Inequality and Economic Growth: A Global View Based on Measures of Pay

This paper discusses two issues in the relationship betweeninequality and economic growth: the data and the econometrics.We first review the income inequality data set of Deiningerand Squire (D&S), which, we argue, fails to provide eitheradequate or accurate coverage, whether through time or acrosscountries. We then introduce our own measures of the inequalityof manufacturing pay, based on UNIDO's Industrial Statistics.These provide indicators of pay inequality that are more stable,more reliable and in our view also more comparable across countries,than the D&S data. Turning to the fabled "Kuznets" relationship between inequalityand economic development, we diagnose several common econometricproblems in the literature, including measurement error andomitted variable bias. By taking steps to account for theseproblems, and by introducing a more complete panel data setbased on pay inequalities, we seek more reliable inferencesconcerning the relationship between inequality, national incomeand economic growth. We find evidence that generally supportsKuznets' specification for industrializing countries: pay inequalitytends to decline as per capita income increases, though withsome tendency for the relationship to curve up for the richestcountries. After 1981 two findings emerge. First, per capitaGDP growth slowed dramatically in most countries, increasinginequality along the augmented Kuznets curve. Second, thereis a global and macroeconomic effect that produces rising inequalityin our data, independent of GDP or its changes. The timing ofthis effect suggests a link to the high real interest ratesand global debt crisis of the period beginning in 1982. (JELC23, D31, J31, O11)