Wealth inequality: measurement, resilience, and an anti-Rentoul hypothesis
Yesterday John Rentoul published an article in which he sought to refute accepted wisdom about the growth of inequality in Britain:
After all, “everyone knows” that the rich are getting richer and the poor are getting poorer, don’t they? Well, this is where it gets interesting, because what “everyone knows” is not what is happening. The gap between rich and poor has not changed significantly for about 20 years, not since the increase in inequality that occurred when Margaret Thatcher was Prime Minister in the 1980s.
The argument now developing, with Alex Hern responding and Rentoul re-responding, is focused entirely on income inequality. This is odd, because Rentoul’s original claim, reflected in the headline, is that wealth inequality, not just income inequality, has remained largely constant.
So I’ll respond to that aspect of Rentoul’s claim.
Rentoul’s evidence is the 2012 ONS Wealth and Assets Survey, and specifically Table 5 in chapter 2. This shows that overall the Gini Coefficient for Wealth reduced very slightly between the 2006-08 and the 2008-10 period.
This data, for Rentoul, is enough to show that:
the degree of inequality of wealth hardly changed from 2006 to 2010, becoming slightly more equal. Unless something dramatic has changed in the past two years, we would expect that trend to continue.
I assume Rentoul’s logic for this statement is that relative wealth is more stable ovet time than relative income because it reflects past accruals amd is therefire unlikely to be changed dramatically in a short space of time. This might be fair enough, but it ignores one vital consideration: richer people not only have more wealth, they are also more resilient to economic shock.
Take financial wealth, one of the four sub-categories of total wealth measured by ONS. Table 1 of chapter 2 shows that in 2006-08 14.9% of households owned UK shares, rising to 15.4% in 2008-10. Table 2 then shows that in 2006-8 the mean average of UK shares per household was worth £25,200, dropping to £22,700 in 2008-10, while the median values were £4,000 and £2,000 respectively. This means that while quite a few people had fairly small share ownership, a relatively small number had a very large amount of their wealth held in shares, and continued to do so through the two periods surveyed (though the table also shows a significant increase in gilt/bond ownership, some of which will have been a shift from shares as safe havens were sought out post-2008.
The key point, though is that UK shares, taking the FTSE 100, actually had 17.84% less value in June 2010 than they did in July 2006, as a result of the 2008 crash, with the major loss made just after the first survey period ending June 2008, before a partial recover from early 2009 onwards. This means that, for the Gini coefficient to remain stable (and on the reasonable assumption that it is the wealthier households who own most of these shares), wealthier households must have accrued other types of wealth during the same period. Yet we also know that substantive wealth is not reflected in the value of shares at any one point in time; rather it is reflected in the liberty to hold on to shares and sell them on at the point of greatest advantage (see als ch.3, annex 1, confirming this).
And the same logic applies to other types of non-cash wealth, held disproportionately by the wealthy, whether it be other financial investments held in less risky but lower value funds like bonds, or property values, which also suffered a dip in the post-crash period (except the very luxurious ones).
In short, the fact that the Gini coefficient remained more or less the same across the two survey periods reflects not the stable inequality that Rentoul would have us acknowledge, but a growing inequality masked by the difficulty inherent to valuing the resilience of wealth. That growing inequality – growing if measured as overall asset base – in turn reflects the ability of wealthier to do well in times of economic crisis.
Meanwhile, we also know from the ONS report (ch3, p.34) that resilience amongst poorer households is falling:
In 2010/11, almost a third of households reported not having any savings at all, up from 24% in 2006/07. almost a third of households reported not having any savings at all, up from 24% in 2006/07.
Moreover, it needs to be recognised that the Gini coefficient, while it is a useful tool, is still a fairly blunt instrument, and tells us relatively little about distributions of inequality. In fact, it may well be that the similar Gini coefficients in the two periods mask quite dramatic changes in the internal composition of that inequality, in ways similar to those shown in this paper about the more refined use of the Gini coefficient to study the aftermath of the 1990s economic crisis in Korea, in which detailed analysis shows that:
the average contribution of between-group inequality during the six quarters is about 122% of the changes in overall inequality. Therefore, it is believed that severe deterioration in between-group inequality has more than offset minor improvement in within-group inequality, resulting in deterioration of overall income inequality. In other words, the worker households in Korea are undergoing a distinct process of income stratification parallel with the concentration of income.
It may be that a similar secondary analysis of the ONS data might reflect some of the Labour narrative around the ‘squeezed middle’. It is beyond the scope of this blog to test the hypothesis but, if that were to be done, we could at least thank John Rentoul for flagging up the ONS data again, even if his own use of it is sub-standard.