Quantcast
Channel: Wealth Management - Trusts & Estates
Viewing all articles
Browse latest Browse all 733

Drafting Trusts for Children After the SECURE Act

$
0
0

Clients need to carefully weigh income taxes against asset protection concerns

On Dec. 20, 2019, the Setting Every Community Up for Retirement Enhancement (SECURE) Act was enacted.1 The new legislation made significant changes to the taxation of individual retirement accounts and other retirement plans subject to Internal Revenue Code Section 401(a)(9). Except for the eligible designated beneficiary (EDB),2 the life expectancy method is no longer the default method for determining required minimum distributions (RMDs).3 The new rules require that RMDs are distributed to non-EDB individual beneficiaries over 10 years beginning at the plan participant’s death, the EDB’s death or the minor child’s attainment of majority.

The SECURE Act played havoc with many clients using the much-touted stretch individual retirement plan technique to optimally dribble out RMDs over the designated beneficiary’s (DB’s) life expectancy. A 10-year (versus life expectancy) payout dramatically diminishes income tax deferral for anyone who isn’t an EDB.

IRC Section 401(a)(9)(E)(i) provides that only an individual can be named a DB but doesn’t explain how a trust or its beneficiaries can qualify. Internal Revenue Service regulations allow so-called “conduit” and “accumulation” trusts to funnel distributions during an applicable distribution period (ADP) to individual beneficiaries.4 A conduit trust requires that RMDs and other plan distributions be distributed to the child, the sole DB. An accumulation trust allows retirement plan distributions to be retained in trust to benefit the child and other beneficiaries.

The trustee of an accumulation trust typically can make discretionary distributions based on subjective impressions of whether the child exhibits spendthrift behavior or is susceptible to drugs, unstable marriages or creditors. The SECURE Act allows the life expectancy method for EDBs in conduit trusts, as well as a narrow range of accumulation trusts. Accumulation trusts are preferable to conduit trusts from an asset protection point of view. However, the former structure often generates more income taxes on undistributed income due to compressed trust income tax brackets.5   

Life Expectancy Method 

The SECURE Act changed the landscape for trusts holding retirement plan assets benefiting children. Because the minor child is an EDB until majority, using the life expectancy method for the pre-majority period results in small RMDs in the 1%-to-2% range. The 10-year rule then kicks in after majority is achieved. The problem is that the definition of “majority” is about as clear as mud. 

Most states define “majority” as the time the child attains age 18 or 21. Still, the regulations paradoxically provide that a child may be treated as having not reached the age of majority if the child hasn’t completed a specified course of education and is under the age of 26.6 Who knows what “course of education” means and the needed academic change to lose EDB status before age 26. What if: there’s a gap in college studies; on graduation, the child takes a few additional courses toward a master’s degree; or the studies are less erudite and mere hobby interests?

The extra income tax deferral gained by dragging out the majority determination is important but not earth-shattering. Most RMDs are distributed during the pendency of the 10-year rule. The younger the child, the longer the tax deferral. Many parents will prefer accumulation trusts and the immediate beginning of the 10-year rule (contrasted with conduit trusts that delay such rule until majority) as the price to pay for enhanced asset protection.  

A trust holding retirement benefits for minor children is an unlikely event that typically occurs if both parents die or on a single parent’s death. The presumed low income tax rates of a non-income-producing minor (possibly without the “kiddie tax”7) encourages examining whether RMDs should be accelerated rather than deferred. Does it always make sense to arrange taxation of the bulk of the retirement plan to 10 of the child’s working years (beginning sometime after ages 18-to-26), when such RMDs could have been taxed over 10 of the child’s earlier, non-working years?8

Disabled and Chronically Ill Children

The SECURE Act added new categories of EDBs for “disabled” and “chronically ill” beneficiaries that can apply to children.9 Qualifying as an EDB under these categories is preferable to the age-of-majority rule, as the life expectancy method presumably continues for the child’s lifetime. EDB status for a disabled/chronically ill beneficiary is determined at the participant’s death.10 However, it appears that a child who becomes disabled before the age of majority can continue using the life expectancy method until the termination of such disability.11

