Compliance Failure Q&A
Q: First things first: what exactly is 401(k) compliance testing and when is it performed?
Sometimes called “non-discrimination” testing, compliance testing is conducted shortly after the close of a plan year, so roughly mid-January through mid-April for calendar year plans. In short, a 401(k) plan must pass these tests each year to verify that it does not benefit highly compensated employees (HCEs) at the expense of non-highly compensated employees (NHCEs) unfairly. Although there are a number of compliance tests, one of the most important is the “actual deferral percentage,” or ADP, test.
Q: What exactly does the ADP test look at?
With the ADP test, we compare the average 401(k) deferral percentage for HCEs to the average 401(k) deferral percentage of NHCEs. If the difference is greater than a certain margin (as shown in the table below), the plan is said to have “failed” the ADP test.
|Average NHCE rate||Maximum HCE Average Rate|
|Under 2%||2 x NHCE Rate|
|2% to 8%||2% + NHCE Rate|
|Over 8%||1.25 x NHCE Rate|
It’s important to note that the average deferral rate for testing purposes takes into account all eligible employees, including both active and terminated employees for the plan year. As a side note, this might differ from how the average deferral rate is defined for plan health purposes, which usually looks at the average deferral rate only of those participants who are actively contributing.
Q: What exactly are the consequences of failing the ADP test?
It’s probably scarier than it sounds because it can be fairly easily corrected. There are two ways to correct the failure. One is to refund excess 401(k) contributions to the impacted HCEs. The refund amount is dependent on the size of the failure, but it is taxable to the affected employees (often owners and senior managers) who will likely be unpleasantly surprised by this turn of events. It’s never an easy conversation for the plan administrator to have.
The second method is to make a corrective contribution (equal to the failed margin) known as a Qualified Non-elective Contribution (QNEC) to all of the non-highly compensated employees. This may be costly to the employer, but if the failed margin is small and the company is on the smaller side, this may be a good alternative to correct the failure. Generally, this correction needs to be completed by two and a half months after the end of the plan year being tested (March 15 for calendar year plans).
Q: OK, before we go any further, let’s make sure everyone is on the same page with respect to the definitions of HCE and NHCE.
Ah, and that’s where some of the 401(k) fun comes in because there are actually different ways that these categories of employees can be defined. And the plan sponsor has some flexibility in choosing the definition that may make it easier for the plan to pass compliance testing. But one thing that’s important to know is that you only have to define who falls within the HCE category since NHCEs are just the residual (i.e., everyone else).
Compensation is understandably one factor in determining whether someone is an HCE or not. But it’s based on prior year compensation data. So if we’re in February of 2021, doing 2020 compliance testing on a plan, we’d need 2019 compensation data from the plan sponsor. That can cause a lot of confusion at first, especially for plans that just started up. For example, a plan that started in June 2020 will understandably question why they need to provide us with company compensation data —before the 401(k) plan was even in existence. It’s because we need to determine who was an HCE, and that’s based on the prior year, also known as a “lookback year.” Obviously, if the company wasn’t even in existence in the prior year, we would then have to rely on more recent data. And in fact, in such a case, we wouldn’t rely on compensation to define HCEs, but just the ownership definition, which we’ll get into.
Q: So it sounds like the next logical question then is: what are the different HCE definitions?
Sure. Certain types of employees are automatically defined as HCEs regardless of their compensation. This can be tricky for businesses, especially those that are small and/or family run. An HCE is an employee who meets one of the following criteria:
- Ownership in Current or Prior Year – regardless of compensation, owns over 5% of (1) outstanding corporate stock, (2) voting power across corporate stock, or (3) capital or profits of an entity not considered a corporation. This includes family members.
- Prior year compensation exceeds IRS definition of HCE. This is regardless of current year compensation. In 2019, this amount was $125,000.
Q: I guess this leads to other methods of defining HCEs and NHCEs.
Exactly. Alternatively, larger plans especially may wish to define HCEs using the Top-Paid Group Election (TPG) method that allows them to limit the number of HCEs to their top 20% of employees based on prior year compensation. This must be defined in the plan document and could be beneficial if high earners who aren’t in that top 20% are contributing significantly, which in turn can help boost the average contribution rate of the other 80%.
One thing to note is that any employees who are considered highly compensated under the ownership definition will still be treated as HCEs, regardless of compensation. So the exact percentage of HCEs using the TPG method may actually exceed 20%.
