Britain has a data problem that is raising fears of policy mis-steps at a critical juncture for the UK economy.In recent months, serious issues have been exposed at the Office for National Statistics and the Bank of England (BoE). Even the government’s own accounts are under scrutiny. For users and observers, the worry is that bad data leads to bad policy decisions.The collapse in standards has come at an inconvenient moment. The BoE is cutting interest rates, the government is issuing near-record levels of debt and Chancellor Rachel Reeves has little budget headroom against a new and untested fiscal target. A misleading dip in growth or a questionable jobs report could provoke a financial markets response with nasty consequences for borrowing costs.“Changing the goalposts for the measure of debt at the same time as other statistics have become unreliable, alongside increased global political volatility, can only increase the political risk premium that should be built into UK assets,” said Helen Thomas, founder of research consultancy Blonde Money and an adviser to George Osborne when he was chancellor of the exchequer.Investors are demanding more yield to hold UK government debt. The rate on 10-year gilts is currently around 10 basis points higher than comparable US debt. That’s historically unusual, with Treasuries averaging a 66-basis-point yield premium over the last 10 years.The most acute problem at the moment is with the Labour Force Survey but consequential errors are not isolated to the ONS. The BoE has been attacked for its “shocking” data infrastructure, the UK spending watchdog refused to sign off on the latest public-sector financial statements and the Office for Budget Responsibility found an £18bn ($23bn) error in its calculation of the chancellor’s new debt-reduction rule.Data quality matters because it determines policy. Like the European Central Bank, the BoE insists it is “data dependent,” taking decisions on rates meeting by meeting. “When the map doesn’t fit the territory, it makes it hard for policymakers to navigate,” said Peel Hunt Chief Economist Kallum Pickering.The LFS poses the most immediate risk. The BoE says the speed at which it will cut rates will be determined in part by inflationary pressure in the labour market. Whether employment is rising or falling is a key piece of information. However, it’s not clear what is happening because the data are in disarray.The government thinks employment is “probably falling,” Tom Younger, deputy director of labour market analysis at the Department for Work and Pensions, told lawmakers this month. The LFS says it is increasing.The UK is not unique. The US has also seen a fall in survey response rates for its jobs numbers, but not on the same scale.Britain’s LFS is so “misleading” it is “not fit to be used for critical decisions like setting rates or making fiscal projections,” Adam Corlett, principal economist at the Resolution Foundation think tank, said after the ONS warned its figures may not be usable until 2027. Resolution analysis based on administrative data such as tax records found the employment rate to be materially higher than officially stated.Frustration has been building at the BoE as officials try to calibrate how quickly to move rates out of restrictive territory. The most stinging criticism has come from rate-setter Swati Dhingra, who said that if India can produce accurate labour market data with a population of more than 1bn, Britain should be able to do so too.The bad jobs data has wider implications than monetary policy. The government is spending millions of pounds on ways to tackle inactivity, which is around 1mn higher than before the pandemic with the bulk of the surge due to long-term sickness. Yet the programme may prove to be a waste of both time and taxpayer money if the numbers turn out to be a mirage. This week, Minister for Employment Alison McGovern raised the question in parliament: “Is our strategy wrong given the revisions?”Companies are also affected. “It’s a big issue in terms of business planning as well as policymaking,” said Oxford Economics Chief UK Economist Andrew Goodwin. “If we don’t know what the true level of unemployment is then it’s very difficult for banks to model how many mortgages might fall into arrears.”Despite its concerns with the ONS, the BoE is no paragon of data virtue. In April, former US Federal Reserve Chair Ben Bernanke said the BoE’s data handling infrastructure was “seriously out of date” and “not adequately maintained.”The deficiencies he identified shocked Cambridge University Economics Professor Petra Geraats into asking at the BoE Watchers Conference in November whether there had been “negligence from a governance point of view.” Bernanke’s revelations were “shocking,” she said. “How on earth did it end up getting so bad?” The BoE has since committed millions of pounds and asked recently appointed Deputy Governor Clare Lombardelli to clean up the mess.The government, too, is having trouble with its numbers. For the first time, the National Audit Office refused to sign off the Whole of Government Accounts because vast tracts of local government data were missing. The gaps were so “large and pervasive” that the NAO could not “give any opinion at all” for 2022-23, it said.Now that the government has adopted a fiscal target based on “public sector net financial liabilities,” it could make forecasting more difficult as the accounts are used by the OBR, the arbiter of the public finances, to measure how much the public sector owns and how much it owes.Even the OBR has had a data wobble. An error in its forecasts for the Conservatives’ final budget turned into a headache for the new Labour government. Reeves switched the gauge of debt in her fiscal rule to PSNFL to allow more room for investment, only to discover the OBR had mistated the figures back in March to the tune of £18bn. It means her margin against fiscal shocks has been reduced to just £15.7bn.The decline in UK data standards can be traced to both a lack of oversight and underinvestment. Bernanke said the deterioration at the BoE was due to cheap “makeshift fixes.” Andrew Sentance, a former rate-setter, said the ONS suffered from “inertia and a lack of willingness to get on with things.”Adding to the list of its shortcomings, the statistics body this month revealed it had undercounted migration in the year to June 2023 by 166,000 - enough to fill a city the size of Oxford. Migration is not only a hot political topic in the UK, but officials require reliable population projections to plan for housing, public services and welfare spending.“There is always something going wrong with official statistics. It’s the nature of the beast because the economy is always changing so methods have to adapt,” Sentance said. “But the ONS is very slow off the mark, so it becomes a much bigger issue. They tend to suppress problems rather than redress them.”Goodwin is equally concerned. “This is a major problem for the whole economy, and I don’t think the ONS is moving anything like quickly enough to resolve it.”