A disabled EDB is unable to engage in any substantial gainful activity by reason of any “medically determinable physical or mental impairment which can be expected to end in death or to be of long-continued and indefinite duration” (following the Social Security definition of “disability”).12 A chronically ill EDB must be unable to perform at least two activities of daily living for 90 days due to a loss of functional capacity or need substantial supervision to protect from threats to health and safety due to severe cognitive impairment.13 This latter definition is more geared toward a loss of functional capacity relating to the elderly, not young children. Under both definitions, effective certification procedures need to be developed.

Section 401(a)(9)(H)(v) defines an “applicable multi-beneficiary trust” (AMBT) as a trust with multiple DBs, of which at least one is disabled/chronically ill. AMBTs have the green light to qualify the disabled/chronically ill beneficiary as an EDB if one of two special rules are met.  

The first special rule requires the AMBT to be divided immediately (a vertical trust slice) at the plan participant’s death into separate trusts for each beneficiary.14 The subtrust for the disabled/chronically ill beneficiary is then tested separately for EDB status (negating the IRS rule discussed below limiting separate account treatment for trusts). The sizzle of this AMBT variant is that it permits the participant’s living trust to be named as the retirement plan beneficiary. If allowed by the trust instrument, the trustee can disproportionately allocate retirement plan assets to the trust benefiting the disabled/chronically ill beneficiary (making these assets more valuable because of the longer income tax deferral).

The second special rule provides that the life expectancy method can be used on behalf of a disabled/chronically ill EDB if such beneficiary is the sole lifetime beneficiary (a horizontal trust slice).15 This rule permits many existing special needs trusts (SNTs) to attain EDB status, although some SNTs will need revisions eliminating non-disabled/chronically ill family members as current beneficiaries.

A nagging statutory construction problem AMBTs face is whether the life expectancy of an older individual beneficiary sets the ADP, rather than the expectancy of the disabled/chronically ill EDB.16 This interpretation seems to be against the statutory intent, yet the ambiguity exists, making it prudent to use drafting techniques to ensure that no DB is older than the disabled/chronically ill EDB.17

Counting Trust Beneficiaries

The regulations explain when and how to count a remainder trust beneficiary as a DB for determining the ADP. A critical rule is that for multiple DBs, the one with the shortest life expectancy is used to determine the ADP. Also, a remainder beneficiary won’t be counted unless they have any right (including a contingent right) to plan benefits “beyond being a mere potential successor.”18 The see-through trust rules19 also impose several mechanical conditions that must be met for trust beneficiaries to count as DBs.

For trusts not meeting the DB or see-through trust qualification rules, RMDs are due within five years of the participant’s death (5-year rule). But, if death occurs after the required beginning date, the participant’s remaining “ghost” life expectancy applies if it produces a longer ADP. For participants dying between (roughly) ages 73 and 80 after the effective date of the SECURE Act, a strategy is to purposefully flunk the see-through trust rules and qualify for ghost life expectancy if it produces a longer ADP.20  

Two regulatory examples form the primary analysis in counting trust beneficiaries. In Example #1, an accumulation trust directs trust income and discretionary distributions to the surviving spouse, with the spouse reserving the power to withdraw retirement plan income. The children, younger than the spouse, are remainder beneficiaries. Because distributions can be accumulated in the trust for the children’s benefit (even though access is delayed until the spouse’s death), both the spouse and children are DBs. The ADP is set by the life expectancy of the spouse, the beneficiary with the shortest life expectancy. Example #2 recites the same facts for a conduit trust requiring all retirement plan distributions to be paid to the spouse. The spouse was deemed the sole DB (meriting the benefit of a special rule21), with the children disregarded as mere potential successors.22

For accumulation trusts, the regulations are ambiguous as to when to stop counting trust beneficiaries down the chain until coming to ones who are mere potential successors. In the regulatory example, the children were younger than the spouse, but the regulations imply that they were adults entitled to outright distribution (because successor beneficiaries weren’t counted). Yet, what if the children were minors who could die before receiving full control of the benefits, with the deceased child’s share then passing to their estate (a non-individual flunking DB qualification)? The regulations are silent as to how successor beneficiaries are determined under these facts.

