April 6, 2015 § Leave a comment


1. Unpredictable Contingent Event Benefits

For plans with an AFTAP under 60%, no enhanced benefits can be paid on account of layoff, plant closing or other unpredictable contingent event. (Code § 436(b).)

2. Increased Benefits Due to Amendment

For plans with an AFTAP under 80%, no amendment increasing plan benefits can take effect. (Code § 436(c).)

3. Funding Limits on Accelerated Benefits

a. For plans with an AFTAP under 60% or a plan sponsor in bankruptcy and an AFTAP under 100%, no lump sums can be paid. Other accelerated benefits are also limited. (Code § 436(d)(1) and (2).)

b. For plans with an AFTAP under 80%, the amount of a lump sum distribution is limited to either half of the value of the lump sum or the value of the lump sum equivalent of the benefit guaranteed by the PBGC, whichever is less. (Code § 436(d)(3).)

c. Accelerated benefits for this purpose include lump sum payments (most commonly), social security leveling options, term certain annuity forms of benefit, benefits with retroactive annuity starting dates and lump sum death benefits.

4. Future Benefit Accruals

For plans with an AFTAP under 60%, no new benefits can accrue under the present benefit formula. (Code § 436(e).)


Whether limits apply depends on the “adjusted funding target attainment percentage” or “AFTAP” of the plan. To determine the AFTAP, the “funding target attainment percentage,” or FTAP is calculated first.

1. Calculation of FTAP

The FTAP for the plan is a ratio expressed as a percentage. The numerator is the value of plan assets for the year, which excludes the amount of pre-funding balance and funding standard carryover balance (unless the plan has a binding agreement with the PBGC that contains certain provisions limiting application of the specified balances). (Code §§ 430(d)(2) and 430(f)(4)(B).) The denominator for the year is the funding target for the plan year, determined without regard to the special rule for at-risk plans in Code § 430(i)(1). The funding target is the present value of all benefits accrued or earned under the plan as of the beginning of the plan year. (Code § 430(d)(1).)

2. Valuation Date

The valuation date for determining FTAP is the first day of the plan year. Assuming a calendar year plan, the valuation date is January 1. (Code § 430(g)(2)(A).)

However, for purposes of the unpredictable contingent event limitation of Code § 436(b) and the amendments increasing benefits in Code § 436(c), the AFTAP is determined as of the time immediately prior to the event or effective date of the amendment.

Timing of the determination of AFTAP for purposes of Code section 436(d) and Code § 436(e) limitations is discussed below.

3. Adjustment of FTAP to Calculate AFTAP

The FTAP is adjusted (as one would expect) to determine the AFTAP. This adjustment calls for adding to both the numerator and the denominator are adjusted by adding back in the value of annuity contracts purchased for non-highly compensated employees during the two preceding plan years. (Code § 436(i).) In her excellent article, “PPA New Benefit Restrictions for Defined Benefit Plans,” Kathryn J. Kennedy states, “I surmise that the AFTAP was adjusted for any prior two-year annuity contracts to reduce fluctuations year-to-year for small employers. Because highly compensated employers tend to elect lump sum distributions in lieu of annuity contracts, the statute adds back in only the annuity contracts purchased for non-highly compensated employees.” (The article is available here: As an aside, I find this interesting because in my experience the vast majority of employees offered a lump sum option – not just highly compensated employees – will take a lump sum, particularly if it is calculated on somewhat favorable terms.

The bottom line for larger plans is that FTAP and AFTAP will likely be the same.


The question that I am asked more than any other regarding Code § 436, is when contributions must be made to avoid imposition of limits. That ties, logically, to when the limits would – absent additional contributions – take effect. The application of benefit restrictions is determined as of a Code “section 436 measurement date,” which is a term defined in Treas. Regs. § 1.436-1(j)(8). The definition in the regulations is somewhat circular, “A section 436 measurement date is the date that is used to determine when the limitations of §§ 436(d) and (e) apply or cease to apply, and is also used for calculations with respect to applying the limitations of paragraphs (b) and (c) of this sections. See paragraphs (h)(1)(i), (h)(2)(iii)(B), (h)(2)(iv)(B) and (h)(3)(i) of this section regarding section 436 measurement dates that result from application of the presumptions under paragraph (h) of this section.”

