[TOTM: The following is part of a digital symposium by TOTM guests and authors on the legal and regulatory issues that arose during Ajit Pai’s tenure as chairman of the Federal Communications Commission. The entire series of posts is available here.
Randy May is president of the Free State Foundation.]
I am pleased to participate in this retrospective symposium regarding Ajit Pai’s tenure as Federal Communications Commission chairman. I have been closely involved in communications law and policy for nearly 45 years, and, as I’ve said several times since Chairman Pai announced his departure, he will leave as one of the most consequential leaders in the agency’s history. And, I should hastily add, consequential in a positive way, because it’s possible to be consequential in a not-so-positive way.
Chairman Pai’s leadership has been impactful in many different areas—for example, spectrum availability, media deregulation, and institutional reform, to name three—but in this tribute I will focus on his efforts regarding “net neutrality.” I use the quotes because the term has been used by many to mean many different things in many different contexts.
Within a year of becoming chairman, and with the support of fellow Republican commissioners Michael O’Rielly and Brendan Carr, Ajit Pai led the agency in reversing the public utility-like “net neutrality” regulation that had been imposed by the Obama FCC in February 2015 in what became known as the Title II Order. The Title II Order had classified internet service providers (ISPs) as “telecommunications carriers” subject to the same common-carrier regulatory regime imposed on monopolistic Ma Bell during most of the 20th century. While “forbearing” from imposing the full array of traditional common-carrier regulatory mandates, the Title II Order also subjected ISPs to sanctions if they violated an amorphous “general conduct standard,” which provided that ISPs could not “unreasonably” interfere with or disadvantage end users or edge providers like Google, Facebook, and the like.
The aptly styled Restoring Internet Freedom Order (RIF Order), adopted in December 2017, reversed nearly all of the Title II Order’s heavy-handed regulation of ISPs in favor of a light-touch regulatory regime. It was aptly named, because the RIF Order “restored” market “freedom” to internet access regulation that had mostly prevailed since the turn of the 21st century. It’s worth remembering that, in 1999, in opting not to require that newly emerging cable broadband providers be subjected to a public utility-style regime, Clinton-appointee FCC Chairman William Kennard declared: “[T]he alternative is to go to the telephone world…and just pick up this whole morass of regulation and dump it wholesale on the cable pipe. That is not good for America.” And worth recalling, too, that in 2002, the commission, under the leadership of Chairman Michael Powell, determined that “broadband services should exist in a minimal regulatory environment that promotes investment and innovation in a competitive market.”
It was this reliance on market freedom that was “restored” under Ajit Pai’s leadership. In an appearance at a Free State Foundation event in December 2016, barely a month before becoming chairman, then-Commissioner Pai declared: “It is time to fire up the weed whacker and remove those rules that are holding back investment, innovation, and job creation.” And he added: “Proof of market failure should guide the next commission’s consideration of new regulations.” True to his word, the weed whacker was used to cut down the public utility regime imposed on ISPs by his predecessor. And the lack of proof of any demonstrable market failure was at the core of the RIF Order’s reasoning.
It is true that, as a matter of law, the D.C. Circuit’s affirmance of the Restoring Internet Freedom Order in Mozilla v. FCC rested heavily on the application by the court of Chevron deference, just as it is true that Chevron deference played a central role in the affirmance of the Title II Order and theBrand X decision before that. And it would be disingenuous to suggest that, if a newly reconstituted Biden FCC reinstitutes a public utility-like regulatory regime for ISPs, that Chevron deference won’t once again play a central role in the appeal.
But optimist that I am, and focusing not on what possibly may be done as a matter of law, but on what ought to be done as a matter of policy, the “new” FCC should leave in place the RIF Order’s light-touch regulatory regime. In affirming most of the RIF Order in Mozilla, the D.C. Circuit agreed there was substantial evidence supporting the commission’s predictive judgment that reclassification of ISPs “away from public-utility style regulation” was “likely to increase ISP investment and output.” And the court agreed there was substantial evidence to support the commission’s position that such regulation is especially inapt for “a dynamic industry built on technological development and disruption.”
