Rate Banding

[wmyer]

I was wondering how practitioners in general are doing rate banding for DC 401(a)(4) testing (non cross tested) for large plans. Is there a standard way of doing this? How much flexibility is there in setting up the rate groups?

For example, do you take the highest HCE allocation rate as the starting point and use that as the minimum of the top rate group or as the midpoint of the top rate group? And then, for the next rate group, do you repeat this procedure based on the next highest HCE allocation rate that does not fall within the first range? Let's say I have HCEs with accrual rates of 3.0%, 2.8%, 2.4%. Would my first rate group be 2.75%-3.25%? And then would my second rate group be 2.25% to 2.75%? Or, could my first rate group be 2.5% to 3.0% and then my next rate group go from 1.9% to 2.4%?

[Tom Poje]

There is a good deal of flexibility as I understand the regs.

bands could actually be less than 5%

obviously no overlap,

and an ee does not have to belong to a specific group.

you listed a range 2.75% - 3.25%. the rule is the numbers must be within 5% of a midpoint.

lets suppose you chose 3% as the midpoint. 5% of 3 = .15

so your range would be 2.85% - 3.15%

the regs frown upon it if allocation rates of HCEs within a range are significantly higher than the NHCEs in that range.

Thus, I wouldn't recomend setting the HCE at the hi point and work backwards to determine the lo point of the range.

I know, I know, that is the way at least one software does it, but that goes against what the regs would say.

To avoid the HCE being significantly higher it is probably a good strategy to have some NHCEs in the group that have higher rates than the hce.

that way, you are bringing some NHCEs up to a midpoint, and bringing some NHCEs down to the midpoint. (I am not saying that the HCE has to be the midpoint, just that you try to avoid the significantly higher issue. There is nothing specified what is considered to be significantly higher so you have to use some common sense.

actually wrote this one up in the coverage and nondiscrimination answer book (q.8:20) based on what I learned years ago.

see 1.401(a)(4)-2©(2)(v)(A)

[wmyer]

Thanks Tom for that answer. FYI, I was using a range of 2.75 to 3.25 because of the alternative de minimus rule for DC plans that you could use .25% instead of 5% of the midpoint (that's why I specified it wasn't cross-tested or DB, because the de minimus rule there is .002%).

[Tom Poje]

ok, I'll but that. my brain is on freeze sometimes.