Is Long-Term Non-employment a Lifetime Disease?

Abstract

Long-term non-employment (which is not long-term unemployment) has been almost neglected in the academic literature, long term here implying up to 15–25 years of absence from the labour market, let alone full and definitive exit. This study takes the lead from a previous paper (2017) in which the magnitude of long term non-employment (LTNE) and its duration are estimated from administrative databases of Italy, Germany and Spain (Contini B, et al., IZA discussion papers, no.11167, 2017). In all three countries long-term nonemployment appears to be a lifetime disease for many workers who drop out of the (official) labour market and never return, left unsheltered from the welfare institutions. The main task of this work is an analytical exploration of the factors leading to LTNE development in Italy, estimated at almost 1.3 million male individuals (about as many as the officially unemployed), average duration exceeding 12 years. An econometric exploration indicates that it is often more profitable for employers to hire new unexperienced young workers in place of confirming individuals already onthe-job, leading to excessive turnover, long-term non-employment and waste of human capital. There are strong policy implications of this result as the EU Commission has for many years advocated low wages for new entrants and high contract flexibility as major instruments to promote youth employment.

This is a preview of subscription content, log in to check access.

Fig. 1
Fig. 2
Fig. 3
Fig. 4

Notes

  1. 1.

    Definitions:

    1. (i)

      “OLF” = out of the labour force (inactive), but willing to work if given the chance;

    2. (ii)

      “long-term unemployment” = unemployment lasting more than 1 year, according to almost all official sources. Occasionally it is indicated as “lasting more than 2 years”;

    3. (iii)

      “LTNE = long-term non-employment” (our) estimate relating to individuals in working age who lose their job (in the official economy) and no longer find a new one.

  2. 2.

    The paper does not report country-based studies loosely related to the rest of the literature, that document a positive, although modest frequency of transformations of temporary contracts into open-ended contracts: Berton et al. (2011, 2012), Bentolila et al. (2010, Bonnal et al. (1997), Booth et al. (2002), Bover and Gomez (2004), Dolado et al. (2002), Ichino et al. (2008), Picchio (2008).

  3. 3.

    For instance, Tatsiramos estimates unemployment duration for a number of EU countries based on official LFS data (2010) and reports durations of 1–2 years.

  4. 4.

    WHIP is Italy’s longitudinal database originating from Social Security records. It is a large random sample representative of the universe of employees in the private sector, of the non-tenured employees in the public sector, the self-employed and the professionals, as well as all workers covered by atypical (non-standard) contracts. WHIP also contains information on unemployment and Cassa Integrazione Guadagni (CIG) episodes. It is integrated by the INPS-Casellario degli Archivi containing the working careers of individuals who moved from dependent work into the public sector. A caveat is necessary on this point: some transitions from private to public employment of individuals with special university training may have eluded the linkage: these individuals will appear as non survivors, downward biasing the relative survival rate. Data on educational attainment are unrecorded in the WHIP database. The sampling design consists of selecting all individuals aged 18-30 at their first labour market entry, born on the 10th of each month, leading to a 1:90 sample/population ratio. For the time being, data on workers’ injuries and professional diseases contained in the INAIL databases are not available in WHIP, but related temporary absence from work is observed. The WHIP database is available to interested researchers who will apply to the INPS, Social Security Administration and guarantee to comply with their standards.

  5. 5.

    Registered unemployment is recorded in the INPS databases, and refers to individuals who report to the Employment Agencies (Centri per l’Impiego) and may, if strict requirements allow, draw unemployment benefits. The unemployment figures reported by ISTAT are estimated instead from the Labour Force Survey, based on individual interviews: these will usually be much higher than the officially registered ones.

  6. 6.

    Men (age 15–65) drawing disability pensions are about 0.7% of overall employment. Temporary illness and disabilities amount to 3–4% of male population in working age (from INPS Database—Archive O1 M, basis for WHIP).

  7. 7.

    see Contini et al. (2017).

  8. 8.

    All these characteristics are available since 1987. Entrants in previous years (starting in 1980) can be observed, but some detailed information is missing.

