Employee Benefits
DCAP Limit Went Up – A Win for Employees or a Nondiscrimination Nightmare? Five Takeaways
Effective for plan years beginning on or after January 1, 2026, the statutory limit on tax-free contributions to a Code Section 129 Dependent Care Assistance Program increased from $5,000 to $7,500 (and $2,500 to $3,750 for married individuals filing separately). Many plan sponsors eager to help their employees shelter more dependent care expenses from tax amended their cafeteria plans at the first opportunity to adopt the higher limit. Although this may seem like a straightforward employee-benefits win, employers would be wise to run projected DCAP nondiscrimination testing before deciding whether to implement the increased limit and should consider doing so.
1. DCAPs Are Subject to Nondiscrimination Testing Under Code Section 129(d)
A DCAP is only tax-favored if it satisfies four nondiscrimination tests each year including: (1) the plan must not favor highly compensated employees (“HCEs”) as to eligibility to participate (“Eligibility Test”); (2) the contributions and benefits actually made available under the plan must not favor HCEs (“Contributions and Benefits Test”); (3) no more than 25% of the dependent care assistance provided during the year can go to employees who own more than 5% of the company (“More-Than-5% Owners Concentration Test”); and, (4) the average benefit provided to non-HCEs must be at least 55% of the average benefit provided to HCEs (“55% Average Benefits Test”).
The Code defines an HCE as anyone who was a more-than-5% owner of the employer at any time during the current or preceding year, or who received compensation above the annually indexed statutory threshold in the preceding year ($160,000 for 2026). An employer may also make a “top-paid group election,” under which the compensation-based HCE category is limited to employees who were both above the compensation threshold and in the top 20% of employees ranked by compensation for that year. If an employer makes the top-paid group election for one plan, it must make the same election for all other plans it maintains that are required to use a consistent testing methodology for that plan year. Making this election can shrink the HCE group, which in turn can affect the DCAP’s testing results.
2. Employers Frequently Fail the 55% Average Benefits Test
As noted above, under the 55% Average Benefit Test, the average benefit provided to non-HCEs must be at least 55% of the average benefit provided to HCEs. It is a direct comparison of dollar amounts contributed by each group, not simply a question of who is eligible to participate.
Generally, HCEs are more likely than lower-paid employees to have the ability to take full advantage of whatever contribution limit the plan allows.
3. Raising the DCAP Contribution Limit Will Likely Increase Failures Under the 55% Average Benefits Test
Due to the above-described behavioral patterns, if HCEs disproportionately increase their DCAP elections to take advantage of a $7,500 limit while non-HCE participation and contribution levels stay essentially flat, the average benefit gap between the two groups widens. In other words, a change intended to help employees save more can end up making an already-difficult test harder to pass.
4. Nondiscrimination Testing Failures Must Be Corrected
When a DCAP fails testing, contributions made by HCEs in excess of the amount the plan can support become taxable income to those employees.
Employers who test at the beginning or middle of a plan year may be able to correct a projected failure prospectively, before it becomes a taxable event, by capping or reducing HCE contribution elections for the remainder of the year, amending the plan to add a sublimit going forward, or increasing non-HCE participation and contribution levels through targeted education and employer contributions (e.g., employer seed, match, or tiered contributions).
Employers who test at the end of, or after, the plan year no longer have the option to prevent the failure and instead face more complex corrections, including reissuing or correcting Forms W-2, adjusting withholding, and explaining to affected employees why money they believed was tax-free is now taxable after the fact.
5. Plan Sponsors Can Minimize DCAP Testing Failures By Testing in Advance of a Plan Year
Employers who run projected nondiscrimination testing before the plan year begins by using current census data and anticipated participation patterns, rather than waiting for actual year-end results, can see how a higher limit is likely to affect their 55% Average Benefits Test and Contributions and Benefits Test before employees have made irrevocable elections.
Like employers who test at the beginning or middle of a plan year, this information will then allow them to make prospective plan design decisions to minimize potential failures, as noted above.
What Plan Sponsors Should Do Now
In summation, before adopting the higher DCAP limit or communicating it broadly to its workforce, employers may want to model the testing impact first and make appropriate plan design changes. Setting realistic expectations with HCEs up front, rather than correcting their W-2s months later, is likely less costly, both administratively and in terms of employee relations, than discovering a testing failure after the plan year has closed.
