The British state is blind

Britain isaonsonsonsgdpmpM bit bigger than it thought. In 2023 net migration stood at 906,000 people, rather more than the 740,000 previously estimated, according to the Office for National Statistics. It is equivalent to discovering an extra Slough. New numbers for 2022 also arrived. At first the thought net migration stood at 606,000. Now it reckons the figure was 872,000, a difference roughly the size of Stoke-on-Trent, a small English city.If statistics enable the state to see, then the British government is increasingly short-sighted. Fundamental questions, such as how many people arrive each year, are now tricky to answer. How many people are in work? The answer is fuzzy. Just how big is the backlog of court cases? The Ministry of Justice will not say, because it does not know. Britain is a blind state.This causes all sorts of problems. The Labour Force Survey, once a gold standard of data collection, now struggles to provide basic figures. At one point the Resolution Foundation, an economic think-tank, reckoned the had underestimated the number of workers by almost 1m since 2019. Even after the rejigged its tally on December 3rd, the discrepancy is still perhaps 500,000, Resolution reckons. Things are so bad that Andrew Bailey, the governor of the Bank of England, makes jokes about the inaccuracy of Britain’s job-market stats in after-dinner speeches—akin to a pilot bursting out of the cockpit mid-flight and asking to borrow a compass, with a chuckle.Sometimes the sums in question are vast. When the Department for Work and Pensions put out a new survey on household income in the spring, it was missing about £40bn ($51bn) of benefit income, roughly 1.5% of or 13% of all welfare spending. This makes things like calculating the rate of child poverty much harder. Labour s want this line to go down. Yet it has little idea where the line is to begin with.Even small numbers are hard to count. Britain has a backlog of court cases. How big no one quite knows: the Ministry of Justice has not published any data on it since March. In the summer, concerned about reliability, it held back the numbers (which means the numbers it did publish are probably wrong, says the Institute for Government, another think-tank). And there is no way of tracking someone from charge to court to prison to probation. Justice is meant to be blind, but not to her own conduct.Crime, immigration and benefits are not the only fraught political debate fought in the statistical gloom. Rows about how to care for the elderly are shaped by the image of grannies forced to sell their homes to pay for a care home. It is the main political obstacle of arguably Britain’s biggest public-policy failure. How many pensioners are forced into this unhappy situation? No one knows.Rather than trying to learn what it can, the state sometimes deliberately compounds its ignorance. Patients can opt out of their medical data being used for research and planning, making the National Health Service harder to improve. Since the option to opt out was introduced, almost 4m patients have done so. Likewise, the future of the census, which has run since 1841, is in doubt. It is old-fashioned, instantly outdated and, at a cost of just shy of £1bn, expensive. Yet it is still useful. Scrapping it is the equivalent of responding to failing eyesight by gouging out an eye.Simple solutions exist. The state could invest in better spectacles. In a system that adores false economies, skimping on the statistics agency is the worst value. Political prioritisation would help. Data-flows between the Home Office and the Ministry of Justice fall off the agenda if officials have to spend their days wondering if it is legal to stick a prisoner in a van and drive him round the 25 due to a lack of cells.What would help more is a shift in attitude. If the state can compel people to sit in a stale room for hours to decide if someone is a thief, it can force people to fill in a form. There is no good reason to pander to the paranoid when it comes to health data. No such opt-out exists for education data, points out Tim Leunig, a researcher at Nesta, also a think-tank. To put it bluntly: the state is always allowed to know about the thick, so why not the sick?A naive idea that more data inevitably lead to better governance should also be abandoned. It is too easy for a government to be sidetracked by what it can see. Britain now has excellent data when it comes to sewage spills (of which there were 464,056 in 2023). Every year brings a new high not because nefarious water companies suddenly decided to pump effluent into waterways but because new laws introduced stricter monitoring. The result? British politics is now obsessed with crap, with 1,700 mentions of “sewage” in Hansard over the past four years.Humility would help above all. Though accurate data are harder to come by, analysis of available statistics has never been easier. The result is often junk. Grand narratives rest on shaky numbers. Why did Britain, alone among rich Western countries, suffer a boom in economic inactivity? Everyone had an answer. For the right, the culprit was a soft-touch welfare state; for the left, it was a broken health service. Few alighted on the most plausible explanation: that it was a statistical artefact of increasingly dodgy data, rather than a moral crisis that confirmed political preconceptions.A scepticism about statistics once so ingrained it became a cliché (“lies, damned lies and statistics”) has been replaced by arrogant credulity just as British data sources have developed serious flaws. Instead, smug sayings such as “the plural of anecdote is not data” abound, as if “data” are anything other than anecdotes, piled high and categorised. Immaculate analysis is impossible when the state is blind. All that is left is for the state to be curious about what it does not know and humble about what it thinks it does.