In any plan year in which a plan is a top-heavy plan (Q 3917), additional qualification requirements must be met.¹Moreover, except to the extent provided in the regulations, all non-exempt plans, whether or not actually top-heavy, must contain provisions that meet the additional top-heavy qualification requirements and that will become effective should a plan become top-heavy.²

Plans established and maintained by the United States, by state governments and political subdivisions thereof, and by agencies and instrumentalities of any of these, are exempt from the top-heavy requirements.³ Also, the top-heavy rules are not applicable to SIMPLE IRA plans (Q 3706), SIMPLE 401(k) plans (Q 3778), safe harbor 401(k) plans (Q 3773), or automatic enrollment safe harbor 401(k) plans (Q 3762).⁴ PPA ’06 Section 902(c) amended IRC Section 416(g)(4)(H) to exempt from the top-heavy rules plans consisting solely of (1) cash or deferred arrangements that meet the requirements of Section 401(k)(12) or 401(k)(13) and (2) matching contributions which meet the requirements of Section 401(m)(11) or 401(m)(12).

As to when a participant is a key employee for purposes of the top-heavy rules, see Q 3931. For the additional qualification requirements applicable to top-heavy plans, see Q 3922.

Where an employer maintains only one qualified plan, that plan is a top-heavy plan with respect to a plan year if the present value of the cumulative accrued benefits under the plan, or the aggregate account balances if the plan is a defined contribution plan, for key employees (Q 3931) exceeds 60 percent of the present value of the cumulative accrued benefits under the plan, or the aggregate account balances, for all employees.¹For purposes of determining the present values of accrued benefits, or the sums of account balances, benefits derived from both employer contributions and nondeductible employee contributions are considered; benefits derived from deductible employee contributions are disregarded. Deductible employee contributions are certain contributions made before 1987; the term does not refer to salary reductions or employee deferrals.

Any reasonable interest rate assumption may be used to calculate these present values, but the IRS automatically will accept as reasonable a rate that is not less than 5 percent or greater than 6 percent. The interest rate used need not be the same as other assumptions used in the plan (e.g., the rate assumed for funding purposes). Where an aggregation group consists of two or more defined benefit plans, the interest rate assumptions used to calculate the present values must be the same in all plans.²

Present values and account balances generally are determined on the last day of the prior plan year, but when testing for top-heaviness with respect to the first plan year (as well as the second) of a new plan, the determination date is the last day of the first plan year.³

In the case of a defined contribution plan, the balance in each account on the determination date is calculated by adjusting the balance of each account as of the most recent valuation date occurring within 12 months prior to the determination date for contributions due as of the determination date.⁴

For defined benefit plans, the present value of an accrued benefit as of the determination date generally is determined as of the most recent valuation date occurring in the previous 12 months. Special rules apply in the case of a new defined benefit plan in its first and second plan years.⁵

The cumulative accrued benefit of non-key employees must be determined under the method used for accrual purposes for all plans of the employer or, if there is no such method, as if such benefit accrued not more rapidly than under the fractional method (Q 3716).⁶

In determining these present values and account balances, any distribution (generally including death benefits) made from the plan with respect to any employee during the one-year period ending on the determination date, and that is not already reflected in the present value or account balance, must be added back to the present value of that employee’s accrued benefit or to his or her account balance, whichever is applicable.⁷ In the case of a distribution made for a reason other than severance from employment, death, or disability, a five-year look-back period applies for this purpose.⁸

If an individual has not performed any services for his or her employer during the one-year period ending on the determination date, the individual’s accrued benefit and account are not to be taken into account for purposes of determining whether the plan is top-heavy.⁹ If an individual was a key employee in a previous plan year but currently is a non-key employee for purposes of the top-heavy test, the individual’s cumulative accrued benefit (or account balance) is totally disregarded.

The terms “key employee” (Q 3931) and “employee” include their beneficiaries, so that the beneficiary of a key employee is treated as a key employee and the beneficiary of a former key employee is treated as a former key employee.¹⁰ This apparently means that for purposes of testing top-heaviness, an individual’s accrued benefit or account balance must be considered in its entirety and not allocated between the individual and his or her beneficiaries. For plan years beginning before January 1, 2002, it also meant that the accrued benefit or account balance of a deceased key employee, even though payable (or paid) to his or her beneficiary, was treated as that of a key employee for four years.¹¹

A plan will not be treated as violating the top-heavy rules merely on account of the making of, or the right to make, catch-up contributions (Q 3761) by participants age 50 or over, under the provisions of IRC Section 414(v), so long as a universal availability requirement is met.¹²

Where an employer maintains more than one qualified plan, some or all of those plans will be aggregated and tested as a group for top-heaviness.Specifically, all qualified plans (including collectively-bargained plans) of an employer that cover at least one key employee (i.e., key employee plans) and any qualified plans that enable an otherwise discriminatory key employee plan to satisfy the nondiscrimination requirements of IRC Sections 401(a)(4) or 410 (Q 3841 to Q 3863, Q 3773 to Q 3802) are required to be aggregated into a single group.

In addition, an employer may designate any other qualified plan or plans (including collectively-bargained plans) not required to be aggregated under the above rules to be included in an existing aggregation group, provided that the resulting group, taken as a whole, would continue to satisfy IRC Sections 401(a)(4) and 410.

