Investigating Short Run Causality between Real GDP and Government Expenditure in India Since 1950s

Abstract

This paper has examined short run causality between government expenditure and GDP in India during 1951-2013 using a Toda-Yamamoto (1995) modified Granger causality approach under VAR environment. Exponentially detrended annual time series data on GDP and government expenditure at constant prices are used.  Structural break point unit root tests are conducted besides the usual unit root tests to determine the order of integration of each variable. Tests for structural breaks reveal significant breaks in both time series around the period 2001-04.  Government expenditure is found to significantly Granger-cause real GDP but the converse is insignificant implying that Wagner’s law is inapplicable. The study thus suggests uni-directional causality from government expenditure to GDP. Moreover government expenditure in India has a long-run co-integrating relationship with real GDP and therefore short run causal relations may be anticipated.

Keywords: Real GDP, government expenditure, Granger Causality, Toda-Yamamoto approach and VAR.

JEL Classification: C32, E50, E52, E59.