FLORIDA TAX REVIEW VOLUME 2 1994 NUMBER 2 Simplification and IRC § 415 Norman P. Steitz I. INTRODUCTION There has been much talk of late about simplifying Subchapter D, the web of rules governing the affairs of tax-qualified pension plans.' Among the targets that the simplification advocates have in their sights is section 415 of the Code.2 Section 415 places limits on contributions to defined contribution * Douglas Arant Professor of Law, University of Alabama School of Law. The author gratefully acknowledges the helpful comments of Don Grubbs, Dan Halperin. Mary Oppenheimer, Mary Radford, Bruce Wolk, and members of the ABA Subcommittee on Section 415, despite, in many cases, their disagreement with the author's conclusions. The University of Alabama Law School Foundation generously provided funding for the article. Finally, this article is dedicated to Don Grubbs, who will retire later this year. and whose career and devotion to public service have provided an exemplary model for me and other members of the employee benefits bar. 1. See, e.g., H.R. 2742, 102d Cong., 1st Sess. (1991); Pension Access and Simplifi- cation Act of 1991, H.R. 2730, 102d Cong., 1st Sess. (1991); H.R. 2390, 102d Cong., 1st Sess. (1991); H.R. 1735, 102d Cong., 1st Sess. (1991); Employee Benefits Simplification Act. S. 2901, 101st Cong., 2d Sess. (1990); The Association of Private Pension & Welfare Plans, Gridlock: Pension Law in Crisis and the Road to Simplification 32-34 (1989) [hereinafter Gridlock]. 2. Simplification of the rules applicable to pension plans may have a range of purposes. These include making plans easier and less costly to administer, and easier for both the plan sponsors and plan participants to understand; easing the IRS's burden of monitoring compliance; reducing the opportunities for plan sponsors to evade legislative purpose through plan design; and improving the ability of professionals to predict accurately the consequences of particular rules to particular situations. The meaning of simplification in the tax law has been subject to extensive scholarly reflection and debate. See, e.g.. Boris I. Bittker, Tax Reform and Tax Simplification, 29 U. Miami L. Rev. 1 (1974); Walter J. Blum, Simplification of the Federal Income Tax Law, 10 Tax L. Rev. 239 (1954); Robert B. Eichholz, Should the Federal Income Tax Be Simplified?, 48 Yale LJ. 1200 (1939); James S. Eustice, Tax Complexity and the Tax Practitioner, 45 Tax L. Rev. 7 (1989): Randolph E. Paul. Simplifica- tion of Federal Tax Laws, 29 Cornell L.Q. 285 (1944); Stanley S. Surrey & Gerard M. Brannon, Simplification and Equity as Goals of Tax Policy, 9 Win. & Mary L Rev. 915 (1968); Laurence N. Woodworth, Tax Simplification and the Tax Reform Act of 1969, 34 Law & Contemp. Probs. 711 (1969). Florida Tax Review plans and on benefits payable from defined benefit plans.' The basic idea justifying section 415 is that an employer should not be able to use the Code's qualified plan provisions to bestow tax deferral on deferred compensa- tion in excess of the reasonable retirement needs of its employees.4 The simplification advocates have proposed repeal of section 415 or at least of its most supremely complex part, section 415(e), which sets combined limits for individuals who participate in both defined contribution and defined benefit plans. They argue that section 4980A, which was enacted in 1986 and which imposes a nondeductible excise tax on excessive distribu- tions from pension plans, has rendered section 415 limitations, or at least the section 415(e) limitation, superfluous.5 They also argue that subsection (e) is impervious to any sort of simplification efforts short of outright repeal.6 3. The limit on benefits payable from a defined benefit plan affects the amount of contributions paid to the plan, since the employer's annual contribution is limited to the appropriate annual actuarial cost of the maximum benefit permitted by § 415(b). For an excellent treatment of the technical operation of the § 415 rules, see Richard B. Stanger et al., Tax-Qualified Retirement Plans After TEFRA: Limitations on Benefits and Contributions and Top-Heavy Plan Rules, 41 Inst. on Fed. Tax'n § 37 (1983) [hereinafter Tax-Qualified Retirement Plans After TEFRA]. Although Congress has since modified § 415, this article remains the most comprehensive treatment of the rules. 4. See infra text accompanying notes 48-54; see also H.R. Rep. No. 807, 93d Cong., 2d Sess. 35 (1974); 2 Dep't of the Treasury, Tax Reform for Fairness, Simplicity, and Economic Growth: General Explanation of the Treasury Dep't Proposals 346, 350 (1984) [hereinafter Tax Reform]. 