Fresh capital expenditure figures for Internet service providers are out, so that means we have another chance to endlessly debate the wisdom of regulating broadband access providers as common carriers! Yay!
The question of the appropriate source of legislative authority with which to regular broadband providers has diverted attention from the more relevant policy question of how to regulate broadband providers. And for that policy question, we have a fairly strong consensus. The sooner we leave the false choice of Title I versus Title II-based rules, the sooner we resolve the net neutrality debate.
An analysis in the Financial Times on Thursday suggests that broadband industry capital expenditures (capex) declined in 2018 by 0.4 percent.
This news was celebrated by proponents of the FCC’s 2015 Open Internet Order, which classified Internet service providers (ISPs) as common carriers. According to their logic, “The findings call into question one of the main arguments the industry used in its successful campaign to have the Federal Communications Commission overturn the Obama-era net neutrality regulations…” Opponents of Title II-based authority but not of net neutrality rules, your fearless blogger included, have argued that capital formation was impaired by the 2015 Order.
But the FT analysis looked at only the top four broadband providers—AT&T, Verizon, Charter and Comcast—and it considered the entirety of their capital expenditures as opposed to their broadband-related expenditures.
A dueling analysis released by trade association USTelecom on the same day suggests that broadband capex across the top six ISPs, including wireless carriers Sprint and T-Mobile, increased by 3.6 percent in 2018. [Disclosure: I have submitted testimony on behalf of USTelecom in several proceedings at the FCC.] Sprint and T-Mobile account for $1.8 billion of USTelecom’s estimated $2.3 billion net increase across the six providers, which implies that even among the same four carriers analyzed by FT, USTelecom estimates a modest increase in capex.
What gives? The discrepancy appears to be that USTelecom excludes the cable programming arm of Comcast (NBCU) and includes AT&T’s expenditure on FirstNet, the carrier’s public-safety broadband network dedicated to police, firefighters, and emergency medical services. While FT notes the sizeable increase in Sprint’s capital expenditures on its networks in the body of its story, this increase curiously was not included in FT’s summary table or headline. I understand that author of the FT report reached out to USTelecom prior to publication, but USTelecom’s survey was not available at that time.
Granted, USTelecom serves as the trade association for carriers with an interest in the fate of Internet regulation. But the benefit of leaning on USTelecom’s survey is that it has a long history dating back to 2011, which allows for consistent comparisons across time. Moreover, in his last speech as Chairman of the FCC, Tom Wheeler cited USTelecom’s survey of broadband capex as the gold standard.
And what is that history? Per USTelecom’s time series, broadband capex across all ISPs (not just the top four or six) declined significantly from 2014 to 2016 (by roughly $3 billion), which coincides with Wheeler’s classification of ISPs as common carriers. In 2017, broadband capex increased by roughly $1.5 billion, which coincides with the repeal of Title II-based regulations. This is hardly proof that Title II caused broadband capex to fall, but the coincidence is very unfortunate for Title II defenders.
Fortunately, there is no special nexus between Title II and net neutrality. It is simply one way among several to authorize the FCC to enforce net neutrality regulations. The DC Circuit in Verizon signaled another way under section 706, which involved putting the burden of proof on edge providers rather than ISPs. And Congress could provide a third way by authorizing the FCC to regulate broadband providers under a new title of the Communications Act.
As former Wheeler policy advisor Gigi Sohn tweeted during the net neutrality hearing last week: “Hey @RepDarrenSoto, any legislation must include, at a minimum, all of the protections of the 2015 Open Internet order & #netneutrality rules. This includes reinstating @FCC authority 2 oversee the broadband market. It can be Title VII, Title X or Title 500.” (emphasis added).
Sohn clearly gets it. And several Republicans on the communications and technology subcommittee signaled strongly that they were ready to codify net neutrality rules outside of the Title II framework. This seems like an important development in the net neutrality wars.
Notwithstanding the views of certain libertarians who think the market (or the FTC’s consumer protection rules) can solve all problems, there is a consensus in the policy community about how to regulate broadband operators: We need bright-line rules to police blatant forms of discrimination, such as blocking and throttling, and we need a standard to deal with milder forms of discrimination that may be good or bad for competition.
Both the 2010 and 2015 flavors of net neutrality regulation reflected this policy consensus. The only difference is that the 2010 Order leaned on a nondiscrimination standard to assess paid priority arrangements, and the 2015 Order used an Internet conduct standard to assess zero-rating plans.
A carrot that can be added to a Title II-free net neutrality compromise is a private right of action. This would allow edge providers who feel like they have been discriminated against to bring a complaint to the FCC’s Administrative Law Judge. That way, not all future enforcement would run through the Chairman’s office—some Chairs might not support the new rules.
It’s long past time to end the net neutrality debate. The faster policymakers remove Title II from the conversation, the faster we will get to the finish line.
After nearly twelve hours of holiday travel—two planes, three airports, tasteless meals, crying babies, vomiting passengers, and incessant announcements in two languages—I can get crotchety.
