Published online by Cambridge University Press: 05 June 2016
I have said that government action might be evaluated by asking whether it increases welfare, promotes autonomy, respects dignity, and promotes self-government. I have also noted that the four values might conflict with one another. In many of the most important cases, however, they do not. Let us explore the relationship between choice architecture and each of the values, focusing in particular on the role of nudges.
If we care about human welfare, the guidance for public officials is simple and straightforward: Act to promote social welfare. But the idea of welfare can be specified in many different ways, and it is not clear how to measure or to promote it. One idea, associated with F. A. Hayek, is that government should create a basic structure, featuring property rights, freedom of contract, and free markets, and expect that the structure will generally promote social welfare. But as Hayek himself acknowledged, that framework is too simple. A government that wants to promote social welfare, however specified, will do other things as well. Above all, it will have to correct market failures, and that will require it to do a great deal, especially (but not only) if behavioral market failures count.
Some form of cost-benefit analysis, meant to capture the consequences of various possible interventions (including none at all), is the most administrable way of measuring the welfare effects of government action. If welfare is the guide, there is no prohibition on mandates and bans, because they can be amply justified on that ground. Air pollution regulations often have health and other benefits that dwarf the costs. Occupational safety and health regulations can save significant numbers of lives without imposing unjustified burdens. Of course it is true that environmental and occupational safety regulations can reduce rather than increase social welfare. What is needed is an accounting of their human consequences, not a celebration of regulation as such. And in some cases, mandates will have far higher costs than choice-preserving approaches, and their benefits will not justify those higher costs. The only point is that if welfare is our guide, our analysis of what government should do will be informed and guided by some form of cost-benefit analysis, and everything will depend on the details.