The problem with economic data is there’s just too much of it
By Simon French, Chief Economist and Head of Research
This is a column about the importance of high quality statistics. As such I offer an apology in advance as it won’t be everybody’s cup of tea. Times readers - when asked to consider the most pressing issues facing the UK economy - might cite global trade, oil prices, or the value of the US Dollar.
But such issues are largely external factors that buffet the UK economy and are hard to predict, and even harder to influence. So a more pressing issue for a mid-sized economy like the UK - which is a price-taker on international markets - is to have accurate and reliable statistics to measure incomes and prices faced by UK consumers and businesses, and inform important economic decisions. Unfortunately for the UK right now there are considerable deficiencies across a range of important data from one of our most important sources: the Office for National Statistics (ONS).
Faced by these deficiencies, the government announced in April a review of performance and culture at the ONS led by Sir Robert Devereux. The launch of this review coincided with a critical report from the Office for Statistics Regulation that concluded that the ONS must “fully acknowledge and address declining data quality”. Last month the head of the ONS, Professor Sir Ian Diamond, stepped down on health grounds - putting the organisation under interim leadership. Reviews, critical reports and leadership flux are symptomatic of the underlying problems facing the UK’s provider of official statistics.
These problems could be dismissed as a relatively costless example of a struggling public service. However poor data has a real cost for Britons. Inaccurate information means the basis on which to make good policy by national and local governments is impeded. It also acts as a barrier to private sector investment - with those deciding where to allocate capital doubting the basis on which they are making those decisions. At a time when investors across the world are questioning the wisdom of owning American and Chinese assets, having a timely and accurate story to tell about the UK is key to attracting new investment.
Of all the data deficiencies the most material surrounds the state of the UK labour market. For more than half a century the most important source of insight has been a household survey called the Labour Force Survey. But the combination of societal change and the COVID-19 pandemic has heavily impacted responses in recent years - with response rates falling from 49% in 2015 to less than 13% by mid-2023. Whilst all international household surveys have faced similar problems, the UK’s decline has been more pronounced.
Timely insight on the UK jobs market is always important, but perhaps no more so than at a time when the Labour government’s decision to levy higher Employer National Insurance, introduce an Employee Rights Bill, raise the National Living Wage at more than twice the rate of inflation, and tighten working visa eligibility are all impacting the jobs market simultaneously. Privately, the government will be asking itself whether it has overcooked its labour market reforms by introducing all these changes in one go given recent sharp declines in payroll employment. Timely alternative data from HMRC’s Pay as you Earn (PAYE) records help in this regard, but this data is subject to considerable revision, systematically underestimate UK employment, and omit almost four-and-half million self-employed workers. Policymakers aren’t flying blind, but there is thick fog.
Whilst there have been signs of a gradual improvement in the Labour Force Survey data over recent quarters, the Bank of England has cited the lack of trustworthy labour market data as impeding confidence on the correct stance for UK interest rates. This is having a tangible impact on the costs of more than nine million mortgages in the UK, and a wide range of business loans.
And when it comes to interest rates it is not just labour market data causing challenges, it is also the evolution of prices. There were audible gasps on City trading floors when the most recent UK inflation data saw prices growing at 3.5% a year, well ahead of market expectations for growth of 3.3%. Private sector economists struggled to work out how the ONS had arrived at that number, despite some hefty increase in costs at the start of the 2025/26 tax year. Then a fortnight later the ONS was forced into a rather embarrassing revision as it identified an error in the Department for Transport’s calculation of Vehicle Excise Duty. This lowered the estimate for April inflation to 3.4%. For investors worrying whether the UK will experience a sustained second spike in price inflation this was not an episode that inspired confidence. It also comes at a time when the ONS has already suspended its estimates for producer prices - a leading indicator of consumer prices - citing methodological problems.
So what can be done? Well the various reviews will run their course and their recommendations will likely trigger improvements in data quality. But there is an aspect to this ONS story that also bedevils large parts of the UK state – an attempt to do too much that is peripheral to the organisation’s core objectives. Two examples help illustrate this point.
Firstly, the ONS has decided in recent years to produce monthly GDP estimates. Admirable as it is to have timely data, only Canada amongst major economies produces equivalent data rather than the more standard quarterly release. If the job of statistics is to provide insight then there is simply too much volatility in monthly output data to provide value. Well-intentioned efforts by the ONS to point to the trends in this data get drowned out by the media’s focus on a single month. This does little to support a better understanding of UK economic trends. The second example is apparent if you scour the ONS website for recent publications. These include “Navigating Demographic Gaps in Hybrid Working” and “Short-term Lets through online collaborative economy platforms”. Both of these are interesting and insightful pieces of data analysis. But at a time when the core mission on accurate data surrounding employment, prices, population and economic growth are open to question, they are also luxuries the ONS cannot afford.
And here is the lesson from the ONS for leaders across the public sector as they face a tight Spending Review for administrative budgets - identifying what is peripheral and needs to be stopped will be central to avoiding core services being underfunded and deteriorating. The challenge facing our national statistics is an illuminating case study for a much wider program of public service reform.