Youth survival on the labour market: Comparative evidence from three Western Balkan economies in The Economic and Labour Relations Review (forthcoming issue) Nikica Mojsoska Blazevski (nikica@uacs.edu.mk) Marjan Petreski Marjan Bojadziev
Outline Motivation Research objective Theoretical background Literature review Labour market data Data and model Results and discussion
Motivation: Some facts for labor markets Macedonia, Montenegro and Serbia shared the same economic and political systems; embarked on similar transition path High overall unemployment, and high youth unemployment Youth face low employment and activity rates School to work transition lasts 2+ years High LTU
Labour Markets: Employment rate of youth, 2000-2015 30.0 28.0 26.0 24.0 22.0 20.0 18.0 16.0 14.0 12.0 10.0 MKD MNE SRB
Unemployment rate of youth 70.0 65.0 60.0 59.7 55.0 50.0 45.0 40.0 35.0 30.0 25.0 35.6 28.3 49.4 37.5 45.2 MKD MNE SRB 20.0
Labour force participation rate of youth 45.0 40.0 35.0 30.0 25.0 20.0 15.0 10.0 5.0 0.0 MKD MNE SRB 2000 2010 2015
Motivation (2) Costs of youth unemployment Large individual and societal costs of youth unemployment Individuals: - Direct costs: lost income - Long-term adverse effects: employment and wage scarring (Arulampalam et al. 2001; Ryan, 2001; Gregg and Tominey, 2004; Fares and Tiongson, 2007; Cruces et al. 2012) Society: Lost growth and development, welfare costs, social misbehavior
The research objective To explore whether the unemployment experience of youth, early in their career, has negative effects on subsequent labour market performance - To assess what factors determine the duration of the unemployment spell of youth, and - To produce survival probabilities
Theoretical background - employment scars - Human capital theory (Backer, 1964) Depreciation of skills and knowledge lead to lower productivity Lower labor market returns Gender differences: females accumulate less human capital and hence will experience lower re-employment wage
Theoretical background - employment scars - Signalling theory (Spence, 1973) Information asymmetry and uncertainty about worker s productivity Statistical screening device based on the group to which the worker belongs Unemployment signals lower productivity Highest among most skilled, male workers
Theoretical background - extension - Theories fit well tight labor markets Socio-economic environment, culture, norms, wage setting and other institutions, informality all different/prevalent among developing economies Different labour markets, unemployment as normal phenomenon Social behaviour and attitudes towards unemployment: greater tolerance and acceptance of unemployment Informality All of which may be driving different patterns of scarring
Literature review - findings - Most studies for developed (OECD) countries find presence of employment scars: Nilsen and Reiso (2011) find that young Norwegians who experience early unemployment spell have 10 p.p. higher chance of being unemployed at year five (decreases to 5 p.p. in year eight) Cockx and Picchio (2011) (Belgium) find that the probability to find a job within two years decreases from 60% (for those unemployed 3 quarters) to 16% (unemployed 7 quarters) and only 6% (unemployed for 11 quarters).
Literature review - findings - Countries/regions with high unemployment: Fares and Tiongson (2007) for Bosnia found significant employment scar: young people that experienced joblessness in 2001 had 11% greater probability of being unemployed in 2004 Lupi and Ordine (2002) find a wage scar of about 8% for the workers that experienced unemployment spell of up to 6 months in Northern Italy, but no scar in Southern Italy
Data ILO s project Work4Youth, School to Work Transition Survey (SWTS) Data for 2014-2015 Representative sample of young people aged 15-29 Ad hoc-module to LFS Cross sectional, employment history Active youth (1,643 observations in Serbia, 1,336 in Montenegro and 1,248 in Macedonia)
Methodological issues Unobserved heterogeneity Some people are always more prone to unemployment because of low education or other less easily observed heterogeneity. (Gregg, 2001, p.631) More motivated, talented, capable and enthusiastic individuals may be more inclined to find a job sooner than the other youth Identified employment scar would actually be a combination of scarring and unobserved characteristics Two potential approaches: IV and fixed-effects
Methodology The unemployment spell has two specifics: first, it measures duration, i.e. elapsing of time; and second, it is grouped in eight distinct categories Duration variable - cannot use linear models unemployment duration not normally distributed some predictions may be left censored at zero, which is unrealistic for the unemployment spell; and OLS would not make a distinction between cases where the unemployment spell still lasts versus the cases where it terminated (and the person is now employed).