Q: So what are some factors that contribute to ADP testing failures?
One of the most common challenges we see happens when plans start late in the year. Often, HCEs who have more discretionary income are so excited about the plan and the ability to maximize their savings and their tax deferrals. So even with just a few payroll periods left in the year, they maximize their contributions, contribute at much higher rates than NHCEs, and cause the plan to fail.
Plans that start late in the year should be aware of this potential problem. If they don’t want to delay starting the plan, they should communicate to HCEs that they will be unlikely to contribute the maximum annual amount (and may risk receiving a taxable refund of contributions after the year ends).
Our message to plan administrators, though, is this: if the sole focus for starting a 401k plan is to allow the owner or other HCEs to max out their contributions, be forewarned that your plan may fail the ADP test. Remember that as a fiduciary, you must operate the plan in the interests of all of your employees.
Q: Any other things to watch out for?
Other plans that may need to be more cautious include small plans, especially where the owner may be the only HCE. If other employees aren’t contributing or contributing enough, that can be difficult.
Another wrinkle can occur when there are HCEs who are earning less than the statutory maximum compensation amount. Among other things, this is the maximum amount that can be used when performing the test calculations. Consider an HCE under age 50 who is contributing the maximum annual 401(k) contribution of $19,500. If they are earning the statutory maximum compensation amount of $285,000 for 2020, that is the equivalent of 6.8% of salary. However, if they are earning just $150,000 (which also qualifies them as an HCE), that same $19,500 translates into a 13% contribution rate. So the range of HCE compensation can have a huge impact on that HCE ADP.
Q: How can plans ensure that they don’t fail the ADP test?
For those who aren’t aware, there’s one very easy solution. The ADP compliance test can be bypassed altogether if the plan adopts a Safe Harbor plan design, which requires a mandatory contribution. Of course, the company needs to weigh the added expense against the benefits of reduced compliance headaches and potentially better funded retirements for their employees. But for plans who are starting late in the year, adopting Safe Harbor is a great way to avoid potential testing failures and having to refund contributions to HCEs.
Q: And what if the Safe Harbor plan design is just too costly?
Short of adopting a Safe Harbor plan design, there are other things plans can do to reduce the likelihood of failing the test. For instance, implementing automatic enrollment for all employees at a rate that is sufficiently high can go a long way. Most people do not opt-out of the plan or change their default contribution amount. So if the plan contributes everyone at, say 6%, there’s probably a much better chance that the plan will pass ADP testing.
In addition, sometimes (but not always) a matching contribution can really motivate employees to save more. For instance, if employees have to contribute 6% to earn the maximum employer contribution, they will be more likely to contribute that amount. Often this requires clear and consistent communication to be sure that employees newly eligible for the plan are also aware of the matching feature and how they can earn the maximum amount. That said, if an employer is willing to take on the expense of a matching contribution, then a Safe Harbor plan design may make more sense since that eliminates the uncertainty associated with compliance testing altogether.
Plans may also decide to use the Prior Year testing method, which allows them to limit HCE plan contributions going forward based on the results of the prior year tests. This is not a guarantee that the plan will pass the ADP test, but it reduces the likelihood.
Q: What’s your advice to plans that have failed the ADP test?
First of all, don’t panic. It’s not uncommon for plans to fail the ADP test. That said, it’s worth analyzing what else is going on with plan design that could be negatively impacting participation or contribution rates. For instance:
- Is the eligibility requirement restricting from contributing to the plan who otherwise might thereby helping boost NHCE engagement?
- Is the plan being made available to all employees who are eligible according to the plan document? Unless it’s written in the plan document that part-time and/or seasonal employees such as interns cannot contribute to the plan until they meet certain eligibility requirements (specific to this employee class), they must be given the opportunity to contribute to the plan.
- Are definitions of plan compensation (excluding pre-participation compensation for instance) skewing the average contribution percentages and impacting testing in unexpected ways?
Of course, we caution everyone that testing can change from year to year, especially for new plans or companies just starting out, so it’s not something that’s one and done. New plans who pass their first year should especially guard against getting too complacent in thinking they won’t have any problems in the future.
In addition, plans should monitor plan engagement, paying attention not just to the participation rate but the average contribution rate of different employee groups, and continue to communicate the benefits of the plan, particularly to those groups who need to hear it most. Betterment can work with plans to develop a strategy for reaching out to their employees.
Q&A was conducted in 2021 and is meant to be educational in nature.
View original post