Essays on the composition of government spending and economic growth

The link between fiscal policy and economic growth has raised a great deal of debate at both theoretical and empirical level. Public expenditure and national income have been the focus of public finance, since the amount of public expenditure has been increasing over time in almost all countries in the world. The aim of this study is to empirically examine the effects of different types of government expenditures on economic growth, with particular attention to the Italian case. The first chapter of this dissertation offers a literature review on the relationship between the size of government and economic growth to show how findings follow a contrasting pattern. Moreover, the more recent strand of the literature attempts to evaluate the influence of functional breakdown of public spending on economic growth, thus going beyond the simple division between productive and unproductive expenditure. Studies that disaggregate public expenditures in healthcare, education, defence, infrastructures, housing sector and other expenditures categories show that the results change according to the specific countries under scrutiny, the analysed period and methodologies applied. Chapter two and three examine the empirical relationship between government spending composition and economic growth within an endogenous growth framework as developed by Ghosh and Gregoriou (2008) with two public goods, one being a priori more productive than the other. In particular, the added value of the second chapter lies in its empirical analysis which is based on a panel of 19 Italian regions over the period 1996-2007. Our interest is to estimate the influence of economic and functional expenditure categories of the general government on real per capita GDP growth rate. The functional breakdown of expenditures includes six components, which have been obtained aggregating the thirty sectoral expenditure components on the basis of the available data. Interestingly, inside functional categories we distinguish between capital and current items. Using dynamic panel GMM estimators, economic and functional components seem not to have any significant effect on growth. However, the results improve when we take into account spatial correlation through a spatial lag model. Although there is no evidence of spatial autocorrelation in our specifications, the results show that the capital and current expenditure behaviour is driven by capital transfers and financial assets. In addition, it is also induced by current expenditure on personnel, purchases on goods and services, interest expense and other unclassified current expenditures. As for the functional components, the main result is given by capital expenditure on economic affairs, such as infrastructures, which plays an important role in augmenting the economic growth rate among Italian regions in the period investigated here. Current expenditure on health, and capital and current expenditure on general public services had a negative and statistically significant effect on growth. The third chapter offers an empirical investigation of the relationship between government spending composition and economic growth by using a historical time-series (for the period 1862-2007) on the Italian case. Our aim is to estimate the effects of economic and functional components of government spending on economic growth both in the short- and long-run. After testing our series for unit roots with structural breaks, we divided the sample period into two main sub-samples and estimate the influence of the different components of public spending by means of an ARDL model. Regarding the first sample (1970-2007), the most surprising results is the negative effect of capital expenditure on economic growth, both in the short- and in the long-run. In particular, capital expenditure appears to have been excessively high during the period under scrutiny, thus resulting unproductive at margin. On the other hand, current expenditure has an insignificant effect on economic growth both in the short- and in the long-run. Among functional components, expenditures on defence and economic affairs have a statistically significant effect on economic growth in the short-run. In contrast, the latter maintains a significant or even negative effect on real per capita GDP in the long-run only. As for the second sample period (1862-1969), we split it endogenously so as to establish two structural break dates. Overall, the results vary across sub-samples, but educational spending shows a positive and statistically influence on per capita GDP in the long-run.