Indeed, the evidence has only become more substantial since the RIF Order’s adoption. Here are only a few factual snippets: According to CTIA, wireless-industry investment for 2019 grew to $29.1 billion, up from $27.4 billion in 2018 and $25.6 billion in 2017. USTelecom estimates that wireline broadband ISPs invested approximately $80 billion in network infrastructure in 2018, up more than $3.1 billion from $76.9 billion in 2017. And total investment most likely increased in 2019 for wireline ISPs like it did for wireless ISPs. Figures cited in the FCC’s 2020 Broadband Deployment Report indicate that fiber broadband networks reached an additional 6.5 million homes in 2019, a 16% increase over the prior year and the largest single-year increase ever.
Additionally, more Americans have access to broadband internet access services, and at ever higher speeds. According to an April 2020 report by USTelecom, for example, gigabit internet service is available to at least 85% of U.S. homes, compared to only 6% of U.S. homes three-and-a-half years ago. In an October 2020 blog post, Chairman Pai observed that “average download speeds for fixed broadband in the United States have doubled, increasing by over 99%” since the RIF Order was adopted. Ookla Speedtests similarly show significant gains in mobile wireless speeds, climbing to 47/10 Mbps in September 2020 compared to 27/8 Mbps in the first half of 2018.
More evidentiary support could be offered regarding the positive results that followed adoption of the RIF Order, and I assume in the coming year it will be. But the import of abandonment of public utility-like regulation of ISPs should be clear.
There is certainly much that Ajit Pai, the first-generation son of immigrants who came to America seeking opportunity in the freedom it offered, accomplished during his tenure. To my way of thinking, “Restoring Internet Freedom” ranks at—or at least near—the top of the list.
Near the end of her new proposal to break up Facebook, Google, Amazon, and Apple, Senator Warren asks, “So what would the Internet look like after all these reforms?”
It’s a good question, because, as she herself notes, “Twenty-five years ago, Facebook, Google, and Amazon didn’t exist. Now they are among the most valuable and well-known companies in the world.”
To Warren, our most dynamic and innovative companies constitute a problem that needs solving.
She described the details of that solution in a blog post:
First, [my administration would restore competition to the tech sector] by passing legislation that requires large tech platforms to be designated as “Platform Utilities” and broken apart from any participant on that platform.
* * *
For smaller companies…, their platform utilities would be required to meet the same standard of fair, reasonable, and nondiscriminatory dealing with users, but would not be required to structurally separate….
* * * Second, my administration would appoint regulators committed to reversing illegal and anti-competitive tech mergers…. I will appoint regulators who are committed to… unwind[ing] anti-competitive mergers, including:
Let’s consider for a moment what this brave new world will look like — not the nirvana imagined by regulators and legislators who believe that decimating a company’s business model will deter only the “bad” aspects of the model while preserving the “good,” as if by magic, but the inevitable reality of antitrust populism.
Utilities? Are you kidding? For an overview of what the future of tech would look like under Warren’s “Platform Utility” policy, take a look at your water, electricity, and sewage service. Have you noticed any improvement (or reduction in cost) in those services over the past 10 or 15 years? How about the roads? Amtrak? Platform businesses operating under a similar regulatory regime would also similarly stagnate. Enforcing platform “neutrality” necessarily requires meddling in the most minute of business decisions, inevitably creating unintended and costly consequences along the way.
Network companies, like all businesses, differentiate themselves by offering unique bundles of services to customers. By definition, this means vertically integrating with some product markets and not others. Why are digital assistants like Siri bundled into mobile operating systems? Why aren’t the vast majority of third-party apps also bundled into the OS? If you want utilities regulators instead of Google or Apple engineers and designers making these decisions on the margin, then Warren’s “Platform Utility” policy is the way to go.
Grocery Stores. To take one specific case cited by Warren, how much innovation was there in the grocery store industry before Amazon bought Whole Foods? Since the acquisition, large grocery retailers, like Walmart and Kroger, have increased their investment in online services to better compete with the e-commerce champion. Many industry analysts expect grocery stores to use computer vision technology and artificial intelligence to improve the efficiency of check-out in the near future.
Smartphones. Imagine how forced neutrality would play out in the context of iPhones. If Apple can’t sell its own apps, it also can’t pre-install its own apps. A brand new iPhone with no apps — and even more importantly, no App Store — would be, well, just a phone, out of the box. How would users even access a site or app store from which to download independent apps? Would Apple be allowed to pre-install someone else’s apps? That’s discriminatory, too. Maybe it will be forced to offer a menu of all available apps in all categories (like the famously useless browser ballot screen demanded by the European Commission in its Microsoft antitrust case)? It’s hard to see how that benefits consumers — or even app developers.
