‘Dissatisfaction drain’ as job satisfaction falls to record low and workers become less willing to go the extra mile, new British Social Attitudes research shows
Press release
Report
BSA | Work in progress
A decade of change in the world of work
This chapter examines how labour market changes over the past decade have influenced attitudes to work and employment.
Report PDF
Summary
This chapter examines how changes in the labour market over the past 10 years have affected attitudes to work, and what these attitudes might mean for three persistent woes in the British labour market: low productivity, low levels of job mobility and new business start-ups, and low levels of labour market participation among some groups.
Workers are less satisfied with their jobs than they were a decade ago and attitudes to work may have contributed to weak productivity growth
- 74% of workers say they are satisfied with their main job, down from 81% in 2015.
- The proportion willing to work harder than required to help their employer succeed has fallen from 69% to 55%.
- Workers are less likely than 10 years ago to say they have an interesting job (63%, down from 74%) or to report opportunities for advancement (26%, down from 35%).
- Those saying a high income is important to them personally in a job has risen from 67% to 78%. But only 23% feel their job provides a high income, a gap of 55 points. The equivalent gap for advancement opportunities is 54 points. Both gaps have widened since 2015.
- 61% say it is important to them that someone can decide their own working hours in a job, up from 46%. But just 31% say they can do that in their current job.
Weak dynamism does not appear to be explained by workers wanting less change
- The proportion of workers likely to look for a new job in the next 12 months has changed little in the past decade (32% in 2015 and 34% now). However, actual job moves fell over the period, suggesting demand for change is going unmet.
- Hybrid workers are less likely than other workers to say they plan to look for a new job. Their apparent reluctance to give it up may be dampening job moves, especially in higher-paying jobs.
While shifting attitudes do little to explain why inactivity has risen among some groups, the desire for full-time work has fallen
- The proportion who would prefer not to work at all (4%) has not changed significantly over the past decade.
- The proportion who would ideally work full time (47%) fell by 10 points between 2015 and 2025. The fall has been entirely driven by men (down from 77% to 62%). But it has not translated into a drop in actual full-time employment, which rose from 57% to 60%.
Introduction
The UK labour market has changed profoundly over the past decade. The National Living Wage was introduced in 2016 and has since risen to become one of the highest minimum wages in the world (Cominetti and McCurdy, 2025). Non-standard working arrangements, including zero-hours contracts and the gig economy, have become more prevalent (Slaughter, 2024). And the COVID-19 pandemic has left a legacy of increased remote working, at least for those in office-based jobs, and prompted debate over work-life balance (Office for National Statistics, 2025). Further change lies ahead: many workers (predominantly those in lower-paying jobs) are set to benefit from stronger rights through the Employment Rights Act (Department for Business and Trade, 2024), but at the same time, questions are growing over the impacts of artificial intelligence (AI) on the labour market, particularly on white-collar jobs (Department for Science, Innovation and Technology, 2026).
Yet for all this change beneath the surface, the labour market’s headline indicators look in some respects much as they did a decade ago: unemployment remains low, and job vacancies, having surged after the pandemic, have since cooled (Office for National Statistics, 2026a). And many of the labour market’s structural weaknesses have changed little over this period. Productivity growth slowed after the financial crisis, and output per hour remains low relative to countries like France and Germany (Pittaway and Slaughter, 2026).1 Low productivity is in part linked to weak economic dynamism – the rate at which new firms start up, grow and close, and at which workers move between them (Oliveira-Cunha et al., 2021). True, overall economic inactivity – the proportion of working-age people who are neither in work nor looking for work – is not especially high: indeed, after falling during the 2010s it was actually at a record low on the eve of the pandemic, and has since returned to around that level after a pandemic-era uptick (Office for National Statistics, 2026b; Resolution Foundation, 2026a). However, it remains high among some groups, notably young people, where the large number who are not in education, employment or training (NEET) has prompted a government review (Milburn, 2026). Meanwhile, previous research has found that people in lower-paying jobs have faced persistently high insecurity – which in part prompted the Starmer-led government to make improving employment rights a central manifesto commitment – and has identified the erosion of the higher job satisfaction they once reported relative to better-paid workers (Shah and Tomlinson, 2021).
In this chapter, we use data from the 2025 wave of the British Social Attitudes Survey (BSA) survey, in which the Work Orientations module was fielded as part of the International Social Survey Programme (ISSP). The questions have been fielded five times, in 1989, 1997, 2005, 2015 and 2025, allowing us to examine long-term trends in people’s attitudes towards, and experiences of, work. We examine how changes in the labour market over the past 10 years have affected attitudes to work, and what these attitudes might mean for the labour market’s three persistent woes – weak productivity growth (and low productivity relative to the UK’s peers), weak dynamism and high economic inactivity among some groups, particularly young people – in the years ahead. In particular, we attempt to answer the following questions:
- Has a wider disengagement with work contributed to increased inactivity?
- Has what people value in their jobs – and how hard they are willing to work – shifted in the wake of the COVID-19 pandemic, helping to account for weak productivity growth in the years that followed?
- Is there any truth in the post-pandemic phenomenon of ‘quiet quitting’ (Hamilton et al., 2025), in which workers disengage and do no more than the minimum their job requires?
- In a less secure environment, have people become less willing to change jobs, or more attached to their current jobs – helping to explain the lack of dynamism in the economy?
- And, as AI becomes ever more prevalent, how worried are people about their jobs being displaced by it – and are those fears sharpest among the workers who are likely to be most exposed?
The changes in the labour market over the past decade have affected different groups of workers unevenly, so shifts in attitudes are likely to be unevenly spread too. For example, we might expect appetite for full-time work to have changed most among those groups that have been most exposed to insecure work, such as young people – though the direction of any change is not clear-cut. Insecurity could weaken attachment to full-time work, by lowering expectations of what it has to offer, or strengthen it, as people seek the security of guaranteed full-time hours that they cannot currently find. And with men historically far more likely than women to want full-time work, it will be worth examining whether any shift has narrowed or widened this gap.
