Educational differences in macro-level determinants of early exit from paid work: a multilevel analysis of 14 European countries

Do early life factors explain the educational differences in early labour market exit? A register-based cohort study

BACKGROUND

Socioeconomic inequalities in labour market participation are well established. However, we do not fully know what causes these inequalities. The present study aims to examine to what extent factors in childhood and late adolescence can explain educational differences in early labour market exit among older workers.

METHODS

All men born in 1951-1953 who underwent conscription examination for the Swedish military in 1969-1973 (n = 145 551) were followed from 50 to 64 years of age regarding early labour market exit (disability pension, long-term sickness absence, long-term unemployment and early old-age retirement with and without income). Early life factors, such as cognitive ability, stress resilience, and parental socioeconomic position, were included. Cox proportional-hazards regressions were used to estimate the association between the level of education and each early labour market exit pathway, including adjustment for early life factors.

RESULTS

The lowest educated men had a higher risk of exit through disability pension (HR: 2.72), long-term sickness absence (HR: 2.29), long-term unemployment (HR: 1.45), and early old-age retirement with (HR: 1.29) and without income (HR: 1.55) compared to the highest educated men. Factors from early life explained a large part of the educational differences in disability pension, long-term sickness absence and long-term unemployment but not for early old-age retirement. Important explanatory factors were cognitive ability and stress resilience, whilst cardiorespiratory fitness had negligible impact.

CONCLUSIONS

The association between education and early exit due to disability pension, long-term sickness absence and long-term unemployment was to a large part explained by factors from early life. However, this was not seen for early old-age retirement. These results indicate the importance of taking a life-course perspective when examining labour market participation in later working life.

The Implication of Physically Demanding and Hazardous Work on Retirement Timing

The need to retain individuals longer in the workforce is acknowledged in many high-income countries. The present study therefore aimed to examine the importance of physically demanding work tasks (PDWT) and physically hazardous work environment (PHWE) in relation to retirement timing among pensionable workers (≥61 years). A particular question was whether PDWT and PHWE increased in importance with age. Six waves (2008–2018) of the Swedish Longitudinal Occupational Survey of Health (SLOSH) were used (n = 5201; 56% women and 44% men; mean age at first survey was 61.0 (SD 2.0) years). Discrete time-event history analysis, stratified by socioeconomic position and gender, showed that among blue-collar workers, PDWT and PHWE were associated with an increased likelihood of retiring within the next two years. With increasing age, high-level PHWE was associated with higher probability of retiring among blue-collar men, whereas heavy PDWT was associated with lower probability of retiring among blue-collar women. Among white-collar workers, having at least some PDWT compared to no PDWT was associated with a lower likelihood of retiring within the next two years. With increasing age, exposure to PHWE was associated with higher probability of retiring among white-collar women. These results suggest that to delay retirements, organizations could offer their older employees, especially blue-collar workers and the oldest white-collar women, alternatives to PDWT and PHWE.

Educational qualification differences and early labor market exit among men: the contribution of labor market marginalization measured across the working life

BACKGROUND

The present study aims to investigate the association between educational qualification and early labor market exit among men and to examine the contribution of labor market marginalization measured across the working life on this association.

METHOD

A register-linked cohort study was conducted including men who completed military service in 1969/70 (born between 1949 and 1951) and were alive at age 55 and not disability pension beneficiaries (n = 40 761). Information on the highest level of educational qualification and the outcome of early exit (disability pension, sickness absence, unemployment, and early old-age pension) was obtained from Swedish nationwide registers between the ages of 55 and 64 years. Labor market marginalization was defined as periods of long-term unemployment and sickness absence over the working life and up to follow-up. Cox regression analyses were used to obtain hazard ratios (HR) with 95% confidence intervals (CI).

RESULTS

Low-educated men were more likely to leave the labor force early due to disability pension or sickness absence (HR: 2.48), unemployment (HR: 2.09), and early old-age pension with- (HR:1.25) and without -income (HR: 1.58). Labor market marginalization across the working life explained a large part of the association for the more involuntary early exit routes (disability pensions, sickness absence, unemployment) and explained very little with regards to the more voluntary early exit routes (early old-age pension with and without income).

CONCLUSION

Exposure to labor market marginalization across the working life was important in explaining educational differences in early labor market exit due to disability pension or sickness absence and unemployment. This study underscores the importance of identifying and implementing preventive measures in the workplace (e.g. adaptions) to prevent new spells of sickness absence and unemployment, especially among low educated individuals.

Does Machine Learning Offer Added Value Vis-à-Vis Traditional Statistics? An Exploratory Study on Retirement Decisions Using Data from the Survey of Health, Ageing, and Retirement in Europe (SHARE)

Do machine learning algorithms perform better than statistical survival analysis when predicting retirement decisions? This exploratory article addresses the question by constructing a pseudo-panel with retirement data from the Survey of Health, Ageing, and Retirement in Europe (SHARE). The analysis consists of two methodological steps prompted by the nature of the data. First, a discrete Cox survival model of transitions to retirement with time-dependent covariates is compared to a Cox model without time-dependent covariates and a survival random forest. Second, the best performing model (Cox with time-dependent covariates) is compared to random forests adapted to time-dependent covariates by means of simulations. The results from the analysis do not clearly favor a single method; whereas machine learning algorithms have a stronger predictive power, the variables they use in their predictions do not necessarily display causal relationships with the outcome variable. Therefore, the two methods should be seen as complements rather than substitutes. In addition, simulations shed a new light on the role of some variables—such as education and health—in retirement decisions. This amounts to both substantive and methodological contributions to the literature on the modeling of retirement.