Google the words “public pensions” and you might get the impression that the sky is falling. That’s because several recent reports conclude that public retirement systems face significantly larger long-term funding shortfalls than pension system administrators claim. While these reports might suggest to casual readers that drastic and immediate action is needed to close those shortfalls, the issue is more complex than it’s been made out to be. By conflating two distinct issues – how to measure retirement systems’ funding shortfalls and how to actually fund those systems – recent reports contribute to the misperception that public pension systems face an immediate crisis. As part of the CBP’s ongoing look at state worker pensions, we address this misperception.
How is a pension system’s funding shortfall measured? Calculating a pension system’s “funded status” – whether it has sufficient resources to provide the retirement benefits promised to workers – requires making an assumption about the investment earnings the fund will earn over time. Public pensions are typically “pre-funded,” meaning that over the course of workers’ careers, employers’ contributions are pooled with employees’ contributions in a trust fund and invested so that they can accumulate investment earnings by the time workers are ready to retire. The lower the assumed earnings, the greater the amount assumed to be needed to provide future retirement benefits. Several recent reports conclude that public pension systems are severely “underfunded” because the authors assume that pension fund investments will earn a very low “rate of return” over time – that is, that they’ll accumulate relatively low investment earnings. A widely cited report authored by a group of Stanford students, for example, suggested that public pension fund administrators ought to assume that their investments will earn an average annual long-term rate of return of just 4.1 percent – a so-called “riskless” rate of return – the interest rate that could be earned on an extremely low-risk investment at the time of their analysis.
While there is some debate around what constitutes an appropriate riskless rate of return – researchers at Boston College’s Center for Retirement Research (CRR), for example, argue that a rate of 4 percent is probably too low – many experts believe that it’s prudent to use a riskless rate to estimate pension systems’ funded status. CRR researchers argue, for instance, that doing so could help policymakers make more informed decisions about changing retirement benefits. However, these experts do not claim that a riskless rate of return should necessarily be used to determine how much employers must contribute to pension funds. Here’s why:
Pension funds are typically invested in a diverse array of assets that generally far outperform a riskless rate of return. Over the past two decades, California Public Employees’ Retirement System (CalPERS) investments, for example, earned an average annual return of 7.91 percent. In fact, CalPERS administrators actually exceeded their expectations of a 7.75 percent average annual long-term return – significant in light of the fact that this period included three recessions and the worst stock market crash in decades. While future investment returns may be lower than historical returns, assuming that pension funds will earn an average long-term rate of return of 3 to 4 percent “probably represents a ‘worse-than-worst case’ scenario.” Rather than drastically changing how pension systems are currently funded based on such an analysis, pension fund administrators should assess actual investment returns over time and gradually increase employer contributions if earnings fall short of expectations. Employer contributions should be based on pension funds’ actual rate of return, not on the assumed rate of return used to provide a conservative snapshot of pension systems’ funded status.
Public pension systems are invested – in principle – for perpetuity, which means they have decades to make up for any investment losses. Since public pension systems pool contributions on behalf of workers who will retire at different times, they can spread any losses over multiple generations of retirees. Over the long-term, below-average investment earnings in some years will likely be offset by years of higher-than-average earnings. As we blogged last week, when states experience investment losses, they can gradually rebuild their pension trust funds over as much as a 30-year period under standards developed by the Governmental Accounting Standards Board. Thus, there is no need to move abruptly to increase employer contributions to public retirement systems.