If commitment to employers, or to work more broadly, has weakened, we might expect this to be most pronounced among workers in low-paid jobs, who previous research suggests have lost the job satisfaction premium they once had over higher-paid workers. The spread of remote and hybrid working, concentrated in professional occupations, might also be expected to have widened the gap in flexibility between these jobs and more routine ones, where non-standard contracts have become more common – although remote working could equally have fostered an ‘always on’ culture in white-collar jobs.
Given that actual job-to-job moves have fallen, it will be telling whether this is matched by a declining appetite to move jobs, which would suggest that the fall has been voluntary, or, conversely, by a greater appetite to move jobs, suggesting pent-up demand for moves that people have been unable to make. We might expect any shift in the appetite to move jobs to be most marked among young people, who tend to move jobs more often early in their careers as they look for work that fits their skills and interests. At the same time, workers who have gained the option to work from home, most of them in higher-paid jobs, may be reluctant to give it up, which could dampen their appetite to move. And we would expect concern about AI to be concentrated among workers in white-collar and graduate occupations, whose jobs research suggests are most exposed.
The chapter begins by looking at preferences over how much people work, and whether shifting aspirations for work (and especially full-time work) can explain rises in inactivity and part-time work among some groups. We then turn to examine how workers feel about their jobs, whether low levels of satisfaction and commitment could help explain the UK’s weak productivity growth in the years after the pandemic, and how access to flexible and remote working differs across occupations. We go on to ask why so few workers are moving jobs, before finally turning to how people view the rise of AI.
Preferences for hours of work
We begin by exploring preferences over working hours. Labour supply has been a central policy concern over the past decade. As already noted, economic inactivity – the proportion of working-age people who are neither in work nor looking for work – fell steadily during the 2010s, largely driven by more women moving into work (Bell and Gardiner, 2019), but then rose during the COVID-19 pandemic. It has since fallen back to around its pre-pandemic level (Resolution Foundation, 2026a). By historical standards, then, overall inactivity is not especially high. But the make-up of the inactive group has been shifting, with the fewer people inactive for caring reasons being offset by rising health-related inactivity. Young people stand out: the proportion who are NEET – and particularly those not working due to a long-term health condition or disability – is continuing to increase (Diniz and Murphy, 2025).2
At the same time, part-time work has been on the rise among low-paid workers, and in particular low-paid men (Murphy, 2022), and new, indicative estimates from the ONS suggest that average working hours may have been falling over the past two decades, rather than flatlining as previously thought (Office for National Statistics, 2026d). There is an important question over whether these reductions in labour supply – both whether people are working at all, and if so, how many hours – reflect a desire to work less, constraints arising from the work available or people’s wider circumstances.
To measure preferences for hours of work, we ask respondents:
Suppose you could decide on your work situation at present. Which of the following would you prefer?
Full time (30+ hours)
10-29 hours
Under 10 hours
No work at all
The answers provided in 2025, and when the question was previously asked in 2015, are presented in Table 1, for the working-age population as a whole and for different age groups within in. At the population level, the proportion who would ideally work full time (defined here as 30 or more hours a week) fell by 10 percentage points between 2015 and 2025, to 47%. This was almost matched by a nine-point increase in the proportion who would prefer to work 10-29 hours, to 39%.3 But there has been no significant change in the proportion who do not want to work at all, which now stands at 4%. While these answers could reflect personal circumstances, such as health status, as well as underlying preferences, this would always have been the case, meaning that changes over time tell us something about how people’s
attitudes to work are shifting.
In the introduction, we suggested that any change in the appetite for full-time work might be most pronounced among young people, who have been particularly exposed to insecure work. As shown in Table 1, the falling preference for full-time work has indeed been largest among the youngest age group (aged 18 to 29): among this group, the proportion who would like a full-time job fell by 19 points, from 61% in 2015 to 42% now. One in ten say they would like to work fewer than 10 hours a week, a rise of seven points since 2015. Both shifts are largely explained by students: once full-time students are excluded, the fall in the desire for full-time work is more consistent across age groups (a nine-point drop among 18 to 29 year olds, compared with a 10-point drop among those aged 30 to 49), with the proportion of young people who would prefer to work fewer than 10 hours a week staying broadly the same.4 This suggests that the reduced desire for full-time work, rather than reflecting a shift in attitudes among young people as a whole, reflects today’s students being less likely than students in 2015 to want to combine paid work with studying, and more likely to prefer to focus on one or the other. This could be good news for educational and long-term labour market outcomes, but it may also make it harder for young people to gain an initial foothold in the labour market after graduation (Gardiner, 2020).5
The introduction also asked whether any shift in the appetite for full-time work has narrowed or widened the gap between men and women. We had two more specific expectations for how men’s and women’s preferences might have changed. First, changes in how childcare is shared between parents might have reduced men’s desire for full-time work. Second, the expansion of childcare funding over the past decade might have increased women’s desire for full-time work.
Figure 1 (which excludes full-time students) shows how the desire for full-time work breaks down by age and sex, with the purple dots showing the proportion of each group who said they would ideally work full time in 2015 and the blue dots depicting the proportion who did so in 2025. It shows that the fall in preference for full-time work has been driven largely by men. The proportion of working-age men wanting full-time work fell from 77% in 2015 to 62% now - but was broadly unchanged among women of the same ages (37% in 2015 and 38% in 2025).
The fall has been especially marked among men aged 30 to 49 – historically the group with the strongest attachment to the labour market – among whom the proportion wanting to work full time has dropped from 86% to 67% (a 19-point fall). Although this is an age group especially likely to have dependent children, the first of our expectations – that changes in how childcare is shared between parents might have reduced men’s desire for full-time work – finds little support. Fathers did become less keen on full-time work, but they did not drive the shift. Among fathers of under-16s, the proportion wanting to work full time did fall by a statistically significant 12 percentage points (from 89% to 77%), although a smaller seven percentage point fall among fathers with a working partner was not statistically significant.6 However, among men of the same age without children under 16, the fall was, if anything, larger: 26 percentage points, rising to 29 points among those who also do not have a partner (due to the small sample sizes involved, the difference between the falls for fathers and other men was not found to be statistically significant).
