Abstract. The objective of this study is to examine the long run as well as short run relationship between government expenditure at aggregate as well as disaggregate level and economic growth in Pakistan. The study uses six functional forms of Wagner’s law and further classify these forms to incorporate disaggregate expenditure. Study uses annual time series data of Pakistan from 1976 to 2015 and applies Engle and Granger cointegration test for long run relationship, while Granger causality test is employed for short run analysis. The study found that no long run relationship exists between GDP and total expenditure as well as its sub-components i.e. expenditure on current subsidies, defence expenditure, current expenditure, and developmental expenditure, while long run relationship exists among GDP and expenditure on social, economic and education services. Unidirectional causality in favor of Wagener’s hypothesis exists from GDP to expenditure on current subsidies, expenditure on social, economic and education services, defence expenditure, and current expenditure, while unidirectional causality in favor of Keynesian hypothesis flows from developmental expenditure to GDP in the short run. Government has to invest more on human resource development to achieve sustainable economic growth in the long run. Government has to reallocate more resources towards developmental projects to achieve higher economic growth in the short run.

Keywords: government expenditure, economics growth, cointegration, causality, Pakistan.