(a) General rule. A plan satisfies the average benefit percentage test of this section for a plan year if and only if the average benefit percentage of the plan for the plan year is at least 70 percent. A plan is deemed to satisfy this requirement if it satisfies paragraph (f) of this section for the plan year.

(b) Determination of average benefit percentage. The average benefit percentage of a plan for a plan year is the percentage determined by dividing the actual benefit percentage of the nonhighly compensated employees in plans in the testing group for the testing period that includes the plan year by the actual benefit percentage of the highly compensated employees in plans in the testing group for that testing period. See paragraph (d)(3)(ii) of this section for the definition of testing period.

(c) Determination of actual benefit percentage. The actual benefit percentage of a group of employees for a testing period is the average of the employee benefit percentages, calculated separately with respect to each of the employees in the group for the testing period. All nonexcludable employees of the employer are taken into account for this purpose, even if they are not benefiting under any plan that is taken into account.

(d) Determination of employee benefit percentages—(1) Overview. This paragraph (d) provides rules for determining employee benefit percentages. See paragraph (e) of this section for alternative methods for determining employee benefit percentages.

(2) Employee contributions and employee-provided benefits disregarded. Only employer-provided contributions and benefits are taken into account in determining employee benefit percentages. Therefore, employee contributions (including both employee contributions allocated to separate accounts and employee contributions not allocated to separate accounts), and benefits derived from such contributions, are not taken into account in determining employee benefit percentages.

(3) Plans and plan years taken into account—(i) Testing group. All plans included in the testing group under §1.410(b)-7(e)(1), and only those plans, are taken into account in determining an employee's employee benefit percentage.

(ii) Testing period. An employee's employee benefit percentage is determined on the basis of plan years ending with or within the same calendar year. These plan years are referred to in this section as the relevant plan years or, in the aggregate, as the testing period.

(4) Contributions or benefits basis. Employee benefit percentages may be determined on either a contributions or a benefits basis. Employee benefit percentages for any testing period must be determined on the same basis (contributions or benefits) for all plans in the testing group.

(5) Determination of employee benefit percentage—(i) General rule. The employee benefit percentage for an employee for a testing period is the rate that would be determined for that employee for purposes of applying the general test for nondiscrimination in §§1.401(a)(4)-2, 1.401(a)(4)-3, 1.401(a)(4)-8 or 1.401(a)(4)-9, if all the plans in the testing group were aggregated for purposes of section 410(b). Thus, if employee benefit percentages are determined on a contributions basis, each employee's employee benefit percentage is the aggregate normal allocation rate that would be determined for the employee under §1.401(a)(4)-9(b)(2)(ii)(A) (if the plans in the testing group include both defined benefit and defined contribution plans), the allocation rate that would be determined for the employee under §1.401(a)(4)-2(c)(2) (if the plans in the testing group include only defined contribution plans), or the equivalent normal allocation rate that would be determined for the employee under §1.401(a)(4)-8(c)(2) (if the plans in the testing group include only defined benefit plans). Similarly, if employee benefit percentages are determined on a benefits basis, each employee's employee benefit percentage is the aggregate normal accrual rate that would be determined for the employee under §1.401(a)(4)-9(b)(2)(ii)(B), the normal accrual rate that would be determined for the employee under §1.401(a)(4)-3(d), or the equivalent accrual rate that would be determined for the employee under §1.401(a)(4)-8(b)(2), depending on whether the plans in the testing group include both defined benefit and defined contribution plans, only defined benefit plans, or only defined contribution plans.

(ii) Plans with differing plan years. If not all the plans in the testing group share the same plan year, §1.410(b)-7(d)(5) would ordinarily prohibit them from being aggregated for purposes of section 410(b). In such a case, employee benefit percentages are determined by applying the rules of paragraph (d)(5)(i) of this section separately to each subset of plans in the testing group that share the same plan year (or the same accrual computation period) and aggregating the results for all plans in the testing group. Thus, an employee's employee benefit percentage is determined as the sum of these separate employee benefit percentages that are determined consistently for all the plans in the testing group (except for differences attributable solely to the differences in plan years).

(iii) Options and consistency requirements. In determining employee benefit percentages under this paragraph (d)(5), any optional or alternative methods or rules available for determining rates in §§1.401(a)(4)-2, 1.401(a)(4)-3, 1.401(a)(4)-8, or 1.401(a)(4)-9, whichever is applicable, may be applied. Thus, for example, employee benefit percentages may generally be calculated using any of the alternative methods of determining average annual compensation or plan year compensation under §1.401(a)(4)-12, and using any underlying definition of compensation that satisfies section 414(s). Except as otherwise specifically permitted, the determination of employee benefit percentages must be made on a consistent basis for all employees and for all plans in the testing group as required by §§1.401(a)(4)-2(c)(2)(vi), 1.401(a)(4)-3(d)(2)(i), 1.401(a)(4)-8(b)(2)(iv), 1.401(a)(4)-8(c)(2)(iv) or 1.401(a)(4)-9(b)(2)(iv).

