The minimum distribution rules and their critical role in controlling the floodgates of...
I. INTRODUCTION
When complete in 2009, the Three Gorges Dam at the Yangtze River will retain trillions of gallons of water.l The scope of this technological feat cannot be overstated. The dam will tower over 600 feet high; it will flood an area almost equal in size to half of the state of Rhode
In the United States, there are no dams the size of the Three Gorges Dam. Yet, there is a vast pool, not of water, but of trillions of dollars of wealth that is contained in tax-free qualified plans (including individual retirement accounts).3 Like the Chinese officials who will oversee water flow at the Three Gorges Dam, Congress has the ability to determine how and when the floodgates of this qualified plan wealth will be released. Commonly referred to as the minimum distribution rules, these "floodgate switches" are embodied in Internal Revenue Code (the "Code") 401(a)(9).4 Upon certain triggering events, these rules require plan participants or their designated beneficiaries to take distributions from their qualified plans.5
To date, Congress appears satisfied that the minimum distribution rules adequately regulate the flow of qualified plan wealth, as evidenced by its lack of proposals to change these rules. However, this sense of satisfaction is misplaced. The minimum distribution rules are an administrative nightmare, and they fail to achieve their intended goal of forcing plan participants to use their retirement assets during their and their spouses' lifetimes. Put differently, the floodgate switches are precariously stuck in the "off" position. While this state of affairs may foster swelling retirement account balances, it restricts the flow of federal tax dollars rippling into the federal coffer to a mere trickle.
The analysis that follows is divided into four sections. Section II sets forth the factors that make an evaluation of the minimum distribution rules timely. Section III provides an overview of the minimum distribution rules and how they regulate the flow of qualified plan wealth. Section IV evaluates the administrability and equity of the minimum distribution rules and whether they should be retained in their current form. Finally, Section V offers various proposals that would make the minimum distribution rules more effective.
11. TIMELINESS OF EVALUATION
The establishment of qualified plans by employers is a fairly recent phenomenon in the United States.6 Various tax incentives, however, have caused their growth to flourish, making them a nearly ubiquitous feature in today's labor market.7 These incentives include the following privileges: employers receive an immediate income tax deduction for contributions made on the behalf of employees to qualified plans,8 employees are not required to declare such contributions as income until they are distributed,9 and earnings on assets held in qualified plans may accumulate free from income tax.10 Along with Social Security and private savings, qualified plan wealth is supposed to enable taxpayers to enjoy the same standard of living preand postemployment.11
Promoting taxpayers' use of qualified plans through various tax incentives comes at a price: taxpayers are able to shelter income that would otherwise be taxed. By sanctioning the deferral of tax on trillions of dollars, Congress has lost (and continues to lose) billions of dollars annually in potential revenue.12 In order to curtail this revenue loss and to ensure that qualified plan wealth is used on behalf of plan participants (and their spouses), Congress instituted the minimum distribution rules.13 At first, these rules applied only to a small number of qualified plans, namely those established by the self employed.14 Eventually, however, Congress expanded their coverage to apply to virtually all qualified plans.15 Upon the passage of time or the occurrence of certain events, these rules require that plan participants remove wealth from the tax-free sanctuary of their qualified plans.16
Until recently, the minimum distribution rules played a marginal role in dictating the outflow of wealth from qualified plans. This was largely due to the fact that most employers traditionally established defined benefit plans.17 Under the terms of these plans, a participant is promised a specific benefit upon retirement, which is determined under a formula contained in the plan.18 Once determined, the benefit is ordinarily distributed in the form of a single life annuity (or joint annuity, in the case of married participants).19 This type of distribution obviates the need for minimum distribution rules since such payments are automatically spread evenly over the life of the participant and the participant's spouse and cease upon their death. Defined benefit plan distributions are rarely made to persons other than plan participants and their spouses.20
But the minimum distribution rules have now begun to play a pivotal role in dictating the flow (or trickle) of wealth out of qualified plans. This is because over the last two decades the retirement planning environment has undergone dramatic changes. The nature of these changes and their implications are discussed in the subsections that follow.
A. Popularity of Defined Contribution Plans
Over the past two decades, defined contribution plans (including IRAs) have evolved to be the qualified plan of choice, eclipsing the once dominant role of defined benefit plans.21 In general, a defined contribution plan is a qualified plan in which the plan participant is not guaranteed a predetermined benefit upon retirement but instead has an individual account that houses his or her retirement wealth.22 Several factors have led to the preference for defined contribution plans over defined benefit plans.
First, over the past three decades, major legislative changes have made defined benefit plans more difficult and expensive to maintain. With the intention of overhauling the private retirement planning system, Congress passed the Employee Retirement Income Security Act ("ERISA") in 1974.23 Among other things, this legislation imposed strict minimum funding requirements on defined benefit plans.24 To comply with the minimum funding requirements, employers with defined benefit plans must retain the services of an actuary who must annually monitor current and projected plan assets and liabilities. ERISA also established the Pension Benefit Guarantee Corporation ( "PBGC" ) to monitor plan solvency and insure the payment of promised benefits.25 Employers who sponsor defined benefit plans must pay insurance premiums to the PBGC26 so that the PBGC can meet the financial obligations of retirement plans that become insolvent.27 Furthermore, ERISA makes plan sponsors liable for any shortfall in defined benefit plan assets when plans become insolvent.28 Thus, the employer using defined benefit plans bears a downside investment risk. In contrast, under a defined contribution plan, the employee bears all of the investment risk. In addition to ERISA, other legislative measures have placed additional administrative burdens on the maintenance of defined benefit plans.29 As a consequence, defined benefit plans have less potential appeal to employers.30
Second, in addition to being less costly, less risky, and less administratively burdensome to employers, defined contribution plans offer plan participants more flexibility and independence.31 More specifically, defined contribution plans often enable employees to make investment choices geared towards their own risk tolerance and individual idiosyncrasies. Given the mobility of today's labor force, employees also have a greater sense of economic security in having control over their retirement destiny rather than having it rest with their long-forgotten former employers.
Third, traditional defined benefit plans often compute pensions on a backloaded basis; that is, benefits are based on a specified percentage of an employee's final salary. In contrast, defined contribution plan benefits are determined on a frontloaded basis-the earlier a contribution is made to a plan, the longer will be the time period of tax-free accumulation. Younger, highly mobile employees who do not envision working thirty or forty years for the same employer particularly value this frontloading feature.
Fourth, various legislative initiatives have spurred the use of defined contribution plans in the labor market. These initiatives include permitting the establishment of qualified plans that do not require (a) employer contributions (e.g., 401(k) plans)32 or (b) employer involvement (e.g., IRAs).33 By the institution of various rollover options, Congress has also facilitated the transfer of retirement wealth between different qualified plans.34
The dominance of defined contribution plans has a number of important public policy implications.35 Most relevant to this analysis is the inherent incentive to retain, rather than distribute, retirement plan wealth. In a defined benefit plan setting, distributions are typically limited to a single life annuity or, in the case of a married plan participant, a joint life annuity. That being the case, plan participants are often anxious to receive as much as possible as quickly as possible from their defined benefit plans because their benefits cease at death. The same philosophy, however, does not permeate the defined contribution plan setting. This is because plan participants often have several distribution options, and they have no financial incentive to deplete their retirement account balances. To the contrary, their retirement account balances continue to accumulate income tax free, which inures to their benefit rather than to the benefit of their employer or other plan participants, as is the case in defined benefit plans. Moreover, upon the death of a plan participant, the retirement account balance is not extinguished, but, rather, it can be directed to the plan participant's beneficiaries.
B. Swelling Retirement Account Balances
In the early 1980s, the number of defined contribution accounts surged. This, along with other factors (e.g., a blisteringly hot stock market and a lack of effective statutory coordination between the maximum defined benefit limitation and the maximum contribution limits for the various types of defined contribution plans), caused plan participants' account balances and vested benefits to swell.36 Their size and cost to the Treasury finally attracted congressional concern.37
In 1986, Congress instituted two different taxes-an excise tax on "excess distributions"3g and an additional estate tax on "excess accumulations"39-in order to discourage the excessive accumulation of tax-favored retirement wealth.4 Each tax amounted to 15%.41 The introduction of these taxes had a significant chilling effect on plan participants' funding efforts. Practitioners' journals warned tax advisers to caution plan participants not to be too ambitious in their sheltering attempts lest they risk exposure to these taxes.42 When it came to funding, moderation was the word; excesses were to be punished.
This state of affairs lasted approximately a decade. The Small Business Job Protection Act of 1996 suspended application of the excise tax on excess distributions from 1997 through 1999.43 Labeled "success taxes" by their critics, both the excise tax and the additional estate tax were repealed the following year by the Taxpayer Relief Act of 1997.44
The Small Business Job Protection Act of 1996 also repealed the so-called " 415(e) limitation."45 This limitation applied to plan participants who participated in a defined benefit plan and a defined contribution plan maintained by the same employer 46 Under this limitation, such plan participants could not simultaneously take full advantage of both the maximum permitted defined benefit under the defined benefit plan and the maximum permitted contribution to the defined contribution plan.47
The repeal of the excise tax, the additional estate tax, and the 415(e) limitation has had an important effect in transforming the qualified plan landscape. Plan participants now have a green light to strive for as much asset growth as possible, even if the amount of wealth they contribute and the asset growth they achieve on such contributions far exceeds their (and their spouse's) retirement needs. Plan participants now realize that even if they do not reap the harvest of their qualified plan wealth, their designated beneficiaries will, no matter how excessive the wealth.48
C. Introduction to Roth IRAs
The Taxpayer Relief Act of 1997 introduced a new qualified plan vehicle entitled the Roth IRA.'9 Roth IRAs are similar to traditional IRAs in that contribution limits are comparable and assets held in both kinds of accounts grow income tax free." However, Roth IRAs are fundamentally different from traditional IRAs. First, only nondeductible contributions can be made to Roth IRAs." Second, any "qualified distribution"52 from a Roth IRA is not includible in gross income. Third, the minimum distribution rules do not apply until the death of the account holder.53
These differences have made Roth IRAs a new favorite among tax planners and their clients.54 There are many reasons for their popularity, but central among these is the fact that plan participants may keep their wealth housed in Roth IRAs longer than in any other retirement planning vehicle. This is because the minimum distribution rules are held in abeyance until the death of the plan participant. 55
There is a two thousand dollar annual limitation relating to contributions made to Ruth IRAs.sb This limitation, however, cloaks the tax-deferral opportunity Roth IRAs offer. If a plan participant's adjusted gross income falls below certain limits, that participant has the opportunity to roll over amounts from a traditional IRA to a Roth IRA.57 Although a rollover to a Roth IRA from an employerprovided qualified plan is not allowed, there is nothing to prevent a participant from first rolling over the qualified plan distribution into a traditional IRA and then making a rollover to a Roth IRA. Of course, the rollover from the traditional IRA to the Roth IRA is fully taxable, but if the resulting tax can be paid from other savings, the overall effect of the rollover is to permit even greater tax subsidies to flow to the plan participant. From a practical perspective, what this rollover ability means is that plan participants may contribute tremendous amounts of wealth into Roth IRAs and stay the application of the minimum distribution rules, at least until their death.
