Behavioral Economics and Tax Policy

William J. Congdon, Jeffrey R. Kling, and
Jeffrey R. Kling Former Brookings Expert, Associate Director for Economic Analysis - Congressional Budget Office
Sendhil Mullainathan
Sendhil Mullainathan Roman Family University Professor - University of Chicago Booth

June 16, 2009


Rahm Emmanuel, the current White House Chief of Staff, has famously been quoted as saying that: “You don’t ever want a crisis to go to waste; it’s an opportunity to do important things that you would otherwise avoid” (Zeleny and Calmes 2008). The recent crisis—the recession and the associated financial emergency—represents such an opportunity for a wide range of economic policies, including tax policy. This is true not just in some political or operational sense: as an opportunity to accomplish policies we would otherwise avoid. But also from a scientific perspective: as an opportunity to reconsider the intellectual foundations of economic policy in ways we might otherwise not be motivated or encouraged to do.

In this paper, we link this opportunity to rethink aspects of the standard economic approach to tax policy with a compelling cause for doing so. We argue that the implications of behavioral economics—the integration of economics and the psychology of preference formation and choice—for public policy, including tax policy, have yet to be systematically explored, and that this oversight leads to both mistaken policy and missed opportunity. Behavioral economists have now accumulated several decades of findings indicating that the standard economic assumptions about individual behavior are not accurate, that people do not act rationally, that they are not perfectly self-interested, and that they hold inconsistent preferences. Moreover, and especially in recent years, policy economists have increasingly come to see that these deviations from the standard assumptions about behavior matter for economic policy. The most celebrated example is the use of defaults in retirement savings: policies encouraging firms to automatically enroll their workers in 401(k) plans, rather than waiting for individuals to sign up on their own, seem to encourage participation and savings in those plans to an extent that is extremely difficult to rationalize under standard assumptions about preference and choice (Madrian and Shea 2001).

Here we take up the question of how to think about incorporating results from behavioral economics into tax policy. Because a complete reconceptualization of tax policy from the ground up is beyond the scope of a single review paper, we take the approach of working through the implications of behavioral economics in a series of extended examples, from each of three distinct levels of analysis for tax policy: understanding the welfare consequences of taxation, using the tax system as a platform for policy implementation, and employing taxes as an element of policy design.

Welfare consequences. Perhaps the central concern of tax policy, from the perspective of economics, is understanding how taxes matter for welfare in order to better design taxes that are maximally efficient and equitable. To do this, economists have developed models of deadweight loss and incidence. And based on these models, derived results for what optimal taxes look like—results along the lines of Mirlees (1971) for taxes on labor, and along the lines of Ramsey (1927) for commodity taxes. In applying these models to the practical matter of policy design, these results are often incorporated by way of rules of thumb for what “good” taxes looks like: they are simple, they impose low rates on wide tax bases, they are imposed on relatively inelastic goods, and so on. Crucially, however, the underlying models that generate these results depend centrally on how individuals respond to taxation. In the standard model, the key factors for understanding both tax efficiency and tax incidence are elasticities. But elasticities are simply a parameterization of a behavioral response. And behavioral economics shows that how people respond to taxes is less straightforward than the standard model supposes. Imperfectly rational people will respond to taxes in a way that is mediated by psychology. The case we review here is for rethinking tax simplicity.

Platform for policy implementation. For a number of reasons, both economic and practical, a wide variety of public policies operate through the tax system. So, for example, some transfer policies, like the EITC, are a part of the tax system. But other platforms are available. Transfers, for example, can be done as standalone programs such as TANF. Determining whether or when it is desirable to implement policy through the tax code depends in part on how individuals behave. That is, it depends not on how individuals respond to the taxes themselves, but how they interact with the features of the system in place for tax collection. The issue we discuss here is the attractive automaticity of the tax system.

Element of policy design. Taxes are one tool among many in the policymaker toolbox. So for example, in discussions of policy options for curbing carbon emissions, one tool policymakers can reach for is a carbon tax. In some sense, the core idea of using taxes as an element of policy design is that tax policy can be used to change behavior. And as behavioral economics has informed how economists understand individual behavior, it also informs how economists understand what levers are more or less effective for changing behavior. As a result, behavioral economics changes standard conclusions about the usefulness and effectiveness of taxes as elements of policy. We discuss this in the context of the problem of understanding how best to use taxes for fiscal stimulus.