  9. 9.

    The survival rate of the two youngest cohorts is 80%. The frequency of entrants in 1987 at age 26++  is about 20% of the sample. The EHCP suggests that about 30% of the late entrants have university degrees, while there is no indication of employment in the public sector. The university graduates are the most likely to have joined the public sector after an initial working episode in private employment. Surprisingly, however, many report to be in difficult economic conditions. See also footnote (21).

  10. 10.

    The WHIP observation window is 1987–2012 (25 years), but INPS provided some additional information on the previous working history of all sampled individuals who were at least 55 years old in 2012. The resulting observation window covers 1980–2012, except for the caveats mentioned in footnote 5.

  11. 11.

    The decline in male and, more generally, youth participation is a long-term phenomenon dating to the mid Seventies and common to all the EU countries, a consequence of increased schooling age and women’s participation. It was the object of investigation in Contini (2011).

  12. 12.

    Reliable micro-data are inexistent. Battistin and Rettore (2008) indicate that people who work in the irregular economy are unlikely to reveal their status in the LFS interviews for fear of being disclosed. More generally, according to these authors, the likelihood of misclassification among the unemployed, the inactives and the irregulars is always extremely high.

  13. 13.

    A plausible, yet untested, explanation for the Italian OLF-exception is that only a small proportion of Italy’s working population is eligible for unemployment benefits: Italy’s recipiency rate is 32%, against 50% in the UK, 60% in France, 65% in Denmark, 73% in Spain, 94% in Austria and 100% in Germany (OECD figures, although these rates do not imply the same degree of generosity). In Italy there is little incentive to self-report one’s true employment status because the opportunity cost is often close to zero. Where unemployment benefits are generously available, as in Germany, the opportunity cost of misreporting is high because the perceived risk of losing the benefits is high as well. If only half of the Italians classified as inactive but willing to work, were (conservatively) counted among the unemployed—as would be the case anywhere else in the EU—Italy’s unemployment rate would be well above the optimistic 13% reported by official sources in 2013. This question will make the object of a separate paper.

  14. 14.

    In a multiple-spell setting mobility is fully endogenous as it is activated at each job change.

  15. 15.

    Unless the complete parametrization of the survival function is known (and estimable), Cox’s regression methods do not provide estimates of the benchmark exit probability, but only hazard ratios indicating the distance from the benchmark.

  16. 16.

    As usual, medians are preferable to means in order to avoid the influence of outliers.

  17. 17.

    MOB-PRED is the OLS first stage predictor of MOB necessary to compute MOB-2SLS. In addition RESIDUALS = MOB − MOB-PRED.

  18. 18.

    This result recalls a similar one obtained in a model of labour demand with permanent and temporary contracts (Contini 2011): small firms hire permanent workers more frequently than larger ones as the investment spent for the enhancement of their employees’ specific human capital and loyalty is often high.

  19. 19.

    An alternative predictor MOB^^^ has been estimated using all the exogenous covariates of the HAZARD equation with the addition of a new exogenous covariate—a string < 0, 1, 2, 0, 1, 2, 0, 1, 2 …… > . The correlation between MOB^^ and MOB^^^ is 0.982. The results of the 2SRI estimation are unchanged.

  20. 20.

    The European Household Community Panel (EHCP) provides some information on the workers’ personal characteristics, unavailable in the administrative data (Contini et al. 2017). The EHCP indicates that the survivors are often better off than the non-survivors on all the items related to one’s wellbeing, education and past unemployment experience. This holds also for the entrants of age 26–30 whose survival rates appeared suspiciously low in ch.3. In addition, there are strong hints of a relatively high participation (not necessarily full-time) to the irregular economy, both among those who report to have worked without a contract, and also among those who do not indicate any contract typology.

  21. 21.

