What Is the 401(k) ACP Test?

ACP uses the ADP math on different money.

That distinction is the whole test.

The Actual Contribution Percentage, or ACP, test generally measures:

• employer matching contributions
• employee after-tax contributions

for highly compensated employees (HCEs) and nonhighly compensated employees (NHCEs).^[1]

It does not ordinarily test pre-tax 401(k) deferrals.

It does not ordinarily test designated Roth 401(k) deferrals.

Those are elective deferrals and belong in the ADP test.^[1][6]

The confusion comes from the phrase after-tax. Roth contributions are made with after-tax dollars for income-tax purposes, but federal retirement-plan rules still classify them as elective deferrals. Employee after-tax contributions used in ACP are a different contribution type.^[6]

Key Takeaways

• ACP means Actual Contribution Percentage.^[1]
• The test generally measures matching contributions and employee after-tax contributions.^[1][2]
• Designated Roth 401(k) deferrals are not employee after-tax contributions for ACP testing; they remain elective deferrals for ADP.^[1][6]
• Each participant's tested ACP contributions are divided by testing compensation to produce an individual contribution ratio.^[1]
• Individual ratios are averaged separately for HCEs and NHCEs.
• The ACP passing formula mirrors ADP:
• 125% of NHCE ACP, or
• the lesser of 200% of NHCE ACP and NHCE ACP plus 2 percentage points,
• whichever produces the greater permitted HCE result.^[1]
• A plan can pass ADP and fail ACP because deferrals can be nondiscriminatory while matching or after-tax contributions are not.
• HCE classification errors affect ACP just as they affect ADP. INV-084 covers HCE status.
• Prior-year and current-year testing methods are available under the governing rules and plan document.^[1]
• Contributions used as QMACs in ADP generally cannot be counted again in ACP.^[5]
• A safe-harbor 401(k) can receive ACP relief only when the applicable Section 401(m) safe-harbor conditions are satisfied.^[3][4][7]
• Extra matching contributions can fall outside that relief and require ACP testing.
• Failed ACP testing creates excess aggregate contributions, which generally must be corrected under the plan and tax rules.^[1]

ACP Starts With the Contribution Source

Before calculating a percentage, classify the money.

That table prevents the most common ACP error: treating every contribution made after income tax as the same type of money.

Roth Is Not ACP "After-Tax"

IRS distinguishes:

Designated Roth contribution

An elective deferral included in current taxable income.^[6]

Employee after-tax contribution

A contribution from compensation that is not a designated Roth contribution and is included in current income.^[6]

Both can reduce take-home pay after tax.

They occupy different testing categories.

Example: 10% Roth Deferral, 0% ACP

Employee earns:

$100,000

Employee elects:

$10,000 Roth 401(k) deferral

Employer provides:

no match

Employee makes:

no non-Roth after-tax contribution

ADP numerator:

$10,000

ACP numerator:

$0

The employee can therefore have:

• 10% Actual Deferral Ratio for ADP
• 0% contribution ratio for ACP

Calling the Roth contribution an "after-tax contribution" in the recordkeeping feed would put the same money into the wrong test.

The Individual ACP Ratio

IRS says ACP is calculated in the same general manner as ADP, except the numerator uses matching and after-tax contributions rather than elective deferrals.^[1]

Simplified:

tested matching + tested employee after-tax contributions ÷ testing compensation

The resulting percentage is the individual's contribution ratio for ACP.

The plan then averages individual ratios by HCE and NHCE group.

Example: Match Only

Employee compensation:

$80,000

Tested employer match:

$2,400

Employee after-tax contribution:

$0

Individual ACP ratio:

$2,400 ÷ $80,000 = 3%

The employee's elective deferral rate can be 4%, 8% or 12%.

That does not change the 3% ACP ratio unless it changes the employer match amount under the plan formula.

Example: Match Plus Employee After-Tax

Employee compensation:

$100,000

Employer match:

$4,000

Employee non-Roth after-tax contribution:

$3,000

Tested ACP contributions:

$7,000

Individual ACP ratio:

7%

The test combines contribution sources that Section 401(m) requires to be tested together, subject to the governing rules.

ACP Averages People, Not Group Dollars

Like ADP, ACP averages individual percentages.

Suppose four NHCEs have:

NHCE ACP:

(0% + 3% + 4% + 5%) ÷ 4 = 3%

The $200,000 employee does not get four times the weight of the $50,000 employee.

The test averages participant ratios.

