David J. Kupstas, FSA, EA, MSPA Chief Actuary

*This is Part 2 in a two-part series on the mechanics of cross-testing.*

In our first installment, we described how cross-testing could be used to prove that a given profit sharing contribution allocation did not discriminate in favor of Highly Compensated Employees (HCEs) under Section 401(a)(4). We illustrated a two-person plan with one HCE and one Non-Highly Compensated Employee (NHCE).

We calculated an Equivalent Benefit Accrual Rate (EBAR) for both employees. The EBAR is a measure of the theoretical retirement benefit a current contribution can buy in the future. By comparing the EBARs and not the current contribution, we showed that the allocation was not discriminatory under the rules. Even though the HCE received a much greater current-dollar contribution, those contributions are expected to provide a bigger benefit at retirement for the NHCE since she has many more years for her money to grow with interest than the HCE does.

In this installment, we will examine the effects of including additional employees in the plan as well as adding a 401(k) feature.

**Section 410(b) and Section 401(a)(4) Intertwined**

Before we get to our example, let us talk for a moment about Internal Revenue Code Section 410(b). This section is called “Minimum Coverage Requirements.” In a teeny nutshell, Section 410(b) says a plan may meet the coverage requirements one of two ways:

- If the ratio of NHCEs benefitting is at least 70% of the ratio of HCEs benefitting. This doesn’t mean 70% of the NHCEs necessarily have to benefit under the plan. If only 50% of the HCEs are benefitting, then 35% (70% × 50%) of the NHCEs have to benefit. If 20% of the HCEs benefit, then 14% (70% × 20%) of the NHCEs must benefit.
- If the 70% test is not met, the plan can still satisfy 410(b) if (i) the ratio is at least as big as a lower “midpoint percentage” found in the regulations and (ii) the average allocation rate or EBAR of the NHCEs is at least 70% of the same measure for HCEs.

Getting back to cross-testing, a plan satisfies the nondiscrimination test of Section 401(a)(4) if each “rate group” in the plan satisfies Section 410(b) as though the rate group were a separate plan.

**An Extended Example**

We have a plan that has two HCEs, 10 NHCEs, a 401(k) deferral feature, and a company profit sharing contribution. For simplicity, we will not concern ourselves with whether the deferrals will pass ADP testing or if a safe harbor 401(k) design should be used. Each HCE is receiving a $36,000 profit sharing contribution. Each NHCE is receiving 5.00% of pay. While this may seem discriminatory in favor of HCEs when you compare the current contributions, remember that with cross-testing we are comparing the benefits these contributions could buy at retirement.EBAR 1 is based on the profit sharing contributions only. For HCE 1, EBAR 1 is equal to

EBAR 1 would also include certain other kinds of annual additions such as safe harbor nonelective contributions. EBAR 1 does not include salary deferrals or matching contributions. EBAR 2, on the other hand, does take into account this plan’s salary deferrals as well as the profit sharing contribution:

In general, EBAR 2 includes all of the contribution types of a plan, including deferrals and matches.

Once the EBARs are determined, a “rate group” is established for each HCE. An HCE’s rate group consists of every employee being tested that has an EBAR 1 greater than or equal to the EBAR 1 for that HCE. For a plan to pass the nondiscrimination test, each rate group must satisfy Section 410(b). There are two ways a rate group may satisfy Section 410(b):

**Ratio test:**If the ratio percentage for the rate group is at least 70%. If this test is met, the rate group passes with no further steps required. If not, you must proceed to the average benefit test.**Average benefit test:**There are two components to this test that must be satisfied. Under the “midpoint test,” the ratio percentage for the rate group must be greater than some lesser percentage found in the regulations, while the “average benefit percentage test” (ABPT) requires the average EBAR for the NHCEs to be greater than or equal to 70% of the average EBAR for HCEs. Note that the ratio percentages are specific to a rate group, while there is one average benefit percentage test for the entire plan.

The notes underneath the exhibit above give a clue as to whether the plan passes these various tests. Sixty percent (6 out of 10) of NHCEs have an EBAR greater than or equal to HCE 1. All of the HCEs (100%) have an EBAR greater than or equal to HCE 1, including HCE 1 himself. Thus, the ratio percentage is 60%. This does not satisfy the ratio test, but it does satisfy the midpoint test. The table in the regulations says a ratio percentage of 22.75% is needed for the midpoint test given the relative numbers of HCEs and NHCEs in the plan. Because the midpoint test is satisfied, the rate group is eligible to go on to the ABPT and still has a chance to pass 410(b).

As we see above, the average of EBAR 2 for the NHCEs is 8.57%. (Note that we have switched from talking about EBAR 1.) The average of EBAR 2 for the HCEs is 13.46%. That HCE average multiplied by 70% is 9.42%. Because this is higher than the NHCE average of 8.57%, this rate group does not pass 410(b), and therefore the entire plan cannot pass nondiscrimination without some adjustments.

While we’re here, we will mention also that the ratio percentage for HCE 2’s rate group is 20%. This is less than the needed percentage of 22.75%. So HCE 2’s rate group fails for two reasons: the midpoint test fails, and so does the ABPT.

**The Test Fails. What Now?**

When a nondiscrimination test fails like this, we don’t just leave the failing result or make the client figure out what to do. We provide options that will bring forth a passing result. Either one or more HCEs need to be cut back, one or more NHCEs need to be increased, or both.

One way to get the ABPT to pass would be to cut back the salary deferrals of an HCE. Let us cut HCE 2 back to a $12,000 deferral. That would bring the average EBAR 2 down to 12.15%, 70% of which is 8.51%, which is less than the NHCE average EBAR 2 of 8.57%, so the ABPT passes. Be advised that this change would have to occur before the deferral is actually made, so probably in December or earlier when running a year-end estimate. Other ways to pass this test would be to decrease HCE profit sharing contributions or increase those of NHCEs.

Reducing HCE 2’s deferral does nothing to help us pass either the ratio test or the midpoint test because those tests are based on EBAR 1 which does not include deferrals. To fix those tests, a good place to start would be giving a little extra profit sharing contribution to NHCE 9 since his EBAR is the closest to HCE 2’s. Indeed, an increase in NHCE 9’s profit sharing from 5.00% to 6.00% puts NHCE 9 in HCE 2’s rate group, so HCE 2’s midpoint test passes, therefore the rate group passes 410(b), and therefore the whole plan passes nondiscrimination testing.

The revisions we just made are shown below: There are other techniques in nondiscrimination testing, such as imputing permitted disparity, grouping accrual rates, and restructuring, that are not being covered here. When looking for fixes of a failed nondiscrimination test, the plan sponsor must be careful about singling out young or short-service NHCEs for additional contributions, as doing so could be deemed a violation of the rules.

**Don’t You Feel Smart Now?**

If you have followed along until the end, give yourself a pat on the back. No one said cross-testing was a straightforward concept. We hope this article gives you some insight on what is going on behind the scenes when we say that your cross-tested plan passes or fails the test. As always, we are happy to provide further details if you have questions about cross-testing.