FSA Nondiscrimination Testing: A Plain-English Guide for Employers

If you offer a health FSA — or run any pre-tax benefit through a Section 125 cafeteria plan — the IRS requires annual testing to prove the plan doesn't disproportionately favor your highest-paid people. Failing doesn't blow up the plan for everyone; it makes the benefit taxable for the favored group. Here's what the tests check, who counts, and how to stay on the right side of them.

Why Testing Exists

A cafeteria plan lets employees pay for benefits with pre-tax dollars — no federal income tax, no FICA. That's a valuable exclusion, and Congress conditioned it on fairness: the tax break can't flow mainly to owners and executives while rank-and-file employees get little. Nondiscrimination testing is how you demonstrate that, once a year, with actual plan data.

This applies to the cafeteria plan as a whole; health FSAs add a similar layer of their own under Section 105(h), and administrators typically run everything together, so we'll treat it as one exercise here. (Employer HSA contributions run through a cafeteria plan are tested here too — outside one they follow comparability rules instead.)

The Three Tests

Test	What it asks	Who it protects against
Eligibility test	Can enough non-highly-compensated employees actually participate?	Plans that quietly exclude lower-paid workers
Contributions & benefits test	Are benefits and employer contributions available on comparable terms to everyone eligible?	Richer benefits or bigger employer dollars for executives
Key employee concentration test	Do key employees receive more than 25% of the total nontaxable benefits under the plan?	Small companies where owners dominate participation

The eligibility test looks at who's allowed in. If your plan covers all full-time employees after a uniform waiting period, you'll generally pass. Trouble starts when eligibility carves along lines that correlate with pay — covering salaried but not hourly staff, or one location but not another.

The contributions and benefits test looks at what participants can actually get. The same election limits, the same employer contribution formula, the same benefit menu for everyone eligible is the safe pattern. An employer FSA contribution that scales with salary, or a benefit tier only managers see, is the classic failure.

The key employee concentration test is pure arithmetic: add up the nontaxable benefits all participants elected under the plan, and check whether key employees account for more than 25% of that total. Small businesses fail this one most often — if two owners max out their FSAs while few of their twelve employees participate, the math gets tight fast. The fix is participation: the more your broader workforce contributes, the smaller the key-employee share becomes.

Who Counts as an HCE or Key Employee

Two overlapping-but-different definitions, and the distinction matters because different tests use different groups:

• Highly compensated employees (HCEs) generally include officers, more-than-5% owners (and their spouses and dependents), and employees whose prior-year compensation exceeds the IRS's indexed threshold.
• Key employees generally include officers with compensation above an indexed threshold, more-than-5% owners, and more-than-1% owners with compensation above $150,000.

The exact dollar thresholds adjust over time, and attribution rules can pull in family members of owners — worth confirming the current figures and your specific roster with your administrator or benefits counsel.

What Failing Actually Means

Failing a test does not disqualify the plan or create a problem for your broader workforce. Instead, the affected HCEs or key employees lose the tax exclusion: their elections get treated as taxable income, reported as wages on their W-2, with income tax and FICA due. Your non-highly-compensated employees keep their pre-tax treatment either way.

That's still a bad outcome. Your executives were promised a pre-tax benefit and got a taxable one, payroll has to re-report contributions, and if the failure surfaces after W-2s go out, you're issuing corrections. The reputational cost inside the leadership team usually exceeds the dollars.

When to Test

The tests are formally measured as of the last day of the plan year — exactly the wrong time to discover a problem, because by then there's nothing left to adjust. Standard practice:

• Mid-year preliminary test (month 6–8 of the plan year): run the numbers on actual elections and flag anything trending toward failure.
• Final test at plan year end, documented and kept with your plan records.

If you've never tested, or your administrator has never mentioned it, that's a genuine gap — testing is your responsibility as plan sponsor even when you delegate the mechanics.

How to Fix a Failing Test

Caught before year end, most failures are fixable:

1. Reduce HCE or key employee elections. The plan can cut the favored group's contributions prospectively (and in some cases refund excess amounts as taxable wages) until the math passes. This is the most common cure for a concentration failure.
2. Broaden eligibility. If the eligibility test is the problem, extend the plan to the excluded group — often cheaper than it sounds, since not everyone will elect.
3. Increase broad-based participation. For the 25% concentration test, the denominator is your friend. Employee education, better understanding of what FSAs cover, and making wellness expenses eligible via Letters of Medical Necessity all raise rank-and-file elections — which passively fixes the ratio without cutting anyone.
4. Fix the design. Salary-scaled employer contributions or management-only benefit tiers should be redesigned before the next plan year, not patched annually.

Caught after year end, the only remedy is including the excess in the favored employees' taxable income for that year. Test early.

How Hammock Helps

Hammock administers FSAs with nondiscrimination testing built into the plan year — preliminary testing while there's still time to act, final testing documented for your records, and a dedicated account manager who flags a drifting ratio instead of surprising you in December.

Because Hammock also makes wellness spending FSA-eligible through provider-reviewed LMNs and surfaces eligible expenses employees are already paying for out of pocket, participation rises across the whole workforce — which is the healthiest possible way to pass a concentration test.

FAQ

Do small employers have to run nondiscrimination testing?

Yes — there's no small-employer exemption for Section 125 testing, and small companies are actually the most likely to fail the key employee concentration test because owner participation looms large against a small payroll.

What happens if we fail and don't fix it?

The favored employees' pre-tax elections become taxable wages for that year, subject to income tax and FICA, and must be reported on their W-2s. On audit, uncorrected failures can also mean penalties and interest on unpaid payroll taxes.

Can we just exclude HCEs from the FSA to avoid testing?

You can limit or exclude HCE participation, and some employers cap HCE elections preemptively — but you still have to run the tests and document the results each year.

Does testing apply to HSA contributions too?

If employer HSA contributions run through your cafeteria plan (most do), they're tested under Section 125 alongside the FSA. Outside a cafeteria plan, they follow comparability rules instead — a different, stricter standard.

When during the year should we test?

Twice: a preliminary run mid-year, when elections can still be adjusted, and a final run at plan year end for your records. A single year-end test leaves you no room to cure a failure.

The Bottom Line

Nondiscrimination testing is a fairness check with a taxable-income penalty, and it lands on the people you least want to surprise. The mechanics are simple — three tests, run twice a year — and almost every failure is preventable if you test mid-year and treat broad participation as your first line of defense. An administrator who runs the tests proactively, and a benefit employees across the pay scale actually want to use, solves this quietly.

Want testing handled — and participation that makes passing easy? Talk to our team.