A New Look at State and Local Pension Liabilities

What if state and local government deficits doubled over night and nobody noticed? That’s what happened last Wednesday when the Bureau of Economic Analysis released its comprehensive revision of the National Income and Product Accounts.

As my TPC colleague Donald Marron noted, the new BEA numbers downsized the federal government relative to GDP.  They assigned economic value to intellectual property (such as a TV show about nothing).  The numbers also brought about some soul searching on what GDP should be measuring.

But few people noticed that the numbers included a new measure of defined benefit (DB) pension obligations, which remain a big deal in the public sector even as most private employers have switched to 401(k)s.  Largely because of this accounting change, what the BEA calls state and local government “net savings”– or the difference between current revenues and expenditures – fell from -$129 to -$252.7 billion.

Given attention to state and local pensions lately, especially in Detroit, one would think this news would have been met with cries of alarm or at least mild curiosity.  But alas, no.  That oversight is unfortunate because the new data shed light on an important issue.

Previously, the National Accounts measured DB pensions on a cash basis.  The only spending that counted was cash out the door when employers made pension contributions.  The only income that showed up for employees was pension fund dividends and interest.  Promises of future benefits didn’t appear anywhere on the ledger, as household income or business and government expenses.

This approach didn’t exactly add up – most importantly because public employee pensions are a form of household wealth.  Many state constitutions affirm this view, although the proposition may be tested in Michigan courts.  What’s more, the level of pension wealth depends on promised future benefits, which might bear little relationship to employers’ actual contributions.  Treating pensions on a cash basis also introduced a lot of volatility into employee compensation measures– because employers tend to cut back on pension contributions when times are tough.

So with a few key strokes last week, the BEA instituted a more consistent, accrual-based measure of DB pensions.  Overnight, we became a nation of savers.  As some commentators noticed, the 2012 personal savings rate climbed to a respectable 5.6 percent from a more modest 3.5 percent because of the increase in pension wealth.

But in the national accounts, as in life, the money had to come from somewhere.  In this case, the money came from business profits and government saving.  As noted above, state and local government “net savings” dropped from -$129 to -$252.7 billion.  Apart from some data updates and technical changes, the difference is mainly imputed interest that governments are now paying, in essence on the unpaid pension bills that they are generating each day.

The imputed interest approach gets around the fact that governments themselves do not always publish accrual-based pension measures, and when they do they use different methods and assumptions.  However, the BEA’s approach is not perfect.  They are using corporate AAA bond rates as the cost of borrowing for both public and private sector pension plans.  But if you believe states and localities can never default on pension promises, then you may prefer a lower interest rate, like the U.S. Treasuries rate.  The truth may lie somewhere in between, especially if employee promises can in fact be renegotiated in Detroit as they have been elsewhere, like Central Falls, Rhode Island.

Critics also may not like that BEA is using an accumulated benefit obligation (ABO) or “termination liability” concept for state and local pension liabilities, or the cost of these plans if governments closed their doors tomorrow and stopped accumulating liabilities for workers’ future service (and assets from workers’ pension contributions). The ABO approach understates the cost of future pension obligations, especially if there is “spiking” or benefits that increase as workers approach retirement age.

Still, the BEA should be applauded for making it clearer what state and local governments owe and to whom.  More transparency is always better.