If an aggregation group is top-heavy, all plans required to be included in the group under the above rules will be considered top-heavy plans; any plan included in the group solely because of the employer’s designation will not be treated as top-heavy. Even though a collectively-bargained plan covering a key employee might be part of a top-heavy aggregation group because it was required to be aggregated, that collectively-bargained plan will be excepted from the faster vesting, minimum benefits, and maximum compensation requirements discussed in Q 3922.¹ If an aggregation group is not top-heavy, no plan in the group will be considered top-heavy, even though one or more plans composing the group would be top-heavy if tested alone.²

The procedure for testing top-heaviness of an aggregation group is the same as that discussed in Q 3917 for a single plan, except that the values tested are the sums of the respective present values and account balances determined for each plan, as of its determination date, composing the group. When plans composing the aggregation group have different plan years, the test is carried out using the determination dates that fall within the same calendar year.³

If only one of the employer’s plans is a key employee plan and that plan, by itself, satisfies the nondiscrimination requirements of IRC Sections 401(a)(4) and 410, that plan will be tested as a single plan unless the employer elects to designate another plan for aggregation with the key employee plan.

For purposes of testing for top-heaviness, a simplified employee pension plan (SEP), including a SAR-SEP (Q 3705), is treated as a defined contribution plan. An employer may elect to use aggregate employer contributions to the simplified employee pension plan, rather than aggregate account balances, for purposes of the top-heavy test.¹

How amounts rolled over (or otherwise transferred) to or from a qualified plan are treated for purposes of determining whether a plan is top-heavy depends on the surrounding circumstances.A related rollover or transfer is initiated other than by the employee, as in the case of a merger or division of plans, or is made between plans of the same employer, or related employers required to be aggregated under IRC Section 414. For a related rollover or transfer, the amount rolled over is counted as part of an employee’s accrued benefits by the receiving plan but is disregarded by the distributing plan.

An unrelated rollover or transfer is initiated by an employee, regardless of who initiated the distribution, and made between plans of unrelated employers. For unrelated rollovers or transfers, the rollover or distribution generally must be added back to the distributing plan for a one year period and generally is disregarded by the receiving plan.¹

Notice 2013-74² provides that an in-plan Roth rollover is considered a “related rollover” and so the accepting plan must include the rollover in determining the present value of accrued benefits for top-heavy status.

Precise top-heavy ratios need not be computed every year so long as the plan administrator knows whether or not the plan is top-heavy. For this purpose, and for the purpose of demonstrating to the IRS that a plan is not top-heavy, an employer may use computations that are not precisely in accordance with the top-heavy rules but that mathematically prove that the plan is not top-heavy. Several such methods are provided in the regulations.¹

In addition to the qualification requirements that apply to qualified plans generally (Q 3838), special requirements are imposed by the IRC on top-heavy plans. In addition, top-heavy simplified employee pension plans are required to meet certain minimum contribution requirements (Q 3924). In meeting these requirements the common control, controlled group, and affiliated service group aggregation rules apply (Q 8964, Q 3935). Under some circumstances, “leased” employees may be imputed to an employer (Q 3929).¹The requirements discussed in Q 3923 and Q 3924 must be met by top-heavy plans in general; top-heavy simplified employee pensions must meet only the minimum contribution requirements discussed in Q 3924.

A top-heavy plan must provide that an employee has a non-forfeitable right to his or her accrued benefit derived from employer contributions in accordance with one of the two following requirements:

1. Three-year vesting. An employee who has completed at least three years of service with the employer must have a non-forfeitable right to 100 percent of his or her accrued benefit.¹
2. Six-year graded vesting. An employee who has completed at least two years of service must have a non-forfeitable right to at least the following: 20 percent of his or her accrued benefit after two years of service, and 20 percent additional for each of the following years of service, reaching 100 percent after six years of service with the employer.²

Except to the extent that they are inconsistent with these fast vesting schedules, the rules that pertain to vesting in qualified plans generally (including years of service and breaks in service, etc.) apply for purposes of the fast vesting requirements.³ Thus, the fast vesting schedules are not safe harbors; even faster vesting may be required by IRC Section 411(d) where there is a pattern of abuse (Q 3869).

When a plan becomes top-heavy, fast vesting under one of the two schedules generally must be applied to all benefits accrued under the plan for the current plan year and all prior plan years (including benefits accrued in years before the plan became top-heavy and benefits accrued before the effective date of the top-heavy rules). The accrued benefit of any employee who does not have an hour of service after the plan became top-heavy, and any accrued benefits that were forfeited before the plan became top-heavy, need not be covered by the fast vesting schedule.⁴

Although the IRC does not require that fast vesting be applied to benefits accrued in future plan years in which a plan is not top-heavy, a return to the plan’s slower vesting when the plan ceases to be top-heavy in many cases may be impractical or impossible. For example, IRC Section 411(a)(10) requires that a change in vesting schedules not reduce a participant’s non-forfeitable percentage in his or her accrued benefit and that participants with at least three years of service be allowed to elect to be covered by the previous vesting schedule (Q 3869).⁵

Integration

Although the IRC does not prohibit integration in a top-heavy plan, the fast vesting and minimum benefit (and contribution) requirements must be satisfied without considering employer payments of FICA taxes or contributions or benefits made or received under any other federal or state law.⁶

For any top-heavy plan year, a plan generally must provide a minimum benefit or contribution for each non-key employee who is a participant.¹ Integration (i.e., permitted disparity) must be disregarded for purposes of determining a minimum benefit or contribution.Defined benefit plans. A top-heavy defined benefit plan generally must provide an accrued benefit derived from employer contributions for each non-key employee participant that, when expressed as an annual retirement benefit, is not less than the participant’s average compensation multiplied by the lesser of 2 percent for each year of service with the employer or 20 percent.²

