Change for the better in pension accounting

Mobile Share Email Facebook Twitter LinkedIn
Headshot of Alicia H. Munnell

is a columnist for MarketWatch and director of the Center for Retirement Research at Boston College.

New procedures for estimating assets and liabilities are a positive development.

Something exciting has happened in government accounting for those interested in defined benefit pensions.  Defined benefit plans provide payments during retirement based on a formula that reflects the employee’s length of service and final average pay.  To fund the promised benefits, the plans rely on contributions from the employer – and, in the public sector, from the employee – and on the returns earned on plan assets.  Until recently, the government accounts have only showed the assets being held in defined benefit plans.  Now they show promised benefits.  That is, the National Income and Product Accounts (NIPA) have changed accounting for defined benefit plans from a cash basis to an accrual basis. 

This change in NIPA accounting has resulted in conforming changes in the Federal Reserve’s Flow of Funds financial accounts.  Previously, only the assets held by defined benefit plans were considered assets of the household; now, the accrued benefits will be treated as assets of the household and, conversely, as liabilities of pension fund sectors (private pension funds, state and local government retirement funds, and federal government retirement funds).  In turn, the unfunded liability – accrued benefits in excess of assets – will be shown as an asset to the pension fund and a liability to the private or government sponsor.  

Reporting pension benefits, rather than pension assets, has led to an increase in household net worth relative to previous publications of both the NIPA and the Flow of Funds.  Promised benefits exceed assets because both private and public plans are underfunded.   

Some questions immediately arise with the new accounting.  The first is the rate used to discount future benefits.  That is, the accrued benefit is a stream of promised future benefits discounted to the present.  What interest rate is the government using to do that calculation?  Our best understanding is that for both private plans and state and local government plans, the discount rate is based on AAA corporate bonds.  This assumption is consistent with the government’s funding standards for private plans and with arguments that fixed obligations should be discounted by something close to the riskless rate.  The corporate bond rate is currently 4.4 percent.   

The next question is the funded status of these plans as reported in the Flow of Funds.  Based on the new tables, as of the second quarter of 2013, private plans were 88 percent funded and state and local plans were 71 percent funded.   

At first, the funded ratio for state/local plans looked much too high.  My colleagues and I at the Center for Retirement Research at Boston College collect very detailed data on state and local plans, and reported a funded ratio in 2012 of 73 percent, using the plans’ assumed discount rates (average 7.75 percent).  Our estimate with a 5-percent discount rate – close to the 4.4 percent used by the government – showed public plans about 50-percent funded.   

One reason for the discrepancy is the liability concept employed.  The numbers reported by state and local plans, which we aggregate, are based on the projected benefit obligation (PBO), as recommended by the Governmental Accounting Standards Board.  The PBO includes pension benefits paid to retired employees, benefits earned to date by active workers based on their actual salaries and years of service, and the effect of future salary increases on the value of pension rights already earned by active workers.  The concept used in the government accounts for private plans and state and local government plans, and preferred by economists, is the accumulated benefit obligation (ABO), which does not include the effect of future salary increases on accrued benefits.  The difference in accrual concept could explain much of the difference in the funded ratio estimates.  In addition, the market value of assets increased between 2012 and 2013.

By the way, I know for sure that neither the federal government nor the Federal Reserve could have 2013 data for state and local plans because the reports have not yet been prepared, much less released.  But that fact should not detract from this very positive development on the pension accounting front!