What this means is that the “section 436 measurement date” occurs whenever an actuary provides an AFTAP certification. In the absence of a certification, various presumptions apply.
If the AFTAP of a plan will be below the 80% or 60% threshold, restrictions apply when the certification is issued to the plan administrator or when the applicable presumption takes effect. There was, several years ago, a great deal of discussion about whether the enrolled actuary for a plan had a duty to issue his certification as soon as his work was completed or whether he could hold off in issuing his certification. The Society of Actuaries Practice Note, “Preparing a Certification of the Adjusted Funding Target Attainment Percentage (AFTAP) for a Pension Plan (Revised December 2009)” states, “No official guidance has been issued on this question, and the EA may wish to consult with legal counsel. However, in general, it appears that the plan administrator controls the timing of AFTAP certifications and the EA will follow the plan administrator’s instructions.” (See Q&A 8; the Practice Note is available here:

This Practice Note contains a series of questions and answers, all of which basically conclude that it is the duty of the actuary to explain the consequences of not issuing the actuarial certification and the duty of the plan administrator to act in the best interests of plan participants and beneficiaries in administering the plan, which would presumably require the plan administrator to obtain a timely actuarial certification. However, there may be legitimate reasons for a plan administrator to ask the actuary to delay issuance of the certification – either to communicate with plan participants or to take steps to ensure proper administration of the plan following the imposition of restrictions.
As the American Academy of Actuaries notes in its February 5, 2015 letter to the Treasury and IRS on “Potential Improvements in IRC § 436 Benefit Restriction Rules” that timing rules are among the more difficult aspects of Code § 436 to administer. Once a certification is obtained, then the benefit restrictions under Code § 436 must be implemented as of the first annuity starting date that falls after the AFTAP certification. The letter states, “This requirement does not allow lead-time to modify election forms and systems and therefore does not allow for the normal election time frames using those modified forms and systems. Election forms must be provided to participants at least 30 days before the ASD. However, plan administrators commonly provide them earlier so that trustees can set up an annuity for payment commencing on the chosen ASD (this typically requires at least two weeks advance notice). If the paperwork is provided to the participant only 30 days before the ASD, there is very little time for a participant to make a thoughtful decision and return the paperwork in time to actually receive a payment on the selected ASD.” (The letter is available here: )

The difficulties of administration may be one reason that a plan administrator would ask an actuary to hold off on providing a certification. For example, the plan administrator might choose to issue a communication to participants that would allow those contemplating retirement to retire and elect lump sums before restrictions took effect. While this could (depending on how the lump sums were calculated) result in some actuarial loss to the plan, it could also arguably be in the best interests of participants and beneficiaries to ensure orderly and predictable plan administration.
Note that a communication to plan participants in advance of a benefut restriction would not satisfy the requirement under § 101(j) of ERISA to provide a notice to participants and beneficiaries within 30 days after the plan becomes subject to restrictions. However, the short time window to provide the § 101(j) notice might be another legitimate reason for a plan administrator to ask an actuary to hold off on issuing an actuarial certification that would trigger benefit restrictions.


The section 436 mesurement date for a calendar year plan is January 1. However, the actuarial certification is unlikely to be available on January 1. Therefore, the AFTAP is presumed to be equal to the prior year’s AFTAP until April 1.

At April 1, if the plan was not subject to benefit limitations in the prior year, but the plan was within 10 percentage points of an applicable threshold, then the AFTAP from April 1 is presumed to be the prior year AFTAP minus 10 percentage points. (Code § 436(h)(3).) This presumption will apply until October 1 or until a current year AFTAP is certified, whichever occurs first.

If no current year AFTAP has been certified by October 1, then the AFTAP is presumed to be less than 60%. (Code § 436(h)(2).)

This encourages the plan administrator to obtain an AFTAP certification before April 1.


