Feature Article: The Fed Calls Time on the United States’ Hidden Debt Crisis: Public Sector Pensions

Normally, the Federal Reserve’s quarterly “Z.1” release of the National Accounts of the United States is not a cause of much excitement, let alone angst. But this year's 3rd Quarter release different, because the Fed finally added its very considerable weight to a critical argument over the true size of America’s unfunded public sector defined benefit plan liabilities.

As one who has been involved with the public pension issue for almost twenty years, I cannot overemphasize the importance of what the Fed has done, even if its consequences remain uncertain.

To be sure, recent years have seen a rising chorus of concern over the potential understatement of US public sector pension liabilities, by, among others, Robert Novy-Marx and Joshua Raugh, Jeffrey Brown and David Wilcox, Andrew Biggs, Steve Malanga, David Crane, Chad Aldeman, the Pew Charitable Trusts, and the credit rating agencies. Yet America’s growing public sector pension crisis has mostly remained in the “grey swan” category, or a recognized and growing risk, with a potentially substantial negative impact, whose consequences are believed to be so far in the future that they have yet to trigger or force significant action in the present.

So what did the Fed do? Technically, it began to report unfunded public sector pension liabilities, as calculated by the US Bureau of Economic Analysis using a new methodology that makes two important changes from the previous one.

First, liabilities are now calculated on a projected basis, rather than simply those that have been accrued to date by current and retired employees. In a world in which salaries and benefit promises continue to grow, and longevity increase, this methodology presents a more accurate estimate of the size of a pension plan’s liabilities.

Second, these future liabilities are discounted to their present value using the yield on AAA rated corporate bonds (about 4.10% at the end of November). This is significantly lower than the discount rate used by most state and local pension plan sponsors.

The Government Accounting Standards Board (GASB) allows these plans to (a) use accrued by not projected liabilities, and (b) discount them at a rate equal to the expected long-term investment return on the assets they hold. Use of this discount rate has been a source of increasing controversy in recent years.

Under the ERISA law, which applies to private but not public sector defined benefit pension plans, the former must discount their future liabilities using the yield on “High Quality Corporte Bonds” (which are further defined as those rated AAA, AA, A by Standard and Poor’s, or the equivalent by Moody’s, Fitch, and other rating agencies). The logic for using this rate is that the risk of default on a private company’s defined benefit pension obligations is equivalent to the risk of default on its bonds. Note that this logic is completely separate from the investment return the plan earns on its assets.

ERISA does not apply to public sector pension plans. Doing so, however, would make the essence of the grey swan issue crystal clear. According to the National Association of State Retirement Plan Administrators, the median public sector defined benefit pension plan currently uses a discount rate of 7.375%. To put this in perspective, at the end of November, the average yield on Single B rated corporate bonds was 7.48%. Using ERISA’s logic, which accords with financial economics theory, public plans’ use of this high discount rate implies that their beneficiaries are exposed to significant default risk (or, put differently, the risk of having their retirement benefits cut, as frequently happens when bankrupt private sector pension plans are transferred to the Pension Benefit Guarantee Corporation).

But this does not align with what public sector plan beneficiaries are told, or the way courts have behaved in the case of recent municipal bankruptcies, where pension obligations have been treated as senior to other forms of debt (including general obligation bonds, which heretofore were thought to be the most senior of all unsecured municipal debt issues). Seven states go so far as to protect public sector pension benefits in their constitutions.

If this is in fact the case, then it makes no sense to discount public sector pension plan liabilities at an unrealistically high 7.375% rate. Rather, they should be discounted at a rate that, at minimum, reflects the true probability of default by the plans’ municipal sponsors. For example, the average S&P credit rating on state government general obligation bonds is AA. At the end of November, the taxable equivalent yield on AA rated municipal bonds is 5.00% (using a 35% marginal tax rate). Yet as we have seen, courts have found that pension obligations are senior to general obligation bonds; hence, the Fed’s use of the lower AAA corporate bond yield is logical. In fact, if one were to claim that no plan default was possible, then the appropriate discount rate would be that on US Treasuries (e.g., the current 3.14% yield on 30 year Treasury bonds).

The initial result of the BEA and Fed’s changes was a very substantial increase in the present value of state and local defined benefits pension fund liabilities to approximately $8 trillion dollars, and a substantial increase in the funding shortfall (estimated liabilities less the current value of pension plan assets) to $4.2 trillion. This is significantly greater than the total value of state and local governments’ outstanding debt, which the Fed places at $3.1 trillion.

