University of Rochester

In Review

Robert Novy-Marx has become something of a canary in the coal mine when it comes to understanding the potential for a budgetary explosion caused by state and local public employee pension programs. Novy-Marx, assistant professor of finance at the Simon School, and his colleague, Joshua Rauh of Northwestern, have been widely called upon to help policymakers understand the magnitude of pension obligations faced by states from New Jersey to California. In January, they were awarded the American Finance Association’s Smith-Breeden Prize for their 2010 study, “Public Pension Promises: How Big Are They and What Are They Worth?”

How deep in the hole are state and local pension funds?

Joshua and I have analyzed data from about 190 individual plans—116 state plans and another 70 local plans. And they’re not standardized in any way, but we’ve done a lot of work to try to back out what the liabilities really are year by year, and calculate what the economic magnitude of the liabilities are. Plan managers claim the plans are maybe $1 trillion underfunded. My view is they’re more like $3 trillion underfunded. The states themselves are maybe $2 trillion underfunded, and sub-state government entities are another half a trillion more underfunded than their own claims.

Is that because state and local governments are not funding their pension plans appropriately?

We’re primarily interested in valuing the liabilities. Whether these things should be fully funded is a separate issue. Some other pension plans in other countries are operated as pay-as-you-go plans, which means that they’re not funded, and that’s not necessarily a bad thing. What I think is a bad thing is having an extra $2 trillion in off-balance sheet unrecognized debt, which the states have.

What does that represent to the average taxpayer?

We estimate that without policy changes, to achieve full funding for public employee pension systems over the next 30 years, each U.S. household would see a tax increase of about $1,398 each year, above and beyond what they would pay because of expected economic growth. In many states, the increases would be higher.

How has it gotten to this point?

The problem with state and local government pension plans is that they’re using an accounting methodology that violates basic tenets of common sense. In particular, one absolutely necessary aspect of an accounting methodology is that if you have two plans that have the exact same liabilities, and Plan A has some assets and Plan B has those assets and some more assets, you should view Plan B as better funded. And it’s better funded by the amount of extra assets it holds. Under the Government Accounting Standards Board (GASB) methodology, it could easily be that Plan B, which has the exact same assets plus more assets and the exact same liability, could have a worse funding status under the government plan.

How does this square with other budgeting and accounting practices in government?

One thing that state governments do is they use these plans very much to circumvent their own balanced budget requirements. Almost every state—I think all the states with the exception of Vermont—have their own balanced budget requirements; they’re not allowed, like the federal government, to run a deficit. They have to pay for what they spend every year. If you’re going to be compensating your workers, you should be paying them for it. And for deferred compensation, you should be putting that money away. By using GASB, which undervalues these liabilities, they’re able to put away significantly less money than they would if they recognized the full extent, the full magnitude of these liabilities they have. So there’s been a way in which they’ve circumvented their own rules by borrowing and not recognizing that borrowing.

But isn’t a liability always a liability?

State and local governments have their obligations, and they’re told how to fund these obligations—by putting money away—but the GASB methodology makes the obligations look very cheap. And not only does it make them look cheap, but it encourages state and local governments to take on lots of investment risk, because the more investment risk they take on with the assets they hold, the smaller their liabilities get under GASB. They don’t really get smaller—they are what they are—but the recognized magnitude of liabilities under GASB gets smaller the more risk they take on.

What happens when the obligations— “the recognized magnitude of liabilities”—come due?

Government is a pass-through entity: if the governments come up short and have to meet their liabilities to their current workers and retired workers, it just means higher taxes or fewer services. Those are the only two possible ways, apart from defaulting on their municipal bonds—provided they don’t default on their obligations to their workers. Another way to look at this is, there are a few trillion dollars of unrecognized liabilities on the table. And it’s a distribution issue at that point. The $3 trillion deficit people don’t want to deal with has to get eaten by someone—and it’s about apportioning it between different groups. So the possible groups are to the taxpayers, either through higher taxes or less services, the beneficiaries through lower benefits, and holders of other public debt through default.

What do you hope happens with your research?

What I want to do, my primary goal, is to get everyone to recognize what the real costs of these plans are, not so that public workers can be compensated less, but so that we can appropriately value what the cost of a worker is, what the costs of these plans are, so we can plan for them going forward. Otherwise, they’re distortionary. I’m most interested in transparency.