Years of service are the same as the “years of service” taken into account for the ordinary vesting rules (Q 3869), but years of service in which non-top-heavy plan years end are not counted for this purpose; years in which no key employee or former key employee benefits under the plan also are not counted.³

Average compensation is a participant’s average annual compensation for the period of consecutive years (not exceeding five) during which the participant had the greatest aggregate compensation from the employer.⁴ Compensation for any year that is not a year of service is disregarded.⁵

Similarly, unless the plan provides otherwise, compensation (Q 3867) for any year beginning after a plan has ceased forever to be top-heavy, is not counted.⁶ Annual retirement benefit means a benefit payable annually in the form of a single life annuity (with no ancillary benefits) beginning at the normal retirement age under the plan.⁷ If a benefit other than a single life annuity without ancillary benefits is provided, the employee must receive an amount that is the actuarial equivalent of a single life annuity commencing at normal retirement age. Similarly, if the benefit starts at a date other than normal retirement age, the employee must receive an amount that is at least the actuarial equivalent of the minimum single life annuity benefit starting at normal retirement age.⁸

For the application of the minimum benefit requirement to a defined benefit plan funded exclusively by level premium insurance contracts, see Treasury Regulation Section 1.416-1.⁹

Collective bargaining units. The minimum contribution and minimum benefit requirements do not apply in the case of any employee covered by a collective bargaining agreement if there is evidence that retirement benefits were the subject of good faith bargaining.¹⁰

For any top-heavy plan year, a plan generally must provide a minimum benefit or contribution for each non-key employee who is a participant.¹ Integration (i.e., permitted disparity) must be disregarded for purposes of determining a minimum benefit or contribution.

Defined contribution plans. For each plan year in which a defined contribution plan or simplified employee pension plan is top-heavy, employer contributions and forfeitures allocated to the account of each non-key employee participant must not be less than the amount that is calculated by multiplying the participant’s compensation by the lesser of 3 percent or the percentage that is the highest contribution rate made for a key employee.²

For purposes of determining the highest contribution rate received by a key employee, employer contributions and forfeitures made on behalf of each key employee under the plan or, if the plan is part of a required aggregation group (Q 3917), all defined contribution plans included in the group, are divided by total compensation for the year (but not more than $345,000 in 2024, $330,000 in 2023 $305,000 in 2022, $290,000 in 2021 and $285,000 in 2020).³

Although employer contributions attributable to salary reduction or similar arrangements may not be disregarded when calculating the minimum contribution requirement for a top-heavy defined contribution plan, these contributions may not be used to satisfy the top-heavy minimum contribution requirement.⁴ Non-elective contributions and employer matching contributions may be used to satisfy the minimum contribution requirement, but such amounts generally cannot then be used in the ACP or ADP test (Q 3802, Q 3804).⁵ For application of the minimum contribution requirement in the case of a plan that has received a waiver of the minimum funding requirements, see Treasury Regulation Section 1.416-1.⁶

If a top-heavy defined contribution plan required to be included in an aggregation group (Q 3917) with a discriminatory defined benefit plan enables that defined benefit plan to satisfy the nondiscrimination requirements of IRC Sections 401(a)(4) and 410, the minimum contribution is 3 percent of the participant’s compensation, and the highest contribution rate for key employees is disregarded.⁷

For any top-heavy plan year, a plan generally must provide a minimum benefit or contribution for each non-key employee who is a participant.¹ Integration (i.e., permitted disparity) must be disregarded for purposes of determining a minimum benefit or contribution. Defined benefit and defined contribution plans. Although an employer that maintains both a top-heavy defined benefit plan and a top-heavy defined contribution plan is not required by the top-heavy rules to provide a non-key employee who participates in both plans with a minimum contribution and a minimum benefit, the non-key employee may not receive less under the combined plans than he or she would if he or she participated in only one of the plans.² The regulations provide four safe harbor rules a plan may use to determine the minimum an employee must receive.³

Collective bargaining units. The minimum contribution and minimum benefit requirements do not apply in the case of any employee covered by a collective bargaining agreement if there is evidence that retirement benefits were the subject of good faith bargaining.⁴

A. Permanence. A qualified plan must be a permanent program. Thus, although an employer may reserve the right to amend or terminate the plan, the abandonment of the plan for any reason other than business necessity within a few years after its establishment will be evidence that the plan from its inception was not a bona fide program for the exclusive benefit of employees in general. This will especially be true if, for example, a pension plan is abandoned soon after pensions have been funded for the highly-paid or stockholder employees. The permanency of a plan will be indicated by all of the surrounding facts and circumstances, including the likelihood of the employer’s ability to continue contributions as provided under the plan. In the case of a profit-sharing plan, other than a profit-sharing plan which covers employees and owner-employees (see section 401(d)(2)(B)), it is not necessary that the employer contribute every year or that he contribute the same amount or contribute in accordance with the same ratio every year. However, merely making a single or occasional contribution out of profits for employees does not establish a plan of profit-sharing. Since 1985, plan sponsors no longer must have profits to make contributions to profit-sharing plans.¹ In the event a plan is abandoned, the employer should promptly notify the district director, stating the circumstances which led to the discontinuance of the plan. ²

B. Benefits after merger. The plan must provide that in the case of any merger or consolidation with, or transfer of assets or liabilities to, any other plan, each participant in the plan would (if the plan then terminated) receive a benefit immediately after the merger, consolidation, or transfer which is equal to or greater than the benefit the participant would have been entitled to receive immediately before the merger, consolidation, or transfer (if the plan had then terminated).³ (This requirement does not apply to certain multiemployer plans.) Shifting assets between funding media used for a single plan (e.g., between trusts and annuity contracts) is not a transfer of assets or liabilities.⁴

