Once and Future Nudges.

Author:Rowell, Arden
Position:Evaluating Nudge: A Decade of Libertarian Paternalism

    The nudge has successfully colonized the regulatory state. (1) The success of nudges has, in turn, changed the regulatory ecosystem in which individual nudges operate.

    To see how the world has changed, consider the environment of the primordial nudge. To succeed--to be adopted as a mechanism for government action--early nudges had to outcompete existing regulatory mechanisms. Those existing mechanisms--command and control, pure market mechanisms, basic disclosure, etc.--had already carved ecological niches for themselves. As entrenched species, these other mechanisms initially had an advantage over the newcomer nudges. Policymakers and stakeholders knew about these older mechanisms, had tested their legal and political welcome, and in many cases, statutes and regulations had already been specifically crafted to be hospitable to them.

    As a practical matter, this meant that adoption of early nudges was relatively costly for policymakers. Not only did policymakers have to know of the nudge to begin with, but they also had startup costs in ensuring that nudges could be used in a way that was legally permissible and politically and practically feasible. As the new kid on the block, these early nudges had much to prove to gain acceptance and to carve a place for themselves.

    This early environment was not all bad for primordial nudges. True, each nudge had to compete with other entrenched types of government action. But structural advantages that nudges hold versus those other mechanisms--including administrability benefits, (2) welfare benefits, (3) cost-effectiveness (4) and autonomy benefits (5)--often stood the nudge in good stead. When a nudge competed with other (even entrenched) regulatory species--that is, when policymakers were introduced to the option of using nudges as alternatives to other forms of government action--the nudge often prevailed.

    One way to understand the explosion of the nudge population, then, is that the competitive advantage nudges offered in many regulatory contexts allowed nudges to outcompete other mechanisms and led to increasing numbers of nudges. (6) As a species, the nudge was successful at replicating itself and at finding an ecological niche within the regulatory ecosystem. In this sense, the captive breeding program that Cass Sunstein and Richard Thaler developed to engineer and breed nudges and release them into the wild (7) has been extraordinarily effective.

    And yet the very success that has led to Symposia like this one has also changed the regulatory ecosystem in which nudges now operate. Not only must modern and future nudges compete with traditional tools of regulatory intervention like mandates and bans, they must also interact successfully with other nudges. And as the nudge population continues to grow, strategic policymakers now need to consider how the size of the population is likely to affect the experience of individual organisms.

    This consideration should address at least two aspects of nudges' evolutionary ecology. First, policymakers should develop observational methods for identifying and classifying possible nudge-nudge interactions. These observations can then form the groundwork for evaluating when and how nudges can be expected to interact with one another. Second, policymakers should consider the institutional tools available for managing the nudge population and for structuring potential interactions between and among nudges. The remainder of this Essay provides starting points for each of these analyses.


    Early nudges were generally justified by reference to their superiority (at least in some circumstances) to entrenched regulatory tools, including mandate, pure market mechanisms, and (non-behaviorally-informed) disclosure. (8) Initial arguments for nudges were thus arguments about the regulatory fitness of nudges in comparison to other mechanisms. (9)

    When nudges were rare, this initial focus on how nudges would compete with other regulatory species was entirely reasonable. The mechanism of nudging is based on the fundamental insight that people have scarce cognitive resources. (10) Why spend policymakers' and scholars' scarce resources addressing nudge-nudge interactions when a nudge might spend its whole life never meeting another of its kind? Yet as nudges continue to propagate, policymakers face a different landscape.

    In natural populations of organisms, successful population growth of this type has predictable implications. In particular, when natural species become established, individual organisms often compete with other similar organisms for scarce resources. Worrisomely, this type of inter-species competition can lead to costly and socially inefficient investments that only capture a marginal benefit for individual organisms. The classic example of this--an example that troubled a young Charles Darwin--is the showy, even over-elaborate tail of the male peacock. Darwin puzzled over peacocks for some years, as he (and critics) saw them as a challenge to the mechanism of natural selection, whereby organisms inherit traits that are differentially fit for their environment. (At one point, an exasperated Darwin apparently wrote to a friend that "[t]he sight of a feather in a peacock's tail, whenever I gaze at it, makes me sick!" (11)) Eventually Darwin realized that apparently maladaptive traits--like peacock tails--could actually confer a reproductive advantage when organisms were required to compete with others of their own kind. The result was his exhaustively specified theory of sexual selection, to which he devoted over half of his late-career book, The Descent of Man, and Selection in Relation to Sex. (12)

    Modern resource economists have reaffirmed that costly sexual selection of this type tends to evolve where species compete with other members of their own species. (13) In such circumstances, sexual selection works like an arms race, as individual organisms invest in energy-draining or risk-creating strategies that do not benefit the species as a whole. Peacocks, for instance, might actually do better as a species if they were better-camouflaged from predators. Yet research with eye-tracking cameras has confirmed that female peahens do indeed spend more time looking at "showy" male peacocks, (14) meaning that individual peacocks still enjoy a comparative advantage in reproducing their genes if they have the plumage that attracts potential mates. Thus, mechanisms that improve the fitness of individual organisms do not necessarily improve the fitness for the species as a whole.

    Only safely-entrenched species can afford to spend resources on highly distinctive species-specific sex selection. By contrast, when an organism must compete with other species for scarce resources, organisms can ill afford costly investments- like showy plumage, elaborate nests, or enormous racks of antler --that will help it (if at all) only in attracting mates of the same species. Other species that develop environment-specific adaptations, or which can spend their energy on food rather than display, will have too much of a comparative advantage. Under conditions of inter-species competition, then, natural selection--selection driven by the environment in which species live--is likely to drive organisms' success.

    What do intra-species competitions in natural species have to do with evaluating nudges? At least two things. First, the comparison can highlight the importance of increasing resource demand. If--because of the success of nudges--we now live in a system of increasing population of both nudges and nudgers, we should be alert to the possibility that nudge(r)s will face increasing competition from other nudge(r)s and that individual responses may not be socially optimal. In fact--as with peacock plumage--as nudges face increasing competition intra-species instead of cross-species, there is an increasing danger of nudge "arms races." The more successful nudges become as compared to competitor species, the more we might want to be alert to the corollary to sexual selection in nudges: selection that is driven by intra-nudge competition rather than by optimal (or even efficient) resource investment.

    Second, the comparison to biological competitions can help to illuminate what happens in competitive environments where resources are scarce. Recall that a foundational insight of behaviorally informed regulation is that people's cognitive resources (attention, processing, etc.) are bounded, and therefore scarce. (15) Agencies--the source of most government nudges--are statutorily required to establish policies in furtherance of particular goals. Statutes are mostly drafted in isolation from one another, without relative prioritization. If we conceive of the public's attention as the scarce resource for which agencies are often competing, many statutory directives implicate potential competition--and the competition for that resource may therefore be illuminated by evaluating biological competitions for scarce resources.


    Thus far the analysis has been theoretical. To illuminate real-world problem, it might be helpful to evaluate an example of where agency nudges interact with other agency nudges. In what kind of landscape might we expect nudge-nudge interaction?

    Much might (and perhaps should) be written in answer to this question. As a starting point, however, it may be reasonable to think that nudge-nudge interactions are most likely to occur where there are multiple potential policy goals that underlie a single consumer choice, and where there are multiple agencies who administer differing statutory requirements to try to shape public behavior regarding that choice. Such contexts provide fertile ground for nudges and thus (perhaps) present a greater chance of fostering nudge-nudge interactions.

    One example of such a policy space is...

To continue reading