A 0% ACP Does Not Mean the Employee Is Inactive

An employee can have a 0% ACP while:

• deferring 15% pre-tax
• deferring 15% Roth
• making catch-up contributions

if no tested matching or non-Roth after-tax contribution is allocated.

ACP is not a general retirement-savings participation rate.

It measures specific Section 401(m) contribution sources.

The Passing Formula

The ACP test is met if the HCE ACP does not exceed the greater of:^[1]

Test A

125% of NHCE ACP

or

Test B

the lesser of:

200% of NHCE ACP

or

NHCE ACP + 2 percentage points

This is the same mathematical framework used for ADP.

The inputs are different.

Example: NHCE ACP Is 4%

Test A:

125% × 4% = 5%

Test B:

Lesser of:

• 200% × 4% = 8%
• 4% + 2% = 6%

Test B result:

6%

Maximum HCE ACP:

6%

If HCEs average 7%, ACP fails.

That can happen even if ADP passed cleanly.

Example: NHCE ACP Is 1%

Test A:

1.25%

Test B:

Lesser of:

• 2%
• 3%

Maximum HCE ACP:

2%

A plan with very little NHCE matching or after-tax activity can face a tight HCE ACP ceiling.

Why ADP Can Pass While ACP Fails

Assume NHCE and HCE elective-deferral rates are balanced.

ADP:

Pass

The employer's match is much richer for HCEs because of an operational or design issue.

ACP:

Fail

The plan did not fail because HCEs saved too much.

It failed because tested Section 401(m) contributions were disproportionately high for HCEs.

The two tests should not be read as one combined result.

Example: Same Deferral Rate, Different Match

HCE and NHCE each defer:

6%

ADP can be perfectly aligned.

Now assume:

NHCE match

2% of compensation

HCE match

6% of compensation

Ignoring other facts, that disparity can create an ACP problem even though the employee deferral behavior is identical.

The match is its own nondiscrimination issue.

A Match Formula Can Be Correct and ACP Can Still Fail

Do not jump from:

ACP failure

to:

the written match formula is illegal.

A plan can fail because:

• HCEs contribute at rates that produce more match
• NHCEs contribute little and therefore receive less match
• employee after-tax contributions are concentrated among HCEs
• the plan uses a non-safe-harbor matching structure
• testing data is wrong

The formula may have been followed exactly.

The resulting contribution pattern can still fail the annual test.

HCE Classification Still Drives the Groups

ACP uses HCEs and NHCEs, not key employees.^[1]

INV-084 covers HCE classification.

Relevant issues include:

• more-than-5% ownership
• family attribution
• prior-year compensation
• top-paid group election where applicable

One misclassified owner can distort both group averages.

Example: Owner's Spouse Misclassified

Employee classified as NHCE:

• ACP ratio: 8%
• spouse of more-than-5% owner

If attribution makes the employee an HCE, rerunning ACP:

• removes 8% from NHCE average
• adds 8% to HCE average

The correction can move both sides against the employer at once.

A passing report is only as good as the HCE list.

ACP Uses Testing Compensation

The denominator must use the correct compensation definition.

IRS warns that wrong compensation data produces false ADP and ACP results.^[1]

Suppose employee receives:

• salary: $100,000
• commission included in testing compensation: $25,000
• match: $5,000

Correct ratio:

$5,000 ÷ $125,000 = 4%

If payroll sends only salary:

$5,000 ÷ $100,000 = 5%

That one-point error can matter in a small HCE group.

Match Calculation Errors and ACP Errors Are Different

A plan can have two separate problems.

Operational match error

The employer failed to make the match required by the plan document.

ACP testing failure

The contributions actually counted in Section 401(m) testing are too concentrated among HCEs.

IRS treats failure to make the correct match as its own qualification problem.^[1][3]

Fixing the match can change the ACP test.

That means the sequence should be:

1. correct plan operations
2. rerun ACP
3. correct any remaining testing failure

Do not test known-bad contribution data.

Payroll-Period Matching Can Distort Annual Results

IRS warns that a plan promising an annual matching formula can be operated incorrectly if payroll calculates each pay period without a required year-end true-up.^[3]

Example:

Plan promises:

50% match on deferrals up to 6% of annual compensation

Employee front-loads deferrals early in the year and later earns a large bonus.

If the system never true-ups the match against annual compensation, the employee can receive less than the written formula requires.

That is a plan-operation error first.

ACP testing should use the corrected match.

