The relationship between income inequality and aggregate saving: an empirical analysis using cross-country panel data

ABSTRACT

If the rich save more than the poor, an increase in income inequality raises aggregate saving. We investigate whether income inequality is positively related to aggregate saving ratio by estimating a fixed-effect model based on a panel data of 48 countries for the period 1991–2010. We find evidence that aggregate saving ratio increases with income inequality using various inequality measures. In particular, the effect of income distribution on saving is greater and statistically more significant with in financially developed, rich and OECD countries. It suggests that the rich save much more than the poor under advanced financial system and in a rich country. We also find that the relationship between income inequality and saving ratio is closer in the 2000s than the 1990s. This finding may result from financial development and the high income level in the 2000s.

KEYWORDS:

• Income inequality
• saving ratio
• GINI coefficient
• marginal propensity to save
• financial development

JEL CLASSIFICATION:

• E21
• D31

Notes

¹ According to the World Bank 2010 standard, we exclude countries of which average income level is below 1005 dollars. Musgrove (1980) argues that income distribution does not affect saving rate in poor countries since many agents are near subsistence. We also consider that poor data quality of low-income countries may bias the result.

² Because the WDI does not offer GINI index for some advanced countries like Germany, UK, US, and so on, we derive GINI data for OECD countries from the OECD data set.

³ We employ only 40 countries in the regressions using income share because of insufficient data availability. Excluded countries are listed in Appendix.

⁴ It is unclear which variable is appropriate. Schmidt-Hebbel and Servén (2000) employ GNS while Smith (2001) uses private saving rate. Recently, Alvarez-Cuadrado and Vilalata (2012) adopt personal saving rate. Because many countries do not offer private or personal saving rate, we do not adopt them to include as many countries as possible in the sample.

⁵ Since King and Levine (1993), the ratio of financial asset value to real production is used as a measure of financial development.

⁶ Eight countries are excluded for M3 data and one country is excluded for private credit data. Excluded countries are listed in Appendix.

⁷ We do not report the results of regressions using H20/L40 and M60 because the results are almost the same.

⁸ At a first glance, this result seems different from Smith (2001) who claims that a negative association between income equality and saving ratio is weaker in financially developed countries. It is, however, noted that he measures equality by the income share of the poorest 40% of the population. The poor tend to have incentive to save because they have limited access to financial system. An increase in the income share of the poor has small impacts on saving behaviour of the poor in developed financial system where the poor do not face serious borrowing constraints. The negative effect of income equality on the aggregate saving is, therefore, weak in developed financial system. In contrast, we claim that the difference in saving behaviour between the rich and the poor is larger, thereby the effect of income inequality on the aggregate saving rate being greater in financially developed countries.

⁹ Many papers including Blinder (1975) and Dynan, Skinner, and Zeldes (2004) offer evidence that the marginal propensity to save increases with disposable income at the micro level.

¹⁰ When we split the sample into OECD and non-OECD groups, we find similar results. Strong relationship between income inequality and saving ratio exists for OECD countries only.