Last month, the Treasury Department and the IRS issued proposed regulations that provide guidance on the nondiscrimination testing rules for dependent care assistance programs (DCAPs) under Section 129 of the Internal Revenue Code (the “Code”).

In a welcome and long-awaited development, the proposed regulations clarify that, under the 55% average benefits test, only employees who receive dependent care assistance benefits are counted when running this test. The proposed regulations also establish a framework for testing eligibility and permit employers to correct certain failures through income inclusion and Form W-2 reporting.

The proposed regulations were issued with guidance on contributions to Code Section 128 Trump accounts (a new type of tax-advantaged savings account for children) because Code Section 128 incorporates several of Code Section 129’s nondiscrimination requirements. This post focuses on the Code Section 129 nondiscrimination rules applicable to DCAPs.

Background: Section 129 Nondiscrimination Requirements

By way of background, Code Section 129 allows an employee to exclude from gross income up to $7,500 per year ($3,750 for a married individual filing separately) of dependent care assistance provided under a DCAP, including amounts contributed through an employee’s pre-tax salary reduction under a cafeteria plan. A DCAP must satisfy the following four nondiscrimination requirements:

• Average benefits test: The average benefits provided to non-highly compensated employees (NHCEs) under all DCAPs maintained by the employer must be at least 55% of the average benefits provided to highly compensated employees (HCEs).
• Contributions and benefits test: Contributions and benefits may not discriminate in favor of HCEs or their dependents.
• Eligibility test: The DCAP must benefit employees in a classification that does not discriminate in favor of HCEs or their dependents.
• 25% Owner concentration test: No more than 25% of the dependent care assistance provided through the DCAP during the year may be provided to individuals who own more than 5% of the employer (including any ownership through their spouses and dependents).

A failure generally causes HCEs, but not NHCEs, to lose the Code Section 129 income tax exclusion. Although these statutory requirements have applied since Code Section 129 was enacted in 1981, employers have had little guidance on how to perform the testing.

55% Average Benefits Test Uses Actual Benefit Recipients

For some time, it has been unclear whether eligible employees who elect not to participate in a DCAP must be counted when calculating average benefits for the 55% test. The proposed regulations would resolve this question. An eligible employee who receives no benefit would not count as receiving a zero-dollar benefit. For each group of HCEs and NHCEs, the average benefit would equal the total dependent care assistance provided during the plan year divided by the number of employees in that group who received an amount greater than zero (an employee would be included if they receive any amount of benefit, even just $1).

Here’s an example: Assume the employer has a DCAP covering 15 HCEs and 15 NHCEs. Eleven HCEs and four NHCEs each elect $7,500, while the remaining employees elect nothing. The HCE average is $7,500 ($82,500 divided by 11 participating HCEs) and the NHCE average is also $7,500 ($30,000 divided by 4 participating NHCEs). The NHCE average is 100% of the HCE average, above the 55% threshold, so the DCAP passes the test even though a greater percentage of HCEs participate.

This approach should make the test easier to pass for many DCAPs, particularly where NHCE participation is low but participating NHCEs and HCEs elect similar amounts. It does not guarantee a passing result, however. A DCAP may still fail if NHCE participants elect materially lower amounts than HCE participants.

The test would be performed as of the last day of the plan year and would include any individual who was employed on any day during the plan year and received dependent care assistance during the year. For benefits provided through a salary reduction agreement, a plan may disregard employees whose compensation (as defined in Section 414(q)(4) of the Code) is less than $25,000. Employees who are not yet 21 years old and have not completed one year of service are disregarded, even if they received benefits under the DCAP.

Same Program Terms Satisfy the Contributions and Benefits Test

A DCAP would satisfy the contributions and benefits test if it provides benefits on the same terms to all eligible employees. For example, a DCAP does not discriminate as to contributions or benefits if NHCEs may elect the same salary reduction amount as HCEs. Different election amounts or utilization rates would not, by themselves, cause the DCAP to fail this test. However, utilization is still relevant under the separate 55% average benefits test, as explained above. It is also relevant for the 25% owner concentration test.

Eligibility Has a New Testing Framework

The eligibility classification determines which employees may participate in the DCAP. Before now, Code Section 129 has required that a DCAP benefit employees under a classification found by the Secretary of the Treasury not to be discriminatory, but there has been no regulatory framework detailing how to evaluate that classification. The proposed regulations would establish specific criteria.

• Reasonable Classification: An eligibility classification must be based on objective business criteria. Examples include job category, salaried or hourly status, and geographic location. A list of employees by name, or criteria that have substantially the same effect, would not be reasonable. An employee would be treated as eligible only if the employee has a meaningful opportunity to receive benefits, whether through salary reduction or otherwise.
• Nondiscriminatory Classification: A reasonable classification would then need to satisfy either a numerical safe harbor or a facts-and-circumstances test:
  • Numerical safe harbor: The percentage of NHCEs who are eligible would be compared with the percentage of HCEs who are eligible. The resulting ratio percentage would need to equal or exceed the employer’s safe harbor percentage, which starts at 90% and is reduced by 0.75 percentage points for each whole percentage point by which the NHCE concentration exceeds 60%.  
  • Facts and circumstances test: If the numerical safe harbor is not satisfied, the classification could still pass under a facts-and-circumstances test based on relevant considerations, including the business reason for the classification, the percentage of the employer’s workforce that is eligible, whether eligible employees are representative of the workforce across salary ranges, and how close the DCAP’s ratio percentage is to the safe-harbor percentage.

55% Average Benefits and 25% Owner Concentration Testing Failures Can Be Corrected After Year-End

Under current law, there is no prescribed correction method that allows an employer to cure a testing failure. Instead, if a DCAP fails a nondiscrimination requirement, the benefits provided to affected HCEs generally lose the Section 129 income tax exclusion.

The proposed regulations would change this result for failures of the 55% average benefits test and the 25% owner concentration test. If either test fails as of the last day of the plan year, the employer may correct the failure by including the excess amount in the income of the affected HCEs (for an average benefits failure) or principal shareholders or owners (for an owner concentration failure) and reporting it on Form W-2 by the applicable Form W-2 furnishing deadline, which is generally January 31st. The proposed regulations include formulas for determining the excess amount and permit reasonable allocation methods.

The correction method has an important limit. It would not apply to a failure of the contributions and benefits test or the eligibility test. Employers therefore should not abandon advance testing or assume that every year-end failure can be mitigated.

Other Clarifications

The proposed regulations would also clarify several other testing rules, including:

• The Code Section 414 controlled group, common control, affiliated service group, and other aggregation rules would apply in determining the employer’s employee population and benefits.
• Self-employed individuals described in Section 401(c)(1) of the Code would continue to be treated as employees for Section 129 purposes.
• Taxable benefits provided outside of a DCAP (for example, a direct taxable employer subsidy for childcare that is not part of the DCAP) would not be taken into account for purposes of the four nondiscrimination tests. This means only benefits provided under the DCAP itself are counted when performing the testing.

Effective Date and Next Steps

The proposed regulations will apply to plan years beginning on or after the date final regulations are published. However, employers may rely on the proposed regulations for plan years beginning before that date. For those curious, this means “Yes”—employers can adjust DCAP nondiscrimination testing and correction processes to reflect the proposed regulations now, instead of having to wait until the proposed rules are finalized. By the same token, employers should review their DCAPs in light of the new testing rules and confirm they comply with the updated eligibility, contributions and benefits test, and other clarified testing rules.