Supreme Court Decision in Quality Stores

March 27, 2014 § Leave a comment

The Supreme Court decision in U.S. v. Quality Stores, Inc. just came out. For once, I think the Court’s analysis is completely right.

First, the severance payments made by Quality Stores, which were based on the level and years of service of the employee, were wages subject to FICA. Well, no kidding. Code section 3402(o) talks about “any supplemental unemployment compensation benefit paid to an individual”. Nothing about the severance payments in Quality Stores was actually tied to whether or not the individual was receiving state unemployment benefits, so the severance payments are not supplemental unemployment compensation benefits within the meaning of Code section 3402(o).

Second, the Court suggests that CSX Corp. v. U.S., 518 F.3d 1328 (CA Fed. 2008), was wrongly decided. The Court states, “In concluding, the Court notes that the IRS still provides that severance payments tied to the receipt of state unemployment benefits are exempt not only from income-tax withholding but also from FICA taxation. See, e.g., Rev. Rul. 90-72, 1990-2 Cum. Bull. 211. Those Revenue Rulings are not at issue here. Because the severance payments were not linked to state unemployment benefits, the Court does not reach the question whether the IRS’ current exemption is consistent with the broad definition of FICA.”

For the last couple of years, I’ve been advising clients that despite the sales pitch they’re receiving from various big name firms (yeah, I’m talking about Morgan Lewis and Crowell & Moring), making claims for refunds was a fool’s errand unless actual, honest-to-goodness supplemental unemployment benefit plans were involved — which was never the case. Plain vanilla severance payments are subject to FICA. Why this was ever in doubt is unclear to me, but I think an 8 – 0 decision (Kagan did not take part) should clear it up for everyone. Being right feels good.

Time Procedures with Two Levels of Claims Review

October 28, 2013 § Leave a comment

If a plan such as a disability plan provides for two mandatory levels of appeal, the participant must have 180 days from the date of the initial determimation within which to file his first level of appeal.

The plan does not have to provide 180 days for the participant to file his second mandatory appeal following the determination upon the first level of appeal. While there is nothing in the DOL regulations that would lead one to this conclusion, the court in Price v. Xerox Corporation, 445 F.3d 1054 (8th Cir. 2006), held that only the initial denial of benefits is to be considered an “adverse benefit determination” and a decision on appeal is not. (Yes, this seems to defy all logic and common sense, but that is, in fact, what the opinion says.) The court then holds, “Accordingly, an ERISA plan need not provide a claimant with at least 180 days to file a mandatory second appeal.”

The Xerox plan in question provided 60 days in which the participant could file the mandatory second level of appeal, which the claimant argued was unreasonable. The opinion states, “Price contends that 60 days is unreasonable, because it restricts new evidence and adequate dialogue with the administrator. However, this argument ignores the fact that he had over 180 days to present evidence and contest the denial in the first appeal.”

While this was true for Mr. Price, who had requested and received a 90-day extension to file his first appeal, what about a participant who files his first appeal promptly? What if the plan only allows 30 days for the second appeal? In that case, a participant who filed his first appeal in 15 days could be limited to a total of 45 days in which to appeal.

Nevertheless, I suspect a court would still apply the holding in Price to preclude that participant from having additional time to prepare a second appeal. As a result, participants would be wise to prepare their initial appeals carefully, and plan sponsors would be wise to shorten the time frame for the second level of appeal to 60 days (for which Price serves as precedent) or even less for plan sponsors who are more aggressive.

Change in Health Plan Elections Due to Incarceration of Spouse

October 25, 2013 § Leave a comment

I had a client call today with a question I’ve never considered before. The client’s employee is married, and the employee’s spouse is about to be incarcerated for about a year. The employee has asked if she can drop the spouse from health coverage. The employee has also asked if she will have a special enrollment right under HIPAA and be able to reenroll her spouse when he is released.

