In 2016, after his regime had survived an attempted military coup, Turkey’s President Recep Erdoğan stood up in the national parliament to denounce an unlikely foe: the American credit rating agency Standard & Poor’s. The agency had downgraded Turkey’s credit score, concluding, perhaps unsurprisingly, that it could be risky to invest in a country facing armed bids to take over the state. This threatened to raise the cost of borrowing money for Erdoğan’s government. “What is your deal with Turkey?” the furious President said. “Turkey is not a member of your agency. Do not mess with us.”
Though rarely expressed so honestly, these are sentiments that leaders around the world could sympathise with. Most governments claim to be answerable to their people, yet if they rely on debt to fund their activities, they will find their sovereignty constrained by analysts sitting in distant financial districts. James Carville, the outspoken former adviser to Bill Clinton, famously quipped that he would like to be reincarnated as the bond market, since “you can intimidate everybody”.
When Labour entered power last year, Chancellor Rachel Reeves faced the same problem that bedevilled Erdoğan and Carville: how to reassure the bond markets that they could keep lending at reasonable rates. Her answer was to commit to a set of “fiscal rules”, pledging that government debt will be falling by the end of the parliament. This was not an original strategy: Reeves was merely adapting what has become a convention for British chancellors since George Osborne created the framework for austerity in 2010. But during Reeves’ hapless tenure, it has become clear that the mantra of fiscal responsibility conceals another tendency: government by metrics.
Bereft of imagination in a rapidly changing world, our political class clings ever more tightly to numerical targets and indicators, hoping to convey an impression of competence and stability. Instead, they have only reinforced the sense that we are ruled by management consultants rather than leaders. Nor have metrics provided the bulwark against chaos that they promised; for the numbers, it turns out, do not always show what people think they show.
An over-reliance on statistical models and measurements comes with two great dangers. The first is that the seemingly objective quality of numbers can seduce us into believing that they present an infallible picture of reality. In fact, data are gathered and modelled by institutions with all the shortcomings and flaws that one would expect of a human undertaking. The second problem is that our behaviour is increasingly determined by what we can measure — or think we can measure — rather than what is actually important or desirable.
“Our political class clings ever more tightly to numerical targets and indicators, hoping to convey an impression of competence and stability.”
Both problems are abundantly illustrated by British politics today. When it comes to her fiscal rules, Reeves is essentially relying on economic forecasts from the Office for Budget Responsibility, an independent body, to tell her how much money she has to spend. The government ties its own hands, inviting the OBR’s economists to rate its finances before the markets get a chance to. The problem is that these forecasts are imprecise, and regularly revised as the various metrics change. Consequently, Reeves’ chancellorship has been an endless drama of budgetary “black holes” and vanishing “fiscal headroom”, as she scrambles to tailor her policies to the shifting estimates. Measuring has displaced thinking, to the point where the government’s decisions are being dictated by a numerical target rather than an actual plan for how to match the public finances to the nation’s needs.
In July, for instance, Reeves tried to tackle the problem of runaway welfare spending. Instead of setting out a strategy for reforming a fiendishly complex and thoroughly broken system of benefits — a problem whose roots go deep into British society — she set about salami-slicing various budgets to satisfy the fiscal rules. This so outraged Labour backbenchers that Reeves and Keir Starmer had to abandon their cuts in any case.
That same month, as if to emphasise the futility of the government’s approach, the OBR suggested it would be reconsidering some of its forecasts, which were too optimistic. The adjustment will likely force Reeves to raise billions of pounds in tax at her upcoming Budget. As the blogger M.F. Robbins has aptly put it, “you cannot micromanage policy based on statistical noise. The more you try to do this, the more you’ll find that, in fact, the noise is managing you.” The Chancellor is now trapped between a bond market waiting to see if she will keep her promises, and a party that wants her to abandon them. Last week, former transport secretary Louise Haigh published a frontal assault on the Chancellor’s strategy, demanding a less prominent role for the OBR in government. A group of 40-odd Labour MPs in Red Wall seats has previously expressed similar frustrations.
Labour’s love of metrics does not end with fiscal rules. Hoping to show a firm, technocratic grip on the nation’s problems, Starmer’s government has issued a blizzard of “missions” and “milestones” based on quantifiable targets. It wants 75% of five-year-olds to reach “a good level of development”. It aims to recruit 13,000 neighbourhood police officers and 6,500 new teachers; to build 1.5 million homes and to achieve 95% clean electricity. The list goes on. At one point, the government even proposed a “digital dashboard”, so that the public could track its progress towards these goals on their smartphones.