Private letter rulings that are theoretically precedential only for taxpayers receiving them have imperfectly filled in the gaps in the analysis. The mainstay IRS viewpoint is that we count trust beneficiaries until an individual remainder beneficiary is found who takes the benefits outright on the prior beneficiary’s death.23 

Assume an accumulation trust provides discretionary distributions benefiting Carter (a child, age 25), with outright distribution at age 35. If Carter dies before age 35, the trust property is retained in trust for his son, George (age 4), with outright distribution at age 35. But, if both Carter and George heartbreakingly die before the all-important age 35 date, the trust property is distributed to charity. In such a scenario, the government’s position seems to be that the charity is a countable beneficiary, thereby disqualifying DB status. The mere potential successor rule applies because plan distributions could (hypothetically) be accumulated in trust for the charity’s benefit. Is this conclusion sound from an actuarial standpoint? The charity only receives an interest in the retirement plan on the double death of Carter and George before age 35. In most circumstances, there’s a meager chance of that happening.

The American College of Trust and Estate Counsel (ACTEC) recently released exhaustive comments to the Treasury critiquing the existing RMD rules and making suggestions on how they could be updated after the SECURE Act. ACTEC proposed that the mere potential successor rule be replaced with a simpler and easier-to-administer approach counting only current beneficiaries or qualified beneficiaries under the Uniform Trust Code. Both of these theories eliminate the charity in the above example from constituting a countable beneficiary. Also advocated was a rule eliminating the counting of permissible appointees under powers of appointment (POAs) and solutions for numerous AMBT construction problems.24

Before the SECURE Act, there was a direct correlation between the ADP and each countable beneficiary’s life expectancy. Except for EDBs, the correlation is busted because the 10-year rule applies to DBs regardless of age (resulting in much simpler drafting for accumulation trusts). No longer does the planner have to worry that listing the participant’s 90-year-old brother as contingent beneficiary will dramatically truncate the ADP (this remains a concern for AMBTs). Unchanged is the risk that naming non-individual estates or charities as remainder beneficiaries will negate DB status unless preceding individuals receive outright distributions. If the client determines that the trouble of complying with these technical tenets aren’t worth it, they can be ignored with the 5-year or ghost life rules applying to nonconforming trusts.

Separate Account Rule

The regulations provide under the separate account rule that when a participant designates different beneficiaries to inherit a retirement plan, the resulting shares are analyzed separately for ADP purposes.25 For example, suppose a minor child and his adult sibling are named DBs on the beneficiary form. In that case, the minor is permitted to use his lifetime (not the adult sibling’s life or the SECURE Act’s 10-year rule) to calculate RMDs up to the age of majority.  

The IRS initially extended separate account treatment to multiple beneficiaries taking under a single funding trust (permitting the funding trust to be named DB on the beneficiary form). The IRS then abruptly slammed the door on this strategy in its 2002 regulations by mandating that separate account treatment wasn’t available for trust beneficiaries with respect to the trust’s interest in the retirement plan.26 Assume that a living (funding) trust named on the beneficiary form creates separate equal trusts for an adult child and a minor child. Under the pre-SECURE Act analysis, the minor’s subtrust was required to use the adult sibling’s older life expectancy to calculate RMDs, even though the subtrust was recognized as a separate entity for all other purposes.  

The planning to get around the separate account limitation (blessed by PLRs) is for the client to expressly name the subtrusts (not the living trust) on the beneficiary designation form. Under such circumstances, the separate account rule applies to each subtrust, with the child’s life expectancy (up to the age of majority) setting the ADP.27 While this on its face seems like an easy fix, clients usually require estate-planning attorneys to draft customized trust beneficiary designations. Even if this is done, many retirement plan custodians resist accepting such designations.  

