The key prediction of the life-cycle models is that consumers make intertemporal plans to smooth their standard of living over the life cycle. However, recent empirical evidence shows that drops in consumption at retirement contradict the consumption-smoothing theory; this phenomenon is known as the “retirement consumption puzzle”(Battistin, Brugiavini, Rettore, and Weber (2009)). Possible reasons for this decrease in consumption include unexpectedly low pensions, increased non-market time, change in preferences, and others.
A majority of the research on the retirement-consumption puzzle focuses on non-durable expenditures, which fails to include expenditures on medical care(Li, Shi, and Wu (2015)). The impact that retirement can have on medical expenses is rarely studied because retirement is an endogenous decision variable. Additionally, it should be noted that concerns about medical expenses could affect an individual’s retirement decision.
Medicare, they found, has a profound impact on retirement behavior. For instance, prior to receiving Medicare at age 65, many individuals receive health insurance only if they continue to work. This incentive to work disappears at age 65, when Medicare provides health insurance to almost everyone. However, retiring before 65 exposes workers who lack retiree health insurance coverage to possible catastrophic medical expenditures. The Social Security system and pensions also provide retirement incentives at age 65. With these factors in mind, it is difficult to determine whether the high job exit rates that are observed at age 65 are due to Medicare, Social Security, pensions or a combination of the three .
Significantly, China has a mandatory retirement age. The exogenous rule forces individuals to exit the labor market immediately upon retirement. Also, health insurance provisions are the same for workers before and after retirement; indeed, there is no switch in health insurance at the extensive margin. This paper, therefore, provides insight into the impact of medical expenditures on relative costs and benefits in China.
The majority of the literature focuses on the effects of changes in health-insurance provisions on health-care utilization at the extensive margin. This paper, however, contributes to the limited literature that focuses on the changes in cost sharing on health insurance at the intensive margin. It should be noted that Baicker and Goldman (2011) offers a review of the difficult task of identifying these effects, due to unobserved characteristics in the presence of self-selection.
For example, Aron-Dine, Einav, and Finkelstein (2013) offer evidence on the effects of cost sharing on health utilization, based on the RAND Health Insurance Experiment (Manning, Newhouse, Duan, Keeler, and Leibowitz (1987)). Chandra, Gruber, and McKnight (2010) also studies a policy change that raised co-payments on the supplemental insurance for retired public employees in California, though the increased cost sharing was restricted to office visits and prescription drugs. Finally, Shigeoka (2014) analyzes a policy in Japan that reduced the cost sharing of patients older than 70 years, and also studies how such cost sharing affects health outcomes and health utilization.
These studies offer some evidence to the effects of cost sharing on health utilization. This paper, however, suggests that a quasi-experiment of change in cost sharing occurs for over the course of an individual’s life cycle. Determining how these retiring individuals respond to such changes in cost sharing will provide information that the literature currently needs.
It is worth noting that females and males not only differ in mandatory retirement ages, but they also exhibit different health-seeking behaviors. Mustard, Kaufert, Kozyrskyj, and Mayer (1998) provides evidence that the gender difference in the use of health-care can be substantial at various stages of one’s life cycle. For example, females tend to spend more when they are pregnant or giving birth, while males consume more during the period right before they die. Hawkes and Buse (2013)also studies gender as an important influence on health behaviors. This paper provides new insight into the gender differences that exist in health utilization, specifically from an economic perspective.