All of this assumes that we want to treat politics like a stock market, assessing the fundamentals of a government before choosing to invest. We measure and rate everything else, so why not our politicians? One reason is that voters have an intuitive understanding of “Campbell’s Law”, which observes that a measure which becomes a target will likely come under “corruption pressures” — that is, it will be fudged or misrepresented to appear more impressive. More importantly, politics, like human experience, is fundamentally qualitative. The number of new houses does not tell us whether they are decent homes in neighbourhoods where people actually want to live, just as the number of new police officers and teachers does not tell us whether the criminal justice and education systems are working as we would like. These are ratings we make with our eyes and ears.
It’s tempting to think that numbers can build trust by helping the public to understand the work of government. But if statistical and subjective reality clash, it has the opposite effect. People do not like being told a graph proves that their perceptions are wrong. In 2024, supporters of then president Joe Biden foolishly tried to convince struggling voters that they were doing great, actually; had they not seen the economic metrics? Something similar is happening with law and order issues in Britain today, as data on falling crime rates fail to assuage anger at low-level lawlessness and police impotence. The Covid pandemic was perhaps the most ambitious attempt in history at doing politics via statistical models, and it was ultimately a disaster for public trust in institutions. People were initially willing to give up their freedoms to “bend the curve” of infections, but as time wore on, the experts who were meant to save the politicians from resentment were themselves contaminated by it.
It is true that our impressions diverge from the numbers in part because we are biased. Political partisanship has long influenced perceptions of the economy in the United States, and now the same appears to be happening in Britain too. Then again, sometimes the official statistics are just spectacularly wrong, as has consistently been the case with immigration. In 2004, Tony Blair’s government estimated that welcoming migrants from Eastern European countries to work in Britain would result in a maximum of 13,000 people arriving annually. Just four years later, the Home Office revealed that 750,000 had come already. Another discrepancy emerged when EU citizens were invited to apply for settled status after Brexit. In 2019, the government thought there were 3.7 million EU nationals in the country, yet closer to 6 million have asked to remain. Similar misjudgements led former prime minister Boris Johnson to relax visa restrictions in 2021, resulting in a surge of new arrivals.
Such errors stemmed in part from a woefully inadequate method for collecting data. The International Passenger Survey, designed in the Sixties for studying tourism, involves little more than questioning people at random at airports and other border crossings. It uses a tiny sample of migrants, and misses out many patterns of movement altogether. But until other data sources were added in the last few years — and despite immigration becoming a persistent political issue — this was the British state’s sole way of measuring it.
At one point, it could be argued that migration was an outlier in a generally solid field of British statistics. Since the Covid pandemic, though, the country’s chief number-crunching body, the Office for National Statistics, has been mired in crisis. Reports have been delayed, revised, and retracted, and it has admitted that some of its most important metrics are no longer reliable. It is a portrait in miniature of British misgovernance. In 2007, the ONS lost 90% of its London-based staff — a devastating flight of expertise — when a misguided attempt at devolution saw the agency moved to Newport in southern Wales. Governments have subsequently cut its funding, which as a cost-saving exercise is a bit like trying to lose weight by gouging out an eye. But the most pressing problem has been that fewer and fewer people are taking ONS surveys, which seems partly to reflect an unwillingness to cooperate with the state, or for that matter to answer phone calls and open the door to strangers.
This erosion of statistical competence has far-reaching implications, given the range of public and private actors who use ONS data. Their uncertainty has knock-on effects for the affordability of mortgages, the value of British businesses and assets, and the quality of government policy. Political discourse will, meanwhile, become increasingly farcical, as politicians stake their authority on numbers that provide only a tenuous picture of reality, and we will criticise them using the same illusory figures. We are already familiar with the seismic importance given to every tenth of a percentage point of GDP growth, even though such changes are far too small to measure accurately, and would tell us little of value to our lives in any case. So we are unlikely to change tack just because the measurement has been declared unreliable as well.
A grasp of metrics is essential for governing a modern society; we are right to expect that the state will make the numbers add up. The fate of Liz Truss, a prime minister effectively toppled by the bond markets when her plans were found wanting, is a reminder of the ruthless will to measurement at the heart of capitalism today. Donald Trump can bully government agencies to confect more favourable economic data, as he appeared to do last month by firing the head of the Bureau of Labor Statistics, but he cannot force businesses and investors to believe that data. Still, politics requires an insight that goes beyond our fixation with the quantifiable. The numbers do not lie, but nor do they tell us everything. It was once widely believed that demographic change in the United States would hand the future to the Democrats, and that the size of the EU’s GDP and population meant it could easily propel Ukraine to victory against Russia. Those claims appear less certain now.
Likewise, Britain’s current regime shows that an obsession with metrics is ultimately a profound weakness, mistaking the means of statecraft with its ends. Starmer and Reeves’ project appears to be falling apart before it is even clear what that project was, teaching us that a government led by managerial bean counters can be as fragile and incoherent as one led by clowns. Leaders who confuse meeting targets with governing well will achieve neither.