D. Miscellaneous Factors
Aside from the changes mentioned in the prior three subsections, there are two other factors that make an evaluation of the minimum distribution rules particularly timely. First, over the past decade, marginal federal income tax rates have gradually risen from a low of 28% in 1986 to where they are today at 39.6%. Higher income tax rates function as qualified plan fertilizer, creating added incentive for plan participants (particularly those whose incomes are taxed in the higher tax brackets) to shelter their otherwise taxable income by making qualified plan contributions.sg Moreover, the imposition of higher income tax rates entices plan participants to hold their accumulated wealth in the tax-free sanctuary of qualified plans for as long as possible.
A final reason to examine the minimum distribution rules now is the coming crisis in funding Social Security and Medicare. Many studies indicate that within the next two or three decades Social Security and Medicare will fall short of being able to sustain themselves financially.59 The question on the minds of commentators and politicians alike is how to maintain their solvency. Given that the largest source of private wealth in the world is currently held by qualified plans, some commentators have raised the idea that plan assets and/or earnings on these assets should be taxed.bo
E. Summary
The recent dominance of defined contribution plans; the repeal of the excise tax, additional estate tax, and limitations that once discouraged bountiful account balances; and the introduction of Roth IRAs have significantly altered the retirement planning landscape. The convergence of these events, along with relatively high income tax rates during a time when Social Security and Medicare are struggling to maintain their solvency, has catapulted the minimum distribution rules to a new level of importance. As the baby boomer generation nears retirement, the tension between the natural desire of taxpayers to maximize tax deferral via qualified plans and the government's need for tax revenue can only increase.
III. MECHANICS OF THE MINimum DISTRIBUTION RULES
This Section, which details the current system for regulating minimum distributions from qualified plans, is apt to prove challenging and frustrating. This reflects the fact that the minimum distribution rules themselves are inherently complex and difficult to comprehend. They are replete with special definitions, general rules, exceptions, and exceptions to the exceptions." But before delving into the rules, an example of their operation illustrates the extraordinary benefit that tax-free compounding offers plan participants under the existing minimum distribution rules.
A. An Example of the Minimum Distribution Rules
This subsection illustrates the distribution effect under the current minimum distribution rules, using data set forth in Table I. Assume that Owner, age seventy, has $1,000,000 in an IRA. Assume further that Owner has a Spouse, age sixty-six, and a Child, age forty. Owner and Spouse both live until age eighty-five and both elect not to recalculate life expectancies. When Owner dies, Spouse rolls over the account into Spouse's own IRA, elects the term certain method, and names Child as the beneficiary. After Spouse's death, Child is permitted to continue to receive distributions for twentyfour more years. (The subsections that follow explain the nonrecalculation or "term certain" option as well as the reasons why Spouse and Child can continue to receive benefits under the IRA. )
The example assumes a pre-tax investment return of 8% on the assets held by the IRA. This percentage times the "Beginning IRA Value" less the "Minimum Distribution" results in the "Ending IRA Value." The example also assumes that the distributions out of the IRA are taxable at a combined federal and state income tax rate of 43%. Finally, once out of the IRA, the example assumes that these proceeds grow at an after-tax investment return of 6.4% (reflecting the fact that some returns will be taxable at lower long-term capital gains rates). These assumptions produce the final column labeled "Accumulated Distributions."
When Owner dies, the minimum distribution drops from $153,417 to $68,079, and when Spouse dies at age eighty-five, the minimum distribution drops from $82,388 to $47,730. The reason for these drops will be explained shortly. If Child dies before age eighty-three, Child's heirs can continue the same distribution scheme; there is no acceleration of distributions due to Child's death.
What is striking about Table I is the last entry of the "Accumulated Distributions" column. It should come as no surprise that the value of an investment can greatly multiply over a forty-four year period given the effects of compounding at a steady rate of return. But compare this figure with what would have been accumulated had the $1,000,000 held in the IRA been initially distributed to Owner at the beginning of Year 1, i.e., had there been no further tax subsidy. Under the same rate-of return and tax-rate assumptions, the $1,000,000 would have been immediately subject to a tax of $430,000, leaving a balance of $570,000. By the end of the fortyfour year period, this balance, rather than growing to $11,154,724 as provided in Table I, would have instead grown to only $8,735,877. The difference between $11,154,724 and $8,735,877, $2,418,847, represents the cost to the government of permitting tax deferral during Owner's retirement as well as over the time period during which Spouse and Child enjoyed Owner's inheritance. Bluntly put, the roughly $2.4 million difference represents a rather hefty government grant to the Owner and his family.
IMAGE TABLE 39Table I
IMAGE TABLE 41B. Lexicon of Important Terms Under the Minimum Distribution Rules
This subsection introduces a set of technical terms central to the comprehension of the minimum distribution rules.
1. Required beginning date
Integral to mastery of the minimum distribution rules is the concept known as the "required beginning date" ("RBD"). Distributions from a qualified plan (other than a Roth IRA) must commence on or prior to the RBD.62 Failure to make the requisite distribution by a participant's RBD results in the imposition of a 50% excise tax, as described infra in subsection III. D.63
The RBD for plan participants who are still actively employed by their employer differs from those who are not. The RBD for any employee who is a participant in an employer's qualified plan (and who is not a 5% owner of the employer at any time) is April 1 following the year of retirement, if such employee retires after age 701/2.65 For all other plan participants, the RBD is April 1 of the year following the year they attain age 70-1/2. 66
2. Minimum distribution amount
Once plan participants determine their RBD, they must next determine the amount the plan must distribute to them in order to avoid the imposition of the 50% excise tax. The minimum distribution amount must be paid each distribution calendar year.67
In the case of a defined benefit plan, distributions are generally made in the form of annuity payments. Starting at the participant's RBD, these annuity payments must be paid periodically at intervals not longer than one year using ( 1 ) the life expectancy of the plan participant or joint life and last survivor expectancy of the plan participant and a designated beneficiary, if any, or (2) a period certain not longer than a life expectancy or joint life and last survivor expectancy of the plan participant and a designated beneficiary, if any.bg To help ensure that the majority of the qualified plan assets inure to the benefit of the plan participant, annuity payments must conform to the minimum distribution incidental benefit requirement.69 This requirement limits the period certain over which annuity payments can be made and the payment percentage that can inure to a designated beneficiary.70 Typically, upon the death of the plan participant (or the plan participant's designated beneficiary), the plan's annuity payment obligations cease.71
In the case of a defined contribution plan (including IRAs), the process of determining the minimum distribution amount involves a simple computation involving a numerator and a divisor. Subject to certain adjustments,72 the numerator is the account balance on the last valuation date in the calendar year before a distribution calendar year.73 The divisor is generally the applicable life expectancy, which is the life expectancy of the plan participant (or the joint life and last survivor expectancy of the plan participant and the plan participant's designated beneficiary):74
Participant's Account Balance Applicable Life Expectancy
The applicable life expectancy is determined using the participant's (and the designated beneficiary's) age on the participant's birthday (and the designated beneficiary's birthday) in the calendar year prior to the participant's RBD.75 Life expectancies are set forth in Tables V and VI of Treas. Reg. 1.72-9.76
Example: Owner turns age seventy on June 17, 1999, and he is married to Spouse, who is four years younger. Owner has an IRA with an account balance of $1,000,000 on December 31, 1998. Owner may determine his minimum distribution amount for 1999 (his first distribution calendar year) using his own life expectancy of sixteen years. This would produce a minimum distribution amount equal to $62,500 ($1,000,000/16). Alternatively, were Spouse named as Owner's designated beneficiary, Owner could use their joint life expectancy of 22.5 years. This would produce a minimum distribution amount equal to $44,444 ($1,000,000/22.5) for the 1999 distribution calendar year. This is precisely how the minimum distribution amount was computed in Table I.
Note that the larger the divisor, the smaller the minimum distribution amount. The smaller the minimum distribution amount, the larger the retirement account balance that remains and the greater the income tax deferral. If the designated beneficiary is more than ten years younger than the plan participant and is not the plan participant's spouse, a smaller number known as the "applicable divisor" is used as the divisor instead of the applicable life expectancy." This is an application of the minimum distribution incidental benefit requirement, which is discussed infra subsection III.B.4.
Unless the terms of a plan provide otherwise or a plan participant elects to the contrary, a plan participant's life expectancy (or the joint life and last survivor expectancy of the plan participant and spouse) is recalculated annually.'$ The Treasury regulation tables supply the applicable life expectancy or joint life expectancies, as the case may be, based on the participant's and spouse's (if applicable) attained ages as of their birthdays in that distribution calendar year. 79 Alternatively, a plan participant may elect not to have his (or his spouse's) life expectancy recalculated.gg Instead, the life expectancy would be decreased by one each year to determine the minimum distribution amount. This is known as the "term certain method."81 In the case of a designated beneficiary other than the surviving spouse, the term certain method must be used to determine the divisor 82
Example: Using the example in Table I, the joint life expectancy of Owner and Spouse in the first distribution year (Year 1) is 22.5. Since in this example the participant has elected not to recalculate life expectancies, each year the divisor will be reduced by one. Thus, in Year 2 it is 21.5, and the minimum distribution is $48,165 ($1,035,556/21.5). By Year 16, the year of Owner's death, the divisor is 7.5. What happens after Owner's death is discussed below.
