Introduction
- Concerns over the performance and labour market opportunities of young people in the UK have been prominent in the public debate for many decades. Since the arrival of mass unemployment in the 1980s it has been known that young people typically are more at risk of a) experiencing unemployment, b) being laid off in recessions and c) finding it harder to find work in a downturn when hiring pauses. In what follows I assess key indicators of youth labour market performance both through the current downturn and from 1975 to the present.
I Measuring Youth Labour Market Performance
- Youth labour market statistics are complicated by the relatively recent rapid growth in the numbers of young people still in education, (since levelled off). This matters because the typical measure of labour market performance – the unemployment rate — is a count of the unemployed relative to the size of the labour force (the employed plus the unemployed).
Rising staying-on rates in education will tend to reduce numbers in the labour force and so inflate the unemployment rate for a given number of unemployed.[1] This makes comparisons of the youth unemployment rate over time harder – if staying-on rates in education change over time. It also makes comparisons of the youth unemployment rate across countries harder – if different countries have different staying-on rates in education.
- If, instead, we measure the number of unemployed youth relative to something that is less influenced by government policy or labour force participation trends, this can give a better metric for comparing performance over time. Measuring unemployment relative to the entire youth population, we get the u-pop rate.
This should be much less susceptible to changes in labour force participation and is another useful indicator of youth labour market performance.
- Some people worry that a count of the unemployed may not capture all aspects of joblessness. The official UK count of unemployment is estimated from respondents’ answers to a set of questions in the Labour Force Survey, (LFS). Some of those surveyed may not satisfy all the criteria to be classified as unemployed, but may express a desire to seek work or identify a constraint that prevents them from starting work. One way of capturing this is the NEET rate, which measures the share of the youth population not in education, employment or training.[2]
In short, it is probably best to look at a range of youth labour market indicators rather than focus on just the unemployment rate.
II The Youth Labour Market In The Coronavirus Pandemic
8. To get a sense of the performance of the youth labour market during the crisis, Figure 1 plots both the youth unemployment (first panel) and NEET rates (second panel) in each week of the pandemic, (through to February 2021), relative to the average of the last five years. The dark grey shading represents a significant departure from recent norms.[3] The unemployment and NEET rates are, coincidentally, very similar at around 15%. However, there is not much in either the unemployment or the NEET rate that have not been seen in the last five years before the COVID downturn. Young people – while again disproportionately affected by a downturn - have not been subject to mass joblessness seen in earlier recessions. The furlough scheme almost certainly underlies much of this observation.
Figure 1. Youth Unemployment and NEET Rates 2020 relative to 2015-2019 average
Unemployment Rate: Age 16-24 |
|
NEET rate: Age 16-24 |
|
Source LFS, author calculations
- Figure 2 shows that the youth labour market has been much worse in each of the previous three recessions, whether measured by the unemployment rate, the unemployment-to-population rate or the NEET rate.
- Using the unemployment rate tends to exaggerate the scale of the 2008-2011 downturn in the run up to this point.[4] But all the metrics of labour market performance confirm that the current youth labour market downturn is not nearly as bad as in the past.
Figure 2. The UK Youth Labour Market Through the Ages 
Source LFS, author calculations
III. Long-Term Unemployment
11. While all unemployment is a worry, we tend to worry even more about the build-up of long-term unemployment. We have known since at least the recession of the 1980s that long-term unemployment is harder to escape from and tends to leave lasting scars on both the individuals and the communities in which they live. For governments, this means more resources are needed to try to alleviate the problem.
- Historically young people were largely spared long-term unemployment. Younger workers were more likely to lose their jobs in a downturn, but more likely than older workers to find new work more quickly and so less likely to become long-term unemployed.
