The following is the first in a new blog series by TOTM guests and authors on the law, economics, and policy of the ongoing COVID-19 pandemic. The entire series of posts is available at https://truthonthemarket.com/symposia/the-law-economics-of-the-covid-19-pandemic/.Continue Reading...
Archives For public choice
This guest post is by Neil Chilson, Senior Research Fellow for Technology and Innovation at Charles Koch Institute and former Chief Technologist at the Federal Trade Commission.
Is it rude to respond to a compliment with a rebuttal? I’m afraid that’s what I’m about to do. The antitrust reformers at the University of Chicago’s Stigler Center recently paid me a generous compliment, and I’m going to reply with an argumentative blog post. I guess that’s just the kind of guy I am.
First, some background. In mid-February the White House released its Annual Report of the Council of Economic Advisors. This CEA Report examines digital platforms and competition issues in some depth. In particular, it criticizes a September 2019 report by the Stigler Center (hereinafter the Final Report or Final Stigler Report). The Final Report recommends that the U.S. create a single, sector-specific Digital Authority to regulate digital companies, but the CEA Report rejects that recommendation.
In response, Luigi Zingales and Filippo Maria Lancieri of the Stigler Center posted “The Trump Administration Attacks the Stigler Report on Digital Platforms.” Zingales and Lancieri push back on the CEA’s conclusions. And here they pay me an unexpected compliment. From their blog post:
The section [of the CEA Report] on the digital economy is centered on an attack on the proposal of the Stigler Report to create a Digital Agency, saying it “raises a host of issues” and “that the downsides of new, far-reaching regulation need to be taken seriously.” In doing so, the Council of Economic Advisers repeats, almost verbatim, a claim already advanced by Charles Koch Institute’s Neil Chilson in the Washington Post, an argument already refuted in an earlier ProMarket post. (emphasis added)
I am flattered that the Stigler Center would connect my op ed with the thoughtful analysis of the experts at the Council of Economic Advisors. I like to think my op ed influenced the CEA’s report. But I suspect this is just a case of independent folks saying relatively obvious things about regulatory capture.
However, I’m a bit put out by their claim that my op ed was “already refuted in an earlier ProMarket post.” It is true that Fiona Scott Morton and Luigi Zingales published a post in November criticizing my op ed. At the time I found it so unpersuasive that I thought it unnecessary to respond with anything more than one tweet:
Still, the November post has occasionally been promoted by various U. Chicago-related twitter accounts. And now the Stigler Center claims that post refuted my op ed (and, by implication, the CEA Report). So, it’s time to respond to the November post more fully.
My Washington Post op ed made two relevant points, one observational and one substantive. First, I observed the irony of a center named after George Stigler proposing a new agency. Stigler was a Nobel-prize winning economist. He popularized the term “regulatory capture” and warned that “regulation is acquired by the industry and is designed and operated primarily for its benefit.” As my op ed points out, it’s a fair bet that he would be skeptical of creating a new agency like that recommended by the Final Stigler Report.
Second, my op ed argues against establishing a single new agency to regulate digital companies, primarily because such single-industry agencies are more vulnerable to regulatory capture. That is, such specialized agencies are more likely than generalist agencies to serve the interests of the regulated companies rather than the public. It would be better to subject the digital companies to an existing economy-wide enforcement agency like the Federal Trade Commission, I argue.
In the November blog post Morton and Zingales did not rebut either of these points. They exaggerate how thoroughly the Final Stigler Report evaluated and addressed such concerns. But more importantly, they do not refute my substantive claim about the capture of single-industry agencies – indeed, they appear to begrudgingly agree with me.
Let’s look at each argument.
First, the Report eagerly recommends a new independent agency but barely acknowledges regulatory capture risks and offers only a last-minute, hardly discussed fix. The body of the Final Stigler Report, which draws from four separate subcommittee reports, enthusiastically endorses a stand-alone, industry-specific Digital Authority (DA). As the Report’s Policy Brief summarizes, “The strongest indication emerging from the four reports is the importance of having a single powerful regulator capable of overseeing all aspects of DPs [digital platforms].” This conclusion represents a common theme throughout the Final Stigler Report:
- “Therefore, the report suggests that Congress should consider creating a specialist regulator, the Digital Authority.” (p 32)
- “For the reasons above, we believe the establishment of a sectoral regulator should be seriously considered.” (p 100)
- “Finally, because the problems we identify may require action beyond antitrust, we also propose the establishment of a new digital regulatory agency, or Digital Authority.” (p 120)
Despite this enthusiastic support for a new agency, the 313-page body of the Final Report offers no institutional design solutions to address regulatory capture concerns. The Final Report does mention “regulatory capture” or similar phrases a few times, mostly in citations or in platitudes like “it is important to remain cognizant of the dangers of regulatory capture.” In one example, the Final Report points to the sector-specific Federal Communications Commission as a model for the Digital Authority even as it acknowledges the FCC’s deep historic problem with regulatory capture. (My op ed dives deep into the lessons for a Digital Authority from the FCC’s history of capture – analysis wholly ignored by Morton and Zingales’s post.) So, the body of the Final Report offers very little to support Morton and Zingales’ claim that regulatory capture concerns were addressed.