(6) Permitted disparity—(i) In general. Permitted disparity may be imputed in determining employee benefit percentages as provided in §§1.401(a)(4)-2, 1.401(a)(4)-3, 1.401(a)(4)-8, or 1.401(a)(4)-9, whichever is applicable. When separate employee benefit percentages are determined for individual plans under paragraph (e)(2) of this section (or for subsets of plans that have the same plan year as described in paragraph (d)(5)(ii) of this section), permitted disparity may be imputed for an employee only in one individual plan (or subset of plans) and may not be imputed for the same employee in another individual plan (or subset of plans). However, if the same average annual compensation or plan year compensation is used to determine employee benefit percentages in more than one plan, the employee's employee benefit percentages for those plans may be summed prior to imputing permitted disparity.

(ii) Plans which may not use permitted disparity. Permitted disparity may be reflected in the determination of rates only to the extent that the plans for which rates are being determined are plans for which the permitted disparity of section 401(l) is available. Thus, for example, if a section 401(k) plan is included in the testing group and permitted disparity is imputed under §1.401(a)(4)-2(c)(iv), then employee benefit percentages are determined by first calculating an adjusted allocation rate (within the meaning of §1.401(a)(4)-7(b)(1)) without regard to the amount of allocations under the section 401(k) plan and adding to it the allocation rate for the section 401(k) plan. See §1.401(l)-1(a)(4) for a list of types of plans for which permitted disparity is not available.

(7) Requirements for certain plans providing early retirement benefits—(i) General rule. If any defined benefit plan in the testing group provides for early retirement benefits in addition to normal retirement benefits to any highly compensated employee, and the average actuarial reduction for any one of these benefits commencing in the five years prior to the plan's normal retirement age is less than four percent per year, then the aggregate most valuable allocation rate, equivalent most valuable allocation rate, aggregate most valuable accrual rate, or most valuable accrual rate must be substituted for the related normal rates in paragraph (d)(5) of this section.

(ii) Exception. Paragraph (d)(7)(i) of this section does not apply if early retirement benefits with average actuarial reductions described in that paragraph are currently available, within the meaning of §1.401(a)(4)-4(b), under plans in the testing group to a percentage of nonhighly compensated employees that is at least 70 percent of the percentage of highly compensated employees to whom these benefits are currently available.

(e) Additional optional rules—(1) Overview. This paragraph (e) contains various alternative methods for determining employee benefit percentages for a testing period.

(2) Determination of employee benefit percentages as the sum of separately determined rates—(i) In general. Employee benefit percentages may be determined as the sum of separately determined employee benefit percentages for each of the plans in the testing group that are aggregated under paragraphs (d)(5) (i) or (ii) of this section, provided that these employee benefit percentages are determined on a consistent basis for all of these plans pursuant to paragraph (d)(5)(iii) of this section.

(ii) Exception from consistency requirement. The consistency requirement of paragraph (e)(2)(i) of this section is not violated merely because employee benefit percentages are not determined in a consistent manner for all of the plans in the testing group and the inconsistencies in determination of rates among plans are described in paragraph (e)(2)(iii) of this section. The exception in this paragraph (e)(2)(ii) applies only if it is reasonable to believe that the inconsistencies do not result in an average benefit percentage that is significantly higher than the average benefit percentage that would be determined had employee benefit percentages been determined on a consistent basis pursuant to paragraph (d)(5)(iii) of this section.

(iii) Permitted inconsistencies. The following inconsistencies between plans are permitted under this paragraph (e)(2)—

(A) Use of different underlying definitions of section 414(s) compensation in the determination of rates;

(B) Use of different definitions of average annual compensation;

(C) Use of different testing ages;

(D) Use of different fresh-start dates;

(E) Use of different actuarial assumptions for normalization; or

(F) Disregard of actuarial increases after normal retirement age and QPSA charges without regard to any requirement for uniformity in the actuarial increases or QPSA charges.

(3) Determination of employee benefit percentages without regard to plans of another type—(i) General rule. Employee benefit percentages may be determined under plans of one type (i.e., defined benefit plans or defined contribution plans) by treating all plans of the other type (i.e., defined contribution plans or defined benefit plans, respectively) as if they were not part of the testing group, using the method provided in this paragraph (e)(3). If this method is used to determine whether a defined contribution plan satisfies the average benefit percentage test, employee benefit percentages under all defined contribution plans in the testing group must be determined on a contributions basis, and benefits under any defined benefit plans may not be included in the employee benefit percentage. Similarly, if this method is used to determine whether a defined benefit plan satisfies the average benefit percentage test, employee benefit percentages under all defined benefit plans in the testing group must be determined on a benefits basis, and allocations under any defined contribution plans may not be included in the employee benefit percentage.