For clients establishing living trusts and subtrusts, optimal beneficiary designations naming subtrusts on the designation form often aren’t made or are bungled. Talk about a trap for the unwary. Maybe after the SECURE Act, this doesn’t make a lot of difference if all trust beneficiaries are DBs (not EDBs), resulting in the 10-year rule applying without regard to the life expectancy method.

But, the stakes are much higher if the subtrusts include both EDBs and DBs. In our example, the minor child is an EDB, and the adult child a DB. If the living trust is named on the designation form and the post-death division creates the subtrusts, which payout method governs? It seems that the ADP should be based on the adult child’s 10-year payout period, because this interval is shorter than the minor child’s life expectancy and post-majority 10-year rule. However, the regulatory yardstick focuses on the DB with the shortest life expectancy, not the shortest payout period. Can it somehow be argued that the minor child’s payout period sets the ADP because it’s the only one that involves a life expectancy? Results vary significantly, depending on which criterion is used.28

Why should one group of clients not seeking expert help in drafting beneficiary designations be punished, compared to the group obtaining such help? Both are in the same position economically, with different tax results. Prohibiting the separate account rule from applying to single trusts holding retirement plan assets is bad tax policy that should be eliminated.29

Drafting Options

The simplest option for the adult child who’s attained majority (and isn’t disabled or chronically ill) is to designate the child as the outright retirement plan DB. Because the child isn’t an EDB, the 10-year rule applies, and income taxes are paid at the child’s income tax rates. The child has immediate control, but who cares if they’re sensible and not a spendthrift? A conduit trust can be used to keep benefits within the trust, possibly with the child as trustee, until all benefits are distributed after 10 years.

Naming the minor child directly as the outright DB ensures the life expectancy method, with unfavorable guardianship costs and remittance of funds to the beneficiary at a young age. Uniform Transfer to Minor Accounts are less costly but also subject to the latter problem.   

Conduit trusts for minor children are beneficial from a tax viewpoint. The child’s life expectancy is used, with the 10-year rule applying after majority status is achieved. Nevertheless, the deferral period shrinks the closer the child gets to the age of majority, making this course less productive as the child ages. It’s unclear if a pot trust for multiple children constitutes a conduit trust30 or whether separate conduit subtrusts qualify when the funding trust is named on the designation form.31 Often, the key is how much dominion and control the client feels comfortable in ceding to the child, with other facts and tax considerations determining whether a conduit or accumulation trust is employed.

Practitioners representing disabled or chronically ill children should pay careful attention that those children qualify under the AMBT special rules and facilitate the stuffing of these trusts with retirement plan assets (such as trustees disproportionately allocating retirement plan assets to AMBTs). Clients should contemplate Roth conversions and AMBTs funded with Roth IRAs to take advantage of non-taxable distributions32 paid over a life expectancy period. Outright beneficiary designations or conduit trusts are discouraged because of the loss of control over distributed funds.  

Practitioners should adopt multiple approaches and not embrace the same solution for all clients. A minor child could be named the beneficiary of a conduit trust paying college costs, with the remaining retirement benefits held by an accumulation trust. Or, an accumulation trust could benefit a spendthrift beneficiary with a conduit trust established for the more responsible sibling, and so on. A creative solution is to form an accumulation trust making distributions to low bracket DBs, with equalizing distributions from a side trust to higher taxed beneficiaries.33 Another adventuresome plan names one or more qualified subchapter S trusts to hold S corporation stock with the S corporation as retirement plan beneficiary to snag individual rates.34

Retirement plan assets aren’t great for funding trusts exempt from the generation-skipping transfer (GST) tax. To avoid wasting GST tax exemption on property that income taxes will devalue, it makes sense to allocate retirement plan assets (excluding Roth IRAs) to GST non-tax-exempt trusts. For GST non-tax-exempt accumulation trusts, this becomes tricky because of the planning objective (without using general POAs) of forcing estate tax inclusion to avoid GST tax. Making the non-tax-exempt trust a conduit trust for retirement plan assets circumvents this concern.35 A fractional “pick and choose” formula should define the retirement plan assets passing to the non-tax-exempt trust (don’t use a pecuniary formula to avoid triggering tax on funding).36 Conduit trusts for healthy adult children may constitute baseline planning, even though conduit distributions leak assets out of the protected GST tax exempt structure.