3. Designated beneficiary
In general, having a designated beneficiary allows the plan participant to use the joint life and last survivor tables to compute the minimum distribution amount, resulting in smaller minimum distribution amounts and maximizing income deferral. For purposes of the minimum distribution rules, only individuals (and individual beneficiaries of certain kinds of trusts)" may be designated beneficiaries." The terms of a plan may name the designated beneficiary, or, if the terms of the plan provide, plan participants may name an individual or trust as a beneficiary.85 For purposes of computing the minimum distribution amount, the participant's designated beneficiary will ordinarily be determined at the earlier of either the participant's death or as of the participant's RBD.86 If there is more than one designated beneficiary, the one with the shortest life expectancy will be used to determine the required minimum distribution amount.87
4. Minimum distribution incidental benefit (`MDIB") requirement The primary purpose of a qualified plan must be to provide retirement benefits or deferred compensation." Any other benefit must be incidental.89
Basically, under the MDIB requirement for computing the applicable life expectancy, any nonspouse who is a designated beneficiary is treated as being no more than ten years younger than the participant.90 In the case of defined benefit plans, the MDIB functions in a similar fashion by requiring that distributions made to plan participants not be too small relative to those that are to be made to their designated beneficiaries.91
There are two important limitations on the application of the MDIB requirement. First, if the participant's spouse is the designated beneficiary, the MDIB requirement does not apply.92 Second, after the death of the participant, the MDIB requirement no longer applies, and the applicable life expectancy that would have been used absent the MDIB requirement is then used to compute the minimum distribution amount.93 As the following example illustrates, the latter restriction has significant income tax deferral repercussions.
Example: Recall from the example in Table I that when Owner dies in Year 16 and Spouse rolls over the IRA into his/her own, Spouse names Child as the designated beneficiary. Without the MDIB requirement, the minimum distribution divisor for Year 17 could be based on Spouse's and Child's joint life expectancy as of that year, which under the Treasury's tables is twenty-eight. But since Child is more than ten years younger than Spouse, the MDIB requirement limits the divisor to sixteen, the "applicable divisor" in the Treasury's table.94 Thus, the minimum distribution for Year 17 is $68,079 ($1,089,259/16). When Spouse dies in Year 20, distributions in future years are no longer subject to the MDIB requirement. In Year 21, the remainder of the applicable life expectancy is twenty-eight (the Child's and Spouse's original joint and survivor life expectancy when minimum distributions commenced to Spouse) less four (the number of years that elapsed from the Spouse's first distribution year to the date of Spouse's death), and the minimum distribution is $47,730 ($1,145,525/24). The divisor is reduced by one in each of the next twenty-three years. 95
Understanding these definitions will clarify the discussion that follows of how the minimum distribution rules operate.
C Application of the Minimum Distribution Rules
Distributions made during the life of the plan participant may be made before or after the participant's RBD. Distributions made after the death of the plan participant will occur either before or after the participant's RBD.
1. Distributions during the lif of the plan participant
a. Before RBD. In general, the minimum distribution rules do not apply with respect to distributions made prior to the participant's RBD.96 Plan participants may, therefore, prolong the time period before which distributions have to be made from their qualified plans. For example, in the case of the active octogenarian who participates in his employer's qualified plan, distributions from such a plan do not have to commence until the employee retires, say at age ninety. Plan participants are thus free to capitalize upon the tax-free status of their qualified plans.
b. After RBD. In order not to jeopardize a qualified plan's taxexempt status, or to prevent the plan participant from incurring a 50% excise tax, distributions of the minimum distribution amount from a qualified plan must commence at the participant's RBD.` Subject to the MDIB requirement, distributions may be made (1) over the life of such participant, (2) over a term certain that does not exceed the actuarial life expectancy of the plan participant, (3) over the lives of the participant and a designated beneficiary, or (4) over a term certain that does not exceed the actuarial life expectancy of the plan participant and a designated beneficiary."
2. Distributions after the death of the plan participant
a. Death of plan participant before the RBD. If the plan participant dies prior to the RBD, distributions must be made under one of two methods. The first method is the five-year rule under 401(a)(9)(B)(ii). This method requires that the entire interest of the participant be distributed within five years of the participant's death (i.e., on or before December 31 of the fifth calendar year after the participant's death)99 regardless of to whom or to what entity the distribution is made.100
The second method is the exception to the five-year rule under 401(a)(9)(B)(iii). This method, which presumes that the plan participant has named a designated beneficiary, requires that any portion of a participant's interest that is payable to (or for the benefit of) a designated beneficiary be distributed, commencing within one year of the plan participant's death (i.e., on or before December 31 of the first calendar year after a plan participant's death)1i and extending over the life of such beneficiary (or over a period certain not extending beyond the life expectancy of such beneficiary),'oz
If the participant's spouse is the designated beneficiary, the surviving spouse has two options. Under the first option, distributions may commence on or before the later of ( 1 ) December 31 of the calendar year immediately following the calendar year in which the plan participant died or (2) December 31 of the calendar year in which the plan participant would have attained age 701/a.loa Under the second option, the surviving spouse may elect to roll over the plan participant's account to an IRA (assuming this is a permissible distribution option),114 thus deferring distributions until April 1 of the calendar year following the year in which the surviving spouse attains age 7-1/2. In addition, the surviving spouse has the ability to name new designated beneficiaries.los
b. Death of the plan participant after RBD. In general, if a plan participant dies after the participant's RBD, an "at least as rapidly" rule applies. That is, the account balance of the plan participant must be distributed "at least as rapidly" as the method being used for distribution as of the date of the participant's death.lo6 How this rule applies and its implications depend upon whether the participant has named a designated beneficiary.
If the participant fails to name a designated beneficiary (e.g., the plan participant names his estate as the beneficiary) and if the participant is recalculating his life expectancy, the entire account balance must be distributed by December 31 of the year following the participant's death.l' Alternatively, if there is no designated beneficiary and the participant had elected the term certain method, distributions may be made over the remaining designated period.loa
If the participant dies after the participant's RBD and the participant has named a designated beneficiary, the remaining portion of such interest must be distributed to the designated beneficiary at least as rapidly as under the distribution method being used as of the date of the participant's death.l9 Because the MDIB requirement, however, does not apply after the participant's death, the applicable divisor no longer applies. Instead, the divisor used to determine the minimum distribution amount is either ( 1 ) the joint life expectancy of the plan participant and the designated beneficiary at the commencement of distributions less the number of years that have elapsed since the RBD (where the plan participant had elected the term certain method) or (2) the designated beneficiary's life expectancy at the commencement of distributions less the number of years that have elapsed since the RBD (where the plan participant had used the recalculation method).'lo
If the participant names the participant's surviving spouse as the designated beneficiary, the surviving spouse has two options. Under the first option, the surviving spouse can continue to receive minimum distributions under the schedule in place at the participant's death.111 Under the second option, the surviving spouse may roll over the participant's account balance into an IRA (assuming this is a permissible distribution option).1'2 In addition, the surviving spouse has the ability to name new designated beneficiaries. (The second option was used by Spouse in the Table I example, which resulted in the deferral of distributions far beyond the joint lives of Owner and Spouse. )
D. Failure to Comply with the Minimum Distribution Rules and the Imposition of the 50% Excise Tax
To encourage compliance with the minimum distribution rules, Congress has instituted a 50% excise tax on the difference between the amount of the minimum distribution and the amount, if any, that was actually distributed to the plan participant.113 However, the tax can be waived if a plan participant can show that the participant's failure to receive the entire minimum distribution is due to reasonable error and reasonable steps are instituted to remedy the shortfall.14
IV. EVALUATION OF THE MINIMUM DISTRIBUTION RULES
An analysis of the minimum distribution rules reveals their flaws. The rules are administratively daunting, making compliance difficult and costly, especially given the large number of taxpayers affected. At the same time, the rules are inequitable because benefits inure primarily to high-income earners.
A. Administration of the Minimum Distribution Rules
1. Complexity
By any standard, the minimum distribution rules are horrendously complex. First, the rules require mastery of special terms of art that have the aura of mystical incantations. People who speak in minimum-distribution-rule parlance have a language of their own. Their conversations are punctuated by cryptic phrases, where legions of acronyms abound.ls Outsiders to the minimum-distribution-rule expert clique are rendered linguistically impotent, unable to communicate and reliant on the expertise of others.
Second, the minimum distribution rules involve numbing detail. Unlike many Code provisions that provide a general rule with one or two exceptions, the minimum distribution rules provide a series of general rules that are riddled with numerous exceptions and exceptions to those exceptions. Indeed, the regulations that elaborate on the technical meaning of the statute span over forty single-spaced pages.lb The outcome of this complexity is that plan participants, tax planners, and brokerage houses all too often commit inadvertent planning errors.ll7
Finally, the minimum distribution rules spawn additional complexity in other areas of tax planning. Consider, for example, the plight of married taxpayers who participate in qualified plans. A married individual may wish to establish trusts that qualify for the estate tax marital deduction for the benefit of the surviving spouse. In order to qualify for the estate tax marital deduction, the terms of such trusts must provide that the surviving spouse annually receives all trust income.118 Coordinating the plan participant's dual objectives of satisfying the estate tax marital deduction and minimum distribution rules is not an easy task. It requires an extraordinary amount of careful (and expensive) tax planning involving the use of special trust provisions and the issuance of special instructions to plan administrators.119 Plan participants who already had a hard time comprehending the dynamics of their estate plans undoubtedly find that the minimum distribution rules add another unwelcome and cumbersome dimension of complexity.lzo
2. Enforceability by the Internal Revenue Service
The complex nature of the minimum distribution rules along with the absence of any congressionally mandated oversight requirements makes enforcement by the Service difficult.121 For example, plan participants make supposedly irrevocable elections in determining their minimum distribution amounts, such as whether to recalculate life expectancies and who is named as the designated beneficiary. Despite the importance of these elections, the Service receives no independent verification of what they are in the form of information returns. This is in stark contrast to other areas of the law where the issuance of information returns plays a vital role in ensuring taxpayer compliance.122 This absence of direct oversight means that plan participants who supposedly make irrevocable, yet illconceived, elections may make changes with virtual impunity.
3. Tactics targeted to maximize tax deferral
The significant economic benefits of tax deferral motivate plan participants and their advisors to exploit the weaknesses of the minimum distribution rules."' Often, however, plan participants fail to comprehend the full consequences associated with the adoption of such strategies, fostering plan participant confusion and frustration. Such strategies for prolonged tax deferral incur costs not only to society in the form of lost revenue but also to plan participants in the form of hefty legal and actuarial fees. Adopting these strategies also contributes to the difficulties of making the minimum distribution rules easier to administer. The following three subsections illustrate some of the tactics used to maximize tax deferral.
a. Naming the plan participant's children as the designated beneficiaries and the plan participant's spouse as the contingent beneficiary. This allows required minimum distributions to be based on the joint life expectancy of the plan participant and a child (subject, of course, to the MDIB requirement) rather than the shorter joint life expectancy of the participant and the spouse. If the participant predeceases the spouse, the children would be expected to disclaim their interest in the plan, allowing the surviving spouse to roll it over into an IRA and name new designated beneficiaries, presumably the very same children.
b. Continuing employment. To gain an extra decade or two of income tax deferral, plan participants may continue their employment, perhaps through a part-time employment arrangement and thereby postpone their RBD.
c. Rolling over account balances. In certain instances, plan participants who have made poor beneficiary designation choices (e.g., naming their estate as the designated beneficiary) can rectify these errors-albeit at an immediate tax cost-by rolling over their account balances into newly formed Roth IRAs that permit them to name new designated beneficiaries.
