Essays on the Intergenerational Welfare State

Research output: Book/ReportPhD thesis

101 Downloads (Pure)


This thesis comprises three chapters on different topics related to the intergenerational welfare state, which is a system of social policies providing economic security across generations. Although the chapters relate to different topics, public pension schemes constitute the unifying theme of the thesis.

Chapter I
The first chapter, entitled Population Aging, Public Finances, and Alternatives for Retirement Reform, is written in collaboration with Tim Dominik Maurer and motivated by advanced-economy pension systems coming under fiscal pressure due to population ageing. Against this background, we conduct counterfactual policy experiments to compare different retirement reforms that ensure sustainable public finances in the face of increasing longevity and decreasing fertility. As our main contribution, we consider a particular reform that in-creases fully-funded (FF) contributions to reduce pay-as-you-go (PAYG) benefits indirectly through means testing. We evaluate this against three standard reforms: increasing the retirement age, cutting public pension benefits, and increasing taxes to finance growing public pension expenditures. For the policy analysis, we build a sophisticated structural life-cycle model incorporating a pension system consisting of two pillars - a public PAYG scheme and a mandatory FF scheme - interacting through means testing. We estimate all unobserved preference parameters using the Simulated Method of Moments targeting Danish micro data on wealth and labor market participation. Our subsequent policy experiments suggest that increasing contributions to mandatory FF pensions is best for welfare. The positive welfare effects arise from the FF scheme giving access to fair annuities that offer higher returns and better longevity risk insurance than voluntary savings. This positive wealth effect allows agents to enjoy more consumption and leisure. Additionally, the expansion of the FF scheme reduces the relative importance of the return-dominated PAYG scheme. These positive wel-fare effects outweigh two negative effects related to the loss of income insurance through the public pension system and a crowding out of bequest income.

Chapter II
The second chapter, Redistribution in Public Pension Schemes: Evidence from Denmark, also with Tim Dominik Maurer, documents a loss of redistribution through Danish public pensions when accounting for inequality in longevity. While we focus on the Danish case, our results are relevant to other countries with pension systems that promote redistribution. As our main contribution, we use a so-called affluence measure combining income and wealth information to allocate individuals into socioeconomic groups. Using this measure, we first predict significant mortality differences across groups. Next, we assess redistribu-tion via the public pension system by computing and comparing the expected net present values of the implicit public pension contract. To this end, we use detailed Danish register data for the entire population, allowing us to estimate mortality rates accurately and track a large set of covariates over the individual life cycle. Focusing on recently retired cohorts, we use a combination of historical and forecasted income and wealth data as inputs to a microsimulation model that includes detailed tax and pension system components. Our results indicate a significant loss of redistribution in public pensions due to inequality in longevity. Specifically, the present value of the pension contract develops non-monotonically with affluence, with males in the middle affluence groups benefiting more than the least affluent males. Only the top 30% males have lower net present values than the least affluent males. Moreover, females of a specific affluence rank have considerably higher net present values than similarly ranked males. Interestingly, the estimated redistribution loss depends on the allocation mechanism by affluence. When we repeat the experiment using only information on lagged income, we find a smaller redistribution loss – although inequality in longevity still implies that net present values develop non-monotonically. Thus, we argue that the results of other papers that use allocation by income or education and report relatively small redistribution losses would likely be different had they used a more informative allocation mechanism.

Chapter III
The third chapter, Improving Welfare through Public Education, is motivated by the student debt crisis in the United States and examines the economic consequences of imple-menting an optimal education subsidy. Here, I address two key questions: i) What is the welfare-maximizing education policy in the long run? and ii) How can policymakers ensure socially responsible implementation of such a policy in the short run? The findings demonstrate that a well-designed education subsidy reduces student debt, mitigates inequality, and enhances social welfare in the long run. The long-run welfare gains arise through three channels: a human capital externality, redistribution, and borrowing effects. Next, I show how the government can ensure socially responsible implementation by utilizing public pensions as an instrument for intergenerational cost sharing. The analysis utilizes a calibrated partial-equilibrium overlapping-generations model, extending earlier models to incorporate a general number of overlapping generations, heterogeneous agents, probabilistic survival, and realistic demographics. The paper also presents two model extensions incorporating length-of-education choices and idiosyncratic income risk to demonstrate that the main insights hold in generalized settings. Thus, it contributes to the literature on public education by providing nuanced insights at both the micro and macro levels.
Original languageEnglish
Place of PublicationFrederiksberg
PublisherCopenhagen Business School [Phd]
Number of pages123
ISBN (Print)9788775682034
ISBN (Electronic)9788775682041
Publication statusPublished - 2023
SeriesPhD Series

Cite this