While private-sector pensions help millions of Americans achieve retirement income security, the number of private defined benefit (DB) plans1 has declined substantially over the past two decades. Recently, those concerned with the viability of the private defined benefit pension system point to significant increases in pension contributions plan sponsors must make and to the fact that most plans are currently underfunded. The underfunding of plans, due in large part to the sharp decline in the stock market combined with a general decline in interest rates, has increased substantially. The Pension Benefit Guaranty Corporation (PBGC), whose single-employer insurance program insures the benefits of over 34 million workers and retirees in private defined benefit plans, estimated that the total underfunding exceeded $350 billion as of September 4, 2003. According to employer groups, plan sponsors face inflated and unpredictable pension contributions that have greatly diminished the attractiveness of maintaining DB plans. As a result, employer groups have suggested that plan sponsors may consider freezing their plans rather than confronting the possibility of increased pension contributions. A plan "freeze" could have adverse consequences for the retirement income security of participants because new employees are precluded from participating in the plan, and current participants might not receive additional benefit accruals. Because timely and accurate information on plan freezes could be important in assessing the overall health of the private DB system, Congress asked us to determine how many DB plans have been frozen since 2000.

We could not determine the number of defined benefit plans frozen since 2000. Currently available public information cannot be used to readily identify and track frozen DB plans. Officials at PBGC and DOL we interviewed considered potential approaches to using Form 5500 filings to determine how many DB plans have been frozen, and we found that such approaches have limitations with respect to the accuracy and completeness of the results. Consequently, PBGC requested that the 2002 Form 5500 Annual Return collect information on DB frozen plans. This information, however, is limited and will not be available until 2004. We reviewed other reports and disclosures that DB plans must provide to the government or participants, but determined that these are not useful in identifying frozen plans. Publicly available information is of limited use in identifying and tracking frozen plans in part because there is no statutory definition of a DB plan freeze. Private surveys from benefits consulting firms seek to estimate the incidence of DB plan freezes in recent years, but the results reflect information from their clients only and do not provide a clear and complete picture of plan freezes among all DB plans. More complete and timely information on frozen defined benefit plans could help assess the potential risks plan freezes pose to participants' retirement income security and the viability of PBGC's single-employer insurance program. Because it is uncertain to what extent DB plan freezes pose risks to participants' retirement income security and PBGC's single-employer insurance program, we believe that additional steps are needed to collect information on such plans and evaluate their potential consequences. Accordingly, PBGC's Executive Director should direct the agency to identify, track, and analyze frozen DB plans on a pilot basis.