So it was when I nearly came to blows with a rental car agent in Lisbon last summer, over his refusal to hand over the keys unless I purchase local car insurance. And so it was again when I was refused my Welcome Drink at a fancy resort in Cartagena over Christmas break.
(The name of the fancy resort will remain anonymous in this post because I really like it, and I don’t want this story to reflect poorly on its service.)
During orientation, the hotel manager gave us a coupon for five “Welcome Drinks” by the pool, as a reflection of the hotel’s gratitude for my gold status. A bellhop showed us to our rooms. I was anxious to redeem my coupon, so I left my bilingual but slow-to-unpack-her-bags wife and raced to the pool with the kids.
In addition to a handful of waiters, the hotel employs several men to stand around the pool. They wear a sign on their shirts that reads “Guardia,” which means what it sounds like. It’s Colombia! One of the guards is tasked with entertaining children. And he’s really good at that: His favorite game involves throwing his keys in the pool while the kids face him, eyes covered, and then making the kids search for the keys.
Of course, I interpreted “Guardia” to mean Man Servant and I proceeded to order my Welcome Drink from him.
“Puedo tener cinco bebidas de bienvenidas?” I asked the key thrower. “Dos con alcohol y tres sin alchohol, para los ninos.” Not too shabby for someone who last studied Spanish in El Centro Bilingue in Cuernavaca in the late 1980s.
To remove any ambiguity, I handed the guard my Welcome Drink coupon, he nodded his head, and disappeared. My twelve hours of travel hell was about to be washed away with a yummy mixed drink. Perhaps infused with a local fruit, like guava or papaya or dragon fruit.
Or so I thought.
Ten minutes passed and no one returned with my God Damn Welcome Drinks. I flagged another Guardia and asked about my drinks. He was clearly confused and promised to fetch an English-speaking waiter. Fifteen minutes in and I could feel my blood pressure rising. Keep it together, Hal. The kids are watching.
“What would you like to drink?” the waiter asked.
“My God. I ordered a Welcome Drink fifteen minutes ago. Surprise me!”
“Can you give me your coupon, sir?”
“No, I can’t. I gave it to the Guardia,” I insisted.
“Well, I can’t give you a free drink without the coupon,” the waiter retorted.
That was not the right answer. Not after twelve hours of travel. Not after the vomiting passenger who refused to vomit in the bathroom like a civilized vomiter. “Do I look like I crawled in off the street?” I asked. My seventeen year old moved to another section of the pool.
Taken aback by my temper, the waiter retreated and huddled with the guards. He came back to my chaise lounge. “I can order you a drink, sir. But without the coupon, I must charge you for them.”
“Send me your manager!” I declared. When a good fifteen seconds passed without the manager, I left my two remaining kids and stormed back to the lobby—shoeless but wearing some fresh Tommy Bahamas Bermuda shorts—to confront the manager myself. “Can you please give me another coupon for five Welcome Drinks?,” I asked the manager. “Your wait staff at the pool won’t serve me unless I present a coupon. Apparently, they think I’m lying about the coupon.”
I took a picture of the coupon in the event the staff misplaced it a second time.
When I returned to the scene of the crime, I found my wife speaking in Spanish to the staff. She had ordered herself a drink. Then broke the news to me: “The Welcome Drink is not a drink.”
“Of course it’s a drink!” I was still hot. “We are greeted with Welcome Drinks all the time. They typically have umbrellas. I can’t believe this place!”
“But the welcome drink here is just a coupon.” She dispassionately explained. “You were supposed to order yourself a drink. And when the waiter presents the bill, you present your coupon. Weren’t you listening during our orientation?”
Got me there. I figured it was best to let things die down. I told her, however, that I didn’t want to interface with any staff at this shithole for the rest of our stay. And maybe not at the next hotel either.
I tried to defend myself at the next few meals, but my family showed no sympathy. After a few restful days, some quality Twitter time, and a John Grisham true-crime book, I realized that the problem was not with the staff–and certainly not with me–but instead was with the coupon.
In a low blood-pressure state, I explained to the hotel manager on day three that, in the year 2018, there should be no physical coupons. If you want to reward a loyal customer with a free drink or a free meal, program that credit into the billing system; when the customer checks out, the credit should be applied automatically. I recounted a similar experience at the Grand Hyatt Kauai, which required (at least a few years ago) customers to present a coupon for a free breakfast. I left the coupon in my room, and when the hostess insisted on the coupon as a condition of eating breakfast, I walked back across the property, cursing the Grand Hyatt Kauai, to retrieve it from my room.
The next time a hotel presents you a coupon during orientation, I suggest you tear it into little pieces. Tell the manager that if they want to reward you for your loyalty, they should add a credit to your bill. And maybe listen to what they say during the orientation.
Two final lessons: A Welcome Drink is not always a welcome drink. And bebida de bienvenidas does not mean Welcome Drink. It means you’re an idiot gringo.