5. See Memorandum from Lee Irish to Elaine Church & Roger Siske 8 (Oct. 25, 1990) (on file with author) [hereinafter Irish Memo]; see also Dianne Bennett, Plan Distribu- tions: A Call for Some Order out of the Chaos, in ALI-ABA Fifth Pension Invitational Conference 33, 39-40 (Oct. 11-12, 1987) (on file with author). Bennett's paper provides an extraordinarily thoughtful approach to the web of rules governing distributions from qualified plans. Bennett, in suggesting that a bolstered version of § 4980A replace § 415, noted that the suggestion, even if rejected, might give rise to "consideration... of simplifications to section 415." Id. at 40. As is noted later, the proposals to eliminate § 415 are not nearly so serious as the proposals to eliminate subsection (e). The Association of Private Pension & Welfare Plans ("APPWP") has argued that § 415(e) should be eliminated if Congress retains the § 4980A tax. Gridlock, supra note 1, at 32-34. However, APPWP would prefer repeal of § 4980A to repeal of § 415(e). The concept for the § 4980A excise tax came from the Department of the Treasury. Tax Reform, supra note 4, at 349-53. The influential proposals led to the Tax Reform Act of 1986. The Department had proposed an excise tax as a substitute for § 415(e), which it stated "may be the primary source of complexity in the retirement plan area." Id. at 350. Congress, although adding the excise tax, retained the § 415(e) limits. 6. The case for elimination of § 415(e) is made articulately in a memorandum prepared by an attorney participating in the American Tax Policy Institute Pension Roundtable. See Irish Memo, supra note 5, at 8. The memorandum also suggests that § 415 itself might be eliminated. Id. [Vol 2:2 Simplification and IRC § 415 Because I agree with the basic idea undergirding section 415 and reject the argument that section 4980A serves adequately as a replacement, I disagree that section 415 is an attractive candidate for repeal. Moreover, in my view even a repeal limited to subsection 415(e) would be an error because of the subsection's importance in maintaining section 415's overall integrity. This article explains the basis for these views and also suggests ways of simplifying section 415(e) short of repealing it. The article is divided into three parts. The first part describes the pertinent provisions of section 415. The second part responds to the argument that section 415 should be eliminated rather than simplified; this part explores the policy goals served by section 415 and considers whether section 415 has anything to add to other Code sections-primarily section 4980A-that serve similar policy objectives. The third part suggests two approaches that Congress might take toward simplification of section 415. I. DESCRIPTION OF SECTION 415 Section 415 includes three limits. One limit (from section 415(b), hereinafter referred to as the "(b) limit" or the "defined benefit limit") restricts the maximum annual retirement benefit an employee can accrue under all defined benefit plans maintained at any time by an employer. The limitation is an annual straight single-life annuity equal to the lesser of $90,000 or 100% of the employee's average compensation over the highest three-year period.7 The dollar figure, however, is indexed to the cost of The American Bar Association Section of Taxation has prepared a report advocating several changes to simplify § 415, including repeal of § 415(e). American Bar Association. Section of Taxation, Possible Simplification Proposals Relating to Limitations on Benefits and Contributions Under Qualified Plans (Sept. 23, 1993) (on file at the University of Alabama School of Law Library) [hereinafter ABA Report]. The author of this article, who was a member of the subcommittee that drafted the report, dissented from the report's recommenda- tion that § 415(e) be deleted. Id. at I. 7. IRC § 415(b)(1). ERISA originally set the dollar limi" at S75,000. but cost-of-living adjustments pushed it up to S136,425 by 1982. I.R.S. Information Rcl. IR-8"2-18 (1982). Congress reduced the limitation to S90,000 in the Tax Equity and Fiscal Responsibility Act of 1982 ("TEFRA"), Pub. L. No. 97-248, § 235(g)(2). 96 Stat. 324, 508 (1982) but generally preserved already-accrued benefits in excess of the reduced limits. § 235(g)(4), 96 Stat. at 508-09. See generally Tax-Qualified Retirement Plans After TEFRA. supra note 3, § 37.03[6]. Congressional lore, as retold by former tax-writing committee staff members, has it that there had been intense lobbying in 1982 from business interests to leave the dollar limits intact, since the limits were an integral part of ERISA, a statute that was then only eight years old. Reducing the limits, they argued, would have an adverse effect on plan formation and also would be unfair to the businesses that had continued plan sponsorship since ERISA's passage in reliance on the limits being indexed. The lobbying efforts appeared to be successful and a majority of members of the House Ways and Means Committee was ready to vote against 19941 Florida Tax Review living and in 1994 is $118,881.8 The dollar limit is adjusted if the benefit begins earlier or later than social security retirement age, or is paid in a form other than a straight-life annuity.9 In addition, the dollar limit is phased in ratably over ten years of actual plan participation."