Academic journal article Quarterly Journal of Business and Economics

Infrastructure Capital and Private Sector Productivity: A Dynamic Analysis

Academic journal article Quarterly Journal of Business and Economics

Infrastructure Capital and Private Sector Productivity: A Dynamic Analysis

Article excerpt


In recent years, a sizable literature has emerged concerning the effect of public infrastructure capital on the productivity of labor or capital in the private sector. Two empirical approaches have been used to study this issue at the regional, national, and international levels. One approach uses a production function that includes the stock of public capital in addition to private inputs (Aschauer, 1989a, 1989b, 1989c; Eisner, 1991; Holtz-Eakin, 1994; Hulten and Schwab; 1991a, 1991b; Moomaw, Mullen, and Williams, 1995; Munnell, 1990a, 1990b; Tatom, 1991). The second approach uses a cost function dual to a production function that includes public capital as an argument (Berndt and Hansson, 1992; Lynde and Richmond, 1992, 1993; Nadiri and Mamuneas, 1991; Conrad and Seitz, 1994).

The findings concerning the productivity effect of public capital have been mixed. Whereas some find that public infrastructure capital contributes to private sector productivity or lowers production costs, others find no discernible effect attributable to public capital. Much of the time-series evidence on the productivity effect of public capital is based on models that suffer from a number of drawbacks.

One problem is that most studies assume unidirectional causality from public capital to private sector productivity. But one-way causation is only one of four possibilities. Another possibility is reverse causation whereby higher levels of output in the private sector lead to the accumulation of public capital. As Eisner (1991, p. 49) puts it,

[s]erious questions remain ... as to which is cause and which is effect. Does public capital contribute to more output? Or do states that have greater output and income, as a consequence of having more private capital and labor, tend to acquire more public capital...?

A third possibility is a two-way causation or feedback between productivity and public capital. Finally, it is possible that there is no causal relationship between these two variables.

A second problem with much of the literature is the potential endogeneity of public capital. This is conceptually different from the issue of causality, as neither concept is a necessary nor a sufficient condition for the other. Almost all empirical studies of the productivity effect of public capital implicitly assume that public capital and private inputs are exogenous. Unless this assumption is satisfied, the single-equation approach that is used prominently in the literature will produce unreliable results.

A third problem is that most studies do not distinguish the short- and long-run effects of infrastructure capital on private sector productivity. In many instances the univariate properties of the data are not examined. Given that aggregate time-series data are typically nonstationary in the level or logarithmic form, not removing the unit root from the data can lead to spurious results. Recognizing this fact, some authors specify their models of productivity in the first-differenced form which means that they lose long-run information in the data. As Engle and Granger (1987) have shown, a first-differenced model of nonstationary variables is misspecified if the elements of the data vector are cointegrated, in which case an error-correction model is the proper specification.

This paper examines the relationship between public capital and private sector productivity while addressing the problems above in the context of the dynamic framework developed by Johansen (1988) and Johansen and Juselius (1990).(1) A similar approach has been undertaken by Flores de Frutos and Pereira (1993).

Using annual data covering the 1948-1987 period, we find that there is a long-run relationship among private sector productivity, private inputs of capital and labor, and core infrastructure capital. Moreover, public capital exerts a positive influence on private sector productivity along this long-run path, although this effect is only statistically significant at low levels of confidence. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.