However, the Final Report’s 17-page “Policy Brief” summary does contain a single paragraph that acknowledges the importance of institutional design and suggests a potential mitigation:
The Dangers of a Digital Authority: As George Stigler would readily point out, a new Digital Authority runs the risk of being captured by industry, becoming a new barrier to entry rather than a promoter of competition. This risk can be minimized, albeit not eliminated, by a careful institutional design. This is one reason why we envision—at least initially—to have the Digital Authority as a subdivision of the FTC, an across-industry authority with a better-than-average record of avoiding capture. Most importantly, the Digital Authority will have to be very transparent in all its activities. The Reports discuss a range of different institutional design mechanisms that can be explored to protect the Digital Authority from capture. (emphasis added)
Morton and Zingales point to this lone paragraph in the Policy Brief as proof that the Stigler Center had considered regulatory capture implications for institutional design, saying:
To avoid the legitimate concern raised by Chilson, the Stigler Center proposal suggests, as one option, that the Digital Authority be placed inside the Federal Trade Commission (FTC), an agency even Chilson appreciates, having worked there. Rotating FTC staff between consumer protection, antitrust, and digital authority could create useful synergies and expertise and minimize the risk that DA employees will be captured by industry. More stringent restrictions to revolving doors would help minimize this risk.
Yet this single paragraph in the Policy Brief is the Final Report’s only proposed mitigation of regulatory capture. And it is the only mention of a FTC-embedded approach in all the reports, including the subcommittee reports. In fact, Morton and Zingales’ blog post, in suggesting rotating FTC staff and revolving door restrictions, provides more detail about the potential institutional design of the DA than do the Policy Brief and the Final Report combined.
This is not strong evidence of a thoughtful approach to regulatory capture. Instead, as I pointed out in my tweet, this tack-on paragraph looks more like someone raised last-minute concerns that George Stigler might have objected to creating an entire new agency.
Morton and Zingales also claim that the four preliminary reports “discuss a range of different institutional design mechanisms.” But there isn’t much in the subcommittee reports, either. Neither the Market Structure and Antitrust Subcommittee’s report nor the Privacy and Data Protection Subcommittee’s report offer any mechanisms to prevent capture. The Media Industry Subcommittee’s report actually agrees with my op ed, noting that “industry regulators are often more prone to capture than the antitrust regulator” and suggests some transparency requirements.
The Political Systems Subcommittee’s report contains four paragraphs with some relevant discussion. Those paragraphs discuss whether the new Digital Authority should be modeled on the Consumer Financial Protection Bureau or instead designed like the Securities Exchange Commission- or FTC. However, the Subcommittee primarily expresses concern about politicization rather than industry capture. And it never makes a recommendation, concluding that “[t]here is probably no single way to balance the concerns for preventing industry capture and those related to political manipulation.”
So, does the Final Report address regulatory capture concerns generally, as Morton and Zingales claim? To sum up, across the hundreds of pages in five reports there are a few mentions of regulatory capture. One report briefly suggests transparency as a remedy. Another considers two different institutional designs for the DA but mostly worries about politicization. Finally, one sentence of the Final Report’s Policy Brief offers a novel alternative to the DA that is counter to the recommendations in the body of the Final Report. Perhaps these brief, contradictory discussions would be enough to address George Stigler’s presumptive concerns about regulatory capture, but I remain skeptical.
Second, Morton and Zingales do not refute my argument that industry-specific agencies are more likely to be captured – indeed, they appear to begrudgingly agree with me.
Most of Morton and Zingales’ blog post furiously battles a strawman, which is why I didn’t feel a strong need to respond last November. For example, they claim that, “Chilson’s argument, taken at face value, eliminates all government agencies.” Not at all – I focus on the downsides of industry-specific regulators. They accuse me of a laissez-faire approach toward tech companies. Yet my piece applauds FTC enforcement. They accuse me of evidence-free fear mongering and lacking concern for our democracy while they hypothesize about TikTok monopolizing social media (odd hypo, given five paragraphs earlier they claimed Facebook has a monopoly on social media) and say I don’t care about disclosure of foreign spending on political ads.
None of this has anything to do with the argument my op ed. All it does is distract from the fact that, as they admit, I raise “a legitimate concern” about regulatory capture. And they must admit this, because the Policy Brief essentially agrees with my op ed when it rejects, at the last minute, the Final Report’s recommendation for a stand-alone industry-specific Digital Authority. Compare the Policy Brief’s “Dangers of a Digital Authority” paragraph quoted above with the final paragraphs of my op ed:
[G]eneralist agencies that broadly regulate many industries are more resistant to capture. Further, agencies that primarily enforce laws (like the DOJ) are far less attractive targets for regulatory capture than those that mostly write rules (like the FCC). … In the United States, we already have an economywide enforcement agency: the Federal Trade Commission. Charged by Congress to promote competition and to protect consumers, the FTC has decades of experience addressing antitrust and consumer protection issues in the tech industry. … [I]f new capabilities are needed to police digital companies, doesn’t it make sense to give such authority to an experienced agency that has been resistant to regulatory capture?