° The compensation limit is phased in over ten years of service with the employer rather than plan participation. A second limit (under section 415(c), hereinafter referred to as the "(c) limit" or the "defined contribution limit") applies to defined contribution plans. The defined contribution limit provides that the annual additions to an employee's accounts in all defined contribution plans maintained by an employer may not exceed the lesser of $30,000 a year or 25% of the reductions to the dollar limits. Before the vote, however, the roll call bell rang and the committee members filed out to the floor of the House, where they participated in a vote to reduce food stamps. The irony of voting to cut food stamps while preserving a $136,425 dollar limit for benefits of the nation's most affluent citizens was apparently too much for a majority of the committee members, who returned to their committee seats and voted to reduce the § 415 limits. 8. I.R.S. Information Rel. IR-94-3 (Jan. 13, 1994). TEFRA delayed indexation until 1986, with indexation reflecting cost-of-living increases occurring after 1984. TEFRA, supra note 7, § 235(g)(1), (2), 96 Stat. at 508. 9. IRC § 415(b)(2)(A)-(D). See Tax-Qualified Retirement Plans After TEFRA, supra note 3, § 37.03 for details on the rules and their operation. 10. IRC § 415(b)(5). The § 415(b)(5) phase-in period for the dollar limit formerly was based on years of service with the employer rather than years of participation in the plan. The Department of the Treasury believed that phasing in the dollar limits based on service permitted a small employer to establish a defined benefit plan close to the time its key employee would retire. Because the benefit could be based on past service with the employer, the key employee generally would have service sufficient for "a fully funded benefit [but the plan could] ... avoid providing benefits to non-key employees," who generally would have less past service credit. Tax Reform, supra note 4, at 351. Thus, the Department proposed, and Congress accepted, a phase-in period for the dollar limit on the basis of years of plan participation rather than service. Section 415(b)(5)(D) provides that to the extent provided in regulations, the phase-in period applies to changes in benefit structure. The Service initially took the position that the phase-in of the dollar limits would apply to a plan amendment improving benefits. Thus, an amendment to a plan could improve benefits by no more than 10% of the dollar limit per year of participation after the plan amendment. I.R.S. Notice 89-45, 1989-1 C.B. 684. The Service later reconsidered its position in light of the regulations it promulgated under § 401(a)(4), which prohibit discrimination in favor of highly compensated employees. See Regs. § 1.401(a)(4)-l(b). In Revenue Procedure 92-42, the Service noted that the legislative history of § 415(b)(5) indicated that Congress was not concerned with phasing in benefit improvements that were not primarily for the benefit of highly compensated employees. The Service therefore concluded that the § 401(a)(4) nondiscrimination regulations eliminated the need for the phase-in period to apply to "changes in benefit structure," since those regulations prohibit amendments that discriminate in favor of highly compensated employees. Rev. Proc. 92-42, 1992-1 C.B. 872. [Vol 2:2 Simplification and IRC § 415 employee's compensation." For these purposes, additions include employer contributions, employee contributions, 2 and forfeitures allocated to the employee's account.' 3 The $30,000 limit is indexed to inflation, but the indexing will not commence until the inflation-adjusted defined benefit limit reaches $120,000, which will likely occur in 1995.4 There is an important conceptual distinction between the (b) and (c) limits. The (b) limit restricts the total career benefit an employee can accrue in defined benefit plans maintained by an employer; the (c) limit restricts the annual additions to a defined contribution plan. A third limit (from section 415(e), hereinafter referred to as the "(e) limit" or the "combined limit") applies when employers maintain both a defined contribution plan and a defined benefit plan, whether simultaneously or seriatim. The limit, which is quite complex, is designed to permit an employer who sponsors or has sponsored both types of plans to provide greater benefits to its employees than can be provided by an employer who sponsors only one type of plan. The increased benefits under the combined limit are, however, less than those determined by combining the full limits applicable to each type of plan. The combined limit's complexity reflects the difficulty of combining a career defined benefit limit with an annual defined contribution limit. The combined limit requires the preparation of both a "defined contribution fraction" and a "defined benefit fraction" indicating the percent- 11. IRC § 415(c)(1). ERISA originally set the limit at S25,000, with indexation to the cost of living. By 1982, the limit reached $45,475, but Congress cut back the limit to $30,000 in 1982. TEFRA, supra note 7. § 235(a)(2), (g)(1), 96 Stat. at 505, 508. The defined contribution limit permits the funding of a larger benefit over an employee's career than does the defined benefit limit. See infra note 52 and accompanying text. The two limits, although far from mathematically equivalent, apparently were regarded in Congress as political equivalents. 