The data on which Figure 1 is based can be found in the appendix to this chapter.
By contrast, the proportion of women who would prefer full-time hours is broadly unchanged since 2015 across all age groups, so our second expectation – that expanded childcare funding would increase women’s desire for full-time work – is not borne out either. Although among women aged 30 to 49, the proportion wanting to work full time rose by five percentage points, and by 10 percentage points among mothers of children under 16 in this age group, in neither case was the change statistically significant.
Taken together, these changes mean that the gender gap in the desire for full-time work is much smaller than in 2015 – but it has narrowed because of the fall in desire among men rather than any rise among women. But there has been a significant fall in the proportion of women aged 30 to 49 who do not want to work at all, concentrated among those with children.7 This may have been helped by policy changes over the past decade, many of which – such as increasing free childcare and flexible working becoming a day-one right – might have been expected to ease constraints on parents, and particularly mothers, who want to work (we return to how this compares with the hours people actually work later in the chapter).8
igure 2 considers how the match between aspirations for and actual full-time working have changed over the past decade. For each group defined by age and sex, the purple bar shows the change between 2015 and 2025 in the proportion of those who say they would like to work full time, and the blue bar shows the change in the share who are actually working full time, according to government data.
The decline in the preference for full-time work does not appear to have translated into a fall in actual full-time employment. As noted above, the proportion of 18 to 64 year-olds who would ideally like to work full time fell by 10 percentage points between 2015 and 2025, from 57% to 47% (and excluding full-time students, it fell from 57% to 50%). But over the same period, the proportion of 18 to 64 year-olds who were actually in full-time work rose slightly, from 57% to 60% (or from 60% to 63% excluding full-time students).9
That said, there is a relationship between the change in the preference for full-time work and actual levels of working. Breaking this down by group, Figure 2 shows that the groups whose stated preference for full-time work has fallen furthest are also those where full-time working has either risen least, or fallen outright.
The data on which Figure 2 is based can be found in the appendix to this chapter.
By contrast, there has been little change in whether people want to work at all: in most groups, the proportion who would prefer not to work at all has not changed significantly, so shifting attitudes do little to explain recent rises in inactivity. But the fall in the desire for full-time work could be a leading indicator of future behaviour, pointing to a weakening of labour supply in the years and decades to come.
Job satisfaction and engagement with work
Having considered how much people want to work, we now turn to examine how they feel about the work they actually do. In particular, what changes, if any, have occurred in the level of job satisfaction and two related measures, that is, work commitment – how central paid work is to people’s lives, over and above the money it brings in – and organisational commitment – how attached workers feel to their own employer, and how willing they are to put in effort beyond the minimum. All three attitudes may be tied to productivity, another of the labour market’s woes: output per hour remains below that of the UK’s peers, and growth was weak for several years after the pandemic, although it appears to have picked up recently.
To measure the centrality of work to people’s lives, we asked respondents to say how far they agree or disagree with the following two statements:
I would enjoy having a paid job even if I did not need the money
A job is just a way of earning money – no more
Responses to these questions collected since the question was first asked in 1989, are presented in Table 2. Starting with work commitment, most people still value work for more than the pay cheque it provides, but there has been a small, statistically significant shift in how central paid work is to people’s lives. The proportion of people who say they would enjoy having a paid job even if they did not need the money has fallen from 66% in 2015 to 59% now, but remains higher than in 2005, when it was 51%. And the proportion who agree that “a job is just a way of earning money – no more” has risen from 29% to 34% – again, not a dramatic shift in the long-term context, but a rise nonetheless.
To measure workers’ organisational commitment – their engagement with their own job and employer, respondents are asked the following question, and are presented with a set of answer options ranging from, “completely satisfied” to “completely dissatisfied”.
All things considered, how satisfied are you with your [main] job?
Respondents are additionally asked to say how far they agree or disagree with the following statement:
I am willing to work harder than I have to in order to help the firm or organisation I work for succeed
As shown in Table 3, there is more movement in workers’ organisational commitment, than in their views regarding the centrality of work to their lives, considered previously. The first column shows that 74% of workers report being “completely”, “very” or “fairly” satisfied with their main job – still a clear majority, but down from 81% in 2015 and 82% in 2005, and the lowest figure to date. Relatedly, the second column shows that the proportion who say they would be willing to work harder than required to help their employer succeed fell from 69% in 2015 to 55% now.10 This fall reverses a rise that took place between 1997 and 2015 (from 58% to 69%), taking willingness to work harder back to around its 1997 level.11
In the introduction, we suggested that if commitment to employers, or to work more broadly, has weakened, this would be most pronounced among workers in low-paid jobs.
The two panels of Figure 3 take a similar approach to Figure 1: for each group, the lighter dot represents attitudes in 2015 and the darker dot attitudes in 2025. The left panel of Figure 3 shows the change in job satisfaction between these two years, while the right panel shows the change in willingness to work harder than is needed.
The fall in job satisfaction has not been uniform across different demographic groups. Satisfaction fell among professionals (from 84% to 77%) and, more sharply, among those in intermediate occupations (from 83% to 70%), but not among those in routine and semi-routine occupations (68% to 70%), although the differences between these groups’ trends are not large enough to be statistically significant.12 This runs counter to our expectation about low-paid workers, for whom routine and semi-routine occupations are the closest proxy available. Meanwhile, non-graduates’ satisfaction fell by nine percentage points, from 81% to 72% (driven by those in intermediate occupations), compared with a smaller, not statistically significant, 5-point fall among graduates – although, again, the difference between the two groups’ trends is not statistically significant.13
Job satisfaction also fell further among men (by nine percentage points, from 79% to 70%) than among women (from 83% to 78%), leaving a statistically significant gender gap.14 This is not a new pattern: women have historically tended to report higher job satisfaction than men, which Clark (1997) attributes to women having lower expectations of work, and a gap of a similar size was seen in 1997.
