Public infrastructure and private economic performance : evidence from U.S. panel data

Abstract

Government can affect private economic activities through many means: the legal system and regulatory mechanisms, the tax and subsidy systems, lending and other monetary activities, and the services from public infrastructure. This dissertation investigates the potential contribution of public infrastructure to private economic activities. Importantly, this research attempts to provide an answer to the question: is public infrastructure undersupplied? The contribution of different types of public infrastructure (i.e.. highways and streets, water and sewer systems, and other public buildings) on private production is investigated using time-series and cross-sectional data of the 48 contiguous states over the period of 1970-1986. A Cobb-Douglas production function is estimated with unobserved state-specific effects. Measurement errors in the total public capital stock and its components are detected and rectified. It is found that water and sewer facilities (and in most cases, highways and streets) consistently provide productive contributions to private productivity. This study also attempts to identify the contribution of different types of public infrastructure to regions and industries. The error component model is specified using the translog production function. The results show that regions and industries do not react uniformly to various types of public investment. For example, transportation infrastructure is more productive with respect to the agricultural sector in the Northeast than in any other regions and industries. To answer the undersupply question, the short-run variable cost function is utilized. The system of first-order conditions is estimated by Iterative Seemingly Unrelated Regressions. The main result shows that both private and public capital have been under-capitalized in the sense that their long-run desired levels are greater than their actual levels. Finally, the internal rate of return to quasi-fixed inputs are provided. In conclusion, the relationship between public infrastructure and private economic performance is a complex one. Nevertheless, the positive contribution of public capital is discovered. Both private and public capital have not reached their long-run equilibrium levels. However, this does not imply that the level of total tangible capital should be raised without a careful cost-benefit analysis.