12. IRC § 415(c)(2)(A), (B). One of the minor complexities in § 415 is that the Tax Reform Act of 1986 expanded the definition of included employee contributions. Before the Act, employee contributions for § 415(c) purposes were limited to the lesser of the amount of the contributions in excess of 6% of compensation or one-half of the contributions. IRC § 415(c)(2)(B) (1985). The current version of § 415 includes all employee contributions. Thus. in determining employee contributions, a plan administrator has to cope with different sets of rules depending on the year in question. 13. IRC § 415(c)(2)(C). 14. The statutory mechanism for indexing the defined contribution limit is § 415(c)(1)(A), which provides that the dollar limit is "S30,000 (or. if greater. 1/4 of the dollar limitation in effect under subsection (b)(l)(A))." The defined benefit dollar limit, in turn. is indexed to the cost of living. IRC § 415(d)(1). The original dollar limit for defined benefit plans was set at $90,000. IRC § 415(b)(l)(A). The indexation of the defined contribution limit is deferred, however, until the defined benefit limit is adjusted under § 415(d) to S120,000. 1994] Florida Tax Review age of utilization of each type of limit.'" The sum of the two fractions must not exceed one.' 6 This description suggests that there would be no advantage to sponsoring two types of plans. The manner in which the fractions are calculated and the effect of the time value of money on the funding of defined benefit plans, however, permit employers who sponsor two types of plans (either simultaneously or seriatim) to provide greater total benefits than can be provided by employers who sponsor only one type of plan. Both the calculation of the combined limit and the implications of the time value of money are discussed below. The numerator of the defined benefit fraction is the amount of the projected annual benefit the employee has accrued under the plan. 7 The denominator of the fraction is the lesser of 1.25 times the dollar limit in effect for the year or 1.4 times the compensation limit applicable to the employee for the year.'8 This means that for an employee who has accrued the maximum defined benefit under the (b) limit, i.e., $118,881 in 1984, the defined benefit fraction is $118,881/(1.25 x 118,881), or 1/1.25, or 0.8. Thus, even though the employee has accrued the maximum defined benefit under section 415(b), he or she also may have contributions made to a defined contribution plan so long as the defined contribution fraction does not exceed 0.2. The defined contribution fraction is more complex because rather than reflect a defined benefit as currently calculated, it reflects the cumulative use of the (c) limit over the employee's service for the employer. The numerator is the sum of all applicable additions to the employee's accounts in all defined contribution plans ever maintained by the employer.' 9 The denomi- nator of the fraction is the sum of variables calculated for each year of ser- vice.2" The variable for each year is the lesser of 1.25 times the dollar (c) limit or 1.4 times the compensation (c) limit for defined contribution plans in effect for the applicable year.2' 15. IRC § 415(e)(l)-(3). 16. IRC § 415(e)(1). 17. IRC § 415(e)(2)(A). 18. IRC § 415(e)(2)(B). 19. IRC § 415(e)(3)(A). 20. IRC § 415(e)(3)(B). 21. Id. A source of needless complexity with respect to § 415(e) is that the definition of compensation under § 415(c)(3) often differs from the definition of compensation actually used by the employer to determine contributions under the plan. Some employers find that a quite burdensome part of § 415(e) computations is determining § 415(c)(3) compensa- tion over the employee's career, which might span 40 or more years. Records of such compensation may not be in plan records, and W-2 information, when it can be unearthed, is not useful if the plan year is other than the calendar year. [VoL 2:2 Simplification and IRC § 415 To illustrate, consider an employee with two years of service. In the first year, the employee's compensation was $64,285.72, and no additions were made to a defined contribution plan for the employee; in the second year, compensation was $125,000, and $30,000 in additions were allocated to the employee's account. In the first year, the compensation limit-25% of compensation ($16,071.43)-is applicable; in the second year, the dollar limit is applicable. Thus, the denominator is ($16,071.43 x 1.4) + (30,000 x 1.25). The numerator of the fraction is 0 + $30,000. The fraction, then, is as follows: 0 + 30.000 (16,071.43 x 1.4) + (30,000 x 1.25) This reduces to 30,000/60,000), or 0.5. If an employee subject to the dollar limits in each year received the maximum additions under the (c) limitation for each year of service, the defined contribution fraction, like the defined benefit fraction for an employee who has accrued the maximum benefit permitted under the (b) limitation, would be 1/1.25, or 0.8. The result of the above calculations is that an employee can com- pletely utilize either the (b) or (c) limit and still have a benefit under plans subject to the other limit. There is one important exception to the above rules: certain plans must substitute a 1 for the 1.25 factor applied to the dollar limit.2- The effect of this is that participants subject to the dollar limit may use only a single maximum limit. Thus, for example, an employee whose defined benefit equaled the dollar limit could not receive any contributions under a defined contribution plan. The exception applies to two types of "top-heavy plans." A top-heavy plan is essentially a plan in which more than 60% of the total benefits are for key employees. 