C. Early retirement benefit. If a plan provides for payment of an early retirement benefit, a vested participant who terminates employment after having satisfied the service requirements, but not the age requirement for the early benefit, must be entitled, upon satisfaction of the age requirement, to receive a benefit not less than the benefit to which the participant would be entitled at normal retirement age, actuarially reduced in accordance with reasonable actuarial assumptions.⁵ In the case of a defined contribution plan, the employee, upon reaching early retirement age following termination after having satisfied service requirements, must be entitled to receive a benefit equal in value to the vested portion of his account balance at early retirement age.⁶

D. Social Security offset. The plan must not permit benefits to be reduced by reason of any increase in Social Security benefit levels or wage base occurring (1) after separation from service, in the case of a participant who has separated from service with nonforfeitable rights to benefits, or (if earlier) (2) after first receipt of benefits, in the case of a participant or beneficiary who is receiving benefits under the plan.⁷ This requirement also applies to plans that supplement benefits provided under state or federal laws other than the Social Security Act, such as the Railroad Retirement Act of 1937.⁸

E. Withdrawal of employee contributions. The plan must preclude forfeitures of accrued benefits derived from employer contributions (whether forfeitable or nonforfeitable) solely because a benefit derived from the participant’s contributions is voluntarily withdrawn by him after the participant has a nonforfeitable right to 50 percent of the accrued benefit derived from employer contributions.⁹

F. Compensation. Generally, a plan will not be qualified unless (for the purpose of any of the qualification rules) not more than $345,000 (for 2024) of annual compensation of any employee is taken into account under the plan for any plan year.¹⁰ (See Appendix E for the amounts for earlier plan years.) This amount is indexed for inflation in increments of $5,000.¹¹

“Employee” generally includes any individual who performs services for a person or entity that has the right to control and direct the individual’s work, not only as to the result to be accomplished but also as to the details and means by which the result is accomplished.¹ These individuals are referred to as common law employees. The U.S. Supreme Court has set forth a 20-factor test for determining whether an individual is a common law employee.² Self-employed individuals, including sole proprietors, partners, and members of LLCs taxed as partnerships operating trades, businesses, or professions, are treated as employees for purposes of participating in qualified plans even though they clearly are not common law employees (Q 3932).³To prevent abuses of the tax advantages of qualified retirement plans through the manipulation of separate employer entities, the IRC provides several special rules that generally must be applied when testing plan qualification, as follows:

(1)    All employees of all corporations that are members of a controlled group of corporations and all employees of trades or businesses under common control must be treated as employed by a single employer (Q 8964).⁴

(2)    All employees of the members of an affiliated service group must be treated as employed by a single employer (Q 3935).⁵

(3)    Leased employees must be treated as employees (Q 3929).⁶

The aggregation rules for controlled groups and trades and businesses under common control also appear in ERISA; the aggregation rules for affiliated service groups do not. Except in the case of employees of an affiliated service group or certain leased employees, employees of a partnership need not be treated as employees of any partner who does not own more than a 50 percent interest in the capital or profits of the partnership.⁷

For the IRC’s qualification requirements, an employer generally treats any individual who is a leased employee as though that individual were the employer’s own employee. To the extent that contributions or benefits provided for a leased employee by the organization from which the employee is leased are attributable to services performed for the employer, these contributions or benefits are treated as if they were provided by the employer under a qualified plan.¹ These two requirements have the effect of requiring most leased employees who have met a recipient plan’s eligibility provisions to be included in the recipient’s plan as an employee of the recipient.A leased employee is an individual who is not an employee of the recipient employer and who performs services for a recipient employer, if (1) the individual’s services are provided to the recipient under one or more agreements with a leasing organization, (2) the individual has performed services for the recipient or related employer on a substantially full-time basis for a period of at least one year, and (3) the services are performed under the primary direction or control of the recipient employer.² For purposes of this definition, the term employee means a common law employee as determined under the 20 factor test set forth in Nationwide Mutual Ins. Co. v. Darden,³ which must be applied before it can be determined whether an individual meets the definition of a “leased employee.”⁴

The fact that an individual is a leased employee does not automatically mean he or she must be a participant in a plan maintained by the employer. A plan may exclude a leased employee from participation in the plan when the plan can satisfy coverage and nondiscrimination testing by including the leased employee with no benefits or the benefits provided by the leasing company plan. At least two circuit courts have held that ERISA does not per se require the inclusion of leased employees in an employer’s plan.⁵ In addition, the IRS addressed this issue in Notice 84-11,⁶ stating that leased employees should be treated as employees, but the plan’s failure to include them as participants in the plan does not result in disqualification of the plan. Despite its issuance prior to TRA ’86, Notice 84-11 was cited favorably in Bronk v. Mountain States Tel. & Tel., Inc.,⁷ as controlling authority on this issue.

The determination of whether services are performed under the primary direction or control of the recipient is based on the facts and circumstances. A finding will be made if the service recipient exercises the majority of direction and control over the individual; for example, whether the individual is required to comply with the recipient’s instructions as to when, where, and how the services are to be performed; whether the services will be performed by a particular person; whether the individual is subject to the recipient’s supervision; and whether the services must be performed in a particular order or sequence set by the recipient.

