Labour Market Outlook Q4 2025

Employment has fallen over the past two years and is substantially lower than it was before the pandemic. Perhaps surprisingly given its central place in policy debates, participation is essentially unchanged compared to pre-pandemic: rising inactivity due to ill-health has been offset by long-term declines in family-related inactivity. Instead, rising unemployment, rather than falling participation, is the counterpart to falling employment. Unemployment has increased steadily over the past year, consistent with sluggish GDP growth as well as increases in labour costs in 2025, in particular following the increase in employer National Insurance Contributions (NICs). As is typical in economic downturns, young people have been hit hardest. With unemployment expected to stay elevated, Government should be cautious about any further increases in labour costs and should strengthen support for young jobseekers.

The UK lacks jobs, not just jobseekers

The working-age employment rate is lower than it was pre-pandemic, and has been falling for the past two years. There has been much focus on ‘economic inactivity’ (meaning people of working age who are neither working or looking for work) as the cause – the Government commissioned a review, and last year the then Secretary of State for Work and Pensions Liz Kendall described economic inactivity as “spiralling”. The UK does indeed have a participation problem, but perhaps surprisingly given how central it has been to recent policy debates, this is one of stalled long-term progress rather than deterioration compared to pre-pandemic levels. Currently, the demand side of the labour market is just as important as changes in supply: with unemployment rising, the UK is lacking jobs as well as jobseekers.

Progress has stalled, but labour market participation is at the same level it was pre-pandemic

The UK’s employment rate is substantially lower than it was pre-pandemic. That’s true whether we use the employment rate from the ONS’ Labour Force Survey (LFS), or the alternative estimate developed by the Resolution Foundation using job counts from HMRC’s tax data and ONS population estimates to make up for the LFS’ lack of reliability post-2020. In the LFS, the 16-64 employment rate reached a high of 76.5 per cent on the eve of the pandemic (in the three months to February 2020). It currently stands at 75.0 per cent in the LFS for the three months to September 2025 (the LFS has become more reliable recently thanks to efforts to boost response rates) and at 75.3 per cent in the Resolution Foundation’s alternative estimate for October 2025.

We have become accustomed to blaming participation for the UK’s post-pandemic employment problems. But in fact, economic inactivity is currently at more or less the same level it was on the eve of the pandemic. Again, that is true according to both the LFS inactivity rate and an estimate derived from the Resolution Foundation’s alternative employment estimate and the LFS’ unemployment rate (which we believe to be more reliable than its employment estimate). In the LFS, economic inactivity averaged 20.8 per cent in 2019, and in the latest data (covering the three months to September 2025) the rate was 21.0 per cent. Using our alternative estimates we put economic inactivity slightly lower at 20.5 per cent.[1]

The big picture, therefore, is that labour market participation – the inverse of economic inactivity – remains historically high. This is shown in Figure 1. The share of the working-age population (aged 16-64) who are either working or looking for work stood at 79.5 per cent in the three months to September, slightly down from a brief high point of 79.9 per cent in early 2023 but above the 2019 average (79.2 per cent) and – because participation has been trending upwards for most of the past 50 years – higher than at any previous point too.

If we extend potential labour supply to also include those who aren’t working or looking for work but who would like to work, there again isn’t a great deal of change over the long term. The share of inactive people who say they would like a job has fallen somewhat (from 27 per cent in the mid-1990s to 23 per cent today). But because the overall inactivity rate has come down in this period, the share of the working-age population who are working, looking for work, or who would like to work is very similar now (84 per cent) to 30 years ago (83 per cent).

Figure 1: Labour supply is historically high…

The reason economic inactivity is unchanged from pre-pandemic levels is that the worrying trend everyone is familiar with – of rising inactivity due to long-term ill-health – has been offset by falling inactivity for other reasons. In particular, there are fewer people looking after family rather than working, and fewer working-age retirees. Ill-health-related inactivity is up 1.2 percentage points as a share of the working-age population compared to January 2020 (reducing the size of the workforce by an equivalent of 530,000 people, based on today’s population size) but family-related inactivity is down 1.0 percentage points and retirements down 0.2 percentage points (adding an equivalent of 430,000 and 170,000 to the workforce, respectively).[2] Both of these latter trends are long-standing: family-related inactivity has been falling since the series began in the mid-1990s, and ‘working-age’ retirements (meaning, here, before age 65) have been falling since 2011, largely due to increases in the State Pension age for women. These trends are set out in Figure 2.