23 The types of top-heavy plans covered by the exception are plans in which more than 90% of total benefits are for key employees and other top-heavy plans that do not satisfy certain optional minimum benefit requirements.24 22. IRC § 416(h). 23. IRC § 416(g)(1). 24. IRC § 416(h)(2). Employers must contribute at least 3% of compensation (or if less, the percentage of compensation contributed for the key employee for whom the percentage is greatest) annually for each non-key employee. IRC § 416(h)(2)(A). Top-heavy defined benefit plans must provide a retirement annuity equal to at least 2% times years of service (up to 10) times average compensation. IRC § 416(c). Section 415(e can remain at 1.25, however, only if 4% of compensation is contributed to a defined contribution plan, or 19941 Florida Tax Review The time value of money affects the amount an employer contributes to a defined benefit plan: the longer the period between contribution and benefit distribution, the smaller the contribution to the plan because the contribution will have more time to produce investment return. To illustrate, assume an employer will make a contribution to a pension plan sufficient to pay the employee a $1,000 lump sum benefit when the employee attains age sixty-five. Assuming an annual 8% rate of return on investment, the employer would have to contribute a single payment of $46 to fund the benefit for a twenty-five-year-old employee or $681 to fund the benefit for a sixty-year-old employee.25 This simply reflects the fact that the contribution for the twenty- five-year-old will produce investment return for forty years, while the contribution for the sixty-year-old will produce a return for only five years. This has important consequences to small employers who wish to benefit a principal employee while sponsoring only one plan at a time.26 When the employee is young, the employer will sponsor a defined contribu- tion plan, for the $30,000 maximum addition to such plan will exceed the maximum contribution that can be made to fund the maximum defined benefit for such employee. However, a switchover point will be reached, where the defined benefit contribution for the employee will exceed the $30,000 maximum defined contribution addition.2 7 Depending on circum- a benefit of 3% times years of service times average compensation is provided by a defined benefit plan. IRC § 416(h)(2)(A). 25. The contributions in the text do not reflect a discount for preretirement mortality. Such a discount would somewhat reduce each of the contributions. The percentage decrease of the contribution on behalf of the younger employee would be slightly greater because actuarialy a 25 year-old is expected to die at an earlier age (82.0) than a 60 year-old (84.2). See Regs. § 1.72-9, tbl. V. 26. The costs of defined benefit plans, plus the § 415(e) limits, especially as applicable to certain top-heavy plans, constrain many small employers from adopting defined benefit plans. If the § 415(e) limits were repealed, small employer sponsorship of defined benefit plans almost certainly would increase. 27. See, e.g., Stanley N. Bergman & David L. Reynolds, Plan Selection-Pension and Profit-Sharing Plans, 350 Tax Mgmt. (BNA) 10 (indicating that larger annual contributions can be made to defined benefit plans than to defined contribution plans for employees near retirement age); Robert E. Madden, Tax Planning for Highly Compensated Individuals 5.02[2] (1983) ("[Ain employer who is contemplating the implementation of a qualified plan may favor a defined benefit pension plan over other options that are available if that employer has older employees it particularly wishes to benefit."); American Bar Association, Senior Lawyers Division, The Lawyers Guide to Retirement 140 (David A. Bridewell ed., 1991) [hereinafter Lawyers Guide to Retirement]. [Vol. 2:2 Simnplification and IRC § 415 stances, this point will generally occur sometime between an employee's early forty's and early fifty's.' For example, assume a fifty-year-old employee, with a defined contribution fraction of 0.4. The defined benefit fraction for the employee is 0.6, which translates into a defined benefit limitation equal to 75% of the defined benefit dollar limitation.29 Assume that the limit is $ 12 0 ,0 00O." Thus, the plan could provide, and the employer could fund, a $90,000 defined benefit, unless the plan were subject to the more restrictive top-heavy limitations."1 The amount of the annual contribution required to fund this benefit will depend on the actuarial cost method and assumptions employed by the plan's actuary.32 The employer probably will contribute between $35,000 28. The literature does