[I have posted my opening remarks from the Great Net Neutrality Debate, which can be viewed here.]
An online telemedicine provider, an online chocolatier, and an ISP walk into a bar. The chocolatier orders some red wine, naturally, and proceeds to get drunk. There’s no purpose of the joke, other than to introduce my cast of characters for a compelling reason why we shouldn’t ban paid priority on the Internet.
In my hypothetical, the telemedicine provider is not content with the best efforts offered by the broadband providers that serve his network of consulting doctors. He wants his packets to get special treatment so that a consulting doctor who is remotely monitoring a life-saving procedure doesn’t have to tolerate a second of jitter. He doesn’t want the same treatment afforded a cat video, and neither do his doctors or patients.
My friend Jonathan wants to step in between that voluntary and highly beneficial arrangement. If he had his way, the telemedicine provider and the ISP either would be forbidden from striking such deals, or would have to go on bended knee to some bureaucrat’s office in the FCC and ask for a special exemption from the blanket ban to innovate this way. If we permitted any such deals and policed them on a case-by-case basis, Jonathan asserts, too many harmful deals would slip past our filter, and innovation would somehow grind to a halt. In the parlance of antitrust, a rule of reason approach like that called for under the commercially reasonable standard would, in his mind, generate too many “false negatives.”
Now clearly no one is harmed in my hypothetical telemedicine arrangement. The patient wins, the app provider wins, and God forbid the ISP wins a little extra revenue. We tried asking the online chocolatier if he cared about the telemedicine guy buying priority, but he was puking in the bathroom.
Jonathan will likely say: “But Hal’s telemedicine example is not representative of the kinds of deals that would be struck under a more permissive case-by-case regime.” But what if Jonathan’s wrong? What if a non-trivial percentage of such deals were done for the right reason? Then Jonathan’s per se rule against priority would generate an unacceptably high amount of what antitrust calls “false positives.”
While Jonathan and I can agree that we don’t want to lean on antitrust to replace the FCC here, we can borrow some important lessons from antitrust when it comes to choosing between a per se prohibition and a rule of reason framework. And here is the first lesson: When the Court can conceive of plausible efficiency motivations—as it did recently in Actavis and Leegin—the Court subjects such conduct to the rule of reason.
Contrast those cases with how antitrust treats price fixing, with a per se prohibition. Why? Because we can’t conceive of any way in which consumers could benefit when suppliers get together and fix prices. When conduct is subject to a per se rule, the court won’t even entertain efficiency defenses—mere evidence that the conduct occurred is sufficient to violate the law. Let’s look at the first of two exhibits.
For Jonathan to win today, he needs to convince you that paid priority can be analogized to price fixing—that is, there can be no compelling efficiency justification for priority. Don’t think about the telemedicine provider, he insists, whose application can’t function properly without priority. Relative to his proposed blanket prohibition on priority, a rule of reason approach can reduce the number of false positives while not permitting the number of false negative to rise—like the whack a mole, we can zap bad deals as they pop up.
Now the second exhibit. When choosing a case-by-case standard, which the FCC did in its 2010 order, you have to decide whether priority is presumptively in violation of the standard or not. Where you set the presumption dictates which party has the initial burden of proof. For two reasons, the optimal default rule here is to presume at least initially that a priority deal is not in violation of the new standard.
First, the D.C. Circuit told the FCC in January that it can’t go down the left path under its 706 authority, as it tried to do in the 2010 Order. This doesn’t mean that Title II is guaranteed to end priority deals either, as Berin will explain, but it does take section 706 off the table, leaving us with the dreaded Title II.
The second reason for assigning the initial burden this way is that edge providers are in the best position to know if they’ve been harmed.
Now Jonathan argued at the FCC Roundtable that placing the burden on edge providers will permit the discrimination to go on too long before any relief has been achieved, and that small websites cannot afford the litigation expenses associated with prevailing under a full blown rule of reason standard.
Here we can lean on antitrust once again. Antitrust courts are increasingly using bright-line tests to streamline the inquiry: If a plaintiff can marshal evidence to trigger a bright-line test, then the initial burden shifts quickly to defendants to rebut the updated presumption that the conduct is anticompetitive. For example, in Actavis, the Court offered a bright-line test to help identify anticompetitive settlements between brands and generics: If plaintiffs can show that the payment associated with the settlement is unexplained, then the burden shifts and “defendants have some explaining to do.”
Applied here, the FCC could establish a bright line test to work in conjunction with the commercially reasonable standard. For example, if a complaining edge provider could show either (1) it suffered a degradation in service by refusing to accept an ISP’s priority offering, or (2) it was denied access to the same terms for priority that were extended to its similarly situated edge rival, then the burden could shift to the ISP to explain what in the heck was going on.
For the foregoing reasons, we should not ban paid priority. Instead, they should be subjected to a rule of reason approach, in which bright-line rules can be used to reduce false negatives, expedite the process, and provide sufficient protection to nurture innovation at the edge of the network, which preserving incentives for ISPs to continue investing at the core.