The fall in the willingness to work harder, depicted in the right panel, is more broad-based. It has taken place across occupational classes – with the largest fall, of 17 percentage points, among managerial and professional workers – among both men and women, and among workers both with and without a degree.15 Willingness to work harder has also fallen across age groups: by a statistically significant 18 percentage points among 30 to 49 year olds and 12 points among 50 to 64 year olds, and by a smaller, not statistically significant, seven points among 18 to 29 year olds – although the differences between age groups in the size of the fall are not themselves statistically significant. Especially taken together with the measures of
broader work engagement shown in Table , this is consistent with the concept of ‘quiet quitting’ discussed in the introduction – workers today put a little less intrinsic value on work than workers did a decade ago – which could have knock-on implications for productivity. If much of this disengagement has emerged since the pandemic, it could be part of the reason why productivity growth was weak in the years that followed.16
The data on which Figure 3 is based can be found in the appendix to this chapter.
There are likely to be many reasons for the decline in satisfaction and engagement, and we do not attempt to unpick all of them here. But we can shed light on part of the answer by comparing the job attributes that workers want with those they actually have. If the gap between the two has widened, whether because workers want more or because jobs offer less, that could help explain the fall in satisfaction.
Respondents are asked, in relation to the attributes listed in Table 4, to answer the following question, with answer options ranging from “very important” to “not important at all”.
For each of the following, please tick one box to show how important you personally think it is in a job.
They are subsequently asked how far they agree or disagree that these descriptions apply to their main job.
For each of the years that the two questions were included on BSA, Table 4 sets out the proportion of workers who say each attribute it is important and the proportion who say their job has it. It shows that the gap between what workers want and what they have has widened since 2015 on several of the attributes they value most, marking a break with the longer-term trend.
Job security remains overwhelmingly the most important attribute: in 2025, 94% of workers rated it as important, unchanged from 2015. But three attributes have become significantly more important: the ability to decide one’s own working hours (which rose from 46% to 61%)17, a high income (from 67% to 78%), and being able to work independently (68% to 75%). On the other hand, two attributes fell in importance, though by smaller amounts: having an interesting job (which fell from 93% to 88% of workers thinking it was important) and personal contact with others (65% to 61%). Overall, the average number of attributes workers rated as important rose from 6.4 to 6.7 out of the nine included in both the 2015 and 2025 surveys – a small but statistically significant increase in the number of demands workers place on their jobs.
It is perhaps unsurprising that the focus on income has risen, given the flatlining living standards of the past decade, particularly among those on lower incomes (Resolution Foundation, 2026b). But the rise in the importance of having a high-paying job has not been concentrated among people in the lowest-income households, who might have been expected to have felt the squeeze most acutely. Between 2015 and 2025, the importance attached to a high income rose across the income distribution, but the only statistically significant increases were in the second and top quartiles (up 14 and 12 percentage points respectively); the four percentage point rise in the bottom quartile was not statistically significant. The increased salience of pay therefore looks like a broad-based reordering of priorities rather than a direct reflection of the ongoing squeeze of living standards.
We might also expect the increased desire for control over working hours to have been driven by working parents. But here, too, the data tells a slightly different story. Among 30 to 49 year olds, mothers with children under 16 are indeed the most likely group to rate the ability to decide their hours and days of work as important – 75%, compared with 66% of fathers, 67% of women without dependent children, and 57% of childless men. But looking at the change between 2015 and 2025, the sharpest increases in the value placed on flexible hours came from women without dependent children (up 24 percentage points) and fathers (up 22 points), suggesting that the preference for flexibility that was previously more concentrated among mothers has become more broad-based.
Even as workers today want more from their jobs than workers did in 2015, they report receiving less in several important respects. The largest declines over the past decade are in the proportion reporting an interesting job (down from 74% to 63%), working independently (from 82% to 73%), and opportunities for advancement (down from 35% to 26%) – and all three are significantly linked to job satisfaction once other characteristics are taken into account.18 The proportion reporting a high income did not change significantly (25% in 2015 and 23% in 2025), so the widening gap on pay reflects more workers wanting a high income rather than fewer jobs offering one. Only one attribute became significantly more commonly reported: doing something useful to society (which rose from 67% to 73%). Taken together, this suggests that the fall in the proportion of jobs offering attributes that are linked to satisfaction provides at least a partial explanation for the fall in job satisfaction.
Figure 4 displays the relationship between the attributes that people want in a job and the ones they feel they have in a visual format, providing a summary of the current picture in relation to different attributes. Using the attribute of a high income as an example, Figure 4 plots how 78% rated this attribute as important and 23% reported having it. Because the dot in this instance sits below the dotted line, where the two proportions would be equal, it is clear that more workers want a high income than have one. Personal contact with others, on the other hand, sits above the line: 61% rate it as important and 85% say their job offers it, meaning that more workers feel their jobs offer it than rate it as important.
The data on which Figure 4 is based can be found in the appendix to this chapter.
Bringing the two measures in Figure 4 together, the largest unmet wants in 2025 (the gap between the proportion rating the attribute as important and the proportion reporting that they have it) were for high income (a gap of 55 percentage points) and advancement opportunities (54 points), followed by flexibility over working hours (30 points). And as Table demonstrates, the gaps on high income and advancement have both widened since 2015, reversing their earlier narrowing.
Flexibility and remote working across occupations
As Table 4 and Figure 4 show, workers increasingly value flexibility: the ability to decide one’s own working hours is one of the job attributes whose importance has risen most over the past decade, and one of the largest unmet wants. The rising value that workers place on flexible working is consistent with a pandemic-related shift, both towards valuing work-life balance and towards more remote working in professional occupations. To measure the extent to which workers experience remote working, we ask respondents:
How often do you typically work remotely, if at all? Please consider your standard working hours.
