served to enhance social welfare.