401(k) Testing: Improving the ADP and ACP Tests

By Fred Reish, Esq., Bruce Ashton, Esq., and Marge Paul, MSPA, MAAA, EA

Plan sponsors have an important decision to make regarding the ADP and ACP tests for their 401(k) plans. This decision, which may help 401(k) plans pass the discrimination tests, must be made prior to the deadline for amending plans to comply with recent law changes, i.e., prior to December 31, 2000 for calendar year plans.

For simplicity, assume a 401(k) plan has a plan year ending December 31. For 1999, the plan must decide which non-highly compensated employees (NHs) to include in its ADP and ACP tests: either (1) all eligible NHs in the current plan year (1999) or (2) all NHs who were eligible employees last year (1998). This decision is important because it can make the difference between passing the tests or failing them or, if the tests are failed, in the amount of refunds the plan will be required to make to highly compensated employees (HCs). Note that HCs are not affected by this change in law; the plan must use 1999 compensation and 1999 eligible employees to determine the ADP and ACP for HCs.

Congress changed the testing rules to make it easier for plans to pass the nondiscrimination tests. For example, if the 1999 ADP and ACP tests are based on 1998 NH data, the plan sponsor will know the ADP and ACP for NHs shortly after the end of the 1998 year end. This means that it can limit the amounts that HCs are contributing in 1999 to either reduce refunds or eliminate them.

Alternatively, if a plan has no difficulty passing the tests using the old rules, i.e., using current year (1999) compensation and eligible NHs, no change in testing method should be made.

This new rule has been in effect since 1997. Therefore the plan's consultants, attorneys and/or third party administrators will be able to provide some insight to the plan sponsor on which method is likely to work better, given the plan's experience.

For years prior to 2001, the plan may select the more favorable method each year; all that is required is for the plan to state in the GUST amendment the method used for each year since 1996. For years after 2000, there are additional requirements. If the testing is performed using current year data for any year after 1999, the plan will not be able to switch methods until current year data has been used for five consecutive years. This increases the likelihood that the plan will choose to use prior year data—if in doubt.

The prior year method does not work well for plans in which the employer makes additional contributions (QNECs, QMACs) to enable the plan to pass the ADP and ACP tests. Therefore, we anticipate that the current year data method will be used for these (usually smaller) plans. The prior year method will generally be preferable for larger plans.

Testing methods have been liberalized. Plan sponsors should consult their advisors to find out how they can use the new, more generous, testing methods to maximum advantage.