I did not find IRS guidance that specifically adresses the issue of changes in coverage related to incarceration, but the prevailing view (based on internet boards such as, which I generally am hesitant to rely too much on) seems to be:

1) When a spouse or dependent is incarcerated, the incarceration itself is not a life event that would permit the employee to change coverage, and the inmate can only be dropped during open enrollment. (Apparently, the government provides medical care to inmates but can seek reimbursement from a plan that covers the inmate, and so the government does not want to encourage dropping coverage.) The fact that incarceration probably coincides with loss of employment doesn’t seem to change this because the employee’s loss of employment would generally not be a reason to drop coverage, and it’s possible that the employee is covered under the spouse’s plan because he either wasn’t employed or his employer didn’t offer him coverage.

2) If an inmate is housed in a facility that is “out of area,” the change in residence could be a change that would permit the employee to drop coverage, but many provider networks (Anthem, BCBS, UHC) are effectively nationwide, so the prison is likely to be in an area that the network covers.

3) When an inmate is released from incarceration, I think that generally would be considered a loss of government coverage, and the employee would be allowed to make a change to cover the spouse or dependent again based on the inmate’s loss of coverage. The authority for this is in Treas. Regs. section 1.125-4(f)(5), which is a final regulation (not part of the 2007 proposed regulations) and refers to loss of “group health coverage sponsored by a governmental or educational institution”.

My client is going to tell the employee that she can’t drop her spouse from coverage (at least until a divorce, which seems like it could happen in these circumstances) but she can decide not to reenroll him at their upcoming open enrollment and then wait until he’s released to add him again.

Has anyone else answered this question differently? If so, why? With the shockingly high percentage of our population that we incarcerate, it would be really helpful if the IRS would issue some guidance on this issue.

Interaction of Code Section 409A and Code Section 457(f)

October 23, 2013 § Leave a comment

I don’t do a lot of work for non-profits, but I recently drafted an employment agreement for a honcho at a fairly well-known non-profit, and I learned a few things about the interaction of Code section 457(f) and Code section 409A that I’d like to note here. My facts were that the honcho was entitled to a pension payment when he separates from service. The entitlement to this payment “vested” after he worked for the entity for five years. Under Code section 457(f), the amount would therefore be taxable to him after five years. As a part of his new contract, he wanted the right to payment at that point without having to wait until he separates from service. My initial sense was that this would be a problem under Code section 409A, but I wanted to look at how the interaction between Code sections 457(f) and 409A played out.

First off, there is some language in the preamble to the final 409A regulations about this, which states:

For purposes of determining the the time of payment, the term ‘payment’ generally refers to an actual or constructive payment of cash or property. However, the final regulations provide that for purposes of the short-term deferral rule, an amount is treated as paid when it is included in income under section 457(f) whether or not an actual or constructive payment occurs. Accordingly, where the income inclusion under section 457(f) stems from the lapse of a substantial risk of forfeiture that is also treated as substantial risk of forfeiture for purposes of section 409A, the amount included in income will be considered a short-term deferral for purposes of section 409A.

(See 72 FR 19235.) At first, I wasn’t sure what this meant, and the language in the actual regulations did not clarify it for me. The regulations themselves address section 457(f) plans in three places. First. Treas. Regs. section 1.409A-1(a)(4) provides in part:

A nonqualified deferred compensation plan under section 457(f) may constitute a nonqualified deferred compensation plan for purposes of this paragraph (a). The rules of section 409A apply to nonqualified deferred compensation plans separately and in addition to any requirements applicable to such plans under 457(f). […] For purposes of the application of section 409A to a plan to which section 457 applies, a payment under the plan generally means the provision of cash orproperty to the service provider, provided that for purposes of application of the short-term deferral rule set forth in paragraph (b)(4) of this section, the inclusion in income of an amount under section 457(f) is treated as payment of the amount.

The short-term deferral rule in Treas. Regs. section 1.409A-1(b)(4) cited above states, “A payment is treated as actually or constructively received if the payment is includible in income, including if the payment is includible in income under section 83, the economic benefit doctrine, section 402(b), or section 457(f).”

