Is participant count on 5500 really that important?

[Guest IluvNewComp]

Is an accurate participant count on the 5500 really THAT important?

I mean, if we had put 34 actives and it was really just 32 or 36?

Would a govn't auditor really get upset in that case?

I'm not talking about a case in which the difference would mean filing as a large plan or not. And I'm not talking a significant delta of like 20% or anything like that.

[david rigby]

Are you looking for the line that you don't want to cross? There is no permitted "range". Answer the question to the best of your knowledge. Later, if you discover an error due to imperfect data, then your choices are to (1) amend the filing, or (b) ignore it and get it right the next year.

If your participant count is in the thousands, errors are much more likely than a count in the range of 30.

[Bird]

uh-oh. Could mess up some guvmint statistics that say average plan size is x.1233456789 when it really is x.12345678.

I always say to be careful with these kinds of questions, because when you start to "go there" you realize how very UNimportant a lot of the "work" we do is!

[BG5150]

because when you start to "go there" you realize how very UNimportant a lot of the "work" we do is!

True. In all my different positions in my pension career, there seems to be an enormous amount of time and effort into determining if there are 85 participants in the plan, or, wait, didn't that lady terminate in December 2007 and January 2008? Maybe we really only have 84 participants. I'd better call, send e-mails and faxes and registered letters and drive over tot he client's office to take a look at the payroll systems just to make sure...

[Peanut Butter Man]

I would be worried about the "due diligence as to accuracy" requirement, the accurate record-keeping requirement AND that the Form 5500 is signed under penalty of perjury.

I agree that if it is incorrect when the Form 5500 is filed, I amend to correct.

[Guest Sieve]

How hard is it to look at an allocation report and count how many active & inactive account balances there are, and how many term'ed during the year, etc.--especailly when you're only counting to 34? (I'm not a TPA, so I assume you'll tell me it's much more difficult than I could ever imagine.)

If you can't tell how many there are, then how do you know if allocations of contributions/earnings are being done properly--or if testing is being done properly? It certainly makes a differenc on ADP & TH testing.

[TPAMan]

I agree, counting folks is a pretty easy except for when it isn't. A few instances come to mind - seasonal type employees, plan status changes (eg, non-union to union), unpaid furloughs, or military leave (or any other approved leaves). These kinds of things can easily muddle a count particularly when an event crosses a plan year end.

Then there was the client that wanted us to tell them how to determine if an employee was terminated.....

It's all easy. That's why we get paid the big bucks!

[david rigby]

Indeed.

Counting participants is not the same as counting account balances.

(This is especially true in a DB plan.)

[PensionPro]

How hard is it to look at an allocation report and count how many active & inactive account balances there are, and how many term'ed during the year, etc.--especailly when you're only counting to 34? (I'm not a TPA, so I assume you'll tell me it's much more difficult than I could ever imagine.)

If you can't tell how many there are, then how do you know if allocations of contributions/earnings are being done properly--or if testing is being done properly? It certainly makes a differenc on ADP & TH testing.

Yes, it is much more difficult than you could ever imagine. Had a large takeover multiple employer plan with 1000s of participants. Same participants had multiple accounts (some had wrong SSNs), multiple accounts were set up for beneficiaries, alternate payees had separate accounts, several were on LOA, etc. Needless to say, various adjustments had to be made.

[Guest Sieve]

For thousands, I understand. But for 34? How long can that take, difficulties notwithstanding? 2-3 hours in the worst-case scenario? Sorry, but in the scheme of things, this just seems like a task that has to be done right, since it's the base for so many other things. I obviously overstated the ease with which this can be accomplished, but I would think that it's the cost of doing the TPA business, and complications have to be overcome (client confusion notwithstanding).

Does it sound like I'm having a bad hair Monday (err, rather, a bad baldy Monday)?

[PensionPro]

I agree that for about 34 participants, the count should be exactly accurate.

[austin3515]

Let's say the client omits 3 term dates from the census file. Personally, I don't see this as a big deal. Even if it came to light next year, I would not do an amended 5500.

[K2retire]

Ideally, we would always have accurate data. But realistically (especially if the employer is small enough that the don't have a full time HR person) what Austin describes happens regularly. The scarier example is the client who doesn't bother to tell us about the new people, because "they don't want to participate anyway" when they really mean that the newly eligible participant decided not to defer at this time.

[BG5150]

The scarier example is the client who doesn't bother to tell us about the new people, because "they don't want to participate anyway" when they really mean that the newly eligible participant decided not to defer at this time.

Got the W2's from a client the other day. "Is this all of them," I ask. "You bet," she says.

Wow! 100% of the employees deferred in 2009! Cool!

*lightbulb*

Call her back: "Hey, did everyone pu into the 401(k) last year?" "No, why?" "Because all of the W2's have amounts in 12D. That indicates they were putting money away." "Oh, I only sent you the ones who were doing 401(k). Do you need the others?"

/facepalm

[PensionPro]

Since we are trading war stories, we had a so-called one-participant plan, a doctor. One year suddenly the client says after the work is completed I hired a few employees last year, but I did not tell you because they did not work long enough. Come to find out these were rehired employees who were immediately eligible upon rehire. Not the first or last time we had to redo work because of client's decision to provide selective information.

[EPCRSGuru]

I had a plan once, while I was still young and naive, in which every employee shared the same last name--or so I thought. Turns out the family-owned corporation was only allowing family members to participate and was threatening all the non-family members with termination if they tried to enroll. (We resigned the account.)

[PensionPro]

I stay awake at night wondering why there are so many safe harbor match plans with only the owners participating ...

[Andy the Actuary]

I agree that for about 34 participants, the count should be exactly accurate.

Archimedes Baby Ox Principle: A newborn baby ox weighs 17 pounds. You lift it everyday. After a year, you are lifting a full grown ox.

So, it you should be exactly accurate with 34, then what about 35? If 35, then what about 36? What is the cutoff for doing the job right versus approximately right? Perhaps it's the Justice Potter Stewart's definition, "I don't know how to define the cutoff but I know it when I see it." How would the PBGC react on a premium audit (Does the PBGC conduct premium audits) if they discover reporting was inaccurate?

It is incumbent upon us to do "the best to our knowledge" and if our knowledge changes, fix whatever if the resulting count is material.

[PensionPro]

Each professional needs to make their own determination how accurate they want to be. By the time you get to the count you would have completed your data checks, discrimination testing, reviews, etc. Not claiming that the participant count is a material component of the tax return, but there should be a good reason for not getting it right. Such as incomplete or inaccurate information provided by the client or their advisors.

If it is okay to have an incorrect (or approximate) participant count for a 34 participant plan? How about a plan with 33? or 32? or 3? or 2? or 1?