    The relative weight of the black/irregular economy on Italy’s GNP is estimated at 24% by ISTAT National Accounts. Alternative methods of estimation make use of different macroeconomic hypotheses Feige (1979). Schneider (2011) estimates the share of irregular activities on GNP for several OECD countries: Italy ranks among the highest at 21.5% and Germany, among the lowest, at 13.5%. The share of irregular employment on overall employment in Italy is estimated at around 16%. Ardizzi et al. (2013) indicate 16.5% of Italy’s GNP as attributable to irregular work in legitimate activities and 21.6% inclusive of all criminal activities. The same authors estimate the share of irregular activities in GNP at 19.2% in Spain, 13.5% in Germany and 11% in France.

  22. 22.

    Battistin and Rettore (2008) indicate that people who work full time in the irregular economy are unlikely to reveal their status in the course of LFS interviews for fear of being disclosed. More generally, the likelihood of misclassification among the unemployed, the inactives and the irregulars is often very high.

  23. 23.

    The collective layoffs from the industrial sectors in the 80s and 90s involved mainly blue collar workers in their 60s by 2012, who could easily find niches in the irregular sectors (construction, maintenance, small trades and public services).

  24. 24.

    The simulation exercise—based on a pseudo-Markovian process with state dependence due to long-term joblessness—provides a rough estimate of the long-run impact of policy changes. Today’s overall employment rate is about 69%, the remaining 31% including unemployment and non-employment. If the annual transition probability from unemployment/non-employment to employment increases by 10%, the long-term (steady state) employment rate is estimated to reach 71% in about 10 years. The ceteris paribus assumption implicit in this simulation is, obviously, very restrictive. Only a general equilibrium approach would improve the credibility of this exploration.

  25. 25.

    The quest for new personnel is well founded: the employment of Italy’s public placement agencies is one tenth that of Germany, with France and the UK not far behind. On the other hand, the impact of a tenfold multiplication of placement agencies is impossible to ascertain via a model like the one implemented here.

  26. 26.

    It is sometimes based on a very rough comparison of average yearly working hours: 1377 in Germany and 1725 in Italy. If Italy were to reach the German level, it would imply a 25% addition to overall employment.

References

  1. Abraham K, Haltiwanger J, Sanduski K, Spetzler J (2018) The consequences of long-term unemployment: evidence from linked survey and administrative data. ILR Rev. https://doi.org/10.1177/0019793918797624

    Google Scholar 

  2. Ardizzi GC, Petraglia M, Piacenza G, Turati (2013) Measuring the underground economy with the currency demand approach: a reinterpretation of the methodology, with an application to Italy. J Income Wealth, 747–772

  3. Battistin E, Rettore E (2008) Ineligibles and eligible non-participants as a double comparison group in regression-discontinuity designs. J Econ 14:715–730

    Article  Google Scholar 

  4. Bentolila S, Cahuc P, Dolado J, Le Barbanchon T (2010) Two-tier labour markets in the great recession: France vs. Spain. CEMFI, W.P. 1009

  5. Berton F, Devicienti F, Pacelli L (2011) Are temporary jobs a port of entry into permanent employment?: Evidence from matched employer‐employee. Int J Manpow 32(8):879–899

    Article  Google Scholar 

  6. Berton F, Richiardi M, Sacchi S (2012) The political economy of work security and flexibility: Italy in comparative perspective. Policy Press, Bristol

    Google Scholar 

  7. Blanchard O (2006) European unemployment. Econ Policy 21(45):6–59

    Article  Google Scholar 

  8. Blanchard O, Summers L (1986) Hysteresis and the European unemployment problem. In: NBER macroeconomics annual, vol 1, Fall 1986. MIT Press, Cambridge, pp 15–78

  9. Bonnal L, Fougere D, Serandon A (1997) Evaluating the impact of French employment policies on individual labour market histories. Rev Econ Stud 64(4):683–713

    Article  Google Scholar 

  10. Booth A, Francesconi M, Frank J (2002) Temporary jobs: stepping stones or dead ends? Econ J 112(480):F189–F213

    Article  Google Scholar 

  11. Bover O, Gomez R (2004) Another look at unemployment duration: exit to an open-ended vs a temporary job. Investig Econ 28(2):285–314

