Institut fr Gerontologie an der TU Dortmund Jrgen

Institut für Gerontologie an der TU Dortmund Jürgen Bauknecht* | Andreas Cebulla^ | Mikkel Barslund° * Institute of Gerontology at TU Dortmund University ^ SA Centre for Economic Studies, University of Adelaide ° Centre for European Policy Studies, Brussels Working conditions and retirement: how important are HR policies in prolonging working life? ‘Work, Age and Employment – Evidence from Longitudinal Studies (WAHE 2016)’ University of Wuppertal, Germany, September 21 st 2016

Institut für Gerontologie an der TU Dortmund Contens § § § 1. Working conditions and longer working lives 2. State of research for 9 work satisfaction components 3. Data 4. Results 1 – 5 5. Evaluation of maximum effects 6. Summary and conclusions 2

Institut für Gerontologie an der TU Dortmund 1. Working conditions and aspects of longer working lives § Macro (primarily national policies) Pensions, unemployment, disability, subsidies/tax breaks § Meso (companies/organisations, social partners) Labour supply (e. g. workability-measures for health, skill development, motivation) § Labour demand Field of action Empirical basis § Micro (not as field of action) Determinants e. g. family status, health status, attitudes Meso level measures → micro level factors → retirement behaviour 3

Institut für Gerontologie an der TU Dortmund HR components → retirement/not 4

Institut für Gerontologie an der TU Dortmund 2. Work satisfaction components My job is physically demanding. I am under constant time pressure due to a heavy workload. I have very little freedom to decide how I do my work. I have an opportunity to develop new skills. I receive adequate support in difficult situations. I receive the recognition I deserve for my work. Considering all my efforts and achievements, my [salary is/earnings are] adequate. My [job promotion prospects/prospects for job advancement] are poor. My job security is poor. 5

Institut für Gerontologie an der TU Dortmund 2. State of research (1/6) 1. Job is physically demanding → early labour market exit (8 longitudinal quant. studies, van den Berg et al. 2010) → disability retirement (Krause et al. 1997) → retirement intentions (for job demands and emotional demands, Burke et al. 2012) → recovery needs and following labour market exit (Schreur et al. 2011) → early retirement (Szubert/Sobala 2005) → disability retirement due to musculosceletal diseases (Lahelma et al. 2012) → weaker effects on early retirement intentions than mentally demanding work (von Bonsdorff et al. 2010) → no evidence for work exit (Carr et al. 2016) → no evidence for retirement intentions (Henkens/Leenders 2010) → no significant effects on labour market participation (Fleischmann et al. 2013) → evidence for any kind of early retirement only for women, for disability retirement men and women (Blekesaune/Solem 2003) 6

Institut für Gerontologie an der TU Dortmund 2. State of research (2/6) 2. Time pressure → lower probability of bridge employment (Wang et al. 2008) → frequent deadlines and time pressue lead to disability exit (Krause et al. 1997) → work overload leads to early retirement intentions (Burke et al. 2012) → autonomy over work pace increases probability of labour market participation (employed or unemployed in contrast to e. g. being retired, Fleischmann et al. 2013) → job demand index (2/4 variables on time pressure) no considerable effects on early retirement for men, and statitsically insignificant for women (Mein et al. 2000) → effects of job pressure only on early retirement intentions but not on early retirement (van Solinge/Henkens 2013) → effects on workload on early retirement intentions not significant, possibly due to the inclusion of ‚burnout factors‘ (Henkens/Leenders 2010) → no effect from ‚psychsocial demands‘ (incl. fast work and time pressure) on labour market exit (Carr et al. 2016) 7