not indicate the age at which a contribution to a defined benefit will exceed the maximum contribution to a defined contribution plan, and indeed, the age will vary depending on a number of factors. See infra text accompanying notes 29-35. The immediately following textual example involves an employee who is age 50. In preparing this article, the author spoke with six pension attorneys, who suggested that the age at which defined benefit plans permit larger contributions than defined contribu- tion plans occurs somewhere between 41 and 51. Of course, the actual point at which contribu- tions to a defined benefit plan, with respect to a particular participant may exceed contributions to a defined contribution plan, will depend on the assumptions the plan's actuary makes con- cerning interest, mortality, assumed retirement age, the history of previous additions to defined contribution plans, and the actuarial method used by the plan to assign benefit costs to each year in which the plan is being funded. See generally I Gary I. Boren. Qualified Deferred Compensation Plans ch. 8 (1993) (discussing the elements of various actuarial methods). 29. A defined benefit fraction with a numerator of .75 and a denominator of 1.25 is .6. 30. The $120,000 dollar limit was selected because indexation of the defined contribution begins when the (b) limit reaches S120,000. See supra note 14. 31. See supra text accompanying notes 23-24. In the example, if the plan were super top-heavy, the defined contribution fraction would have been .5 rather than .4 (i.e., .5/1.25=.4). and the defined benefit fraction also would have been .5. Thus, the maximum defined benefit that could be funded would have been $60,000 if the plan had been super top-heavy. 32. See Jerome Mirza & Assocs. v. United States, 882 F.2d 229 (7th Cir. 1989). In Jerome Mirza & Associates, the employer's contribution was based on an interest rate assumption of 5% and an actuarial method that allocated costs related to past service entirely to the year of plan adoption. The Service successfully argued that the 5% interest rate assumption was unreasonable and that the costs attributable to past years should have been allocated to those years and then amortized over a ten-year period. The court adopted the Service's 8% interest rate assumption. The plan itself had invested in government securities that yielded between 11.65% and 15.75%. The court reduced the taxpayer's original deduction of $625,925 to $115,953. Id. at 230-31. However, more recently, the Service was unable to persuade the Tax Court that a 5% interest rate or an assumed retirement age of 55 was unreasonable. Citrus Valley Estates, Inc. v. Commissioner. 99 T.C. 379 (1992): Vinson & Elkins v. Commissioner, 99 T.C. 9 (1992). aff'd. 7 F.3d 1235 (5th Cir. 1993); Wachtell. Lipton, Rosen & Katz v. Commissioner, 64 T.C. Memo 1992-392 (CCHI) 1992. 19941 Florida Tax Review and $85,000, 33 depending on how aggressive it wishes to be (i.e., how much risk of dispute with the Service it wishes to assume). 34 Moreover, the amount of the permissible benefit for the employee will increase each year because the employee's defined contribution fraction will decline to reflect the fact that no further contributions are being made to the defined contribu- tion plan. An employer whose sole motive was to maximize tax deferral, however, would sponsor defined contribution and defined benefit plans simultaneously rather than seriatim. Simultaneous funding would permit full utilization of the defined contribution fraction and permit earlier funding of 33. The low-end figure assumes retirement at age 65, an interest rate of 8%, and a life expectancy of 85. Funding at a level-dollar amount over 15 years would yield a contribution of $36,491.30. The high-end figure assumes a retirement age of 60, a 6% interest rate, and a life expectancy of 90. These assumptions would yield a first year contribution of $86,244.90. The latter assumptions are less conservative than the assumption challenged in Jerome Mirza & Associates, or in the Tax Court cases cited supra note 32. It should be noted that the size of the contribution depends not only on the plan's actuarial assumptions, but also its actuarial method, which assigns portions of the cost of benefits to each year in which the benefits are funded. See Boren, supra note 28, ch. 8. 34. The Service has engaged in an audit of small defined benefit pension plans. The American Society of Pension Actuaries has opposed this program vigorously. See, e.g., American Society of Pension Actuaries, IRS Small Plan Actuarial Audit Program (1990); Ellin Rosenthal & Herman Ayayo, PBGC Told to Wait in Line; Small Plan Actuarial Audit Program Feud Continues, 47 Tax Notes 1157 (June 4, 1990); Ellin Rosenthal, IRS v. Actuaries: The Feud Continues over Small Plan Audits, 47 Tax Notes 140 (Apr. 9, 1990). The Service apparently routinely challenged interest rate assumptions lower than 8% and age-65 minimum retirement ages. See Rosenthal, supra, at 140. The positions that the Service took in the audit program were rejected by the Tax Court in a trio of 1992 decisions. See supra note 32. A relatively recent