Internet Search. Or take search. Calls for “search neutrality” have been bandied about for years. But most proponents of search neutrality fail to recognize that all Google’s search results entail bias in favor of its own offerings. As Geoff Manne and Josh Wright noted in 2011 at the height of the search neutrality debate:
[S]earch engines offer up results in the form not only of typical text results, but also maps, travel information, product pages, books, social media and more. To the extent that alleged bias turns on a search engine favoring its own maps, for example, over another firm’s, the allegation fails to appreciate that text results and maps are variants of the same thing, and efforts to restrain a search engine from offering its own maps is no different than preventing it from offering its own search results.
Nevermind that Google with forced non-discrimination likely means Google offering only the antiquated “ten blue links” search results page it started with in 1998 instead of the far more useful “rich” results it offers today; logically it would also mean Google somehow offering the set of links produced by any and all other search engines’ algorithms, in lieu of its own. If you think Google will continue to invest in and maintain the wealth of services it offers today on the strength of the profits derived from those search results, well, Elizabeth Warren is probably already your favorite politician.
And regulatory oversight of algorithmic content won’t just result in an impoverished digital experience; it will inevitably lead to an authoritarian one, as well:
Any agency granted a mandate to undertake such algorithmic oversight, and override or reconfigure the product of online services, thereby controls the content consumers may access…. This sort of control is deeply problematic… [because it saddles users] with a pervasive set of speech controls promulgated by the government. The history of such state censorship is one which has demonstrated strong harms to both social welfare and rule of law, and should not be emulated.
Digital Assistants. Consider also the veritable cage match among the tech giants to offer “digital assistants” and “smart home” devices with ever-more features at ever-lower prices. Today the allegedly non-existent competition among these companies is played out most visibly in this multi-featured market, comprising advanced devices tightly integrated with artificial intelligence, voice recognition, advanced algorithms, and a host of services. Under Warren’s nondiscrimination principle this market disappears. Each device can offer only a connectivity platform (if such a service is even permitted to be bundled with a physical device…) — and nothing more.
But such a world entails not only the end of an entire, promising avenue of consumer-benefiting innovation, it also entails the end of a promising avenue of consumer-benefiting competition. It beggars belief that anyone thinks consumers would benefit by forcing technology companies into their own silos, ensuring that the most powerful sources of competition for each other are confined to their own fiefdoms by order of law.
Breaking business models
Beyond the product-feature dimension, Sen. Warren’s proposal would be devastating for innovative business models. Why is Amazon Prime Video bundled with free shipping? Because the marginal cost of distribution for video is close to zero and bundling it with Amazon Prime increases the value proposition for customers. Why is almost every Google service free to users? Because Google’s business model is supported by ads, not monthly subscription fees. Each of the tech giants has carefully constructed an ecosystem in which every component reinforces the others. Sen. Warren’s plan would not only break up the companies, it would prohibit their business models — the ones that both created and continue to sustain these products. Such an outcome would manifestly harm consumers.
Both of Warren’s policy “solutions” are misguided and will lead to higher prices and less innovation. Her cause for alarm is built on a multitude of mistaken assumptions, but let’s address just a few (Warren in bold):
“Nearly half of all e-commerce goes through Amazon.” Yes, but it has only 5% of total retail in the United States. As my colleague Kristian Stout says, “the Internet is not a market; it’s a distribution channel.”
“Amazon has used its immense market power to force smaller competitors like Diapers.com to sell at a discounted rate.” The real story, as the founders of Diapers.com freely admitted, is that they sold diapers as what they hoped would be a loss leader, intending to build out sales of other products once they had a base of loyal customers:
And so we started with selling the loss leader product to basically build a relationship with mom. And once they had the passion for the brand and they were shopping with us on a weekly or a monthly basis that they’d start to fall in love with that brand. We were losing money on every box of diapers that we sold. We weren’t able to buy direct from the manufacturers.
Like all entrepreneurs, Diapers.com’s founders took a calculated risk that didn’t pay off as hoped. Amazon subsequently acquired the company (after it had declined a similar buyout offer from Walmart). (Antitrust laws protect consumers, not inefficient competitors). And no, this was not a case of predatory pricing. After many years of trying to make the business profitable as a subsidiary, Amazon shut it down in 2017.
