TA Re-examination of Wagner’s Law Based on Disaggregated U.S. State-Local Government Expenditure

We used U.S. state-local data on real government expenditure and personal income over the period 1957-2006 to test the validity of Wagner’s law. Our informal analysis of level data provided prima facie evidence in favor of Wagner’s law. In particular, total expenditure and several of its sub-categories grew at rates (significantly) above the rate of growth of personal income. However, our formal analysis- based on two cointegration techniques- provided consistent evidence suggesting that, with the exception of social services and income maintenance spending to income ratio (ssim), no other spending ratio was part of a cointegrating relationship with real per capita personal income (pcpi) and, thus, error-corrected over time. Moreover, ssim was found to have an income elasticity coefficient consistent with Wagner’s law and part of a (bidirectional) causal relationship with pcpi. Our findings imply that the ability to counter cyclically adjust “public welfare” spending during the current economic downturn may be limited and federal assistance may bring a welcome relief to state and local governments.