The audit by the Labor Department’s Office of Inspector General recommended that the Pension and Welfare Benefits Administration focus more attention on protecting pension benefits in cash balance plans, according to the Bureau of National Affairs.
Additionally, a report on that OIG audit faulted PWBA for not directing ‘significant enforcement resources’ to reviewing the way plans calculate accrued benefits for those who leave their jobs before normal retirement age, and challenged the PWBA to work with the Internal Revenue Service to develop improved guidance for cash balance plan administrators in determining participants’ accrued benefits.
That report was based on an audit of 60 converted cash balance plans, 13 of which were found to have failed to pay workers who left employment before normal retirement age the benefits to which they were legally entitled – underpayments that added up to about $17 million a year, the report estimated.
However, the OIG went on to extrapolate from those findings to estimate that workers whose benefits are in cash balance plans ‘may be underpaid between $85 million and $199 million annually.’ Those numbers were based on the application of the sample finding to an estimated 300-700 traditional defined benefit plans converted to cash balance plans since the mid-1980s, affecting over 8 million workers and involving pension assets of more than $334 billion.
The PWBA questioned whether the audit team’s sampling methodology ‘was appropriate for reaching such a broad conclusion and whether the assumptions used to extrapolate the error from the sample to the overall population were correct,’ according to BNA, citing comments by Ann L. Combs, assistant secretary of Labor. She did, however, appear conducive to investigating the 13 plans identified in the audit as having underpaid workers, according to the report.
For its part, OIG defended its audit, saying that, ‘despite the lack of a statistical approach, we found 13 plans with underpayments to participants out of 60 plans reviewed. Under any sampling or targeting method, statistical or judgmental, this is a disturbing finding,’ the report said.