Prior-Year vs Current-Year Testing

IRS permits ACP testing to use either current-year or prior-year NHCE data under the applicable rules.^[1]

The election belongs in the plan document.

Prior-year method

Current HCE ACP is generally compared with prior-year NHCE ACP.

Current-year method

Current HCE ACP is compared with current NHCE ACP.

The same structural trade-off seen in ADP applies here:

• prior-year data improves predictability
• current-year data is more responsive to current contribution patterns

Example: Method Changes the Result

Assume:

• prior-year NHCE ACP: 5%
• current-year NHCE ACP: 2%
• current-year HCE ACP: 6%

Prior-year method

NHCE ACP of 5% permits maximum HCE ACP of:

7%

Result:

Pass

Current-year method

NHCE ACP of 2% permits maximum HCE ACP of:

4%

Result:

Fail

Same HCE match.

Different documented testing method.

QMACs Cannot Do Double Duty

A qualified matching contribution, or QMAC, can under specified conditions be treated as an elective deferral for ADP testing.^[5]

If a QMAC is counted in ADP, IRS says it is disregarded in ACP.^[5]

That prevents the same dollar from improving both tests.

This is a critical data-tagging issue.

Example: $50,000 QMAC

Employer makes:

$50,000 of qualifying matching contributions

and elects to use those amounts in ADP under the applicable rules.

The same $50,000 cannot then be inserted again into ACP as ordinary matching contribution merely because its source code says "match."

The testing designation matters.

QNECs Can Also Receive Special Treatment

Qualified nonelective contributions can under specified conditions be used in ADP or ACP correction and testing.^[1][5]

A QNEC is generally:

• employer funded
• fully vested when allocated
• subject to distribution restrictions comparable to elective deferrals.^[1][5]

The employer cannot simply rename an ordinary profit-sharing contribution after year-end.

The contribution must satisfy the regulatory requirements for the treatment claimed.

Safe Harbor: The Subtle ACP Issue

A safe-harbor 401(k) can be deemed to satisfy ADP.

That does not mean every possible matching contribution added to that plan is automatically exempt from ACP.

Section 401(m) has its own safe-harbor conditions.^[2][4][7]

This is where short explanations often become inaccurate.

Safe-Harbor Matching Conditions

IRS guidance states that matching contributions can receive ACP safe-harbor treatment when the plan satisfies the applicable safe-harbor framework and the match meets conditions such as:^[4]

• the match does not take into account deferrals or employee contributions above 6% of safe-harbor compensation
• the matching rate does not increase as the participant's contribution rate rises
• an HCE does not receive a higher match rate than an NHCE at the same contribution level
• if the match is discretionary, the discretionary amount does not exceed 4% of safe-harbor compensation.^[4]

The exact safe-harbor design should be checked against the governing plan terms and current regulation.

The 4% Discretionary Match Limit Matters

Assume a safe-harbor plan uses a 3% nonelective safe-harbor contribution.

The employer adds a discretionary match.

Discretionary match of 4%

If the contribution satisfies the other Section 401(m) safe-harbor conditions, IRS guidance says it can be deemed to satisfy ACP.^[4]

Discretionary match above 4%

The additional match can fall outside the safe-harbor limit and require ACP testing.^[4]

That is why:

"We are safe harbor, so ACP does not apply"

can be wrong.

Safe Harbor Is Contribution-Specific

A safe-harbor plan can contain:

• safe-harbor contribution
• discretionary match
• employee after-tax contributions
• profit sharing

Each source needs its own testing analysis.

The plan's label does not automatically exempt every source.

The question is:

Does this contribution satisfy the applicable Section 401(m) safe-harbor rule?

Safe-Harbor Match Vesting Is More Nuanced Than ADP Safe Harbor

IRS notes that some matching contributions in a safe-harbor 401(k) can receive ACP safe-harbor treatment even when they are subject to a permissible vesting schedule, provided the Section 401(m) conditions are satisfied.^[4]

That is distinct from the vesting requirements attached to the contribution used to satisfy the 401(k) ADP safe harbor itself.

Do not transfer one vesting rule across every contribution source.

Employee After-Tax Contributions Can Create ACP Exposure

Many modern 401(k) plans allow voluntary after-tax employee contributions beyond ordinary elective deferrals, often as part of a strategy that may later involve Roth conversion.

INV-047 covers the mega backdoor Roth concept.

