Key findings

We have now published a video summarising the key findings and recommendations. This is also followed by expert commentary by Gail Irvine (Carnegie Trust UK), Magdalena Soffia (What Works Centre for Wellbeing), and Jonny Gifford (Chartered Institute of Personnel and Development). This post covers the highest level findings, which are:

  1. The quality of work is highly structured by occupation

The sharpest divide is between managerial and professional occupations on the one hand (such as HR managers, academics, and physiotherapists) and routine and manual occupations on the other (such as sales assistants, cleaners, and waiters/waitresses). Managerial professional occupations not only pay more, they also tend to do well in the other areas of job quality too – particularly those that we (and decades of social science research) found matter a great deal for job satisfaction – intrinsic job quality – such as skill-use, task variety, and control over job tasks.

However, the correspondence between how much an occupation pays and its overall quality is not perfect. These discrepancies are outlined in more detail in Research Brief 2. These really are the exception to the general class divide we find, however. Moreover, the class divide appears to be strengthening, while we also find that mobility across occupations appears to be falling since the mid-1980s.

  1. The occupational quality structure has generally been upgrading

In terms of how the occupational quality structure has been evolving since the mid-1980s, we find that there has been a general growth in higher quality ones with a steady decline in the lower quality ones. This is in contrast to the popular narrative that the labour market is polarizing into ‘good’ and ‘bad’ jobs. This discrepancy is explained by the different way we conceptualise the occupational quality structure – in terms of overall quality – incorporating more intrinsic aspects of work as well as how much an occupation pays.

Many of the declining middle-paying occupations tend to fall at the lower end of the occupational quality spectrum. These include some related to manufacturing and production such as assembly line work but also some clerical work such as database clerks and stock control clerks. These declining lower quality occupations are also more likely to be automated according to the best guesses on automation probability data produced by the Office for National Statistics.

  1. The quality of work is deteriorating for all workers in some critical respects

While the occupational quality structure has been generally upgrading, we find work is getting more routine, more controlled, and more intense across the occupational quality spectrum. This implies that technology’s main effect on the quality of work is not through its shaping of the structure of jobs, but more directly through how work and workplaces are organised – for all kinds of jobs.

  1. Upward occupational mobility is associated with an upward trajectory in job satisfaction

Following on from key finding 1, we find that upward occupational mobility is associated with a boost to job satisfaction. While workers eventually return to their baseline after the switch due to adaptation (as with most pleasant life events), this is still an important means for individuals to improve the quality of their working lives. By contrast, those who experience downward mobility display a steady deterioration in job satisfaction that they cannot recover from even several years after the transition.

We also find that personality moderates the magnitudes of these patterns. Downward transition is particularly problematic for individuals high in neuroticism. These findings demonstrate that individual factors can interact with situational factors to influence subjective well-being. Mobility across the occupational structure has the potential to deteriorate, as well as improve, job satisfaction trajectories.

  1. Experience of unemployment moderates the effect of downward mobility

Building upon the worrying effects of downward mobility on job satisfaction, and potentially lasting scars, we further explored how the life event of unemployment (which often accompanies downwards mobility) moderates job satisfaction trajectories for the downwardly mobile. Here, we find when downward mobility is preceded by a spell of unemployment, it has no significant effect on post-turnover job satisfaction. By contrast, those who move down the occupational ladder without unemployment experience a severe deterioration in job satisfaction.

Similarly, those who reside in high unemployment areas experience little change in  job satisfaction when they make a downward career transition, whereas the same event does hurt those who live in low unemployment areas. These findings imply that individuals evaluate downward career mobility in the context of their career trajectory and social environment. Self-comparison and social comparison significantly affect how people judge their lives.

  1. Workplace policies can moderate the effects of occupational quality

Finally, given this project’s emphasis on between occupational differences in the quality of work, we also explored how workplace policies can moderate the experience of work within occupations. Here, we find that policies related to teamworking and participation (high involvement management) tend to boost wellbeing directly – independent of occupation. This suggests if employers want to foster higher levels of wellbeing, the effects of such practices are likely to be positive no matter the occupational mix.

However, the magnitude of the effects depends on the occupational mix. For example, teamwork works best in occupations that require constant effort and attention, whereas participation is most effective in boosting employee well-being among occupations that do not make much use of skills. These findings reinforce that high-quality management is critical to the wellbeing, but also that its effects are differentiated by occupational quality.

A research brief summarising the findings and recommendations can be accessed here.

Back to blog