A computational model of labor market participation with health shocks and bounded rationality

Abstract

This paper presents a computational agent-based model of labor market participation, in which a population of agents, affected by adverse health shocks that impact the costs associated with working efforts, decides whether to leave the labor market and retire. This decision is simply taken by looking at the working behaviors of the other agents, comparing the respective levels of well-being and imitating the more advantageous decision of others. The analysis reveals that such mechanism of social learning and imitation suffices to replicate the existing empirical evidence regarding the decline in labor market participation of older people. As a consequence, the paper demonstrates that it is not necessary to assume perfect and unrealistic rationality at the individual level to reproduce a rational behavior in the aggregate.

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

Fig. 1

Source: first wave of the English Longitudinal Study of Ageing (ELSA)

Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6

Notes

  1. 1.

    This is an Herculean task requiring, among other things, to estimate transition probabilities from one state to all possible future states based on beliefs which, in turn, depend on 26 parameters to be determined. Other questionable assumptions include “rational expectations,” a notion according to which all individual subjective probabilities equate the “objectively estimable population probability” and that requires the capability to anticipate regulatory changes. The paper acknowledges several times that only some justification for such assumptions can be provided.

  2. 2.

    The data used to draw the curves in Fig. 1 are taken from Table 4A.3 in the Annex 4.1—tables on work and retirement of the first wave of the ELSA survey. http://www.elsa-project.ac.uk/reportWave1.

  3. 3.

    As mentioned previously, pensioners and pension income can also be interpreted as people who decide to leave the labor market and receive a subsidy and the corresponding disability benefit, respectively.

  4. 4.

    Simulations are performed using the BehaviorSpace tool in NetLogo, which allows to specify the grid of parameter values and the number of simulations for each point in the grid.

  5. 5.

    These seven classes of age correspond to the ones presented in Table 4A.3 in the Annex 4.1—tables on work and retirement of the first wave of the ELSA survey.

  6. 6.

    In this setup, the term “equilibrium” differs from the concept of Nash equilibrium because our agents, keeping fixed the choices of their peers, are always willing to change their working status as long as they meet another similar agent with a higher well-being.

  7. 7.

    On the contrary, the pension income \(y^\mathrm{p}\) is assumed fixed to an arbitrary value in the model simulations.

References

  1. 1.

    Alonso-Ortiz J (2014) Social security and retirement across the OECD. J Econ Dyn Control 47:300–316

    Article  Google Scholar 

  2. 2.

    Axtell RL, Epstein JM (1999) Coordination in transient social networks: an agent-based computational model of the timing of retirement. In: Aaron HJ (ed) Behavioral dimensions of retirement economics. Brookings Institution Press, Washington, DC, pp 161–183

    Google Scholar 

  3. 3.

    Bound J, Stinebrickner TR, Waidmann TA (2010) Health, economic resources and the work decisions of older men. J Econom 156:106–129

    MathSciNet  Article  MATH  Google Scholar 

  4. 4.

    Burkhauser RV, Butler JS, Gumus G (2004) Dynamic programming model estimates of social security disability insurance application timing. J Appl Econom 19:671–685

    Article  Google Scholar 

  5. 5.

    Camerer CF (1997) Progress in behavioral game theory. J Econ Perspect 11:167–188

    Article  Google Scholar 

  6. 6.

    Epstein JM (2014) Agent_Zero: toward neurocognitive foundations for generative social science. Princeton University Press, Princeton

    Google Scholar 

  7. 7.

    French E (2005) The effects of health, wealth, and wages on labour supply and retirement behaviour. Rev Econ Stud 72:395–427

    Article  MATH  Google Scholar 

  8. 8.

    Gilleskie DB (1998) A dynamic stochastic model of medical care use and work absence. Econometrica 66:1–45

    Article  MATH  Google Scholar 

  9. 9.

    Goudet O, Kant JD, Ballot G (2015) Forbidding fixed duration contracts: unfolding the opposing consequences with a multi-agent model of the french labor market. In: Amblard F et al (eds) Advances in artificial economics. Springer, Berlin, pp 151–167

    Google Scholar 

  10. 10.

    Gustman AL, Steinmeier TL (2002) Retirement and the stock market bubble. In: NBER working paper no. 9404. National Bureau of Economic Research, Cambridge

  11. 11.

    Gustman AL, Steinmeier TL (2015) Effects of social security policies on benefit claiming, retirement and saving. J Public Econ 129:51–62

    Article  Google Scholar 

  12. 12.

    Haan P, Prowse V (2014) Longevity, life-cycle behavior and pension reform. J Econom 178:582–601

    MathSciNet  Article  MATH  Google Scholar 

  13. 13.

    Heyma A (2004) A structural dynamic analysis of retirement behaviour in the Netherlands. J Appl Econom 19:739–759

    Article  Google Scholar 

  14. 14.

    Laun T, Wallenius J (2015) A life cycle model of health and retirement: the case of Swedish pension reform. J Public Econ 127:127–136

    Article  Google Scholar 

  15. 15.

    Laun T, Wallenius J (2016) Social insurance and retirement: a cross-country perspective. Rev Econ Dyn 22:72–92

    Article  Google Scholar 

  16. 16.

    Lettau M (1997) Explaining the facts with adaptive agents: the case of mutual fund flows. J Econ Dyn Control 21:1117–1147

    Article  MATH  Google Scholar 

  17. 17.

    Lurie NH (2004) Decision making in information-rich environments: the role of information structure. J Consum Res 30:473–486

    Article  Google Scholar 

  18. 18.

    Malhotra NK (1982) Information load and consumer decision making. J Consum Res 8:419–430

    Article  Google Scholar 

  19. 19.

    Moro A, Pellizzari P (2016) An agent-based model of labor market participation with health shocks. In: Demazeau Y et al (eds) Advances in practical applications of scalable multi-agent systems. The PAAMS collection. Springer, Cham, pp 157–168

    Google Scholar 

  20. 20.

    OECD (2013) How’s life? 2013: measuring well-being. OECD Publishing, Paris

  21. 21.

    Pelgrin F, St-Amour P (2016) Life cycle responses to health insurance status. J Health Econ 49:76–96

    Article  Google Scholar 

  22. 22.

    Rabin M (1998) Psychology and economics. J Econ Lit 36:11–46

    Google Scholar 

  23. 23.

    Reyna VF, Brainerd CJ (2007) The importance of mathematics in health and human judgment: numeracy, risk communication, and medical decision making. Learn Individ Differ 17:147–159

    Article  Google Scholar 

  24. 24.

    Rust J, Phelan C (1997) How social security and medicare affect retirement behavior in a world of incomplete markets. Econometrica 65:781–831

    Article  MATH  Google Scholar 

  25. 25.

    Wilensky U (1999) NetLogo: center for connected learning and computer-based modeling. Northwestern University, Evanston. http://ccl.northwestern.edu/netlogo/

Download references

Author information

Affiliations

Authors

Corresponding author

Correspondence to Alessandro Moro.

Additional information

The views expressed in the paper are those of the authors and do not involve the responsibility of the Bank of Italy.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Moro, A., Pellizzari, P. A computational model of labor market participation with health shocks and bounded rationality. Knowl Inf Syst 54, 617–631 (2018). https://doi.org/10.1007/s10115-017-1096-3

Download citation

Keywords

  • Labor market participation
  • Health shocks
  • Bounded rationality
  • Agent-based modeling