disincentive to some small employers making aggressively large contributions to a defined benefit plan is the excise tax that § 4980 imposes on surplus assets reverting to the employer when an overfunded pension plan terminates. The excise tax is 50%, unless the employer establishes a "qualified replacement plan" or shares the plan surplus with the employees, in which case the tax is reduced to 20%. IRC § 4980(d). Assume a small business will come to a close when a principal employee reaches retirement age and that the business will then have an overfunded pension plan. When the plan terminates, the residual assets will be taxed at punitive rates under § 4980(d) unless they are partially used to provide additional benefits to employees. The principal employee, however, will not be able to share in the additional benefits if already at the § 415 limits. Thus, the employer will either have to pay the tax or give 20% of the surplus to employees other than the principal. This scenario will discourage some small employers from funding their plans aggressively. There may, however, be ways to plan around this problem to some extent; for example, a plan might provide for a lump sum distribution option, with a relatively low interest rate assumption. See IRC § 415(b)(2)(E)(i). 35. In effect, a zero is added to the numerator in each year, to reflect a zero contribution, while 1.25 times the (c) dollar limit is added to the numerator. Forfeitures added to the employee's account, however, are also treated as additions. IRC § 415(c)(2). Thus, any forfeitures allocated to an employee's account would increase the numerator. [Vol 2:2 Simplification and IRC § 415 a portion of the same defined benefit that can be funded at a later stage.36 Earlier plan funding is less costly than later plan funding because earlier funding enjoys the benefits of the tax-deferred plan funding vehicle for a longer period of time. However, there are reasons why small employers might prefer not sponsoring defined benefit plans while the favored employee is young. The most evident reason is that defined benefit plans are relatively expensive to administer and may not be justified until the sponsor can make substantial contributions on behalf of the favored employee.3 ' Additionally, given the relative illiquidity of pension wealth prior to the time an employee retires, not all younger employees wish to maximize contributions to qualified plans." Thus, many small employers will sponsor only a defined contribution plan when the favored principal is young and later will switch to only a defined benefit plan. This strategy has an additional benefit to small employ- ers that wish to minimize the cost of providing benefits for employees other than the principal-provided such other employees are younger than the principal during the defined benefit sponsorship. There are three reasons why defined benefit plans can minimize costs attributable to young employees. The first reason is the time value of money's effect on contributions. Section 401(a)(4) of the Code, which proscribes discrimination in favor of highly compensated employees, generally permits an employer to fund a 36. Section 404j), however, does not permit a deduction for a contribution to a defined benefit plan to the extent that the benefit being funded exceeds the benefit permissible under § 415. IRC § 4040). This prevents an employer from anticipating inflation-adjusted increases in the § 415 limits for purposes of plan funding. Thus, maintaining two plans would permit early funding of only a portion of the complete benefit that ultimately can be funded under a defined benefit plan. 37. Defined benefit plans must utilize actuarial services, which can add costs. In addition, many defined benefit plans must pay premiums to Pension Benefit Guarantee Corporation ("PBGC") for each participant. There are important exceptions to PBGC coverage. including an exception for professional service employer plans which never have had more than 25 active participants. ERISA § 4021(b)(13) (codified at 29 U.S.C. § 1321(b)(13)). Finally, because defined benefit plans are considerably more complex than defined contribution plans, associated legal and accounting costs also may be higher. See, e.g., Madden. supra note 27, 5.02[2]. 38. Section 72(t) imposes a 10% excise tax on most distributions from a qualified plan to a participant younger than age 59.5. IRC § 72(t). In addition, the Service has long taken the position that a pension plan (as opposed to a profit-sharing plan) may not make distributions to an employee prior to the employee's separation from service, death, or disability. Rev. Rul. 56-693, 1956-2 C.B. 282. To some extent, this can be mitigated by an employee's access to plan loans, but plan loans are subject to many restrictions and are not available to owner-employees who sponsor Keogh plans or to shareholder-employees in Subchapter S plans. See generally Boren, supra note 28, § 11:18. In addition, loan programs do not work well in defined benefit plans because plan assets are not allocated to individual participants. 19941 Florida Tax Review defined retirement benefit for each employee equal to a uniform percentage of the employee's compensation. 