Measuring the Impact of Fiscal Policy on Economic Growth and Income Inequality

Abstract

Globally increasing income disparity, both across and within countries, suggests attention be paid to the role of fiscal policy in mitigating the problem. However, conventional wisdom suggests this can be achieved only at the cost of economic efficiency. This suggests fiscal policy has two incompatible major, long-term targets: economic growth and income equality. Thus economic growth, income inequality and fiscal policy variables are neither independent nor separately determinable. The implication is that it is essential to determine empirically the impact of fiscal policy on economic growth and income inequality and to do so in a way that allows for their interdependence. Consequently a conceptual framework is developed in which economic growth, income inequality and fiscal policy are interdependently determined. Within this framework, and based on a large body of empirical research, a system of simultaneous equations (SSEs) is constructed, with each endogenous variable treated as a function of the other endogenous variables, and also as a set of controlled variables taken from the theoretical and empirical literature. The following six hypotheses are developed from the literature in such a way that they can be parameterized and tested as individual system coefficients of the SSEs. The first hypothesis is that redistributive government expenditures involve a trade-off between economic growth (efficiency) and income equality (equity). The second hypothesis is that direct taxation receipts reduce income inequality, while indirect taxation receipts and non-redistributive expenditures increase income inequality. Within hypothesis 3, redistributive government expenditures are assumed to largely reduce income inequality when they are financed by direct taxes rather than by debt. Hypothesis 4 posits that non-redistributive expenditures reduce economic growth when they are financed by direct taxes rather than by debt. Outside the policy arena, hypothesis 5 is that income inequality reduces economic growth, while hypothesis 6 is that economic growth reduces income inequality. The SSEs are then specified as an empirically measurable simultaneous equations model (SEM), taking country- and time-specific unobservable factors into account. The SEM is estimated using the three-stage least-squares method (3SLS) to test the validity of the hypotheses using the three-year averages of data for 1995–2015 for a balance panel of 19 high-income OECD countries. A reduced-form version using a structural vector autoregressive (SVAR) model and a structural vector error correction (SVEC) model is also estimated using country-specific annual time series data from Australia and Sri Lanka for 1965–2014 and 1981–2013 respectively. A small SVAR model for an open economy is constructed for contemporaneous identification. Based on evidence of one cointegrating vector among the variables, a SVEC model is specified for the long run and is estimated to examine the impact of three permanent fiscal shocks (i.e. direct and indirect taxation receipts and government expenditure) on aggregate output and income inequality. Estimation of the SEM provides statistically significance evidence for all except the first and fifth hypotheses. An increase in redistributive expenditures reduces income inequality but is without a statistically significant impact on economic growth. The total net effect of direct taxation receipts, indirect taxation receipts and non-redistributive expenditures on income inequality is positive: increases inequality. Financing non-redistributive expenditures by debt is less harmful for economic growth than financing them by direct taxes, while lowering non-redistributive expenditures increases economic growth and decreases income inequality. The Australian SVEC model shows that an increase in direct taxation receipts permanently reduces per capita real GDP but has no statistically significant impact on income inequality. Conversely, an increase in deficit-financed government expenditures does not affect per capita real GDP but permanently reduces income inequality. An increase in indirect taxation receipts permanently increases income inequality, while the adverse effect of indirect taxation receipts on income inequality is greater than the redistributive effect of government expenditure. The Sri Lankan SVEC model shows that an increase in deficit-financed government expenditures permanently reduces both income inequality and per capita real GDP. An increase in indirect taxation receipts permanently increases income inequality. However, indirect taxation receipts have a statistically significant positive impact on long-run per capita real GDP. From these results a range of fiscal policy implications are drawn. For the SEM model this includes increasing redistributive expenditures financed by direct taxes to reduce income inequality, while cutting non-redistributive expenditures to help achieve both efficiency and equity objectives. The SVEC models imply that Australian fiscal policy did little to either influence the efficiency–equity trade-off or ameliorate its effects in the long run, while for Sri Lanka continued reliance on indirect taxes exacerbates equity problems.