Thus, Morton and Zingales don’t refute my op ed at all. I’m honestly not sure why they claim to. Perhaps I poked a sore spot. After all, there is a clear tension between the body of the Stigler Center Final Report, which recommends a stand-alone agency, and the Policy Brief, which in one sentence rejects that approach because of concerns over regulatory capture of an industry-specific regulator.
I share the concerns of that one sentence in the Policy Brief. So does the President’s Council of Economic Advisors. It remains to be seen where the Stigler Center settles on this point.
FCC Commissioner Rosenworcel penned an article this week on the doublespeak coming out of the current administration with respect to trade and telecom policy. On one hand, she argues, the administration has proclaimed 5G to be an essential part of our future commercial and defense interests. But, she tells us, the administration has, on the other hand, imposed tariffs on Chinese products that are important for the development of 5G infrastructure, thereby raising the costs of roll-out. This is a sound critique: regardless where one stands on the reasonableness of tariffs, they unquestionably raise the prices of goods on which they are placed, and raising the price of inputs to the 5G ecosystem can only slow down the pace at which 5G technology is deployed.
Unfortunately, Commissioner Rosenworcel’s fervor for advocating the need to reduce the costs of 5G deployment seems animated by the courageous act of a Democratic commissioner decrying the policies of a Republican President and is limited to a context where her voice lacks any power to actually affect policy. Even as she decries trade barriers that would incrementally increase the costs of imported communications hardware, she staunchly opposes FCC proposals that would dramatically reduce the cost of deploying next generation networks.
Given the opportunity to reduce the costs of 5G deployment by a factor far more significant than that by which tariffs will increase them, her preferred role as Democratic commissioner is that of resistance fighter. She acknowledges that “we will need 800,000 of these small cells to stay competitive in 5G” — a number significantly above the “the roughly 280,000 traditional cell towers needed to blanket the nation with 4G”. Yet, when she has had the opportunity to join the Commission on speeding deployment, she has instead dissented. Party over policy.
In this year’s “Historical Preservation” Order, for example, the Commission voted to expedite deployment on non-Tribal lands, and to exempt small cell deployments from certain onerous review processes under both the National Historic Preservation Act and the National Environmental Policy Act of 1969. Commissioner Rosenworcel dissented from the Order, claiming that that the FCC has “long-standing duties to consult with Tribes before implementing any regulation or policy that will significantly or uniquely affect Tribal governments, their land, or their resources.” Never mind that the FCC engaged in extensive consultation with Tribal governments prior to enacting this Order.
Indeed, in adopting the Order, the Commission found that the Order did nothing to disturb deployment on Tribal lands at all, and affected only the ability of Tribal authorities to reach beyond their borders to require fees and lengthy reviews for small cells on lands in which Tribes could claim merely an “interest.”
According to the Order, the average number of Tribal authorities seeking to review wireless deployments in a given geographic area nearly doubled between 2008 and 2017. During the same period, commenters consistently noted that the fees charged by Tribal authorities for review of deployments increased dramatically.
One environmental consultant noted that fees for projects that he was involved with increased from an average of $2,000.00 in 2011 to $11,450.00 in 2017. Verizon’s fees are $2,500.00 per small cell site just for Tribal review. Of the 8,100 requests that Verizon submitted for tribal review between 2012 and 2015, just 29 ( 0.3%) resulted in a finding that there would be an adverse effect on tribal historic properties. That means that Verizon paid over $20 million to Tribal authorities over that period for historic reviews that resulted in statistically nil action. Along the same lines, Sprint’s fees are so high that it estimates that “it could construct 13,408 new sites for what 10,000 sites currently cost.”
In other words, Tribal review practices — of deployments not on Tribal land — impose a substantial tariff upon 5G deployment, increasing its cost and slowing its pace.
There is a similar story in the Commission’s adoption of, and Commissioner Rosenworcel’s partial dissent from, the recent Wireless Infrastructure Order. Although Commissioner Rosenworcel offered many helpful suggestions (for instance, endorsing the OTARD proposal that Brent Skorup has championed) and nodded to the power of the market to solve many problems, she also dissented on central parts of the Order. Her dissent shows an unfortunate concern for provincial, political interests and places those interests above the Commission’s mission of ensuring timely deployment of advanced wireless communication capabilities to all Americans.
Commissioner Rosenworcel’s concern about the Wireless Infrastructure Order is that it would prevent state and local governments from imposing fees sufficient to recover costs incurred by the government to support wireless deployments by private enterprise, or from imposing aesthetic requirements on those deployments. Stated this way, her objections seem almost reasonable: surely local government should be able to recover the costs they incur in facilitating private enterprise; and surely local government has an interest in ensuring that private actors respect the aesthetic interests of the communities in which they build infrastructure.
