Anyone an expert on Average Benefits testing?

[AKconsult]

We are testing 2 plans in a control group. One plan has a match. The other plan is deferrals only, no match. Pretty large population - about 1200 people total (after using otherwise excludable rule).

The Ratio Percentage test fails any way we test it. ADP/ACP for the plan with the match both pass. ADP for plan without the match fails and refunds have already been issued. Had the ADP/ACP testing been done on a permissive aggregation basis, the whole testing would have failed and even more refunds would have been necessary.

BRF testing passes. The Average Benefits Percentage test fails.

Now I am somewhat stumped...How do I determine how to correct for the test failure? Is the answer to give a match to the employees of the plan without a match? to give a higher match to the nonHCEs in the plan with a match?

Thanks. any thoughts are appreciated:)

[Tom Poje]

well....caution (or at least ignoring the following):
1.401(a)(4)-11(g)(3)(iv) corrective amendment supposed to be in place by 10/15
...................
1.401(a)(4)-11(g)(2) ....may retroactively increase allocations for those who benefited during the year, or may grant allocations to individuals who did not benefit

I suppose there is nothing to stop you from provided a profit sharing (e.g. QNEC not used in ADP testing) as well to select NHCEs to bump up the avg ben pct test.

depending on how close you are to passing avg ben pct test use comp - deferrals for 414s comp

of course, if document has fail safe language it is a moot point as you have to bring people in and no corrective amendment needed.

or probably other minor things that could be tried that help

[AKconsult]

Thanks Tom. We have tried average ben pct test using comp less deferrals and still failing. We also pulled in 3 years of compensation and tried the 3-year average comp testing, still failing.

Document does not contain fail safe language.

Obviously, we are now past the 10/15 amendment deadline.

We are considering a couple of options.

1. give a match to the employees in the plan that didn't get matched, using same formula as other plan. That document allows for a discretionary match, so I am assuming it would be OK to make a match and fund before year-end for 2013, they just take the deduction a year later. I don't see any reason this would require an amendment, do you? this will make the ratio percentage test pass, on an aggregated basis. We then just have to go back and retest ADP/ACP on an aggregated basis, and do additional refunds.

2. give a profit sharing allocation to the people in the plan who didn't get a match. That document has language already to allow for a discretionary profit sharing, so I am assuming that as long as we follow the allocation requirements, no amendment necessary, we just declare the profit sharing and fund before 12/31. Obviously, problem with this is it creates a lot of small balances for people who don't otherwise have an account, so may not be palatable to client.

Do you see a reason either of these approaches would have required a 10/15 amendment? Thanks!!

[Tom Poje]

if you have doc language that says each person is in their own group, then I could see going with the profit sharing and picking and choosing people to receive.

discretionary does not imply 'pick and choose' so unless the document language is such I'm not sure you can just give to some and not others. you have to follow the terms of the document

[Mike Preston]

Have you run the ABPT on a cross-tested basis? If not, make sure the plan's ERISA counsel agrees that you can (some are misguided into thinking that doing so invokes gateway: it does not) and once you get approval, do so. I can't believe it will fail. If this were left to fester for two years and the IRS audited and if I was brought in to see what could be done the first thing I'd do is run the ABPT on a cross-tested basis.

While you don't say it explicitly, I am presuming the plan would satisfy 410(b) if the ABPT was passed.