by Josh Rauh and Robert Novy-Marx
If you are a homeowner, you have probably estimated your net worth. You know the balance left on your mortgage, and you have a sense of the market value of your home. Perhaps you also have some financial assets. If you are very cautious, most of these are in cash or bonds. If you prefer more risk, you own stocks. But whether your savings is in cash, bonds or stocks does not affect your net worth today, or the reported balance on your debts.
Unless, that is, you are a public pension system following Government Accounting Standards Board (GASB) rules. GASB has recently held hearings on proposed changes to their public pension accounting standards. Unfortunately, the changes under discussion maintain the flawed framework. State and local governments will still be able to hide trillions of dollars of borrowing in the form of improperly valued pension promises.
Under current standards, state and local governments value pension liabilities using the expected return on pension fund assets. Arguing that diversified investment portfolios have an 8% expected return, governments typically use a discount rate of around 8%. Under these procedures the 50 states report unfunded liabilities of around $1 trillion.
But this practice misrepresents the cost of the government promises. The value of a financial obligation has nothing to do with the allocation of the debtor’s assets. If a state wanted to pay an investor to take over the pension liability, the amount the investor would accept would not depend on the state’s asset allocation. If a state tells its employees that their pensions are secure — not risky like the stock market — then it should use the yields on safe government securities such as Treasury bonds to discount the benefits.
Do the GASB rules make sense? Let’s consider an underwater homeowner’s financial situation under common-sense financial logic, and then under GASB rules.
Suppose you are substantially underwater on your mortgage. For example, imagine you own a house worth $200,000 on which you have a $300,000 mortgage, and that you also have $40,000 in a money market account. If you have no other assets, then by any reasonable calculation your liabilities exceed your assets, and your net worth is negative $60,000. If you decide to sell the money market funds and buy stock, nothing has changed. You are still $60,000 in the red. Reducing your cash holdings and increasing your stock market exposure has no impact on your wealth today.
But under the logic of GASB, moving your savings of cash into stocks, while having no effect on the value of your assets, magically reduces the reported amount that you owe on your mortgage. Since the financial assets you own now have a higher expected return, you can mark down your mortgage on your personal balance sheet. GASB effectively says that by moving some savings from your left pocket to your right pocket you instantly make yourself richer. If you take on enough risk, you can claim positive net worth — and even take out a second mortgage on the same home.
The GASB magic is even more impressive for a recent college graduate with $100,000 in student loans, a mere $100 in money market funds, and no other assets. By taking that $100 out of the bank and buying stocks, the graduate would be able to write down the value of his debts by as much as one-third, even though his savings is trivial in comparison to his debts. Under this logic banks hit by the 2008 financial crisis could even “recapitalize” by taking on more risk in their loan portfolios and writing down debt accordingly.
The proposed GASB changes maintain this discounting to the extent that plan assets are projected to be sufficient to make benefit payments. The remainder would be discounted at a high-grade municipal rate. However, plans are given wide discretion to include projected future contributions from all sources in determining when that depletion point comes. Without actually doing anything about it, states can claim that they have planned sufficient contribution increases in future budget years to cover benefits.
As such, the proposed GASB rule changes will have little effect on stated liabilities. In fact, Alan Milligan, the chief actuary of CalPERS, has even written that “the liability that would be reported under the proposed rules would be the same as CalPERS currently reports.” It seems that even the graduate with only his $100 equity investment could claim that his future earnings will allow him to pay off his loans.
If the GASB logic appeals to you then you can sleep easy, trusting that state pension plans are underfunded by “only” the $1 trillion that they themselves report. If you doubt, however, that moving funds from cash into stocks reduces the amount you owe on your mortgage, then states have underreported their pension liabilities. The true unfunded liability using Treasury rates is around $3 trillion for state governments, plus an additional $0.6 trillion in local governments. Whether or not your mortgage keeps you up at night, your future tax bills and the ability of state and local governments to provide essential services certainly should.
[...] Rauh and Robert Novy-Marx have an excellent blog explaining the wrong-headed logic behind how state and local pensions are valued. Under current [...]
Let us follow your examples to their logical conclusion.
Over the last 20 year period including the recent financial disaster of the last couple of years Calpers earned 7.79% against their estimate of 7.75%. I don’t know what T-bills earned during the same period, but lets say 4%.
