Nondiscrimination Module

The NHCE concentration % may be incorrect.
Terminees with less than 500 hours are being included in the concentration test, but not in the ratio percentage test.

example:

3 nhces, 1 hce. one NHCE terminated with less than 500 hrs, one nhce terminated > 500 hrs. terminees do not receive allocation.
while cross testing I get a ratio % of
1/2 divided by 1/1
but my concentration % is showing as 3/4 = 75%.
It should be 2/3 = 66%, or at least that is my understanding of the regs.
This may cause a plan to say it passes, when in reality it fails.
This is the rate group report on an HCE.
..............
create new plan from existing plan (including employees)

have a profit sharing plan. created a money purchase plan, and copied existing employees.
Status code and status date do not get copied, so have to re-key terminees status code and date!
also, participants who I deleted from the
prior year magically reappeared! so I had to delete them from the money purchase.
.........
census request

I am sometimes 'rented' out, so to speak, to do Quantech training. No, I am not working for Quantech, this started as part of something I do for the Southern Users Group. Anyway, I was discussing the top-paid group calculation for HCEs.
One of the items involved is hours - the system checks to make sure ees worked an average of 17 1/2 hours a week.
(e.g. an ee who quits 1/5 with 20 hours is considered in the top paid group count because he averaged over 17 /12 hours a week. [assuming he is over age 21, and has worked at least 6 months in prior year])
One of the trainees pointed out that the census request asks for the following codes: 0 = <500 hrs
1=>500 but < 1000 hrs
2= >= 1000 hrs

I hadn't thought about it before, be he is correct, if all you go by is a simple code like that rather than actually hours, your top paid group count will be incorrect. Most of the time this won't make a difference, but be careful because it might!
see the trainer gets trained!
this is not a bug, but something to watch out for.

The other problems have been reported to Quantech, but I figure this is a good spot to make other people aware of 'possible' bugs. I call them possible bugs, because there is always a chance I coded something wrong, although in the examples cited this seems doubtful.