Thus, amounts included in income under Code section 457(f) are not treated as deferred for purposes of Code section 409A, which seems fairly straightforward until you reach the exceptions from the antiacceleration rules. Treas. Regs. section 1.409A-3(j)(4)(iv) provides:

A plan subject ot section 457(f) may provide for an acceleration of the time or schedule of a payment to a service provider, or a payment may be made under such a plan, to pay Federal, state, local and foreign income taxes due upon a vesting event, provided that the amount of such payment is not more than an amount equal to the Federal, state, local and foreign income tax withholding that would have been remitted by the employer if there had been a payment of wages equal to the income includible by the service provider under section 457(f) at the time of vesting.

If the amounts included in income under a 457(f) plan are not treated as deferred compensation, then why should there be an exception to the anti-acceleration rules for tax withholding when such amounts are included in income? I was honestly mystified.

I discussed this with a colleague who does significantly more non-profit work than I do, and his explanation is that the definition of “substantial risk of forfeiture” under Code section 457(f)(3)(B) is limited to future performance of substantial services. Thus, once future performance of substantial services is no longer required — for example where the participant has satisfied a five-year vesting requirement (which was the case for the executive I was working with), the amount is includible in income for purposes of section 457(f). However, it is possible to conceive of vesting requirements that would not satisfy section 457(f), such as vesting upon attainment of certain performance commitments, that would require an amount to be included in income under Code section 457(f) but also not payable until a future date.

If an amount ceases to be subject to a substantial risk of forfeiture under section 457(f)(3)(B) and is included in income, then the amount is no longer treated as deferred for purposes of Code section 409A, but rather as paid. As a result, the actual time of payment could be moved up from separation from service to the effective date of the new contract.

This was a surprising outcome for me, but a happy one for the honcho.

Quick update: while looking for something else last night, I came across a brief article about the interaction of Code sections 409A and 457(f) here: The conclusion (as much as there is one because the writing is muddy at best) seems to be, “Such ‘vested’ [for purposes of 409A but not 457(f)] amounts might still be eligible for deferral of taxation under Section 409A, provided that distribution cannot occur until a statutory permissible event. However, if amounts are vested under both Section 409A and Section 457(f), taxation under Section 457(f) will be required.” This article either breezes by (or perhaps misses entirely) the bigger issue that an amount is not subject to Code section 409A after having been included in income under Code section 457(f). But perhaps someone else will see value that I don’t here? If I’ve missed true genius in this article, please let me know.

Differential Wages

October 22, 2013 § Leave a comment

This post is just a few short thoughts about differential wages, which are the wages that some employers will pay to reservists called to active duty in order to make up the difference between a reservist’s military pay and his or her civilian pay.

Section 3401(a) of the Code defines “wages” for income tax withholding purposes, and the FICA and FUTA definitions in Code sections 3121(a) and 3306(b), respectively, contain very similar definitions. Prior to 2008, differential pay did not fall within the definition of “wages” because the payments were made to idividuals who were not performing services for the employer. The Heroes Earnings Assistance and Relief Tax Act of 2008 (the “HEART Act”) added section 3401(h) to the Code. Section 3401(h) defines the term “differential wage payment” and provides that differential wage payments are treated as a payment of wages from the employer to the employee for income tax withholding purposes. However, no similar Code sections were added to cause differential wage payments to be treated as “wages” for FICA and FUTA purposes.

Rev. Rul. 2009-11 provides that differential wage payments are supplemental wages because they are not payment for services in the current pay period. (This may be true for some pay periods, but it would be possible for someone to work one week of a two-week pay period and to receive differential wages for the second week.) Still, on the basis of Rev. Rul. 2009-11, the employer can (and should) treat the differential wage payment as supplemental wages for income tax withholding purposes. As a result, the employer can withhold at either the aggregate W-4 rate or the supplemental wage withholding rate, which is 25% unless supplemental wages for the calendar year exceed $1,000,000, which seems unlikely. Rev. Rul. 2009-11 also provides that differential wage payments must be reported as income on Form W-2.

That’s the black-letter law as it stands, and the Revenue Ruling is very helpful. Going a step further, into more interesting territory, I have a client that continues to calculate 401(k) contributions and match on the employee’s aggregate wages so that employees’ retirement savings will not be disrupted by deployment. I’m not sure that this works as a technical matter, but their plan, which obviously provides for this deferral and match, just received a determination letter. The employer feels strongly about this issue because if only the differential wages were counted, the employee would be eligible for a substantially lower match, and they view the point of differential wages as allowing the employees called to active duty to avoid financial disruption.

