Do governments always need to enforce laws?
Coercive power is government’s most frightening weapon (one that libertarian paternalists fear as much as libertarians), and the conventional wisdom inside and outside of academia says that bureaucracies that use this power to implement and enforce a given regulation will be more successful than those that do not.
Ironically, when looking at the data, there are a number of cases where coercion is not necessary for a successful policy (though no universal rules about when these cases occur). For example, feelings of duty and an (probabilistically inaccurate) fear of getting caught lead most people to file honest tax returns. In other words, it is the attitudes of individual taxpayers – not the actual probability that that the IRS will audit them – that makes tax collection a relative success.
Another example is workplace safety and the Occupational Safety and Health Administration (OSHA). Inspections and penalties imposed by OSHA inspectors only lead to small drops in worker accident rates, typically on the order of less than 5 percent, according to a by study John Scholz and Wayne Gray. To some conservative think tanks, this is a satisfactory reason to abolish OSHA through a gradual phase-out. However, bureaucracies can play a role in achieving policy success by acting either as facilitators of cooperation, or as signalers of information to less-informed actors that help shape their behaviors. Similar to the IRS example, it is the reactions of company mangers to OSHA inspections, and inspections initiated by workers that have more positive effects on workplace safety than the actual inspections themselves.
These differences may seem irrelevant to some readers. The fear of coercion, rare or not, is grounded in the potential for actual coercion, which still puts coercion at the heart of the individual’s decision-making calculus. This is not a philosophical point we want to debate right now. Rather, we only want to stress that it is in the areas of signaling and facilitation where we think nudges show the most promise for sustaining “self-enforcing” public policies. Nudges are likely to be much less expensive than coercive methods; they are preferable in those situations where more traditional penalties fail (ie. issues surrounding additions); and they are easily adaptable to situations where social norms readily exist.
One skepticism of some of our nudges is that they are unlikely to be effective because they lack teeth. The lessons of research on citizen compliance and bureaucracies suggest that teeth may not always be the best instrument or the first priority. Despite James Madison’s doubts about the power of parchment barriers, there can situations when they are enough to achieve their authors’ intended ends.
For more on coercion, signaling, and facilitation, and for a third empirical example where success can be achieved without enforcement (indoor clean air laws) see Michael Licari’s excellent chapter “Bureaucratic Discretion and Regulatory Success without Enforcement” from Politics, Policy, and Organizations, edited by George Krause and Kenneth Meier.