State dependence in youth labor market experiences, and the evaluation of policy interventions

Abstract

The finding of strong duration dependence in explaining the length of unemployment spells has influenced the design of many labor market policy reforms. However very little work has been done on more complex effects of labor market experiences and in particular on the causal effects of past outcomes involving other labor force states. In this paper we use longitudinal data to investigate the extent of state dependence in labor market outcomes for young Australians. The econometric model estimates the effects of past outcomes in three labor force states, employment, unemployment and out of the labor force, on current transitions between any two states allowing for observed and unobserved heterogeneity. Our findings suggest strong state dependence in all three states and we use the estimates to simulate various policy experiments.Keywords: Transition data, event history analysis, state dependence, unobserved heterogeneity, labor force states, policy effectiveness.J.E.L. Classification Numbers: C33, C41, J64, J68. Acknowledgment: We wish to thank seminar participants at the University of Melbourne, the University of Copenhagen, and the University of Aarhus. Tue Gørgens gratefully acknowledges the financial support of the Department of Family and Community Services. The views expressed in this paper reflect those of the authors and do not represent the opinions of the Department. * Address: School of Economics, University of New South Wales, Sydney NSW 2052, Australia. E-mail: D.Doiron@unsw.edu.au.† Address: SPEAR Centre RSSS, Australian National University, Canberra ACT 0200, Australia. E-mail: Tue.Gorgens@anu.edu.au. 1 Executive summary• In this paper, we extend the literature on youth labor market outcomes (employment, unemployment and being out of the labor force) by modeling the effects of labor market experiences in a general and flexible manner. The model is estimated on longitudinal data for young Australians.• Our estimates suggest that the effects from one's past experiences on the current and future labor market outcomes are large. Consequently periods of employment (unemployment) have long-term effects on the future probability of employment (unemployment) and the length of future employment (unemployment) spells. These effects are over and above the impact of many other personal characteristics and the state of the labor market.• From the perspective of society, the benefits of early intervention may therefore be substantial. If bad labor market outcomes increase the probability of future bad outcomes, a government program which is successful in placing young people in employment early would benefit the affected individuals in the immediate future and would have persistent effects in the longer-term as well.• Our findings have implications for the policy evaluation. The usual methods of policy evaluation which compare labor market outcomes before and shortly after a policy intervention provide an incomplete and possibly misleading estimate of the impacts of programs since they do not include t...

Youth unemployment and active labor market policies in Europe

“…On the one hand, job changes may allow an upward movement in the career-and wage-ladder and allow youth to become aware of their skills and preferences (Topel and Ward 1992). On the other hand, it has been found that job-changes promote further job-changes (occurrence dependence) and that early job instability significantly reduces later wages (see, e.g., Doiron and Gørgens 2008;Neumark and Wascher 2006). Consequently, ALMP programs should also be measured by their ability to improve the job match quality and stability of accepted employment.…”

Comparing Persistence of Product and Process Innovation: A Discrete-Time Duration Analysis of Innovation Spells

“…We expect a positive relationship between the duration of the previous spell and persistence. On the other hand, path dependence, proxied by the number of previous innovative spells, measures the influence of previous episodes of innovation regardless of their duration (Doiron and Gørgens, 2008). A high number of previous spells should indicates a large number of short innovative episodes, therefore we expect a negative relationship between the number of previous spells and persistence.…”

Working while studying: does it lead to greater attachment to the regional labour market?

“…For example, for graduates from universities of applied sciences, a poor match between education and job after graduation is found to lead to income penalty and worse job satisfaction (Diem, 2015). Unemployment at a young age or after graduation has a significant negative effect on future earnings (Doiron and Gørgens, 2008;Mroz and Savage, 2006;Oreopoulos et al, 2012;Wachter and Bender, 2006). In contrast, stability in the early career and a successful transition from school to work increases future labour-market activity (Neumark and Joyce, 2001) and adult earnings (Neumark, 2002).…”