Section 415 Limits and refunds on earnings

In 2025, the 415 limits are $70,000. Here is the scenario:

The employee has a profit-sharing plan because of an agreement that can not produce refunds.  The employee also has a 401k plan with employee deferrals only. Weekly payroll on the deferrals where the monies go in within 3-5 days after the pay date, is consistent and timely. 

The 2025 year ends and we realize that some 401k participants are over their 415 limits. Let's also assume that over the age of 50 and between 60-63 the catch-up and super catch-up contributions were backed off to the 415 limits.  We now have a scenario where refunds have to occur. 

In the first scenario, a total of $9,000 has to be refunded. This $9,000 came in over 9 weeks ($1,000 per week) and applied to payroll periods in November and December. The $9,000 has to be refunded (This is a given)

Now, the $9 million question:  What about earnings on the $9,000 over 9 weeks, plus lead time to perform the calculations and set up the refunds?  It is nearly impossible to capture 10 investments in different asset classes from the day the monies were invested, plus earnings to the day it was refunded.  

How does a TPA or Plan Administrator/Trustee calculate the earnings on the refunds?  One can use this calculator and refund the monies by March 1st of the following year as the pooled investments are held in individual accounts from a major recordkeeper, etc. 

https://www.askebsa.dol.gov/vfcpcalculator/webcalculator.aspx

Do we use the online VFCP calculator to determine refunds on earnings? I see no guidance if we had market losses, like in 2022 (unless you were all in guaranteed investment contracts).  

I doubt a TPA has sophisticated earnings software, and they probably perform this manually.  Again, only for 415 limits here. The ADP test is a separate issue that one can monitor and know upfront based on the census of this plan. 

Earnings Calculator VFCP.pdf

If the plan uses one of the major TPA/recordkeepers, they should be able to handle the earnings calculations.  They do it all the time. I assume you have asked them.  Assuming you did and they said they can’t do it, then the big questions are whether the majority of the affected participants are NHCEs and was there an overall gain in plan investments during for the period of correction?  If the answer to both is yes, you do not have to do any earnings calculations. 

First, note that you have to make sure you are looking at the right way to calculate earnings.  Under EPCRS, corrections that require a reduction of a participant’s account balance have different rules for determining earnings than corrections that require a contribution/allocation to plan accounts.  Here, you are reducing a participant’s account so you use (or first start with) those rules.

If the majority of affected participants are NHCEs and there has been an overall gain for the period of correction, an adjustment for earnings is not required for the affected participants.  If the plan sponsor decides it wants to remove the earnings from participant accounts, the reduction to the account balances may be adjusted by the lowest rate of return of any fund available for the correction period for administrative convenience.  Hopefully, the plan sponsor will not want to do this but at least you should be able to determine that fund.   

However, if the majority of affected participants are not NHCEs (i.e., HCEs) or there has been an overall loss for the period of correction, reductions to earnings may be required for corrections (and using the lowest rate of return is not permitted).

EPCRS doesn’t address losses in earnings for corrective distributions or reductions in a participant’s account balance. The common view is that the safest method to calculate earnings in such instances is to make corrective distributions or reductions in account balances without adjustment for negative losses in earnings if the majority of the participants affected are HCEs. (So the affected participant is funding the loss.  That is, you take the full amount of the overcontribution and don't offset it for any loss.)   If most of the participants affected are NHCEs, a plan sponsor may choose to apply the loss in earnings to the amount to be recovered (i.e., the Company will fund the loss, which is what we normally do if it is NHCEs, but the problem here is you have to determine the loss so the plan sponsor would not choose to apply the loss and thus make the participants fund that loss).

When the majority of affected participants are HCEs we go back to the earnings calculation rules for corrections that require a contribution/allocation.  Distilling the rules based on a majority of affected participants being HCEs and assuming, because you state there are 10 investments, that the plan allows participant direction of investments, it would be best to use the actual rates of return for the period of correction.  Alas, the instant TPA cannot make that calculation.

Our rule is never to use the CFVCP calculator...  If the affected participants are primarily HCEs, of course, they will want to use the DFVCP calculator, as it actually gives really low earnings (at least from our experience)).  That said, be careful when you start to get into the administrative convenience rules or the rules that allow you to use alternative methods as you have to show various things.  For instance, to use the DFVCP calculator that it is not feasible to make a reasonable estimate of what the actual investment results would have been.  If this is the way you go, you should require the TPA to memorialize all the reasons as to why it is not feasible for them not to be able to do it.  Our concern is that feasibility will be from the perspective of the IRS not ours.  If it is a $9M question (and not hyperbole), you may wish to go to VCP.  

