The real black hole is not in public finances

NHS workers
NHS productivity is clearly slipping - Jeff Moore/PA Wire

The Nobel Prize-winning economist, Paul Krugman, once famously said that “productivity isn’t everything, but in the long run it is almost everything”. If so, the UK has a huge problem – and that problem is particularly acute in the public sector.

Official estimates released last week suggest that public service productivity was an astonishing 8.5pc lower in the second quarter of this year than the pre-Covid peak in the final three months of 2019.

Moreover, even before the pandemic, this measure had barely improved over several decades. The latest annual estimates suggest that public service productivity was about 5pc lower in 2023 than in 1997.

What does this mean? Since 2019, public service output has grown by a little under 10pc, but the inputs used to produce that output (mainly labour and materials) have increased by nearly 20pc. The difference between these two numbers is the shortfall in productivity.

Indeed, data from NHS England suggest that hospital productivity was 11pc lower in 2023-24 than pre-Covid levels.

And why is this so important? Even without thinking about the missed opportunity to improve the quality of services, this is an appalling waste of money.

This year alone the Government (that is, the taxpayer) is expected to spend £453bn on the day-to-day running of public services, including £190bn on health, £89bn on education, and £38bn on defence. The total will soon top £500bn.

Closing the productivity gap of 8.5pc that has opened up since 2019 would therefore allow the Government to provide the same services for about £40bn less. Which, by coincidence, is the same size as the “black hole” that Rachel Reeves is attempting to fill with tax increases instead.

Admittedly, it is difficult to measure the output of public services. The main benefits are particularly hard to quantify, such as the “health of the nation”, the “defence of the realm”, or “law and order”.

Instead, statisticians count specific outputs, such as the numbers of GP appointments and elective operations in the NHS, the number of hours of classroom teaching in state schools, and the number of cases dealt with by courts and tribunals.

There are two further statistical challenges. First, there were substantial shifts in supply and demand during the pandemic, partly due to the constraints of social distancing.

For example, the NHS cut back many regular activities in order to prioritise the prevention and treatment of Covid, while schools switched to online learning. Simply counting the regular activities as before could understate the value actually being added. Nonetheless, the impact of Covid should largely have washed out of the data by now.

The second challenge is the problem of adjusting for improvements (or declines) in the quality of services being delivered. This is typically easier in the private sector, where quality is reflected in the price someone is willing to pay. In the public sector, of course, most services are not priced, or the state can use its monopoly power to charge whatever it likes.

On top of this, there is a fundamental economic problem. It is generally harder to boost productivity in face-to-face services than in, say, manufacturing.

Digging deeper, the biggest drags on public service productivity are in health and education, which can partly be explained by the nature of these activities. They are inherently labour intensive and require lots of personal contact.

For example, it is hard to raise the productivity of an individual teacher, given the constraints on class size, or for a surgeon to perform many more operations at the same time. Think of the difficulties in raising the productivity of a hairdresser, or a musician.

Nonetheless, productivity has improved in many areas where services are mainly delivered by the private sector, so it can be done. For example, output per hour worked has increased since 1997 by more than 10pc (and sometimes by much more) in retail, distribution, professional services, and office support.

The reasons for the shortfalls in the public sector are not entirely clear. It may be tempting to blame the trend towards civil servants “working from home”. But this does not explain the problems in healthcare or education, or the fact that public sector productivity was lagging behind well before Covid struck.

Instead, I would focus on four factors. The first is the lack of competition. It is no surprise that sectors where market pressures are stronger also tend to be those where productivity gains are greater.

Second, public services are too reliant on the Treasury for much needed capital spending. We can debate the appropriate amount of public investment in new hospitals or school buildings, but any business model that depends on political choices will always be vulnerable to under-funding – or wasteful spending.

Third, the public sector is hamstrung by higher rates of trade union membership – and more militant unions. This is reflected in greater resistance to change, even though higher productivity is the best way to justify bigger pay rises. The doctors’ unions are particularly obstructive.

Fourth, and related to all the first three, the public sector has been slower to adopt new working practices and technologies. The scope for AI to transform the provision of public services is surely huge.

The biggest challenge of all, of course, is fixing the NHS. The UK now has the shortest life expectancy in Western Europe, and some of the worst health outcomes of any comparable economy, despite spending a similar amount of national income.

Any serious analysis has to start from the recognition that the NHS itself is part of the problem and that it needs fundamental reform to deliver a better service.

Encouragingly, NHS productivity did start to improve in the final years of the last government, partly thanks to pressure from then-chancellor Jeremy Hunt.

The new Government is at least making the right noises here too. Over to you now, Wes Streeting.


Julian Jessop is an independent economist. Jeremy Warner is away