Dynamic Output and Employment Effects of Public Capital

Some of my previous research investigates the static, or short run impacts of changes in the public capital stock on economic performance. For instance, in Aschauer (1997a) I use state level data for the period 1970 to 1990 and find that the public capital stock is an important determinant of the rate of growth of output per worker. Specifically, a one standard deviation increase in public capital (relative to private capital) induces an increase in the growth rate of output per worker of some 1.4 percent per year. Similarly, in Aschauer (1997b) I determine that the public capital stock is also a key factor lying behind the rates of growth of output and employment, with a one standard deviation rise in public capital generating an increase in the growth rate of output and employment, respectively, of about 1.6 and 0.5 percent per year. However, these findings leave open the question of the dynamic, or long run effects of public capital on the economy. To answer this question, it is just as important to understand the dynamic interrelationship between productivity, output, and employment as the economy evolves over time as it is to know the effect of public capital on the initial growth rates of these variables. For example, depending on the persistence of the increase in the productivity growth rate, any particular static increase in productivity growth can translate into a rather small of large increase in the long run level of output per worker. This paper explores these persistence concerns by stimulating the dynamic, long run effects of public capital on output and employment. Section II lays out a dynamic model relating output and employment growth to public capital, initial output, and initial employment--a minimalist model capable of capturing in a compact fashion the interrelationship between output and employment as the economy evolves over time. Section III presents empirical estimates of the model based on fixed effects regression analysis of U.S. state level date over the period 1970 to 1990. Section IV employs the estimated model to stimulate the long run impact of public capital under two scenarios-- where the state unemployment rate and the labor force, respectively, are assumed to be exogenous. Section V concludes the paper with some suggestions regarding future research.