403(b) Plans: Things to Know about Average Contribution Percentage Testing

November 17, 2009 (PLANSPONSOR (b)lines) - The arrival of the final 403(b) regulations has transformed many compliance issues; however, the regulations are just as noteworthy for what has not been altered.

For example, 403(b)s that provide for employer matching or employee after-tax contributions have always been required to perform the Average Contribution Percentage (ACP) test. However, plan sponsors of new 403(b) plans or plan sponsors who have modified their plans to add employer matching or after-tax contributions as a result of the changing 403(b) landscape may be facing ACP testing for the first time this year.


Which Plans Must Test?


A matching contribution is defined as an employer contribution that is made to a plan on condition of an elective deferral. Some plan sponsors fall into a common trap whereby voluntary pre-tax deferrals are “required” in a plan but fail to satisfy the Code requirements for a mandatory contribution (e.g., as a condition of employment, or pursuant to a one-time irrevocable election to participate in the plan). They provide what they believe is a non-matching contribution, but require that the participant defer to the plan.


For example, the plan may provide for a 4% base employer contribution, but require an employee to contribute 5% in order to receive the employer contribution. If the 5% deferral is not a condition of employment, or not pursuant to a one-time election to participate, the 4% “base” contribution is actually a 4% matching contribution subject to ACP testing. (That might occur if, when the employee is first eligible, he either participates in the plan or does not, and is excluded from any future right to participate.)


Also note that after-tax contributions do not include Roth elective deferrals.


There are a host of exemptions to the ACP test requirements. All plans except for the following types are subject to the ACP test requirements:


  • Plans that do not benefit any Highly Compensated Employees (HCEs), who are generally defined in 2009 as those employees who earned in excess of $105,000 in 2008. The $105,000 figure is subject to indexing each year.
  • Church plans under section 3121(w) of the Code. Note that this exemption only applies to so-called “steeple” churches and qualified church controlled organizations (QCCO), which means that only the churches themselves and QCCOs such as church elementary and secondary schools are exempt from the ACP test requirements. Church hospitals, nursing homes, and other religious organizations under 414(e) must satisfy ACP testing.
  • Governmental plans under section 414(d) of the Code, including public education institutions.
  • Plans or portions of plans that benefit collectively bargained union employees.
  • Plans that qualify for the ACP safe harbor. To qualify, the plan must contain special provisions, including the following:            
    • Minimum required employer contribution of i) a 4% maximum match (100% of first 3%, 50% of next 2% of employee elective deferral) or ii) a 3% discretionary base contribution that would not require an employee elective deferral
    • Contribution must be immediately vested
    • Contribution would apply to all employees, existing and new
    • Contribution must not be eligible to be distributed in the event of hardship
    • Participant notice requirement of at least 30 days prior to beginning of plan year
  • Plans that qualify for the QACA, or Qualified Automatic Contribution Arrangement safe harbor. To qualify, the plan must contain several provisions, including the following:
    • Minimum required elective deferral of 3%, increasing to 4% after two plan years and each plan year thereafter, up to a maximum of 6%. Employees can opt out within 90 days, and contributions made on their behalf can be withdrawn within that timeframe should they do so.
    • Minimum required employer contribution of i) a 3.5% maximum match (100% of first 1%, 50% of next 5% of employee elective deferral)  or ii) a 3% base contribution that would be made even to those who opt out of the automatic contribution
    • Contribution must vest within two years
    • Contribution would apply to all employees, existing and new, who have not made an affirmative election to defer a percentage of pay
    • Two participant notice requirements, including  i) annual notice requirement at least 30 days prior to beginning of plan year and ii) notice to newly hired employees on first day of employment


 If your plan falls into one of these categories, congratulations, the rest of this article does not apply to you!

The test calculates an “actual contribution percentage” for each participant. Basically, that amount is the matching contribution an employee receives plus the after-tax contribution an employee makes in a plan year, divided by his compensation as defined in the plan document. Thus, if an individual receives a matching contribution of $2,000, and his plan compensation is $40,000, his actual contribution percentage is $2,000/$40,000, or 5%.

Note that the plan document definition of compensation must satisfy a nondiscriminatory definition under 414(s) in order to be used in ACP testing. Suffice it to say that the plan compensation definition could not include/exclude types of pay that would favor HCEs as defined above. If, for example, only overtime pay was excluded, and overtime pay was only earned by Non-Highly Compensated Employees (NHCEs), such a definition of compensation would be discriminatory and could not be used.

If an employee is eligible to receive the match or to make after-tax deferrals, but fails to participate, then the average contribution percentage would be zero. The “zeros” will in fact have a negative impact on the testing, so the greater the voluntary participation in the plan, the less likely it would be for the ACP test to fail.

Once an actual contribution percentage is calculated for each employee, the employees are then divided into two categories: HCEs (as defined above for 2009), and Non Highly Compensated Employees (all other eligible employees).

The average of each category is taken and compared as follows:

If average NHCE% (ACP) is:

Then average HCE% (ACP) cannot exceed:

2% or less

2 x ADP of NHCEs

More than 2%, but less than 8%

ADP of NHCEs + 2

8% or more

1.25 x ADP of NHCEs


For example, if the Average Contribution Percentage of the NHCEs amounts to 1.5%, and the Average Contribution Percentage of the HCEs is 3.5%, the ACP test would fail, since, according to the chart above, the ACP for HCEs could not exceed 2 x 1.5%, or 3%.

In addition, for plans with less than a one-year waiting period to receive the matching contribution, NHCEs with less than a year of service may be excluded from ACP testing under what is called the “early participation” rule.

ACP testing is extremely sensitive to participation level. If employees fail to voluntarily participate in the plan in significant numbers, the test may fail year after year. There are safe harbors to avoid such testing, but they require the built-in expense of a required minimum contribution.

What happens if the plan fails ACP?

The primary method of addressing the test failure is to distribute the excess aggregate contributions to the HCEs by the end of the year following the plan year of the failure. The distribution is taxable to the employee in the year distributed, and is not eligible for rollover. If the plan provides, non-vested contributions should be forfeited rather than distributed.

An alternative method of correction is to make additional contributions to NHCEs in lieu of HCE distributions, but this solution is not frequently utilized in practice, due to its added cost.

Controlled Group Issues

Suppose that multiple tax-exempt entitles are under common control, so that they are treated as one employer under Code Section 414(c).  Different matching formulas among plans in the same controlled group may necessitate a separate test under 401(a)(4) benefits rights, and features testing that takes into account the right to a particular form of matching contribution. However, depending on other nondiscrimination tests, such as 410(b) coverage, it is possible to perform separate ACP tests for each plan with a matching contribution within the controlled group.


ACP testing can be an administrative headache for plan sponsors, but fortunately, providers and advisers are willing and able to help facilitate the testing process.


- Michael Webb, Vice President, Retirement Services at Cammack LaRhette Consulting

NOTE: This article is for general informational purposes only and is not intended to be taken as legal advice or a recommended course of action in any given situation.  Readers should consult their own legal adviser before taking any actions suggested in this article.