Compliance

IRS Proposes Rules for Dependent Care FSA Nondiscrimination Testing

The proposed guidance offers long-awaited clarity on how employers can satisfy testing requirements.

September 15, 2026

On August 11, 2026, the IRS issued proposed rules addressing nondiscrimination testing requirements for dependent care flexible spending accounts (FSAs). This marks the first set of regulatory guidance on the mechanics of nondiscrimination testing for dependent care FSAs. Importantly, the proposed rules do not create any new nondiscrimination requirements for dependent care FSAs; they simply clarify how decades-old statutory rules should be applied.

Employers with dependent care FSAs often struggle to pass nondiscrimination testing, especially the 55% average benefits test, because lower-paid employees are less likely to participate in these plans. The proposed rules would make the following key changes:

Clarify how the 55% average benefits test applies to dependent care FSAs, including the methodology for calculating average benefits for highly compensated employees (HCEs) and non-highly compensated employees (non-HCEs);
Allow employers to correct a failed 55% average benefits test by including excess benefits in HCEs’ gross income by the deadline for furnishing Form W-2 for the testing year (i.e., January 31 of the following year); and
Establish a clear safe harbor for passing the eligibility test through a percentage-based approach, rather than a facts and circumstances analysis.

While the rules have not been finalized, employers may rely on the proposed guidance for plan years beginning before final rules are issued.

Dependent Care FSAs

Internal Revenue Code (Code) Section 129 allows employers to provide dependent care assistance benefits for their employees on a tax-free basis. These benefit plans are referred to as dependent care FSAs or dependent care assistance programs. Most dependent care FSAs are structured so that employees make pretax contributions through a Code Section 125 cafeteria plan. Married employees who file a joint tax return and unmarried employees may contribute up to $7,500 each year to their dependent care FSAs. The annual limit for married employees who file separate tax returns is $3,750. These limits do not receive annual adjustments for inflation.

In general, benefits that an employee receives from their dependent care FSA are nontaxable if:

the expenses are for the care of one or more qualifying individuals (for example, a child under the age of 13); and
the employee incurs the expense in order to enable the employee (and the employee’s spouse, if applicable) to be gainfully employed.

Nondiscrimination Requirements

Code Section 129 imposes nondiscrimination requirements on dependent care FSAs to make sure they do not discriminate in favor of HCEs. An employee is generally an HCE if they are a more-than-5% owner at any time during the current or prior year, or if their prior-year compensation exceeded the applicable dollar threshold for that year ($160,000 for 2025 and 2026). In general, employers with dependent care FSAs have had difficulty with nondiscrimination testing, largely because non-HCEs tend to participate at lower rates, while HCEs are more likely to elect the maximum contribution.

Four Different Tests

To avoid adverse tax consequences for HCEs, a dependent care FSA must satisfy four nondiscrimination tests under Code Section 129. The proposed rules are intended to make the testing requirements clearer and easier to administer. The following chart describes each of these tests and summarizes the IRS’s proposed corresponding guidance:

Eligibility Test: Code Section 129(d)(3)
Statutory Description	Proposed Rules
A dependent care FSA cannot discriminate in favor of HCEs as to eligibility to participate. The following employees are excluded for testing purposes: Employees who have not attained age 21 and completed one year of service; and Collectively bargained employees who are not included in the dependent care assistance program.	This nondiscrimination test requires both that the employer's eligibility classification be reasonable and that the classification be nondiscriminatory in operation. Reasonable classifications generally include specified job categories, nature of compensation (salaried or hourly), geographic location and similar bona fide business criteria. Listing employees by name, or by criteria having substantially the same effect, is not a reasonable classification. A classification can establish nondiscriminatory operation in either of two ways: Facts-and-circumstances test: This test evaluates factors including the business justification for the classification, the percentage of the workforce covered, whether coverage is representative across salary ranges and how close the plan comes to the numerical safe harbor described below. In general, the greater the business justification for the classification, the broader the coverage under the plan, the more representative the classification is across salary ranges, and the smaller the difference between the plan's ratio percentage and the employer's safe harbor percentage, the more likely the classification is to be nondiscriminatory; or Numerical safe harbor: A dependent care FSA satisfies the safe harbor if its "ratio percentage" (the percentage of eligible non-HCEs compared to the percentage of eligible HCEs) is at or above the employer's "safe harbor percentage." The safe harbor percentage starts at 90% and is reduced by 0.75 percentage points for every whole percentage point by which the employer's non-HCE concentration percentage exceeds 60%. The non-HCE concentration percentage is the percentage of all the employer's employees who are non-HCEs. A classification that satisfies the safe harbor is treated as nondiscriminatory without the need to establish, based on all the relevant facts and circumstances, that the classification is nondiscriminatory.
Contributions and Benefits Test: Code Section 129(d)(2)
Statutory Description	Proposed Rules
The contributions or benefits provided under a dependent care FSA cannot discriminate in favor of HCEs.	To satisfy this qualitative test, a dependent care FSA cannot provide more favorable terms for HCEs than for other employees. However, a dependent care FSA that provides contributions and benefits on the same terms to all eligible employees satisfies this test, even if employees receive different amounts of contributions and benefits due to differing elections or utilization.
Owner Concentration Test: Code Section 129(d)(4)
Statutory Description	Proposed Rules
Not more than 25% of the amounts paid or incurred by the employer for dependent care assistance during the year may be provided for the class of individuals who are shareholders or owners (or their spouses or dependents), each of whom (on any day of the year) owns more than 5% of the stock or of the capital or profits interest in the employer.	The proposed rules restate the statutory requirement and do not provide additional guidance on this test.
55% Average Benefits Test: Code Section 129(d)(8)
Statutory Description	Proposed Rules
The average benefits provided to non-HCEs under the dependent care FSA must be at least 55% of the average benefits provided to HCEs. The following employees are excluded for testing purposes: Employees who have not attained age 21 and completed one year of service; Collectively bargained employees who are not included in the dependent care assistance program; and For benefits provided through a salary reduction agreement, employees whose compensation is less than $25,000.	The proposed rules provide a framework for applying this test. In general, the average benefits provided to a group of HCEs or non-HCEs for a plan year equals the total dollar amount of such contributions and benefits provided during the plan year to employees in that group, divided by the number of employees in that group to whom such contributions and benefits in a dollar amount greater than zero are provided during the plan year, via salary reduction or otherwise. For purposes of this calculation, an employee is taken into account in the denominator only if the employee is provided contributions and benefits under a dependent care FSA in an amount greater than zero during the plan year. Employees who were eligible but did not elect or receive any benefits are not included in the denominator. Compliance with this test is determined as of the last day of the plan year, taking into account any individual employed on any day of the plan year who is not an excluded employee and who was provided dependent care FSA benefits, via salary reduction or otherwise, on any day during the plan year.

Testing Failures

If a dependent care FSA fails nondiscrimination testing, the benefits provided to HCEs will be taxable, but benefits for non-HCEs will not be affected. To avoid tax issues, employers often test their dependent care FSAs early in the plan year and reduce HCEs’ pretax contributions, as necessary, to get the plan to pass by the end of the year.

The proposed rules would also provide a correction method for failures of the 55% average benefits test and the owner concentration test. If a dependent care FSA fails either of these tests, the plan may nonetheless be treated as satisfying that testing requirement if, on or before the deadline for furnishing Form W-2 for the year in which the benefits were provided, the employer includes the amount of excess benefits in the gross income of affected HCEs. For example, corrections for 2026 must be made no later than January 31, 2027, and included on HCEs’ 2026 Forms W-2.

In addition, the proposed rules would allow dependent care FSAs to allocate excess benefits for HCEs as follows:

Average Benefits Test: In general, if all HCEs have benefits in excess of the amount that would satisfy the 55% average benefits threshold, the excess benefit amount for each HCE is determined by reference to that threshold. If not all HCEs have benefits in excess of that amount, the employer would be permitted to allocate the excess benefit and required reduction among HCEs in any reasonable manner.
Owner Concentration Test: A similar allocation would be permitted when a dependent care FSA fails to satisfy the owner concentration test. In that case, the permitted concentration amount is subtracted from the benefit provided to participating shareholders or owners to determine the amount to be included in income. The permitted concentration amount is 25% of the total dependent care benefits provided by the employer to all participants during the year, divided by the number of participating shareholders or owners.

Action Items

The proposed rules may make it easier for dependent care FSAs to pass nondiscrimination testing, especially the 55% average benefits test. Employers who have not completed this year’s testing should check in with their vendors to confirm their methodology will reflect the new guidance. Employers whose plans have already failed may wish to retest under the proposed rules.

If you have questions, please reach out to your Hylant representative for further information. Don’t have one? Contact us here.

The above information does not constitute advice. Always contact your employee benefits broker or trusted advisor for insurance-related questions.

Related Insights

Changes to Pediatric Vaccine Recommendations: Impact on Health Plan Coverage

September 16, 2026

Department of Labor Releases Updated FMLA Forms

September 15, 2026