Sixteen per cent of workers report being fully remote and 34% hybrid. But this is highly uneven: 67% of those in managerial and professional occupations work remotely at least some of the time, compared with just 15% of those in routine and semi-routine occupations, more than three-quarters of whom (79%) never work remotely.19
In the introduction, we suggested that the uneven spread of remote and hybrid working across occupations might have widened the gap in flexibility between professional occupations and more routine jobs, where a rise in non-standard forms of work (such as zero-hours contracts or variable-hours contracts where the employer decides hours of work; see Slaughter, 2024) may have pushed in the other direction. But there is an alternative possibility: that remote working has instead fostered an ‘always on’ culture in white-collar jobs, blurring the boundary between work and home and making it harder, rather than easier, for these workers to step away from work during the day. Figure 5 allows us to test which of these accounts, if either, the data support, using two measures of flexibility: whether workers would find it difficult to take an hour or two off during working hours to deal with personal or family matters, and whether they can decide their own working times.
On the first of these measures, the results lean towards the ‘always on’ account. Contrary to the first expectation, and as shown in the left hand panel, it is those in managerial and professional occupations who report the largest rise (a statistically significant 11 percentage points, from 25% to 36%) in the proportion who find it difficult to take an hour or two off for personal reasons, whereas there has been no significant change among routine and semi-routine workers.20 Levels still differ – (semi-)routine workers remain the most likely to say they would find it difficult to take an hour or two off (47%, compared with 42% in intermediate and 36% in managerial and professional occupations) – but the rise among managerial and professional workers suggests that perceived expectations of being ‘always on’ in white-collar jobs may have been exacerbated by the pandemic.21
The second measure, control over working time (the lower panel of Figure 5), offers no sign of improvement either: managerial and professional workers remain the most likely to be able to decide their own working times (54%, compared with 49% in intermediate and 25% in (semi-)routine occupations), but this proportion has not risen among managerial and professional workers (it was 61% in 2005 and 56% in 2015), and none of the changes between 2015 and 2025 are statistically significant. Overall, then, there is no sign that flexibility has improved for managerial and professional workers as remote working has spread, and the one statistically significant change – the rise in the share who would find it hard to take time off – points towards the ‘always on’ explanation.
This is despite remote workers having considerably more control over their time than other workers. Today, 70% of fully remote workers and 66% of hybrid workers can decide their own start and finish times, compared with 28% of those who never work remotely. Just 15% of fully remote and 30% of hybrid workers would find it difficult to take an hour or two off, compared with 57% of those who never work remotely. That makes it more striking that flexibility for managerial and professional workers has not improved as remote working has spread.
The data on which Figure 5 is based can be found in the appendix to this chapter.
Taken together, the findings in this section point to a weakening in workers’ engagement with their jobs over the past decade. Paid work has become slightly less central to people’s lives, job satisfaction is at its lowest level to date, and willingness to work harder than required has fallen by 14 percentage points, reversing the rise between 1997 and 2015. The fall in willingness to work harder has been broad-based, while the fall in satisfaction has been concentrated among men and among those in professional and intermediate occupations, although the differences in trends between occupational groups are not statistically significant. At the same time, workers want more from their jobs but report getting less of several of the attributes most closely linked to satisfaction, and, after narrowing between 1989 and 2015, the gap between what they want and what they have has widened. Flexibility, one of the attributes whose importance has risen most, shows the same pattern: rather than improving flexibility for managerial and professional workers, the spread of remote working has coincided with a significant rise in the proportion of these workers who would find it hard to take time off during working hours, and no improvement in their control over when they work. This is consistent with ‘quiet quitting’, but suggests that it may reflect, at least in part, a decline in what jobs offer rather than simply a change in workers’ attitudes. If so, the prospects for improving productivity may depend as much on the quality of the jobs on offer as on workers’ attitudes to work.
Intentions to move jobs and actual job moves
The labour market’s third persistent weakness is weak dynamism. Since the early 2000s, job-to-job moves, particularly moves to a different sector, have declined, and this has contributed to productivity being lower than it would otherwise have been (Davies et al., 2023). But does this reflect a lack of appetite for change among workers, or barriers that prevent them from moving jobs?
The left panel of Figure 6 shows the proportion of workers who say they are likely to look for a new job in the next 12 months. This proportion is little changed since 2015 (32%, compared with 34% now), and it has risen slightly over the longer term (up from 26% in 1997, when the question was first asked). But this has not translated into job moves. As the right panel shows, following a short-lived post-pandemic uptick, the proportion of workers moving from one job to another each quarter is at a similar level to the early 2010s and lower than in the early 2000s. According to ONS data, on average 2.6% of workers moved job each quarter in 2005, 2.4% in 2015 and 2.2% in 2025 (the ONS series begins in 2002, so there is no comparable figure for 1997).
The decline in job-to-job moves, and associated weakness in labour market dynamism, does not therefore appear to be explained by workers wanting less change. Instead, demand for change appears to be increasingly going unmet: workers may be unable to find another job, the jobs available may not offer what they want from work, or they may face personal circumstances that prevent them from moving.
The data on which Figure 6 is based can be found in the appendix to this chapter.
