How to Regulate: An Overview
So I’ve just finished writing a book (hence my long hiatus from Truth on the Market). Now that the draft is out of my hands and with the publisher (Cambridge University Press), I figured it’s a good time to rejoin my colleagues here at TOTM. To get back into the swing of things, I’m planning to produce a series of posts describing my new book, which may be of interest to a number of TOTM readers. I’ll get things started today with a brief overview of the project.
The book is titled How to Regulate: A Guide for Policy Makers. A topic of that enormity could obviously fill many volumes. I sought to address the matter in a single, non-technical book because I think law schools often do a poor job teaching their students, many of whom are future regulators, the substance of sound regulation. Law schools regularly teach administrative law, the procedures that must be followed to ensure that rules have the force of law. Rarely, however, do law schools teach students how to craft the substance of a policy to address a new perceived problem (e.g., What tools are available? What are the pros and cons of each?).
Economists study that matter, of course. But economists are often naïve about the difficulty of transforming their textbook models into concrete rules that can be easily administered by business planners and adjudicators. Many economists also pay little attention to the high information requirements of the policies they propose (i.e., the Hayekian knowledge problem) and the susceptibility of those policies to political manipulation by well-organized interest groups (i.e., public choice concerns).
How to Regulate endeavors to provide both economic training to lawyers and law students and a sense of the “limits of law” to the economists and other policy wonks who tend to be involved in crafting regulations. Below the fold, I’ll give a brief overview of the book. In later posts, I’ll describe some of the book’s specific chapters.
At the book’s outset, I have to clarify what I mean by regulation. That’s trickier than it sounds. Many common law doctrines are “regulatory” in that they address welfare-reducing market failures. The law of nuisance, for example, addresses negative externalities. So does negligence doctrine, which aims to induce efficient precaution-taking. Privately ordered arrangements like Airbnb’s reciprocal rating system may also serve regulatory purposes (in that case, combatting adverse selection-inducing information asymmetry). When most people talk about regulation, though, they aren’t thinking of the common law or private ordering. Instead, they have in mind the largely post-New Deal laws—mainly statutes and agency rules—that supplement traditional common law prohibitions.
To constrain the book’s scope, I define regulation as any threat-backed governmental directive aimed at correcting a defect in private ordering (i.e., the world that would exist if people did their own thing without government intervention beyond enforcing common law rights to person, property, and contract) where the defect causes total social welfare to be less than it otherwise would be. This admittedly narrow definition focuses the book’s inquiry on governmental directives aimed at reducing market failures (and market failure-like situations) and excludes from consideration the common law, privately ordered responses to market failures, and governmental directives aimed solely at redistribution. (I don’t deny that those things may be “regulation” in some sense, just not in the sense used by the book.)
If regulation is defined in this fashion, then one thing is immediately clear: It will almost always be dealing with “mixed bag” conduct—i.e., behavior that is sometimes good and sometimes bad. That’s because the common law has long managed to condemn all “always bad” behavior (e.g., battery, rape, fraud), and regulation, as defined for purposes of the book, takes over where the common law leaves off.
If regulation governs mixed bag conduct, then it can always err in two directions: It may fail to condemn bad instances of the conduct at issue (false acquittals), or it may wrongly condemn good instances of the conduct (false convictions). With either type of error, some welfare loss—some “error cost”—results. To minimize total error cost, regulators might seek to make their directives more nuanced, stating with greater precision the circumstances in which the conduct at issue will and will not be allowed. Nuance, though, isn’t free. Adding nuance raises a directive’s “decision costs”—i.e., the costs business planners and adjudicators must incur in reaching a decision as to what behavior is permitted and what condemned.
False acquittal error costs, false conviction error costs, and decision costs are in inexorable tension. Shrinking the scope of a rule’s prohibition to avoid false convictions enhances the risk of false acquittals. Conversely, expanding the prohibition to avoid false acquittals threatens an increase in false convictions. Attempting to minimize both false convictions and false acquittals simultaneously by adding in exceptions and other nuances raises decision costs.
