INTRODUCTION

There is an enormous amount of literature on the effectiveness of state and local economic development policies. Most of the empirical literature by economists [very nicely summarized by Bartik, a major contributor (1991)] has been devoted to measuring how effective various kinds of policies are, with the effects measured in terms of employment, business starts, and new branch plants, among other things. Put baldly, what I want to argue here is that, with a few notable exceptions, the existing literature reflects a great deal of effort that could have been better spent asking different questions. What we should seek to measure in our assessments of local economic development policies is changes in the level and distribution of economic welfare.

The connection between welfare on the one hand and jobs, branch plants, investment, etc. on the other is by no means obvious or straightforward. What is straightforward are some standard propositions about the welfare economics of government intervention in the economy. These may be summarized in an entirely familiar way—unless there is either market failure or dissatisfaction with the income distribution generated by market outcomes, there is no persuasive rationale for government intervention. There is, of course, plenty of market failure that government may be able to ameliorate, and there are many reasons to want to change the distribution of income.