Once again, there is variation between different groups. As in 2015, young people are more likely than average to want to move jobs (54%, compared with 34% of all workers) – unsurprisingly, given that it can take time for people to find a job that is the right fit for their skills and experiences at the start of their career. The group that stands out as having a notably greater appetite to move than 10 years ago is 18 to 29 year old men, among whom the proportion intending to move rose by 21 percentage points, from 42% in 2015 to 63% in 2025 – a rise significantly larger than the change among young women, whose intention to move fell over the same period (from 56% to 45%, although this fall is not statistically significant).22 The increased wish to change jobs among young men has not been matched by higher mobility, however: the proportion of workers in this group who had made a job-to-job move in the past three months fell from 3.2% to 2.4% over the same period.23
In addition, how people work matters for whether they want to move jobs and for their wider attachment to work. Even after taking account of differences in age, sex, occupation and education, hybrid workers are the least likely to say they will look for a new job in the coming year.24 (This is a measure of stated intentions, however: the survey tells us that hybrid workers say they are less likely to look for a new job, not whether they actually move jobs less often.) Remote working may well boost labour supply by allowing more people to work at all. But since access to hybrid working is unevenly distributed, workers’ reluctance to give it up may be dampening job-to-job moves, especially in the higher-paying jobs where remote working is more common.
Attitudes to AI and its impact on jobs
The 2025 survey is the first in the ISSP Work Orientation series to ask directly about how people feel about AI and its impact on jobs. To measure potential concerns about AI, we asked respondents the following two questions:
New types of digital technologies, such as robots, computer programmes, and artificial intelligence (AI) are changing the way we live and work.
(By computer programmes we mean any kind of software programmes and apps.)
Thinking about the impact that the use of digital technologies will have on jobs of people in Britain. Would you say that …
More jobs will be created than lost
A similar number of jobs will be created and lost
More jobs will be lost than created
Can’t choose
Overall, how worried, if at all, are you that in the next 5 years digital technologies will take over many of the jobs done by humans?
Very worried
Somewhat worried
Not very worried
Not worried at all
Can’t choose
As shown in Table 5, the picture is one of widespread worry about jobs in the abstract: three in five working-age adults (60%) are worried that, over the next five years, digital technologies (including, but not only, AI) will take over many of the jobs currently done by humans, and 57% think that more jobs will be lost than created by digital technologies.
But people are less concerned about their own jobs. We asked respondents who are currently in work:
How worried, if at all, are you that in the next 5 years your own job will be taken over by robots, computer programmes, or artificial intelligence (AI)?
Very worried
Somewhat worried
Not very worried
Not worried at all
Can’t choose
Only one in five workers (22%) are worried that their own job might be taken by robots, computer programs or AI over the same period. These worries are greater among younger workers: a quarter (25%) of workers aged 18 to 29 and of those aged 30 to 49 are worried for their jobs, compared with 15% of those aged 50 to 64. But there is little difference by occupational class or by educational qualifications, and to a commonly held understanding that AI is coming for graduate and white-collar jobs. Instead, the greatest levels of anxiety are among those not currently in work: among non-workers aged 18 to 64, 44% worry that new technology will limit their chances of getting a job, double the proportion of workers who worry about losing theirs.
Although this is the first time the ISSP module has covered the subject, earlier BSA surveys have addressed it via other questions. These are not directly comparable, but there does appear to be more concern now about jobs being lost to technology than there was in 2019, when 18% of workers said they were worried about their jobs being taken by robots in the next 10 (rather than five) years. But there is a tentative silver lining to this heightened anxiety about the impacts of technology. Among people currently in work, those who describe their job as interesting are less worried about losing it to AI than those who do not (19% versus 26%) – perhaps reflecting the fact that more routine tasks are more likely to be automated. This offers some grounds for optimism that the jobs lost to AI may be those that workers do
not enjoy in the first place (although, of course, there is a risk that technology removes menial jobs only to create worse ones in their place, as has been discussed elsewhere (O’Connor, 2026)).
Conclusion
This chapter set out to ask what a decade of upheaval has done to the way people in Britain feel about work – and what those feelings might mean for the labour market’s deeper problems. Some of what we find is as expected. Over the past decade, flexibility has risen up the list of what workers want, without a matching rise in the proportion of workers getting it. And, amid talk of ‘quiet quitting’, workers today are indeed a little less committed to their employers, and a little less intrinsically drawn to work in general, than workers were a decade ago.
But other findings are more surprising. Appetite for full-time work has converged between the sexes, but this convergence has come from a sharp downward shift among men, rather than from more women wanting more full-time work. And attitudes to AI are somewhat paradoxical: workers are pessimistic about its impact on jobs in the abstract, but fear its impact on their own jobs far less, while worries about AI do not seem to track how exposed different groups of workers are likely to be, according to other research.
The evidence in this chapter suggests that the problems with Britain’s labour market cannot be accounted for by the attitudes of workers. On labour supply, attitudes and behaviour appear largely disconnected. In most groups, the proportion who would prefer not to work at all has not changed significantly – so attitudes do little to explain recent rises in inactivity. Desired hours show the reverse mismatch: desire for full-time work has fallen, even as actual full-time employment has risen slightly. On dynamism, the attitudinal evidence presented in this chapter rules out one explanation: workers are no less keen to move jobs, even as realised job-to-job moves have fallen, which suggests that demand for change is increasingly going unmet. However, on productivity, there are several signs that attitudes may have contributed to the weak growth over the past 10 years – not least that willingness to go beyond the minimum has dropped heavily.
Stepping back, the falls in job satisfaction and engagement are also important in their own right. In part, they may reflect a re-evaluation of the role of work in people’s lives following the pandemic – but they also have important implications for policy. The Starmer government’s focus on ‘good jobs’ was well placed. But, given the broad-based falls in engagement with work, policy makers and employers should be thinking about how to improve the quality of work more generally.
Acknowledgements
The National Centre for Social Research is grateful to the Economic and Social Research Council (grant reference UKRI912) for their financial support which enabled us to ask the questions reported in this chapter. The views expressed are those of the authors alone.