Figure 2: …but labour supply would be higher still were it not for the increase in sickness-related inactivity

It is certainly bad news that rising ill-health has undone progress on participation that would otherwise have come about through these other factors. Rising labour market participation pre-Covid-19 was an important engine of living standards growth, especially in the absence of productivity growth since the financial crisis. But we should nevertheless be accurate about how the UK’s labour market has changed post-pandemic: the economic inactivity ‘problem’ we face is one of stalled progress, not of overall deterioration.

Rising unemployment, not falling participation, is the counterpart to the recent fall in the employment rate

When it comes to unemployment, on the other hand, things are getting worse in absolute terms, not just relative to pre-pandemic trends. Unemployment has been rising for most of the past year, from 4.1 per cent in the three months to August 2024 to 5.0 per cent in the three months to September 2025. The employment rate has fallen by a roughly equivalent amount over this period – from 76.2 per cent in September 2024 to 75.3 per cent in October 2025. This means that the rise in unemployment can fully explain the fall in employment over the past year. That is also true taking a longer view – the difference between the employment rate today and the pre-pandemic level is fully accounted for by unemployment rather than inactivity, where positive and negative contributions to participation post-pandemic have offset each other. This is shown in Figure 3.

Figure 3: The employment rate falloff relative to pre-pandemic levels is explained by higher unemployment

Rising unemployment reflects sluggish growth, but also the effects of tax increases

The recent rise in unemployment partly reflects the state of the economy more broadly. As Figure 4 shows, changes in unemployment tend to follow changes in GDP growth fairly closely. GDP growth was slow from late 2023 to mid-2024, and unemployment started rising from that point – consistent with the fact that GDP movements are known to affect unemployment with a lag.

Figure 4: Rising unemployment partly reflects a slowdown in the wider economy

Given this lagged relationship between GDP and unemployment, stronger growth in early 2025 provides some cause for optimism on unemployment – although less positively, this stronger growth did not last, and has been non-existent in the most recent months of data. But it is also not as simple as: ‘where GDP goes today, unemployment follows’. In particular, the relationship between the two can change if there are labour market-specific shocks – and that is what happened with the £26 billion increase in employer National Insurance contributions (NICs) introduced in April this year. Employment growth in 2025 has underperformed what would be expected based on the strength of GDP growth.

As with most economic downturns, young people are bearing the brunt

The current episode of rising unemployment is not yet of equal scale to previous labour market downturns, but it appears to bear the familiar hallmark of young people coming off worst. Figure 5 shows the change in employment by age group in the current downturn as well as during Covid-19 and the financial crisis (the latter using LFS employment and the former two using payrolled jobs). During the first year of both the financial crisis and Covid-19, the fall in employment among 18-24-year-olds was around four times bigger than the fall among 25-49-year-olds. In the current episode, the overall fall is much smaller but – at least in the payrolled jobs data (and overlooking the ‘flash’ data point for October, which is liable to be revised) – it shows a jobs downturn highly concentrated among the young.[3] Payrolled jobs among 18-24-year-olds in September were down 1.0 per cent compared to July 2024, while payrolled jobs among 25-49-year-olds have actually scarcely changed at all (down just 0.2 per cent in the same period).[4]

Youth employment suffers during downturns because hiring slows down, making it harder for young people to get their foot on the job ladder. The vacancy rate fell by 37 per cent in the first year of the financial crisis, by 60 per cent in the first few months of Covid, and by 13 per cent over the past year (Aug-Oct 2024 to Aug-Oct 2025). Worse still, there is evidence that the hiring that does happen during downturns becomes more skewed towards older workers.

Figure 5: Employment downturns are always more heavily felt by the young

Young people doing worst is a general feature of downturns, but the specific circumstances this time may also be contributing. The increase in employer NICs and the minimum wage changes introduced in April 2025 both served to raise the labour costs of younger workers relative to older workers. The employer NICs increase affected the cost of employing young workers aged 21 and above (employers do not need to pay NICs on the salaries of workers aged under 21) because it had a bigger impact on low than high earners, and young people earn less than older workers.[5] The labour costs of workers aged below 21 weren’t affected by the employer NICs increase, but they were by the big youth minimum wage increases. In April, the minimum wage rose by 16 per cent for 18-20-year-olds (and by 18 per cent for 16-17-year-olds), compared to a 7 per cent increase in the adult rate.