    Google Scholar 

  12. Contini B (2011) A nutshell model of labour demand with permanent and short-term contracts. unpublished manuscript

  13. Contini B, Garcia Perez JI, Pusch T, Quaranta R (2017) New approaches to the study of long-term non-employment duration in Italy, Germany and Spain. IZA discussion papers, no.11167

  14. Dolado JJ, Garcìa-Serrano C, Jimeno JF (2002) Drawing lessons from the boom of temporary jobs in Spain. Econ J 112(480):F270–F295

    Article  Google Scholar 

  15. Farber HS (1993) The incidence and cost of job loss: 1982–1991. Brooking Pap Econ Act Microecon 1:73–132

    Article  Google Scholar 

  16. Feige EL (1979) How big is the irregular economy? Challenge 12:5–13

    Google Scholar 

  17. Gomes P (2012) Labour market flows: facts from the United Kingdom. Labour Econ 19(2):165–175

    Article  Google Scholar 

  18. Güell M, Petrongolo B (2007) How binding are legal limits? Transitions from temporary to permanent work in Spain. Labour Econ 14(2):153–183

    Article  Google Scholar 

  19. Hausman JA (1978) Specification tests in Econometrics. Econometrica 46:251–1271

  20. Ichino A, Mealli F, Nannicini T (2008) From temporary help jobs to permanent employment: what can we learn from matching estimators and their sensitivity? J Appl Eco 23(3):305–327

    Article  Google Scholar 

  21. Krueger AL (2015) How tight is the labour market?. NBER Report 2015, no.3

  22. Krueger AL, Cramer J, Cho D (2014) Are the long-term unemployed at the margin of the labor market?. In: Brookings Papers on Economic Activity, Spring 2014

  23. Lucifora C, Mcknight A, Salverda W (2005) Low-wage employment in Europe: a review of the evidence. Socio-Eco Rev 3(2):259–292

    Article  Google Scholar 

  24. Machin S, Manning A (1999) Causes and consequences of longterm unemployment in Europe. In: Ashenfelter O, Card D (eds) Handbook of labour economics. North Holland, Amsterdam

    Google Scholar 

  25. Mroz P, Savage (2006) The long-term effects of youth unemployment. J Human Resour XLI:259–293

    Article  Google Scholar 

  26. Mussida C, Sciulli D (2015) Flexibility policies and re-employment probabilities in Italy. BE J Econ Anal Policy 15(2):621–651

    Google Scholar 

  27. Nickell S, Layard R (1999) Labour market institutions and economic performance. In: Ashenfelter O, Card D (eds) Handbook of labour economics. North Holland, Amsterdam

    Google Scholar 

  28. Picchio M (2008) Temporary jobs and transition to stable jobs in Italy. Labour 22(1):147–174

    Article  Google Scholar 

  29. Saint-Paul G (2004) Dual labour markets. MIT Press, Cambridge

    Google Scholar 

  30. Schneider F (2011) The shadow economy, theoretical approaches, empirical studies and political implications. Cambridge University Press, Cambridge

    Google Scholar 

  31. Terza JV, Basu A, Rathouz PJ (2008) Two-stage residual inclusion estimation: addressing endogeneity in health econometric modeling. J Health Eco 27(3):531–543

    Article  Google Scholar 

  32. Torelli N, Trivellato U (1993) Modelling inaccuracies in job search duration data. J Eco 59:1–2

    Article  Google Scholar 

Download references

Author information

Affiliations

Authors

Corresponding author

Correspondence to Bruno Contini.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

This paper has been in the pipeline for many years. It has benefited from numerous insightful suggestions of two excellent referees. The authors also wish to thank the Editor for guidance in a lengthy submission process. D. Card read an earlier version of this work and offered useful comments.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Contini, B., Quaranta, R. Is Long-Term Non-employment a Lifetime Disease?. Ital Econ J 5, 79–102 (2019). https://doi.org/10.1007/s40797-019-00083-2

Download citation

Keywords

  • Labour markets
  • Participation
  • Non-employment
  • Duration

JEL classification

  • J00
  • J01
  • J08
  • J1
  • J6
  • J64