In light of this unhappy situation, a policy maker seeking to maximize social welfare should follow a simple rule: Craft policies so as to minimize the sum of error and decision costs. Regular TOTM readers will be quite familiar with that rule of thumb. Several of us have promoted it in the antitrust context, where courts are routinely called upon to regulate mixed bag business practices. A key point of How to Regulate is that all behaviors subject to regulation (as defined above) are mixed bag practices, so the “decision-theoretic” strategy we regularly use in crafting antitrust standards should really apply to all regulation.
It’s all well and good to set forth an overarching goal like this, but what specific steps should policy makers take to achieve it? Most obviously, they would need to bring “on screen” the key costs and benefits that will determine the magnitude of a proposed policy’s error and decision costs. Moreover, it would not be enough to evaluate regulatory proposals in isolation. To assess the opportunity cost of selecting one regulatory option over another, policymakers would need to have some sense of how net welfare would differ under all regulatory alternatives. Of course, the cost of performing all this analysis must be kept in check; at some point, the incremental cost of further investigation by policy makers exceeds the incremental benefit the additional study provides, so additional investigation just isn’t worth it. The key, then, is to gather as much welfare-related information as can be cost-effectively gathered on all potential regulatory options, and then select the option that minimizes the sum of error and decision costs.
That sounds like a daunting task, but decision-makers routinely perform similar analyses in other contexts. Take physicians. A doctor seeing a patient with a new ailment should aim to select not just a remedy that will leave the patient better off (i.e., will create more benefit than cost) but the remedy that creates the greatest expected net benefit for the patient. And the doctor needs to act quickly and efficiently; she can’t dither incessantly over potential remedies, suffering “paralysis by analysis.”
To perform the task before her, the physician marches methodically through a number of steps. First, she identifies the patient’s symptom, the adverse effects experienced by the patient but not by healthy people. Next, she seeks to diagnose the cause of the symptom: What is it about this patient that leads him to suffer those adverse effects? After that, she catalogues the available remedies for correcting the cause of, or perhaps just alleviating, the symptom. She then assesses the potential side effects of each potential remedy. She ultimately selects the remedy that will provide the greatest net benefit to the patient.
How to Regulate’s central claim is that policy makers, when crafting regulations, should follow the lead of physicians. When confronted with a call to regulate, they should first identify the symptom—the adverse effect citizens confront within the scheme of private ordering. They then should seek to determine the cause of that symptom: Why has private ordering failed in this case to maximize social welfare? The regulator should then catalogue available remedies. Some may be “palliative” only, alleviating the symptom but failing to address the underlying disease. Others will seek to address not just the symptom but also the reason for it. Having catalogued the remedies, the policy maker should investigate their implementation costs and side effects and make an informed judgment as to the net benefits each offers. This process will bring on screen the information needed to minimize the sum of decision and error costs.
The bulk of How to Regulate aims to assist policy makers as they proceed through the steps described above. The book examines five sets of circumstances—“diseases”—that often lead to adverse effects and have traditionally been invoked to justify regulation. Those are the classic market failures of externalities, public goods, market power, and information asymmetry, as well as a long-recognized defect (agency costs) resulting from two of those market failures. The book then considers a novel, but increasingly prominent, non-market failure-based justification for regulatory intervention, people’s apparent cognitive and volitional limitations (i.e., the insights of behavioral economics). For each of these six defects, the book considers the symptoms they present, why those symptoms appear (the disease), what remedies are available for addressing each, and the remedies’ implementation costs and potential side effects.
This may sound old hat to TOTM’s sophisticated readers. How to Regulate’s primary contribution is simply to bring together insights of legal theorists and economists of various stripes—neoclassical, Austrian, public choice, behavioral—and systematize their ideas into a unified, practical approach to regulating. While the ideas are put together in a novel and, I believe, useful fashion, I break little new ground in legal theory or economics. At times during the drafting, I wondered whether the book was “original” enough. And in those times, I took comfort in the words of C.S. Lewis, whose Mere Christianity systematized the ideas of scores of philosophers and theologians, spoke plainly, and became a classic. On the closing page of that book, Lewis writes,
Even in literature and art, no man who bothers about originality will ever be original: whereas if you simply try to tell the truth (without caring twopence how often it has been told before) you will, nine times out of ten, become original without ever having noticed it.
That is what I endeavored to do with How to Regulate. I look forward to sharing more about the book over the next few weeks.