I was surprised by the level of emotion that the employer mustered on this topic. Certainly, I see his point. I also suspect that he’s right when he says the Service will never, ever, ever want to pick this fight with an employer so committed to his employees and his country. I felt like an unpatriotic ninny for advising against this. I just can’t find the basis for it in Code.

Limits on Catch-Up Contributions Are Personal Limits

August 29, 2013 § Leave a comment

This is a post where I will confess to being wrong and learning something new. This time I learned about catch-up contributions.


I was asked the following question about a senior executive starting a new job – can he make catch-up contributions in the new employer’s plan if he has already contributed the maximum catch-up contribution in his former employer’s plan for the plan year?


Well, I looked at Code § 414(v) and the accompanying Treasury Regulations and saw what looked like a plan limit, not an individual limit. I also checked the Tax Management Portfolios, where I read, “Unlike the § 402(g) limit, the § 414(v) catch-up limit does not apply on a taxpayer-by-taxpayer basis, but rather on an employer-by-employer basis. [Cite to Code § 414(v)(2)(D) and Treas. Regs. § 1.414(v)-1(f)(1).] As a result, a catch-up eligible participant that works for two unrelated employers may make catch-up contributions that in total exceed the general dollar limit on catch-up contributions.”

Based on this, I gave advice to my client, and was – SMACK – contradicted by the guy’s advisor, who said that the limit is a personal limit like the Code § 402(g) limit. This made me go back and look again, and what I found was very interesting, so I’d like to share it here in the hope that either (1) now I understand it and this will be helpful to someone else or (2) I’m wrong and someone will explain to me exactly why. There are definitely respected practitioners on both sides of this issue, but at this point, I am inclined to agree with the advisor that the limit is a personal limit and is note merely like the Code § 402(g) limit — it is the the Code § 402(g)(1)(C) limit.

First, let me set out the argument in favor of allowing the executive to make a second catch-up contribution in his new employer’s plan.

  • The limit in Code § 414(v)(2)(A) is specifically a plan limit. However, under Code § 414(v)(2)(D), all plans other that 457 plans maintained by a single employer are aggregated, and all 457 plans maintained by an employer are aggregated. Employer for this purpose is the controlled group, so that if an employer had two separate 401(k) plans in the controlled group (for example for Division A and Division B), they would be treated as a single plan, and no employee could contribute more than $5,500 (or whatever the annual limit for the year in question is) total catch-up contributions to both plans. There is nothing about aggregation of plans by unrelated employers. Thus, if I worked for three unrelated employers and participated in a 401(k) plan at each employer, it looks like I could make $5,500 in catch-up contributions to each plan.
  • However, if an individual is employed by two separate employers and makes catch-up contributions to two plans, the amount that is excludable from taxable income is limited to the Code § 402(g)(1)(B) limit, which is what we all think of as the Code § 402(g) limit ($17,500 for this year) plus the dollar amount under Code § 414(v)(2)(B)(i). (See Code § 402(g)(1)(C).) Thus, an individual who makes catch-up contributions to two plans of two separate employers will not be able to defer more than $5,500 of catch-up contributions on a pre-tax basis.
  • The Technical Explanation of the Job Creation and Worker Assistance Act of 2002, prepared by the Staff of the Joint Committee on Taxation, provides the evidence that I found most persuasive in support of this side of the argument. The Technical Explanation states:

Additional salary reduction catch-up contributions.–Under the Act, an individual aged 50 or over may make additional elective deferrals (“catch-up contributions”) to certain retirement plans, up to a specified limit. A plan may not permit catch-up deferrals in excess of this limit. The provision clarifies that, for this purpose, the limit applies to all qualified retirement plans, tax-sheltered annuity plans, SEPs and SIMPLE plans maintained by the same employer on an aggregated basis, as if all plans were a single plan. The limit applies also to all eligible deferred compensation plans of a government employer on an aggregated basis.