B. Equity of the Minimum Distribution Rules
Aside from administrative difficulties, the minimum distribution rules exacerbate a weakness that is already part of our system of taxfavored retirement saving: the tax benefits are skewed toward highincome individuals. The tax-favored treatment of qualified plans costs the Treasury billions of dollars annually and constitutes the single largest tax expenditure of the federal budget."' Despite these costs, study after study indicates that those who capitalize most on the advantages of tax deferral are the highly compensated."25 The minimum distribution rules perpetuate this inequity by allowing the account balances of the highly compensated to grow virtually unabated over a significant time period, far beyond the lifetime of the plan participant.
During retirement, wealthy plan participants are more likely to be able to draw upon financial resources other than their qualified plans, thus prolonging plan distributions and resultant income taxes for many years. Their designated beneficiaries (children, grandchildren, or great-grandchildren) are likewise apt to have alternative financial resources other than qualified plan assets and could afford to prolong plan distributions as long as possible. In contrast, plan participants who are less financially fortunate often tap their qualified plans sooner and commonly exhaust plan assets well before their death.126
Consider the example in Table I, supra Section III.A. By the time Child inherits the IRA, there is already $1.7 million accumulated outside the IRA, making it more likely that Child can afford to take out only the minimum required distribution each year. Now suppose that Spouse's designated beneficiary had not been Child, but instead Grandchild, age two in Year 19. The joint and last survivor life expectancy of Spouse and Grandchild is 76.6 years, which allows distributions to be spread out over nearly ninety-three years from Owner's RBD.127
There seems little justification for a system that, on one hand, allows the highly compensated to amass significant tax-favored wealth on the theory that it was needed for retirement but, on the other hand, permits them to perpetuate their own financial dynasties as this wealth moves across multiple generations, retaining its tax-favored status.laa
V. PROPOSALS To RENDER THE MINimum DISTRIBUTION RULES MORE EFFECTIVE
This analysis sets forth two alternative proposals, each of which is based upon the following simple proposition: the benefits of tax deferral should inure solely to plan participants (and their spouses). The notion that such benefits should inure primarily to participants and spouses is already well embedded in the MDIB requirement and minimum distribution rule legislative history.129 But unless one believes that all saving should be tax-favored (which would essentially transform our income tax system into a consumption tax system), there seems little justification for extending the tax benefits to other beneficiaries.
These proposals, therefore, aim at protecting plan participants' ( and their spouses') standard of living during their retirement years while simultaneously curtailing the ability of their beneficiaries to gain unwarranted tax benefits. Both proposals also aim to make the minimum distribution rules simpler and more equitable.lao
A. Proposal One: Modify the Existing Minimum Distribution Rules
1. A joint life expectancy should be used only if the designated beneficiary is a spouse
Under this proposal, required minimum distributions would generally be based only on the life expectancy of the plan participant. The only exception to this rule would be if the plan participant's spouse were the designated beneficiary, in which event minimum distributions would be based on the joint and survivor life expectancy of the plan participant and the spouse."'
2. In the case of defined contribution plans and IRAs, life expectancies must be recalculated
When determining their minimum distribution amounts, plan participants (and their spouses) would have to use the recalculation method rather than the term certain method. This means that the minimum distribution for a year is determined simply by dividing the account balance by the current life expectancy of the plan participant (or the participant and a spouse). The relevant life expectancies are already set forth in Treasury Regulation tables.132 The only information needed to use the table is the attained age of the plan participant (and the spouse, if the spouse is the designated beneficiary). Determining a new life expectancy each year guarantees that the minimum distribution requirement will not force the plan account to be liquidated prior to the death of the plan participant and the spouse.133
3. Plan administrators would be responsible for distributing minimum distribution amounts
This proposal would make plan administrators responsible for annually distributing minimum distribution amounts from all plans (including traditional IRAs and Roth IRAs).134 In satisfying the minimum distribution requirement, IRA owners could no longer pick and choose between and among their IRAs.135 To enforce compliance, plan administrators would be liable for an excise tax (equal, for example, to 20% of what they fail to distribute) for any failure to comply with this distribution mandate.
As more and more retirement savings find their way into IRAs via rollovers, the need to make IRA custodians responsible for making minimum distributions has become more urgent. Strong policy grounds support the change. First, plan administrators likely have better resources to monitor compliance with the minimum distribution rules and can do so more efficiently. Second, plan participants are usually not well-versed on the complex minimum distribution requirements and are therefore prone to commit computational errors. Third, the introduction of an excise tax on plan administrators, rather than plan participants, would encourage compliance by those in the best position to do so.136 Finally, by monitoring the centralized efforts of plan administrators rather than the decentralized ef forts of plan participants, the Service will be in a better position to police minimum distribution rule compliance.
4. Accounts must be distributed within the year following the participant's (and spouse's) death
Upon the death of plan participants (and their spouses), qualified plan assets (including those held in Roth IRAs) would have to be distributed by December 31 of the year following the date of death. Except in the case of Roth IRAs or nondeductible contributions, any amounts distributed would immediately be subject to income tax.
The logic of this requirement is straightforward. The reason for the special, highly favorable tax treatment of qualified plans is to create an incentive to provide for one's retirement. When the participant and the participant's spouse are dead, there is no fiu-ther justification for costly tax incentives. 131
5. Limited income averaging for post-death distributions would be available
Under a progressive income tax regime, income bunching occurs when the receipt of income earned over several years is received in one taxable year. In some cases, income may be exposed to a higher marginal tax bracket than it otherwise would have been had it been received in smaller increments over longer periods of time. Although marginal tax rates are much lower today than in decades past, we are nowhere near a flat tax regime, and income bunching can cause increased taxes, albeit in a much more modest fashion than in prior years. Since the beneficiaries of a deceased participant will be compelled to withdraw the participant's account over a short time period (i.e., by the year following the death of the plan participant or his or her spouse), the distribution may be subject to an abnormally high tax bracket due to this bunching.
To minimize potential inequity and defuse a possible critique of this proposal, income averaging should be made available to the beneficiaries of a deceased participant's account. Now repealed, tenyear averaging, later reduced to five-year averaging, was generally available for any lump sum distribution at one time."' It would therefore not be difficult to restore either five- or ten-year averaging for post-death distributions. Five-year averaging seems the most reasonable option, but even ten-year averaging would be a small price to pay for establishing the principle that tax deferral must end after the participant's (or spouse's) death.
6 Reporting requirements would be increased
To enable the Service to enforce the minimum distribution rules effectively, plan administrators would be required to submit information returns to the Service notifying the Service of the commencement of minimum distributions and specifying the birthdates of the participant (and the participant's spouse, if the spouse is the designated beneficiary). In addition, the executor (or estate administrator) of a plan participant's estate would have an affirmative duty to notify the plan administrator of the plan participant's death (and/or the death of the plan participant's spouse, if named as the designated beneficiary).
7. Roth IRAs would be subject to the same minimum distribution rules as traditional IRAs
Currently, the minimum distribution rules do not apply to Roth IRAs until the death of the plan participant."' The purpose of the tax subsidy for qualified plans and IRAs is to encourage saving for retirement so that participants and their spouses can maintain their standard of living, not to encourage the creation of wealth for their heirs. Roth IRAs should not enjoy special tax status. Starting at the plan participant's RBD, distributions from Roth IRAs should commence. The minimum distribution rules should apply in a universal fashion to all tax-favored retirement saving.
B. Proposal Two: Eliminate the Existing Minimum Distribution Rules and Replace Them with a Date ofDeath Distribution Mandate
Some politicians may find the first proposal distasteful because in many cases it will accelerate the receipt of qualified plan distributions during the lives of the plan participants (and their spouses) and does not afford any tax deferral following their deaths. Their perception may be that this proposal, despite its inherent equity, amounts to a hidden tax increase, and, for that reason alone, they would not support it.
We, therefore, offer a second proposal in response to those who might harbor such political misgivings. Under this proposal, no distributions would have to be made during the life of plan participants (or their spouses). However, as in the first proposal, upon the death of the plan participant (and his or her spouse), all the assets held in the qualified plan would have to be distributed by December 31 of the year following such death. Income averaging would be available as in the first proposal.14
This simple system has numerous virtues. First, plan participants (and their spouses) can easily comprehend this rule. Second, there are virtually no compliance issues during the lives of participants (and their spouses), relieving plan administrators and the Service of much of their respective oversight responsibilities. Enforcement of the minimum distribution rules would occur only once, when the participant and spouse have died. The structure of this minimum distribution system, too, would continue to encourage savings over consumption during the lives of the participant and spouse.
But the virtues of this simple rule come at a steeper price than that associated with the first proposal. More specifically, finds that the minimum distribution rules currently force out during the lives of plan participants (and their spouses) would be permitted to accumulate tax free until their deaths. Initially, this represents a temporary increase in the current tax expenditure for qualified plans.'4' Moreover, because the wealthy are most able to capitalize upon the use of qualified plans,142 the additional tax subsidy would most likely accrue disproportionately in their favor.
Despite its added cost, the second proposal is still an improvement over the present system. Exchanging the allowance of lifetime accumulation for the elimination of post-death deferrals amounts to a significant reduction in the tax expenditure in the Long run. This is due to the fact that under the present rules when children or grandchildren are named designated beneficiaries the permitted deferral period after the death of the plan participant is typically much longer than the period between the participant's RBD and the participant's death. For example, in Table I, there are twenty years between the RBD and death of the last to die of Owner and Spouse, but an additional twenty-four years of deferral to Child. Had the designated beneficiary named by Spouse been Grandchild (or Greatgrandchild), age two when Owner died, the deferral could have produced an additional seventy-one years of deferral!
