One of the reasons that pension crises sneaked up on state and local governments over the past few years is that pensions are complicated and their finances are easily misunderstood. If I promise to pay you $100 in 20 years, what sort of expense should I say I incurred this year? Governments have been allowed to come up with all sorts of answers to that question, including “zero.” Two developments this spring push us toward more honest accounting -- but will they matter?
One is a statement from the Bureau of Economic Analysis, which produces the national income and product accounts measuring the size and characteristics of the economy. A major component of the NIPAs is employee compensation, including pensions. The BEA had measured pensions on a cash basis: If a company or a government deposited $10,000 into its pension fund, it was counted as spending $10,000 on pension compensation. But that’s a bad method -- the cash contributions might not line up with actual accrued costs for pensions because, for example, entities sometimes underfund their pension systems.
Starting now, the BEA will use an accrual basis: If, this year, you promise your employees benefits with a present value of $10,000, you are recorded as incurring a $10,000 compensation expense, regardless of how the amount you actually deposited in your pension fund. Even more important, the BEA will use a standard rule for calculating that present value that is much better than the ones governments generally use.
Pensions are bond-like promises: Workers are promised fixed payouts regardless of the financial performance of the assets that are set aside to meet those promises. For that reason, private pension funds are required to establish the present value of benefits by using discount rates that are tied to the interest rates on high-quality bonds. Today, that would mean a rate of about 4 percent. Governments, however, are allowed to use higher discount rates, usually more than 7 percent, linked to the expected rate of return on invested assets.
This artificially depresses the recorded cost of promised pension benefits: The higher the discount rate, the lower today’s reported cost. That math depends on a future high, equity-driven return on assets. It doesn't record any cost for the insurance taxpayers provide to cover the risk that pension-fund assets will underperform their projections.
This methodology has been criticized by financial economists, and BEA has apparently been listening. It will adjust the reported accrual costs of state and local pension plans by using the “same discount rate series used for private plans” -- a methodology that will reflect a higher, more accurate cost. This method has been hotly resisted by public pension administrators, but endorsed by the Congressional Budget Office.
Last week, Representative Devin Nunes, a California Republican, introduced related legislation, the Public Employee Pension Transparency Act, which would require pension funds to make disclosures using approximately the same method that BEA will use to calculate economic data. I wrote about this bill when a version was introduced in the last Congress; it has been improved to add a “current cost” provision, which would improve clarity about not just total pension liabilities but also the annual value of pension compensation provided to employees.
Added transparency along the lines of Nunes's bill would send a valuable message to municipal officials -- they are spending more on pensions than they think -- and might lead to a critical re-evaluation of public employee compensation systems, which tend to be heavy on noncash compensation. Then again, it might not.
Over the past few years, the Government Accounting Standards Board has forced major improvements in the disclosure of governments’ liabilities for retiree health-care benefits. State and local governments must now disclose the present value of the benefits they have promised employees in retirement, and in most cases they must use a low discount rate. This has led to the discovery that state governments alone have run up more than $630 billion in such unfunded liabilities, and that number is much higher if local authorities' obligations are included. Yet the reaction, from bond markets and elsewhere, has been muted.
So opacity, while a problem with retiree benefits, isn't the crucial problem. Over the past few years, bondholders, taxpayers and other stakeholders have become much more informed about the magnitude of unfunded liabilities for pensions and retiree health care. Even if the financial statements of pension funds haven't themselves improved, the readers’ interpretations have.
But bondholders seem to have judged that actual municipal insolvencies will be rare. And when they happen, bondholders will probably end up ahead of pensioners in priority. Rhode Island, which has some of the worst pension funding problems, even passed a law explicitly putting the bondholders first in line for repayment. And taxpayers and pensioners still face years of fighting and uncertainty about who will be forced to bear the costs when promises become too large to keep. Before then, we aren’t going to see big changes in the way states and cities handle retiree benefits.
That’s because even when costs are transparent, the main reason that lawmakers overpromise remains intact: If you pay an employee more wages, you need cash today; if you promise more benefits in retirement, you can pay later. A requirement to disclose a higher cost might not matter much unless it comes with a requirement to actually set aside the extra cash.
This is how the Employee Retirement and Investment Security Act of 1974 drastically changed corporate pensions. Companies weren’t just required to disclose pension costs accurately, but to fund them promptly and adequately. This has made pensions safer but also more rare; deprived the option to promise now and pay later, some companies decided to stop promising. (This recognition that employers aren’t ideally positioned to manage retirement savings, which may belatedly come to the public sector, is all the more reason we need a federal solution to the retirement crisis.)
Cost transparency is a first step, but to produce meaningful policy change it will probably have to be followed by a funding mandate. An ERISA for the public sector would make government pensions more secure and more rare by taking away the option for lawmakers to make pension promises without funding them.
(Josh Barro is lead writer for the Ticker. Follow him on Twitter.)