From John Mauldin via Thoughts from the Frontline
The past 10 years have seen a growing number of economists and financial analysts questioning the propriety of the methods used to forecast pension fund liabilities. This is more than an academic exercise, as the numbers you choose to base your models upon make massive differences in the projected outcomes. As we will see, those differences can run into the trillions of dollars and can mean the difference between solvency and bankruptcy of municipalities and states. The implicit assumption in many actuarial forecasts is that states and cities have no constraints on their ability to raise money. If liabilities increase, then you simply raise taxes to meet the liability. However, fiscal reality has begun to rear its head in a few cities around the country and arrived with a vengeance in Detroit this summer. It seems there actually is a limit to how much cities and states can raise.
“Aah,” cities assure themselves, “we are not Detroit.” And it must be admitted that Detroit truly is a basket case. But it may behoove us to remember that Spain and Italy and Portugal and Ireland and Cyprus all said “We are not Greece” prior to arriving at the point where they would lose access to the bond market without central bank assistance.
In response to growing concerns over public pension debt, the Governmental Accounting Standards Board (GASB) and Moody’s have both proposed revisions to government reporting rules to make state and local governments acknowledge the real scope of their pension problems. (While it is possible to ignore Moody’s, based on the fact that it is just one of three private rating agencies, it is impossible to ignore GASB, which is the official source of generally accepted accounting principles (GAAP) used by state and local governments in the United States.
Under the new GASB rules, governments will be required to use more appropriate investment targets than most public pension plans have been using, bringing them more in line with accounting rules for private-sector plans. Pension plans can continue to use current investment targets for the amounts the plans have successfully funded; but for the unfunded amounts, pension plans must use more reasonable investment forecasts, such as the yield on high-grade municipal bonds, currently running between 3 and 4 percent. From my perspective, not requiring reasonable investment forecasts on already funded accounts is still unrealistic, but the new GASB rules are a major step in the right direction, and I applaud GASB for taking a very politically difficult stance.
Moody’s has also proposed new rules to require states to use more appropriate investment targets. Their new rules require pension plans to use investment targets based on the yield of high-grade, long-term corporate bonds, currently just over 4 percent. (Source: http://illinoispolicy.org/uploads/files/Pension_debt_more_than_doubles.pdf)
What difference does a more “realistic” forecast make? According to the survey done by Moody’s, it makes a difference of more than $3 trillion, or more than double the total actual assets of the 255 largest state-funded pension plans. This is illustrated in the chart below.
Current official reporting suggests that states have funded 73% of their pension liabilities. The fair-market-value approach used by Moody’s and GASB suggests that funding is only at 39%. The difference is almost entirely due to the assumptions one uses about the discount rate for future expected returns.
The next two charts provide an illustration. I’m simplifying a bit, but the principles are correct. If you are a pension plan manager, you have to be thinking over very long periods of time. Someone retiring today at age 60 will likely require almost 30 years of pension payments. Someone aged 40 paying into your pension program will likely be getting his or her pension returns 50 years from now. Let’s look at a few scenarios of what might happen to $1 billion over the next 40 years under various assumptions of investment returns.
Many state-funded pension plans today assume an 8% nominal return for the indefinite future. Some are beginning to forecast lower returns, but very few would forecast lower than 7%. Moody’s argues that somewhere in the range of 4% nominal is more realistic. Notice that the difference after 40 years is well over four times. Even if you assume that magic returns to the markets after 2020 and returns go up to 8% thereafter (the green line in the chart), there is still a gap of $5 billion after 40 years. On assets of $2 trillion, that is a gap of $10 trillion. If you assume only a 4% nominal return for the entire 40 years, the gap is $30 trillion. For the mathematically challenged, that is not a rounding error.