My Remarks at the FTC’s Competition and Consumer Protection Hearings: Understanding Exclusionary Conduct in Cases Involving Multi-Sided Platforms, Issues Related to Vertically Integrated Platforms
Dominant tech platforms have the incentive and ability to leverage their platform power into ancillary markets by vertically integrating and then favoring their affiliated content, applications, or wares in their algorithms and basic features. A platform owner should be concerned for the overall health of its ecosystem, which in theory would discourage it from squeezing complementors; but that calculus goes awry when a platform enjoys monopoly power and can take its customers for granted.
Dominant tech platformsalso can exploit the vast amount of user data made available only to them, by monitoring what their users do both on and off their platforms, and then appropriating the best-performing ideas, functionality, and nonpatentable products pioneered by independent providers. If these practices are left unchecked, the resulting competitive landscape could become so inhospitable that independents might throw in the towel, leading to less innovation at the platforms’ edges.
In a recent issue of The Economist, venture capitalists referred to the area around the tech giants in which startups are squashed as the “kill-zone.” Classic examples of new ventures that have flown too close to the sun include Diapers.com,BareBones WorkWear, and Beauty Bridge (in Amazon’s orbit); Foundem.com, TripAdvisor, Shopping.com (in Google’s orbit); and Snapchat, Timehop, and Grubhub (in Facebook’s orbit). A 2017 survey of two dozen Silicon Valley investorssuggests that Facebook’s appropriation of app functionality from edge rivals is “having a profound impact on innovation in Silicon Valley.”
Some new findings are consistent with independents throwing in the towel or not getting funded. Per Crunchbase data, venture investing inside of tech, as measured by the number of deals, has declined since 2015 on average by 23 percent in the U.S. and by 21 percent globally; in contrast, venturing investing outside of techincreased over that same period, suggesting the problem might be tech-specific. And new research using PitchBook data reveals that broadly defined industries in the Amazon/Google/Facebook orbit experienced a collapse in first financings since 2015, a reduction not observed in comparable tech sectors.
There are three basic approaches to dealing with this threat to edge innovation. First, we could lean on antitrust enforcement to police discrimination pursuant to the consumer-welfare standard. Second, we could police these episodes on a case-by-case basis pursuant to a nondiscrimination standard. Or third, we could erect structural barriers via legislation to prevent dominant platforms from annexing ancillary markets.
I am on Team Nondiscrimination, but before I defend its merits, let me briefly discuss the demerits in the approaches of Team Antitrust and Team Structural Relief. The antitrust path leads to underenforcement because judges increasingly interpret the consumer-welfare standard to require demonstration of a tangible, short-run harm to consumers, and because most episodes of discrimination will not produce a price, quality,or output effect. Moreover, the snail’s pace of antitrust adjudication ensures that edge innovation would be dead by the time relief could be administered.
On the other side of the spectrum, structural separation is a messy undertaking—how one draws the boundaries around a platform’s core mission is not straightforward. Not all ancillary offerings require the same level of ingenuity or creativity, and thus not all verticals present the same welfare tradeoffs. Barring Google from incorporating a commodity feature such as answering a math problem, while beneficial to rival math apps, would likely reduce the welfare of users in the short run without any offsetting innovation gain. Finally, structural separation can always be imposed after less invasive, behavioral remedies have been deployed without success.
The problem from an economic perspective is not vertical integration per se. The problem arises when vertical integration is followed by discrimination in a vertical that entails innovative or creative energies—that is, in verticals where the best source of content is likely from independents. Under a nondiscrimination regime, Amazon would be free to sell private-label masks, and Google would be free to collect and attempt to organize its own restaurant reviews.
But as soon as these platforms vertically integrate, they would be subjected to a nondiscrimination standard. This standard would be enforced via a complaint-driven process, initiated by the party alleging discrimination. The standard would prevent Google from limiting its search results for local doctors or local restaurants to Google-affiliated web properties; instead, Google would be required to run its PageRank algorithm across the entirety of the public web for local searches. Under a nondiscrimination standard, a vertically integrated Google could discriminate in its organic search results in everydimension save one—whether the results are affiliated with Google.
An added benefit of my approach is that borrows from the solution to a nearly identical problem concerning vertical integration by a dominant platform in the late 1980s and early 1990s. The dominant platform of that era was owned by cable operators, many of whom vertically integrated into programming. Based on a handful of compelling anecdotes, which revealed the vulnerability of independent cable networks operating at the “edge” of the cable platform, Congress created a specialized, dispute-resolution process that operated outside of antitrust and provided a forum for independent networks to lodge discrimination complaints against vertically integrated cable operators. The protections were not supported by an econometric proof of diminished edge innovation owing to discrimination, but instead were motivated by a simple political preference—that independent networks were an important source of innovation in content and were deserving of protection.
To create similar protections for independent content providers of the Internet era, Congress would have to pass a law with a private right of action. It could give the FTC power to resolve these matters administratively, or private parties could develop federal common law on this issue by trying cases before Article III judges.
