What’s Your Problem? Building an Evidence-Based System of Regulatory Analysis from the Bottom-Up

August 15, 2023

From his penchant for the gaudiest of wardrobe options to his unique take on regulatory economics, Jerry Ellig was, in addition to being one of the deepest thinkers writing at the confluence of law and economics, a true iconoclast. When Jerry and I wrote our first paper together, which dealt with judicial review of regulatory impact analyses, I’ll admit that I was a tad intimidated to work with an economist of Jerry’s standing, having never taken more than a macro class in undergrad. As an administrative law expert who had worked in the field of regulatory economic analysis, I had picked up some of the lingo here and there, and I could at least give the false impression of comprehension when chatting about things like quality-adjusted life years or Pigouvian taxes.

But Jerry was never interested in discussing anything like that.  Indeed, of the many conversations we had regarding agencies’ regulatory impact analyses (RIAs), what has most stuck with me was one regarding the famous “Regulatory Report Card” that Jerry developed during his time at the Mercatus Center.[1] Jerry was never one to assign an “A for effort,” and he regularly gave failing grades to 300+ page tomes rife with complex formulae and economic jargon that, to me, seemed like the infallible dicta of the high priests and priestesses of the dismal science. The reason for the poor marks, Jerry explained to me, was an inattentiveness to the fundamentals. For all of their arcane prognostications concerning projected benefits and costs, agencies often ignored the most fundamental problem of all: what is the problem you are attempting to solve?

Defining the Problem: Who Should Act?

This is not, of course, to suggest that agencies regularly skip the first step of an RIA, defining the regulatory problem. Every agency economist worth her salt will include at least some statement to that effect. But it often consists of little more than pointing to the authorizing statute, claiming that the mere fact that Congress said the agency could act means that it should act, and that it should act in precisely the way that the agency has in mind. Very seldom will an agency provide any proof of an actual market failure, explaining why regulatory intervention is necessary in order to improve upon a supposedly suboptimal state of affairs.

And though the occasional RIA may contain a reference to externalities or excess market concentration, agencies almost never confront an equally important aspect of defining the regulatory problem: if regulatory intervention is truly necessary, must it occur at the federal, as opposed to the state or local, level?  

Of course, to suggest that anything other than a highly centralized, top-down system of regulation is worthy of consideration has become increasingly heretical on both the left and right sides of the aisle. Conservatives’ traditional affection for federalism has largely fallen by the wayside in an era in which prominent Republicans call for a nationwide strategy on not only economic but also social issues.[2] But the virtues of so-called subsidiarity, which derives from Catholic social teaching and holds that power should be pushed down to the lowest possible level, are hard to deny.

The Virtues of Regulatory Subsidiarity 

A subsidiarity-based approach to regulatory policymaking offers numerous advantages over the prevailing, heavily centralized one.  For one thing, a patchwork approach to regulation promotes competition.  Progressive fears over a “race to the bottom,” in which states trip over each other to despoil their natural environments and exploit their workers in order to attract greedy corporate overlords looking to save a buck, have not exactly played out as feared.[3]

Of course, recent years have witnessed some movement of corporations and people from states with heavy regulatory burdens, such as California and New York, towards states with more laissez-faire approaches, such as Florida and Texas.  But a state that completely gutted its regulatory code would likely do so at its peril, especially given modern corporations’ penchant for “social responsibility.”

For another thing, regulatory variation promotes policy tailoring.  A state that heavily regulates fracking activities, for instance, may endure a loss of jobs in the petroleum sector.  But it may attract green energy companies along with environmentally minded citizens who consider it important to live in a state that promotes their values.  Companies and people are increasingly willing to vote with their feet, and allowing for a wide variety of approaches maximizes the probability that every citizen (human or corporate) will find a place that matches his preferences.

Finally, even if you believe that a nationwide approach is always best, anyone who has paid even the slightest bit of attention to national politics over the last 20 years has to concede that the prospect of major regulatory reform in Congress has become effectively hopeless.  Republicans oppose just about every effort of Democrats to create new regulatory programs, and Democrats oppose basically every Republican effort to get rid of existing ones.  And in a nation that has been and will likely remain almost perfectly divided into two warring camps for the foreseeable future, the only meaningful opportunities for either side to accomplish much of anything are likely to reside at the state and local level.

U.S. Precedent for Regulatory Subsidiarity 

Of course, our collective consciousness has become so fixated on national politics that it can be difficult to conceive of what a subsidiarity-driven approach to regulatory policymaking might look like.  But if you look carefully, the key pieces are already in place.

For most of our history as a nation, the states wielded far more power than the federal government, which tended to focus on a handful of issues of nationwide importance such as immigration and national defense.  The Progressive Era witnessed an increasing centralization of power at the federal level that went into overdrive during the New Deal, but the concept of a massive federal bureaucracy controlling almost all aspects of citizen’s lives is a relatively recent invention.

Partly for this reason, our regulatory system features a variety of programs that are at least theoretically designed to draw upon the comparative strengths of state and local governments.

For instance, both the Medicaid and Social Security programs involve so-called “cooperative federalism” schemes in which the frontline administrators reside at the state level.

Even outside of cooperative federalism, the key instruments governing the regulatory state at least nominally acknowledge the importance of state, tribal, and local interests.  For example, Executive Order 12,866 and Circular A-4, which have for roughly 30 years provided the overarching framework for regulatory economic analysis, both explicitly indicate that agencies should consider the possibility of state or local solutions before deciding that federal intervention is necessary.  And Executive Order 13,132, which was issued by President Clinton and remains in effect, requires federal agencies to consult with their state counterparts prior to acting in a way that displaces state law.

