Push and pull factors in early retirement

Most public old age pension systems are financed (in part) by payroll and income taxes, and define individual benefits as functions of earnings histories. Taxation of labor earnings is therefore partially reflecting forced savings rather than ordinary taxation. The increase in future pension entitlements resulting from an increase in labor earnings can constitute a substantial part of total marginal taxes on earnings. In the paper "Salience and social security benefits" (published in the Journal of Labor Economics), we analyze effects of salience in benefit systems, by comparing labor supply effects of compensation for work in terms of ordinary net income to the labor supply effects of compensation in terms of future pension entitlements, the latter working through complicated accrual rules. Within the context of the Norwegian pension system, we study responses to a kink in the rewards to work because of earnings testing of pension benefits for earnings exceeding a certain amount, and a jump in the rewards to work associated with earning more than one basic amount, which generated a discrete jump in future pension entitlements. We show that incentives in the form of current net earnings have a substantial labor supply effect, while incentives in the form of future pension entitlements do not have a similar effect. In the paper "Excess early retirement? Evidence from the Norwegian 2011 pension reform", we study behavioral responses to pension incentives in the context of the recently reformed public pension system in Norway. For workers covered by the early retirement scheme AFP (AFP workers), the 2011 reform brought major changes in the net returns to continued work past the early retirement age. These changes are driven by the removal of a strict earnings test on AFP pension benefits and the introduction of actuarially neutral pension adjustments for postponed claiming. For AFP workers, the early retirement age remained constant at age 62. Workers not covered by AFP (nonAFP workers), on the other hand, got access to public pensions from age 62 rather than from age 67, at actuarially neutral terms, without any change in the net returns to continued work past age 62. We document reform effects on earnings, employment, and receipt of disability and unemployment benefits using a difference-indifferences framework. For AFP workers, we find large positive effects on employment and annual earnings, but the compensated labor supply elasticity implied by the difference-in-differences estimates nevertheless suggests that individual retirement decisions are rather insensitive to pension incentives: In light of the magnitude of the incentives changes and in a lifetime labor supply perspective, the employment responses are quite moderate. For nonAFP workers, we find no effects on the extensive margin and only moderate intensive margin responses, suggesting that binding liquidity constraints is not an important driver of the retirement behavior of nonAFP workers. In the current Norwegian public pension system, claiming of benefits and retirement from the labor force are largely decoupled: Individuals can decide to claim old age pensions at any point between ages 62 and 75, and annual benefits are subject to actuarial adjustments based on longevity measures specific to each birth cohort, but not earnings tested. Hence, individuals with short expected longevities have incentives to claim pensions at the earliest possible age, regardless of when they plan to retire. With the paper "Life expectancy and claiming behavior in a flexible pension system" (forthcoming in the Scandinavian Journal of Economics), our main objective is to provide an answer to the following question: How, and to what extent, is early claiming of public pensions associated with short expected longevity? We approach this question in two steps. First, we measure the overall association between claiming at age 62 and expected longevity in the flexible Norwegian pension system. Second, we study the extent to which individuals claim pensions early in part because they act on information about their expected longevity. Having characterized the nature of the association between expected longevity and early claiming, we are also able to study the impact of selection in claiming age on the costs of flexible pension provision. We find that individuals who choose to claim pensions early are to a large extent those who gain in terms of expected lifetime pensions from claiming early. We also demonstrate that some of this relationship is causal; some individuals claim pensions early because they gain from doing so. Despite the strong relationship between early claiming and expected longevity, the additional costs to public budgets arising from this selection are rather modest.

In a process accelerating from around age 50, many workers are permanently withdrawing from the labour market. Considerable fractions of workers enter health dependent benefit programmes and after some years receive disability pensions, which in practice is an absorbing state. This withdrawal process is further advanced when workers reach their early sixties, when a majority of the workers get access to the early retirement scheme AFP; less than 20 percent of the workers remain in the labour market until the notional retirement age of 67. It is not at all clear whether this large scale attrition from the labour market before the notional retirement age is due to voluntary choices or to firms pushing workers into disability pensions or early retirement. Our project speaks directly to the issue of what determines labour force participation and retirement behaviour of workers before or in their early sixties, and to the effects of institutional features and policy on retirement choices, including early retirement schemes and mandatory retirement. With a combination of a comprehensive modelling framework and several self-contained empirical studies we will be able to study a considerable number of outcomes with direct policy relevance, including the effects of mandatory retirement age (over and above the direct effects stemming from the small fraction of workers who work beyond the age of 67 under the current system); the labour supply effects of flexible retirement provision; peer effects in the timing of retirement; distributional effects and insurance features of early retirement provision; the labour supply effects of (non-)salience in pension systems; the relative importance of push and pull factors in early retirement; and the value of early retirement provision to firms.