“In the 1990s, Microsoft — the tech giant of its time — was trying to parlay its dominance in computer operating systems into dominance in the new area of web browsing. The federal government sued Microsoft for violating anti-monopoly laws and eventually reached a settlement. The government’s antitrust case against Microsoft helped clear a path for Internet companies like Google and Facebook to emerge.” The government’s settlement with Microsoft is not the reason Google and Facebook were able to emerge. Neither company entered the browser market at launch. Instead, they leapfrogged the browser entirely and created new platforms for the web (only later did Google create Chrome).
Furthermore, if the Microsoft case is responsible for “clearing a path” for Google is it not also responsible for clearing a path for Google’s alleged depredations? If the answer is that antitrust enforcement should be consistently more aggressive in order to rein in Google, too, when it gets out of line, then how can we be sure that that same more-aggressive enforcement standard wouldn’t have curtailed the extent of the Microsoft ecosystem in which it was profitable for Google to become Google? Warren implicitly assumes that only the enforcement decision in Microsoft was relevant to Google’s rise. But Microsoft doesn’t exist in a vacuum. If Microsoft cleared a path for Google, so did every decision not to intervene, which, all combined, created the legal, business, and economic environment in which Google operates.
Warren characterizes Big Tech as a weight on the American economy. In fact, nothing could be further from the truth. These superstar companies are the drivers of productivity growth, all ranking at or near the top for most spending on research and development. And while data may not be the new oil, extracting value from it may require similar levels of capital expenditure. Last year, Big Tech spent as much or more on capex as the world’s largest oil companies:
Warren also faults Big Tech for a decline in startups, saying,
The number of tech startups has slumped, there are fewer high-growth young firms typical of the tech industry, and first financing rounds for tech startups have declined 22% since 2012.
But this trend predates the existence of the companies she criticizes, as this chart from Quartz shows:
The exact causes of the decline in business dynamism are still uncertain, but recent research points to a much more mundane explanation: demographics. Labor force growth has been declining, which has led to an increase in average firm age, nudging fewer workers to start their own businesses.
Furthermore, it’s not at all clear whether this is actually a decline in business dynamism, or merely a change in business model. We would expect to see the same pattern, for example, if would-be startup founders were designing their software for acquisition and further development within larger, better-funded enterprises.
Will Rinehart recently looked at the literature to determine whether there is indeed a “kill zone” for startups around Big Tech incumbents. One paper finds that “an increase in fixed costs explains most of the decline in the aggregate entrepreneurship rate.” Another shows an inverse correlation across 50 countries between GDP and entrepreneurship rates. Robert Lucas predicted these trends back in 1978, pointing out that productivity increases would lead to wage increases, pushing marginal entrepreneurs out of startups and into big companies.
It’s notable that many in the venture capital community would rather not have Sen. Warren’s “help”:
If big companies like Google, Facebook, and Amazon are prevented from acquiring startups, that actually reduces competition. The reason is that if there is less M&A due to legal uncertainty, there is a reduced incentive for angels & VCs to fund those startups in the first place.
just to sustain constant growth in GDP per person, the U.S. must double the amount of research effort searching for new ideas every 13 years to offset the increased difficulty of finding new ideas.
If this assessment is correct, it may well be that coming up with productive and profitable innovations is simply becoming more expensive, and thus, at the margin, each dollar of venture capital can fund less of it. Ironically, this also implies that larger firms, which can better afford the additional resources required to sustain exponential growth, are a crucial part of the solution, not the problem.
Warren believes that Big Tech is the cause of our social ills. But Americans have more trust in Amazon, Facebook, and Google than in the political institutions that would break them up. It would be wise for her to reflect on why that might be the case. By punishing our most valuable companies for past successes, Warren would chill competition and decrease returns to innovation.
Finally, in what can only be described as tragic irony, the most prominent political figure who shares Warren’s feelings on Big Tech is President Trump. Confirming the horseshoe theory of politics, far-left populism and far-right populism seem less distinguishable by the day. As our colleague Gus Hurwitz put it, with this proposal Warren is explicitly endorsing the unitary executive theory and implicitly endorsing Trump’s authority to direct his DOJ to “investigate specific cases and reach specific outcomes.” Which cases will he want to have investigated and what outcomes will he be seeking? More good questions that Senator Warren should be asking. The notion that competition, consumer welfare, and growth are likely to increase in such an environment is farcical.
RSS - Posts