Unemployment isn’t expected to fall for some time…

The good news is that, if the NICs increase is the main cause of the recent increase in unemployment, we’d expect the impact to fade over time as employers are able to pass on more of the tax increase to workers in the form of lower earnings. In November, some businesses told the Bank of England’s agents that “much of the adjustment” has taken place. But of course, that’s not true of the minimum wage increases – there are further rises coming in April, especially for workers aged 18-20, for whom the wage floor is rising by 8.5 per cent.

The UK’s two main official forecasters – the OBR and the Bank of England – certainly don’t take the view that rising unemployment is a brief response to a one-off tax change. They both think that unemployment will remain elevated for some time, and their unemployment forecasts are shown in the right panel of Figure 6. The Bank of England expects unemployment to remain at 5 per cent for the next two years.  The OBR is somewhat more optimistic – it has unemployment hovering just below 5 per cent throughout 2026, but falling from early 2027. In both cases, the main reason for expecting unemployment to remain high in the coming years is that they think the economy will be operating substantially below capacity (see the left panel of , which plots estimates of the ‘output gap’ – the difference between actual GDP and potential GDP).

Figure 6: The Bank of England has a worse unemployment rate forecast than the OBR

… but there is some hope that the Bank’s forecast is overly pessimistic

What happens next to the unemployment rate will depend on what this higher rate is doing to inflation. The best-case scenario is that stronger competition for jobs pushes down on wages. In that case, firms’ costs will fall, inflation will come down, the Bank of England will cut interest rates, that will stimulate spending, and employers will boost hiring to meet the extra demand.

The alternative, less appealing scenario is that wage growth and price inflation settle at stubbornly high rates in the face of this higher unemployment rate. There are several reasons why that might turn out to be the case. First, it could be that the mix of jobseekers looks very different from the mix of jobs available, and so firms still struggle to hire and high unemployment does not push down on wages so much. In that situation economists would say the ‘structural unemployment rate’ had risen. There is not currently much evidence for that, though. The UK’s current position on the ‘Beveridge curve’ (plotted in the left panel of Figure ) suggests the relationship between vacancies and unemployment is normal. So the problem isn’t that the job vacancies are of the wrong kind, but that there just aren’t enough of them – a problem of the overall level of labour demand, not mismatched demand.

Second, it could be that workers expect large wage increases to compensate for high past or expected price inflation. This would reflect an increase in the ‘non-accelerating inflation rate of unemployment’ (NAIRU), and in this scenario, as with unemployment increases stemming from efficiency problems, the Bank would have less space to cut rates and bring unemployment back down again. The Bank of England discussed this possibility in its latest Monetary Policy Report but did not reach a firm conclusion. There does not yet seem to be strong evidence of this having happened. During the cost of living crisis, the level of wage growth (a key driver of inflation) was indeed high relative to unemployment based on the historic relationship between the two. But in recent months, wage growth has slowed, so currently the pace of wage growth is roughly what we would expect given the level of unemployment. This is expressed in the ‘Phillips curve’ chart shown in the right panel of Figure . This bodes well for the ability of the Bank of England to boost labour demand with lower interest rates.

Figure 7: Labour market efficiency, and the relationship between unemployment and wage growth, both currently look normal by the standards of the past 25 years  

Policy needs to adjust to the reality of rising unemployment

Even if the OBR’s somewhat more benign outlook is correct – with inflation returning to target alongside falling unemployment – elevated unemployment looks set to be a defining feature of the UK economy over the next couple of years. And as in previous episodes this will be felt most by those young people looking to take their first steps into the world of work. The recent preoccupation with inactivity is understandable – the UK has fallen off its pre-Covid-19 trend of rising participation, and this increase has been accompanied by rising health-related benefit spending. This is a real and worrying problem – higher participation would boost household incomes as well as the Treasury’s finances. But the UK also now faces a new problem: too many jobseekers chasing too few jobs.

A problem on the demand side of the labour market has different policy implications than if falling employment was purely a participation problem. Unfortunately, we cannot go back in time and choose a more jobs-friendly tax increase than the employer NICs hike. But the Government can be careful about any further increases in labour costs, especially for young people. On minimum wages, the Government should give the Low Pay Commission a clear steer that convergence with the adult rate can happen slowly (or, ideally, change tack altogether and retain the youth rates). Finally, the Government should expand its Youth Guarantee – the offer of a work guarantee should be extended to 22-24-year-olds, and not limited to young people who are claiming out-of-work benefits.