The recipient may be a single employer or a group consisting of employers required to be aggregated under the controlled group, common control, or affiliated service group rules (Q 8964, Q 3935).⁸ Employers are related if a loss on a sale of property between them would be disallowed as a deduction under IRC Sections 267 or 707(b) or they are members of the same controlled group of corporations, using a 50 percent rather than 80 percent ownership test.⁹

Safe Harbor

Even though an individual is a leased employee, he or she may be disregarded by the employer for purposes of determining qualification if the individual is covered by a qualified money purchase pension plan maintained by the leasing organization and the following requirements are satisfied:

(1)    The plan provides for employer contributions by the leasing organization at a nonintegrated rate which is not less than 10 percent.

(2)    The plan provides for immediate participation on the first day an individual becomes an employee of the leasing organization unless (x) the individual’s compensation from the leasing organization in each plan year during the four year period ending with the plan year is less than $1,000, or (y) the individual performs substantially all of his or her services for the leasing organization.

(3)    The plan provides for full and immediate vesting of all contributions under the plan.

(4)    Leased employees do not constitute more than 20 percent of the recipient’s nonhighly compensated work force.¹⁰

This safe harbor applies only for purposes of the leased employee provision; it does not permit an employer to disregard a common law employee who otherwise meets the definition of a leased employee.¹¹

A recipient’s nonhighly compensated work force is the aggregate number of individuals who are not highly compensated (Q 3930) but who are common law employees of the recipient and have performed services for the recipient on a substantially full-time period of at least one year or who are leased employees with respect to the recipient.¹²

A money purchase pension plan of a leasing organization is not qualified if it covers any individuals who are leased by the leasing organization to the recipient but who are not themselves employees of the leasing organization. That is because the plan would not meet the “exclusive benefit” rule (Q 3839).¹³

Status as a highly compensated employee is determined by focusing on the determination year (i.e., the plan year for which the determination is being made) and the immediately preceding 12-month period (the “look-back” year).

An employee is a highly compensated active employee with respect to a plan year (i.e., the determination year) if the employee (1) was a 5 percent owner, as defined for top-heavy purposes (Q 3931), at any time during either the determination year or look-back year, or (2) received compensation for the preceding year in excess of $155,000 in 2024 from the employer.¹

The compensation element of this determination can be limited to the top 20 percent of employees ranked by compensation (the “top-paid group”).² The income threshold ($155,000 for 2024) is indexed at the same time and in the same manner as the Section 415 defined benefit dollar limitation.³

The applicable dollar amount for a particular determination or look-back year is the dollar amount for the calendar year in which the determination year or look-back year begins.⁴

Employers may identify the employees who are highly compensated employees under IRC Section 414(q) using the same snapshot testing that is used for the nondiscrimination requirements (i.e., test results for a single day during the plan year, provided that that day is representative of the employer’s workforce and the plan’s coverage throughout the plan year).⁵

The IRS has stated that a fiscal year plan may make a calendar year data election. If the election is made, the calendar year beginning with or within the look-back year will be treated as the employer’s look-back year for purposes of determining whether an individual is a highly compensated employee on account of his or her compensation. This election will not apply in determining whether a 5 percent owner is highly compensated. The effect of this election is that even though an employer maintains a plan on a fiscal year basis, it uses calendar year data. Once made, the election applies for all subsequent years unless changed by the employer.⁶

---

Planning Point: For a plan that is maintained on a fiscal year basis, making a calendar year data election can simplify plan administration. Calendar year compensation information is easily available because Form W-2 requires the use of calendar year information. Using fiscal year information would require the employer to make a special calculation to determine if an individual is highly compensated.

---

Top-Paid Group

An alternative way to determine highly compensated employees is to make a “top-paid group” election. The top-paid group election must be made in the plan document. The top-paid group of employees for a year is the group of employees in the top 20 percent, ranked on the basis of compensation paid for the year.⁷

Once made, a top-paid group election remains in effect until the employer changes it via plan amendment.⁸ Former employees are not included in the top-paid group. Also, employees who are excluded under the collective bargaining agreement exclusion in determining the number of employees in the top-paid group also are excluded for purposes of identifying the members of the top-paid group.⁹

In determining the number of employees in the top-paid group (but not for the purpose of identifying the particular employees in the group), the following employees may be excluded:¹⁰

(1)    employees with less than six months of service, including any service in the immediately preceding year;

(2)    employees who normally work fewer than 17½ hours per week, if certain requirements are met;¹¹

(3)    employees who normally work during not more than six months in any year, determined on the basis of the facts and circumstances as evidenced by the employer’s customary experience in the years preceding the determination year;¹²

(4)    employees under the age of 21 at the end of the year; and

(5)    employees covered by a collective bargaining agreement if 90 percent or more of the employees of the employer are covered under the agreement and the plan being tested covers only employees who are not covered under the agreement.¹³

An employer may elect to use a shorter period of service, smaller number of hours or months, or lower age than those specified above (including no age or service requirement exclusion).¹⁴ Also, an employer may elect not to exclude members under the collective bargaining exclusion.¹⁵

No special notification or filing of a top-paid group election or a calendar year data election is required, although certain plan amendments may be necessary to incorporate a definition of highly compensated employees that reflects the election.¹⁶ Furthermore, a consistency requirement states generally that an election made by an employer operating more than one plan must apply consistently to all plans of the employer that begin with or within the same calendar year.¹⁷

Nonresident aliens who receive no earned income from sources within the United States are disregarded for all purposes in determining the identity of highly compensated employees.¹⁸ An employer may adopt any rounding or tie-breaking method that is reasonable, nondiscriminatory, and uniformly and consistently applied.¹⁹ An employee who is highly compensated as a result of meeting two or more of the tests above is not disregarded for the purpose of applying any of those tests to other individuals.²⁰

