How To Fix The Pension Mess

Over 10 years, federal liability for guarantees is $87 billion. Let's spread out the risks

By Glenn Hubbard

The Pension Benefit Guaranty Corporation has grabbed headlines in recent months by taking over the pension plans of bankrupt steel and airline companies. And now auto-parts giant Delphi Corp. may be headed its way. Questions about the agency's worsening financial condition create an opportunity to consider reforms -- and, more important, rethink public policy relating to how Americans prepare for retirement.

The PBGC insures (up to a limit) defined-benefit pension plans, a type of employee compensation earned during one's working years but paid during retirement. To make good on this promise, employers are required to set aside funds toward this liability and buy insurance from the PBGC. The agency takes over pension plans if the sponsor becomes insolvent. The PBGC is then responsible (again, within limits) for paying the promised retirement benefits.

Two big problems have led to the PBGC's current financial woes. First, complex pension-funding rules have allowed many plans to become underfunded, especially since 2000, putting taxpayers at risk. Second, the PBGC's insurance is too cheaply priced, increasing the bite taxpayers may one day have to swallow.

THIS URGENT SITUATION must be addressed. The nonpartisan Congressional Budget Office (CBO) estimates that the present value of the PBGC's net costs for defined-benefit pension insurance -- the present value of the unfunded insured liabilities of terminated plans minus the present value of premiums received -- over the next 10 years is $87 billion. About $63 billion of that cost reflects losses not yet incurred -- in other words, losses that might be reduced with reform. To avoid these costs, Congress would need to base premiums charged on the underlying risk of insured plans. The CBO estimates that a fair premium would be a stiff 6.5 times as much as current rates.

In addition, Congress should more closely match expected cash inflows from premium plan assets with the outflows from premium plan liabilities. In practice, this could be done by limiting the share of assets invested in equities, as opposed to bonds. Congress could also tighten rules to give companies less time to close deficits. All these reforms reduce future taxpayer liabilities. But they also raise the costs of offering defined-benefit pensions, probably reducing the number of companies that will make them available.

Any reform should increase the transparency of taxpayers' exposure to PBGC liabilities. Not only should increases in the PBGC's shortfall be accrued in the federal budget but past PBGC losses for plans that have been terminated could be funded with a one-time catch-up appropriation to the PBGC.

Finally, Congress could consider transferring the PBGC to private owners or even put in place private policies bought by plan sponsors from competing insurers. That would inject into the system some innovation in risk management found in the private sector. Taxpayers still must address current losses and, assuming insurance is mandatory, the government would probably regulate terms of the insurance. Yet what would almost surely emerge is more costly insurance -- appropriately so -- and, again, reduced reliance on defined-benefit plans.

But is this consequence really so troubling? In 25 years, participation in private defined-contribution pension plans, such as 401(k)s, has increased, from 37% to almost double that today. Such plans allow workers to take pension assets with them as they change jobs, but employees bear the risk of changes in the value of their pensions. Proponents of defined-benefit plans warn of this greater asset-price fluctuation risk in defined-contribution plans. But defined-benefit plans can leave workers in worse shape given job turnover and wage fluctuation.

So rather than favoring one pension form by concentrating on PBGC reform, Congress should think more broadly about retirement savings policy. Included in that discussion should be a strong minimum benefit in Social Security for lower-income retirees (a de facto defined benefit), slower growth in Social Security benefits for higher-income households to control costs, and increased incentives for defined-contribution plans to help younger households bolster pension savings.

Large deficits faced by the PBGC are a political challenge that Washington will be forced to fix. Doing so without sound economic analysis and thinking more generally about retirement income would be a missed opportunity.

Glenn Hubbard is dean of Columbia Business School. He chaired the Council of Economic Advisers from February, 2001, to March, 2003 (

Before it's here, it's on the Bloomberg Terminal.