Can a trust be built to account for unforeseen possibilities? Consider auto switch provisions providing that if the beneficiary attains EDB status after the estate plan is signed, the trust magically converts to an AMBT or conduit trust to capture the life expectancy method. While this strategy makes sense for beneficiaries who become disabled/chronically ill, there’s debate on whether it works well for other EDBs.37 Conceivably, an auto switch could establish a conduit trust for a minor child if the participant dies before a certain age, funding the trust with a stipulated portion of retirement plan assets deemed sufficient to fulfill educational needs. Deliberate whether a trust protector or independent trustee may also have the power before the Sept. 30 beneficiary elimination date38 to cancel auto switches, eliminate beneficiaries, discard DB and see-through trust restrictions or make other changes.   

A tactic having a lot of promise involves granting a child the power to withdraw principal and income, thereby qualifying the trust as a grantor trust under IRC Section 678. The benefit is that all RMDs (even if retained in trust) are taxed at the child’s lower tax rates. Refinements to curtail abuse by the child are lapsing and hanging withdrawal rights, combined with the trust protector’s ability to suspend such rights.39 

Critics tend to characterize Section 678 trusts as too risky because they aren’t expressly sanctioned in the regulatory examples. Is this too narrow a vantage? In developing solutions to unanswered questions in the post-SECURE Act world, the gorilla in the room is the 50% penalty for late RMDs40 and whether IRS guidance (whenever it comes) will be prospective or retroactive. 

Endnotes

1. The Setting Every Community up for Retirement Enhancement (SECURE) Act, Section 401, Title IV of “Division O” of Pub. L. No. 116-94.

2. Internal Revenue Code Section 401(a)(9)(E)(ii).

3. Treasury Regulations Section 1.401(a)(9)-5, A-1.

4. Treas. Regs. Section 1.401(a)(9)-5, A-4, A-5.

5. For 2021, the ordinary income of a trust over $13,050 is taxed at 37%. Individual taxpayers must have substantially greater income ($523,601 for single taxpayers; $628,301 for married couples filing jointly) to reach the 37% bracket.

6. Treas. Regs. Section 1.401(a)(9)-6, A-15.

7. IRC Section 1(g). 

8. James Blase, Estate Planning for IRAs and 401Ks: A Handbook for Individuals, Advisors and Attorneys, at p. 80 (2020).

9. IRC Section 401(a)(9)(E)(ii)(III) and (IV).

10. Section 401(a)(9)(E)(ii)(V). This rule becomes problematic for chronic conditions such as autism that arguably may exist at death, but not be diagnosed until years later. Upon discovery, are taxpayers required to amend prior tax returns, go to court or otherwise tangle with the IRS to prove that the disability existed at death?

11. Treas. Regs. Section 1.401(a)(9)-6, A-15.

12. Section 401(a)(9)(E)(ii)(III), referencing IRC Section 72(m)(7).

13. Section 401(a)(9)(E)(ii)(IV), referencing Section 7702B(c)(2).

14. Section 401(a)(9)(H)(iv)(I).

15. Section 401(a)(9)(H)(iv)(II).

16. Natalie B. Choate, Planning for Retirement Benefits, Recent Developments: CARES: The Act + IRS Notices 2020-50 and -51, Even Newer Life Expectancy Tables for 2021 and SECURE, SECURE, SECURE! (Version 2020-3, Nov. 20, 2020), at p. 54. Seehttps://secureservercdn.net/198.71.233.107/741.9f4.myftpupload.com/wp-content/uploads/2020/11/NewDev2020-3.pdf.

