Here Comes the Sun:
Shedding Light on the Effects of Controlled Groups on Plan Administration
Adrienne I. Moore, Esq.
George owns three companies: Day Tripper Travel Agency, The Weeping Guitar Brewery, and Strawberry Fields Natural Foods. All three businesses are corporations. George is the sole owner of the brewery but has partial ownership in the other two entities. Day Tripper is split between George, owning 90 percent, and Paul, who owns 10 percent. Strawberry Fields is owned 85 percent by George and 15 percent by Pete. Over the past several years, Day Tripper’s revenues have grown steadily, and George decides he wants to both provide greater benefits to the Day Tripper employees and encourage them to stay with the company. George contacts a TPA, Brian, about setting up a retirement plan. George does not mention the other businesses to Brian. The Day Tripper 401(k) Plan (the “Plan”) is established effective January 1, 2018. All four employees of Day Tripper make elective deferral contributions to the Plan and Day Tripper provides a discretionary employer matching contribution of 100 percent of deferrals up to 4 percent of compensation for the 2018 and 2019 plan years.
When it comes time for Brian to complete the 2020 annual administration, he requests census data from George, like he does every year. George provides data for all four employees of Day Tripper: George, Paul, John, and Ringo. Again, George does not mention the other entities on the annual questionnaire. When Brian does his weekly grocery shopping at Strawberry Fields, however, he runs into his old college classmate, Pete, and learns that George is the majority owner. Brian calls George first thing on Monday to carefully explain the controlled group rules to George and learns that Day Tripper, Weeping Guitar, and Strawberry Fields form a controlled group of corporations. What does the existence of a controlled group mean for plan administration, both historically and going forward?
Get Back(ground)
Internal Revenue Code (the “Code”) Sections 414(b) and (c), 1563, and the related regulations, define what happens when a group of entities constitutes a controlled group of corporations (as addressed in Code Section 414(b)) or a group of partnerships, proprietorships, etc., which are under common control (as addressed in Code Section 414(c)). Code Section 1563 specifically tells us how to determine whether a controlled group of corporations exists. Without getting too much into the nitty gritty, we can establish that a brother-sister controlled group of corporations exists between George’s three entities. A brother-sister controlled group happens when five or fewer persons, estates, or trusts own stock possessing at least 80 percent of the total combined voting power of all classes of voting stock or at least 80 percent of the total value of shares of all classes of stock of two or more corporations. In this case, just one person causes the controlled group to occur, as George controls, at the least, 85 percent or more of each business. (There is actually a second requirement for brother-sister entities, called the “identical ownership” test, but it is not relevant when only one owner is involved.)
Note that this is a deliberately simple example and that this area of law is incredibly nuanced. The smallest facts (for example, the existence of purchase options, changes of entity structure, relationships between the individuals, etc.) can change the outcome. If there is any question about the existence of a controlled group (or affiliated service group, which we do not address here), you should engage a knowledgeable attorney or Certified Public Accountant (“CPA”) familiar with the rules to make a determination.
As dictated by Code Section 414(b), the existence of a controlled group means that all employees of all corporations in the controlled group are treated as being employed by a single employer. Importantly, this specifically means that all employees are treated as being employed by a single employer for purposes of all Code sections related to plan qualification and nondiscrimination testing (Code Sections 401, 410, 411, 415, and 416). Brian, then, did not have all the information he needed to perform coverage and nondiscrimination testing for the prior years and may have to redo testing, this time including the employees of Weeping Guitar and Strawberry Fields. He will also need to determine if any employees of the other entities should have been offered participation in the Day Tripper Plan.
Time to Face the Music
The controlled group affects all aspects of testing and operations. Brian will need to consider the employees of Weeping Guitar and Strawberry Fields for coverage testing, and potentially also the Actual Deferral Percentage (“ADP”) test, and the Actual Contribution Percentage (“ACP”) test. He will also need to review the plan documents to determine if there are operational failures that need to be addressed caused by how the plan defines eligible employees. He needs to determine whether the language of the plan includes the employees of the other companies (which, of course, Brian did not know existed when he drafted the plan document).
While each plan of the controlled group is not required to include all employees of the other companies for plan contributions and benefits, the coverage testing rules require that a plan cover a reasonable cross-section of nonhighly compensated employees (“NHCEs”). If Day Tripper’s employee group is not sufficient to meet these rules when we consider the other companies’ employees, there is a problem.
Operational Failures in Controlled Groups. Plan documents can take different approaches to addressing controlled groups. Under some, the plan document will dictate that all employees of a controlled group may be eligible employees under the plan unless specifically excluded under the terms of the plan. Others might require a controlled group member to adopt the plan as a participating employer to become an eligible employee. In either case, the plan remains at the mercy of the coverage and nondiscrimination testing rules.