Lifting the lid | Three trends in focus

The Government is pushing up youth minimum wage rates

Since the National Minimum Wage was introduced in 1999, there have been lower rates for younger workers, aimed at protecting employment among a group at greater risk of worklessness. Over time, the rate for 18-20-year-olds has drifted downwards relative to the main adult rate, as shown in Figure 8. But the current Government has changed course: its manifesto committed to aligning the 18-20-year-old rate with the adult rate. As a result, although the Government has taken a relatively cautious approach to raising the National Living Wage since it came into office – uplifting it in line with median pay in both 2025 and 2026 – the uplifts in the youth rates have been far more ambitious, and in 2026 the 18-20-year-old rate will be at its highest level relative to the adult rate since 2002. The plan appears to be to completely converge the two rates by the end of the Parliament.[6] But given a high and rising proportion of young people are not in employment, education or training, the Government should consider taking a more cautious approach to protect their employment prospects.

Figure 8: The 18-20 minimum wage rate is set to converge on the main adult rate by 2029

Fewer quits and more redundancies mean workers’ bargaining power has fallen

One consequence of a weaker labour market is that workers have reduced bargaining power: when fewer jobs are available and more people are looking for work, it becomes harder for workers to negotiate higher pay or better conditions, either with their current employer or a prospective new one. Figure 9 shows the ‘labour leverage ratio’, an indicator developed by Aaron Sojourner at the Upjohn Institute that captures workers’ bargaining power, calculated as the number of workers voluntarily leaving their jobs for another one (suggesting they can find better opportunities elsewhere), divided by the number of redundancies (reflecting weaker employer demand). This measure spiked as the economy reopened after the Covid-19 pandemic and employers went on a hiring spree, but has been falling since, and in Q3 2025 was at its lowest level (outside the Covid-19 pandemic) since 2012.

Figure 9: Workers’ bargaining power has fallen considerably over the past two years

Real wage growth has reverted to the post-2008 norm of stagnation

The labour market legacy of the 2008 financial crisis was a decade and a half of wage stagnation. As Figure 10 shows, inflation outpaced nominal wage growth between 2008 and 2014. And even when real wage growth resumed, it was sluggish – with real growth stalling in the aftermath of the Brexit vote and disrupted by the Covid-19 pandemic – so that, adjusted for CPI inflation, real wages did not get back to their 2008 level until 2024 (strictly speaking they did so in 2020 but this was partly a big composition effect rather than due to underlying wage growth). Workers have seen some welcome real wage growth in recent years, with real wages up 4.7 per cent between January 2023 and January 2025. But over the last year, wage growth has once again ground to a near halt – and the latest OBR forecasts suggest that this stagnation is set to continue, with wages set to grow by a paltry 2.0 per cent in total between now and 2031.

Figure 10: Real wage growth has ground to a halt over the past year and the outlook remains poor

[1] From here on, all estimates of the employment or inactivity rate for 2020 onwards are based on the Resolution Foundation’s alternative employment estimates. These produce an employment rate estimate by dividing the number of employee and self-employed jobs by the size of the population. Population estimates are derived from the ONS’ newly published mid-year 2025 estimates, which are projected forwards using the ONS’ 2022-based migrant category variant population projections, in turn adjusted for the latest outturn net migration data (which shows lower net migration than the ONS expected in that projection variant). Our Q3 2025 Labour Market Outlook argued that the LFS’ unemployment rate estimate is reliable, so we use this alongside our alternative employment rate estimates to derive estimates of the economic inactivity rate.

[2] The changes in participation in numerical terms given here apply the change in the percentage of the working-age population who are inactive for the given reason to the current working-age population size.

[3] The October payrolled jobs data showed a scarcely credible 1.1 per cent month-on-month seasonally-adjusted increase in the number of payrolled jobs held by 18-24-year-olds. This is a bigger absolute change than any other in the dataset outside of the pandemic period.

[4] It is worth noting that the LFS data on unemployment by age does not give the same picture of a downturn concentrated on young people as is the case in the payrolled jobs data. In the latest LFS data, unemployment among 18-24-year-olds is down slightly on the year, meaning the overall unemployment rise comes from increases among older groups (as well as from 16-17-year-olds). We are placing more weight on the payrolled jobs data due to ongoing concerns about LFS reliability.

[5] For example, according to the ONS’ Labour Force Survey, in Q1 2025, median weekly pay among 21-24 years was one-third (33 per cent) lower (£462) than for workers age 41-44 (£692).

[6] The Government did not specify a date for convergence in their remit to the Low Pay Commission, but the options the Low Pay Commission drew up in their 2025 consultation on how to achieve convergence all had convergence happening in 2029.