Those employee after-tax contributions can enter ACP testing.^[1][6]

A plan can therefore have:

• ordinary elective deferrals that pass ADP
• after-tax contributions concentrated among HCEs
• ACP failure

The after-tax feature needs compliance modeling before HCEs use it aggressively.

Example: Mega Backdoor Roth ACP Problem

Plan permits employee after-tax contributions.

NHCEs rarely use the feature.

HCEs contribute heavily.

Assume:

• NHCE ACP: 1%
• HCE ACP: 6%

Maximum HCE ACP:

2%

The plan can fail badly even though:

• every participant stayed within Section 415 limits
• ADP passed
• the after-tax feature is permitted by the plan

A legal contribution type can still produce a nondiscrimination failure.

The Conversion Does Not Erase the ACP Test

Suppose an HCE makes an employee after-tax contribution and immediately converts it to Roth inside the plan.

The later conversion does not retroactively change what the original contribution was for ACP testing.

The source began as an employee after-tax contribution.

Tax conversion and nondiscrimination classification are separate questions.

What Happens When ACP Fails?

A failed ACP test produces excess aggregate contributions.^[1]

IRS says the law generally treats these similarly to excess contributions from an ADP failure, subject to the contribution source and correction mechanics.^[1]

Possible correction paths can include:

• distributing or forfeiting excess aggregate contributions allocated to HCEs
• making qualifying employer contributions for NHCEs
• correcting under EPCRS when the statutory correction period has passed.^[1]

The plan document and current correction rules control the exact method.

Vested and Unvested Matching Contributions Can Separate

This is one place ACP correction differs operationally from a simple cash refund.

IRS states that if excess aggregate contributions consist of matching contributions that are not fully vested, the unvested portion is generally handled through forfeiture and reallocation or a suspense account under the plan's terms rather than paid to the HCE.^[1]

The vested portion and earnings may be distributed as required by the correction.

The participant's vesting schedule therefore matters to the correction mechanics.

Example: HCE Match Is 60% Vested

Assume an HCE has:

$10,000

of excess aggregate matching contributions.

Vested percentage:

60%

The correction cannot be analyzed as if the entire $10,000 were an ordinary distributable account balance.

The vested and unvested portions can receive different treatment under the plan and correction rules.

That is why an ACP correction file needs:

• source
• vesting
• earnings
• participant status

not just a dollar amount.

The 2½-Month and 12-Month Clocks

IRS applies the same statutory correction framework to ADP and ACP failures.^[1]

The plan generally has:

2½ months

after the end of the plan year to correct without triggering the ordinary Section 4979 excise-tax consequence associated with late correction, subject to applicable exceptions.^[1]

Certain EACAs receive six months.^[1]

The broader statutory correction period generally extends:

12 months

after the end of the tested plan year.^[1]

Calendar-Year Example

Plan year:

2026

March 15, 2027

Ordinary 2½-month deadline for a calendar-year plan.

December 31, 2027

End of the ordinary 12-month statutory correction period.

A sponsor should verify the exact tax treatment for the chosen ACP correction method rather than assume every ADP-specific exception applies identically.

The Fix-It Guide expressly identifies ACP excesses as excess aggregate contributions and says they are generally treated like ADP excess contributions.^[1]

A Failed ACP Test Is Not a Participant Withdrawal Decision

An HCE receiving an ACP correction may see money removed from the account.

That is not the same as:

• requesting a distribution
• taking a hardship withdrawal
• rolling money to an IRA

It is a plan qualification correction.

Tax reporting and rollover eligibility depend on the correction type.

The participant should use the plan's Form 1099-R and tax guidance rather than treating the payment like a voluntary distribution.

Timing Matters More With Matching Data

A final ACP test can be delayed because employer matches are sometimes:

• funded after year-end
• trueed up after payroll closes
• corrected after compensation review

Testing before the final match is known can produce a false result.

The sponsor should identify:

• which match is accrued
• which match is funded
• which match is required under the plan
• which amount belongs in testing

"Year-end match report" is not enough unless everyone means the same number.

Contribution Source Codes Are a Control

A clean recordkeeping file should distinguish:

• pre-tax elective deferral
• Roth elective deferral
• catch-up source where separately tracked
• safe-harbor match
• discretionary match
• QMAC
• employee after-tax contribution
• QNEC
• profit sharing

If those sources are collapsed into:

employee contribution

and:

employer contribution

the ACP test becomes harder to audit.

Source coding is not cosmetic.

It decides what enters the numerator.

