Jim Dexter

Most of these are legal questions rather than actuarial (I'm an actuary specializing in consulting to employers on multiemployer issues) but I'll take a shot at a few of the questions. The situation is actually one I have seen before. First, my understanding is that a special agreement under which nonunion employees contribute is generally viewed the same way as a collective bargaining agreement. Thus, ceasing to have an obligation to contribute under the special agreement would appear to trigger a partial withdrawal under ERISA § 4205(b)(2)(A)(i). And, depending on what your contribution base unit history has been, the liability may be a much greater portion of the total liability than one might expect based on the size of the group (or it could be much less). Estimating that would require a bit of actuarial analysis. Second, before you attempt to pull out some (but not all) of the nonunion employees, you would want to make sure that didn't violate the special agreement which presumably defines which nonunion employees are included. Also, in all of this, you need to be careful not to do something that the plan can successfully contend was a transaction with a principal purpose of avoiding or evading withdrawal liability (see ERISA § 4212©).

The Opinion Letters that PBGC issued for many years (but now appears to have discontinued) often include useful information. Unfortunately, they aren't very well indexed. To help remedy that, we have put together a summary of PBGC Opinion Letters related to multiemployer issues. This report includes a brief summary of the Opinion Letters related to multiemployer plans along with the text of the Opinion Letters. See http://www.dexhof.com/PBGCOpinionLetterSummary.pdf

The benefit suspension is disregarded in calculating withdrawal liability in the same way that the critical status benefit cuts under PPA are disregarded. In other words, suspending benefits does not reduce the employer's withdrawal liability. However, if the employer's withdrawal occurs more than ten years after the effective date of the benefit suspension, then the withdrawal laibility is calculated based on the reduced benefits. This issue is covered in § 305(g)(1) of ERISA as amended by the Multiemployer Pension Reform Act of 2014.

There's no blanket exemption for multiemployer plans but there is a rule that allows you to exclude bargained employees from testing (see Internal Revenue Code § 410(b)(3) which is referenced by § 401(a)(4)). That generally makes such testing unnecessary. However, multiemployer plans do need to worry about discrimination testing with respect to any nonunion employees covered under the plan (e.g., union officials covered under a special agreement).

The wording in the bill is "No benefits based on disability (as defined under the plan) may be suspended under this paragraph." I assume that they're referring to disability benefits already in pay status rather than benefits for those who haven't yet become disabled. But we may need to wait for IRS guidance before we know exactly what was intended.

Here is a link to a summary of the Multiemployer Pension Reform Act of 2014 that Dexter Hofing LLC has prepared. This summary focuses primarily on the changes of interest to contributing employers. http://dexhof.com/DexterHofingMPRA2014.pdf DexterHofingMPRA2014.pdf

I'm not an attorney but if you're paying the full dollar amount of the liability determined under § 4211, I don't see how the plan could reject it if you pay within 60 days of receiving the demand letter. As a practical matter, most plans will negotiate some sort of discount if you pay in a lump sum and such negotiations typically take months to complete. I've seen plans put in ridiculous deadlines like the 10-day one you cite. I don't think there's any legal basis for that deadline, however. In any event, if there's any potential risk of a mass withdrawal, you really don't want to pay in a lump sum without having a settlement agreement that covers what happens if you later get included in a mass withdrawal. Otherwise you can end up in the position where you essentially lose the value of the lump sum you paid.

From the facts you've presented, I don't see why the sale of Unit A will trigger a partial withdrawal (unless it's expected to trigger a 70% decline). Under which of the partial withdrawal rules do you expect a partial withdrawal to be triggered?

This is really a question for the lawyers and not for an actuary but I'll give my thoughts. To comply with the Taft-Hartley Act they simply must have employer trustees. If there are no employer trustees, any employer contributions to the plan would be a violation of Taft-Hartley potentially subject to penalties (criminal penalties, I believe). Some plans I'm aware of have gone the route of finding people such as retired executives from the industry, retired bankers, etc. to serve as paid trustees.

It's very common for plans to settle withdrawal liability for a discounted lump sum. This is particularly common in cases in which the 20 year payment cap applies. ERISA § 4219©(4) gives the employer the right to settle the withdrawal liability for the value of the unpaid withdrawal liability payments although it doesn't specify the interest rate to be used for determining the present value. Some plans are willing to settle for the present value of the payments determined at their funding interest rate. Others will discount at a somewhat lower rate -- sometimes as low as the PBGC rate. In general, I am not aware of any plans that have formal policies regarding discounting lump sum payments. Instead they tend to view each offer based on the facts and circumstances. As a general rule, the trustees have a fiduciary obligation to collect the full withdrawal liability. However, there are generally arguments for discounting the liability if the employer is willing to pay an immediate lump sum. If there are any disputes pending, the lump sum will eliminate litigation risk and expense. The lump sum payment eliminates credit risk whether or not litigation risk exists. And there is less expense involved for the plan in collecting one lump sum as opposed to quarterly or monthly payments for many years. I'm not aware of any situations in which the validity of a discounted withdrawal liability settlement has been challenged. ERISA § 4224 would seem to give the trustees pretty broad latitude to make any settlement that was consistent with their fiduciary responsibilities.

For funding purposes, you don't count the contribution as an asset until it's received (see Internal Revenue Code § 431(b)(7)(A)). For withdrawal liability purposes, you need to look at the rules for the specific withdrawal liability method. For example, if you're using Rolling-5, you reduce the unfunded vested benefits by "the value as of the end of such year of all outstanding claims for withdrawal liability which can reasonably be expected to be collected..." (see ERISA § 4211©(3)(A)). Note that this would be the present value of expected payments -- not the withdrawal liability amount under § 4211 which could be significantly different, particularly if you're hitting the 20-year payment cap. On the other hand, if you're using the presumptive method you don't count withdrawal liability contributions in the assets until after the payments are made.

Yes, you do add the expense loading - see 29 C.F.R. § 4281.13(e).

I've never been involved in a situation quite like this but I've heard about situations where the plan concluded that 4204 had been complied with when the assets sold were pretty minimal. For example, the incoming contractor purchased a few desks, etc. from the outgoing contractor. Whether that complies with 4204 is a bit gray but if you do that at least you can argue that there has been an asset sale somewhere in the picture.

From what I've seen, auditors differ in the way they handle such amounts. Many auditors include the value of expected withdrawal liability payments in the financial statements (presumably reduced for payments that are expected to be uncollectable). Others don't count any such payments until received by the plan. But note that regardless of what the auditor does, for minimum funding purposes withdrawal liability payments shouldn't be counted until they're actually received by the plan (see Internal Revenue Code § 431(b)(7)(A)). For withdrawal liability purposes, the treatment of receivables depends on the withdrawal liability method being used. For Rolling-5, for example, the unfunded vested benefits are reduced by the amount of outstanding claims for withdrawal liability which can reasonably be expected to be collected. On the other hand, under the presumptive method, withdrawal liability payments aren't taken into account until actually received by the plan.

Unless the partial withdrawal was very recent, they're presumably past the deadlines for contesting the prior partial withdrawal so, right or wrong, they're probably stuck with it. The subsequent complete withdrawal should have no effect on the prior partial. The liability payments for the prior partial will simply continue unchanged.