Compensation is the compensation received by the participant from the employer for the year, including elective or salary reduction contributions to a cafeteria plan, cash or deferred arrangement, or a tax sheltered annuity.²¹

A highly compensated former employee for a determination year is any employee who had a separation year prior to the determination year and was a highly compensated active employee for either his or her separation year or any determination year ending on or after his or her 55th birthday.²²

A separation year is any year during which the employee separates from service with the employer. For purposes of this rule, an employee who performs no services for the employer during a determination year is treated as having separated from service with the employer in the year that he or she last performed services for the employer. An employee will be deemed to have a separation year if, in a determination year prior to attainment of age 55, the employee receives compensation in an amount less than 50 percent of his or her average annual compensation for the three consecutive calendar years preceding the determination year in which the employee received the greatest amount of compensation from the employer (or the total period of the employee’s service with the employer, if less).

Because an employee who is deemed to have a separation is still performing services for the employer during the determination year, the employee is treated as an active employee and the deemed separation year is relevant only for purposes of determining whether the employee will be a highly compensated former employee after he or she actually separates from service. An employee with a deemed separation year will not be treated as a highly compensated former employee by reason of the deemed separation year if the employee later has a significant increase in services and compensation and, thus, is deemed to have a resumption of employment.²³

The controlled group, common control, and affiliated service group aggregation rules, as well as the employee leasing provisions, are applied before applying the highly compensated employee rules (Q 3929, Q 8964, and Q 3935).²⁴ The entity aggregation rules are not taken into account for purposes of determining who is a 5 percent owner. The separate lines of business rules also are not applicable in determining the highly compensated group.²⁵

A key employee for purposes of the top-heavy rules is any employee or, in some cases, a former or deceased employee who, at any time during the plan year containing the determination date for the plan year to be tested, is:

(1)    an officer of the employer whose annual compensation from the employer exceeds $220,000 for 2024 ($215,000 for 2023, $200,000 for 2022, $185,000 for 2020-2021,¹ this amount is indexed for inflation in increments of $5,000);

(2)    a more-than-5 percent owner of the employer; or

(3)    a more-than-1 percent owner of the employer having annual compensation from the employer for a plan year in excess of $155,000; this amount is not indexed for inflation.² (As to when the determination date occurs, see Q 3917.)

The determination as to whether an individual is an officer is made on the basis of all facts and circumstances; job titles are disregarded. An officer is an administrative executive who is in regular and continuous service, not a nominal officer whose administrative duties are limited to special and single transactions. A partner of a partnership will not be treated as an officer for purposes of the key employee test merely because he owns a capital or profits interest in the partnership, exercises his voting rights as a partner, and may, for limited purposes, be authorized and does in fact act as an agent of the partnership.³ Unincorporated associations, including partnerships and sole proprietorships, may have officers.⁴

In any case, the number of individuals treated as key employees because of their officer status is limited to the greater of three individuals or 10 percent of all employees, but in any event, not more than 50.⁵

Those employees who can be excluded when determining the number of employees in the top-paid group for purposes of identifying an employer’s highly compensated employees (Q 3930) also can be disregarded in determining the number of officers to be taken into account in identifying key employees.⁶ It is unclear how ties in compensation should be resolved. Whether an individual is a key employee because of his or her officer status is determined without regard to whether the individual is a key employee for any other reason.⁷

An individual owns more than 5 percent of a corporate employer if the individual owns more than 5 percent of the outstanding stock of the corporation by value or stock possessing more than 5 percent of the total combined voting power of all stock of the corporation. In determining stock ownership, the attribution rules of IRC Section 318 apply, but stock is attributed from a corporation if a 5 percent rather than 50 percent ownership test is met. Only ownership in the particular employer is considered; the controlled group, common control, and affiliated service group aggregation rules of IRC Section 414 are disregarded. An individual owns more than 5 percent of a noncorporate employer if he or she owns more than 5 percent of the capital or profits interest in that employer. Rules similar to the attribution rules of IRC Section 318 apply for purposes of determining ownership in a noncorporate employer.⁸ The aggregation rules of IRC Section 414 are disregarded.⁹

The rules discussed in the previous paragraph also apply to determine whether an individual is a more-than-1-percent owner of the employer.¹⁰ All employers who are under common control or who are members of a controlled or affiliated service group (Q 8964, Q 3935) are treated as one employer for the purpose of determining whether a more-than-1-percent owner has annual compensation from the employer in excess of $215,000 (as indexed for 2023).

Compensation, for purposes of identifying key employees generally, is the compensation considered for purposes of the IRC Section 415 limits on contributions and benefits. Any elective or salary reductions contributions made on behalf of an employee to a 401(k) cash or deferred plan, simplified employee pension, 403(b) tax sheltered annuity, or cafeteria plan are included as compensation for purposes of IRC Section 415.¹¹

For purposes of determining an employee’s ownership in the employer, the attribution rules of IRC Section 318 and the aggregation rules of IRC Section 414 apply. If two employees have the same ownership interest in the employer, the employee who has the greater annual compensation from that employer will be treated as owning the larger interest. ¹²

The terms employee and key employee include their respective beneficiaries.¹³

The term key employee is applied under various provisions of the IRC (e.g., IRC Sections 401(h), 415(l), and 419A). For these purposes, the term does not include any officer or employee covered by a governmental plan.¹⁴ Thus, the separate accounting and nondiscrimination rules under those provisions do not apply to employees covered by governmental plans.

