The Linkage Between Government Expenditure and National Income: the Case of Bangladesh

Abstract

Objective: This study examines the applicability of Keynesian and Wagner's law in terms of gross national expenditure (GNE) and gross national income (GNI) in Bangladesh. Theoretical Framework: According to Wagner's Law, a country's government spending rises in tandem with its national income. The opposite is true, according to the Keynesian Law, which claims that government spending increases GDP. Methodology: This study is a quantitative and secondary data based analysis. It makes use of time series data spanning the years 1981 to 2020. All variables' data were gathered from the World Bank's compilation of global development indicators. In order to exclude years with outlier data, the current study did not include variables' data from 1971 to 1980 (The abnormality occurred in 1971 for the war of liberty; in 1974 for the Famine and in 1975 for the political violence). Once the sequence of integration with ADF had been determined; the ARDL bounds testing technique and Granger causality test were used. Results and Discussion: According to the findings of the ARDL bounds test, there is a long-term correlation between government spending and gross domestic product. Government spending likely contributes significantly to the causes of national income, according to the slope coefficients from long-run equations. Granger causality findings show that government spending and national income have a one-way causal link but not the other way around. Research Implications: One possible explanation for this result is that recent government spending on development projects boost up the national income. Originality/Value: We contribute to the literature in two important ways. The first contribution is that we focus on more robust estimation methodologies. Unlike previous studies of Bangladesh context, we apply ARDL bounds test approach to co-integration. In addition to having advantageous small sample characteristics, the ARDL limits testing approach does not necessitate the same order of integration in the variables. With these considerations, this technique suits better in case of Bangladesh given that data would be of a small span. And second, we examined both of the Wagner’s Law and Keynesian Law for Bangladesh with two separate equations, unlike the previous studies on Bangladesh who investigated the Wagner’s equation only.