Ian Ayres

81 U Chi L Rev Dialogue 72 [Essay PDF]

     In Libertarian Paternalism, Path Dependence, and Temporary Law, Professors Tom Ginsburg, Jonathan S. Masur, and Richard H. McAdams (GMM) present an attractive theory of “temporary law” as law that expires after a set period of time.1 Their theory builds a persuasive normative case that in limited circumstances, temporary law might usefully dislodge a preexisting equilibrium. Like the archaic Alka-Seltzer ads, there may be times when a legally induced “try it, you’ll like it” strategy produces a superior, new separating equilibrium.2

Their simple smoking model raises profound questions about how libertarians should react to the possibility of multiple equilibria and to government-dislodging actions that merely move private choice from one equilibrium to another.3 While they characterize learning as one of the principle benefits of temporarily forcing individuals to take alternative actions,4 their model might generate multiple equilibria even in a world of perfect information—in which each actor knows of the existence of alternative equilibria but has no individual incentive to move toward them.

Their article also succeeds in creating a new category of regulation that helps organize a variety of existing practices, from the Supreme Court’s mandated twenty-five-year sunset for affirmative action to President Franklin D. Roosevelt’s bank holidays.5 Indeed, it is easy to conjure additional examples. There is much more that could be said about sunset laws as premeditated temporary law.6 And while GMM mention temporary curfews, police COMPSTAT-based choices to temporarily flood particular neighborhoods with officers are also related.7 After reading their article, I could imagine the possible utility of temporarily decertifying unions or temporarily unionizing unorganized employees. Their article might even rationalize parents calling for temporary breakups to see if a teen relationship was meant to last. My employer requires that my computer passwords are only temporary and my human subject approvals all have expiration dates. It would be useful to try to join these examples into an even more general theory of the temporary (or the durable).

In this Essay, I will suggest additional ways that the law might temporarily dislodge the ex ante equilibrium, and I will also suggest additional ways that the law might regulate the world ex post the temporary dislodgement.

I.  Dislodging Alternatives

GMM’s motivating examples are about temporary restrictions on liberty (followed by a return to liberty).8 In these examples, the laissez-faire behavior is temporarily dislodged by perfectly enforced restrictions that mandate that private actors do or refrain from doing certain things. But while mandatory restrictions and duties enforced by property rules will usually be sufficient to dislodge an ex ante equilibrium,9 there may be a variety of other temporary dislodging interventions. It might, for example, be possible to dislodge a preexisting equilibrium by imposing a duty that can be avoided by payment of a liability-rule-like amount or by limiting the right to do something unless a payment is made. Indeed, GMM’s example of traffic-congestion pricing represents this kind of a liability rule.10 And the law might temporarily deploy financial carrots as well as liability-rule sticks to bring about the benefits of changed behavior.11 The law might also perturb an initial equilibrium with judicious use of default and altering rules.12 An immediate implication of GMM’s approach is that temporary penalty default rules might be sufficient to force the revelation of information without incurring the long-term transaction costs of forcing all contractors to contract around a permanent default rule. These softer dislodging mechanisms might impose less of a burden on private liberty and, relatedly, might provide more opportunities for political compromise. Indeed, even softer interventions, such as the government just asking, rather than imposing a law, might be sufficient to dislodge a preexisting equilibrium. Informally or formally bestowing some sort of honor on homeowners who alter their lawn watering behavior during a water shortage, for example, might be sufficient to change behavior in the long run.13

II.  Ex Post Alternatives

Many of GMM’s examples contemplate a world in which law reverts to a laissez-faire norm after the temporary period of restriction. But just as there are softer alternatives to the hard-edged dislodging strategies during the period of restriction, the law might produce a range of firmer ex post restriction regulations. For example, after a period of prohibiting employment discrimination on the basis of sexual orientation, the law might transition to an opt-out regime in which employer nondiscrimination is presumed unless employers contracted around the default by publically announcing their desire to retain the right to discriminate on that basis.14 Lawmakers could impact the stickiness15 of the default by imposing altering rules that vary the necessary conditions for contracting around it.,16 What I have called “impeding altering rules” might be a particularly effective way of maintaining a greater degree of individual autonomy than traditional command-and-control mandatory rules.17 Indeed, ex post the temporary restriction, a system of licenses with price or quantity controls might be seen as a special kind of altering rule that requires regulated entities to secure the license or variance as the legally required procedure for opting out of the presumed regulation.

The requisites for opting out might also be made by regulators to turn on a collective choice of the regulated. One could imagine that bars would be allowed to return to permitting smoking only upon a plebiscite of a majority or supermajority of the bar owners. I recently proposed that Yale Law School adopt a supermajority transition rule of this kind with regard to a calendar experiment. The law school has, for about the last one hundred years, had its fall semester exams in January. The experiment was to see on a trial basis of a few years what it would be like to harmonize our calendar with the rest of the university and move the exam period to December.18 The transition rule after the trial would likely be especially important. Should the default be that we return to our prior practice, or that we retain December exams? I argued that to overcome status quo bias we should require not just an affirmative faculty vote to retain December exams, but also that two-thirds of faculty should favor the change for it to continue.