- Figure 3 plots the percentage chances of losing a job and the percentage chance of finding a job for 16–24-year-olds over time, relative to the same chances for individuals aged 25-64. The figure, (left panel), confirms that young people are much more likely to lose their jobs than older workers at any time – but much more so in recessions. However, the right panel confirms that young people are also much more likely than older workers to find a job if not in work. The job finding rate for younger workers rebounds much faster and further during economic recovery. It is true that the relative chances of finding a job have narrowed since 2010. Job finding rates among older workers have risen during this time, but there was no large rebound in job finding rates among younger workers during the recovery after 2011. Some of this reflects a) continued rises in participation of older women and b) the automatic effect that increased staying on rates in education will have on chances of staying in non-employment, rather than a fall in the chances of young people finding work. It may, however, be something to monitor going forward.
Figure 3 Job Loss and Job Finding Rates by Age
|
|
Source LFS, author calculations
- Figure 4 plots the share of 16–24-year-olds in long-term unemployment between 1975 and 2021 alongside the headcount of all long-term unemployed.[5] There have been three economic cycles over this period prior to the COVID downturn. In general, the share of younger workers in long-term unemployment tends to rise at the onset of a recession. Firms stop hiring in times of uncertainty and this affects younger workers disproportionately. As the recession persists, older unemployed workers find it harder to find new work and the youth share of long-term unemployment falls. The 2008-2011 downturn was a little different. The youth share of long-term unemployment grew faster earlier in the cycle and did not fall back as much in the subsequent recovery. During the COVID downturn the share of youth has reached 26% of all long-term unemployed workers – but, as Figure 4 also shows, the number of long-term unemployed workers is currently as low as it has been for nearly 40 years.
Figure 4. Share of Younger Workers in Long-Term Unemployment
Source LFS, author calculations
IV. Jobs for Life
15. Concern is often raised that jobs for life have disappeared and that young people will have to prepare for a world of work involving many more changes of job over a lifetime than in the past. In such an environment, acquired skills need to be more transferable across jobs. However, there is reason to think that concern over the end of a job for life may be premature.
- Table 1 estimates the number of jobs the average person can expect to hold by a given age.[6] Figure 5 plots the change in this prediction over time. The trend is clearly downward. In the 1990s the total number of jobs held was around 11 – with more than half of those held by people before the age of 25. Over time the expected number of lifetime jobs has gradually fallen back – at each age. This means that young people are changing jobs less often than in the past. With hiring seriously compromised under the Coronavirus lockdown, the numbers dip again in 2020. This is perhaps the key issue going forward. The end of lockdown should open up more opportunities for young people – as at the end of previous recessions. But the labour market looks far away from a future of rapidly changing jobs.
Table 1. Expected Number of Jobs Over a Working Life
| 1993 | 2003 | 2013 | 2019 | 2020 |
By Age |
|
|
|
|
|
16-17 | 2.5 | 1.3 | 1.6 | 1.5 | 1.1 |
18-20 | 4.8 | 2.8 | 3.2 | 2.9 | 2.1 |
21-24 | 6.3 | 4.3 | 4.6 | 4.1 | 3.2 |
: |
|
|
|
|
|
60-64 | 11.1 | 9.2 | 8.7 | 8.1 | 6.4 |
Source LFS, author calculations
Figure 5. Expected Number of Jobs Held by a Given Age
Source LFS, author calculations
V. What Jobs do Younger Workers Do?
17. The process of finding a good job match takes time and often involves moves across occupations and industries. Figure 6 outlines the sectoral composition of young people in work - counting for each industry the numbers employed in one of nine occupations.[7] The youth labour market is dominated by just two categories: sales workers in retail and “elementary” occupations in accommodation and food services. One quarter of all 16–24-year-olds work in these two sectors.
- In contrast the sectoral composition of older workers is much more disperse (Figure 7). This means that younger workers and older workers often do not do the same jobs. Many young people will change industries, and indeed occupations, as they get older before finding a settled job match.
Figure 6. Occupation and Industry Composition of 16–24-year-olds in work (2019)
Figure 7. Occupation and Industry Composition of 25–64-year-olds in work (2019)
VI. Averages Hide Different Experiences
19. The analysis above displays the average labour market prospects for youths. Averages often hide many differing experiences of individuals. In reality, an individual’s labour market chances depend on a variety of factors, including age, education, region, gender and ethnicity, but also the interaction of these features.