What to Review Before Accepting ACP PASS

If those inputs are not reviewable, the reported percentage is not enough.

ADP vs ACP

The mathematics look almost identical.

The legal classification of the money is not.

Frequently Asked Questions

What does ACP mean in a 401(k)?

ACP means Actual Contribution Percentage. It is a nondiscrimination test under Section 401(m) that generally compares matching and employee after-tax contribution rates of HCEs and NHCEs.^[1][2]

How is ACP calculated?

For each participant, divide tested matching and employee after-tax contributions by testing compensation. Average those individual percentages separately for HCEs and NHCEs.^[1]

Is the ACP formula the same as ADP?

The passing formula uses the same 125% / 200% / plus-2 framework. The contribution sources are different.^[1]

Do Roth 401(k) deferrals count in ACP?

Generally no. Designated Roth contributions are elective deferrals and are tested in ADP rather than treated as employee after-tax contributions in ACP.^[1][6]

Are employee after-tax contributions the same as Roth contributions?

No. IRS explicitly distinguishes non-Roth after-tax employee contributions from designated Roth elective deferrals.^[6]

Does employer match count in ACP?

Generally yes, subject to safe-harbor rules and special treatment for contributions such as QMACs.^[1][4][5]

Can a plan pass ADP and fail ACP?

Yes. ADP tests elective deferrals; ACP tests different contribution sources. One result does not determine the other.

Does ACP use HCEs or key employees?

HCEs and NHCEs. Key employees are used for top-heavy testing, not ACP.^[1]

Can QMACs count in both ADP and ACP?

Generally no. IRS states that QMACs used as elective deferrals in ADP are disregarded in ACP so the same contribution does not do double duty.^[5]

Does every safe-harbor 401(k) automatically avoid ACP?

No. The matching contribution must satisfy the applicable Section 401(m) safe-harbor conditions. Extra matching contributions can remain subject to testing.^[2][4][7]

Can a safe-harbor plan make a discretionary match?

Yes, but ACP safe-harbor treatment depends on the regulatory conditions. IRS guidance includes a 4% of safe-harbor compensation limit for discretionary matching contributions seeking that relief.^[4]

What happens if the ACP test fails?

The plan generally must correct excess aggregate contributions under the plan and applicable tax rules, which can involve distributions, forfeitures and/or qualifying employer contributions.^[1]

What if the excess match is not fully vested?

IRS states that the unvested portion of excess aggregate matching contributions is generally forfeited and handled under the plan's reallocation or suspense-account provisions rather than distributed to the HCE.^[1]

How long does the plan have to correct?

IRS describes a 2½-month period tied to excise-tax consequences and an ordinary 12-month statutory correction period, with six months for certain EACAs.^[1]

Can employee after-tax contributions used for a mega backdoor Roth cause ACP problems?

Yes. Employee after-tax contributions can enter ACP testing even if they are later converted to Roth. Heavy HCE use with little NHCE use can create an ACP failure.^[1][6]

The Source-Code Check

Before reviewing the ACP percentage, export contributions by source.

You should be able to identify separately:

1. pre-tax deferrals
2. Roth deferrals
3. employer match
4. employee non-Roth after-tax contributions
5. QMACs
6. QNECs
7. other employer contributions

Then ask:

Which of these dollars did the ACP test include, and why?

If the answer is unclear, do not start with the passing formula.

Fix the contribution classification first.

Sources & References

1. IRS: 401(k) Plan Fix-It Guide — Failed ADP and ACP Nondiscrimination Tests
2. IRS: A Guide to Common Qualified Plan Requirements
3. IRS: 401(k) Plan Overview
4. IRS: Issue Snapshot — Vesting Schedules for Matching Contributions
5. IRS: Issue Snapshot — Plan Forfeitures Used for QNECs and QMACs
6. IRS: Retirement Topics — Contributions
7. IRS: Notice Requirement for a Safe Harbor 401(k) or 401(m) Plan
8. IRS Publication 560: Retirement Plans for Small Business

Educational Disclaimer

ROIStreet publishes educational content about retirement-plan testing and administration. This article is not legal, tax, fiduciary or plan-administration advice. ACP results depend on contribution source, employee classification, compensation, testing method, safe-harbor design, vesting, correction timing and current law.

The ROIStreet Reader Promise

We strive to explain before we evaluate, present evidence before opinions, discuss risks alongside potential benefits, distinguish facts from analysis, and correct material errors transparently.

Our purpose is to help readers better understand investing—not to tell them what to do.