C Transition Rules
As with any proposed change in the tax laws, the issue arises as to whether there should be some transition rules to cushion the effect of the change. In connection with either proposal, the case for a transition rule is extremely weak. The major change in both proposals that is detrimental to taxpayers is the loss of tax deferral after the death of the participant and the participant's spouse. Thus, a huge part of the tax subsidy is retained. Moreover, the proposals can be viewed as a refinement, hardly unprecedented, of the longestablished concept that nonretirement benefits (e.g., benefits to heirs) are supposed to be "incidental" to pension plans. It is extremely relevant that when Congress first imposed the minimum distribution requirements on all plans, it saw no need to provide a transitional rule, other than deferring the effective date a few years for governmental and collectively bargained plans.l4a Just as there may have been many at that time who had planned to accumulate their retirement benefits well after age seventy, today there may be many who had planned on their heirs accumulating benefits well after their death. Neither expectation is particularly worthy of protection,
D. Benefits Associated with the Adoption of Either of These Proposals Consider the effects were Congress to adopt either of these proposals,144 No longer would the minimum distribution rules be a source of plan participant confusion and malaise. Armed with a sensible set of rules that they could comprehend, brokerage houses, tax planners, and the Service would benefit as well. Finally, under either proposal, although certainly more so under the first, unnecessary and inequitable tax subsidies would be reduced. The simple truth is that tax-free compounding already provides more than adequate incentive for taxpayers to participate in and contribute to qualified plans. Were either of these proposals adopted, few, if any, plan participants would forfeit their participation in qualified plans or reduce their contributions. Under current law, the benefits of tax-free compounding that serendipitously inure to designated beneficiaries (other than the plan participant's surviving spouse) are thus entirely unwarranted and unnecessary to spur qualified plan participation.
Yet, congressional passage of either of these proposals would not necessarily be easy.145 The proposal to modify the existing minimum distribution rules would likely face stiff opposition from plan participants who have large retirement benefits as well as from their beneficiaries who would lose a significant tax deferral advantage.146 Despite this potential opposition, the shortcomings of the minimum distribution rules are too weighty to be ignored; they must be scrapped for something better.
VI. CONCLUSION
The current minimum distribution rules threaten to turn the pool of wealth held by qualified plans into a vast and burgeoning ocean. Following the model set by Chinese officials in constructing the Three Gorges Dam, Congress should properly harness the power of this ocean of wealth. Adoption of either proposal set forth in Section V would establish effective floodgates for the retirement system, ensuring that the tax-subsidized wealth housed in the sanctuary of qualified plans is used primarily to meet the retirement needs of plan participants (and their spouses) and not to create windfalls for their heirs. Immense amounts of revenue would be generated while leaving intact more than sufficient incentives to save for retirement. From an administrative and equity perspective, the outcome of this harnessing process would be awe-inspiring, creating a structure as magnificent as the Three Gorges Dam.
FOOTNOTE* Jay A. Soled is a professor at Rutgers University, and Bruce A. Wolk is a professor at the University of California, Davis.
1. See Arthur Zich, China's Three Gorges Before the Flood, NAT'L GEOGRAPHIC MAG., Sept. 1997, at 2.
2. See id. at 10-11.
3. See U.S. DEFT OF COMMERCE, 1998 STATISTICAL ABSTRACT OF THE UNITED STATES 534 (stating that qualified plans, including assets held by state and local governments, held $6.946 trillion at the end of 1997). Treasury officials recently reached similar estimates. Lubick's Summary of Administration's Pension Reform Proposals at W&M Hearing, TAx NOTES TODAY, Mar. 24, 1999, at 39-56.
FOOTNOTE4. Unless otherwise indicated, all references are to the Internal Revenue Code of 1986, as amended.
5. For purposes of this analysis, the term "qualified plans" refers to the entire gamut of tax-favored retirement plans, all of which are subject to the minimum distribution rules. These include employer-provided qualified pension and profit-sharing plans (LRC. 401), tax deferred annuities (LRC. 403), unfunded deferred compensation plans of state and local governments (LRC. 457), individual retirement accounts ("IRAs") (LRC.RlC. 408), and Roth IRAs (LRC. 408A).
6. For an overview of the early historic stages of private pension plans, see WILLIAM C. GREENOUGH & FRANCIS P. KING, PENSION PLANS AND PUBLIC POLICY 27-47 (1976). For an overview of the economic motivations that inspired businesses to implement private pension plans, see STEVEN A. SASS, THE PROMISE OF PRIVATE PENSIONS: THE FIRST HUNDRED YEARS 18-37 (1997).
FOOTNOTE7. See JOHN H. LANGBEIN & BRUCE A. WOLK, PENSION AND EMPLOYEE BENEFIT LAw 25-29 (2000) (outlining the characteristics of those employees who participate in qualified plans).
8. See LRC. 404(a)(1)-(3). In the case of traditional IRAs, there is no employer deduction, but the contribution is directly deductible by the employee. See id. 219.
9. See id. 402-403.
10. See id. 408(e)(1), 501(a). As in the case with exempt organizations, qualified plans are taxed on "unrelated business income," which is essentially the income from any trade or business regularly carried on by the trust or by a partnership of which it is a member. See id. 511-513(6).
11. See Michael J. Graetz, The Troubled Marriage of Retirement Security and Tax Policies, 135 U. PA. L. REV. 851, 855 (1987) ("Thus, while there are no doubt disagreements at the margin, replacement of some significant portion of preretirement wages must be the fundamental goal of retirement security policy."). A series of other law reviews express similar views. See Nancy Altman, Rethinking Retirement Income Policies: Nondiscrimination, Integration and the Quest for Worker Security, 42 TAx L. REV. 433 ( 1987); Michael A. Oberst, A Perspective of the Qualified Plan Tax Subsidy, 32 BUFF. L. REV. 603 (1983); Bruce A. Wolk, Discrimination Rules for Qualified Retirement Plans: Good Intentions Confront Economic Reality, 70 VA. L. REV. 419 (1984).
12. See Robert L. Clark & Elisa Wolper, Pension Tax Expenditures: Magnitude, Distribution, and Economic Effects, in PUBLIC POLICY TOWARD PENSIONS 41-84 (Sylvester J. Schieber & John B. Shoven eds., 1997) (pointing out that revenue losses attributable to pension plan provisions constitute the single largest tax expenditure in the federal budget); tee infra Table I and accompanying text (detailing the magnitude of the tax expenditure).
13. See 401(a)(9).
FOOTNOTE14. See Self Employed Individuals Tax Retirement Act of 1962, Pub. L. No. 87-792, 2, 76 Stat. 809, 809 ( 1962 ). Even prior to the introduction of the minimum distribution rules, the Internal Revenue Service ("Service") took the position that plan participants' efforts to provide death benefits would violate the exclusive benefit rule. The exclusive benefit rule requires that qualified plans primarily benefit employees rather than their designated beneficiaries. See Rev. Rul. 56-656, 1956-2 C.B. 280 (stating that an arrangement does not qualify under LR.C. 401 (a) if the benefits it provides are not payable to an employee but only to his beneficiary upon his death); Rev. Rul. 74-360, 1974-2 C.B. 130 (holding that because the participant's beneficiary or beneficiaries may receive concurrent payments from the plan prior to the death of the plan participant, the profit-sharing plan fails to qualify under LRC, $ 401(a)).
FOOTNOTE15. See Tax Equity and Fiscal Responsibility Act of 1982, Pub. L. No. 97-248, 310(b)(1), 96 Stat. 324, 327 (1982).
16. See infra Section III.
17. See SASS, supra note 6, at 147-78; Olivia S. Mitchell & Sylvester J. Schieber, Defined Contribution Pensions. New Opportunities, New Risks, in LIVING WITH DEFINED CONTRIBUTION PENSIONS: REMAKING RESPONSIBILITY FOR RETIREMENT 1-15 (Olivia S. Mitchell & Sylvester J. Schieber eds., 1998) [hereinafter DEFINED CONTRIBUTION PENSIONS.
18. The amount of this benefit is usually based upon a formula that takes into account factors such as the participant's age, years of service, and salary. For a general description of defined benefit plan dynamics, see MICHAEL J. CANAN, QUALIFIED RETIREMENT AND OTHER EMPLOYEE BENEFIT PLANS 3i 3.51-.59 (1997).
19, See RICHARD A. IPPOLITO, PENSIONS, ECONOMICS AND PUBLIC POLICY 103-05 ( 1986).
20. This is because (i) spousal consent is required to name a designated beneficiary other
FOOTNOTEthan the spouse, see LRC. 401(a)(11), and (ii) upon the death of the plan participant, under the terms of most defined benefit plans the surviving spouse cannot name a new designated beneficiary.
21. See ANGELA CHANG, EXPLANATIONS FOR THE TREND AWAY FROM DEFINED BENEFIT PENSION PLANS, CONGRESSIONAL RESEARCH SERVICE REPORT FOR CONGRESS (1991); KELLY OLSEN & JACK VANDERHIE, EMPLOYEE BENEFIT RESEARCH INSTITUTE, SPECIAL REPORT SR-33/EBRI ISSUE BRIEF NO. 190, DEFINED CONTRIBUTION PLAN DOMINANCE GROWS ACROSS SECTORS AND EMPLOYER SIZES, WHILE MEGA DEFINED PLANS REMAIN STRONG (1997).
22. For a general description of defined contribution plan dynamics, see CANAN, supra note 18, at 31 3.11-.19.
23. Pub. L. No. 93-406, 88 Stat. 829 (1974).
24. See LRC. 412.
FOOTNOTE25. See ERISA SS 4002-4003.
26. See id. S 4006(a)(3).
27. See id. S 4022(a) (the PBGC "shall guarantee [the] payment of all nonforeitable benefits [under] a single employer plan").
28. See id. S 4062.
FOOTNOTE29. For an overview of these legislative changes, see Edwin C. Hustead, Trends in Retirement Income Plan Administrative Expenses, in DEFINED CONTRIBUTION PENSIONS, supra note 17, at 167-70. In addition, Hustead comments that "[the cost of administering retirement plans has continuously increased since the enactment of ER.ISA in 1974. This is particularly true for defined benefit plans." Id. at 175-76.
30. Even many of those employers who have been maintaining defined benefit plans have been replacing them with a special form of such plans known as a cash balance pension plan. See, eg., Lee Sheppard, Pension Downsizing, Continued, 83 TAx NOTES 1107 (1999) (discussing the use of cash balance plans as a mechanism to avoid penalties associated with a pension plan termination); IBM Retools Pensions: New Plan Sweeping Corporate America, U.S.A. TODAY, May 4, 1999, at Al (estimating that cash balance plans comprised over 12% of all defined benefit plans in 1998 and that this is a growing trend). Cash balance pension plans bear a close resemblance to defined contribution plans. One benefit treatise highlights their fundamental characteristics as follows: "Each participant has an account that is credited with a dollar amount that resembles an employer contribution and is generally determined as a percentage of pay. Each participant's account is also credited with interest. The plan usually provides benefits in the form of a lump-sum distribution or annuity." EMPLOYEE BENEFIT RESEARCH INSTITUTE, FUNDAMENTALS OF EMPLOYEE BENEFIT PROGRAMS 111 (5N ed. 1997). For an excellent discussion of the reasons that underlie the rising popularity of cash balance plans, see Anna M. Rappaport et al., Cash Balance Pension Plans, in POSITIONING PENSIONS FOR THE TWENTY-FIRST CENTURY 29-44 ( 1997).
