Nondiscrimination Question 410(b)

[Mister Met]

Employer with several DB and DC plans. The last coverage test was done and passed as of 12/31/08. Will need to do another one as of 12/31/11. Data is collected as of 12/31. There is concern that the 12/31/11 test won't pass coverage. If the ratio percentage test at 12/31/11 fails and the average benefits test at 12/31/11 fails as well, what corrective action could be taken, and when? I'm confused since we won't know the testing result until after the end of the plan year. Could an amendment be made after 12/31/11?

Thanks

[AndyH]

Employer with several DB and DC plans. The last coverage test was done and passed as of 12/31/08. Will need to do another one as of 12/31/11. Data is collected as of 12/31. There is concern that the 12/31/11 test won't pass coverage. If the ratio percentage test at 12/31/11 fails and the average benefits test at 12/31/11 fails as well, what corrective action could be taken, and when? I'm confused since we won't know the testing result until after the end of the plan year. Could an amendment be made after 12/31/11?

Thanks

Yes, see 1.401(a)(4)11-(g)

[Tom Poje]

and this reg cite will note that correction must be made within 9 1/2 months after the end of the plan year.

that being said,

I'd have other concerns.

my old notes from CCH describes the 3 year cycle (you can only use this method if it is reasonablly expected there were no significant changes subsequent to the test)

but now you say you expect the plan to fail this year. that sounds to me like there has been a possible significant change.

what about last year? if you didn't test because you were relying on the 3 year cycle, but suspected there might be problems because of changes I think you have problems.(ok, as with anything in life, only if you get caught)

Almost forgot I even had these notes stashed away. ugh, this goes back to 1992, but at least it talks about the three year cycle. Its Announcement 92-117, IRB 1992-33,21 August 17, 1992I see

I made a notation on these notes many years, also indicating under certain condition you'd have to bump the 70% to 77% just to be on the 'safe side'.

since I've never done 3 year testing I never bothered with, but it's there stuck in the back of my mind.

[AndyH]

All good points of course.

Couple of add-ons. It is possible that the 9.5 months could be longer if there is a determination letter pending and you are in the remedial amendment period.

The 3 year cyle is much more common in DB world. There are adjustments as Tom refences depending upon whether or not you use shortcuts, of various types.

But overall I agree that the expectation of failure should cause questioning of the validity of the three year testing cycle itself.