Over the last three decades, after the publication of Aschauer ́s (1989) empirical paper on the productivity of public capital and Barro ́s (1990) theoretical paper on the effects of government spending on economic growth, the analysis of the macroeconomic effects of public investment has attracted research interest in growth economics. The theoretical research was mainly focused on analyzing how public spending and public capital may enhance productivity and promote economic growth. At the same time, the empirical literature was trying to confirm the existence of a positive empirical relationship between productive public spending and economic growth, by employing a variety of econometric techniques, data sets, and measures of public spending. In principle, public spending enhances economic growth through its external effect in the production function of private firms. This effect can be modeled by adding into the production function either the aggregate flow of public spending, following Barro (1990), or the aggregate stock of public capital, as in Turnovsky (1997). A new line of theoretical research recognizes the possibility that different types of public spending (e.g., infrastructure, education, health, military expenses) may exert a different effect on economic growth. In this line of research, Devarajan, Swaroop and Zou (1996) develop an endogenous growth model using a production function which includes two types of government spending, Kalaitzidakis and Kalyvitis (2004) into infrastructure and maintenance expenditure, and Chen (2006) into productive and consumptive spending.
Largely overlooked in this strand of the literature has been the question of the economic growth impact of military spending. This oversight is surprising given the significant portion that military spending absorbs from public capital in both developed and developing countries. This question has received substantial attention in the early economic growth literature especially in developing countries initiated by Benoit’s (1973; 1978) empirical study. Benoit (1973; 1978), in her pioneering work in this area using data from 44 developing countries, found that defense spending does indeed stimulate economic growth. Government provision of national defense which maintains property rights can be viewed as maintaining both internal and external security, increasing hence the incentives to accumulate private capital and attract foreign investments. On the other hand, the defense sector also provides a variety of public infrastructure like roads communication networks etc, while also enhances human capital accumulation through training and provision of educational services. These are at least some ways that military spending can foster economic growth domestically. However, military spending also diverts resources that could more productively be employed in sectors of the economy other than defense. In that sense, defense spending may degrade long run economic growth. Hence the net effect that military spending has on economic growth appears to be ambiguous. This results has been confirmed by many subsequent empirical studies devoted in addressing the validity of Benoit hypothesis utilizing a variety of empirical models and different data sets.
The empirical research on the productive effects of public capital has followed two basic lines of research. The one line of research, following Benoit (1973; 1978) directly estimates aggregate production functions to identify the productive types of public capital. The most indicative examples in this strand of the literature are the studies by Baffes and Shah (1998) and Evans and Karras (1994). The empirical results often validate Benoit’s initial assumption on the productivity of military spending. The other line of research estimates Barro-type growth regressions in which different types of public expenditure may serve as explanatory variables, as in Barro and Sala- i-Martin (1995, ch. 12). In most cases the different types of public expenditure enter linearly into the growth regression, as, for example, in Devarajan, Swaroop and Zou (1996). The results usually indicate a non significant or negative effect of public expenditure variables on economic growth. These results are interpreted as an over- provision of public capital.
Based on this conflicting empirical evidence, recently Shieh et al., (2002) developed an intertemporal optimizing endogenous growth model to analyze how the government’s resource allocation between the defense and the non-defense sectors affect economic growth and social welfare. They concluded that there is indeed an optimal share of defense expenditures that maximize the economic growth rate but this rate is smaller than the welfare maximizing share. In the same line of research and building upon Shieh et al., (2002) theoretical development, the present paper derives the growth-maximizing values of the shares of public investment allocated to the two different types of public capital (i.e., military spending and public infrastructure), as well as the growth-maximizing tax rate. Then the paper proceeds with an empirical investigation of the theoretical implication of the model assuming that both the effect of the share of public investment and that of the tax rate on the long-run growth rate are non-linear, following an inverse U-shaped pattern.
Using data of public capital and military capital formation from a sample of both developed and developing economies, we derive empirical estimations that confirm the theoretical implications of the model. The innovative feature of our empirical analysis is that we test a specific non-linear functional specification implied by the theoretical model. In addition tackling the problem of spurious regression arising from the non-stationarity of the data set, the empirical part of the paper makes use of panel unit-root tests and cointegration analysis to conclude that there is a fairly strong evidence in favor of the hypothesis that long run causality exists between the optimum allocation of government expenditures among military spending and public infrastructure.