Good morning from New Economy Brief.
Local election post-mortems are flooding in, with lots of commentary blaming Labour’s losses and Reform’s record gains on the decisions to cut the Winter Fuel Payment and disability benefits. (See More in Common’s presentation for a good roundup.) But what if the decisions behind these cuts were based in part on inaccurate data?
This week's New Economy Brief explores ongoing problems with some of the most important economic indicators collected and calculated by the Office for National Statistics (ONS). Getting them wrong can have far-reaching consequences because they are used to develop the forecasts that directly inform large parts of government policymaking.
–
The UK’s macroeconomic institutions are flying blind.
Economic forecasts play a critical role in setting policy, and two of the most important official forecasters – the Bank of England and the Office for Budget Responsibility (OBR) – rely heavily on ONS data. But concern has been steadily mounting about that data’s reliability, with the Labour Force Survey (LFS) coming under particular scrutiny.
The LFS is the UK’s official employment data series, and has been struggling with declining response rates since the pandemic, hitting a low of 17.4% in 2023. So low, in fact, that the LFS lost its accreditation as an official statistic in November 2023. The Resolution Foundation estimated that the LFS may have “lost” up to 930,000 workers, and accused the ONS of “leaving policymakers in the dark”. Chair of the Treasury Select Committee Meg Hillier warned that “these delays will make some of the most consequential decisions taken by the Treasury and Bank of England challenging at best and misinformed at worst”. ONS Chief Executive Professor Sir Ian Diamond has said it may take until 2027 to introduce a more reliable replacement.
This employment data informs decisions taken by the Bank of England’s Monetary Policy Committee about the rate of inflation and appropriate path of interest rates, as well as government decisions on welfare policy. If it is inaccurate, the political and economic implications of inaccurate data are significant. Such is the concern that the Governor of the Bank of England himself, Andrew Bailey, has called it a "substantial problem".
But this substantial problem doesn’t end with the Labour Force Survey. The ONS’s Living Costs and Food Survey, used to help calculate GDP figures, also suffers from low response rates. And in March this year the ONS uncovered further issues, this time related to its producer price index and services producer price indices, which "gauge changes in the price for goods and services bought and sold by manufacturers and service sector companies.” Both are used to estimate the rates of economic growth and inflation.
–
Is inflation being calculated wrong?
Of course, the ONS understands the importance of getting its statistics right, and it has been trying to improve. One example is its approach to the monthly Consumer Prices Index (CPI) – the UK’s headline inflation statistic. The ONS plans to input more real-time data from supermarket scanners, for a more accurate picture of how prices change across the economy. This is because scanners record the prices that people actually pay and include discounts from loyalty cards like Tesco’s Clubcard. This is important because just under a quarter of all UK groceries are bought at Clubcard discounted prices.
This new data was meant to become part of the CPI calculation in March this year, but has now been delayed until 2026. But its absence could be distorting our understanding of how high inflation actually is. As FT’s Alphaville pointed out last week, the ONS itself says using scanner data to recalculate CPI for March 2024 shows inflation was just 2.8% – 0.4 percentage points below the 3.2% it thought at the time.
Overstating inflation has huge implications for the government. If the Bank of England’s Monetary Policy Committee had better data about the true rate of inflation, would interest rates be lower? It is difficult to speculate, but overstating the rate of inflation has other effects too. Around a quarter of the UK’s £2.8 trillion national debt is linked to inflation, so if the ONS overstated inflation just a little, forecasts of higher borrowing costs could eat into projections of the government’s ‘headroom’ under the government’s fiscal rules. FT Alphaville cites one calculation which suggests that overstating inflation by 0.2% could have cost the government £1.4 billion unnecessarily, nearly as much as the savings made from cutting the winter fuel allowance.
–
Political risks in making cuts based on uncertain forecasts.
The ONS’s exclusion of supermarket scanner data on changing food prices is just one example of how bad data could end up causing bad decisions on major policies – especially when so much weight is put on staying on the right side of a highly uncertain forecast from the OBR. For instance, a briefing from Invest in Britain argued that: "The haphazard decisions around welfare spending at the Spring Statement exemplify the perverse effect of making policy to meet targets based on uncertain forecasts."
Cuts driven by short-term bookkeeping. It is important to remember what drove the benefit cuts announced by the government since coming to power. UK borrowing costs rose earlier this year due to worsening global instability, which prompted the Chancellor to find £4.8 billion in savings by cutting disability benefits (which the government’s impact assessment notes could push 250,000 people into poverty). How this decision was made was the subject of a widely reported controversy. Essentially, it was clear that the policy was being decided for accounting reasons – in order to find enough savings to restore the government’s headroom to £10 billion – not because it was good policy. Or put another way, ministers were tweaking policy to meet the forecast, rather than tweaking forecasts to meet the policy, as was normal before the OBR’s establishment in 2010. Meanwhile, the Winter Fuel Payment cut announced last year was part of a package to fill what Labour called the ‘black hole’ in the public finances left by the Conservatives.
‘Budgetarianism’ leads to bad policy. Jagjit Chadha, former head of NIESR, has previously critiqued this approach as ‘budgetarianism’: where government policy is decided upon based on whether an arbitrary fiscal rule will be met, and the focus of national debate is distorted from what fiscal policy should accomplish, to “whether there is any room for a sliver of expenditure one way or the other.” We have previously covered the risks of cutting spending to meet the OBR’s highly uncertain forecasts of the government’s income and spending needs several years in the future. Now there are concerns about the underlying data these forecasts are based on, and the harm this could be doing to policy, particularly in politically sensitive areas such as benefits.
The Guardian's Pippa Crerar and Jessica Elgot report “growing anxiety at the top of government” and say “pressure from MPs for a major rethink of economic strategy is likely to mount in the coming days”, with the cuts to the Winter Fuel Payment and disability benefits seen as important drivers of Labour’s losses at the local elections last week. More accurate data probably wouldn’t have been enough on its own to prevent these decisions. And many of the downsides of short-termist, budget-driven policymaking would still apply, even with better data. But it’s also clear that the Chancellor’s last-minute decision to impose extra cuts to disability benefits at the Spring Statement was entirely driven by a desire to meet her self-imposed fiscal rules. And with fiscal headroom so tight, small changes to the data underlying economic forecasts can have an outsized impact on whether these rules are met or missed – which in turn can drive the kind of back-to-front and damaging policy decisions we saw at the Spring Statement. Better data can’t fix this, but it can help.