References
Bell, T. & Gardiner, L. (2019), Feel poor, work more: Explaining the UK’s record employment, London: Resolution Foundation, available at: https://www.resolutionfoundation.org/publications/feel-poor-work-more/
Clark, A. E. (1997), ‘Job satisfaction and gender: Why are women so happy at work?’, Labour Economics, 4(4): 341-372, https://doi.org/10.1016/S0927-5371(97)00010-9
Cominetti, N. & McCurdy, C. (2025), Low Pay Britain 2025: Where next for the Government’s employment reforms?, London: Resolution Foundation, available at: https://doi.org/10.63492/bjv537
Davies, R., Hamdan, N. & Thwaites, G. (2023), Ready for change: How and why to make the UK economy more dynamic, London: Resolution
Foundation, available at: https://economy2030.resolutionfoundation.org/reports/ready-for-change/
Department for Business and Trade (2024), Employment Rights Bill: factsheets, available at: https://www.gov.uk/government/publications/employment-rights-bill-factsheets
Department for Science, Innovation and Technology (2026), Assessment of AI capabilities and the impact on the UK labour market, available at: https://www.gov.uk/government/publications/assessment-of-ai-capabilities-and-the-impact-on-the-uk-labour-market
Diniz, J. & Murphy, L. (2025), False starts: What the UK’s growing NEETs problem really looks like, and how to fix it, London: Resolution Foundation, available at: https://doi.org/10.63492/kvz546
Elliott, J. & Kelly, E. (2025), The hidden work in health care: unpaid overtime in the NHS, London: The Health Foundation, available at: https://www.health.org.uk/features-and-opinion/blogs/the-hidden-work-in-health-care-unpaid-overtime-in-the-nhs
Gardiner, L. (2020), Never ever: Exploring the increase in people who’ve never had a paid job, London: Resolution Foundation, available at: https://www.resolutionfoundation.org/publications/never-ever/
Hakim, C. (2000), Work-Lifestyle Choices in the 21st Century: Preference Theory, Oxford: Oxford University Press
Hamilton, O., Jolles, D. & Lordan, G. (2025), ‘Does the tendency for “quiet quitting” differ across generations? Evidence from the UK’, Frontiers in Behavioral Economics, 4: 1539771, https://doi.org/10.3389/frbhe.2025.1539771
Hunsaker, S. & Pittaway, S. (2026), The Macroeconomic Policy Outlook: Q3 2026, London: Resolution Foundation, available at: https://doi.org/10.63492/peqr754
McKay, S. & Simpson, I. (2016), ‘Attitudes and experiences of work in a changing labour market’, in Curtice, J., Phillips, M. & Clery, E. (eds.), British Social Attitudes: The 33rd Report, London: NatCen Social Research, available at: https://natcen.ac.uk/publications/british-social-attitudes-33
Milburn, A. (2026), Young people and work: interim report, London: Department for Work and Pensions, available at: https://www.gov.uk/government/publications/young-people-and-work-interim-report/young-people-and-work-interim-report
Murphy, L. (2022), Constrained choices: Understanding the prevalence of part-time work among low-paid workers in the UK, London: Resolution Foundation, available at: https://www.resolutionfoundation.org/publications/constrained-choices/
O’Connor, S. (2026), We Are Not Machines: The Fight for the Future of Work, London: Penguin
Office for National Statistics (2025), Who has access to hybrid work in Great Britain?, available at:
https://www.ons.gov.uk/employmentandlabourmarket/peopleinwork/employmentandemployeetypes/articles/whohasaccesstohybridworkingreatbritain/2025-06-11
Office for National Statistics (2026a), Labour market overview, UK: August 2026, available at: https://www.ons.gov.uk/employmentandlabourmarket/peopleinwork/employmentandemployeetypes/bulletins/uklabourmarket/august2026
Office for National Statistics (2026b), LFS: Economic inactivity rate: UK: All: Aged 16-64: %: SA (series LF2S), available at: https://www.ons.gov.uk/employmentandlabourmarket/peoplenotinwork/economicinactivity/timeseries/lf2s/lms
Office for National Statistics (2026c), LFS: Employment rate: Aged 18-24: In full-time education: UK: All: %: SA (series AIXA), available at:
https://www.ons.gov.uk/employmentandlabourmarket/peopleinwork/employmentandemployeetypes/timeseries/aixa/lms
Office for National Statistics (2026d), Component approach to measuring labour productivity, UK: 1997 to 2024, available at: https://www.ons.gov.uk/economy/economicoutputandproductivity/productivitymeasures/methodologies/componentapproachtomeasuringlabourproductivityuk1997to2024
Oliveira-Cunha, J., Kozler, J., Shah, P., Thwaites, G. & Valero, A. (2021), Business time: How ready are UK firms for the decisive decade?, London: Resolution Foundation, available at: https://economy2030.resolutionfoundation.org/reports/business-time/
Pittaway, S. & Slaughter, H. (2026), Revision of labour: Unpacking the ONS’s new estimates for productivity and hours worked, London: Resolution Foundation, available at: https://doi.org/10.63492/hfg9936
Resolution Foundation (2026a), Estimates of UK employment, available at: https://www.resolutionfoundation.org/our-work/estimates-of-uk-employment/
Resolution Foundation (2026b), Unsung Britain: A portrait of the country’s poorer half, available at: https://doi.org/10.63492/ndtl708
Shah, K. & Tomlinson, D. (2021), Work experiences: Changes in the subjective experience of work, London: Resolution Foundation, available at: https://economy2030.resolutionfoundation.org/reports/work-experiences/
Slaughter, H. (2024), Firm foundations: Understanding why employers use flexible contracts, London: Resolution Foundation, available at: https://doi.org/10.63492/hkkg822
Thwaites, G., Cominetti, N. & Slaughter, H. (2025), Labour Market Outlook Q3 2025, London: Resolution Foundation, available at: https://doi.org/10.63492/wvk136
Appendix
-
New, indicative estimates from the ONS, which measure hours worked using a ‘component approach’ (Office for National Statistics, 2026d), suggest that output per hour grew by 1.0% a year between 2007 and 2019, double the previous estimate of 0.5%, but still below the pre-financial crisis trend. Productivity growth weakened again in the years after the pandemic, at 0.7% a year between 2019 and 2024. These data revisions do not affect the UK’s level of output per hour as compared to other countries. For further discussion, see Pittaway and Slaughter (2026). Separate analysis (Hunsaker and Pittaway, 2026) suggests that productivity growth may have picked up again in 2025.
