“It is generally recognized that more saving takes place in communities in which the distribution of wealth is uneven than in those in which it approaches more closely to modern conceptions of what is just.” T.S. Ashton (1948) The Industrial Revolution 1760-1830, p. 7.
I came across this quote while reading Ashton’s account of the industrial revolution and was taken by this sweeping assertion especially in light of current public debate such as Obama’s State of the Union pledge to tackle inequality as well as research on economic inequality and how it is harming economic growth. If new investment and capital formation is a function of saving, then if inequality fosters greater saving, it should also play some role in fostering investment and economic growth.
Economists like Kuznets seemed to think that income inequality was associated with economic growth at least when economies were starting to develop – remember the Kuznet’s Curve? You can get a pretty good overview of the literature and the issues in Growth, Inequality and Globalization by Philippe Aghion and Jeffrey Williamson (2001). There is an argument that inequality is good for economic growth because it provides an incentive to get rich – I suppose this is Kevin O’Leary’s line in response to the recent report by Oxfam that 85 people control one half of the world’s wealth.
Yet, there is also the view that rising inequality can be bad for economic growth. See for example work by Jared Bernstein or this IMF report by Berg and Ostry. Whether inequality somehow undermines consumer spending or leads to under-investment in human capital that undermines economic growth is an empirical question. What is also an empirical question is whether there is more saving when there is greater economic inequality. T.S. Ashton certainly seemed to think so.
So, I got together some data on household saving rates, economic growth rates, and Gini coefficients for both disposable household income and household wealth for the OECD countries (sources below). The Gini coefficients for household income are for the late 2000s, the Gini coefficients for wealth are for the year 2000 and the household savings rates out of disposable income and real GDP growth rates are an average for the period 2001 to 2011. This data is difficult to get perfect time period correspondence for. Nevertheless, here are some plots. The results are interesting but not definitive in providing an answer.
Figure 1 and 2 rank the Gini Coefficients for both household disposable income and household wealth for the OECD countries. Figures 3 and 4 plot the respect Gini coefficients for wealth and income against the average annual saving rate for the period 2001 to 2011. Figures 5 and 6 plot the Gini coefficients for wealth and income against the annual average GDP growth rate (real GDP) for that same period.
The results for inequality by rank show that income inequality (Figure 1) was the highest in Chile and the lowest in Slovenia (Canada was 12th and the USA 4th). As for wealth inequality, it was the highest in Denmark and the lowest in Japan (with Canada ranked 10th and the United States 3rd). As is generally known, wealth is more concentrated than income. When wealth inequality is plotted against the saving rate (Figure 3) and a linear trend fitted, there is indeed a slight positive relationship but there is plenty of variation around the trend. There is even less of a fit when income inequality is plotted against the saving rate (Figure 4). As for economic growth, the relationship between wealth inequality and real GDP growth rates (Figure 5) is at best slightly negative. There is a more pronounced positive relationship between income inequality and economic growth (Figure 6).
Greater wealth inequality does appear to be associated with higher savings rates but greater income inequality does not appear to be very strongly related to savings rates. Wealth inequality does not appear to be positively related with GDP growth but higher income inequality appears to be positively related with higher GDP growth. Of course, these are only scatterplots and the results anything but definitive. There you have it.
Household net saving as a percent of disposable income and GDP growth rates, average for period 2001 to 2011. Source: OECD National Accounts at a Glance 2013. http://dx.doi.org/10.1787/na_glance-2013-en
Gini Coefficient for disposable household income in late 2000s. Source: Table A1.1 Trends in different income inequality measures, An Overview of Growing Income Inequalities in OECD Countries: Main Findings 2011 www.oecd.org/els/social/inequality
Gini Coefficient for household wealth in 2000. Source: J. Davies, S. Sandstrom, A.B. Shorrocks, E.N. Wolff (2009) The Level and Distribution of Global Household Wealth, NBER Working Paper 15508.