So, I have a modest proposal: The FTCshouldpursue a Section 2 case against a tech platform when the harms manifest as a price, output, or quality effect. In the absence of a tangible consumer injury, the FTC couldpursue a Section 5 case by treating discrimination as an unfair practice. In any event, at the end of its competition hearings, the FTC should ask Congress to give the agency a new source of authority to adjudicate complaints against vertically integrated tech platforms pursuant to a nondiscrimination standard. The FTC already has an administrative law judge. Now it just needs a mandate from Congress and some complaints to protect edge innovation.
See, e.g., Feng Zhu & Qihong Liu, Competing with Complementors: An Empirical Look at Amazon.com, Strategic Management Journal(forthcoming 2018) (finding that “affected” sellers on Amazon’s platform, against which Amazon competes directly, reduce the number of products offered on Amazon by 24.1 percent relative to unaffected sellers); Wen Wen & Feng Zhu, Threat of Platform-Owner Entry and Complementor Responses: Evidence from the Mobile App Market (October 2017). Harvard Business School Working Paper No. 18-036 (finding that prior to Google’s entry, the affected app developer, against which Google competes directly, reduces updates on an affected app by 5 percent, and increases the prices of affected apps by 1.8 percent when the entry threat increases; once Google enters, the affected developer reduces updates on the affected app by 8 percent and increases the prices of affected apps by 3.6 percent, consistent with entry accommodation).
An exception to this general rule is when a search platform degrades its search results so as to promote its affiliated content. See, e.g., Michael Luca, Sebastian Couvidat, William Seltzer, Timothy Wu, and Daniel Frank, Does Google Content Degrade Google Search? Experimental Evidence, Harvard Business School Working Paper 16- 035 (finding that hen Google was induced to revert back to its organic search results, the rankings of competing independent properties were elevated in Google’s search, and users were 40 percent more likely to engage with the search results, as measured by click activity).
The prospect of a conservative ideologue such as Judge Brett Kavanaugh—Donald Trump’s second nominee for the Supreme Court—exposes the increasing politicization in our appellate system. Through the Executive and Congress, we already have two layers of politics in our legal system. We don’t need a third.
When considered alongside Neil Gorsuch, Trump’s judicial nominees appear to be selected, not on some technocratic standard, but instead on the nominee’s willingness to abide by the policy preferences of the Federalist Society—hardly a recipe for an impartial jurist.
This isn’t meant to imply that Trump started the politicization of the federal courts or that it’s comparatively recent phenomenon. In the last half-century in judicial nominations, both parties have dirty hands, and for a long time it’s been a very cynical game of realpolitik. Both parties would actually like to find some way out of this deadlock, but they are locked in a terrible, extremely difficult stalemate, in which the worst problem is the purely strategic logjam; they know they’ve broken the federal judiciary, but they also know that any solution that wouldn’t give one side an unacceptable advantage seems unavailable.
There is no place for an ideologue in the appellate system
To understand how to break through this logjam, it’s worth reviewing the basic rules and purpose of an appeal. Appeals are meant to give the losing party at the trial court an opportunity to correct bias in fact-finding and the misapplication of the law to the established facts. Everyone has a right of appeal from a district court decision, as there needs to be a place to go with runaway juries, testimony flaws missed, or wrong application of the law.
Deference is historically and routinely given to the fact-finding of assumed unbiased lower-court judges, who observe firsthand the demeanor of “fact witnesses” and “expert witnesses” as they testify confidently, hesitantly, or inconsistently, to introduce facts and expert opinion. The appeal should not be a place where political ideology—from either side—trumps a dispassionate reading of the facts by the trial court judge as they apply to the law.
If only Kavanaugh showed such deference to the original factfinder. In one opinion overturning the FCC’s finding that a vertically integrated cable operator violated the Cable Act’s nondiscrimination standard, Kavanaugh wrote that, because Congress used a single antitrust phrase in a regulation meant to plug a gap in antitrust enforcement, the higher antitrust evidentiary standard—and not the nondiscrimination standard—should apply.
Kavanaugh has has proven through his competition-related opinions in Anthem-Cigna, Whole Foods-Wild Oats, and Comcast v. FCC, that he is willing to play fast and loose with facts and science, disregarding the record below, to achieve an outcome that suits his free-market ideology. This is the opposite of deference to the original factfinder, and it perversely introduces bias in a process designed to achieve the unbiased pursuit of justice. (Disclosure: I was Tennis Channel’s expert in the discrimination case against Comcast.)
Kavanaugh’s judicial demeanor in this regard represents an abuse of the appeals process from a good-governance perspective. It is, ironically, the use of judicial activism in pursuit of a conservative agenda. And like most governmental abuses, it can be adopted by the opposition when it has power, leading to a government that is no longer ordered by normative behavior. The loser in such a tit-for-tat game, no matter who is in power, is the American citizen.
Kavanaugh is a symptom of a larger problem relating to the appeals system
Although Judge Kavanaugh was not involved in the recent American Express antitrust litigation, the appeal turned on a disregard of settled facts by the Second Circuit in much the same manner as in Anthem-Cigna and Whole Food-Wild Oats; this occurred in American Express despite the district court judge’s having the benefit of hearing all the testimony regarding the intricacies of the credit card market. It’s the equivalent of displacing the first base umpire’s unobstructed view of a bang-bang play at first base with the perspective of commentators sitting in the stands who hold preexisting views about whether the runner should have been safe or out given his running style or where he hit the ball.