GMM were wise to focus on clean theoretical examples of temporary restrictions that extinguish choice with regard to the regulated activity, followed by a period of uninhibited autonomy. Their limiting case makes clear that a temporary restriction dislodging a preexisting equilibrium might have long-lived effects even after the restriction expires.19 But having realized this possibility, lawmakers have additional ways to exploit the benefits of temporary law. Most importantly, there is a tradeoff between the deviations from laissez-faire that the law imposes during and after the period for which the restriction is in effect. Lawmakers can soften the temporary restriction by reducing the impositions on autonomy during the period of restriction and might instead impose a firmer-than-laissez-faire regulation ex post the restriction. The alternatives that I envision would still move from a period of greater to a period of lesser restriction and hence could still partake of the same kinds of benefits and costs as GMM’s limiting examples.


GMM have written a fine article that inventively theorizes a new strategy—temporary law—for responding to a variety of inefficiencies that may impact unregulated behaviors. By identifying and naming this strategy, the article allows the authors to organize a variety of existing practices and to suggest other contexts in which temporary law might provide a net benefit. Their model also reminds us of a crucial temporal limitation of paternalism (or what I prefer to call “parentalism”20). Children are only temporarily subject to restrictions that limit their autonomy. They literally age out of these legal restrictions. Many of the temporary parenting restrictions have similar learning justifications. What’s more, as I have suggested in this Essay, parents utilize a range of methods that vary the harshness of the temporary restriction as well as the ex post transitions to greater autonomy.21 All in all, GMM have helped develop a device that deserves its place in the behaviorally informed policymaker’s tool kit.

          William K. Townsend Professor, Yale Law School.

      1     Tom Ginsburg, Jonathan S. Masur, and Richard H. McAdams, Libertarian Paternalism, Path Dependence, and Temporary Law, 81 U Chi L Rev 291 (2014).

      2     Stuart Elliott, Try It; You Liked It Once, and Alka-Seltzer Hopes You Do Again, NY Times Business: Media & Advertising (NY Times June 19, 2006), online at http://www.nytimes.com/2006/06/19/business/media/19adco.html?_r=0 (visited July 18, 2014).

      3     Ginsburg, Masur, and McAdams, 81 U Chi L Rev at 310–25 (cited in note 1).

      4     See id at 327.

      5     See id at 349–50, 355–56.

      6     See generally, for example, Richard E. Myers II, Responding to the Time-Based Failures of the Criminal Law through a Criminal Sunset Amendment, 49 BC L Rev 1327 (2008) (arguing that imposing twenty-five-year limits on all criminal legislation would increase legislative oversight and reduce courts’ incentives to address perceived enforcement failures); Manoj Viswanathan, Note, Sunset Provisions in the Tax Code: A Critical Evaluation and Prescriptions for the Future, 82 NYU L Rev 656 (2007) (criticizing sunset provisions in tax legislation as a means to bypass budgetary constraints and enact permanent law under the guise of temporary law).

      7     See James J. Willis, Stephen D. Mastrofski, and David Weisburd, Making Sense of COMPSTAT: A Theory-Based Analysis of Organizational Change in Three Police Departments, 41 L & Society Rev 147, 169–70 (2007). COMPSTAT is a system designed to help police organizations improve responsiveness to management as a means of more effectively reducing crime. Id at 148.

      8     See Ginsburg, Masur, and McAdams, 81 U Chi L Rev at 335–37 (cited in note 1) (discussing temporary smoking bans as an example); id at 347–49 (discussing the possibility of exclusively temporary car seat belt requirements); id at 351–52 (describing temporary curfews enacted to reduce crime).

      9     See, for example, id at 348 (cited in note 1) (noting that seatbelt regulations have contributed to “historically high rates” of seatbelt use).

      10     See id at 353–55 (cited in note 1). See also generally Abraham Bell and Gideon Parchomovsky, Pliability Rules, 101 Mich L Rev 1 (2002).

      11     See generally Lee Anne Fennell, Forcings, 114 Colum L Rev (forthcoming 2014), online at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2351358 (visited July 19, 2014).     

      12     See Ian Ayres, Regulating Opt-Out: An Economic Theory of Altering Rules, 121 Yale L J 2032, 2036 (2012) (explaining altering rules as “rules that govern how one contracts around the default”).

      13     This would be in contrast to command-and-control methods like those outlined in Douglas S. Kenney, Roberta A. Klein, and Martyn P. Clark, Use and Effectiveness of Municipal Water Restrictions during Drought in Colorado, 40 J Am Water Res Assoc 77 (2004).

      14     See generally Jennifer Gerarda Brown, Facilitating Boycotts of Discriminatory Organizations through an Informed Association Statute, 87 Minn L Rev 481 (2002).

      15     For a discussion of “sticky” default rules, see Ayers, 121 Yale L J at 2086–87 (cited in note 12).

      16     See id at 2084 (cited in note 12).

      17     Id. 

      18     See Lavinia Borzi, Law School Revamps Academic Calendar, Yale Daily News (Jan 16, 2014), online at http://yaledailynews.com/blog/2014/01/16/law-school-revamps-academic-cal... (visited July 19, 2014). 

      19     See Ginsburg, Masur, and McAdams, 81 U Chi L Rev at 314 (cited in note 1).

      20     Ian Ayres and Robert Gertner, Filling Gaps in Incomplete Contracts: An Economic Theory of Default Rules, 99 Yale L J 87, 88 (1989).

      21     For an extreme example, see generally Amy Chua, Battle Hymn of the Tiger Mother (Penguin 2011).