The Relationship Between Government Size and Economic Performance with Particular Application to New Zealand

This paper argues that the ambiguous relationship found between government size and economic growth in the empirical growth literature can be attributed at least in part to the different time series characteristics of the two series. For most countries economic growth is stationary while government size has almost always been non-stationary, implying that the finding of a negative correlation between the two series is likely to be spurious. From a time series perspective, economic performance and government size should be approached by first looking for evidence of cointegration between the two non-stationary series: government size and per capita real output. Then, should evidence of cointegration exist, the analysis can be extended to look for evidence of short run movement about the implied long run model through an error correction model. When this approach is applied to New Zealand, evidence of a long run, inverted U-shaped relationship between government size and private per capita output is revealed. The corresponding error-correction models reveal that while transitory short run increases in government expenditure size are generally ineffective in increasing real output, increases in government spending relative to taxation are. On the other hand, the concomitant increase in national debt is found to decrease real output in both the long and the short run. Since the net effect of fiscal expenditures will be the sum of the latter three (and conditional on its initial position), it is not surprising to find that conclusions on fiscal effectiveness will be highly problematic.