An owner-employee is an employee (Q 3827) who owns the entire interest in an unincorporated trade or business or, in the case of a partnership, owns more than 10 percent of either the capital interest or the profits interest in the partnership.¹ Even if a partnership agreement does not specify a more than 10 percent interest in profits for any partner, if the formula for dividing profits (e.g., based on a partner’s earnings productivity during the year) in operation produced a distribution at the end of the year of more than 10 percent of profits to a partner, the Tax Court has ruled that he or she is an owner-employee for the year.²An individual who owns the entire interest in an unincorporated trade or business is treated as his or her own employer.³ Thus, a proprietor or sole practitioner who has earned income (Q 3827) can establish a qualified plan under which he or she is both employer and employee.

A partnership is treated as the employer of each partner who is an employee (Q 3827).⁴ Thus, partners individually cannot establish a qualified plan for a firm or solely for themselves, but the partnership can establish a plan in which the partners can participate.

Persons who are shareholder-employees in professional corporations or associations or in business corporations (including S corporations) are not self-employed individuals. These people participate in a qualified plan of the corporation as regular employees of the corporation.⁵ This is true even of a shareholder-employee who is sole owner of the corporation. S corporation pass-through income may not be treated as self-employment earnings for purposes of a Keogh plan deduction, even where the shareholder performed services for the corporation.⁶

A common law employee is an employee under common law rules, as distinguished from a self-employed individual who is considered an employee only for qualified plan purposes. An individual generally is considered an employee under common law rules if the person or organization for whom the individual performs services has the right to control and direct his or her work not only as to the result to be accomplished, but also as to the details and means by which the result is accomplished (Q 3928).⁷

The common law rules also apply generally in determining whether an individual is an employee for Social Security purposes. Ordinarily, therefore, an individual who is an employee under Social Security is a common law employee for self-employed plan purposes.

A person’s status for self-employed plan purposes is determined by the definition of employee under the Social Security law, irrespective of whether or how the person’s earnings are covered under Social Security.

Thus, if a person is an employee under the common law rules, it is immaterial that his or her earnings are treated as self-employment income under the Social Security law. For example, a minister or other clergy who is employed by a congregation on a salaried basis is a common law employee, and not a self-employed individual, even though for Social Security purposes the person’s compensation is treated as net earnings from self-employment. Amounts received by the minister directly from members of the congregation, such as fees for performing marriages, baptisms, or other personal services, represent earnings from self-employment.

Full-time life insurance salespersons are treated as common law employees for both Social Security and qualified retirement plan purposes even though, under the common law rules, they are self-employed. This is because of special statutory provisions in the Social Security Act and the IRC. Thus, a full-time life insurance salesperson under the Social Security law is prohibited from establishing a qualified plan for himself or for herself.⁸ Depending on the salesperson’s contractual arrangements, he or she may be considered a self-employed individual for some sales or services, allowing them to establish a qualified plan. These provisions do not appear to apply to general agents and most general lines insurance agents and brokers. As a result, based on their situation, they may be considered self-employed individuals and are eligible to establish qualified plans for themselves and their employees.

Attorneys with a law firm, depending on the circumstances, can either be self-employed or have the status of an “employee” of the firm.⁹

An individual may participate in a qualified plan as a self-employed person even though the individual performs work as a common law employee for another employer. For example, an attorney who is a common law employee of a corporation and who in the evenings maintains an office in which he or she practices law is eligible to establish a plan as a self-employed person with respect to the law practice.

An individual may be self-employed with respect to some services the individual sells to a business even though he or she also provides other services to the same business as an employee. In either case, the individual may participate in a qualified plan as a self-employed person with respect to his or her self-employed earnings, even though the employer maintains a qualified plan under which the individual is covered as a common law employee.¹⁰ A tenured university professor who conducted seminars in a separate capacity at the university with which he was employed was determined by the Tax Court to be self-employed, despite objections by the IRS. As a result, he was permitted to establish a Keogh plan with amounts earned from his self-employment.¹¹

The term controlled group is used to determine who makes up the group of employees that will be subject to the IRC’s coverage, nondiscrimination testing, and most qualification requirements that apply to qualified retirement plans. All employees of a single employer generally are included in this testing. The controlled group rules aggregate several entities (e.g., partnerships, sole proprietorships, and corporations) into a single employer for purposes of meeting various qualification requirements of the IRC. All employees of a group of employers that are members of a controlled group of corporations or, in the case of partnerships and proprietorships, are under common control will be treated as employed by a single employer.¹ In general, the determination of whether a group is a controlled group of corporations or under common control is based on stock ownership by value or voting power.A controlled group may be a parent-subsidiary controlled group, a brother-sister controlled group, or a combined group.²

A parent-subsidiary controlled group is composed of one or more chains of subsidiary corporations connected through stock ownership with a common parent corporation. A parent-subsidiary group exists if at least 80 percent of the stock of each subsidiary corporation is owned by one or more of the other corporations in the group and the parent corporation owns at least 80 percent of the stock of at least one of the subsidiary corporations. When determining whether a parent owns 80 percent of the stock of a subsidiary corporation, all stock of that corporation owned directly by other subsidiaries is disregarded.