The problem for Commissioner Rosenworcel is that the Order explicitly takes these concerns into account:
[W]e provide guidance on whether and in what circumstances aesthetic requirements violate the Act. This will help localities develop and implement lawful rules, enable providers to comply with these requirements, and facilitate the resolution of disputes. We conclude that aesthetics requirements are not preempted if they are (1) reasonable, (2) no more burdensome than those applied to other types of infrastructure deployments, and (3) objective and published in advance
It neither prohibits localities from recovering costs nor imposing aesthetic requirements. Rather, it requires merely that those costs and requirements be reasonable. The purpose of the Order isn’t to restrict localities from engaging in reasonable conduct; it is to prohibit them from engaging in unreasonable, costly conduct, while providing guidance as to what cost recovery and aesthetic considerations are reasonable (and therefore permissible).
The reality is that localities have a long history of using cost recovery — and especially “soft” or subjective requirements such as aesthetics — to extract significant rents from communications providers. In the 1980s this slowed the deployment and increased the costs of cable television. In the 2000s this slowed the deployment and increase the cost of of fiber-based Internet service. Today this is slowing the deployment and increasing the costs of advanced wireless services. And like any tax — or tariff — the cost is ultimately borne by consumers.
Although we are broadly sympathetic to arguments about local control (and other 10th Amendment-related concerns), the FCC’s goal in the Wireless Infrastructure Order was not to trample upon the autonomy of small municipalities; it was to implement a reasonably predictable permitting process that would facilitate 5G deployment. Those affected would not be the small, local towns attempting to maintain a desirable aesthetic for their downtowns, but large and politically powerful cities like New York City, where the fees per small cell site can be more than $5,000.00 per installation. Such extortionate fees are effectively a tax on smartphone users and others who will utilize 5G for communications. According to the Order, it is estimated that capping these fees would stimulate over $2.4 billion in additional infrastructure buildout, with widespread benefits to consumers and the economy.
Meanwhile, Commissioner Rosenworcel cries “overreach!” “I do not believe the law permits Washington to run roughshod over state and local authority like this,” she said. Her federalist bent is welcome — or it would be, if it weren’t in such stark contrast to her anti-federalist preference for preempting states from establishing rules governing their own internal political institutions when it suits her preferred political objective. We are referring, of course, to Rosenworcel’s support for the previous administration’s FCC’s decision to preempt state laws prohibiting the extension of municipal governments’ broadband systems. The order doing so was plainly illegal from the moment it was passed, as every court that has looked at it has held. That she was ok with. But imposing reasonable federal limits on states’ and localities’ ability to extract political rents by abusing their franchising process is apparently beyond the pale.
Commissioner Rosenworcel is right that the FCC should try to promote market solutions like Brent’s OTARD proposal. And she is also correct in opposing dangerous and destructive tariffs that will increase the cost of telecommunications equipment. Unfortunately, she gets it dead wrong when she supports a stifling regulatory status quo that will surely make it unduly difficult and expensive to deploy next generation networks — not least for those most in need of them. As Chairman Pai noted in his Statement on the Order: “When you raise the cost of deploying wireless infrastructure, it is those who live in areas where the investment case is the most marginal — rural areas or lower-income urban areas — who are most at risk of losing out.”
Reconciling those two positions entails nothing more than pointing to the time-honored Washington tradition of Politics Over Policy. The point is not (entirely) to call out Commissioner Rosenworcel; she’s far from the only person in Washington to make this kind of crass political calculation. In fact, she’s far from the only FCC Commissioner ever to have done so.
One need look no further than the previous FCC Chairman, Tom Wheeler, to see the hypocritical politics of telecommunications policy in action. (And one need look no further than Tom Hazlett’s masterful book, The Political Spectrum: The Tumultuous Liberation of Wireless Technology, from Herbert Hoover to the Smartphone to find a catalogue of its long, sordid history).
Indeed, Larry Downes has characterized Wheeler’s reign at the FCC (following a lengthy recounting of all its misadventures) as having left the agency “more partisan than ever”:
The lesson of the spectrum auctions—one right, one wrong, one hanging in the balance—is the lesson writ large for Tom Wheeler’s tenure at the helm of the FCC. While repeating, with decreasing credibility, that his lodestone as Chairman was simply to encourage “competition, competition, completion” and let market forces do the agency’s work for it, the reality, as these examples demonstrate, has been something quite different.
The Wheeler FCC has instead been driven by a dangerous combination of traditional rent-seeking behavior by favored industry clients, potent pressure from radical advocacy groups and their friends in the White House, and a sincere if misguided desire by Wheeler to father the next generation of network technologies, which quickly mutated from sound policy to empty populism even as technology continued on its own unpredictable path.
* * *
And the Chairman’s increasingly autocratic management style has left the agency more political and more partisan than ever, quick to abandon policies based on sound legal, economic and engineering principles in favor of bait-and-switch proceedings almost certain to do more harm than good, if only unintentionally.
The great irony is that, while Commissioner Rosenworcel’s complaints are backed by a legitimate concern that the Commission has waited far too long to take action on spectrum issues, the criticism should properly fall not upon the current Chair, but — you guessed it — his predecessor, Chairman Wheeler (and his predecessor, Julius Genachowski). Of course, in true partisan fashion, Rosenworcel was fawning in her praise for her political ally’s spectrum agenda, lauding it on more than one occasion as going “to infinity and beyond!”