31. See Mitchell & Schieber, supra note 17, at 9.
FOOTNOTE32. SeeI.R.C. S 401(k)k).
33. See id. 408.
34. See id. 402(c).
FOOTNOTE35. These implications include various concerns such as risk, liquidity, and savings rates. These implications are analyzed in William G. Gale & Joseph M. Milano, Implications of the Shift to Dined Contribution Plans for Retirement Wealth Accumulation, in DEFINED CONTRIBUTION PENSIONS, supra note 17, at 115-35.
FOOTNOTE36. See U.S. DEFT OF THE TREASURY, TAX REFORM FOR FAIRNESS, SIMPLICITY, AND ECONOMIC GROWTH 328-32 (1984).
37. See H.R. REP. NO. 99-426, at 740 (1985).
38. These were distributions made to plan participants in excess of the greater of (i) $150,000 or (ii) the indexed amount referred to as the annual threshold amount. See LRC. 4980A(a) (repealed 1997).
39. In general, this amount was equal to the difference between the value of a hypo thetical annuity (i.e., a single life annuity based on the decedent's age at his or her death multiplied by an amount equal to the annual threshold amount in effect in the year in which the decedent died) and the participant's account balance. See id. 4980A(d) (repealed 1997).
40. See Tax Reform Act of 1986, Pub. L. No. 99-514, 1133(a), 100 Stat. 2085, 2481 ( 1986). For a detailed discussion of the purpose and operation of these excise taxes, see Bruce A. Wolk, The New Excise and Estate Taxes on Excess Retirement Plan Distributions and Accumulations, 39 U. FLA. L. REV. 987 (1987).
41. See LRC. 4980A(a), (d) (repealed 1997).
42. See, eg., Thomas A. Kirschbaum 8t Louis Kravitz, Minimizing Taxes on Excess Retirement Distributions and Accumulations, 17 TAX'N FoR LAW. 216 ( 1989).
FOOTNOTE43. See Small Business Job Protection Act of 1996, Pub. L. No. 104-188, 1452(6), 110 Stat. 1755, 1816 (effective after Dec. 31, 1999). The moratorium on the excess distribution excise tax was declared in hopes of raising revenue as taxpayers had added incentive to withdraw large amounts from their qualified plans. The theoretical justification offered was that the same legislation also repealed LR.C. 415(e), but not until the year 2000. This Code section imposed a combined limit on benefits from defined benefit and defined contribution plans maintained by the same employer. See S. REP. NO. 104-281, at 91-92 (1996), reprinted in 1996 L7.S.C.C.A.N. 1474, 1565-66.
44. Pub. L. No. 105-34, 1073, 111 Stat. 788, 948 (1997). In repealing both excise taxes, the Senate offered many justifications. "The Committee believes that the limits on contributions and benefits applicable to each type of vehicle are sufficient limits on tax-deferred savings. Additional penalties are unnecessary, and may also deter individuals from saving. The excess accumulation and distribution taxes also inappropriately penalize favorable investment returns." SENATE COMM. ON FIN., REVENUE RECONCILIATION ACT OF 1997, S. REP. NO. 105-33, at 199-200 (1997).
45. Pub. Law No. 104-188, 1452(a), 110 Stat. at 1816 (1996).
46. See LRC. 415(e) (repealed 1996).
47. See id.
48. To be sure, there may be estate taxes due upon the death of the plan participant, but these would have been payable if the same wealth had been accumulated outside of the qualified plan. The estate tax recovers some of the tax subsidy but still leaves much of it in the hands of the plan participant's designated beneficiaries.
FOOTNOTE49. pub. L. No. 105-34, 302(a), 111 Stat. at 825-28.
50. See LRC. 408A(a).
51. See id. 408A(c)(1).
52. A "qualified distribution" means any payment or distribution (a) made on or after the date on which the individual reaches age 5912, (b) made to a beneficiary (or to the estate of the individual) on or after the individual's death, (c) attributable to the individual's being disabled, or (d) which is a qualified "special purpose" distribution (i.e., for any qualified first-time homebuyer to which LRC. 72(t)(2)(F) applies). See id. 408(d)(4)(A).
53. See id. 408A(c)(4).
54. See, eg., Lynn Asinof, Stuck for a Gift Idea for a Favorite Child? A Surprising Choice May be a ROTH IRA, WALL ST. J., Dec. 16, 1998, at CI; Laura Saunders, In Roth We Trust, FORBES, Apr. 20, 1998, at 466. For a stinging critique of the fiscal damage the use of Roth IRAs will wreak on the treasury, see Daniel Halperin, 1 Want a ROTH IRA for Xmas, 82 TAx NoTES 1567 (1998).
55. See supra note 53.
56. Sec 408A(c)(2)(A).
57. See id. 408A(c)(3)(B).
FOOTNOTE58. See generally, Robert L. Clark & Sylvester J. Schieber, Factors Affecting Participation Rates and Contribution Levels in 401 (k) Plans, in DEFINED CONTRIBUTION PENSIONS, supra note 17, at 69-97.
59. See SOCIAL SEC. AND MEDICARE BDS. OF TRUSTEES, STATUS OF THE SOCIAL SECURITY AND MEDICARE PROGRAMS 6-10 (1996.
60. Before the recent era of projected budget surpluses, some commentators and members of the General Accounting Office and Congressional Budget Office maintained that the wealth of qualified plans should be used to help erase the federal deficit and simultaneously
FOOTNOTEfacilitate Social Security funding. See CONGRESSIONAL BUDGET OFFICE, REDUCING THE DEFICIT: SPENDING AND REVENUE OPTIONS 299 (1994) (estimating that a 5% tax on investment income of pension plans and IRAs would raise revenue by $6.9 billion in year 1 and $13.6 billion in year 5); Alicia Munnell, Current Taxation of Qualified Pension Plans: Has the Time Come?, NEW ENG. ECON. REV. Mar.-Apr. 1992, at 12 (advocating a 1590 tax on all employer contributions to pension and profit sharing arrangements as well as a 15% levy on the annual earnings of such arrangements).
61. In describing the difficulty of explaining the minimum distribution rules to friends, one author of this analysis makes the following comparison: Attempting to master the minimum distribution rules is like going into a field and trying to catch grasshoppers in a glass jar. You have to make numerous attempts to catch one. Each successful catch makes the next more difficult, and when the hunt is over, what's left in the glass jar is not exactly a pretty sight.
FOOTNOTE62. Distributions from a Roth IRA may be suspended until the plan participant's death, see LRC. 408A(c)(5), at which time the minimum distribution rules then apply, see id. 401(a)(9)(B).
63. See id. 4974.
FOOTNOTE64. The determination of whether or not an employee is a 5% owner is made in accordance with LR.C. 416 but is made without regard to whether the plan is top-heavy. See id. 401(a)(9)(C)(ii)(I); Prop. Treas. Reg. 1.401(A)(9)-1, Q&A B-2(d)(2), 52 Fed. Reg. 28,070, 28,076 (1987).
65. See LR.C. 401(a)(9)(C)(i)(II). To illustrate, suppose A is a rank-and-file employee of company X. If A participates in X's defined contribution plan and retires at the end of 2001 at the age of 75, a minimum distribution out of the plan would have to commence on or before April 1, 2002.
66. See id. 401(a)(9)(C)(i)(I). To illustrate, suppose B retires at age 65 from company Y and he soon thereafter rolls over his retirement account balance into an IRA. If B turns 70 in 2001, the minimum distribution out of the IRA would have to be made on or before April 1, 2002.
FOOTNOTE67. See Prop. Treas. Reg. 1.401(a)(9)-1, Q&A F-1(b), 52 Fed. Reg. at 28,084. The first distribution calendar year is the later of the year (a) a plan participant attains age 70/ or (b) the employee retires or becomes a 5%0 owner of the employer. See id. This rule applies on a plan-by-plan basis (i.e., a plan participant who participates in more than one qualified plan may have different distribution calendar years for each). See id.
68. See Prop. Treas. Reg. 1.401(a)(9)-1, Q&A F-3(a), 52 Fed. Reg. at 28,084.
69. See Prop. Treas. Reg. 1.401(a)(9)-I, Q&A F-4A, 52 Fed. Reg. at 28,086; infra subsection III.B.4.
FOOTNOTE70. To illustrate this latter point, consider what would happen if a plan participant names her granddaughter as the designated beneficiary and the granddaughter is 40 years junior to the plan participant. In this case, annual annuity payments to the granddaughter cannot exceed 54% of the amount of annual annuity payments made to the plan participant. The table that sets forth these percentage limitations is found in Prop. Treas. Reg. 1.401(a)(9)-2, Q&A 6(b)(2), 52 Fed. Reg. at 28,100.
71. If distributions from a defined benefit plan are not in the form of an annuity, the plan participant's benefit will be treated as an individual account for purposes of determining the minimum distribution amount. See Prop. Treas. Reg. 1.401(a)(9)-1, Q8zA F-3(e), 52 Fed. Reg. at 28,085.
72. See Prop. Treas. Reg. 1.401(a)(9)-1, Q8rA F-5(b)-(c), 52 Fed. Reg. at 28,086. For example, increasing for allocations of contributions or forfeitures and decreasing for distributions.
73. See Prop. Treas. Reg. 1.401(a)(9)-1, Q&A F-5(a), 52 Fed. Reg. at 28,026.
74. See Prop. Treas. Reg. 1.401(a)(9)-1, Q&A F-1(a), 52 Fed. Reg. at 28,083.
75. See Prop. Treas. Reg. 1.401(a)(9)-1, Q8tA E-1(a), 52 Fed. Reg. at 28,080.
76. See Prop. Treas. Reg. 1.401(a)(9)-1, Q&A E-3 & 4, 52 Fed. Reg. at 28,081.
FOOTNOTE77. See Prop. Treas. Reg. 1.401(a)(9)-2, Q&A 4 & 5, 52 Fed. Reg at 28,098-99.
78. See Prop. Treas. Reg. 1.401(a)(9)-1, Q&A E-8(a), 52 Fed. Reg. at 28,082; See generally, David Johnson, Recalculating the Life Expectancy Election, TR 8r EST., Nov. 1990, at 8.
FOOTNOTE79. See Prop. Treas. Reg. 1.401(a)(9)-1, Q8tA E-1(a), 52 Fed. Reg. at 28,080; supra tent accompanying notes 75-76.