For the non-ideological slice of Americans in the political center, we are not rooting for plaintiffs or defendants in any particular case. The optimal outcome for those without a political dog in the fight is that meritorious cases prevail and weak cases fail. This is achieved by having an unbiased, technocratic fact-finder, dutifully apply the law to the facts at hand.
If hyper-ideological appellate judges such as Kavanaugh merely serve to inject a layer of politicization into what should be an apolitical inquiry, what can be done?
Weakening the grip of the ideologues
For starters, we should tighten the requirements for reversal. Under current law, a successful appeal requires merely two of three reviewing judges to see things according to the losing party. In today’s hyper-political climate, the number two seems too low. In particular, the need to compromise under the current “win-by-one-vote” regime is exceedingly weak, leading to extremist views. Assuming that the number of ideologues is a minority of the judiciary, ideology could get watered down a fair bit just by having bigger panels.
Consider a hypothetical appellate circuit of nine judges, two of which are conservative ideologues who are willing to overturn any decision—regardless of the fact pattern—in favor of an antitrust plaintiff. For a given random panel of three judges, the two ideologues will be selected (and thus have a voting majority) with a probability of 8.33 percent. There are three paths to get the two ideologues on the three-judge panel: (1) miss, hit, hit; (2) hit, hit, miss; and (3) hit, miss, hit. The probability of each path is 2.78 percent. Summing the probabilities across the three paths yields 8.33 percent. That probability might seem acceptably low, but it means that nearly ten percent of pro-plaintiff decisions will be reversed in such circuits for no reason other than politics.
To give more deference to the original fact-finders, appeals should require the consent of three of five or even four of seven judges. This would reduce the chances of drawing a majority of ideologues. A larger appellate bench will also assure claimed legal errors are based on mainstream interpretations of statutes and the Constitution, rather than readings from the fringe.
One might be concerned that this higher burden for reversals would make it more difficult to reverse ideologically driven mistakes at the district-court level. But the trial judge can only be as much of an ideologue as her controlling appellate court opinions allow. The trial judge will not, as a rule, overtly cross the appellate court’s views, and if she does, she’ll get reversed. So the chance of having a rogue district court judge having the opportunity to mess up a valid case via ideology is not only rarer compared to an appellate court, but also subject to reversal as a matter of right. In contrast, the Supreme Court provides much less guidance to the appellate courts because there is no appeal as of right from them, and because the Supreme Court handles fewer cases.
Another potential concern is that larger panels would entail greater resource costs in the form of additional federal judicial personnel. The courts already claim that they are overwhelmingly overworked, but Congress stubbornly refuses to add more judges. Under the current regime, adding more judges gives one’s opponents more seats that they can fill. To the extent that more judges per panel would depoliticize appeals, however, it might not be so hard to convince Congress just to add more judges.
Under this higher burden for reversals, fewer district court decisions would be overturned, but it would not change the opportunity to launch an appeal. If the appeals courts were so constrained by this higher standard, then the stakes for presidential appointments would be lowered considerably. “Call-ups” from the Circuit courts to the Supreme Court could be based on technical brilliance, not on one’s political ideology. Courts could emerge as an unbiased but supple interpreter of the laws handed down by Congress.
Stopping Kavanaugh would be a good thing. But it wouldn’t solve the politicization of our appellate courts. If Trump wants to stack the courts with ideologues, the best protection for liberals is to reduce an ideologue’s power to inject his or her politics into judicial rulings. Recognizing that they could soon be the party out of power, conservatives just might go along with such a compromise.
The Comcast-NBCU protections served an important objective when implemented: The protections sought to fill a gap between (a) what was needed for online video distributors (“OVDs”) and independent networks, and (b) existing protections in program access and program carriage protections under the Cable Act. New protections included, but were not limited to, a provision that allowed OVDs to get access to Comcast-affiliated online programming at fair market value, as well as protections for independent news networks such as Bloomberg to be placed in the same neighborhood as NBCU-affiliated news programming. Considering the concept of online video and independent news networks were at best nascent when the Cable Act was written, such provisions are better viewed as plugging an existing statutory hole rather than addressing a need unique to the transaction.
The Comcast-NBCU protections are still important. These incremental protections are needed today because the gap between (a) and (b) still exists. Combining NBCU’s must-have programming with Comcast’s large in-region video distribution shares created fresh opportunities for Comcast to discriminate against rival distributors and independent cable networks. Nothing in the competitive landscape—including Netflix’s successful OVD offering and Facebook’s soon-to-be launched online video service, Watch—has caused this gap to disappear. Comcast still has very large video distribution shares (although shrinking in a mobile first world), and NBCU programming is still must-have, particularly the localized content available through the owned-and-operated broadcast affiliates.
