Why might the rich be indifferent to income growth of their own countries?

The paper shows using empirical evidence frommore than 100 countries’ household surveys, that income gains that the rich can realize through amore unequal distribution are oftenmuch larger than the realistic gains from a distribution-neutral growth. The rich are thus more likely to support policies that increase inequality than be concerned about income growth of their countries. © 2016 Published by Elsevier B.V. It is often said that in very unequal societies the rich compose a group apart. Not only are their social mores and consumption patterns different, but their fortunes seem dissociated with those of the middle class or the poor. More recently, there has been an argument that the rich (the top 1%) from different nations form a group apart, a global ‘‘superclass’’ (Rothkopf, 2008; Freeland, 2012; van der Weide and Milanovic, 2014). The objective of this note is to consider whether there is empirical substance to the claim that in high inequality societies income of the rich is in some sense decoupled from the income of the rest of society. I do not mean it in an obvious sense that the rich simply have higher income than the others. What I mean is to look at the income gains that the rich can make from an overall increase in national income (while keeping the distribution unchanged) versus the gains that they can make from a further widening of income distribution (while keeping mean income the same). I will show that this particular trade-off varies in function of income class, and that especially for the top income classes, the gains from greater inequality tend to be E-mail address: [email protected]. http://dx.doi.org/10.1016/j.econlet.2016.08.027 0165-1765/© 2016 Published by Elsevier B.V. disproportionately high compared to the gains from an increased overall income without a change in the distribution. Income distributions in the nations of the world differ a lot. Table 1 shows, using the data from 116 countries around the year 2008, the average ventile shares and their standard deviations. (All ventile shares are calculated from micro data provided by nationally-representative household surveys. The data are available at https://thedata.harvard.edu/dvn/dv/restat/faces/ study/StudyPage.xhtml?studyId=114484&versionNumber=1More information about the database in available in Milanovic, 2015). Consider the first ventile (the poorest 5% of population ranked by income per capita). On average, across countries, the poorest ventile receives just slightly above 1% of total national income (1.064%). Inmore equal counties, the share of the bottom is greater (almost 2%), in less equal, it is under 1/2 of 1%. The standard deviation of the bottom ventile share is 0.45 percentage point (column 2). Thus the gain that an average person placed in the bottom ventile would make from moving from a distribution that ‘‘allocated’’ to him the average worldwide share of the bottom ventile to a distribution that would be more favorable to the bottom ventile (by one standard deviation of that ventile share), would be 42% B. Milanovic / Economics Letters 147 (2016) 108–111 109 Fig. 1. Distribution-neutral growth rate needed to make people from a given income fractile indifferent between growth and ‘‘conceivable’’ favorable distributional change. Table 1 Share of total income received by each ventile of national income distribution. Source:Household surveys from 116 countries around year 2008. Unweighted data. World Income Distribution (WYD) database. Ventile Mean ventile share in total income (in %) Standard deviation of ventile share (in %) Income gain from 1 standard deviation increased share (in % of own income) First 1.06 0.45 42.2 Second 1.59 0.53 33.3 Third 1.90 0.56 29.6 Fourth 2.17 0.57 26.4 Fifth 2.42 0.58 23.9 Sixth 2.67 0.58 21.6 Seventh 2.90 0.57 19.8 Eighth 3.15 0.56 17.8 Ninth 3.40 0.55 16.1 Tenth 3.68 0.53 14.5 Eleventh 3.98 0.50 12.6 Twelfth 4.30 0.47 11.0 13th 4.67 0.43 9.1 14th 5.09 0.38 7.4 15th 5.60 0.32 5.8 16th 6.22 0.28 4.5 17th 7.04 0.32 4.5 18th 8.22 0.58 7.1 19th 8.00 0.93 11.6 Twentieth (top) 19.51 5.65 29.0