80. See Prop. Treas. Reg. 1.401(a)(9)-1, Q&A E-7(b), 52 Fed. Reg. at 28,082.
81. See, eg., Priv. Ltr. Rul. 9712032 (Mar. 21, 1997).
FOOTNOTE82. See LRC. 401(a)(9)(D); Prop. Treas. Reg. 1.401(A)(9)-1, Q&A E-8(b), 52 Fed. Reg. at 28,083. However, the incidental benefit requires a recalculation of the designated beneficiary's life expectancy, at least until the death of plan participant. This rule is reflected in the tables under Prop. Treas. Reg. 1.401(A)(9)-2, Q8u16, 52 Fed. Reg. at 28,099-100.
83. If a trust is named as beneficiary of a plan participant, all beneficiaries of the trust may be treated as having been designated as beneficiaries of the participant. See Prop. Tres. Reg. 1.401(A)(9)-1, Q&A D-5(a), 52 Fed. Reg. at 28,080 (as amended by 62 Fed. Reg. 67,780, 67,783 (1997)). This will happen if, as of the later of the date on which the trust is named as a beneficiary of the participant or the participant's required beginning date, and as of all subsequent periods during which the trust is named as a beneficiary, the four following requirements are met: (1) the trust is a valid trust under state law, or would be but for the fact that there is no corpus; (2) the trust is irrevocable or will, by its terms, become irrevocable upon the death of the plan participant; (3) the beneficiaries of the trust are identifiable from the trust instrument; and (4) the plan administrator must be provided with either a copy of the trust instrument or a certified list of all beneficiaries (including contingent and remainderman beneficiaries with a description of the conditions on their entitlement). See id. 1.401(A)(9)-1, Q&AD-5(b), 52 Fed. Reg. at 28,080 (as amended by 62 Fed. Reg. at 67,783). In addition, within nine months after the plan participant's death, the trustee must provide the plan administrator with a copy of the trust or provide the plan administrator with a final list of trust beneficiaries (including contingent and remainderman beneficiaries with a description of the conditions on their entitlement). See Prop. Treas. Reg. 1.401(A)(9)-1, Q&A D-7, 62 Fed. Reg. 67,780, 67,784 (1997).
84. See Prop. Treas. Reg. 1.401(A)(9)-1, Q8rA D-2A(a), 52 Fed. Reg. at 28,079. Note that the members of a class of beneficiaries capable of expansion or contraction (e.g., the participant's children) will be treated as being identifiable if it is possible at the applicable time
FOOTNOTEto identify the class member with the shortest life expectancy. See Prop. Treas. Reg. 1.401(A)(9)-1, Q&A D-2(a)(1), 52 Fed. Reg. at 28,079.
85. See Prop. Treas. Reg. 1.401(A)(9)-1, Q8rA D-2(a), 52 Fed. Reg. at 28,079.
86. See Prop. Treas. Reg. 1.401(A)(9)-1, Q8tA D-3(a), 52 Fed. Reg. at 28,079.
87. See Prop. Treas. Reg. 1.401(A)(9)-1, Q8tA E-5(a), 52 Fed. Reg. at 28,081.
88. See Treas. Reg. 1.401-1(b)(1)(as amended in 1976); supra note 11 and accompanying text.
89. See LRC. 401(a)(9)(G); Prop. Tress. Reg. 1.401(a)(9)-2, 52 Fed. Reg. at 28,098. Prior to the Tax Reform Act of 1986, the Treasury Department had set forth a MDIB requirement in a number of revenue rulings interpreting the term "incidental" as applied to death benefits. See supra note 14. The purpose of this requirement was to limit the portion of the participant's benefit that would be paid after the participant's death. The policy justification for tax subsidized death benefits is much weaker than for subsidized retirement benefits. The MDIB requirement has now been made a part of the minimum distribution requirements. Thus, distributions must satisfy not only the regular minimum distribution rules discussed above but also the MDIB requirement as well.
90. See LRC. 401(a)(9)(G); Prop. Treas. Reg. 1.401(A)(9)-2, 52 Fed. Reg. at 28,098.
91. See Prop. Treas. Reg. 1.401(A)(9)-2, Q&A 6(b)(2), 52 Fed. Reg. at 28,100.
FOOTNOTE92. See Prop. Treas. Reg. 1.401(A)(9)-2, Q8tA 7(a), 52 Fed. Reg. at 28,100.
93. See Prop. Treas. Reg. 1.401(A)(9)-1, Q&AF-3A(b)(1), 52 Fed. Reg. at 28,085.
94. The MDIB requirement establishes a table specifying a limit on the divisor, which changes each year based on the plan participant's age during the distribution year. See Prop. Treas. Reg. 1.401(A)(9)-2, Q8rA-4, 52 Fed. Reg. at 28,098-99.
95. See Prop. Treas. Reg. 1.401(A)(9)-1, Q&A E-8(c), Exs. 1 & 2, F-3A(b)(1), 52 Fed. Reg. at 28,083, 28,085. To illustrate the more powerful effects this rule has when a plan participant names an even younger designated beneficiary, consider the following fact pattern. Assume that an employee, age 70, names his grandchild, age 10, as his beneficiary under his profit-sharing plan. The employee elects installment payments over their joint life expectancy, which is 71.8 using Table VI of Treas. Reg. 1.72-9. This would permit the employee to receive as little as 1/71.8 of his account for the first distribution year. The MDIB requirement limits the divisor. For an employee age 70 the divisor is 26.2, which is essentially just the joint life expectancy of a 70 year old and' a 60 year old. The payments can still be spread out over the joint life expectancy of the employee and the grandchild, but they will be considerably skewed toward the earlier years when the employee is more likely to be alive. Once the employee dies, however, the MDIB requirement no longer applies. Thus, if the employee dies at
FOOTNOTEage 72 1/2 (during the third distribution year), the payments to the grandchild can be spread out over the remainder of their joint life expectancy, i.e., 71.8 less 3, or 68.8 years. See id.
96. See Prop. Treas. Reg. 1.401(A)(9)-1, Q&A B-3A, 52 Fed. Reg. at 28,077.
97. SeeI.RC.401(a)(9).
98. See Prop. Treas. Reg. 1.401(A)(9)-1, Q&A B-1(a), 52 Fed. Reg. at 28,076.
FOOTNOTE99. See Prop. Treas. Reg. 1.401(A)(9)-1, Q&A C-2, 52 Fed. Reg. at 28,077.
100. See LRC. 401(a)(9)(B)(ii); Prop. Treas. Reg. 1.401(A)(9)-1, Q&A C-1(a), 52 Fed. Reg. at 28,077. To illustrate, suppose A, a participant of Plan X, dies on July 15, 2000, at age 65 and his designated beneficiary is his estate. In accordance the five-year rule, the entire account balance of Plan X must be distributed to A's estate by Dec. 31, 2005.
101. See Prop. Treas. Reg. 1.401(A)(9)-1, Q&A C-3(a), 52 Fed. Reg. at 28,077-78.
102. SeeI.RC. 401(a)(9)(B)(iii); Prop. Treas. Reg. 1.401(A)(9)-1, Q8tA C-1(a), 52 Fed. Reg. at 28,077.
103. See Prop. Treas. Reg. 1.401(A)(9)-1, Q8tA C-3(b), 52 Fed. Reg. at 28,075.
104. See LRC. 402(c).
105. The ability to name a new designated beneficiary is an important tax deferral device. The selection of a designated beneficiary who is younger than the plan participant (or his or her spouse) has the direct effect of reducing the minimum distribution amount.
FOOTNOTE106. See I.IC. 401(a)(9)(B)(i).
107. See Prop. Treas. Reg. 1.401(A)(9)-1, Q&A E-8(a), 52 Fed. Reg. at 28,082.
108. See Prop. Treas. Reg. 1.401(A)(9)-1, Q&A B-4, F-3A(a), 52 Fed. Reg. at 28,077, 28,085.
109. See id.
110. See Prop. Treas. Reg. 1.401(A)(9)-2, Q&A 7(a), 52 Fed. Reg. at 28,100.
FOOTNOTE111. See Prop. Treas. Reg. 1.401(A)(9)-1, Q&A E-7(c), 52 Fed. Reg. at 28,082.
112. See, eg., Priv. Ltr. Rul. 9534027 (Aug. 25, 1995) (under the authority of Prop. Treas. Reg. 1.408-8, Q8tA A-4(b), 52 Fed. Reg. at 28102-03, the surviving spouse may claim the participant's IRA as the surviving spouse's own).
113. See LRC. 4974(a). To illustrate, suppose A is a participant in Plan Z. If A's minimum distribution from Plan Z is $10,000 and A withdraws $7,000, then the excise tax would be equal to $1,500 (.50 x ($10,000 - $7,000)).
114. See id. 4974(d).
FOOTNOTE115. For example, you might hear a minimum-distribution-rule pundit say, "You're nearing your RBD. Did you receive your minimum distribution during this distribution calendar year?" Alternatively, you might hear another such pundit say, "I'm using the recalculation method based on my ALE and my spouse is using the term certain method. We have chosen a qualifying trust as our contingent designated beneficiary."
116. See 6 Stand. Fed. Tax Rep. (CCH) x1 17,724-17,725 (2000).
117. See, eg., Lynn Asinof, Oops . . . How a Variety of Basic Foul-Ups Are Bedeviling the Beneficiaries of IRAs, WALL ST. J., Mar. 29, 1999, at Cl; Mary Rowland, Who's Advising Financial Advisors?, N.Y. TIMES, May 3, 1993, at B2.
118. See I.R.C. S 2056(b)(7)(B)(ii).
119. The Service has fortunately issued guidance in this perplexing and evolving area of the law. See Rev. Rul. 2000-2, 2000-3 LRB. 305. The ruling clarifies that the trustee of the marital deduction trust (usually what is known as a QTIP trust) does not have to withdraw
FOOTNOTEIRA income annually. All that is required is that the spouse have the power, exercisable annually, to compel the trustee to withdraw the IRA income and pay it to the spouse.
Note that the trustee must still withdraw the required minimum distribution amount. The issue considered in the ruling is whether the trustee has to withdraw even more if the IRA income exceeds this amount. It is therefore crucial to draft the trust to specifically give the spouse the requisite power over the IRA income.
120. See Marcia Chadwick Holt, Retirement Planning: A Practical Guide to Making the Tough Choices, 29 MIAMI INST. ON EsT. PLAN., 11 400-410.10 (1995); Louis A. Mezzullo, Serving an Ace Without a Foot Fault When Planning for Qualified Plan Benefits: Spousal Rollovers, Excise Taxes, Charitable Bequests, and IRAs, 32 MIAMI INST. ON EsT. PLAN., I[ 1100.11103.6 (1998).