Institut für Gerontologie an der TU Dortmund 2. State of research (3/6) 3. Very little freedom to decide how to do work → job control negatively affects early retirement intentions (Elovainio et al. 2005) → job control foster work enjoyment which lower early retirement intentions (Schreur et al. 2011) → Low job control increases likelihhod of depressive symptoms which fosters early retirement (Lunau et al. 2013) → decision authority negatively affects work exit (Carr et al. 2016) → job control lowers probability of disability due to musculoskeletal diseases and especially mental disorders (Lahelma et al. 2012) → autonomy over work fosters probability to be in work or unemployed (Fleischmann et al. 2013) → no significant association between low autonomy and expected retirement age, but low-n (Beehr et al. 2000) → no effects of autonomy on later retirement (Henkens/Lenders 2010) → effect of autonomy on willingnes to continue to work not signficant, but low-n (Shacklock/Brunetto 2008) 8

Institut für Gerontologie an der TU Dortmund 2. State of research (4/6) 4. Opportunity to develop new skills → no results found so far 5. Adequate support in difficult situations → negative effect of co-worker support on disability retirement insignificant but signficant with supervisor support (Krause et al. 1997) → effect on early retirement intentions but weak on actual retirement behaviour (van Solinge/Henkens 2013) → low social support fosters labour market exit (disability physical/mental see above) (Lehalma et al. 2012) → low social support fosters early labour market exit, but unstable significance (Carr et al. 2016) 9

Institut für Gerontologie an der TU Dortmund 2. State of research (5/6) 6. Recognition considered adequate and 7. Earnings considered adequate → negative effort-reward imbalance and negative assessment of justice harms health (Kivimäki et al. 2007, Fischer/Sousa-Poza 2007, Lunau et al. 2013 for depression). Since health affects retirement behaviour (e. g. Wang et al. 2008, Behr et al. 2000, Szubert/Sobala 2005) subjective negative injustice foster labour exit 8. Promotion prospects poor → also part of justice in 6. and 7. , Kivimäki et al. 2007) → Subjective promotion prospects, opportunities for growth and absence of feeling of dead end delays retirement (van Solinge/Henkens 2013) → negative effects of opportunities for growth on early retirement intentions (Henkens/Leenders 2010) 10

Institut für Gerontologie an der TU Dortmund 2. State of research (6/6) 9. Job security is poor → job insecurity as main driver of early retirement (Winter-Ebmer et al. 2011) → fear of losing job delays retirement (Debrand/Sirven 2009) 11

Institut für Gerontologie an der TU Dortmund HR components → job satisfaction → retirement/not 12

Institut für Gerontologie an der TU Dortmund 3. Data § SHARE § § 1 2 4 5 2004 2007 2011 2013 13

Institut für Gerontologie an der TU Dortmund 4. Results (1/5) How well does HR component predict overall job satisfaction? (Odd ratios going from worst to best, ordered logit [<1 neg]). IV turned to positive direction Physical demanding 0. 9 Time pressure 1. 45** Little freedom 2. 26*** Develop new skills 2. 73*** Support 1. 98*** Recognition 5. 31*** Salary Prospects for promotion Job security 2. 9*** 1. 08 2. 22*** N=23, 791, 75 > age >=50, employed. **: p = 0. 05, ***: p = 0. 001 14

Institut für Gerontologie an der TU Dortmund 4. Results (2/5) Most people ‚satisfied‘ or ‚very satisfied‘ with job Satisfied with job High education Low education 50 -54 55 -59 Strongly agree 48 51 39 40 Agree 46 41 51 52 Disagree 5 6 7 6 Strongly disagree 1 2 3 2 SHARE wave 1, 2004. 15

Institut für Gerontologie an der TU Dortmund 4. Results (3/5) Average number of years worked after 2004, 50 -54 (descriptive, no controls) Maximum 9 years (for data reasons 2004 – 2013) Satisfied with job High edu Low edu 50 -54 55 -59 Strongly agree 6. 8 5. 2 6. 3 4. 4 Agree 6. 4 4. 7 5. 9 4. 3 Disagree 4. 9 5. 2 4. 0 Strongly disagree 5. 2 6. 2 5. 2 2. 7 SHARE various waves. Low number of observations in ‚disagree‘ and ‚strongly disagree‘ cells 16