A brother-sister controlled group consists of two or more corporations in which five or fewer persons, individuals, estates, or trusts own stock consisting of 80 percent or more of each corporation and more than 50 percent of each corporation when taking into account each stockholder’s interest only to the extent he or she has identical interests in each corporation. For purposes of the 80 percent test, a stockholder’s interest is considered only if he or she owns some interest in each corporation of the group.³

A combined group consists of three or more corporations, each of which is a member of a parent-subsidiary group or a brother-sister group and one of which is both a parent of a parent-subsidiary group and a member of a brother-sister group.⁴

Special rules apply for determining stock ownership, including special constructive ownership rules, when determining the existence of a controlled group.⁵ Community property rules, where present, also apply.⁶ For purposes of qualification, the test for a controlled group is strictly mechanical; once the existence of a group is established, aggregation of employees is required and will not be negated by showing that the controlled group and plans were not created or manipulated for the purpose of avoiding the qualification requirements.⁷

Under the regulations, trades or businesses are under common control if they constitute a parent-subsidiary group of trades or businesses, a brother-sister group of trades or businesses, or a combined group of trades or businesses. The existence of these groups is determined under rules similar to those discussed in Q 8964 for controlled groups of corporations. “Trades or businesses” include sole proprietorships, partnerships, estates, trusts, and corporations.¹

For purposes of certain qualification requirements, as well as for the vesting requirements, top-heavy rules, Section 415 limits and the requirements for SEPs, all employees of the members of an affiliated service group generally are treated as employed by a single employer.¹An affiliated service group is a group consisting of a service organization (referred to as “FSO” for “first service organization”) and:

(1)    an additional service organization (referred to as an “A” organization) that is a shareholder or partner in an FSO and that regularly performs services for the FSO or that is regularly associated with the FSO in the performance of services for third parties, or

(2)    an organization (referred to as a “B” organization) if a significant portion of the organization’s business is the performance of services for an FSO or for an A organization (or both) and the services are of a type historically performed by employees in the service field of the FSO or A organization; and

(3)    10 percent or more (in the aggregate) of the interests in the B organization is held by highly compensated employees of an FSO or an A organization or certain common owners.²

The term affiliated service group also includes a group consisting of (1) an organization the principal business of which is performing management functions for another organization on a regular and continuing basis, and (2) the organization for which the functions are performed.³

An organization is a service organization if its principal business is the performance of services in one of the fields enumerated in the regulations (e.g., health, law, etc.) or if capital is not a material income-producing factor for the organization.⁴

The performance of services for a first service organization or for an A organization (or both) will be assumed to constitute a significant portion of a B organization’s business if 10 percent or more of its total gross receipts from all sources during the current year, or two preceding years, was derived from performing services for such organization or organizations. It will be assumed that the performance of services for such organization or organizations is not a significant portion of a B organization’s business if less than 5 percent of its gross receipts derived from performing services during the current year and two preceding years was derived from performing services for such organizations.⁵ Services are of a type historically performed by employees in a particular field if it was not unusual for the services to be performed by employees of organizations in that service field in the United States on December 13, 1980.⁶

The principles of the constructive stock ownership rules of IRC Section 318(a) apply. Thus, ownership generally will be attributed to an individual from the individual’s spouse, children, grandchildren and parents, between a partner and the partnership, between a trust or estate and its beneficiaries, and between a corporation and a more-than-50 percent shareholder, including a corporate shareholder.⁷

Two or more affiliated service groups will not be aggregated merely because an organization is an A organization or a B organization with respect to each affiliated service group. All organizations that are A organizations or B organizations with respect to a single FSO, together with that FSO, must be treated as a single affiliated service group.⁸

Taxpayers may rely on the proposed regulations covering service-type affiliated groups until final rules are published. Examples explaining the tests for a service-type affiliated service group can be found in Revenue Ruling 81-105.⁹

Yes. The IRS may retroactively revoke a plan’s determination letter by notice to the taxpayer if:

(1)    there has been a change in the applicable law;

(2)    the organization omitted or misstated a material fact;

(3)    the organization has operated in a manner materially different from that originally represented; or

(4)    in the case of organizations to which IRC Section 503 applies (which includes qualified plan trusts), the organization engaged in a prohibited transaction with the purpose of diverting corpus or income of the organization from its exempt purpose, and such transaction involved a substantial part of the corpus or income of such organization.¹

The IRS has broad discretion to determine the extent to which rulings and regulations will be given retroactive effect.² The wide array of correction procedures established by the IRS in the past decade offers a choice of less severe remedies, however, and suggests that the IRS is reluctant to disqualify plans except under the most egregious circumstances.

The IRS also may retroactively correct its own mistaken application of law, even where a taxpayer may have relied to the taxpayer’s detriment on the IRS’s mistake.³ Concerning determination letters, the IRS has voluntarily limited its authority by the issuance of revenue procedures stating the standards by which the continuing effect of a determination letter will be judged. In substance, there are two standards.

First, if a published revenue ruling is issued that is applicable to a previously approved plan, to retain its qualified status the plan must be amended to conform with that ruling before the end (and effective at least as of the beginning) of the first plan year following the one in which the ruling was published.⁴ Thus, with respect to the approved plan, the revenue ruling is not given retroactive effect and becomes effective only at the beginning of the next plan year.⁵

Second, if no applicable published ruling affecting the qualification of the plan has intervened between the approval and the revocation of the approval, the revocation ordinarily will not have retroactive effect with respect to the taxpayer to whom the ruling was originally issued or to a taxpayer whose tax liability was directly involved in such ruling if:

(1)    there has been no misstatement or omission of material facts;

(2)    the facts subsequently developed are not materially different from the facts on which the ruling was based;

(3)    there has been no change in the applicable law;

(4)    the ruling originally was issued with respect to a prospective or proposed transaction; and

(5)    the taxpayer directly involved in the ruling acted in good faith in reliance on the ruling and a retroactive revocation would be to the taxpayer’s detriment.⁶