Like it says below… Just my thoughts…

Thank you for the detailed response. The TPA does not wish to calculate earnings. However, we will not know if the 415(c) limits have been exceeded until the 4th quarter of 2025. That being said, it is all based on the amount of hours a person works, because the profit sharing is based on total hours. There are 8,736 maximum hours in one calendar year.  Please note that every single one of these people with potential refunds is HCE's, with no exception. If they were NHCE, they would never reach the 415(c) limits. 

Only a few employees can reach the maximum. Everyone else is based on actual time worked, which is favorable. A limited number of employees have had stops placed on deferrals, which are elective not to reach the maximum. Some are over 50 years old, and the age 50 catch-up of $7,500 will not apply, which helps. The age of 50 is scheduled to go away in 2026.

If refunds are due, we will know in November and December, mostly December with almost 99% accuracy, and within the first week of January, with 100% accuracy.

Hypothetical: We take one person who has to refund $5,000, and the 4th quarter gained 10%. Can we safely assume that the withdrawal is $500 in earnings, plus the $5,000?

The Plan Document, not the SPD states the following:

The IRS-approved prototype document with an opinion letter “states reasonable method”, regarding earning calculation refunds. 

In the end, we have notified every participant of the possible conflict. Adjustments have been made in elective deferrals to try to avoid this potential conflict. 

I have also been advised that the W-2 needs to be amended in 2026 if 415 refunds occur for the 2025 calendar year. 

Any further guidance would be highly appreciated. Thank you. 

Why would the W-2 be affected? He should get a 1099-R from the Plan for the refund.

Any reasonable method to determine earnings can be used. I believe we assume deferrals ratebly over the year and apply the annual earnings rate for 1/2 the year if that makes sense.

There is conflicting information online. If a participant, between his profit-sharing plan (company monies based on hours), and his 401k employee-only deferral, is over $70,000 in 2025 (As an example).

In 2026, we have to refund monies as he is over $5,000. Does the refund in 2026 (The distribution) out of his 401k plan (as this is where it needs to come from) not count towards 2025 income? As he went over the limits in 2025? 

The Record Keeper (who holds the plan assets) is willing to issue a 1099-R (I believe for 415 refunds, but I must gain further insight).

Thank you in advance for your help.  There is a 99% probability no refunds have to occur in 2026 as we have developed a detailed tracking system moving forward.  Thank you. 

Online is a fine starting point, depending on your resources, but personally, I don't rely on information I find online unless those sources are citing IRS authorities and I have tracked it back to those authorities.  Here, I would simply look at Rev. Proc. 2021-30 and the 1099-R instructions.  My recollection is that Rev. Proc. 2021-30 has special rules for excess additions/allocations that specifically state that a 1099-R is to be used with 415 corrections and the instructions for the 1099-R have specific rules for distributions under EPCRS.  Of course, don't rely on my statements....

Based on the facts you provided, it appears that there are no matching contributions, so in my view the correction should be first, to distribute 2025 salary deferral contributions (adjusted for earnings), then, if any excess remains, to forfeit 2025 employer profit-sharing contributions (adjusted for earnings if necessary) until the annual additions no longer exceed the 2025 415(c) limits.  This priority order is used so the participant retains as much of the employer monies that they can.

The corrective distribution (not the forfeiture, if any) made to the participant, presumably in 2026, should be reported on a 2026 Form 1099-R. The participant should include the distribution as income in 2026 but does not have to pay the 10% additional tax on early distributions under IRC Section 72(t). I think this is a Code E but not certain.  The distribution is not eligible for rollover (the 1099-R should reflect this and we usually send a letter of explanation making that clear).  The forfeited employer contributions (plus earnings) should be transferred to an unallocated plan account, which must be used to reduce employer contributions in subsequent periods.  No additional employer contributions are to be made to the plan until the unallocated plan account balance is reduced to zero. 

This is not an instance where the participant has the double whammy of income in the year earned and income in the year distributed (which applies to late distributions of excess deferrals or 402(g) busts).

The profit sharing can not exceed $61,000 as it is based on a flat dollar amount per hour worked. Only a few people will reach this amount, and everyone else will fall between $30,000-$55,000 in profit sharing.

Regarding the 401k plan, not everyone maxes out. The issue of "amending returns" could be a nuisance, as well as calculating returns. There may be no profits in 2026, as the market has started quite shaky.  

Thank you for the detailed information. 

In the correction I describe in my prior post, there would be no amending of returns.

Thank you Artie:

I will report back to you at the end of the 2nd Quarter to see how the contributions are going. Once we finish the first half of 2025, we will know better about potential conflicts as once hours are not worked, we can with 100% accuracy figure out how much more each person can contribute which occurs in the 3rd and 4th quarters. 

The IRS should have a simple formula 5% with their calculator for guidance in all markets. FINRA has these  Analyzers and they use 5%, so why can't the IRS do the same? This would make administration so much easier for the USA and retirement plan limits.