Andrew Gilligan: Britain is in a mess, but not as much as you might think

Is Britain in a mess? Yes.

Is it in as much of a mess as we think? Not necessarily.

Significant parts of the political debate, on left and right, are based on official statistics which are wrong. And the wrongness more often seems to be in one direction – towards the gloomy, the pessimistic and the negative.

In 2022, the Office for National Statistics (ONS) produced numbers showing UK growth was so weak that the economy was still 1.2 per cent below its pre-pandemic size. These were understandably used to denounce Boris Johnson (who I served as a special adviser) for leaving Britain in the “economic slow lane” and producing the “worst recovery from the pandemic in the G7,” since that is indeed what they showed.

In the early autumn of 2023, after Boris was out of office, the growth figures for that same period were dramatically raised to, in effect, find an extra 1.8 percentage points, or £45 billion, of GDP down the back of the sofa. Britain’s economy, it turned out, had not shrunk at all. It had grown by 0.6 per cent, faster than Germany’s and about the same as France’s. Of course, far fewer people noticed the new figure than noticed the original one, and even fewer gave us credit for it (and 0.6, to be fair, is still hardly stellar.)

The GDP numbers were estimates, so the changes were a revision, not an error, as better data came in. Data during the pandemic and since was and is harder to collect. But revisions to GDP, as in this case, have been upwards more often than they have been downwards. Further vital ONS economic data, for instance on the labour market and producer price inflation, had to be stopped altogether because it was so unreliable. And in other key official statistics there is a provable bias towards saying that things are worse than they are.

Take the official figures on child poverty. These could hardly be more politically sensitive. The statistic that 31 per cent of British children grow up in poverty is endlessly used by Labour MPs. It featured in probably the key event in Keir Starmer’s demise, the successful backbench rebellion against welfare cuts.

But is it true?

It is based on defining as poor anyone whose household income is less than 60 per cent of the median average. With allowances made by the statisticians for the costs of having children, this means that for a family with two young children an income of £39,000 a year makes you poor. Meanwhile, the recent rise in child poverty (from 27 per cent) under the wicked Tories turns out to be substantially due to higher immigration; recent migrants have more children.

The figures also rely for their estimates of income on responses to the so-called “family resources survey,” where people are asked to say how much money they have coming in. But this survey significantly under-reports people’s income. It reports that they get a total of £191 billion in welfare benefits. The actual amount paid out, £234 billion, is more than 20 per cent greater. If you adjust for this, a significant chunk of Britain’s “child poverty” disappears.

This is now what seems likely to happen: according to reports last month, the data will start using the actual value of benefits paid, meaning that the number and proportion of children in poverty is revised down to reflect, well, reality.

Then there are the official statistics for police use of force. Incredibly, a police officer handcuffing someone – even if they are compliant, as around two-thirds are – counts as a use of force. So does drawing a baton, even if you don’t use it. Yes, NOT using a baton is a “use” of force. This has allowed activists to issue press releases attacking the police for “using force” 812,000 times in 2024/5. But 433,000 of those incidents involved either compliant handcuffing, or drawing but not using a weapon such as a baton.

Only, perhaps, in dear, cock-eyed old Britain would official statistics slander the country and its people as poorer, and its police as more violent, than they actually are. Is it a plot by lefty civil servants and pressure groups? Not directly, I shouldn’t think – but maybe indirectly, sort of.

Last year’s Devereux review, into the failings of the ONS, tells us that the bad data was part of a programme called “Ambitious, Radical, Inclusive Economic Statistics” and drily notes: “Ensuring the relevance of ONS activity to wider political debate… has had the (unintended) effect of de-prioritising the less exciting, but nonetheless crucial, task of delivering core economic statistics of sufficient quality to guide decision making.” Perhaps they tried harder to be radical and inclusive than they tried to be, you know, right.

I foresee, too, golden new opportunities for statistics-related political baloney around Labour’s soon-to-begin mandatory ethnicity pay gap reporting. Will the figures compare like with like? Will they, for instance, take account of the fact that British people of colour are, on average, younger, and that younger people of all races are paid less? Will they reflect other nuances and complexities – for instance, that some ethnic groups are wealthier on average than whites, and that there are big differences in average income between people of the same race (such as between British Indians and British Pakistanis; or British Africans and British Caribbeans?)  Or will the new figures be crude, the cue for yet more handwringing, self-flagellation and activist rhetoric about our blanket, undiminishing ‘raaaacism’?

You might say: yes, but bad statistics don’t change the reality.

GDP is what it is, whether the ONS reports it right or not. I disagree. Bad statistics may be self-fulfilling – reducing business or consumer confidence, for example – and thus having real effects on the economy. They may lead policymakers to make the wrong decisions. And they create a political climate where we may be talking about the wrong issues in the wrong way.

Any new right-wing government will need to get to grips with all this, as part of its reform programme for the British state, as part of standing up better against unrepresentative pressure groups and lobbyists, and as part of resetting the baseline to reflect the truth of the country’s actual problems.

Those problems are great enough without needing to exaggerate them, and without giving our opponents official falsehoods to use against us.