Compounding both of these numbers over 20 years yields a new value of principle after 20 years of 4.25 vs 2.19 times the original principle more than double. I know these are just ballpark figures, but you get the idea. Actual EARNINGS on 7.75% vs the risk free rate of 4% would be 273% more.
The result of using the 4% risk free rate to make future predictions on the pension funds value would be vastly under estimate its current value.
Net result, Calpers would have demanded much larger additional State money to keep it solvent.
The fund at the end of the twenty years would be overflowing with money and State government and the media would be demanding an investigation.
So you can’t win.
These are highly misleading analogies.
Mortgages and student loans are debt – a fixed dollar amount of money is given by a lender to a borrower. Pensions are a series of future expected payments – there is no intrinsic current value for a pension liability, it is simply a number created by applying a discount rate to a future series of payments. A lower discount rate creates a larger present value. So what?
The issue with pensions is not the size of the liability but the adequacy of current assets and expected funding to pay future benefits. Future benefits are a function of the future wage growth of public employees. This has begun to decrease as shown by the reason decision in arbitration that held salary increases to 2% annually for five years for Chicago police officers. If state and local governments are able to control future salary increases, pensions will become more affordable, and the reported value of the liability will also decrease as 5% to 8% future wage increases are often assumed in the calculation of the liability.
Pension costs will also be driven by the expectation of current employees on the security of their pensions. State employees are likely to postpone retirement if they believe that their pensions will be pro-rated because of funding shortfalls – which in itself will lower pension costs and improve the solvency of the fund.
Pensions are a complex topic. Making misleading analogies does not help the public debate.
I strongly disagree with the premise that “there is no intrinsic current value for a pension liability.”
States have promised benefit payments to workers. In most states those payments have special legal protections, and in some states such as Illinois the state constitutions are clear that the promise may not be diminished or impaired. That’s debt in my book, and would be debt in any corporate books as well.
I would add that it is much more likely that state employees will rush for the exits than delay their retirement. That is what is happening in New Jersey (http://bit.ly/c2woIc). The employee calculation seems to be that those who are still working are much more likely to be affected by pension cuts or changes than those who are already retired.
I would like to compliment Charles Sainte Claire and Daniel Kaplan on the two previous comments. Both present a more realistic analysis of the pension promise liability than the authors of this article. Such analysis of pension liabilities as presented by Mr. Rauh and Novy-Marx rush to create a crisis “the sky is falling” reaction to current reduced funding levels. There are definately long-term problems with some public funds that require thoughtful, long-term solutions. First, employers need to fund annual required employer contributions. Second, employers should manage pension cost with the “normal annual cost” of providing the benefit they have promised in mind and not build budgets based on reduced contributions resulting from a few good years of investment returns.
“Josh Rauh
I strongly disagree with the premise that “there is no intrinsic current value for a pension liability.”
I am not sure what you mean. A current bill is static. Your possible means of paying it in the future don’t apply to any analysis of the actual debt which doesn’t change and since it is due at a specific date.
Actuarial funds are not set up that way. The future debt depends on unknowns, for instance, how long will people live on average thirty years from now? How much will they earn? How will the proportion of active employees change compared to retired persons? What will the economy perform over, say, 30 years?
Agreeing with Mr. Daniel Kaplan, discount rates, future employee numbers and other factors which must necessarily impact pensions funds are best guesses. Calpers best guesses were quite good over the long haul, not just twenty years but far more. It is interesting to take a snapshot in time (today)and extrapolate it in both directions for decades.
Your past extrapolation is wrong. Your future extrapolation, then, is certainly questionable.
When a household or business accrues liabilities, they have to give an accurate representation of their debts to potential future creditors. I don’t agree with the idea that “actuarial funds” are somehow exempt from this responsibility. Households and businesses also have future income sources, and that doesn’t exempt them from having to account for the financial value of their debts.
If you take a look at our academic work, you’ll see that our focus is on benefits that have already been promised based on today’s levels of service and salary. Thus, the variables that you invoke about future longevity, earnings, the economy, etc. are not relevant for those calculations.
Thanks for your remarks.
[...] to an even larger downside since the pensions have to be maintained regardless. (For more, see “Wipe Away Your Debts with State Government Accounting” and“Return Assumptions: It’s All Just Borrowing in [...]