What is far less common is any attempt to share the actual policymaking function with state governments.  The most prominent examples are likely the Clean Air Act and Clean Water Act, both of which erect a regime in which the U.S. Environmental Protection Agency sets an overall goal but the state governments are responsible for devising individual plans for achieving that goal.  So long as the states are meeting the federally defined objectives, they enjoy considerable flexibility in determining the exact approach they are going to take.

Foreign Precedents for Regulatory Subsidiarity

Interestingly, this delegated approach to policymaking is actually far more common in Europe than it is here.  Of course, the thought of looking “across the pond” for insights into regulatory policy may strike many as far-fetched: Americans tend to think of Europe as a part of the world whose glory days have long passed, a land of pretty castles and delectable cheeses but very little in the way of innovation, especially in the governmental sector.  But when it comes to the concept of subsidiarity, the Old World has a thing or two to teach the New.

In particular, the governing treaties of the European Union enshrine the doctrine of subsidiarity as a core principle of EU law: prior to acting, the EU government must ask whether the member states may be better positioned to address any given problem.[4]

Once it decides that a Europe-wide approach is necessary, the EU government then faces a second choice.  It can either issue a regulation, which governs EU citizens directly, or it can issue a directive, which functions much like the Clean Air Act and Clean Water Act regimes in the U.S.: Brussels sets the overall goal, but the member states are responsible for deciding how exactly they will achieve it.  Finally, as part of the so-called REFIT initiative, the EU government periodically revisits its existing laws to determine if a Europe-wide approach is still necessary.

From this perspective, it is much easier to conceive of how a subsidiarity-informed approach to policymaking might work in the U.S.  As part of the first step of conducting an RIA, agencies should ask if federal intervention is even necessary or if the state or local governments (or, importantly, the private sector) might be better positioned to address a problem.  If an agency concludes that a nationwide solution is indeed needed, it should then ask whether a cooperative federalism approach may be preferable: allowing states to decide how to address a problem still allows for the experimentation and tailoring that make a subsidiarity-based approach to policymaking so attractive.  And once they have acted, federal agencies should periodically reconsider whether or not a nationwide approach continues to make sense.  Even if state or local governments may not be equipped to handle a problem right now, they may later develop that capability and should have the opportunity to prove themselves.

State and Local Governments are Up to the Task

Speaking from firsthand experience, having spent over a decade working on regulatory policy at the federal level before accepting my current position at the state level, I can attest to the high capability of state regulators and their capacity to address the most vexing regulatory problems that agencies encounter.  In addition to possessing the same level of technical expertise as their federal counterparts, state officials are usually closer to the problem and can more easily design a sophisticated solution that accounts for all the complexities at play.  And though I have never had the privilege of working at the local level, my guess is that local officials would enjoy similar advantages, having an even more granular understanding of the problems they confront.

In short, state and local governments are well positioned to address regulatory issues but seldom receive the opportunity since their federal counterparts so thoroughly occupy the field.  Even though longstanding federal executive branch policy calls for federal regulators to consider deferring to state and local governments prior to acting, they seldom do so.  And when they do, they typically rely upon state regulators to carry out federal policy but fail to give them the chance to say what that policy should be.  This all would change, however, if either the White House or Congress became serious about demanding that agencies do a meaningful analysis at the first step of the RIA process and consider both whether there is truly a problem requiring intervention and, if so, whether the federal government itself must act.

Perhaps some day another economist will pick up Jerry’s important work and start handing out grades to federal agency RIAs.  If that day comes, I would humbly recommend that she or he give serious thought to scoring agencies on their efforts to work with state and local partners to decide who is best positioned to act.  But in the interim, I hope that the President and federal agency officials increasingly come to perceive the virtue of a bottom-up style of regulatory policymaking. 

In a nation as bitterly divided as ours, trying to hammer regulatory policymaking into a one-size-fits-all model is a formula for underperformance and resentment.  The Founders of the EU perceived early on that a region as diverse as the European Continent required a decentralized model to achieve any level of popular acceptance.  And though the United States lacks Europe’s linguistic diversity, I would submit that the ideological gap between California and Alabama is at least as vast as that separating, say, France and Romania.  In that respect, perhaps agencies have an incentive to start taking subsidiarity seriously and looking for ways to draw upon the highly underutilized capabilities existing at the state and local levels.  And though regulatory economists still have a long way to go before they’d ever start earning As from Professor Ellig, I’m quite certain that Jerry, who was ever the optimist (and who, as a practicing Catholic, was undoubtedly quite familiar with the virtues of subsidiarity), would be very happy to see the improvement.


[1] Jerry Ellig, Evaluating the Quality and Use of Regulatory Impact Analysis, Mercatus Working Paper (July 6, 2016).

[2] Paul Waldman, On Abortion and Guns, Republicans Are Ready to Abandon Federalism, The Washington Post (May 31, 2022).

[3] Ronald Brownstein, America Is Growing Apart, Possibly for Good, The Atlantic (June 24, 2022).

[4] European Parliament, The Principle of Subsidiarity, https://www.europarl.europa.eu/factsheets/en/sheet/7/the-principle-of-s… (last visited July 19, 2023).