17. See Steven E. Trytten, “The Zen of Drafting See-Through Trusts,” Trusts & Estates (June 2014) for a good analysis of “age restriction” and other drafting motifs under pre-SECURE Act law.

18. Treas. Regs. Section 1.401(a)(9)-5, A-7 (a) and (c). 

19. Treas. Regs. Section 1.401(a)(9)-4, A-5(b). 

20. Choate, supra note 16, at pp. 38-39.

21. Section 401(a)(9)(B)(iv) permitting the spouse to delay required minimum distributions (RMDs) to when the participant would have attained age 701/2. The spouse as sole designated beneficiary (DB) would also qualify for life expectancy recalculation.

22. Treas. Regs. Section 1.401(a)(9)-5, A-7(c)(3).

23. Natalie B. Choate, Life and Death Planning for Retirement Benefits (8th ed. 2019), at para. 6.3.08.

24. The American College of Trust and Estate Counsel (ACTEC), “Request for Guidance from Treasury Regarding Section 401 of the SECURE Act, Part 2” (July 29, 2020). Seewww.actec.org/assets/1/6/2020-07-29_ACTEC_Request_for_Guidance_from_Treasury_Regarding_Section_401_of_the_SECURE_Act,__Part_2.pdf.

25. Treas. Regs. Section 1.401(a)(9)-8, A-2, A-3.

26. Treas. Regs. Section 1.401(a)(9)-4, A-5(c). Choate, supra note 23, at para. 6.3.02. 

27. Choate, supra note 23, at para. 6.3.01. Private Letter Rulings 200537044 (Sept. 16, 2005) and 200607031 (Feb. 17, 2006).

28. ACTEC, supra note 24, at p. 31.

29. Other commentators have joined the call to eliminate the “no separate accounts under a single trust rule.” See Natalie B. Choate, “IRS Guidance Needed for SECURE’s Changes to the Minimum Distribution Rules,” LISI Employee Benefits and Retirement Planning Newsletter #754, at p. 20 (Feb. 15, 2021), www.leimbergservices.com.

30. ACTEC, supra note 24, at pp. 44-45.

31. Alan S. Gassman, Christopher J. Denicolo, Brandon L. Ketron and John N. Beck, Planning for Ownership and Inheritance of Pension and IRA Accounts and Benefits in Trust or Otherwise, at p. 71 (2020).

32. IRC Section 408A(d)(1).

33. Alan S. Gassman and John N. Beck, “Is the TEA POT Trust Right for Your Clients?” Trusts & Estates (April 2020).

34. Jonathan G. Blattmachr, Ladson Boyle, Mitchell M. Gans and Diana S. C. Zeydel, “The SECURE Act, Trusts, Corporations and CRTs,” Estate Planning Journal (July 2020), at pp. 24-32. 

35. Trytten, supra note 17, at pp. 54-55. A general power of appointment (GPA) in a generation-skipping transfer (GST) non-tax-exempt trust avoids GST by causing estate tax inclusion, but may add non-individuals such as creditors or the estate as countable beneficiaries. This flunks DB status, with default to the 5-year or ghost life rules. For conduit trusts, the GPA is irrelevant for RMD purposes.

36. Gassman, Denicolo, Ketron and Beck, supra note 31, at p. 89.

37. Steven B. Gorin, “Drafting under the SECURE Act,” at pp. 4-11, available as part of the LISI Webinar, “Drafting under the SECURE Act—Expanded and Updated with Scenarios & Sample Language” (Sept. 18, 2020), www.leimbergservices.com.

38. Treas. Regs. Section 1.401(a)(9)-4, A-4.

39. Blase, supra note 8, at pp. 35-54; Blattmachr, Boyle, Gans and Zeydel, supra note 34, at pp. 27-28; Edwin P. Morrow, “Using BDOTs for Optimal Asset Protection and Income Tax Minimization After Passage of the Secure Act,” LISI Income Tax Planning Newsletter #192 (Feb. 18, 2020), www.leimbergservices.com.

40. IRC Section 4974.


Viewing all articles
Browse latest Browse all 733

Trending Articles