There are two ways to ensure that this gap between (a) and (b) remains covered. First, and most obviously, the Comcast-NBCU protections could be extended. Alternatively, the program access and program carriage protections can be strengthened and extended to cover broadband distribution (e.g., extending the forum to permit OVDs to bring access complaints, and to permit content providers/websites to bring carriage complaints against ISPs); under the second approach, the merger-specific behavioral remedies would no longer be needed. The interesting policy questions are: Which of these two paths makes the most sense? And which is most feasible to achieve in the short run?
Regarding feasibility, I am skeptical that DOJ’s new antitrust chief, Makan Delrahim, would extend behavioral remedies in light of his campaign to end the agency’s reliance on “regulation,” as he refers to it, as well as his belief that behavioral remedies have been ineffective. I respectfully happen to think he is wrong, on both scores.
First, under the NBCU protections, complainants were two and an half out of three: Bloomberg secured a favorable neighborhood decision on Comcast’s channel lineup; an unnamed OVD secured access to Comcast-affiliated content per the DOJ’s annual report; and Project Concord—disclaimer, my client—won a favorable ruling from an arbitrator that its offer price was closer to fair market value than was Comcast’s. That’s somewhere between a .667 and .833 batting average, which would get you to the All Star game. (To be fair, Comcast has appealed the Concord decision, which has delayed the requested relief, and I will return to this correctable defect in the post-adjudicatory process later on.) Outside of the context of the NBCU protections, NFL Network and MASN—disclaimer, my clients—have used the program carriage protections to secure favorable carriage on Comcast’s platform. (Tennis Channel, also my client, secured a favorable ruling on discrimination from the FCC’s ALJ, only to lose on appeal at the DC Circuit.)
Second, Delrahim’s claim that adjudication of these cases imposes large costs on the DOJ or FCC is simply false; a single ALJ or arbitrator (plus her legal assistant) is required. Notwithstanding that he is wrong on both scores, I have little hope of persuading Delrahim of the efficacy of behavioral remedies to address discrimination by a vertically integrated platform provider. This suggests that strengthening and extending the program access and program carriage provisions, despite requiring an intervention by Congress, might be the more feasible path to filling the aforementioned gap in statutory protection remedied by the Comcast-NBCU protections.
In short, if we look at the facts, rather than the popular narrative promoted by Delrahim’s office (a diversionary tactic used by other arms of this Administration and broadly criticized by many in this room), we see that the Comcast-NBCU gap filler prompts few complaints, and those that require adjudication place less demand on agency resources than, say, a weekend trip to Florida causes other arms of the Executive Branch.
Regarding policy considerations, I believe that regulating vertically integrated platforms via generalized rules (flowing from Congress, then promulgated by agencies) is better than merger-specific patches. The reasons are myriad. First, merger-specific patches expire, which is what brings us here today. Second, merger-specific patches do not cover other vertically integrated distribution platforms; if a non-merging vertically integrated cable operator were to withhold affiliated content from an OVD, or were to place Bloomberg in a non-news neighborhood, there is currently no forum in which the target of the discrimination could lodge a complaint. Third, merger-specific regulation encourages rent seeking during the merger-review process; merger opponents can extract a concession that may not be justified on cost-benefit grounds, and (Tunney Act objections aside) may not flow from the merger theory of harm. (A good example was the obligation for Charter to invade a rival’s territory in the Charter-Time Warner merger order.) Fourth, regulations that applied to any distributor that elects to vertically integrate—the logic behind the program access and carriage protections—would serve as a fresh deterrent to vertical integration (a “tax” on vertical integration); after all, vertical integration is what gives rise to the discriminatory impulse. But unlike a bright-line rule against vertical integration, general protections would allow procompetitive integration while policing it.
Finally, strengthening and expanding the scope of the program carriage and program access protections leads to two interesting follow-on topics: (1) nondiscrimination protections for broadband start to look a lot like net neutrality protections, which could be codified into law via a new (broadband-specific) title of the Communications Act; and (2) if the protections are extended to broadband, and if non-cable firms such as Amazon, Google and Facebook move into online video, shouldn’t such dominant tech platforms be subjected to the same regulations?
A final thought: Given that the purpose of the protections would be to preserve edge innovation in a layer-neutral way, it is critical that relief for meritorious cases arrives quickly; online platforms would preserve their rights to appeal an adverse decision, but a finding of discrimination by an ALJ (or arbitrator or the “Net Tribunal”) would (typically, using a balance of the harms test) be met with an immediate remedy—the end of discrimination—which would last throughout the appeal process.
Imagine if this conversation led us down a path of solving the “platform privilege problem” and net neutrality all at once!
My Comments on Delrahim’s Speech at the University of Chicago’s Antitrust and Competition Conference
There is something—how can I put this gently—incoherent in Makan Delrahim’s views on antitrust, as evidenced by his speech at the University of Chicago. His opening shout-outs to Robert Bork and the Chicago School, which largely neutered antitrust enforcement in single-firm conduct cases based on a hidden assumption that supported the single-monopoly-profit theory, can be chalked up to currying favor with the home crowd. But the cognitive dissonance that follows is unmistakable.
