The information comes from the Wealth and Assets Survey (WAS) 2008/10. Comparisons can also be made with the WAS undertaken in 2006/08. This is a valuable data source. More detailed data from the latest survey will be published later this year.
The first thing that stands out is how much greater wealth inequality is than income inequality. The Gini coefficient (0 = perfectly equal; 1 = perfectly unequal) for total wealth was 0.62 in 2008/10 and 0.61 in 2006/08. This is almost twice the Gini measure for income inequality of 0.35 for Scotland and 0.36 for the UK in 2009/10. We also note that the degree of income inequality in Scotland and UK was considerably higher than the OECD average of 0.31.
The stability of the Gini coefficient between the two surveys suggests that household wealth inequality while high did not change much over the two years. However the same cannot be said for wealth inequality across the regions and countries of the UK.
The chart below shows the change in regional and country median household wealth Compared to Britain between 2006/08 and 2008/10 in percentage points.
What these data show is that there was a marked regional/country divide in Britain in the change in median household wealth. The median household in London dramatically improved its wealth position relative to the median household in Britain. Median households in Scotland, Yorkshire & the Humber, North West, South East and England as a whole also improved their relative position but by a much lower order of magnitude than London.
In contrast, households in Wales lost out compared to their counterparts in the rest of Britain. And in England, households in North East, East Midlands, South West, and West Midlands lost out considerably relative to the median household in Britain.
What accounts for these changes over the two years?
It is not clear.
There has been a large rise in pension wealth between the two surveys due, it would appear, to improved measurement. But this should not affect median household income given the concentration of pension wealth. It would affect the (arithmetic) mean, however. In addition, detailed pension and financial wealth data, which might help answer this question, will not be available until later in the year.
It must be remembered that the period between the two surveys coincides with the Great Recession and fall in property and other asset prices. Net property values fell in most regions except South West and Scotland. And as the Report notes:
The largest decreases in the mean net property wealth occurred in the North East, Wales and the North West, which showed decreases of 8.8, 8.7 and 8.6 per cent respectively.
Change in net property values might account for some of the relative deterioration in North East and Wales but runs counter to the change in the North West and South West.
Changes in physical wealth also seem unable to account for the regional variation since physical wealth is a small proportion of overall household wealth and the movements were not well correlated with the change in regional wealth.
So, we are left with change in financial wealth as the likely factor accounting for the regional variation in median household wealth. This begs several questions and among them what is the link to median household wealth from
- rising salaries in banking and financial services in London, Scotland and North West?
- rising CEO and senior officer salaries in many companies headquartered in London?
There is evidence - see here - that labour's share of income is falling relative to capital in the US and other OECD countries. But higher incomes, whether from labour earnings or not, allow higher savings and accumulation of assets and in this way income and wealth inequality are linked. But Thomas Piketty and Emmanuel Saez in this 2006 paper found
that the dramatic recent increase in income concentration is a primarily a labor income phenomenon and this has not yet translated into a dramatic increase in wealth concentration.
But maybe there are regional effects if neutral at the national level.
We need further data before we can begin to provide answers to these questions.