New Comparability testing does not ask whether everyone received the same contribution percentage. It asks whether a proposed allocation can be tested on an equivalent-benefits basis and, if so, whether the resulting rate groups satisfy the applicable nondiscrimination rules. The gateway is only one part of that process.
What New Comparability testing is actually proving
A New Comparability profit-sharing formula can assign different employer contribution rates to different participants or groups. That flexibility does not create an automatic pass. Section 401(a)(4) still requires the employer contributions or benefits provided under the plan to be nondiscriminatory in favor of highly compensated employees.
Cross-testing is the method that lets a defined contribution allocation be evaluated on a benefits basis. Instead of stopping at today’s contribution percentage, the testing converts the relevant allocation into an equivalent retirement benefit and compares those equivalent benefit accrual rates across employees.
Before that benefits-basis comparison is available, the arrangement must satisfy the applicable rules for cross-testing. For many individualized New Comparability designs, the common route is the minimum-allocation gateway. The regulations also recognize other routes, including broadly available allocation rates in appropriate designs. That is why “everyone received 5%” is not the definition of a passing New Comparability test.
Can this allocation use benefits-basis testing? If yes, do the resulting HCE rate groups satisfy the applicable coverage and nondiscrimination requirements?
The testing flow, step by step
Population → proposed allocation → cross-testing gateway → EBARs → HCE rate groups → coverage testing → pass or redesign.
| Step | Question | Why it matters |
|---|---|---|
| 1. Population | Who is an employee for testing, who is an HCE or NHCE, and who benefits under the allocation? | The wrong population can produce the wrong answer before any math begins. |
| 2. Allocation | What employer contribution is actually being tested for each participant? | The test needs the final contribution sources, compensation and plan-year data. |
| 3. Gateway | Is the design eligible to use benefits-basis cross-testing under an available regulatory route? | For many New Comparability designs, the minimum-allocation gateway is the first screen. |
| 4. EBARs | What equivalent benefit accrual rate results from each tested contribution? | This is where age and time to the testing age can change the comparison. |
| 5. Rate groups | For each nonexcludable HCE, which employees have an EBAR at least as high as that HCE? | Each HCE creates a separate rate group that must be evaluated. |
| 6. Coverage | Does each required rate group satisfy the applicable section 410(b) coverage route? | A group can pass the gateway and still fail the general test. |
| 7. Result | Does the final tested design pass, or must the allocation be revised? | The answer should tie back to specific census data, assumptions and contribution amounts. |
EBAR means equivalent benefit accrual rate. In a cross-tested defined contribution plan, the contribution is projected and converted under the regulatory framework into an equivalent retirement benefit. An older employee generally has fewer years for the contribution to grow before the testing age, which is one reason the same contribution percentage can produce different equivalent benefit rates for employees of different ages.
After those rates are calculated, each nonexcludable HCE forms a rate group. That group includes the HCE and the other nonexcludable employees whose EBAR is equal to or greater than the HCE’s. The rate group is then tested under the section 410(b) coverage rules. A common path is the ratio-percentage test; where available, the average-benefit test can provide another route and involves additional conditions beyond one percentage.
A worked allocation example
Consider a fictional professional firm with one owner and three employees. Assume all four participants are in the relevant testing population and the compensation shown below is the applicable compensation for this simplified illustration.
| Participant | Age | Compensation | Employer contribution | Allocation rate |
|---|---|---|---|---|
| Owner | 55 | $300,000 | $45,000 | 15% |
| Manager | 45 | $100,000 | $5,000 | 5% |
| Employee A | 35 | $60,000 | $3,000 | 5% |
| Employee B | 30 | $40,000 | $2,000 | 5% |
The highest HCE allocation rate in this simplified example is 15%. Under the common minimum-allocation gateway, one-third of 15% is 5%, so the three NHCEs shown at 5% clear that gateway screen. That tells us the design can move to the next part of the analysis under this assumed route. It does not tell us that the design has passed section 401(a)(4).
The next step converts the tested allocations into EBARs. We intentionally do not invent EBAR percentages here: the calculation depends on the regulatory assumptions and the participant data used by the testing engine. Once calculated, the owner’s rate group is built and tested for coverage. If the rate group passes, this allocation may support the design. If it fails, the firm may need a larger contribution for one or more NHCEs, a lower owner target, a different allocation structure or another permitted testing approach.
A 15% owner / 5% staff design can clear a common gateway and still require a full cross-test. The employee cost shown on a design illustration is therefore a starting point until the rate-group testing is complete.
What can change a result?
New Comparability testing is sensitive to more than the contribution percentages. Two census files with the same total payroll can produce different outcomes because the individual facts are different.
- Ages and dates of birth. Cross-testing converts current contributions into equivalent benefits, so the age pattern matters.
- HCE and NHCE status. Ownership and the applicable compensation rules determine who is tested as highly compensated; job title alone does not.
- Who benefits. Eligibility, allocation conditions, termination status and the plan’s terms affect the population used for testing.
- Compensation. Testing compensation, section 415 compensation and the compensation used by the allocation formula are not automatically interchangeable.
- Contribution sources. Safe harbor nonelective contributions can matter differently from matching contributions, and the amounts already funded need to be identified correctly.
- Related employers. Controlled-group or affiliated-service-group issues can change which employees belong in the analysis.
- Multiple HCEs. Each relevant HCE can create a separate rate group, so a design that works for one owner may not work identically for another.
Separate compliance checks can also affect implementation even when the section 401(a)(4) result is satisfactory. Depending on the plan, that can include section 410(b) coverage outside the rate-group analysis, annual-additions limits, top-heavy requirements, ADP/ACP testing, plan-document provisions and other plan-specific rules. A New Comparability pass should not be presented as a blanket certification of every qualified-plan requirement.
What a useful testing report should show
A decision-ready result should let the advisor, TPA and plan sponsor understand exactly what was tested and exactly which contribution amounts the conclusion supports.
| Report element | What to look for |
|---|---|
| Data version | Plan year, census date, compensation fields and the source of any corrections. |
| Population | HCE/NHCE classification, exclusions, related-employer assumptions and who benefits. |
| Final allocation | Employer contribution by participant and by source, including amounts already funded. |
| Gateway basis | The route used to establish eligibility for benefits-basis testing and the supporting result. |
| Cross-test | The EBAR methodology, HCE rate groups and applicable coverage results. |
| Exceptions | Any unresolved plan-document, compensation, eligibility or other compliance question that requires separate review. |
The practical question at the end should be simple: “Are these the final employer contribution amounts supported by the completed test, or are we still looking at a design illustration?” That distinction matters when the plan sponsor is deciding what to fund.
NewComparability.com is building toward this type of testing workflow. The public New Comparability calculator remains an educational scenario tool: it models contribution economics, not EBARs, rate groups or a formal pass/fail result. For the individual concepts behind this process, read the New Comparability gateway guide, the cross-testing guide and the section 401(a)(4) testing guide.
Sources & further reading
- IRS: Correcting gateway test failures
- Treasury Regulation §1.401(a)(4)-8: cross-testing and gateway rules
- Treasury Regulation §1.401(a)(4)-2: nondiscrimination in contributions and rate groups
- Treasury Regulation §1.410(b)-2: minimum coverage tests
- IRS training material: rate groups and equivalent benefit accrual rates
Educational content updated September 23, 2026. Examples are fictional and simplified; they explain the testing sequence and do not constitute a plan-specific compliance result or funding instruction.