Latest / The Fiscal Policy Podcast with Fexingo: Government Budgets, Stimulus, and Public Spending / How Government Pension Assumptions Hide Trillion-Dollar Gaps
Transcript
- Lucas: Let's start with a number that doesn't get nearly enough airtime: 7 percent. Luna: The assumed rate of return on public pension funds? Lucas: Exactly. Most state and local pension plans in the US assume their investments will earn roughly 7 percent a year. The California Public Employees' Retirement System, CalPERS, uses 6.8. The Illinois Municipal Retirement Fund, IMRF, uses 7. And that number, more than almost any revenue forecast or spending line, determines whether a pension is labeled 'funded' or 'underfunded'. Luna: Because if you assume higher returns, you need to set aside less money today to pay future benefits. Lucas: Right. It's the central lever. Move the assumption up by one percentage point and a plan that looked seventy percent funded suddenly jumps to ninety percent. Move it down and the deficit balloons. So the question is: what should that number be? Luna: And who decides? Lucas: The plan's board of trustees, often with input from actuaries. But there's a structural tension. Lower assumed returns mean higher required contributions from the government employer today — which means less money for schools, roads, or tax cuts. So there's a built-in incentive to keep the assumption high. Luna: Which is why most plans cluster right around 7 percent, even though the ten-year Treasury yield is around four and a quarter. Lucas: And that gap — roughly three percentage points between the risk-free rate and the assumed return — is the source of a trillion-dollar debate. Let's walk through the actual math using IMRF as our case study. Luna: IMRF covers about 300,000 current and former employees of local governments in Illinois — police officers, school janitors, town clerks. Lucas: As of their latest actuarial valuation, they reported about $52 billion in total liabilities and $42 billion in assets. That gives them a funded ratio of about 80 percent. But that liability figure is calculated assuming a 7 percent discount rate. If you revalue those same promised benefits using a rate closer to the Treasury yield — say 4 percent — the liability jumps to something like $70 billion, and the funded ratio drops to 60 percent. Luna: So the same promises, the same workers, the same checks being written in the future — but the reported gap nearly doubles depending on which number you pick. Lucas: Exactly. And this isn't just an accounting curiosity. The discount rate determines how much money the employer — your town, your county — has to wire to the pension fund every year. At 7 percent, a municipality might contribute 12 percent of payroll. At 4 percent, that could be 25 percent or more. Luna: Which is why the assumption becomes a political decision, not just a financial one. Lucas: The Governmental Accounting Standards Board, GASB, sets the rules for public pension accounting. And since 2012, they've required plans to discount liabilities at the expected rate of return on assets — essentially letting the plan choose its own number. Critics, including many economists, argue that pension promises are liabilities, not investments, and should be discounted at a risk-free rate, the way corporate pensions are. Luna: But if you forced every public plan to use the Treasury rate overnight, you'd instantly add a trillion dollars or more to reported state and local debt. Lucas: Which would trigger rating downgrades, higher borrowing costs, and potentially a fiscal crisis for some states. So GASB's approach is a kind of pragmatic fiction — it keeps the numbers manageable, but it also lets the real problem fester. Luna: Let's zoom in on IMRF's portfolio for a second. What are they actually investing in to try to hit that 7 percent? Lucas: A mix of about 30 percent domestic equities, 20 percent international equities, 25 percent fixed income, and the rest in real estate, private equity, and alternatives. They're taking significant risk to get that return. And risk means there's a real chance they underperform. Luna: And when they do, the assumption doesn't get lowered right away. Plans use smoothing methods — averaging returns over five years — so a bad year doesn't immediately trigger a higher contribution requirement. Lucas: Which is exactly what happened after 2008, and again after 2022. The losses get spread out, the assumption stays at 7 percent, and the unfunded liability gets kicked down the road. Luna: Is there any plan that has voluntarily lowered its discount rate to something more realistic? Lucas: A few. The State of Rhode Island's pension system dropped from 8.25 percent to 7 percent over several years, but that was part of a broader reform. Most plans are very reluctant. Once you lower the rate, you're asking politicians to write bigger checks — and that's never popular. Luna: So the 7 percent assumption is less a forecast and more a negotiated compromise between fiscal reality and political convenience. Lucas: If today actually helped you see that number more clearly — and I hope it did — the way these episodes stay ad-free and independent is listener support. If that's something you value, you can find us at buy me a coffee dot com slash fexingo. Luna: It's a small thing that makes a big difference for us. Lucas: Now, back to that gap between assumptions and reality. One economist who's been particularly vocal about this is Alicia Munnell at the Boston College Center for Retirement Research. She's argued that the 7 percent assumption is a relic of a higher-return era. In the 1990s, plans earned 10 percent or more on average. But since 2000, the average return for public funds has been closer to 6.5 percent. Luna: So even the plans that hit their assumed number are falling short on a year-by-year basis. Lucas: Right. And that shortfall compounds. If you assume 7 percent but earn 6.5 percent over twenty years, you end up with about 10 percent less money than you planned for. That gap has to be made up by higher contributions from taxpayers. Luna: Which is another way of saying that the pension assumption is effectively a hidden tax increase waiting to happen. Lucas: That's a framing I hadn't heard before, but it's exactly right. A too-high assumption defers the true cost of today's benefit promises to future taxpayers. And those future taxpayers don't get a vote on the assumption. Luna: So what would happen if GASB changed the rules tomorrow and required a risk-free discount rate? Lucas: It would be chaotic. Many plans would suddenly appear critically underfunded. Bond markets would reprice municipal debt. Some cities might face a liquidity crunch. But over time, it would force honest budgeting. You'd either have to raise contributions or cut benefits. And that's the debate we're not having because the 7 percent assumption lets us pretend. Luna: Are there any proposals to gradually transition to a lower assumption? Lucas: Some have suggested a glide path — lowering the assumption by 25 basis points per year until it reaches something like 5 percent. That would give plans and governments time to adjust. But politically, even a quarter-point cut is a fight. CalPERS tried to lower its rate from 7.5 to 7 percent a few years ago and it took a year of negotiations. Luna: And the irony is that the higher the assumption, the more risk the fund has to take to hit it, which makes the outcome even more uncertain. Lucas: Exactly. If you assume 7 percent, you need a heavy equity allocation. That means the fund is more exposed to market downturns. When the market drops 30 percent, as it did in 2022, the pension fund loses billions. And the assumption doesn't budge. So the risk is asymmetric — the fund takes the downside, but the assumption stays high. Luna: It's like setting your GPS for a destination 30 percent farther away than you can actually drive, then flooring the accelerator to make up time, and being surprised when you run out of gas. Lucas: That's the analogy I'm stealing for my next conversation. Let's wrap with one more concrete number. The total unfunded pension liability across all US states and localities, if discounted at a risk-free rate, is estimated at roughly $4 trillion. That's about 15 percent of GDP. If discounted at the current assumed rates, it's about $1.5 trillion. The difference is $2.5 trillion — hidden in plain sight, buried in a single assumption. Luna: So the next time you hear that a pension fund is 80 percent funded, ask a simple question: 'At what discount rate?' Lucas: That single question cuts through more fiscal fog than almost any other. And it's a reminder that in public finance, the numbers we see are never just numbers — they're choices.