Model - unemployment spell - Use the discrete-time duration models Discrete-time duration models - Box-Steffensmeier and Jones (2004) provide a textbook explanation, while a neat application can be found in Muthén and Masyn (2005) Time-to-event data, i.e. time to employment (survival time) Duration of unemployment composed of two parts: the time to employment and the employment status, i.e. if employment occurred or not Hence, the two functions can be estimated, the survival and hazard functions
Model (cont.) Hence, to estimate the employment scars, we fit a logistic regression model in the following form: logit[h ik (t)] = β 1j Σindiv jik +γ 1j Σsocio jik +δ 1j Σwpa jik +μσcountry jk +ε i - indiv jik stands for a set j of individual characteristics of person i, in country k; - socio jik stands for a set j of socio-economic characteristics of the household; - wpaj jik stands for a set j of working preferences and attitudes; - while country jk k stands for a set j of country characteristics (log of the GDP per capita, the share of LTU and the LFPR). The later play the role of country fixed effects. - ε i stands for the individual heterogeneity.
Model (cont.) Capture the effects of previous unemployment duration on the current employment probability (duration dependence) In the second step, we produce the survival probabilities for the eight categories of unemployment spell duration Unobserved heterogeneity is not of our concern since: - Interested in the differences in the scarring effects among the three countries, and not in the level/intensity per country - Unobserved heterogeneity broadly understood as values, mentality and culture of youth is considered to have been shared even within the entire WB region (e.g. Vujadinović, 2004; Pejovich, 2006)
Unemployment spell in the three transition economies Length of seeking a job before the current one (now employed) Length of seeking a job until the present moment (still unemployed) No spell Less than a week Week to month 1-3 months 3-6 months 6 months to a year 1-2 years Over 2 years 30 25 20 15 10 5 0 5 10 15 20 25 30 SRB MNE MKD SRB MNE MKD
Probability 0.00 0.25 0.50 0.75 1.00 Results (1) All persons 0 1 2 3 4 5 6 7 8 Unemployment duration
0.00 0.25 0.50 0.75 1.00 Results (2) By employment status at the time of interview 0 1 2 3 4 5 6 7 8 Unemployment duration Employed Unemployed
Probability 0.00 0.25 0.50 0.75 1.00 Results (3) By country 0 1 2 3 4 5 6 7 8 Unemployment duration MKD SRB MNE
Results (4) We run two tests: 1. log-rank test whereby the null hypothesis states that the distribution of the unemployment duration among the three countries is the same; and 2. LR-test (Chow test for discrete-probability functions) whereby the null states that all coefficients of a model do not vary between disjointed subsets of the data, being the pooled dataset versus the individual countries. Both suggest that the that the three countries follow different employment scarring patterns
Unemployment spell (marginal probabilities)
Survival probabilities (1-hazard rate)
Conclusion The results suggest presence of employment scar even in countries with high unemployment Differences in employment scarring of young persons due to peculiar developments during transition The position of youth in Serbia is worse compared to the adults, which is also reflected in higher scars Recent flexibilization of Macedonian labour market Large internship program in Montenegro for the graduates (0.3% of GDP)
Policymaking Persistent employment scarring calls for measures preventing young people from falling into unemployment at the exit from education Early interventions Interventions can range from internship programs, flexible forms of employment (for instance, trial work for a limited time period), subsidized employment, training measures and so on. Support for larger scale interventions by policymakers (EU youth guarantees)