A £2.2tn fall in the valuation of UK household wealth following a methodology change by the country’s statistics agency was based on a “fundamentally flawed” analysis of pension wealth, according to a leading think-tank.
The Institute for Fiscal Studies on Friday accused the Office for National Statistics of “jumbled economic reasoning” after an adjustment by the agency last year knocked one-third off estimates of British household private pension wealth between 2018 and 2020 overnight.
The think-tank’s report is the latest setback for the ONS, which last week postponed the release of trade data after identifying an error and has come under fire from politicians and the Bank of England for long-running problems with a key survey on the labour market.
In December, the ONS made several changes to how it values pensions as part of its official estimates of household wealth.
The biggest change was to move away from the use of market interest rates to value future pension benefits and towards greater use of a rate-based on predicted economic growth, known as the superannuation contributions adjusted for past experience (Scape) rate.
The agency’s change in methodology was “fundamentally flawed” because Scape was not in any meaningful sense a rate of interest and did not reflect the true value of a pension’s benefits, the IFS said in its report.
The change cut the ONS estimate of private pension wealth from £6.4tn to £4.2tn, equivalent to a 14 per cent reduction in total household wealth, the IFS added.
It followed a recommendation from the Government Actuary’s Department to “minimise undue volatility” in the value ascribed to defined benefit pensions, which promise guaranteed pensions based on salary and length of service.
However, the IFS said the “misguided” valuation approach was “a mistake, making an already flawed methodology substantially worse”.
“It is quite possible that the ONS’s revised estimates of household wealth are actually further from the truth than those it started with,” it added.
Carl Emmerson, IFS deputy director, said that since the second phase of the government’s pension review would examine “the adequacy of retirement saving, good quality data are needed to understand what the status quo looks like and inform decisions around whether reforms, for example to automatic enrolment, are needed to improve things”.
A collapse in responses to the ONS labour force survey, which is not expected to be replaced until at least 2026, has left the BoE and ministers without reliable data on employment for almost 18 months.
It also triggered the Office for Statistics Regulation, the regulatory arm of the UK Statistics Authority, late last year to warn that 14 connected data sets could no longer be classified as “official statistics”.
The agency has started using Scape — now set at consumer price inflation plus 1.7 per cent a year — to value the post-retirement phase of a pension, having previously used a market-based annuity rate.
Previously Scape, which is applied by the government to calculate the cost of providing public sector pensions, was used to value defined benefit pension promises only for those not yet retired.
The IFS added that the ONS had not revised historic data to reflect errors it had identified, meaning official estimates of DB pension wealth between April 2010 and March 2020 “should be regarded as unreliable”.
But the think-tank welcomed two other changes the agency made to the way it values pensions in December, one to account for inflation before a worker retires and the second to more accurately reflect retirement age.
The ONS said in a statement that any method must make “broad assumptions” about future economic growth. “We sought advice from a range of experts and users before finalising the method used. We outlined these decisions at the time of publication,” it added.
