WIREDGOV

Flawed official pension wealth statistics have overstated inequality between old and young while understating it between graduates and non-graduates.

Private pensions are the single largest component of UK households’ wealth. Official estimates of private pension wealth, produced by the Office for National Statistics (ONS) using the Wealth and Assets Survey, have undergone significant methodological changes in recent years. While the ONS changes rectified some longstanding problems, they were not applied to official estimates for past years (meaning that the official statistics are not comparable over time) and introduced important new problems. 

Until now, this has left policymakers without reliable data on how wealth is distributed among British households and how this distribution has evolved over time.

A new IFS report, released yesterday, addresses these issues. It sets out a new measure of private pension wealth that is calculated coherently and consistently over time, aligning the valuation of pensions with the way other forms of wealth are measured.

Key findings from the report include:

• Under our methodology, 54% of total household wealth was held in pensions in 2020–22, compared with only 38% according to official statistics. This difference results from improving the way in which promised future pension incomes are ‘discounted’ to convert them into today’s terms. This means that, at the time of the last survey, people who held more of their wealth in pensions were better off than previously thought relative to people who held more of their wealth in other assets.
• Our new methodology reveals that the share of household wealth held by people aged 20–39 increased significantly over the 2010s, from 10% in 2010–12 to 18% in 2020–22. In contrast, in the official statistics, this share was constant at 8% between 2010–12 and 2018–20 (when the ONS estimates are comparable over time), increasing to 11% in 2020–22. In other words, the gap in wealth between old and young has been overstated by official statistics since 2010. This difference arises because, during this period of low interest rates, the ONS methodology particularly undervalued private pension wealth for those further from retirement. 
• The share of total household wealth held by the top 10% fell gradually from 45% in 2006–08 to 42% in 2018–20 (and remained at 42% in 2020–22) according to our estimates. Between 2006–08 and 2018–20 (when the (flawed) official ONS estimates are comparable over time), the current ONS estimates instead suggest that the top 10% share fell only marginally, from 46% to 45%.
• People with degrees are more likely to hold private pension wealth whose value was understated by the ONS approach. As a result, gaps in wealth by education were wider in 2020–22 under our methodology than suggested by official statistics. Median wealth for graduates was 50% higher under our methodology than under the ONS methodology in 2020–22; for people with no formal educational qualifications, the difference was less than 25%.
• The differences in methodologies have particularly important effects during periods of ultra-low interest rates. As market interest rates have risen since 2022, the two approaches are likely to produce more similar estimates in the forthcoming 2022–24 data release. However, they could diverge again going forwards should market interest rates change once more.

Laurence O’Brien, Senior Research Economist at IFS and an author of the report, said:

‘Good policy starts with good data. Methodological changes and inconsistencies have made it difficult to get a clear picture of wealth inequality in Britain, and how this has changed over time. Addressing these issues reveals that the share of wealth held by the top 10% has fallen by more since 2006–08 than suggested by official estimates. In addition, young adults held a much higher share of wealth than previously thought, while gaps in wealth between educational groups were wider than official estimates show.’

How is wealth distributed across British households? Reassessing the valuation of pensions