Local government expenditures and regional growth in Sweden

Abstract: This thesis consists of five papers, which concern expenditure decisions and economic growth within Swedish local government.Paper [1] explores the hypothesis that local fiscal shocks have short run effects on revenue and expenditure decisions made within local government. It is found that although fiscal shocks do not affect local authorities' revenue decisions, they do induce municipalities to change expenditures and financial costs along with short term loans. Local authorities are also found to respond more powerful to unfavorable fiscal shocks (deficit shocks) than to favorable fiscal shocks (surplus shocks).In Paper [2], we study the hypothesis that local (municipal) expenditures, in part, can be explained by regional (county) expenditures. We formulate and estimate a demand model for municipal services that is defined conditional on the county expenditures. The results imply a positive dependency between the provision of county and municipal services. Moreover, the results suggest that the hypothesis of weak separability between the provision of county and municipal services can be rejected. In addition, we cannot reject the hypothesis that the regional expenditures are weakly exogenous in the local expenditure equation.In Paper [3] the existence of spillover between Swedish municipalities in the provision of recreational and cultural services is analyzed. A representative .voter model is derived and the demand for recreational and cultural services is estimated using spatial SUR techniques. The results suggest a negative relationship between recreational and cultural expenditures provided by neighboring municipalities, which indicates that these services are substitutes.Paper [4] concerns the regional growth pattern in Sweden by analyzing what factors might determine the growth rate of regional average income levels and the net migration rates. Our results suggest a negative dependence between the initial average income level and the subsequent income growth, which supports the conditional convergence hypothesis. Among other things, we also find that the initial endowments of human capital have a positive effect on subsequent net migration while the initial unemployment rate is found to have a negative impact on net migration.Paper [5] complements the analysis made in Paper [4] by studying which factors determine average income growth and net migration at the local level of government. The conditional convergence hypothesis cannot be rejected. Local government investments are found to have a positive effect on the subsequent net migration while leaving the growth in mean income unaffected. This may indicate that the net migration caused by these investments does not significantly affect the proportion of skilled and unskilled labor.