The retirement decisions of mature age Australians

Abstract

In Australia, as in most OECD countries, an aging population has given rise to concerns about whether, in the future, there will be enough people in the workforce to support the growing proportion of the population who have retired. A key policy issue for the Australian Government is how to increase mature age labour force participation. Understanding how retirement decisions are made is crucial to achieving this aim. The research in this thesis provides insights into the main factors associated with decisions about when, and how, mature age Australians move from work to retirement.

In the first chapter, Australian and international research on retirement decisions is reviewed. The chapter highlights the strengths and limitations of economic models used to analyse these decisions. Also described is the main data source used for the thesis, the Household, Income and Labour Dynamics Australia (HILDA) Survey. Chapter 2 describes the current retirement system in Australia, and provides a summary of the recent reform initiatives designed to increase mature age labour force participation.

In Chapter 3, binary response models are used to evaluate the impact of the financial incentives available to mature age Australians who are deciding whether to retire or continue in paid employment. The results suggest that for many men, and also for single women, the Australian retirement system does create financial incentives to retire early, and that many actually respond to these incentives. Building on these models, simulations are undertaken to assess the likely effects of policy changes recently introduced or announced by the Australian Government. These simulations indicate that increasing Age Pension eligibility age by two years will provide a relatively small incentive to remain in employment. However, the ‘Simpler Super’ reforms introduced in 2007 may create an incentive to retire early.

Chapter 4 highlights the importance of taking into account the family decision-making context when modeling the retirement decisions of partnered men and women. Two complementary estimation approaches are used to model the retirement decisions of couples. First, a single risk duration model provides insights into the influences of a spouse’s characteristics on the retirement decision of his or her partner. In the second approach, a competing-risks framework is used to examine the joint retirement behaviour of couples exiting employment. The results provide evidence of coordination of retirement among mature age couples, largely due to complementarities in leisure and, for women, because of caring responsibilities.

In Chapter 5 the retirement transitions of mature age men and women are modeled using standard and dynamic multinomial logit models. The results confirm that age, health and human capital are important determinants of retirement behaviour and provide further evidence of coordinated retirement by mature age couples. Accounting for unobserved heterogeneity confirms the existence of both true state dependence and spurious state dependence in the labour force decisions of mature age men and women. Therefore, there are likely to be considerable benefits from policies that discourage mature age individuals from leaving the labour force.