In this instance, the Day Tripper Plan includes all employees of the controlled group unless specifically excluded. Brian, who was unaware of the other entities at the time he drafted the Plan, did not exclude the other corporations. Any employees of Weeping Guitar and Strawberry Fields who have met the eligibility requirements should have been offered the opportunity to contribute deferrals. The failure to offer them the Plan is an operational failure to follow the terms of the Plan that must be corrected.
Brian checks it out, and it is not as bad as it could have been. The Day Tripper Plan required age 21 and one year of service (using 1,000 hours) for employees to be eligible to enter. The Plan, however, also provided that all employees who had attained age 21 at the time the Plan was adopted were able to begin participating immediately.
Brian requests census data from George for 2018, 2019, and 2020 for all three entities. We know that Day Tripper has only four employees, all of whom participated in the Plan, both contributing elective deferrals and earning the match. Only three employees at Weeping Guitar have satisfied the eligibility requirements and only one employee at Strawberry Fields, Pete, satisfied the requirements. The four Weeping Guitar and Strawberry Fields employees who have met eligibility requirements should have been offered the opportunity to defer. All in all, we have problems, but they could have been much worse.
Coverage and Nondiscrimination Testing for Controlled Groups. Suppose, instead, that the Day Tripper Plan was drafted to exclude employees of other members of the controlled group from the definition of eligible employees unless the member entity has specifically adopted the plan. If a member of the controlled group is excluded from participating in the plan, its employees are treated as excluded as a class and must still be considered in coverage testing. As a quick reminder, under the ratio percentage test for determining coverage, the test is satisfied if the percentage of NHCEs (who, but for any exclusions other than age and service would have been in the plan) benefitting under the plan, divided by the analogous percentage of highly compensated employees (“HCEs”), is greater than or equal to 70 percent. As noted above, eight employees meet the eligibility requirements (four at Day Tripper, three at Weeping Guitar, and one at Strawberry Fields). Two NHCEs benefit under the Plan (John and Ringo) of the five NHCEs total who have satisfied eligibility requirements. Two HCEs benefit under the Plan (George and Paul) of the three HCEs total (Pete is the only non-benefitting HCE). The ratio percentage test, then, would look like this:
|
NHCE |
HCE |
||
| Benefitting Employees |
2 |
2 |
|
|
Total Non-Excludable Employees |
5 |
3 |
|
| Benefitting Percentage |
40% |
66.67% |
|
|
Ratio Percentage |
60% |
The ratio percentage is not at least 70 percent, meaning the test fails to meet coverage requirements under the ratio percentage test. Additional employees will need to be let into the Plan to correct this failure. (Brian, of course, may still see if the Plan can pass using the average benefits test before taking corrective action.)
Brian would need to look at the entire eligible employee population (or those who should have been eligible, after satisfying the coverage test) to determine whether there are additional failures. The eligible Weeping Guitar and Strawberry Fields employees, then, need to be included in the ADP and ACP tests, and, as they have deferred nothing, they become additional zeroes when calculating the ADP and ACP for the NHCE group – making the test harder to pass.
I’ve Got to Admit It’s Getting Better
Brian and George will need to approach the necessary corrections looking both forward and backward. Looking prospectively, George may decide to amend the Plan to exclude Weeping Guitar and Strawberry Fields (although the plan will need to meet coverage testing after such exclusions, as was discussed above), or only Strawberry Fields (to keep Pete, the HCE, from participating and thereby improve the testing results). George could also exclude only HCEs from participating or adopt a safe harbor provision to make failures less likely in the future. He has plenty of options going forward, which he can discuss with Brian. George also needs to ensure that the failures do not compound by offering the Plan to all eligible employees today. George notifies the four eligible employees about the Plan and provides them with deferral election forms. Pete begins making deferrals on March 1, 2021. Only one employee at Weeping Guitar, Pattie, begins making deferrals at that time. George can now address the prior years.
Correcting Operational Failures. George’s previous failure to offer the Plan to the eligible employees can be corrected following the procedures in Revenue Procedure 2019-19, Appendix A.05, which addresses the exclusion of an eligible employee from participation. Typically, to correct for an improper exclusion of someone from a 401(k) plan, the plan sponsor will have to identify the affected participants, calculate their Missed Deferral Opportunity, and fund a corrective contribution to the plan, adjusted for lost earnings. Any associated match the participants should have earned on those deferrals must also be funded, adjusted for earnings. If the four employees who were improperly excluded were eligible to participate on January 1, 2018, they are entitled to a Missed Deferral Opportunity and the associated matching contribution for January 1, 2018, through March 1, 2021 (the date they were finally offered the Plan). It does not matter that two of them are electing not to participate going forward. If you fail to offer the opportunity to employees, the IRS assumes they all would have participated. Lost earnings will need to be calculated through the date the Missed Deferral Opportunity and missed match are funded, as well.
For more detailed information on correcting improper exclusions and calculating the Missed Deferral Opportunity, see our previous Solution here. For information on calculating lost earnings, see our previous Solution here.
