First there was the crisis, then the long downturn, and now—let us hope—a labour market recovery. And as Britain and America enter the third phase of a painful economic story, a new question is coming to define political debate: how will the proceeds of the recovery be shared? The answer to this question will depend partly on the policy choices that are made in coming years. But it will also depend on underlying labour market dynamics. What kinds of jobs are mature economies creating in the early 21st century? And how have jobs markets changed since 2008? Last week a new report from the Resolution Foundation threw new light on these questions, suggesting that the downturn may have intensified an erosion of middle-skilled jobs that was already apparent before the crisis struck.
The new analysis looks at how the UK and US jobs markets were reshaped from 2008 to 2012 in a number of ways. First, it looks at how the employment shares of different occupations rose and fell in these years. From 2008 to 2012 the crisis appears to have hit middle-skilled occupations harder than others, accelerating a trend of occupational polarisation already established in academic literature. Figure 1 shows how occupations, ranked from low-skilled to high-skilled, saw their share of UK employment change over the four years. The pattern is clear: a relative decline of middle-skilled jobs while low- and high-skilled jobs expanded. Similar results are seen in the US.
Figure 1: The changing shape of the UK jobs market after the crisis
Change in Log (Employment) across the skill-distribution of UK occupations, 2008-2012
Notes: Skill percentiles are defined on the basis of mean wage in 2002. See Box 1 and Annex A in the full report for full methodology.
Source: Resolution Foundation and Centre for Economic Performance analysis, Labour Force Survey
Second, for the UK, the analysis reveals how the crisis has affected jobs differently depending on the types of tasks they involve. It finds that, in the UK at least, the downturn continued an erosion of routine occupations that was apparent before the crisis struck. Figure 2 shows levels of employment in routine and non-routine occupations in the UK from 2007 to 2012. While employment fell in routine occupations throughout the period, employment in non-routine roles actually rose throughout the downturn. Interestingly, real wages show the opposite pattern; they fell faster in non-routine jobs than in routine jobs. While the reasons for this are not entirely clear, it may be that employers sought to hold on to non-routine workers, doing so partly by squeezing their pay.
Figure 2: Impact of the crisis on routine and non-routine jobs
Employment, millions, 2007-2012
Notes: ‘Routine’ occupations are those in the top third of routine-intensity, ‘middle routine intensity’ relates to occupations in the middle third of routine-intensity, and ‘non-routine’ relates to occupations in the bottom third of routine-intensity. See Annex B of the report for full methodology for defining routine-intensity.
Source: Resolution Foundation and Centre for Economic Performance analysis, Labour Force Survey and O*Net dictionary of occupational information
Third, in both the US and UK, the downturn has played out very differently in different industrial sectors. In both economies, the middle-paying third of sectors appear to have seen employment fall from 2008 to 2012 while low- and high-paying sectors saw employment rise. In both cases, these broad patterns mask significant variation in the performance of specific sectors. In the UK in particular, there have been standout successes at the top and bottom of the labour market. At the top, Business Activities grew 15.5 per cent from 2008 to 2012 as net employment rose by 460,000. Meanwhile at the bottom, Hotels and Restaurants, the UK’s lowest paying sector, grew faster than any other sector from 2008 to 2012, seeing employment rise by 17.1 per cent.
Figure 3 shows equivalent analysis of patterns in employment by industry for the US, where similar patterns to the UK can be seen. Employment fell in all six of the middle-paying sectors of the US economy from 2008 to 2012, while it rose in all but one of the six lowest paying sectors. Meanwhile, in both the US and UK, there were also clear impacts from the collapse in demand between in 2008 and 2012, most notably sizeable falls in employment in Construction and Manufacturing.
Figure 3: The shifting industrial make-up of US employment after the crisis
Source: Resolution Foundation and Centre for Economic Performance analysis, Current Population Survey, National Bureau of Economic Research
How should we interpret these findings? Certainly it is easy to overstate what is happening. Middle-skilled jobs are not ‘disappearing’ from the UK and US labour markets, either before or after the crisis—they are simply declining as a share of employment. Moreover, even as middle-skilled occupations see a relative decline, demand for middle-skilled workers will remain high as new cohorts are required to replace those retiring each year. And it is also important not to assume that a polarising labour market necessarily means rising wage inequality. This link is far from straightforward. A polarising labour market could mean rising wage inequality, as more workers are pushed into the two ends of the labour market. But it could also mean falling inequality if, for example, rising demand for low-skilled workers pushed up their pay, and the supply of high-skilled workers rises fast enough to meet growing demand.
But the bigger debate initiated by these results is over their causes. On the one hand, there will be those who argue that these findings simply reflect where we are in the economic cycle. We might expect top jobs to have been protected more than others from the downturn, while low-skilled jobs may be the first to have gained from a gradual return to growth. Middle-skilled jobs may yet bounce back. And of course, big falls in industries like construction are likely to reflect a temporary collapse in demand. On the other hand, there is also a deeper, more structural explanation for some of these trends. The sectors of health and social care are growing, and the collapse of routine jobs in the UK is hard to explain without some reference to the impact of technology. There may yet be deeper trends at work here, even if they are dominated by demand in the short-term. As the UK and US economies enter a third, recovery phase, these questions will become key to understanding how the proceeds of that recovery will play out.