Institut für Gerontologie an der TU Dortmund 4. Results (4/5) Effect of overall work satisfaction on number of years worked (2004, 2007, 2011, 2013) Working years (one of several specifications with similar results) Average number of years worked for 50 -54 y/o: 7. 1 50 -54 y/o Base 55 -59 y/o -3. 7*** 60 -64 y/o -6. 6*** Women -0. 3*** Strongly dissatisfied Base Dissatisfied -0. 01 Satisfied 0. 41* Strongly satisfied 0. 63** Controls for interview year and country N=20, 201, 50<= age <65. SHARE various waves. + 0. 63 years = 7. 5 months 17

Institut für Gerontologie an der TU Dortmund 4. Results (5/5) Percentage wanting to retire ‘as early as possible’ by education, work satisfaction and age groups, age 50 -75. Very low n in ‘disagree’ ‘strongly disagree’ espc. Older groups → even amongst those satisfied a high share wants to quit asap 18

Institut für Gerontologie an der TU Dortmund 5. Evaluation of maximum effects (1/4) Rise of the legal retirement age § DE: + 2 years: Simulation analysis: ~ 11 months later retirement [Fehr et al. 2010] § § DE: Rise of statutory retirement age and of early retirement age by 2 years: Projection: men + 9 months (based on old system) [Berkel/Börsch-Supan 2004] NL: Rise (public sector workers) by 1 year: Retirement expectations + ~ 7 months by 2 year: Retirement expectations + ~ 11 months (men 7 months, women 17 months, largely due to higher incomes amongst men, who can afford pension cuts due to earlier retirement. Also weak effects amongst those with pension entitlements in the 2 nd and 3 rd pillar [de Grip et al. 2013] § ES: Rise of statutory retirement age from 65 to 68: Projection: + 20 months [Díaz-Giménez/Díaz-Saavedra 2009, 2014]. 19

Institut für Gerontologie an der TU Dortmund 5. Evaluation of maximum effects (2/4) Rise of the legal retirement age § US: + 2 months: Projection: + 1 month or 12/18 months for 2/3 years [Mastrubuoni 2006] § § FI: Rise of entry age in regular pension, part-time pension and the unemployment tunnel by 2 years: Projection: + 6 months [Määtänen 2014] FI: Rise of the retirement corridor 63 -68 → 65 -70. Projection: + 1 -2 years [Tikanmäki et al. 2015] § Benefit rises for later retirement ES: Rise of retirement benefits due to longer working lives actuarilly fair, so not 3% at age 65 but ~ 7% annually and rising to 15. 6% at age 80: Projection: + 6 working months [Díaz-Saavedra 2015] 20

Institut für Gerontologie an der TU Dortmund 5. Evaluation of maximum effects (3/4) Retirement: Other issues § ES: Complete removal of restrictions to combine work with pension income: Projection: + 2. 7 working years [Díaz-Saavedra 2015] Unemployment benefits § FI: Higher age threshold for extended unemployment benefits (55 → 57): Large drop in unemployment between 55 and 59 (25 000 → 10 000) between 2005 and 2009, (Braconier 2010), later labour market exit by + 1. 3 months [Kyyrä 2010] 21

Institut für Gerontologie an der TU Dortmund 5. Evaluation of maximum effects (4/4) Disability schemes § § DE: Closure of disability insurance exit pathway: Projection: + 7 months for men and 2. 5 months for women [Jürges et al. 2014] FI: Abolishment of disability scheme ‚individual early retirement‘ (has been done for those born after 1943): Projection: + 3. 4 working months [Kyyrä 2015] 22

Institut für Gerontologie an der TU Dortmund 6. Summary and Conclusions § § Strong effects for small group of dissatisfied Weak effects for large group of satisfied workers § § Effects dwarfed by policy reforms But in contrast to policy reforms´ effects, positive HR effects important in themselves and do not come with disadvantages 23

Institut für Gerontologie an der TU Dortmund Dr. Jürgen Bauknecht jbauk@post. tu-dortmund. de ++49 (0) 231 728488 -15 Dr. Andreas Cebulla andreas. cebulla@adelaide. edu. au ++61 (0)8 83130013 Dr. Mikkel Barslund mikkel. barslund@ceps. eu ++32 2 229 39 49

