The 401(k) plan has emerged as the most popular form of retirement plan in the United States. This trend will likely continue for some time for a number of reasons. One is the cost savings to employers, since deferral contributions are paid by employees. Another is the fact that 401(k) plans are more easily understood than traditional retirement plans and consequently more appreciated by employees.

One aspect of 401(k) plans that is not so easily understood is the annual contribution nondiscrimination testing. This article will review the mechanics of this required testing and correction methods for failed tests.

Highly Compensated Employees

Every 401(k) plan, other than “SIMPLE” plans, “Safe Harbor” plans or “Qualified Automatic Contribution Arrangements,” requires an annual test to prevent discrimination in favor of the group of employees referred to as “highly compensated employees” (HCEs). Employees who fall into the following two categories are considered to be HCEs:

• An owner of more than 5% of the employer in the testing year or the previous year (family stock attribution rules apply which treat an individual as owning stock owned by his spouse, children, grandchildren or parents), or
• An employee who received compensation in excess of a specified limit from the employer in the previous year (e.g., employees who earned more than $100,000 in 2007 will be considered HCEs in 2008). The employer may elect that this group be limited to the top 20% of employees based on compensation.

401(k) Nondiscrimination Testing

The nondiscrimination rules require average deferrals and average contributions for the HCE group to be within a certain range of the average deferrals and contributions for the “non-highly compensated employee” (NHCE) group.

Testing of employee deferrals is referred to as the ADP test (Average Deferral Percentage). The ACP test (Average Contribution Percentage) includes the employer match contributions, employee voluntary after-tax contributions and certain forfeitures allocated on the basis of deferrals or matching contributions.

Each participant’s deferral or contribution percentage is determined by dividing the applicable deferral or contributions by the compensation defined in the plan document. Averages are then determined for the HCE and NHCE groups by dividing the sum of the deferral or contribution percentages by the number of employees in the group. Below is an example of the ADP determination:

The HCEs’ average may only exceed the NHCEs’ average (for both the ADP and ACP tests) by specific limits summarized as follows:

In the above example, the maximum ADP of the HCE group is 6.25% (the NHCE average of 4.25% plus 2%). The test passes since the ADP of the HCE group is 5.50% which is less than the 6.25% maximum.

Catch-up contributions (available to participants who are age 50 or older if permitted by the plan) that exceed a statutory limit or plan-imposed limit are not included in performing the ADP test. Also, compensation for plan purposes is subject to an annual limit ($225,000 for 2007 and $230,000 for 2008). For example, assume Harry earned $300,000 in 2007 and deferred $20,500 (the maximum deferral of $15,500 for 2007 plus the maximum catch-up contribution of $5,000). His deferral percentage is calculated by dividing $15,500 (his deferral without the catch-up contribution) by $225,000 (the compensation limit for 2007).

Employees Included in Testing

In performing the ADP test, all active and terminated employees eligible to defer at any time during the plan year are included, whether or not they actually made a deferral.

The following employees are included in the ACP test, regardless of whether they received matching contributions or made after-tax contributions:

• Active employees who have met the plan’s requirements to receive a match as of the plan year-end being tested (e.g., if the plan requires active employees to have more than 500 hours of service during the plan year in order to receive matching contributions, employees with less than 501 hours are not included);
• Employees who terminated during the plan year being tested if they met the plan’s requirements to receive a match (e.g., if the plan requires 1,000 hours of service and/or employment on the last day of the plan year, employees who have not met these requirements are not included); and
• All employees eligible to make voluntary after-tax contributions at any time during the plan year.

Plans that allow employees to participate before they reach age 21 or complete one year of service are permitted to exclude such employees from the test if they are NHCEs.

If the plan covers both union and non-union employees, each group must be tested separately.

Related Companies

Employees who work for a “related” company may also have to be considered. Related companies are either part of a “controlled group of corporations” or an “affiliated service group.” Whenever an individual who owns any portion of the sponsoring employer buys into another business, the plan’s advisors should be notified so a controlled group determination can be made. The same applies if another company works together with the employer to provide services to each other or to third parties which could constitute an affiliated service group. These circumstances create important issues that could affect the qualification of the plan.

Testing Methods

Two testing methods are permitted. The first method is current year testing where current year deferral and contribution percentages are used to compare the percentages of both HCEs and NHCEs.

The other method is prior year testing where the deferral and contribution percentages for NHCEs in the prior year are compared with HCE deferral and contribution percentages in the current year. The prior year testing method gives employers the ADP and ACP limits for the HCEs in advance, which reduces the chance of a failed test at year-end and the need for taxable refunds or other corrective measures.

Whichever testing method is chosen, regulations require it to be specified in the plan document. The testing method may only be changed by amendment, subject to certain restrictions on changing from current year to prior year testing.

Mechanics of Prior Year Testing

In the first year of a 401(k) plan, or the first year 401(k) provisions are effective in an existing plan, a special rule applies since there are no prior year percentages to use for the test. The employer can assume a prior year percentage for the NHCEs of 3% for both the ADP and ACP tests or use the actual results of the first year’s test.

The second year, the maximum HCE percentage will be based on the NHCE percentage from the first year. At the end of the second year the test will be performed which will be used for two purposes:

1. The average HCE percentage will be compared to the maximum permitted average percentage (based on the NHCE percentage from the first year) to verify that the maximum was not exceeded, and
2. The NHCE average percentage will be used to determine the maximum average HCE percentage for the third year.

Correcting Test Failures

Plans that do not pass the ADP and/or ACP tests must take some action, such as making corrective distributions or additional employer contributions.

Refund Deferrals/Matching Contributions

The most common method used to correct a failed ADP or ACP test is to make corrective distributions of the excess deferrals or contributions, plus earnings (in some cases, forfeiture of matching contributions may be required).

Corrective distribution amounts (determined by a required leveling method) are allocated among the HCEs based on the dollar amount of their deferrals or contributions. If the plan permits catch-up contributions and the participant is 50 or older and has unused catch-up contributions remaining, the ADP refund is first offset by the unused catch-up contributions.

These distributions must be made within 2½ months of the plan year-end in order to avoid a 10% penalty (this deadline is extended to six months for plans that meet the eligible automatic contribution arrangement requirements). The final deadline for making corrective distributions with the penalty is the last day of the following plan year. For plan years beginning on or after January 1, 2008, these distributions are taxable in the year in which they are distributed (for plan years prior to 2008, distributions made before the 2½-month period are taxable in the prior year).

QNECs and QMACs

In some situations, a failed nondiscrimination test can be corrected by having the employer make a “qualified nonelective contribution” (QNEC) or “qualified matching contribution” (QMAC). These additional employer contributions are made to NHCEs to increase their ADP or ACP to the level needed for the HCEs to pass the test. QNECs and QMACs are required to be immediately 100% vested and subject to withdrawal restrictions. These contributions must be deposited by the last day of the following plan year.

Conclusion

Nondiscrimination testing sounds complex–and it is. Gaining a better understanding of this testing should help plan sponsors appreciate the importance of providing complete and accurate census data to their advisors.

Census information, including all employees, should be compiled as soon as possible after the plan year ends so that nondiscrimination testing can be performed accurately and any corrective distributions can be made in a timely manner in order to avoid the 10% penalty.

It is also important for business owners to be aware that related companies may impact nondiscrimination testing and to disclose to their advisors any and all ownership interests or service affiliations. This will help ensure that the plan maintains its qualified status.