On the one hand, he pays lip service to the leaders of the New Brandeis movement, citing by name Lina Khan (for offering an “interesting note” and “diverse thinking”) and Franklin Foer (credited for characterizing data as the “new oil”) to note their important contributions to our collective understanding of the threats posed by platform providers that currently escape antitrust scrutiny. On the other hand, Delrahim rejects the New Brandeis’s teachings that industry concentration, spurred by under-enforcement, has led to “rising economic inequality, stagnant wages, unemployment, and a concentration in political power in the hands of private market actors.” He insists that for platform markets, “[a]ntitrust enforcers may need to take a close look to see whether competition is suffering and consumers are losing out on new innovationsas a result of misdeeds by a monopoly incumbent.” Yet he vehemently defends the consumer-welfare standard, using an “evidence-based” approach that has demonstrably proven blind to innovation harms. Delrahim wants to tap into the populist wave, but simultaneously rejects where the populists want to take us.
This brief comment identifies four errors in Delrahim’s understanding of antitrust and consumer protection.
First, Delrahim reveals a misplaced faith in markets to solve the privacy problem. Relying on our newly “revealed preference” for privacy and a hyper-rational homo economicus, he asserts that consumers “may opt out of certain networks because they simply don’t see the bargain as working in their favor.” But this misunderstands the fallacies of human beings, which is to fall prey to the “status quo bias” and accept the default settings of their preferred data vendors, thereby permitting massive data exploitation. These newly revealed tastes for privacy will spur profit-maximizing firms, in Delhahim’s market nirvana, to solve the data exploitation themselves—a form of self-regulation—via new and innovative offerings: “As a result, the overwhelming incentives to design technologies to maximize the collection and use of personal information may be shifting, and with that shift companies are designing technologies that respond to our revealed preferences for privacy.” This worldview ignores Facebook’s business model as well as those of other digital platforms peddling targeted advertisement—namely, to surveil users and monetize their every movement across the web.
Second, Delrahim ignores how an “evidence-based” approach to antitrust would turn a blind eye to anticompetitive conduct that harmed innovation absent price or output effects. He insists that “[t]aking an evidence-based approach to antitrust law should not be mistaken for an unwillingness to bring enforcement actions.” Actually, when an econometrician cannot empirically establish harm to innovation that manifests itself in the form of an entrepreneur throwing in the towel because she perceives the playing field to be uneven, a rigid evidentiary requirement of short-run harm to consumers is tantamount to no enforcement. Without any sense of irony, Delrahim states “In the context of digital platforms, an evidence-based approach is critical to protecting innovation.” This is not to say that antitrust standards should be bent to accommodate innovation-based theories of harm. I’m a proponent of policing exclusionary conduct by a vertically integrated platform provider with a non-discrimination standard outside of antitrust. But at minimum, Delrahim should acknowledge the gap in antitrust enforcement here and in labor markets (described below) created by strict adherence to the consumer-welfare standard. Yet he can’t conceive of a single deficiency under the antitrust status quo. He claims that the “The Microsoft case proved that an evidence-based antitrust enforcement approach can be flexible in its application to new types of assets and markets.” Fair enough, but why haven’t antitrust enforcers brought a pure innovation-based theory of harm case since Microsoft two decades ago? The likely answer is that the evidentiary requirements under the consumer welfare standard make such a showing impossible. (Google’s discriminatory search algorithm, which steers users to its own properties and away from independent content creators, is an example of a pure-innovation theory of harm, in the sense that prices for users or advertisers are not affected.)
Third, Delrahim also fails to recognize that the consumer welfare standard misses monopsony harms. He claims that “No available data demonstrate increasing concentration of markets, as economists and antitrust enforcers define them.” Citing a new piece by Carl Shapiro, DOJ’s expert in the AT&T-Time Warner case, Delrahim argues that concentration at the industry level might not correspond to concentration in some relevant antitrust market. Delrahim apparently forget this important caveat from Shapiro’s article:
Nothing in this section should be taken as questioning or contradicting separate claims regarding changes in concentration in specific markets or sectors, including some markets for airline service, financial services, health care, telecommunications, and information technology. In a number of these sectors, we have far more detailed evidence of increases in concentration and/or declines in competition.
Even if none of these increases in concentration identified by Shapiro (that Delrahim ignores) were correlated with increases in price or price-cost margins in some relevant output market, so long as concentration has been correlated with wages or wage shares in a relevant input market, then Delhahim’s claim of no enforcement gaps under the consumer welfare standard is false. He ignores new evidence by Autor et al. (2017) (finding that each percentage point rise in an industry’s concentration index predicts a 0.4 percentage point fall in its labor share); Azar et al (2017) (finding that move from the 25th percentile of labor market concentration to the 75th percentile would lower (advertised) pay level in a metro area by 17 percent); and Barkai (2016) (estimating that If competition increased to levels last observed in 1984, wages would increase by 24 percent). Yet Delrahim says he is “skeptical that competition policy can be isolated as a contributing cause to any of these wider trends” in the economy, including “stagnant wages.” Despite the huge attention monopsony has received in antitrust circles, Delrahim’s speech mentions monopsony only once, and even there it is in the context of a user selling her data or “digital currency” to Facebook. Workers be damned.