121. See Jay A. Soled, When Will Congress Police the Minimum Distribution Rules, 86 TAx NOTES 1003 (2000).
The complexity of the minimum distribution rules has perhaps led the Service to embrace taxpayer-friendly positions in developing responses to several private ruling requests. Consider, for example, the outcome reached in Private Letter Ruling 199915063 (Apr. 16, 1999). In this ruling, when the plan participant reached his RBD, he named his designated beneficiaries. This designation gave the plan participant the opportunity to use the dual life expectancy tables in determining the minimum distribution amount. The plan participant, however, instead elected to have the minimum distribution amounts computed using only his recalculated life expectancy. Upon the plan participant's death, in accordance with the regulations, the plan participant's life expectancy was reduced to zero. Despite this fact, the Service ruled that the "at least as rapidly rule" would serve to penalize the plan participant just because he had chosen to take distributions in a more rapid fashion than he was required. Instead, because the plan participant had timely chosen his designated beneficiary by his RBD, "the applicable life expectancy [for purposes of determining the minimum distribution amounts) is the life expectancy of the designated beneficiary." Priv. Ltr. Rul. 199915063 (Apr. 16, 1999 ). This ruling and others like it show incredible tolerance on the part of the Service. See, eg., Priv. Ltr. Rul. 199908063 (Feb. 26, 1999) (surviving spouse is deemed designated beneficiary where IRA named estate as the beneficiary and surviving spouse was the sole beneficiary of the estate). This tolerance is probably attributable to the fact that the Service empathizes with the plight of plan participants and their advisors in comprehending the complexities of the minimum distribution rules.
FOOTNOTE122. See generally, Piroska Soos, Self Employed Evasion and Tax Withholding: A Comparative Study and Analysis of the Issues, 24 U.C. DAVIS L. REV. 107 (1990); American Bar Association, Report of the Second Invitational Conference on Income Tax Compliance, 42 TAx LAw 705 (1989); American Bar Association, Report and Recommendation on Taxpayer Compliance, 41 TAx LAw 329 (1988).
123. For an illustration of the significance of the tax dollars that can be saved, see supra Table I and accompanying text.
FOOTNOTE124. See JOINT COMM. ON TAXATION, JCS-13-99, ESTIMATES OF FEDERAL TAX EXPENDITURES FoR FISCAL YEARS 2000-2004, at 23 ( 1999 ) (for fiscal year 2000, the tax expenditure for the net exclusion for pension contributions and earnings is estimated to be $93.2 billion).
FOOTNOTE125. See David E. Bloom & Richard B. Freeman, The Fall in Private Plan Coverage in the United States, 82 AMER. ECON. REV. 539 (1992) (showing how less educated males experienced particularly severe declines in pension coverage in the 1980s); Craig J. Langstraat, The Individual Retirement Account: Retirement Help for the Masses, or Another Tax Break for the Wealthy?, 60 ST. JOHN'S L. REV. 437 (1986) (arguing that IRA provisions favor wealthy taxpayers); Bruce A. Wolk, Discrimination Rules for Qualified Retirement Plans: Good Intentions Confront Economic Realty, 70 VA. L. REV. 419, 430-31 (1984) ("[L]ow paid employees . . . may be less willing to save than others because of pressing current consumption needs.").
FOOTNOTE126. See Angela E. Chang, Tax Policy, Lump-Sum Pension Distributions, and Household Savings, 49 NAT'L TAX J. 235 (1996) (presenting empirical evidence that those with low incomes are less likely to roll over qualified plan distributions than are the highly compensated).
127. The minimum distribution rules are inequitable from another perspective. They cause otherwise similarly situated plan participants to bear different tax burdens. This inequity is due to such factors as the governing terms found in qualified plan documents that limit distribution modes and marital circumstances that necessitate the use of marital trusts, both of which can curtail deferral opportunities.
128. See John A. Herbers, leveraging an IRA's Tax Deferral Into Multi-Generational Wealth, TR. & EsT., Jan. 1997, at 10.
129. See STAFF OF THE JOINT COMM. ON TAXATION, GENERAL EXPLANATION OF THE TAX REFORM of 1986, at 710.
Uniform minimum distribution rules which establish the permissible periods over which benefits from any tax-favored retirement arrangement may be distributed ensure that plans are used to fulfill the purpose that justifies their tax-favored status
FOOTNOTEreplacement of a participant's preretirement income stream at retirement-rather than for the indefinite deferral of tax on a participant's accumulation under the plan. Id.
FOOTNOTE130. Adoption of either of these proposals would have the immediate effect of significantly increasing the number of taxable lump sum distributions. In recognition of this fact and to make these proposals more politically palatable, this analysis advocates returning to a system of income averaging for such lump sum distributions.
131. The complex MDIB requirements (and their 10 single-spaced pages in the regulations) would no longer be necessary, since they only apply when the life expectancy of a nonspousal designated beneficiary is used. Note that participants could name anyone as the actual beneficiary of the plan, but, unless the designated beneficiary were a spouse, minimum distributions could only be made over the life of the plan participant. Thus, plans would not be permitted to pay survivor annuities to anyone other than a spouse.
FOOTNOTE132. See Treas. Reg. 1.72-9, Tables V (one life) 8t VI (two lives) (1999).
133. It also means that at the death of one spouse minimum distributions will be increased since they will be based solely on the life of the survivor.
134. Currently, only administrators of employer-provided plans have responsibility for ensuring that the plans they administer distribute the minimum distribution amounts to plan participants. See LRC. 401(a)(9). A plan that fails to meet the minimum distribution requirements would be disqualified, although the regulations provide that the disqualification sanction will not be imposed for isolated failures. See Prop. Treas. Reg. 1.401(A)(9)-1, A-5, 52 Fed. Reg. 28,070, 28,099 (1987). In contrast, with respect to traditional IRAs and Roth IRAs, the responsibility for withdrawing the minimum distribution amounts rests with the plan participants themselves. An IRA that fails to make minimum distributions would cease to be tax-exempt, see LR.C. 408(a)(6), but since each IRA is a separate plan, it is only the plan participant who is affected.
135. Although the required minimum distribution must currently be calculated separately for each IRA, see Prop. Treas. Reg. 1.408-8, A-1, 52 Fed. Reg. at 28,102, the Treasury Department permits these amounts to be totaled and the resulting total amount taken from any one or more of the plan participant's IRAs. See Notice 88-38, 1988-1 C.B. 524.
FOOTNOTE136. This proposed change raises a potential problem regarding plan participants who, during the course of a year, roll over their retirement accounts into new retirement accounts with a different plan administrator. In the case of a direct rollover from one plan to another, the plan administrator of the transferring plan, prior to the transfer, would have the responsibility to distribute the minimum distribution amount to the plan participant. If the plan participant receives a distribution directly, the plan would provide a form specifying how much could be rolled over and how much was a minimum distribution that could not be rolled over. The plan administrator of the recipient plan would accept a rollover only if accompanied by the form. The recipient plan would, of course, be required to make a minimum distribution with respect to the rolled over amounts in the following year, based on their value as of December 31 of the year of the rollover.
137. Note that paying the tax presents little difficulty to the beneficiaries. Plan accounts are highly liquid, consisting overwhelmingly of stocks, bonds, bank accounts, and mutual fund shares. (Some qualified plans, particularly plans of large employers, will invest in less liquid assets, such as real estate, but the plan administrators have a fiduciary duty to make sure that the plan has sufficient liquidity to meet the need for distributions to beneficiaries.) The taxes generated by the distribution will simply be paid out of the distribution itself. The heirs may complain about the "outrageous" amounts the government takes, but this complaint is no different from those made by anyone else who receives a lot of income and has to pay tax on it. Also, it should be remembered that the funds in the account should really have been taxed long ago, but, in effect, the government has made an interest-free loan to the participant in the amount
FOOTNOTEof the tax that should have been paid. At some point, it is appropriate to repay that loan. What better time than after the participant's (or spouse's) death? From a moral perspective, any claim by the beneficiaries for a continuation of the tax subsidy is fairly weak. They, unlike the participant, have done nothing to earn these funds; they simply had the luck to be born to parents who could accumulate wealth. Why should others pay higher taxes so that these lucky ones can have a larger windfall? It also bears repeating that the benefits of tax-favored retirement saving are already skewed to high-income earners.
138. See LRC. 402(d)(4)(A). Congress repealed the five-year averaging rules for lumpsum distributions made after Dec. 31, 1999. See Small Business Job Protection Act of 1996, Pub. L. No. 104-188, 1401, 110 Stat. 1755 (1996).
FOOTNOTE139. See supra note 53 and accompanying text.
FOOTNOTE140. See David A. Pratt & Dianne Bennett, Simplifying Retirement Plan Distributions, 57 N.Y.U. INST. ON FED. TAX I 5.06(1](b] ("Any benefits remaining at the death of the owner or surviving spouse must be distributed, in full, by the due date for filing the federal estate tax return, regardless of whether a return is actually required.").
141. The tax deferral incentive of this proposal would probably lead plan participants to hold their retirement assets in qualified plan solution longer. Because assets held at the death of plan participants are includible in the plan participant's estate, see LRC. 2039(a), the revenue cost of instituting this proposal may be less then anticipated due to the likely increase in estate tax revenue generated.
142. See supra Section IV.B.
FOOTNOTE143. See Tax Reform Act of 1984, Pub. L. No. 98-369, 521(d)(2)-(5), 98 Stat. 494, 868 (1984).
144. At least one commentator has suggested that the minimum distribution rules be reformed or repealed. See Mark J. Warshawsky, Minimum Distribution Requirements: Reform or Remove Them, 82 TAx NOTES 1133 (1998). He, however, offers few constructive suggestions to make the minimum distribution rules more administrable or equitable.
FOOTNOTE145. Indeed, a new bill in Congress would exacerbate the dysfunctional nature of the minimum distribution rules. The bill, entitled "Financial Freedom Act of 1999" proposes that the minimum distribution rules be further relaxed; it would essentially eliminate the "at least as rapidly rule" under LRC. 401(a)(9)(B)(i).
146. The proposal to replace the minimum distribution rule with a distribution at death mandate would be scored as a revenue loser, even though in the long run it raises revenue. (This is because the current revenue scoring system used in the budget process looks forward a maximum of ten years.) Given the present budget surplus, the revenue scoring process might actually work in favor of the second proposal since Congress could appear to be cutting taxes when in fact in the long run it was raising them. This is the precise opposite of the scoring of Roth IRAs, which in the short run increase revenue (since contributions are not deductible, as is the case with traditional IRAs) but in the long run their utilization by taxpayers results in tremendous revenue losses.
AUTHOR_AFFILIATIONJay A. Soled and Bruce A. Wolk"