Under the Act, catch-up contributions up to the specified limit are excluded from an individual’s income. The provision also clarifies that the total amount that an individual may exclude from income as catch-up contributions for a year cannot exceed the catch-up contribution limit for that year (and for that type of plan), without regard to whether the individual made catch-up contributions under plans maintained by the more than one employer.

(See page 40 of the Technical Explanation, available here: ) The last sentence of the second paragraph suggests to me that an individual with more than one employer might make catch-up contributions to multiple plans, but that individual could only exclude one amount. In addition the first sentence says, “… catch-up contributions up to the specified limit are excluded” which implies that there can be catch-up contributions beyond the specified limit.

Yeah, all that sounds so reasonable that one could even be forgiven for thinking that it’s right. Can you see why I was thinking this way?


As I was still feeling good about my answer, I came across a client alert from the Groom law group (available here: that cites Proposed Treas. Regs. § 1.414(v)-1(g)(2), which stated, “Maximum excludable amount. If a catch-up eligible participant participates in two or more applicable employer plans during a taxable year, the total amount of elective deferrals under all plans that are not includable in gross income under this paragraph (g) because they are catch-up contributions shall not exceed the applicable dollar catch-up limit under paragraph (c)(2)(i) of this section for the taxable year.” (The proposed regulations are here: )

This sent me back to the final regulations, and this language is not in the final regulations. I think the drafters of the regulation thought it was unnecessary by then.

Catch-up contributions were introduced by the Economic Growth and Tax Relief Reconciliation Act of 2001, or “EGTRRA”. EGTRRA said participants could make catch-up contributions in 401(k) plans. However, EGTRRA did not amend Code § 402(g) to exclude the contributions from income. Initially, Code § 414(v)(3) (A)(i) included Code § 402(g) in the list of limits that did not apply in the case of catch-up contributions.

That didn’t last long. Section 411(o) of the Job Creation and Worker Assistance Act of 2002 (which is Technical Corrections) amended Code § 402(g) to add the current Code § 402(g)(1)(C), excluding catch-up contributions from income. The text of the Job Creation and Worker Assistance Act of 2002 (P.L. 107-147) is here: .

And this caused me to go back and look more closely at the statute, which is when I had my epiphany: Code § 414(v)(1) permits catch-up eligible employees to make additional “elective deferrals”.

For this purpose, “elective deferral” is defined in Code § 414(u)(2)(C), which provides, “… the term elective deferral has the meaning given such term by section 402(g)(3); except that such term shall include deferral of compensation under an eligible deferred compensation plan (as defined in section 457(b)).”

Code § 402(g)(3) defines elective deferrals as the sum of (1) employer contributions “not includible in gross income for the taxable year” under Code § 402(e)(3) plus (2) employer contributions “not includible in gross income for the taxable year” to a simplified employee pension plus (3) employer contributions used to purchase an annuity under Code § 403(b) plus (4) elective employer contributions to a simple retirement account.

Because the amounts deferred under a 401(k) plan must be “not includible in gross income for the taxable year” in order to count as elective deferrals, the limits on exclusion from income in Code § 402(g)(1) mean that amounts in excess of the annual limit in Code § 414(v)(2)(B)(i) cannot be catch-up contributions by definition.


First, there are no dire consequences for the plan(s) if a participant makes multiple catch-up contributions to multiple plans of unrelated employers. If a participant makes catch-up contributions in multiple plans of different employers, then each plan is fine.

Second, the amounts that exceed the Code § 402(g) limit will be excess deferrals and will be identified because of how they will be reported on the participant’s Form W-2. Thus, if the participant doesn’t ask the plan to distribute them, he or she will owe additional taxes.

Third, it is not possible to avoid this result by having after-tax catch-up contributions since catch-up contributions are, by definition, “not includible in gross income” or, alternatively, may be designated as Roth contributions, but in either case, they are subject to the Code § 402(g) limit.

So the advisor was right and I was wrong – the executive could not start over with his catch-up contributions when he changed employers.