The FSA non-discrimination testing deadline is not a date the IRS publishes on a form. Testing must be finished before the last day of the plan year, because that is the last day the employer can still correct a failure. For a calendar-year plan, final testing needs to be complete by December 31; a plan running July 1 through June 30 has to be done by June 30. Miss that window and the correction tools are gone, leaving affected employees with taxable income for benefits they already spent.1The IRS does not require employers to file non-discrimination test results on a specific date. There’s no Form 5500 equivalent for FSA testing, and no agency collects the results annually.
Why There’s No Filing Date
Nothing gets filed. There is no Form 5500 equivalent for FSA non-discrimination testing, and no agency collects results on an annual schedule. What the IRS requires is that the plan actually operate in a non-discriminatory manner across the plan year. If it doesn’t, the tax consequences apply retroactively to that whole year.2What the IRS does require is that the plan operate in a non-discriminatory manner for each plan year. If it doesn’t, the tax consequences apply retroactively to that plan year.
That retroactive quality is what turns the plan year-end into a hard deadline in practice. Once the year closes, the plan’s results are what they are. The administrator can still run the numbers in January, but the number they produce is a diagnosis, not something they can act on. Whatever failure the test reveals has already happened, and the only remaining step is reporting the taxable amounts on the affected employees’ W-2s.
Preliminary vs. Final Testing Timing
Most plan administrators split the work into two rounds so the deadline doesn’t sneak up on them.
The first round is preliminary testing, typically run during or shortly after open enrollment. It uses projected enrollment data and known compensation figures to flag potential problems early. If projections show highly compensated employees electing a disproportionate share of benefits, the administrator has time to adjust before the plan year progresses too far.
The second round is final testing, run with actual data from the full plan year: every enrollment change, termination, new hire, and election modification. For a calendar-year plan, this final analysis needs to be finished before December 31. Running it on January 2 is technically too late, because the window to correct failures for that plan year has already closed.3Final testing uses actual data from the full plan year. For a calendar-year plan, this final analysis should be complete before December 31. Running it on January 2 is technically too late because the window to make corrections for that plan year has already closed. A fiscal-year plan follows the same logic against its own year-end date.
Some administrators add a mid-year check between the two rounds, especially if the workforce is volatile or if a preliminary test showed a close margin. There is no rule about how often to test during the year. There is only the outcome the IRS cares about: a plan that ended the year in compliance.
Fixing a Likely Failure Before the Year Ends
The reason preliminary testing exists is that failures are fixable while the plan year is still open. Once it closes, they aren’t.
p>The most common in-year correction is reducing highly compensated employee elections to bring the plan back into compliance. Employers can also expand eligibility or push enrollment among non-highly-compensated employees to shift the ratios. These moves work because the tests measure what actually happened across the plan year, and there is still year left to change what happens.
After the last day of the plan year, those tools vanish. The employer cannot retroactively reduce an HCE’s elections, cannot retroactively enroll someone who was not enrolled, and cannot reallocate benefits that have already been reimbursed. The only remaining action is reporting the additional taxable income.4Once the plan year closes, these tools vanish, and the only remaining step is reporting additional taxable income on affected employees’ W-2s.
This is why the working deadline for most administrators sits well before December 31, not on it. If preliminary testing in October shows a likely failure, the employer needs weeks to implement election reductions, communicate changes, and let the corrected pattern play out across enough of the year to actually move the final numbers. A test run on December 30 that shows a failure is a test run too late.
What Happens if You Miss the Window
A failed test does not blow up the entire FSA plan. Non-highly-compensated employees keep their tax-free benefits regardless of the outcome. The consequences fall on the highly compensated employees and key employees whose excess benefits triggered the failure.5A failed non-discrimination test does not blow up the entire FSA plan. Non-highly-compensated employees keep their tax-free benefits regardless of the outcome. The consequences fall on the highly compensated employees and key employees whose excess benefits triggered the failure.
Under Section 125, if the plan discriminates in favor of highly compensated participants as to eligibility or benefits, the tax-free treatment under subsection (a) stops applying to those participants. Their FSA benefits for that plan year become taxable income. If key employees receive more than 25% of total qualified benefits, the tax exclusion stops applying to those key employees for benefits attributable to that plan year.6Under Section 125, if the plan discriminates in favor of highly compensated participants as to eligibility or benefits, the tax-free treatment under subsection (a) simply stops applying to those participants.
Under Section 105(h), the calculation depends on which test failed. A benefits test failure treats the full amount of the discriminatory benefit received by each highly compensated individual as taxable income. An eligibility test failure uses a fraction: total benefits paid to all highly compensated individuals divided by total benefits paid to all participants during the plan year. Each highly compensated individual’s reimbursements are multiplied by that fraction to determine the taxable amount.7For a benefits test failure, the full amount of the discriminatory benefit received by each highly compensated individual is treated as taxable income. For an eligibility test failure, the excess reimbursement is calculated using a fraction: the numerator is total benefits paid to all highly compensated individuals, and the denominator is total benefits paid to all participants during the plan year.
The employer reports the taxable amount on the affected employee’s Form W-2 for the year the failure occurred. Federal income tax and employment taxes apply to what the employee thought was a tax-free benefit. The employee may have spent the FSA funds months earlier, creating an unexpected tax bill they can’t easily absorb.
One outcome is much worse and worth flagging as a boundary, even though it is not what routine testing failures produce. Complete disqualification of the cafeteria plan happens when the plan fails a structural Section 125 requirement, like not maintaining a written plan document or not operating according to its terms. Full disqualification makes every participant’s benefits taxable, not just those of highly compensated employees.8Full disqualification makes every participant’s benefits taxable, not just those of highly compensated employees. The distinction matters: routine non-discrimination test failures are painful but targeted, while structural failures are catastrophic for the entire workforce. Missing the testing deadline produces the targeted outcome, not the catastrophic one, but the catastrophic one is why the plan document and its operational rules deserve their own attention.
Data to Have Ready in Advance
Missing the deadline usually traces back to data that arrived late, not to testing that took too long. Gathering the inputs early is the practical way to make sure final testing can actually run before year-end.
The administrator needs a complete employee census for the plan year, showing every person employed on any day during the year, not just current employees at year-end. Former employees who participated before leaving must also be counted.9The administrator needs a complete employee census for the plan year, showing every person employed on any day during the year — not just current employees at year-end. Former employees who participated before leaving must also be included in the count.
For each employee, the administrator needs compensation data sufficient to determine highly compensated and key employee status, plus ownership information going back to the prior year. The ownership data catches anyone who was a 5% owner at any point in either the current or preceding year, which is easy to overlook for employees who sold their stake mid-year.
The testing itself requires enrollment and election data: which employees were eligible, which enrolled, and what dollar amount each person elected. For dependent care FSAs, the administrator also needs actual utilization data, not just elections, because the Section 129 55% average benefits test looks at benefits provided rather than merely offered.10For dependent care FSAs, the administrator also needs actual utilization data — not just elections — because the Section 129 55% test looks at average benefits provided, not merely offered.
Consistency across years matters. The plan document defines how compensation is measured for HCE purposes and which employees are counted in the eligible population. Applying those definitions differently from year to year makes the entire testing process unreliable, and the plan document itself sets the plan year dates that determine when final testing must be complete.11The plan document also sets the plan year dates that determine when final testing must be complete.