A Quantitative Overlapping Generations Model and The Norwegian Pension Reform

The thesis is a part of the evaluation of the Norwegian pension reform from 2011. The two first chapters evaluates the joint retirement in couple. Here, both quantitative methods in empirical studies as well as macroeconomic modelling is used. First, the thesis evaluates whether Norwegian couples are affecting their spouses' timing of retirement. Here, it is exploited that the reform heavily incentivized continued labor force participation combined with early take-up of AFP benefits in the private sector. This allows studying whether the spouses of those individuals who are affected directly by the reform also postpone retirement. The thesis shows that women are particularly affected by their husbands postponing retirement, while it is statistically unclear if men are affected by their wives. Furthermore, the thesis has developed a macroeconomic model of households living in couples. The couples may experience several of the important developments over the lifecycle, such as divorce, good or bad wage shocks and death of the spouse. The individuals choose how much to consume and save over the lifecycle, and they choose when to retire from the labor market. The individuals may work in either the private or the public sector, and have wage profiles corresponding to real lifecycle profiles from those sectors in Norway. This gives a precise image of the Norwegian economy, with the possibility to evaluate counterfactual reforms and predictions for long-run implications of the actual reform. From both a researcher as well as a policy maker perspective, it is also highly interesting to find the fraction of the total labor market response from the reform that stems from complementarity in leisure between the spouses. The thesis shows that about two thirds of the total labor market response for women stems from spillovers from their husbands, while about one fifth of the labor market response of men stems from spillovers from their wives. The final contribution investigates the effect of losing a safe income stream on the take-up of other public benefits. Here, we use the strict rules for eligibility in the private sector AFP scheme. We use data on bankruptcies of firms. If a firm goes bankrupt just too soon, so that an individual does not have sufficient accrual time, she loses her entire early retirement benefit. On the other hand, if the firm goes bankrupt just late enough, she is still eligible for the benefits. Since the bankruptcy is exogenous to the workers, we can exploit this to study otherwise equal workers where one person loses the benefits while the other retains it. The paper shows that those who lost their AFP benefits to a large extent substitute their lost pension benefits with other public benefits, and most of all disability insurance benefits. We estimate that about 80 percent of the lost AFP benefits are substituted with other public benefits, and among those benefits, about 75 percent of it is disability insurance. The remaining part of the lost benefits are, even if it is statistically uncertain, fully replaced by increased labor market income from a new job.

Jeg viser hvordan endringer i pensjonssystemet og trygdesystemene kan gi ringvirkninger ved at personer påvirker hverandre. Jeg studerer par som lever sammen, fordi dette er godt identifiserbart i data og gir de sterkeste effektene. Men det kan også godt tenkes at slike ringvirkninger finnes mellom andre relasjoner som kollegaer, søsken, venner, foreldre og barn. Dette er bidrar til en mer kompleks forståelse av økonomien og hvordan individuelle insentiver kan spille over på andre. Individer kan veksle mellom stønader når en ordning blir vanskeligere å kvalifisere for eller blir mindre gunstig av andre årsaker. Dette kan være svært kostbart, slik som for de ordningene som er evaluert her: når personer som mister privat sektor AFP som er delfinansiert av privat sektor, og heller går over på uføretrygd som i sin helhet er finansiert av det offentlige, vil dette være kostbart for staten og relativt gunstig for det private, uten hensyn til de individuelle kostnadene personen har.

This thesis will study the implications of dual-earner households in retirement ages. It will provide insight into how leisure complementarity affects labor market decisions in old-ages. This is particularly important in terms of aggregation of labor market responses to policy changes and evaluation of the importance of incentives. Another important topic is how long-run elasticities are affected of asymmetric leisure complementarity. This can provide insight into how policy changes affect the Norwegian labor market in the long run when we explicitly consider leisure complementarity as an important feature of an OLG model. This allows us to avoid the systematic underestimation of long-run elasticities that plague most of the current literature, which often only consider male bachelor households. We therefore offer a closer-to-reality description of the behavior of individuals reaching retirement ages, as a large fraction of these are dual-income households. A final feature that will be studied is the effect of health shocks to individuals in dual-earner households reaching retirement ages. This is of particular importance since a bad health shock can leave the household with a tighter budget constraint if no measure is taken to avoid loss of income. A tighter budget constraint may lead the other spouse to increase labor market intensity. However, when one gets sick, it is also likely that the partner has a motive to nurse or spend valuable time together with their spouse, which will lower the labor market intensity or even force the spouse to exit the labor market entirely. Hence, the outcome of a sick spouse on the labor market intensity of their partner is inherently ambiguous. Using detailed register data with information on a weekly basis of the health status of individuals, this thesis will attempt to disentangle these effects and evaluate the importance of health among old-age workers.