↑ -
Issues with Labour Force Survey (LFS) data in recent years mean that for several years from 2020 onwards, headline employment and inactivity rates did not reflect trends in administrative data (although the unemployment rate remained more reliable; see Thwaites et al., 2025). However, by 2025, following ONS efforts to improve the LFS’s sample size and address survey bias, the headline employment rate from the LFS had broadly converged on that implied by administrative data (Resolution Foundation, 2026a). Given these improvements, and since we cannot break down admin data-based estimates of employment and inactivity rates by
↑
characteristics like age and sex, this chapter uses LFS data for 2025 to compare with 2015. -
The increase in the share who wanted to work fewer than 10 hours a week, from 4 per cent to 6
↑
per cent, was statistically significant at the 10 per cent level but not at the 5 per cent level. -
The overall distribution of non-student 18-29-year-olds’ preferred hours did shift significantly between 2015 and 2025, but the fall in the proportion of the group wanting full-time work is too imprecisely estimated to be statistically significant on its own (the 95 per cent confidence interval runs from a fall of 22 percentage points to a rise of 3 percentage points).
↑ -
There appears, however, to have been little change in the proportion of full-time students actually working over the past 10 years. Looking at 18-24-year-olds (Office for National Statistics, 2026c), the employment rate among people in full-time education was 34.2% in 2015; it was higher than this (36.7%) in 2025, but lower in 2024 (33.2%) and has begun to fall again in early 2026. The small sample sizes among this group, combined with the recent challenges with the underlying data (see fn 2), mean we should not put too much weight on a short-term fluctuation.
↑ -
There were too few men with non-working partners in the sample to report results for this group
↑ -
The proportion who would prefer not to have a paid job at all fell from 4% in 2015 to 1% in 2025
↑
among all women aged 30-49. It fell by 4 points among mothers of children under 16 (a statistically significant change), and by 2 points among other women in this age group (not significant). -
This also speaks to a long-running debate over the extent to which men and women want different things from paid work versus the extent to which they are responding to different constraints. For further discussion, see Hakim (2000).
↑ -
Figures on actual rates of employment, participation and full-time work in this section come from RF analysis of ONS, Labour Force Survey. In the LFS, full-time and part-time work are based on respondents’ own classification of their main job rather than a set number of hours.
↑ -
Falling agreement with this question could reflect weaker commitment to employers. But it could also reflect a perception that work has become more intense, making the idea of working beyond the required level less realistic.
↑ -
The 2025 figure is not significantly different from that in 1997.
↑ -
Results by occupational class in this chapter should be interpreted with some caution. The underlying occupational class variable may have been more affected than other variables by the change from face-to-face interviewing: between 2015 and 2025, the proportion of respondents self-reporting that they worked in managerial and professional occupations rose from 37% to 52%, having only risen from 35% to 37% between 2005 and 2015.
↑ -
Satisfaction among non-graduates in intermediate occupations fell by 18 percentage points between 2015 and 2025, whereas the changes for non-graduates in professional and routine occupations were not statistically significant.
↑ -
The gap in job satisfaction between men and women was not statistically significant in 2015, but significant at the 5 per cent level in 2025. Across the full series, the gap was also statistically significant in 1997 (7 percentage points), but not in 1989, when women were more satisfied than men, or in 2005, when men were slightly more satisfied than women. So there is no evidence that the gender gap is narrowing steadily as women’s working lives have become more similar to men’s.
↑ -
In 2025, hybrid workers were more likely than other workers to be satisfied with their job (81%,
↑
compared with 74% of fully remote workers and 70% of those who never work remotely) and to be willing to work harder than they have to (62%, compared with 52% of both fully remote and never-remote workers). Both differences between hybrid and other workers are statistically
significant. But this does not necessarily imply that hybrid working raises satisfaction or
commitment in itself: it is a snapshot for 2025, and hybrid workers are disproportionately in
professional jobs. -
There is specific evidence that this is happening in the NHS, where workers are less willing to put in unpaid overtime since the COVID-19 pandemic, and this reduction in discretionary effort has been linked to lower productivity (Elliott and Kelly, 2025).
↑ -
The question on whether respondents’ own jobs allow them to decide their times or days of work was asked for the first time in 2025; the question on how important this is was also asked in 2015.
↑ -
Results of a logit regression of whether 18-64-year-old workers are satisfied in their job on having each of the job attributes shown in Table. We test two specifications: one pooling 2015 and 2025 data, and one using only 2025 data. Both regressions control for age, sex, whether in a professional occupation, and whether they have a degree. The pooled regression additionally includes a year dummy variable, and the coefficients on having an interesting job, opportunities
↑
for advancement and a high income are all statistically significant at the 5 per cent level. The
2025 regression additionally controls for earnings (and includes the remote working attribute,
which is not available in 2015), and the coefficients on having a high income and an interesting job are statistically significant, but that on opportunities for advancement is not. Full results are shown in the Appendix. -
The remaining six per cent responded “can’t choose” to the question.
↑ -
There was also a smaller rise of nine percentage points, significant at the 10 per cent level, among those in intermediate occupations.
↑ -
The evidence is tentative, however: the difference between this change and the change among (semi-)routine workers is just short of statistical significance.
↑ -
The 95 per cent confidence interval for the rise among 18-29-year-old men runs from 2 to 41 percentage points. It is wide because the samples of young men in work are small (around 60 in each year).
↑ -
Source: RF analysis of ONS, Labour Force Survey.
↑ -
Based on a logit regression of whether workers aged 18-64 in 2025 say they are likely to look for a job with another employer in the next 12 months, controlling for age, sex, professional occupation and degree. Both fully remote workers (p=0.024) and workers who never work remotely (p=0.010) are significantly more likely than hybrid workers to say they will look. Full results are shown in the Appendix.
↑
