The relevance of Wagner's law to Zambia(1980-2013)

Abstract

Wagner (1883), predicted that economic growth would be accompanied by an increase in state activity (growth of government Spending). Thus, the causality according to Wagner‟s law is running from economic growth to government spending. This study focuses on how relevant the theory postulated by Wagner can apply to the Zambian situation. The statistical software package used for all the analysis in the study is STATA (12.1). The study made use of annual secondary time series data from the World Bank database for the period 1980 to 2013. The variables used are all in real terms and were logged so as to achieve stationarity in lower levels and allow for the easy interpretation of the elasticities obtained. The first step in the analysis was the determining of the stationarity of the variables and we found the data was non-stationary but became stationary after first differencing. We then proceeded to test for cointegration; the results showed that our variables are not cointegrated. This allowed us to proceed in using the ordinary least squares method for estimation. The study used four versions out of the six versions of Wagner‟s law. The results show that economic growth is significant in influencing government expenditure in two versions of Wagner‟s law while the other two versions had invalid results. Furthermore, Granger causality test were used to show the direction of the relationship between the variables and the results indicate that economic growth Granger causes government spending in the Peacock-Wiseman and Goffman models. There is a one-direction relationship from economic growth to government spending in the analysis. The results also proved that Wagner‟s law is valid for Zambia as the coefficient of economic growth was greater than one as postulated by the Peacock-Wiseman and Goffman models. It is important to note that Wagner stated that this relationship would hold irrespective of the prevailing situation hence the use of just two variables only. Emphasis must be made that the author tried introducing other variables that could affect government expenditure like population, structural adjustment programmes (SAPs), fluctuations in copper prices, even the use of dummy variables. All these variables did not significantly influence the results or change them by greater margins. This therefore strengthened Wagner‟s belief that the use of just the two variables was enough irrespective of the prevailing situation (in the appendix there are regression tables where population was included as an independent variable and the result prove this point). The above results indicate that Wagner‟s law can be used for policy analysis and justification of increasing government expenditure with the increase in economic activity. In addition, the model stipulates that increase in government expenditure is due to increase in social spending on education, health and social protection, a condition suited for Zambia. Furthermore, the study recommends further research on this topic using a variety of appropriate econometric procedures so as to have a sufficient pool of knowledge from which policy analyst can consult when instituting certain policies to do with public expenditure and economic growth. The paper clearly shows that Wagner‟s law is relevant to Zambia.