Three essays on pay-as-you-go pensions

Thumbnail Image
Liu, Qing
Major Professor
Bhattacharya, Joydeep
Cordoba, Juan Carlos
Li, Jian
Singh, Rajesh
Zhang, Wendong
Committee Member
Journal Title
Journal ISSN
Volume Title
Research Projects
Organizational Units
Journal Issue
Is Version Of
As a system, pay-as-you-go (PAYG) pension has been around for a long time. And many aging countries experience the reform of their PAYG pension system. My dissertation tries to rationalize the existence of PAYG pensions and examine the welfare effects of pension reform uncertainty in China. The classical Aaron-Samuelson result states that there is no theoretical rationale for PAYG pensions in a representative stationary overlapping generations (OLG) economy if the return of pension is dominated by private savings. Later work has shown that it may be possible for pensions to exist: when agents have “myopic” or “time inconsistent” preferences, the government is paternalistic, and all retirement saving is held by the pension fund because of a binding borrowing constraint. My first paper generalizes the Aaron-Samuelson discussion to the reference-dependent utility setup of Kőszegi and Rabin (2009). It demonstrates that in this case it is possible to offer a non-paternalistic welfare rationale for return-dominated PAYG pensions and not all retirement saving has to be held by the pension scheme. The intergenerational redistributive role of pay-as-you-go (PAYG) pensions is often used to justify their existence. My second paper asks, would this role survive if the government could legislate intragenerational transfers as well? The answer is no in a simple, dynamically efficient, life-cycle economy with standard preferences, ex-ante heterogeneity, and no uncertainty. Sup-pose pensions also take on the paternalistic role of ensuring enough retirement consumption for myopic agents and governments prevent borrowing against them. Then, both intra- and inter-generational redistribution may be justified if agents are sufficiently myopic, and if the borrowing constraint binds. Notably, even then, it is the paternalism that justifies pensions not their redistribution function. The upshot is that heavy-lifting in the task of redistribution is better done by intra not intergenerational transfers. In the years ahead, budgetary pressures are projected to strain the Chinese pension sys-tem. Such stresses will likely precipitate further reform, but the scale and timing of such re-form remain uncertain. My third paper studies the welfare effects of future pension-reform risk in a multi-cohort overlapping-generations economy characterized by an intergenerational human capital externality and endogenous, altruistically-motivated, within-family intergenerational transfers. In the model economy, parents optimally decide how much to invest in their children’s education, and the latter decide how much financial support to offer their retired parents. The market, left to its own, under-provides education due to the externality. Faced with policy-reform uncertainty, parents rely on their children’s altruism to self-protect by investing heavily in their education, rationally expecting the children’s reform-contingent transfers. For example, children transfer more (less) when the reform reduces (raises) the pension benefit. Overall, the within-family contingent transfers significantly remove pension risk and the uncertainty-driven investment internalizes the human capital externality. Such effects may generate additional downstream welfare gains. The analytical framework is calibrated to the Chinese economy, wherein a stylized, uncertain policy reform is studied. Our numerical analysis shows that the welfare gains from internalizing the human capital externality dominate the welfare losses from consumption fluctuations by the policy uncertainty; the earlier generations not directly relevant to the reform uncertainty may, however, be hurt. The big takeaway is that limited doses of policy uncertainty may sometimes have unintended but desirable consequences.