Author:
Castles Francis G.,Dowrick Steve
Abstract
Recent findings concerning the impact of government spending on economic growth have been as diverse as the theories underpinning them. Proceeding from a methodological critique of prior work in the field and drawing on a variety of methodological innovations suggested in recent studies, we seek to design rigorous tests using cross-section data from 18 OECD countries over four business cycles between 1960 and 1985. Government expenditure is decomposed into various categories and economic growth is corrected for cyclical disturbances. We also control for the effects of technological catch-up, institutional sclerosis and population growth along with the mediating influences of investment and employment. A variety of diagnostic statistical tests are applied, including tests for parameter stability across countries and across periods and tests for heteroscedasticity, functional form, outliers and endogeneity. We reject the hypothesis that government expenditures reduce economic growth and indicate some areas of social expenditures which may possibly tend to increase growth. We are also able to indicate the methodological assumptions and restrictions which have led to previous contrary findings.
Subject
Sociology and Political Science
Cited by
61 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献