The impact of youth unemployment on adult employment trajectories

Dr Stefan Speckesser and Vahe Nafilyan, from the Institute for Employment Studies, look at patterns of young people's movement into and within the labour market over the last 40 years

Long-term trends and effects of the economic cycle

In a recent project for the Department for Business Innovation and Skills, we examined how the transitions of young people to and within the labour market have changed over the last four decades.[1] These changes were analysed for all people in the UK born between 1959 and 1997 with descriptions of some broad trends in the labour market and education activity using all available years of the Labour Force Survey (LFS) from 1975 until 2013.

Our video below shows how much education participation and patterns of entry to the labour markets have changed in the UK since the 1970s. It shows both the continuing trend of education expansion as well as effects of the economic cycle, resulting in varied initial labour market transitions and potential impacts on adult employment trajectories for different cohorts.







Overall, increasing education participation suggests that young people start employment with higher formal levels of qualifications today than people born in earlier cohorts. This goes along with a delayed entry into the labour market.

Figure 1, which displays employment rates over the life cycle of selected birth cohorts, shows that there is a tendency for younger cohort to delay labour market entry. For instance, among individuals born in 1959, around 70% were working by the age of 18. For the cohort born in 1990, this figure was only 50%. However, looking into the employment rates when people are in their twenties, labour market trajectories across cohorts consistently converge. The employment rate at age 30 stands between 70 and 75% for the three selected cohorts shown. There is virtually no difference in employment rate at age 40 between the cohorts born in 1959 and 1970. We can also observe that recessions tend to affect the younger cohorts more severely.


Figure 1 Employment rates (as % of the total cohort’s population) of different birth cohorts

Source: Labour Force Surveys 1975-2013.

The ‘scarring’ effects of youth unemployment

We then analysed how young people’s labour market entry during a recession affected subsequent adult employment trajectories (‘scarring’ effects of youth unemployment) exploiting the differences in labour market conditions people face when leaving education. We estimated the effects of entering the labour market in a recession using linear regression models separately for men and women [2]. Estimated employment probabilities by age (and surrounding confidence intervals) are graphed in Figure 2. The scarring effect of high youth unemployment was identified based on periods of high youth unemployment [3] affecting several birth cohorts on labour market entry. [4]

We focused on differences in employment rates (as a percentage of the total labour force) and found significantly reduced employment for young people initially affected by adverse labour market conditions. As the top panel of Figure 2 shows, only 72% of 18 year-old men who entered the labour market at a time of high unemployment were employed, compared to more than 80% of those leaving school at a more favourable time. Differences are less pronounced for women than men, as the difference in employment rate at age 18 is estimated to be around 5 percentage points (compared to 8 percentage points for men).

Our analysis confirms the ‘scarring’ found in Gregg’s (2001) [5] analysis based on the 1958 National Child Development Study (NCDS). However, in contrast to Gregg (2001), our analysis found that:

• Trajectories differed only in early adult life (we only find an effect for males up to the age of 22 and for females up to the age of 26) and converged later on, meaning that the scarring effect seems to be only of temporary nature;
• Effects are less substantial for men, but this could result from not including inactivity in our analysis, which was decided because the inactivity status in LFS is not as informative as in the NCDS.

Figure 2 Effects of high youth unemployment on labour market trajectories

Source: LFS and own calculations

Conclusion

We found evidence of a ‘scarring’ effect resulting from youth unemployment in early adult life until people are in their 20s - that is, higher unemployment compared with people making ‘good’ initial transitions to the labour market. This has clear implications for individual wellbeing well beyond the experience of youth unemployment itself, including likely reduced life time earnings and further effects, e.g. on family formation, savings and pensions. In the light of this, we believe Gregg’s (2001) main conclusion that improving (vocational) education investment as the key variable to help people make good transitions remains a key policy conclusion.

The provision of high quality vocational education is the main instrument to achieve this.

* * * 

[1] https://www.gov.uk/government/publications/young-people-and-their-movement-between-education-and-work


[2] Using a pooled dataset of almost 40 years of LFS data, we estimated linear regression models of the age profile (second order) on employment outcomes, controlling for the education composition of the cohorts (four levels) and a time trend (second order), which accounted for the generic employment growth having affected the UK aggregate since the 1970s


[3] Defined as >75th percentile of all unemployment rates observed for the cohorts 1959-1997 when aged 18


[4] Similar effects for particular cohorts such as the 1967 birth cohort, which was affected by the structural weaknesses of the youth labour market in the 1980s, cannot be obtained due to the identity cohort, age and time effects for individual cohorts.


[5] Gregg P (2001), “The impact of youth unemployment on adult unemployment in the NCDS”, The Economic Journal, 111, F626-F653.