Cass Sunstein is passionate about smart government's capacity to improve its citizens’ decision making and welfare. Steeped in the growing field of behavioral economics, Sunstein and colleague Richard Thaler made a splash in 2008 by arguing that public and private institutions could reduce cognitive errors in decision making and improve human welfare in their best-selling book, Nudge: Improving Decisions About Health, Wealth, and Happiness. That next year, Sunstein was given an opportunity to put theory into practice upon appointment as head of the Office of Information and Regulatory Affairs . His latest book, Simpler: The Future of Government, is the product of those experiences. Equal parts memoir and manifesto, Simpler takes the tools of behavioral economics and applies it to government decision making. With simpler government and smarter regulation, Sunstein argues, both citizens and policy makers will make better decisions and enjoy better outcomes.
Simpler is grounded in the premise that humans have preferences about their preferences. A person might demonstrate a preference for smoking, but internally, many smokers harbor a preference to stop smoking. People struggle to fulfill these second-order preferences because of our common cognitive limitations. Sunstein reviews the insights of behavioral economics and demonstrates how they can lead to suboptimal outcomes. Our decisions are often made by the cognitive impulses of our knee-jerk, intuitive “System 1” instead of our logical, deliberative “System 2,” popularized by Kahneman (2011). We procrastinate. We succumb to default settings and framing effects. We are blind to the invisible gorillas that lurk in our midst. We evaluate risks based on how readily examples come to mind. We labor under the paradox of choice. To help people help themselves, Simpler repeats the mantra offered in Nudge: regulation should aim to “influence decisions while preserving the freedom of choice” (p. 38).
Ready to put his regulatory theories to the test, Sunstein arrived in Washington to discover an unexpected roadblock. Regulation, Sunstein saw first-hand, was bloated, inefficient, and in dire need of a scientifically grounded makeover. What's more, regulators were often trapped by the same cognitive errors plaguing those they sought to help. How could regulators be expected to devise superior choice sets for their behaviorally challenged constituents if they could not even get their own houses in order? Before he could turn to architecting the choices of the regulated, Sunstein first had to architect the choices of the regulators.
To do this, Sunstein started playing what he calls “Regulatory Moneyball.” Just as the Oakland Athletics’ Billy Beane replaced superstitious intuitive judgments with statistically informed decisions, Sunstein emphasized System 2-based regulatory decisions. To curb regulators’ reliance on gut feelings, Sunstein promoted cost-benefit analyses. Agencies were directed to simplify or rescind cumbersome forms. When possible, regulators were instructed to automate processes. With these tools in place, Sunstein believed that regulators were ready to nudge.
Sunstein offers several examples to demonstrate the efficacy of his vision. Informational disclosure labels, like the joint Department of Transportation/Environmental Protection Agency crowd-sourced label on car fuel economy, presented what regulators determined to be the most relevant information in the most intuitive way. Graphic warning labels on cigarette cartons adorned with rotting lungs and tracheostomies, although currently in legal limbo, are presented as a smart nudge that shocks a smoker's System 1 into giving that next pack a second thought. Public-private partnerships, like Michelle Obama's collaboration with Walmart to promote healthier choices and labeling options, aim to combat childhood obesity. These examples are presented to show how our lives are better, less stressful, and our government is less costly because of Regulatory Moneyball.
Many of Sunstein's regulations will appeal to those with limited government sympathies. Compared to the alternative of outright bans, choice architecture regulation maintains individual autonomy and choice. Cost-benefit analyses can prevent inefficient regulations. Simplicity and automation can reduce the scope of government. Retrospective review can help clear regulations that have outlived their usefulness. Sunstein ultimately hopes to steer government toward behaving more like a business – albeit one that tells its customers what they should want.
Sunstein's proposals hold implications for the future of entrepreneurship. Kirzner (1978) stresses the role of the entrepreneur as an engine of growth. Empirical analyses of regulation, including by van Stel et al. (2007) and Nyström (2008), suggest that regulation impedes entrepreneurship. Conceptually, the entrepreneurial opportunities forgone by clumsy regulation are elusive and significant, notes Benson (2008). Sunstein's writing demonstrates an awareness of these problems. He decries the regulatory “sewer talk” of rent-seeking identified by Stigler (1971) and Tullock (1967). He hopes to temper the institutional inertia discussed by Niskanen (1971) with cost-benefit analyses and retrospective review. He wants to maximize the benefits and minimize the costs of regulation.
To the extent that regulators live up to their new processes, Regulatory Moneyball can limit the costs of regulation on entrepreneurship. A chapter titled “Removing red tape” provides a list of regulations that were eliminated or reformed when the costs were found to exceed the benefits (pp. 173-189). When entrepreneurship clashes with the goals of choice architects, however, Regulatory Moneyball will be more of an obstacle than an aid. Entrepreneurs of tobacco or soft drink products, for instance, will face formidable regulatory costs under the new regime.
Sunstein is no supporter of a smaller state. A primary goal of Simpler is “to suggest that without a massive reduction in its current functions, government can be far more effective, far less confusing, far less counterproductive, and far more helpful if it opts, wherever it can, for simplicity” (p. 11, emphasis original). Sunstein hopes to extend the simplification of our interactions with government to the simplification of all decisions deemed to be against our interests.
Are nudges paternalistic? Sunstein examines the question in regards to welfare and autonomy. If nudges help individuals to achieve their expressed second-order preferences, Sunstein argues, then they would actually improve welfare. And autonomy? Sunstein believes that so many of our choices are already externally influenced that adding the state's benevolent inputs does little to tip the scales. He summarizes: “No sensible person believes that public officials should be making people's choices for them. They shouldn’t. Freedom of choice and free markets are great engines for prosperity and human liberty […] The only claim I am making here is that in identifiable contexts, human welfare can be promoted by approaches that count as paternalistic” (p. 199). The question is then of which “identifiable contexts” warrant intervention.
Sunstein presents a guideline to judge when paternalism is appropriate: consumers must find their second-order preferences hindered by cognitive errors and express a need for external assistance. The first criterion is easily satisfied, but is the second criterion achievable? Sunstein suggests that public comment could serve as a proxy for gauging consumers’ wishes; however, this would represent a tiny portion of the politically engaged population. It also requires a belief that consumers become less prone to cognitive biases when submitting cheap talk for regulatory consideration than when revealing cost-constrained preferences through market activity. Unless regulators find a way to retrieve and rank each person's changing preference set and tailor unique welfare-enhancing choice sets for each person, choice architecture will run afoul of its own rules. It will continue to be more grounded in what regulators think consumers want rather than what they actually want.
Despite these shortcomings, there is much to like about Simpler. Choice architecture is perhaps best suited to guide the choice architects. Government actors are affected by both cognitive biases and the poor incentives that accompany non-market decision making. They face similar, measurable standards of success. By putting constraints on their actions, choice architecture prompts regulators to limit the destructive unintended consequences of their decisions on third parties.
When these tools are turned to improving the performance of private actors, however, their practical attainability is diminished. These proposals add another layer of complexity in decision making. Consumers must first verify they are being nudged in the direction they want before acting. Sunstein's goal to improve human welfare by backwards-engineering optimal choice sets may be desirable to those who share the values being optimized, but for those who do not, this approach is anything but “simpler.”
