"Model Risk Management for Pension Funds Investment Practitioners"
This project examines model risk management for pension funds, focusing on the use of Economic Scenario Generators (ESGs) in investment-return projections and asset-liability modeling. Pension plans rely heavily on models to guide asset allocation and funding decisions, yet these models are imperfect and subject to risks stemming from model choice, parameter uncertainty, and misuse. Different ESGs and parameter calibrations can produce materially different outcomes, with high-equity portfolios being especially sensitive to assumptions. The report explores how choices related to ESG model structure and parameterization—such as selecting one model form over another or varying key parameters—can lead to materially different conclusions about pension risks, including the likelihood of asset losses and the need for increased contributions. Using simple ESGs linked to a basic pension model, the analysis demonstrates how different modeling assumptions influence projected outcomes and, in turn, decisions such as asset allocation. The report provides practitioners with guidance on what questions to ask about their ESGs and pension models to better recognize and manage model risk in decision-making.
"Universal Public Pension Actuarial Modeling Tool: Development and Initial Modeling Phases"
The Universal Public Pension Actuarial Modeling Tool, developed by Rockefeller College in partnership with the Reason Foundation, is a simulation model designed to forecast benefits, liabilities, and funding for public pension plans across the United States. By allowing users to incorporate their own plan data, the model enables testing of alternative plan designs, funding and amortization strategies, investment-return scenarios, and actuarial assumptions. This helps policymakers, executive officials, and plan members evaluate how proposed reforms would affect pension fund solvency, government contribution requirements, and retirement income security. The tool’s ultimate goal is to provide transparency and data-driven analysis in pension policy, equipping stakeholders to assess reforms and strengthen the long-term financial health and resiliency of public pension systems.
"Examining the Impact of Legacy Costs on New York State Rural Schools"
The project, funded by the Spencer Foundation, investigates how long-term financial obligations—such as pensions, retiree health insurance, and capital maintenance and debt service—affect the budgets and educational capacity of rural school districts facing enrollment declines. Led by researchers at the Rockefeller College, the study uses detailed state-level financial and demographic data to measure the scale, trends, and crowding-out effects of these legacy costs compared to non-rural districts. By analyzing how rising fixed obligations interact with shrinking revenues and student populations, the research aims to inform policymakers and school administrators about the fiscal pressures confronting rural schools and to provide insights into strategies for maintaining educational quality despite mounting financial stress.
Recently Completed Projects
Analyzing Risks and Issues Relating to the New York City Teachers’ Retirement System
The State and Local Government Finance (SLGF) team recently has embarked on a new project, Analyzing Risks and Issues Relating to the New York City Teachers’ Retirement System. This project will analyze costs and risks of the New York City Teachers’ Retirement System (TRS) under current benefit, funding, and investment policies and under potential alternative policies. It will place this analysis in the context of the New York City budget and in relation to other New York City public pension funds. This broad view is important because when TRS suffers downside risks, or benefits from upside risks, other City pension funds are likely to face similar risks, and so the aggregate implications for the City budget are likely to be greater than those for TRS alone.
In addition to the analytic work, the project entails considerable outreach to and engagement with stakeholders in the New York City Teachers’ Retirement System, including stakeholders in New York City more broadly.
The project is being led by New York University’s Marron Institute of Urban Management, with Rockefeller College as a sub-awardee to conduct pension fund modeling. SLGF co-director Don Boyd is leading the analytic work under a direct arrangement with the Marron Institute, and co-director Gang Chen is the Rockefeller College principal investigator on the project. The work is supported financially by the Equable Institute, a new 501(c)(3) foundation that is focusing its early efforts on retirement security issues. Equable is a nonpartisan non-profit organization that seeks to educate and inform employees, retirees, labor, management, and taxpayers about how to create retirement plan sustainability and accountability without sacrificing future income security.
Read the full report.
Analyzing the Interplay between Public Pension Finances and Governmental Finances: Lessons from Linking an Economic Model to a Pension Fund Model
Public pension funds invest in stock, bonds, and other assets in an effort to keep the costs to governments of funding pensions low. Researchers have examined the investment-related risks to public pension funds, to the governments that contribute to them, and to stakeholders in pension funds and governments, using stochastic simulation models of pension fund finances.
These models generally use simple investment return assumptions, such as that returns are drawn from a normal distribution, are independent from year to year, and are not correlated with the governmental tax revenue needed to pay pension contributions. Because investment returns and tax revenue may both be correlated with underlying economic conditions, this could understate the risks that future increased contributions will be difficult to afford: a poor economy may dampen investment returns and cause tax revenue to fall short, compounding fiscal pressures on governments if the increases come when tax revenues are low.
This paper addresses this issue by using a regime-switching macroeconomic model of recession and growth to drive (1) a model of pension fund returns driven by stock and bond returns, and (2) models of state tax revenue driven by real GDP and stock returns, which influence the capital gains component of income taxes. It is parameterized using retrospective and prospective estimates. It shows that the correlated risk is real and significant, meaning that political risks related to pensions are greater than often understood. The models show that these risks are greatest for governments that rely heavily on income taxes.
Read the full report.