To put this into perspective, compare the total liabilities of state and local governments ($7.3 trillion) with others shown in the Fed’s 2018/third quarter report

• Federal Government Debt = $17.8 trillion

• Residential Mortgage Debt = $10.3 trillion

• Non-Financial Corporate Debt = $9.6 trillion

• Student Loan Debt = $1.6 trillion

While the initial impact of the Fed’s was shock, the longer term effects will likely be much more severe.

The first impact may come in disclosure statements made in collection with state and municipal bond issues. Now that the Fed has put a stake in the ground about their true size, continuing to use the GASB methodology to calculate unfunded public pension fund liabilities will almost certainly increase the risk of future litigation, against issuers, bond underwriters, and disclosure counsel. Once disclosed, however, the size of unfunded pension liabilities may end up reducing some issuers’ access to municipal bond markets.

The second impact will come in state and municipal pension plan annual reports, where the same issue will arise. Disclosure of substantially larger unfunded liabilities will inevitably lead to political concerns as to how they should be addressed.

The third impact will come at budget time, specifically in the calculation of sponsors’ “Annual Required Contribution” (ARC) to their public pension plans. If the Fed’s higher estimate of the unfunded liability is used, then ARCs will be significantly higher. In the absence of benefit cuts, this leaves state and local politicians with a no-win political choice between large cuts in other spending programs, or large tax increases. When some politicians inevitably try to make it, the case for the latter will likely be weakened by the presentation of data showing how few public sector employees actually stay on long enough to receive full benefits. Moreover, because of long vesting periods for new employees, many public sector pension plans serve to transfer potential benefits from employees who leave relatively early, to those who hang on until they qualify for full benefits. “Shining the light of day” on these aspects of public sector pension plan operations does not seem likely to boost political support for them.

Which is not to say that such support will not be offered by (usually Democratic) politicians who are heavily dependent on public sector unions for their reelection. Thus the fourth impact is likely to be heightened political conflict in many states, perhaps accompanied by calls for a federal bailout of state and municipal pension plans, which would convert state conflicts into a national one, in which taxpayers across the nation would be asked to bailout public pensions in the states which have been most irresponsible in their management. These include (with the Fed’s funding ratios in parentheses):

• Illinois (25%)

• New Jersey (30%)

• Kentucky (32%)

• Massachusetts (33%)

• Connecticut (34%)

• South Carolina (36%)

• Pennsylvania (37%)

Moreover, this inevitable national debate over state and local pension bailouts will come at the very time that the federal budget itself will likely be under severe pressure from a combination of structural forces (e.g., an ageing population, increased conflict with China, higher interest payments on growing debt, etc.), and cyclical ones (e.g., automatic stabilizer payments, like social safety net benefits, during an extended economic downturn).

This brings us to the final impact, which was well described by Rob Arnott and Lisa Meulbroek in their Wall Street Journal article, “The Stealth Pension Mortgage on Your House” (5Aug18). The essence of their argument is that when it comes to raising taxes to fund public sector pensions, the least mobile tax base is real estate, as taxpayers can find various ways (including moving) to shift earnings and purchases to other states, and thus reduce local income and sales tax revenue. In fact, as others have argued, in some locations a substantial portion of recent tax increases is already being used to fund higher pension contributions (e.g., see, “Pensions Make Illinois Property Taxes Among the Nation’s Most Painful”, by Divounguy, Hill, and Tabor).

Moreover, the size of this “stealth pension mortgage” is non-trivial. As they authors note, “on average nationwide, unfunded state and local pension burdens represent 20% of real estate values. This often rivals or exceeds owners’ equity in their homes…If real estate prices eventually adjust to reflect unfunded pension obligations, many homeowners’ equity could be at risk.”

Last but certainly not least is the predictable response by some public pension advocates when confronted with the above arguments. “There’s not need to worry. High investment returns will solve the problem.” In fact, as AEI’s Andrew Biggs recently pointed out, because of reductions in the future rate of inflation they expect, in real terms the future returns public pension funds expect to earn are at an all time high (“Public Sector Pensions Assume Record High Investment Returns” published by AEI).

The key question is where they expect these returns to come from. Real interest rates are at near-record lows. Most equity markets are substantially overvalued. And if, as we currently forecast, the global economy is poised to enter the Persistent Deflation Regime, high real equity returns are unlikely.

To be sure, many public pension funds have, since the 2008 financial crisis, shifted a substantial portion of their portfolios into private equity, hedge funds, and other strategies in the hope of earning higher returns. Unfortunately, there is abundant evidence that in most cases these hopes are not being realized, even before the high expenses associated with these allocations are taken into account.

In sum, higher investment returns will almost certainly not solve the dire public pension funding situation the Fed has now brought into painful and unavoidable focus. The grey swan has finally arrived.

If you have any questions about anything we have written in this issue, please don’t hesitate to get in touch, at contact@indexinvestor.com