We just took over the admin of what appeared to be a nice, clean cross-test 401(k). However, we just discovered over $60,000 in unallocated forfeitures that apparently have been building over the years. Document says to reduce plan expenses, then contributions, then reallocate. According to prior administrator, these forfeitures were held in suspense "for future plan expenses". Although it appears they have been deducting quarterly fees as well!

Looking for opinions here....due to the multitude of plan years involved, it would cost a fortune to go back and reallocate, make distributions, etc. Not to mention time I don't have. Appears that the most practical solution is to use the forfeitures to fund the $50,000 contribution for 2003 that hasn't been funded yet and reallocate the rest. Of course, I would bring this to the attention of the client and let them know of the potential problem upon audit.

Refusing to take the case is not an option, so please don't offer that opinion! Thanks for any input.

It appears unlikely that there would have been any additional allocations in prior years if the plan had been operated correctly? It appears that the forfeitures for any one year would have been in the range of $10,000 or less (dividing $60,000 by "several"). So if the prior years' contributions would have been similar to this year ($50,000) or even much less, then there probably would not have been any excess to allocate? So no issue there?

But can you clarify what has been happening with fees? They have been deducted out of forfeitures? Or deducted directly out of accounts? So some people have lower account balances than they would have had the expenses been properly paid out of forfeitures?

My only warning is that fee problems are becoming the big issue of the day, so make sure the warning is adequate...

Fortunately, upon closer look, it does look like the fees came from the forfeiture account.

Now I'm concerned that the reason there is so much money in the forfeiture account is because of a layoff in 2001. So we have possible termination in operation issues to consider now.

But assuming there is not a termination in operation, I think there could be potential issues in years in which the employer funded their 404 limit and the excess was never reallocated. There is a DB plan, too. And even if there is not a termination in operation, and the forfeitures are legitimate, I still see a significant problem upon audit, even if 404 limit wasn't reached. If the employer chose to fund a contribution (and take a deduction) before allocating forfeitures, the forfeitures had to be reallocated in that plan year. Wouldn't you agree?

UPDATE - Historical valuation for 2001 shows that forfeitures had already accumulated to $35,000 as of 12/31/00. On the surface, it does not appear that we have a partial termination issue. For now I am concluding that forfeitures have been building up year after year beginning in 1999.

Fellow practitioners out there - please provide your opinion. To review the issue, I have inherited a 401(k) plan with over $60,000 in unallocated forfeitures. I need to provide a recommendation to the client on the best way to fix this problem. In 2001 there were over $15,000 generated in new forfeitures and there was no employer contribution allocated, therefore nothing to reduce anyway. I see this as more than a deduction problem. What would you do? Am I overly concerned?

Difficult to say without seeing specific document language. I'm only guessing, but I suspect the document does not allow forfeitures to simply accumulate - in other words, the forfeitures for any given year pay fees, reduce contributions (whatever that means in a PS plan) and are then reallocated - right away. But not accumulated for years.

If that's the case, you have operational errors stretching back apparently several years. If they are deemed "significant" then you may have lost the window to self-correct, and may have to correct under VCP - possibly a "John Doe" approach.

I'd consult someone experienced in the correction procedures under Rev. Proc. 2003-44. There's enough money involved, with potentially serious ramifications, to make it worth the client having to pay for some good tax/legal advice.

Yes!

I know you don't want to hear this, but...at my previous place of employment (recordkeeper), we got in major trouble for forfeiture issues. (can't remember what type of audit, but it was discovered at audit)

We spent countless MONTHS (YEARS!!) cleaning up many plans! We actually had to go back 5-6 years and do reallocations. Here is how it went....

Reallocate 1998 forfeitures (with earnings funded by ER or recordkeeper), process residual distributions. Forfeitures from these residual distributions had to be allocated to the correct "bucket" for the year in which they should have been forfeitures had they been allocated correctly. (Of course, it was great fun tracking down the census data for those years.)

Move on and reallocate 1999, same follow up as 1998....And you can guess the rest.

I am not sure how correct the following practice was, but I will throw it out there for you to decide....

We did adopt the practice of "carrying forward" amounts that would have resulted in participants receiving less than $1.00 per allocation. For example, you have $10,000 in 1999 forfeitures that should have been allocated, and you have 10,000 participants. Since the average amount a participant would receive is less than $1.00 we carried the balance forward and added it to the forfeitures from 2000.

It was time consuming, labor intensive, and generally STUNK, but it was the right thing to do.....

Have you thought about submitting it to the IRS anonymously for an opinion?

Thank you Belgarath and FundeK! Actually, that was what I wanted to hear. Not that I have the time or energy to do all that work, but I don't think it's right just to simply make it "go away" with the 2003 funding. I believe the participants are due additional money. And yes, it is significant since there are only around 30 participants. The problem appears to be at least 5 years old and possible 7, way outside the self-correction time period.

Thanks again.