تجزیه و تحلیل چرخه عمر از خاتمه دادن به بازنشستگی اجباری

In this paper a life-cycle model is constructed to study the macroeconomic effects and welfare implications associated with eliminating mandatory retirement. Our short run analysis reveals that changes in welfare during the transition depend on the dynamic nature of the wage rate adjustment process. We distinguish between transitions in which the wage rate clears the labor market and transitions with a sticky wage and youth unemployment. We also examine political feasibility by measuring the popular support that this type of policy might have under the two labor market scenarios. Finally, we identify the effects that the policy has on welfare in the long run.

In this paper we study the macroeconomic effects and welfare implications associated with eliminating mandatory retirement. The analysis is performed using a large scale life-cycle model of the type developed by Auerbach and Kotlikoff (1987). The model features heterogeneous individuals with a lifetime labor/leisure choice and endogenous retirement decisions. Model parameters are calibrated with data from the Canadian economy and policy experiments are performed by removing mandatory termination of work. The reason for applying the model to study the Canadian experience is that in recent years several of Canada's provinces have abolished the policy of mandatory retirement.1 Our analysis deals explicitly with policy short and long term macroeconomic effects and potential welfare gains and losses. In addition, we are also able to capture age specific outcomes that may arise in the course of the policy implementation process. This paper is related in both methodology and consideration of a change in the statutory retirement age as a policy option to a large body of literature that has studied reforms to social security systems. These reforms have been proposed in the hope of alleviating pressures that arise from aging populations. Auerbach and Kotlikoff (1984) were the first to employ a large life-cycle model to study social security reforms in the U.S. economy. However, many studies followed, simulating the model by using parameters and demographic patterns specific to different countries. For example, extending retirement as part of pension reforms has been proposed and studied in life-cycle models by Hviding and Merette (1998) for a number of OECD countries; De Nardi et al. (1999), and Conesa and Garriga (2003) for the U.S.; Hirte (2002) for Germany; Lassila and Valkonen (2002) for Lithuania; Henin and Weitzenblum (2003) for France; Beetsma et al. (2003) for the Netherlands; Keuschnigg and Keuschnigg (2004) for Austria; and Hongxin and Merette (2005) for China.2 Unemployment effects are examined in only two of these studies. Hirte (2002) models under-employment that persists over the life-cycle due to a constant difference between the cost of labor and its marginal product, resulting to part of the individual time being employed and part of it being unemployed at each stage of the life-cycle. Keuschnigg and Keuschnigg (2004) look at pension reforms in a model that captures long run search unemployment. We depart from the literature mentioned above in several respects. First, we model transitional unemployment for new entrants in the labor market, which in the model economy correspond to young generations. In this way we address one popular concern related to the policy of ending mandatory retirement which is the possibility of creating youth unemployment. Unemployment at early stages of the life-cycle is important for welfare outcomes as it may prevent individuals to build up necessary assets for retirement and induce them to work longer for the rest of their lifetime. Thus we examine transitional periods that follow two types of wage adjustment scenarios: a flexible wage which clears the labor market and a sticky wage which follows a slower adjustment and creates unemployment. We are able to compare welfare gains and losses in these two types of transitional structures, and in the long run equilibrium. Second, we investigate age specific welfare outcomes at the start of each transitional path with clearing and non-clearing transitional labor markets. Any welfare changes of generations alive at this period will affect public support for the new policy. Thus we examine whether banning involuntary retirement is a feasible political equilibrium that is supported by a majority vote. Third, we examine the effects of banning mandatory retirement in Canada. In provinces which previously allowed mandatory retirement, the normal age of retirement was typically at the age of 65. Although six percent of workers continue to work full-time after the normal age of retirement, the current average retirement age in Canada is 62. Thus, it would appear that the policy might have small or no effect on economic outcomes and individual welfare. However, data from labor force surveys in the period 1997 to 2006 suggest that a trend to retire early, particularly prevalent in the 1990s, may be reversing (see Burbidge and Cuff, 2008). It is likely that this trend may continue to reverse due to several reasons: firstly, an aging population due to low fertility rates and combined with the baby boom generation will increase the number of old workers in the workforce (Martel et al., 2007); secondly, improvements in health and longevity allow for the possibility to work longer (Hogan and Lise, 2003); and thirdly, younger people now spend more time studying than in the past, and thus might be shifting working years to later in life (Beaujot, 2004). To preview our main results, we find that when comparing long run equilibrium outcomes with and without mandatory retirement, new entrants in the labor market would actually prefer to be born in an economy with mandatory retirement. Welfare outcomes are also lower for individuals born during the transition to a voluntary retirement economy, and the reduction in welfare varies with the type of the wage rate adjustment process. In particular, individuals who enter the economy in a transition with flexible wages experience a lesser reduction in welfare than individuals born in the long run with voluntary retirement or a transition where wage rates are slow to adjust. Despite the fact that welfare is lower for all agents born after the policy change, for a majority of the current population welfare can be improved by removing the mandatory retirement rule. As a result, our measures of political feasibility concerning voters alive at the time of the policy announcement to end mandatory retirement indicate that the policy is supported by a majority vote. The rest of the paper is organized as follows: the model is outlined in Section 2 and calibrated in Section 3. Policy experiments are performed and discussed in Section 4. Concluding remarks are provided in Section 5.

In this paper we have looked at aggregate and age specific outcomes, as well as welfare considerations of removing mandatory retirement. Unlike similar literature on pension reforms which extend the date of retirement, we analyze this policy independently, focusing our attention on transitional and welfare outcomes. Perhaps our most interesting finding is that although the policy is preferred by a majority of voters alive at the time of the policy implementation, it eventually leads to a decline in individual welfare. The model also reveals that welfare differences and responses of aggregate variables and prices during the transitional phase will depend on the degree of frictions in the labor market. We observe following a short period of wage stickiness more pronounced aggregate changes and slower convergence to a new equilibrium. Agents born in the transition with a slower wage adjustment also experience a higher welfare reduction than agents born in the transition with flexible wages and the equilibrium with voluntary retirement.