Institut für Gerontologie an der TU Dortmund 7. References (1/5) § § § § Beehr, T. A. , Glazer, S. , Nielson, N. L. , Farmer, S. J. (2000), ‘Work and Nonwork Predictors of Employees´ Retirement Ages’, Journal of Vocational Behavior 57, 206 -225 Berkel, B. , Börsch-Supan, A. (2004): Pension Reform in Germany. The Impact on Retirement Decisions, MEA Discussion Paper 62 -2004. Mannheim. Blekesaune, M. , & Solem, P. (2005). Working conditions and early retirement: A prospective study of retirement behaviour. Research on Ageing, 27(1), 3 − 30. Braconier, H. (2010): Coping with the Job Crisis and Preparing for Ageing: The Case of Finland. OECD Economics Departmnet Working papers, No. 7778, OECD Publishing. Burke, R. , Dolan, S. L. , Fikesenbaum, L. (2012), ‘Predictors of the decision to retire among nurses in Spain: A descriptive survey’, International Journal of Nursing, 1(2), 2012, 25 -32. Carr, E. , Hagger-Johnson, G. , Head, J. , Shelton, N. , Stafford, M. , Stansfeld, S. , Zaninotto, P. (2016): Working conditions as predictors of retirement intentions and exit from paid employment: a 10 -year follow-up of the English Longitudinal Study of Ageing. European Journal of Ageing, Vol. 13 (1), 39 -48. Debrand/Sirven 2009 de Grip, A. /Fouarge, D. /Montizaan, R. (2013): How Sensitive are Individual Retirement Expectations to Raising the Retirement Age? De Economist (2013), 161, 225 -251. 25

Institut für Gerontologie an der TU Dortmund 7. References (2/5) § § § § Díaz-Giménez, J. /Díaz-Saavedra, J. (2009): Delaying retirement in Spain. Review of Economic Dynamics 12 (2009), 147 -167. Díaz-Giménez, J. /Díaz-Saavedra, J. (2014): The Future of Spanish Pensions. Dpto. Teoría e Historia Económica Universidad de Granada, Working Paper n. 14/03, May 2014. Díaz-Saavedra, J. (2015): Tax and Transfer Programs, Retirement Behavior, and Work Hours Over the Life Cycle. Journal of European Policy Reform. Elovainio et al. 2005 Fehr, H. /Kallwit, M. /Kindermann, F. (2010): Pension reform with Variable Retirement Age – A Simulation Analysis for Germany. Journal of Pension Economics and Finance, 11 (3), 389– 417. Fischer, J. A. V. , Sousa-Poza, A. (2007), ‘Does Job Satisfaction Imrpove the Health of Workers? New Evidence Using Panel Data and Objective Measures of Health’, IZA Discussion Paper No. 3256. Fleischmann, M. , Koster, F. , Dykstra, P. , Schippers, J. (2013), ‘Hello pension, goodbye tension? The impact of work and instituions on older workers´ labor market participation in Europe. International Journal of Ageing and Later Life, 2013, Vol. 8 (2), 33 -67. Henkens, K. , Leenders, M. (2010), Burnout and older workers´ intentions to retire. International Journal of Manpower, 31, (3), 306 -321. 26