Meanwhile, Rosenworcel has taken virtually every opportunity to chide and castigate Chairman Pai’s efforts to get more spectrum into the marketplace, most often criticizing them as too little, too slow, and too late. Yet from any objective perspective, the current FCC has been addressing spectrum issues at a breakneck pace, as fast, or faster than any prior Commission. As with spectrum, there is an upper limit to the speed at which federal bureaucracy can work, and Chairman Pai has kept the Commission pushed right up against that limit.
It’s a shame Commissioner Rosenworcel prefers to blame Chairman Pai for the problems she had a hand in creating, and President Trump for problems she has no ability to correct. It’s even more a shame that, having an opportunity to address the problems she so often decries — by working to get more spectrum deployed and put into service more quickly and at lower cost to industry and consumers alike — she prefers to dutifully wear the hat of resistance, instead.
But that’s just politics, we suppose. And like any tariff, it makes us all poorer.
A recent exchange between Chris Walker and Philip Hamburger about Walker’s ongoing empirical work on the Chevron doctrine (the idea that judges must defer to reasonable agency interpretations of ambiguous statutes) gives me a long-sought opportunity to discuss what I view as the greatest practical problem with the Chevron doctrine: it increases both politicization and polarization of law and policy. In the interest of being provocative, I will frame the discussion below by saying that both Walker & Hamburger are wrong (though actually I believe both are quite correct in their respective critiques). In particular, I argue that Walker is wrong that Chevron decreases politicization (it actually increases it, vice his empirics); and I argue Hamburger is wrong that judicial independence is, on its own, a virtue that demands preservation. Rather, I argue, Chevron increases overall politicization across the government; and judicial independence can and should play an important role in checking legislative abdication of its role as a politically-accountable legislature in a way that would moderate that overall politicization.
Walker, along with co-authors Kent Barnett and Christina Boyd, has done some of the most important and interesting work on Chevron in recent years, empirically studying how the Chevron doctrine has affected judicial behavior (see here and here) as well as that of agencies (and, I would argue, through them the Executive) (see here). But the more important question, in my mind, is how it affects the behavior of Congress. (Walker has explored this somewhat in his own work, albeit focusing less on Chevron than on how the role agencies play in the legislative process implicitly transfers Congress’s legislative functions to the Executive).
My intuition is that Chevron dramatically exacerbates Congress’s worst tendencies, encouraging Congress to push its legislative functions to the executive and to do so in a way that increases the politicization and polarization of American law and policy. I fear that Chevron effectively allows, and indeed encourages, Congress to abdicate its role as the most politically-accountable branch by deferring politically difficult questions to agencies in ambiguous terms.
One of, and possibly the, best ways to remedy this situation is to reestablish the role of judge as independent decisionmaker, as Hamburger argues. But the virtue of judicial independence is not endogenous to the judiciary. Rather, judicial independence has an instrumental virtue, at least in the context of Chevron. Where Congress has problematically abdicated its role as a politically-accountable decisionmaker by deferring important political decisions to the executive, judicial refusal to defer to executive and agency interpretations of ambiguous statutes can force Congress to remedy problematic ambiguities. This, in turn, can return the responsibility for making politically-important decisions to the most politically-accountable branch, as envisioned by the Constitution’s framers.
A refresher on the Chevron debate
Chevron is one of the defining doctrines of administrative law, both as a central concept and focal debate. It stands generally for the proposition that when Congress gives agencies ambiguous statutory instructions, it falls to the agencies, not the courts, to resolve those ambiguities. Thus, if a statute is ambiguous (the question at “step one” of the standard Chevron analysis) and the agency offers a reasonable interpretation of that ambiguity (“step two”), courts are to defer to the agency’s interpretation of the statute instead of supplying their own.
This judicially-crafted doctrine of deference is typically justified on several grounds. For instance, agencies generally have greater subject-matter expertise than courts so are more likely to offer substantively better constructions of ambiguous statutes. They have more resources that they can dedicate to evaluating alternative constructions. They generally have a longer history of implementing relevant Congressional instructions so are more likely attuned to Congressional intent – both of the statute’s enacting and present Congresses. And they are subject to more direct Congressional oversight in their day-to-day operations and exercise of statutory authority than the courts so are more likely concerned with and responsive to Congressional direction.
Chief among the justifications for Chevron deference is, as Walker says, “the need to reserve political (or policy) judgments for the more politically accountable agencies.” This is at core a separation-of-powers justification: the legislative process is fundamentally a political process, so the Constitution assigns responsibility for it to the most politically-accountable branch (the legislature) instead of the least politically-accountable branch (the judiciary). In turn, the act of interpreting statutory ambiguity is an inherently legislative process – the underlying theory being that Congress intended to leave such ambiguity in the statute in order to empower the agency to interpret it in a quasi-legislative manner. Thus, under this view, courts should defer both to this Congressional intent that the agency be empowered to interpret its statute (and, should this prove problematic, it is up to Congress to change the statute or to face political ramifications), and the courts should defer to the agency interpretation of that statute because agencies, like Congress, are more politically accountable than the courts.