(ii) Restriction on use of separate testing group determination method. A plan does not satisfy the average benefit percentage test using the method provided in this paragraph (e)(3) unless each of the plans in the testing group of the other type (i.e., defined benefit plan or defined contribution plan) than the plan being tested satisfies the average benefit test of §1.410(b)-2(b)(3) using the method in this paragraph (e)(3) or satisfies the ratio percentage test of §1.410(b)-2(b)(2).

(iii) Treatment of permitted disparity. Although under the general rule of this paragraph (e)(3) plans of another type are disregarded in determining employee benefit percentages, the permitted disparity used by those plans (including any permitted disparity that is used by those plans to satisfy §1.401(a)(4)-1(b)(2)) is nonetheless taken into account in determining the extent to which permitted disparity may be used in determining employee benefit percentages.

(iv) Example. The following example illustrates the rules of this paragraph (e)(3):

(4) Simplified method for determining employee benefit percentages for certain defined benefit plans—(i) In general. An employee's employee benefit percentage with respect to a plan may be determined under the simplified method of paragraph (e)(4)(ii) of this section, provided the following conditions are satisfied:

(A) The only plans included in the testing group are defined benefit plans, and employee benefit percentages under these plans are determined on a benefits basis.

(B) Employee benefit percentages under the plans in the testing group are not required to be determined by taking into account early retirement benefits under paragraph (d)(7) of this section.

(C) The plan is a safe harbor defined benefit plan described in §1.401(a)(4)-3(b).

(ii) Simplified method—(A) Section 401(l) plans. Under the simplified method of this paragraph (e)(4)(ii), an employee's employee benefit percentage with respect to a section 401(l) plan described in §1.401(a)(4)-3(b)(3) (i.e., a unit credit plan) may be deemed equal to the employee's excess benefit percentage or gross benefit percentage (as defined in §1.401(l)-1(c) (14) or (18), respectively), whichever is applicable under the plan's benefit formula in the plan year. In the case of a section 401(l) plan described in §1.401(a)(4)-3(b)(4) (i.e., a fractional accrual plan), an employee's employee benefit percentage with respect to that plan may be deemed equal to the rate at which the excess or gross benefit, whichever is applicable, accrues for the employee in the plan year, taking into account the plan's benefit formula and the employee's projected service at normal retirement age. The use of this simplified method will be treated as an imputation of permitted disparity. See paragraph (d)(6) of this section for a restriction on multiple use of permitted disparity.

(B) Other plans. Under the simplified method of this paragraph (e)(4)(ii), an employee's employee benefit percentage with respect to a plan described in §1.401(a)(4)-3(b)(3) that is not a section 401(l) plan and that is not imputing permitted disparity may be deemed equal to the employee's benefit rate in the plan year under the plan's benefit formula. In the case of a plan described in §1.401(a)(4)-3(b)(4) that is not a section 401(l) plan and that is not imputing permitted disparity, an employee's employee benefit percentage with respect to that plan may be deemed equal to the rate at which the benefit accrues for the employee in the plan year, taking into account the plan's benefit formula and an employee's projected service at normal retirement age.

(5) Three-year averaging period. An employee's employee benefit percentage may be determined for a testing period as the average of the employee's employee benefit percentages determined separately for the testing period and for the immediately preceding one or two testing periods (referred to in this section as an averaging period). Employee benefit percentages of a particular employee that are averaged together within an averaging period must be determined on a consistent basis for all testing periods within the averaging period.

(6) Alternative methods of determining compensation. Employee benefit percentages may be determined on the basis of any definition of compensation that satisfies §1.414(s)-1(d) (without regard to whether the definition satisfies §1.414(s)-1(d)(3)), provided that the same definition is used for all employees and it is reasonable to believe that the definition does not result in an average benefit percentage that is significantly higher than the average benefit percentage that would be determined had employee benefit percentages been determined using a definition of compensation that also satisfies §1.414(s)-1(d)(3).

(f) Special rule for certain collectively bargained plans. A plan (as determined without regard to the mandatory disaggregation rule of §1.410(b)-7(c)(5)) that benefits both collectively bargained employees and noncollectively bargained employees is deemed to satisfy the average benefit percentage test of this section if—

(1) The provisions of the plan applicable to each employee in the plan are identical to the provisions of the plan applicable to every other employee in the plan, including the plan benefit or allocation formula, any optional forms of benefit, any ancillary benefit, and any other right or feature under the plan, and

(2) The plan would satisfy the ratio percentage test of §1.410(b)-2(b)(2), if §§1.410(b)-6(d) and 1.410(b)-7(c)(5) (the excludable employee and mandatory disaggregation rules for collectively bargained and noncollectively bargained employees) did not apply.

[T.D. 8363, 56 FR 47646, Sept. 19, 1991; 57 FR 10817, 10954, Mar. 31, 1992, as amended by T.D. 8487, 58 FR 46840, Sept. 3, 1993]