Correcting Testing Failures. If a plan fails to satisfy the coverage requirements under the Code, and the failure is not corrected within 9½ months following the plan year end, the plan has a demographic failure. Under EPCRS, the demographic failure cannot be self-corrected, meaning a plan sponsor must file the correction with the IRS under the Voluntary Correction Program (“VCP”) to fully and properly correct the failure. This more formal procedure requires the payment of a user fee to the IRS. When Brian redoes the testing for the 2018 and 2019 plan years, he finds the Plan failed coverage and both ADP and ACP testing for both years. The issue is discovered well after the 9½-month mark, meaning the coverage correction will have to be submitted through VCP.
The coverage failure is corrected in the same manner as if it were done timely. Either greater allocations must be given to employees (if using the average benefits test to meet coverage requirements), or employees must be given allocations as new participants, as needed to satisfy the ratio percentage test. The Plan will also need to be retroactively amended, using the same correction method as that found in Treasury Regulation Section 1.401(a)(4)-11(g), to provide these greater benefits. The IRS will need to approve the amendment before it can be adopted.
For the ADP and ACP failures, the rules are not the same. If an ADP failure is discovered timely, it is corrected by refunding amounts to HCEs as necessary to lower the ADP of the HCEs until the test passes. This must be completed within 2½ months of the plan year end to avoid owing excise taxes to the IRS, and within 12 months of the end of the plan year being corrected to correct through refunding alone. Alternatively, the failure may be corrected by funding a Qualified Nonelective Contribution (“QNEC”) or Qualified Matching Contribution (“QMAC”) to the accounts of the NHCEs to raise the ADP of the NHCE group until the test passes (and no reduction to the HCE accounts is needed). Once the one-year correction period passes, however, the QNEC option is still available, but the refund method is modified. The refunds are still made to the HCEs of their excess amounts, but the employer must also fund a QNEC equal to the total dollar amount refunded to the HCEs. That QNEC is then allocated to the NHCEs. This is called the “one-to-one method.”
The same correction methodology applies to ACP failures (except that excess amounts are distributed to the HCEs if vested, or forfeited if unvested). For these failures discovered in early 2021, the 2018 and 2019 plan years can be corrected by funding a QNEC or QMAC sufficient to pass the ACP test or using the one-to-one method. Plan sponsors tend to prefer the one-to-one method, as it is typically less expensive than funding the QNEC.
For a detailed discussion on correcting ADP and ACP failures, see our prior Solution here.
The Yoko Factor: Can We Break It Up?
Although George can correct the prior failures and administer the Plan correctly going forward, he is now left with a plan that is providing benefits to people he did not intend to include. Over the course of the correction, Pete has requested weekly updates on when his QNEC will be funded, and the relationship has soured. Observing this, Brian discusses with George that it might be worth talking to someone experienced to see if there is a way to break the controlled group to get rid of Pete. George reaches out to John’s girlfriend, Yoko, a CPA.
Although it is by no means a simple or easy decision, George does have options. By divesting his ownership in the other organizations, he can break the brother-sister controlled group as best suits his needs. After considerable discussion with Yoko, George decides that Pattie is really a great employee (there’s also something in the way she moves, but that’s another story), and he wants to continue covering Weeping Guitar employees under the Plan. Because the Weeping Guitar employees have less disposable income, he also amends the Plan to provide a 3 percent safe harbor nonelective contribution to help pass future testing. Finally, George sells 35 percent of his interest in Strawberry Fields to Ringo. There are no options or other restrictions on Ringo’s ownership interest. (Yes, the fact that Ringo is an employee of Day Tripper is important to the analysis! This is what we mean when we say it is nuanced.) Because George’s ownership in Strawberry Fields has dropped to 50 percent, there are no longer five or fewer persons who own at least 80 percent of the stock in both Strawberry Fields and Day Tripper. The controlled group is broken, and Strawberry Fields employees can be freely excludible from the Plan.
Ob-la-di, Ob-la-da, Life Goes On.
The existence of a controlled group can profoundly impact a plan’s annual testing and regular operations. It is important to be certain that all entities and all employees are being considered to ensure the continued compliance of the plan. George had to learn this lesson the hard way, but his understanding of plan operations and his fiduciary duties is better for it. Brian, realizing that the best thing he can do for his clients is to get ahead of these problems, adds additional questions to his annual data request designed to elicit information that would flag a possible controlled group issue. He also makes a point of following up with additional questions during his annual meetings with plan sponsors. By asking the right questions and educating the plan sponsors, Brian can be sure that everything in the future will come together.
Additional Guidance:
Treas. Reg. §1.401(k)-2
Treas. Reg. §1.401(k)-6
Treas. Reg. §1.401(m)-2
Treas. Reg. §1.401(m)-5
Treas. Reg. §1.410(b)-2, et seq.