Chevron has always been an intensively studied and debated doctrine. This debate has grown more heated in recent years, to the point that there is regularly scholarly discussion about whether Chevron should be repealed or narrowed and what would replace it if it were somehow curtailed – and discussion of the ongoing vitality of Chevron has entered into Supreme Court opinions and the appointments process with increasing frequency. These debates generally focus on a few issues. A first issue is that Chevron amounts to a transfer of the legislature’s Constitutional powers and responsibilities over creating the law to the executive, where the law ordinarily is only meant to be carried out. This has, the underlying concern is, contributed to the increase in the power of the executive compared to the legislature. A second, related, issue is that Chevron contributes to the (over)empowerment of independent agencies – agencies that are already out of favor with many of Chevron’s critics as Constitutionally-infirm entities whose already-specious power is dramatically increased when Chevron limits the judiciary’s ability to check their use of already-broad Congressionally-delegated authority.
A third concern about Chevron, following on these first two, is that it strips the judiciary of its role as independent arbiter of judicial questions. That is, it has historically been the purview of judges to answer statutory ambiguities and fill in legislative interstices.
Chevron is also a focal point for more generalized concerns about the power of the modern administrative state. In this context, Chevron stands as a representative of a broader class of cases – State Farm, Auer, Seminole Rock, Fox v. FCC, and the like – that have been criticized as centralizing legislative, executive, and judicial powers in agencies, allowing Congress to abdicate its role as politically-accountable legislator, abdicating the judiciary’s role in interpreting the law, as well as raising due process concerns for those subject to rules promulgated by federal agencies..
Walker and his co-authors have empirically explored the effects of Chevron in recent years, using robust surveys of federal agencies and judicial decisions to understand how the doctrine has affected the work of agencies and the courts. His most recent work (with Kent Barnett and Christina Boyd) has explored how Chevron affects judicial decisionmaking. Framing the question by explaining that “Chevron deference strives to remove politics from judicial decisionmaking,” they ask whether “Chevron deference achieve[s] this goal of removing politics from judicial decisionmaking?” They find that, empirically speaking, “the Chevron Court’s objective to reduce partisan judicial decision-making has been quite effective.” By instructing judges to defer to the political judgments (or just statutory interpretations) of agencies, judges are less political in their own decisionmaking.
Hamburger responds to this finding somewhat dismissively – and, indeed, the finding is almost tautological: “of course, judges disagree less when the Supreme Court bars them from exercising their independent judgment about what the law is.” (While a fair critique, I would temper it by arguing that it is nonetheless an important empirical finding – empirics that confirm important theory are as important as empirics that refute it, and are too often dismissed.)
Rather than focus on concerns about politicized decisionmaking by judges, Hamburger focuses instead on the importance of judicial independence – on it being “emphatically the duty of the Judicial Department to say what the law is” (quoting Marbury v. Madison). He reframes Walker’s results, arguing that “deference” to agencies is really “bias” in favor of the executive. “Rather than reveal diminished politicization, Walker’s numbers provide strong evidence of diminished judicial independence and even of institutionalized judicial bias.”
So which is it? Does Chevron reduce bias by de-politicizing judicial decisionmaking? Or does it introduce new bias in favor of the (inherently political) executive? The answer is probably that it does both. The more important answer, however, is that neither is the right question to ask.
What’s the correct measure of politicization? (or, You get what you measure)
Walker frames his study of the effects of Chevron on judicial decisionmaking by explaining that “Chevron deference strives to remove politics from judicial decisionmaking. Such deference to the political branches has long been a bedrock principle for at least some judicial conservatives.” Based on this understanding, his project is to ask whether “Chevron deference achieve[s] this goal of removing politics from judicial decisionmaking?”
This framing, that one of Chevron’s goals is to remove politics from judicial decisionmaking, is not wrong. But this goal may be more accurately stated as being to prevent the judiciary from encroaching upon the political purposes assigned to the executive and legislative branches. This restatement offers an important change in focus. It emphasizes the concern about politicizing judicial decisionmaking as a separation of powers issue. This is in apposition to concern that, on consequentialist grounds, judges should not make politicized decisions – that is, judges should avoid political decisions because it leads to substantively worse outcomes.
It is of course true that, as unelected officials with lifetime appointments, judges are the least politically accountable to the polity of any government officials. Judges’ decisions, therefore, can reasonably be expected to be less representative of, or responsive to, the concerns of the voting public than decisions of other government officials. But not all political decisions need to be directly politically accountable in order to be effectively politically accountable. A judicial interpretation of an ambiguous law, for instance, can be interpreted as a request, or even a demand, that Congress be held to political account. And where Congress is failing to perform its constitutionally-defined role as a politically-accountable decisionmaker, it may do less harm to the separation of powers for the judiciary to make political decisions that force politically-accountable responses by Congress than for the judiciary to respect its constitutional role while the Congress ignores its role.