- Thus, it is equally true to say, for example that the graduate unemployment rate among 16–24-year-olds in 2021 is half that of non-graduate 16–24-year-olds, but also that the graduate unemployment rate for 22-year-olds is twice the unemployment rate for 24-year-old graduates, (source LFS). It is also true to say that these gaps are not constant but evolve over time.
- Small sample sizes in the LFS often preclude estimation of unemployment rates for multiple interactions (for example., graduate women aged 23 living in Yorkshire). Comparisons of unemployment rates across groups are also affected by the interaction of characteristics. For example, many graduates live and work in London where graduate jobs are relatively plentiful. Simply comparing graduate unemployment with non-graduate unemployment rates without taking account of these regional differences would tend to accentuate these differences. One way to take account of these features is by “regression analysis”. Regressions can pick out the effect of a particular attribute while controlling for (“netting out”) the influence of a set of other characteristics that also influence the chances of being jobless.
- The graphs below are taken from regressions and show how the relative chances of being NEET (not in employment, education or training) vary with certain attributes over time – net of other characteristics among 16–24-year-olds. The figures focus on three factors that historically accounted for the largest variation in NEET rates: gender, ethnicity and age.
- The Figures all show evidence of notable convergence in a young adult’s chances of being jobless. Figure 8 shows that young women, net of other characteristics, are now less likely to be observed as NEET than men. Twenty five years ago they were 5 percentage points more likely to be observed as NEET.[8]
Figure 8. Relative Chance of Being NEET aged 16-24: Women
Source: LFS. Author calculations
- Historically youth NEET rates were focussed on individuals in their twenties. Figure 9 shows that this is still the case, but that the gap between young people in their teens and young people in their twenties, net of other characteristics has narrowed.
Figure 9. Relative Chance of Being NEET aged 16-24: Age
Source: LFS. Author calculations
- Historically, young people from many ethnic minority groups were much more likely to experience joblessness than young whites. Figure 10 shows that, in the 1990s, variation in joblessness by ethnicity, controlling for other characteristics, was the largest single factor behind joblessness. Since then, these gaps have narrowed noticeably. Gaps still remain, notably among Black Caribbean but the chances of being observed NEET are now much more even across different ethnicities.
Figure 10. Relative Chance of Being NEET aged 16-24: Ethnicity
Source: LFS. Author calculations
- The factor that shows the largest variation in the chances of being jobless is education. Figure 11 tracks the chances of being jobless for 16- to 24-year-olds for a given level of educational attainment relative to being a graduate, holding other things constant. The Figure shows that the difference in the chances of joblessness for any young person with Level 3 or Level 4 qualifications relative to graduates is very similar and has been so for a long time. There is evidence that the benefit of level 3 vocational qualifications has narrowed the employment gap between graduates noticeably over the past 10 years.
- However, for young people with Level 2 qualifications or below, the employment gap – while falling somewhat over the last 10 years – is still very high. An individual with Level 2 qualifications is some 10 percentage points more likely to be observed NEET than a graduate, other things equal. An individual with no qualifications is some 30 % points more likely to be NEET.
Figure 11. Relative Chance of Being NEET aged 16-24: Education
VII. Conclusions
28. Young people always bear the brunt of downturns. They are often the first to be fired and also suffer if hiring stalls. The COVID-19 downturn was no exception, but the manifestation of the downturn has been such that fewer young people have been spared joblessness. The furlough scheme is probably helping to keep the majority of (young) people in work. Mass long-term unemployment among younger workers and the economic scarring it often induces currently looks unlikely to emerge. Care needs to be paid when establishing appropriate metrics of youth unemployment, but it does seem that youth unemployment this time around is unlikely to be as extensive as in previous downturns.
- While averages often disguise very different experiences of younger workers, there has been a noticeable convergence in labour market prospects over the past 30 years. Educational attainment seems to be the most important characteristic in mitigating the risk of joblessness among young people.
- Younger workers spend the initial part of their careers in different sectors and occupations to older workers before finding a good job match. While some job turnover is an inevitable part of that process, it also seems that the idea that jobs are no longer for life and the associated pressures to find requisite skills are somewhat premature.
10th May 2021
18