Institut für Gerontologie an der TU Dortmund 7. References (3/5) § § § Jürges, H. /Thiel, L. /Bucher-Koenen, T. / Rausch, J. / Schuth, M. / Börsch-Supan, A. (2014): Health, financial incentives, and early retirement: Micro-simulation evidence for Germany. Schumpeter Discussion papers, No. 2014 -003, urn: nbn: de: hbz: 468 -20140206 -112013 -6. Kivimäki, M. , Vahtera, J. , Elovainio, M. , Virtanen, M. , Siegrist, J. (2007), ‘Effort-reward imbalance, procedural justice and relational injustice as psychosocial predictors of health: complementary or redundant models? Occupational & Environmental Medicine, 64, 659 -665. Krause, N. , Lynch, J. , Kaplam, G. A. , Cohen, R. D. , Goldberg, D. E. , Salonen, J. T (1997): ‘Predictors of disability retirement’, Scandinavian Journal of Work, Environment Health, 1997, 23 (6), 403 -413. Kyyrä, T. (2010): Partial unemployment insurance benefits and the transition rate to regular work. European Economic Review 54 (2010), 911 -930. Kyyrä, T. (2015): Early retirement policy in the presence of competing exit pathways: Evidence from pension reforms in Finland. Economica, 82, 46– 78. Lahelma, E. , Laaksonen, M. , Lallukka, T. , Martikainen, P. , Pietiläinen, O. , Saastamoinen, P. , Gould, R. , Rahkonen, O. (2012). , ‘Working conditions as risk factors for disability retirement: a longitudinal register linkage study. BMC Public Health. 2012; 12: 309. http: //dx. doi. org/10. 1186/1471 -2458 -12 -309. 27

Institut für Gerontologie an der TU Dortmund 7. References (4/5) § § § Lunau, T. , Wahrendorf, M. , Dragano, N. , Siegrist, J. (2013), ‚Work stress and depressive symptoms in older employees: impact of national labour and social policies. BMC Public Health, Vol. 13, 1086, [Pub. Med: 24256638] Määtänen, N. (2014): Evaluation of Alternative Pension Policy Reforms Based ona Stochastic Life Cycle Model. In: Lassila, Jukka/Määtänen, Niku/Valkonen, Tarmo (eds. ): Linking retirement age to life expectancy – what happens to working lives and income distribution? Finnish Centre For Pensions Reports ETLA B 260, 2/2014, 17 -52. Mastrubuoni, G. (2006), Labor Supply Effects of the Recent Social Security Benefit Cuts: Empirical Estimates Using Cohort Discontinuities. CEPS Working Paper No. 136 Mein, G. , Martikainen, P. , Stansfeld, S. A. , Brunner, E. J. , Fuhrer, R. , Marmot, Mi. G. (2000), Predictors of early retirement in British civil servants, Age and Ageing 2000, (29), 529 -536. Schnalzenberger, M. , Schneeweis, N. , Winter-Ebmer, R. , Zweimüller, M. (2011), Job Quality and Employment of Older People in Europe, The Austrian Center for Labor Economics and the Analysis of the Welfare State Working Paper No. 1105. Schreurs, B. , de Cuyper, N. , van Emmerik, I. J. H. , Notelaers, G. , de Witte, H. (2011), ‘Job demands and resources and their associations with early retirement intentions through recovery need and work enjoyment’, SA Journal of Industrial Psychology/SA Tydskrif vir Bedryfsielkunde, 37 (2), Art. #859, 11 pages. 28

Institut für Gerontologie an der TU Dortmund 7. References (5/5) § § § § § Shacklock, K. , Brunetto, Y. (2008), ‘A model of older workers´ intentions to continue working’, Personnel Review, 40(1), 252 -274. Szubert, Z. , Sobala, W. (2005), ‘Current determinants of early retirement among blue collar workers in Poland’, International Journal of Occupational Medicine & Environmental Health, 18(2), 177 -184. Tikanmäki, H. /Sihvonen, H. / Salonen, J. (2015): Distributional Effects of the Forthcoming Finnish Pension Reform – a Dynamic Microsimulation Approach. International Journal of Microsimulation, 8 (3), 75 -98. van den Berg et al. 2010 Van Solinge, H. , Henkens, K. (2013): Work-related factors as predictors in the retirement decision-making process of older workers in the Netherlands. Ageing & Society 34 (9), 11511174 von Bonsdorff et al. 2010) Wang et al. 2008, Wang, M. , Zhan, Y. , Liu, S. , Shultz, K. S. (2010), ‚Antecedents of Brdige Employment: A Longitudinal Investigation‘, Journal of Applied Psychology, 2008, Vol. 93, No. 4, 818 -930. Winter-Ebmer et al. 2011 29