ADP Testing Compensation

[Guest Robk]

For ADP Testing when disaggregating, can we exclude compensation earned prior to July 1st for Participants who would not have entered the Plan until July 1st had there been a year of service requirement?

Specifically, even though the Participant was deferring for the entire year, can you still exclude the compensation earned through June 30th?

As an example, say John was hired on July 1, 2005 and begins deferring in 2005. For the 2006 ADP Test, John "enters" the Plan July 1, 2006. For his Actual Deferral Ratio for 2006, can we use TOTAL 2006 deferrals and compensation for the period 7/1/2006 through 12/31/2006?

[buckaroo]

I do not think you can do this, but I have a differnet question. Why not just carve out the NHCEs when running the ADP test?

[Beltane]

Disaggregation only determines who can be separated when you split the test under 401(k)(3)(f). Once you have your 1 year of service/21 years of age group, you use compensation and elective deferrals for the entire plan year for that group.

In your example, John would fall into the main group, and his ADP for the year would be his total plan year deferrals divided by his total plan year compensation.

[Guest Robk]

I've been using total compensation when peforming Tests in this situation.

However, I was just at a session of the Western Benefits Conference where the compensation exclusion was offered as an "Alternative Example". As that would be helpful, I'd love to use it, but obviously I am not very comfortable without futher confirmation.

[Mike Preston]

Do you have a handout that specifies the mechanics or the citation? Somehow I feel there was a disconnect somewhere because I don't really think such an option exists.

[Bob R]

Personally, I'd like to see a cite supporting either position. I believe the regulation provides that you can treat them as separate plans. On that basis I don't know why you can't treat them as separate plans for purposes of being eligible to defer (thus disregarding comp prior to entry in the hypothetical second plan).

However, from a practical perspective, getting this information from the employer is an entirely different matter. That's why I think most people will divide it up based on the plan year in which the 1 year & age 21 condition is met. All deferrals and all comp are recognized for the entire year in which the requirement is met. And, if you want to use the entry date rather than the first day of the PY in which the requirements are met, I think it's consistent to only use deferrals and compensation after that hypothetical entry date.

[Mike Preston]

Well, I remain unconvinced at the moment. There are a couple of obstacles. First, the document should provide a methodology for determining ADPs. There is no need to have the document specify, one way or the other, regarding the use of permissive disaggregation. However, once it is applied, I would think that the document's provisions on determination of ADPs would still apply. And I don't know any documents that define compensation as anything other than one of two things:

1) Compensation while a participant

2) Compensation during the entire plan year

Neither of these allow for "compensation while a participant but ignoring compensation earned prior to a theoretical entry date not actually specified in the plan".

As I said, I remain unconvinced at the moment.

[Mike Preston]

I emailed the author of the handout at the conference and she made it clear to me that the materials she handed out need to be read with the caveats she identified at the beginning of her presentation. Specifically, she caveated her presentation by saying that she was merely communicating techniques that she had seen in use by others and that she was specifically not endorsing any of the concepts mentioned. Further, if anybody wanted to use any of the techniques identified, they were on their own to research the validity or lack thereof of said techniques.

With that said, I'm more comfortable than I was before in saying that I see no justification in the regulations for any sort of testing which limits testing compensation to "compensation while a theoretical participant", but allows deferrals to be based on the entire plan year.

If anybody wants to present a citation for an alternative view, I'd be glad to revisit this.

[Bob R]

Just to clarify - I misread the original question and was responding to Beltane's comment.

I agree that counting all deferrals for the year but only using comp from the hypothetical entry date is impermissible (or at least I'd consider it to be an unreasonable reading of the regulation). But, if you disaggregate the deferrals and test each component plan separately, then I believe the compensation for each respective component could be based on compensation while in such component plan. I think that's a reasonable interpretation.

The only other argument is that all deferrals count in the year in which you reach the hypothetical entry date and compensation is for the entire year. I see nothing wrong with this. But, if this is the only correct interpretation, then we've been struggling for years in trying to determine what hypothetical entry date can be used - and in many cases it would be irrelevant. It would only make a difference where the hypothetical entry date would be in a later plan year (i.e., first day of the following PY). Otherwise, who cares when the hypothetical entry date is within the PY if the entire PY must then be taken into account for testing.

[Mike Preston]

The entry date provision is heavily debated not for purposes of bifurcation of either deferrals or compensation. Rather, it is debated for purposes of determining whether an individual is, in fact, excludable in the year in question. If excludable, they are in one of the permissively disaggregated plans. If non-excludable, they are in the other of the permissively disaggregated plans. Take an individual hired 8/1/2005 in a plan that has quarterly entry dates. If you take the plans entry date provisions into account, this individual is not excludable in 2006. If you don't, the individual is excludable in 2006.