Fourth, Delrahim thinks the biggest mistake of antitrust enforcement is the reliance on behavior remedies in enforcement actions, rather than the more obvious deficiency—namely, a failure to enforce the antitrust laws at all in certain cases: “Enforcers do indeed deserve some blame, particularly for their willingness to settle for ineffective behavioral (or regulatory) fixes to mergers rather than challenging them when necessary.” In what world can the FTC’s failure to bring a case against Google in countless would-be merger or exclusion cases be considered less important than the DOJ’s willingness to embrace a behavioral condition in Comcast-NBCU? This gratuitous attack on behavioral remedies comes of out nowhere—recall the speech is concerned with defending the consumer welfare standard—and appears to be a backhanded defense of the DOJ’s lawsuit to block the transfer of CNN and other Time Warner media properties to AT&T. It also flies in the face of the DOJ’s 2011 remedies guidelines, which note that “[in certain circumstances, depending on what information is available regarding competitive prices in the relevant market, the Division will consider employing a non-discrimination clause requiring the upstream firm to offer the same terms to all three downstream competitors.”
My advice to Delrahim is to open a formal channel to the community of “diverse thinkers” he acknowledges in his talk. He doesn’t have to embrace New Brandeis hook, line, and sinker. But he needs to find a coherent view of antitrust that recognizes some of the clear gaps in enforcement left unremedied by the dutiful allegiance to the consumer welfare standard.
(He should also steer clear of pronouncing to the media “We’re not going to lose” in any ongoing trial, particularly when financial markets and experts think otherwise. But that’s the subject of another blog.)
Let’s divorce the source of legal authority and the optimal net neutrality policies for a moment.
With regard to policy considerations, we had a fairly savvy political and economic compromise in the FCC’s 2010 Order that effectively dealt with the two forms of net neutrality discrimination: It used rules against blatant forms of discrimination (blocking, throttling), and it applied standards to police mild forms of discrimination (at the time, paid priority). Economists prefer standards to rules whenever conduct can be motivated for efficiency reasons. And so should you! Alas, the DC Circuit largely took away the FCC’s legal authority used in support of the 2010 compromise.
The remaining source of legal authority (Title II) is not palatable to the Tech Right. Indeed, it is not palatable to most economists, as it exposes Internet service providers (ISPs) to regulations that were never intended to address net neutrality concerns. So thanks to the DC Circuit, we are in a regulatory hellhole in which Title II-based net neutrality rules (Obama’s 2015 order replaced the 2010 standards with rules) will be repealed with effectively no protections whenever a Republican occupies the White House. This cannot be acceptable to anyone who truly cares about net neutrality.
The ISPs have strongly signaled their willingness to go back to the 2010 compromise. They have offered to live by rules against blocking and throttling, and to abide by standards for the mild forms of discrimination such as zero rating.
The solution to this dilemma is simple: Congress needs to create a new source of legal authority to allow the FCC to do effectively what it wanted to do in the 2010 order. Namely, apply rules against the blatant forms of discrimination, and apply (nondiscrimination) standards to the mild forms.
My only tweak to the 2010 Order would be to move from the guilty-before-proven-innocent presumption to an innocent-before-proven-guilty presumption when adjudicating disputes under the standard—that is, put the burden of proof on the complainant, as we do for program carriage disputes. But this is a detail that can be hammered out. We could even use a burden shifting regime, in which the complainant has to meet an initial evidentiary standard, after which the burden flips to the ISP.
Now I recognize that moving from Title II-based net neutrality rules—with disruptions every time the GOP takes the White House—to a 2010 framework grounded in a new source of authority might be considered a loss for some on the Tech Left. So I propose two left-leaning “sweeteners” to make this compromise go down a bit easier:
(A) Attach opt-in privacy protections to the bill and have them apply to both ISPs and edge providers symmetrically; and
(B) Extend the ambit of the nondiscrimination standard to include conduct by a dominant tech platform (e.g., Google, Amazon, Facebook).
There is growing sentiment on both sides of the aisle that the dominant tech platforms threaten edge innovation (and our privacy) to the same degree or even more than do ISPs. It makes no sense to have two sets of standards and two sets of regulators for identical threats to edge innovation that emanate from two different layers of the Internet ecosystem. We should aspire to have a layer-neutral approach to regulation.
I also recognize that bringing tech platforms into the ambit—by attaching symmetric privacy and nondiscrimination standards—raises tricky issues of jurisdiction. Should the FCC be allowed to police bad acts by tech platforms? But this is a healthy discussion. And it is more important to get the policy right than to fight about which regulator should enforce the policies. My preference would be to port this policing/adjudication of disputes to a different agency, or even to a tribunal that operates outside of an agency (so as to remove any influence of politics during the appeal of the tribunal’s decision on a particular case). Others might prefer to see the scope of the FCC broadened. But again these are details to be hammered out.
Let’s end this net neutrality quagmire once and for all!