Before going too far down this road, I should pause to label the reframing of the debate that I have impliedly proposed. To my mind, the question isn’t whether Chevron reduces political decisionmaking by judges; the question is how Chevron affects the politicization of, and ultimately accountability to the people for, the law. Critically, there is no “conservation of politicization” principle. Institutional design matters. One could imagine a model of government where Congress exercises very direct oversight over what the law is and how it is implemented, with frequent elections and a Constitutional prohibition on all but the most express and limited forms of delegation. One can also imagine a more complicated form of government in which responsibilities for making law, executing law, and interpreting law, are spread across multiple branches (possibly including myriad agencies governed by rules that even many members of those agencies do not understand). And one can reasonably expect greater politicization of decisions in the latter compared to the former – because there are more opportunities for saying that the responsibility for any decision lies with someone else (and therefore for politicization) in the latter than in the “the buck stops here” model of the former.
In the common-law tradition, judges exercised an important degree of independence because their job was, necessarily and largely, to “say what the law is.” For better or worse, we no longer live in a world where judges are expected to routinely exercise that level of discretion, and therefore to have that level of independence. Nor do I believe that “independence” is necessarily or inherently a criteria for the judiciary, at least in principle. I therefore somewhat disagree with Hamburger’s assertion that Chevron necessarily amounts to a problematic diminution in judicial independence.
Again, I return to a consequentialist understanding of the purposes of judicial independence. In my mind, we should consider the need for judicial independence in terms of whether “independent” judicial decisionmaking tends to lead to better or worse social outcomes. And here I do find myself sympathetic to Hamburger’s concerns about judicial independence. The judiciary is intended to serve as a check on the other branches. Hamburger’s concern about judicial independence is, in my mind, driven by an overwhelmingly correct intuition that the structure envisioned by the Constitution is one in which the independence of judges is an important check on the other branches. With respect to the Congress, this means, in part, ensuring that Congress is held to political account when it does legislative tasks poorly or fails to do them at all.
The courts abdicate this role when they allow agencies to save poorly drafted statutes through interpretation of ambiguity.
Judicial independence moderates politicization
Hamburger tells us that “Judges (and academics) need to wrestle with the realities of how Chevron bias and other administrative power is rapidly delegitimizing our government and creating a profound alienation.” Huzzah. Amen. I couldn’t agree more. Preach! Hear-hear!
Allow me to present my personal theory of how Chevron affects our political discourse. In the vernacular, I call this Chevron Step Three. At Step Three, Congress corrects any mistakes made by the executive or independent agencies in implementing the law or made by the courts in interpreting it. The subtle thing about Step Three is that it doesn’t exist – and, knowing this, Congress never bothers with the politically costly and practically difficult process of clarifying legislation.
To the contrary, Chevron encourages the legislature expressly not to legislate. The more expedient approach for a legislator who disagrees with a Chevron-backed agency action is to campaign on the disagreement – that is, to politicize it. If the EPA interprets the Clean Air Act too broadly, we need to retake the White House to get a new administrator in there to straighten out the EPA’s interpretation of the law. If the FCC interprets the Communications Act too narrowly, we need to retake the White House to change the chair so that we can straighten out that mess! And on the other side, we need to keep the White House so that we can protect these right-thinking agency interpretations from reversal by the loons on the other side that want to throw out all of our accomplishments. The campaign slogans write themselves.
So long as most agencies’ governing statutes are broad enough that those agencies can keep the ship of state afloat, even if drifting rudderless, legislators have little incentive to turn inward to engage in the business of government with their legislative peers. Rather, they are freed to turn outward towards their next campaign, vilifying or deifying the administrative decisions of the current government as best suits their electoral prospects.
The sharp-eyed observer will note that I’ve added a piece to the Chevron puzzle: the process described above assumes that a new administration can come in after an election and simply rewrite all of the rules adopted by the previous administration. Not to put too fine a point on the matter, but this is exactly what administrative law allows (see Fox v. FCC and State Farm). The underlying logic, which is really nothing more than an expansion of Chevron, is that statutory ambiguity delegates to agencies a “policy space” within which they are free to operate. So long as agency action stays within that space – which often allows for diametrically-opposed substantive interpretations – the courts say that it is up to Congress, not the Judiciary, to provide course corrections. Anything else would amount to politically unaccountable judges substituting their policy judgments (this is, acting independently) for those of politically-accountable legislators and administrators.
In other words, the politicization of law seen in our current political moment is largely a function of deference and a lack of stare decisis combined. A virtue of stare decisis is that it forces Congress to act to directly address politically undesirable opinions. Because agencies are not bound by stare decisis, an alternative, and politically preferable, way for Congress to remedy problematic agency decisions is to politicize the issue – instead of addressing the substantive policy issue through legislation, individual members of Congress can campaign on it. (Regular readers of this blog will be familiar with one contemporary example of this: the recent net neutrality CRA vote, which is widely recognized as having very little chance of ultimate success but is being championed by its proponents as a way to influence the 2018 elections.) This is more directly aligned with the individual member of Congress’s own incentives, because, by keeping and placing more members of her party in Congress, her party will be able to control the leadership of the agency which will thus control the shape of that agency’s policy. In other words, instead of channeling the attention of individual Congressional actors inwards to work together to develop law and policy, it channels it outwards towards campaigning on the ills and evils of the opposing administration and party vice the virtues of their own party.
