Compensation used for valuation

[ac]

Does anyone know of any guidance regarding what compensation must be used for valuation of a defined benefit plan (i.e. for a 1/1/2002 valuation date, should we use 2001 plan compensation or 2002 plan compensation).

[pmacduff]

What does the Plan Document say? My thought is that it would usually be the 2001 comp because in theory you are valuing the plan on 01/01/02 and would not yet have any 2002 compensation figures, right?

[ac]

The plan document defines compensation, it does not state what should be used to project the retirement benefit at the assumed termination date.

We are valuing a small plan where the owners can control the amount of their plan compensation. For 2001, one owner made $10,000, for 2002 the same owner made $150,000. In calculating the 2002 minimum required contribution, what compensation amount should we use to project the retirement benefit?

[Guest Keith N]

What is your valuation date? If 1/1/02, than I don't think you can use the 2002 compensation since it is paid after the valuation date. If your valuation date is 12/31/02, than I think you should use the 2002 compensation.

I have seen many valuations that used a BOY val date, but waited until the EOY to do the val and used the current year's comp. I have heard the IRS on several occasions state that this was improper, but as far as I know, they have never challenged it.

I don't really think it's a document issue. You’re doing a valuation, not calculating a benefit. The compensation used for valuation purposes is an actuarial assumption that must be reasonable based on the actual definition in the document, but it doesn't have to be the same. In larger plans it is very common to estimate compensation for valuation purposes due to data quality issues.

Consistency is very important. Once you decide how you are going to do it, it becomes part of your method and shouldn't be changed from year to year.

[Mike Preston]

The logic, I believe, that is used to support using end of year compensation when doing a beginning of year valuation is that projecting compensation beyond the valuation date is an assumption. Assumptions must be reasonable. How reasonable is an assumption that ends up matching experience precisely?

[david rigby]

Not trying to read something sinister into the original question, but it seems so simple as to be surprising. I wonder if there is something else going on.

For example, is someone trying to arrive at a particular answer?

Is this a new plan?

[Guest Keith N]

Mike,

I tend to agree with you, but I have been told by several high ranking IRS types at several EA meetings that it is not appropriate.

I was passing on a conservative answer to "ac"

[ac]

Yes the plan is new for 2002, business commenced January 2001. The client had low pay, $10,000, for first year. In 2002 he paid himself $150,000. If business conditions allow, he will pay himself $150,000 each year until retirement. I do not think it is reasonable to fund for a benefit based on $10,000 compensation. for 2002.

Has anyone had a similar situation?

[Blinky the 3-eyed Fish]

Sounds like you need to either use the $10,000 as estimated 2002 compensation or use an end of the year valuation date. I would not have a beginning of the year valuation and use actual compensation for the year.

[Mike Preston]

Blinky, what is it about experience matching one's assumptions that bothers you?

[Blinky the 3-eyed Fish]

It's not that that bothers me, it's the fact that I am using information after the valuation date. Irrespective of a plan termination date during the year where the normal costs and amortization bases would be prorated, I do not consider knowledge past that magical valuation date.

If you are using the actual compensation for the plan year for a beginning of the year valuation, why stop there? Should you then consider distributions, asset changes, employee terminations, etc? Why not just have an end of the year valuation? Personally, I see a consistency in the methodology I use that is not there with using actual events.

I remember a similar discussion before, but couldn't find it.

[mwyatt]

Do an EOY valuation for first year as Blinky suggested (you'll come up with the same result as you have no assets yet), and then change to BOY valuation in 2003 if desired by changing the funding method. You get automatic approval on the change (remember the 4 year requirement between changes doesn't mean that the plan necessarily had to have been in effect).

[Mike Preston]

I would think that any of the items you mentioned might be taken into account in a BOY valuation with specific assumptions. Think of the concept of select and ultimate. Lots of things work this way, don't they? For example, one can take into account, but needn't have to, an amendment made after the valuation date and before the end of the year. One can include new entrants that enter the plan in the year after the valuation date.

Has anybody ever heard of the IRS actually challenging the use of anything that takes place during the year in a BOY valuation?

We know that you can't take into account things that happen after the end of the valuation year, like amendments, etc.

But that sort of acts to define the limits, I would think.

[david rigby]

Does it matter? For example, if the projected benefit is the 415 limit (either one), it may not matter how you got there.

Let's be reasonable out there.

[mwyatt]

I think this is a simpler situation than is being debated. 2002 is the first year of the Plan. Run the val @ 12/31/02, taking into account salaries paid for 2002 (which would logically happen with an end of year valuation). Then, if you wish, change the funding method to beginning of year valuation in 2003 (allowed as the four-year lookback doesn't mean that the plan had to be in existence in the four years) and life goes on. Everyone's happy, and nothing untoward has been done.

[Mike Preston]

There is no question that your suggestion works, mwyatt, but is giving up one's automatic change in val. date the best course of action? In the end, that is the decision that must be made.

[mwyatt]

Actually, what are you giving up, since there is no automatic approval to change to EOY val date, and changing the val date has no impact on waiting period for any other forms of change in funding method (i.e., asset valuation, cost method, etc.). If anything, starting with EOY val date gives you flexibility in future years to change to BOY in any future year, while avoiding this compensation problem for the first year contribution in this case.

[Guest lisbetf]

My firm does the administration for many small plans. All our sole proprietor plans have beginning of the year valuation dates, and use the Schedule C compensation earned during the plan year for income to calculate the benefits used for the valuation. We have to wait until after the end of the plan year to do the valuation.

Our prototype plan document allows you to specify the compensation to be earned during the current plan year. As long as you are consistent from year to year, it shouldn't matter what you use.

[mwyatt]

I remember also back in the mid 80s doing these types of valuations (ie @ BOY but using comp during the year but assets as of the beginning of the year). We steered away from this approach as it tends to take away the advantage of doing a BOY valuation in the first place of knowing the contribution amount during the year, rather than getting a nasty surprise after the year has ended.

I really don't think that your document has too much bearing here as to the definition of compensation; this is more a point of the actuarial cost method.

[David MacLennan]

Isn't the method discussed here, the "prospective" comp method with BOY val date, more or less equivalent to an EOY val date, but using a asset valuation method that lets EOY Assets = BOY Assets x (1+i)? Is this an unreasonable asset valuation method? It may be unusual but it is not unreasonable. For this reason, I don't subscribe to the absolute valuation date "shapshot" idea. The actuary is given latitude in deciding the valuation method. (Changing it with automatic approval is another issue.)

Also, I use DATAIR software and it allows one to use current year comp with BOY val date - they call it "prospective" comp. I know a software company like DATAIR is not authoritative in any way, but they are a useful repository of accumulated experience/methods. Based on my takeover cases, I would say a large percentage (maybe nearly half) of small DB plans use DATAIR for actuarial valuations, and a significant number of these use "prospective" comp method. Regardless of what IRS folks say, I think this method is grandfathered in.