The virtue of judicial independence, of judges saying what they think the law is – or even what they think the law should be – is that it forces a politically-accountable decision. Congress can either agree, or disagree; but Congress must do something. Merely waiting for the next administration to come along will not be sufficient to alter the course set by the judicial interpretation of the law. Where Congress has abdicated its responsibility to make politically-accountable decisions by deferring those decisions to the executive or agencies, the political-accountability justification for Chevron deference fails. In such cases, the better course for the courts may well be to enforce Congress’s role under the separation of powers by refusing deference and returning the question to Congress.
I find it interesting that many on the left, so intent on maintaining their anti-market narratives, distort reality so badly that black is white and up is down–and “government” is “corporations.”
I’ve highlighted this before when discussing the misdirected criticisms (and solutions) of self-described privacy advocates who point the finger at Google when really they should be concerned about the government.
Now comes Brian Leiter referring us to an article on “Corporate Attacks on Law School Clinics.” That’s the title of his post which contains nothing more than a heated admonition to read a linked article, so the title says it all: Corporations are attacking law school clinics (and this is a huge problem that should concern everyone). And I have no doubt many corporations are upset with many law school clinics. But what’s so fascinating is how, when you click through to the article, you discover that the actual attacks on law school clinics are, in every single example adduced in the story, actually emanating from governments. It’s pretty amazing. Here are the relevant snippets from each example in the article, but I recommend reading the whole thing:
In spring 2010, a law-clinic lawsuit against a $4 billion poultry company triggered a legislative effort to withhold state funds from the University of Maryland unless its law school provided the legislature with sensitive information about clinic clients and case activities.
The attack plan included the introduction of legislation that would forfeit all state funding if a university offered certain types of law-clinic courses.
The first occurred in 1968 at the University of Mississippi, where the appointments of two untenured professors were terminated following complaints that their new clinical program participated in a desegregation lawsuit.
In efforts to terminate the program, clinic opponents sponsored a bill in the legislature to withdraw state funding for the entire law school.
In 1993, then-governor Edwin Edwards was so upset at statements the clinic’s director made that the governor threatened to deny financial assistance to state residents attending the university and to prohibit Tulane medical students from working in any state hospital unless the director was fired.
A few years later, the clinic’s success in representing a low-income, minority community opposed to a proposed chemical plant led then-governor Mike Foster and business interests to threaten to revoke Tulane’s tax-exempt status and deny it access to state education trust-fund money, to organize an economic boycott of Tulane, and to refuse to hire its graduates.
When the university still refused to terminate the course, clinic opponents successfully persuaded the Louisiana Supreme Court to impose restrictions on whom law school clinics can assist and what kinds of representation students can provide.
When state legislators expressed disapproval of a law school clinic’s representation of citizens concerned about a proposed highway, university officials began charging the clinic for the university’s overhead costs, prevented it from approaching funders unless it agreed to avoid certain cases that might upset legislators, and pressed it to separate from the school and move off campus.
The clinical program at Rutgers University is defending itself against a lawsuit brought by a developer, who was defeated in a clinic case and is now seeking to use the state’s public records law to gain access to internal clinic case files that would otherwise be beyond the reach of a party to a lawsuit
A dispute in Michigan this past winter demonstrates that attacks also can occur when students get in the way of powerful government interests. The district attorney in Detroit, upset with the efforts of a University of Michigan innocence clinic to exonerate a man it alleged was wrongfully imprisoned for ten years, sought to force the students to testify at trial against their client, an unprecedented effort to interfere in the students’ attorneyclient relationship.
Perdue persuaded legislators to attach a rider to the university’s appropriations that conditioned $750,000 in funding on submission of a report detailing clinic cases, clients, expenditures, and funding, much of which is confidential information.
An even harsher attack occurred in Louisiana this past spring, where the Louisiana Chemical Association (LCA) pushed for legislation, subject to narrow exceptions, that would forfeit all state funds going to any university, public or private, whose clinics brought or defended a lawsuit against a government agency, represented anyone seeking monetary damages, or raised state constitutional claims. The bill also would have made clinic courses at the state’s four law schools subject to oversight by legislative commerce committees.
This isn’t cherry-picking. Unless I made a mistake, this is every single example of “attacks on law school clinics” in the article. And every single one involves government actions or the threat of government actions. Wow. How on earth could anyone read this article and feel comfortable calling this a problem of corporations? Don’t get me wrong–I understand that there are often corporate interests behind these actions, spurring them on. But to call this a “corporate” problem rather than a “government” problem–with the implicit call for government to do something about the problem–is to fail so utterly to understand the problems of government power that it boggles the mind.
Like Brian Leiter, I find this list troubling. I am appalled at how much inappropriate government interference this represents. But it is simply delusional to call this a problem of corporations. You want to fix the problem? Rein in the ability of governments to interfere to thoroughly with private life that special interests don’t have access to such a powerful and, often, invincible bludgeon.