39 The cost of funding this benefit is lower for younger employees than for older employees because the period between contribution and benefit payment is longer. What this means is that the employer will contribute a smaller percentage of pay to fund the benefits of younger employees than for older employees.4" In a garden variety defined contribution plan, however, this will not be the case, for the employer generally will contribute the same percentage of each employee's compensa- tion to the plan. The second reason defined benefit plans are more attractive than defined contribution plans when the goal is to favor older employees is the ability to combine a flat benefit formula with the fractional rule of accrual.42 The fractional rule permits a plan to define a benefit for all employees and have the benefit accrue over the remaining period of an employee's service with the employer. Thus, for example, a plan might promise a benefit of 30% of compensation for all employees, to accrue over each employee's period of projected future service. A fifty-five-year-old doctor would (but for antidiscrimination regulations noted below) thus accrue 1/10 of the benefit in a plan with normal retirement age of sixty-five, or a benefit accrual of 3% of compensation, each year, while the twenty-five-year-old nurse would accrue 1/40 of the benefit, or .75% of compensation, each year. When the doctor 39. A defined benefit plan can be integrated with social security, such that the benefit for lower-compensated employees will be a smaller percentage of their compensation than for higher-compensated employees. IRC § 401(1). See generally Nancy J. Altman, Rethinking Retirement Income Policies: Nondiscrimination, Integration, and the Quest for Worker Security, 42 Tax L. Rev. 433 (1987). 40. This does not, however, mean that the ultimate benefit being paid to the younger employee will be a smaller percentage of pay, but only that the cost to the employer of funding the benefit is less. See supra note 25 and accompanying text for an illustration. Assume further that each employee in the example eams $10,000; thus, each of them accrues a benefit equal to 10% of pay. However, the employer would contribute .46% of the younger employee's compensation, compared to 6.81% of the older employee's compensation. See generally Lawyers Guide to Retirement, supra note 27, at 140 (noting that "[blecause the contribution formula considers the number of years remaining to retirement for all participants, a defined benefit plan works best when your employees are somewhat younger than you"). 41. The defined contribution plan, like a defined benefit plan, can be integrated with social security, resulting in larger contributions for employees whose compensation exceeds the plan's integration level. IRC § 401(l)(2)(a); see supra note 39. Also, an employer, at the cost of added complexity, can sponsor a defined contribution plan that permits contributions that are larger as a percentage of pay for older employees. See Regs. § 1.401(a)(4)-8(b)(3) (describing safe harbor nondiscrimination rules for target benefit plans); Regs. § 1.401(a)(4)- 8(b)(1)-(2) (providing rules to cross-test defined contribution plans on the basis of benefits). However, contributions to such plans may run up against the defined contribution limitation of $30,000. IRC § 415(c). 42. IRC § 411(b)(1)(C). [Vol. 2:2 Siniplification and IRC § 415 retires in ten years, the doctor will have accrued a benefit equal to 30% of compensation while the nurse during this same period of time will have accrued a benefit equal to 71/2% of compensation.13 Relatively recent regula- tions under section 401(a)(4) put limits on this technique but do not eliminate it. 44 The third reason is that a defined benefit can be based in part on past-service credit.45 In many cases, the favored principal will have substan- tially more past service than other employees. Thus, the benefit being funded for the principal can be larger (as a percentage of compensation) than the benefits for employees without equivalent past service credit. Newly finalized regulations under section 401(a)(4) limit but do not eliminate an employer's ability to exploit the grant of past-service credits to favor a principal.6 This conflicts with a purpose of the tax subsidy for qualified plans, which is to encourage employers to establish plans that provide benefits to rank-and-file employees that are proportionate (as a percentage of compensa- tion) to the benefits provided to highly compensated employees. 7 43. It is probable that the plan will terminate when the doctor retires. Thus, the nurse will not have the opportunity to increase her benefit accruals to 30% of pay. 44. Regs. § 1.401(a)(4)-3(b)(2)(iv), (4). The regulations create a safe harbor in which a plan's fractional rule formula will be considered nondiscriminatory, and thus nonviolative of § 401(a)(4), if the denominator of the fraction is at least 25 for each employee. Regs. § 1.401(a)(4)-3(b)(4). 45. For example, an employer might adopt a plan on January 1. 1994, providing a benefit equal to 1% of final pay multiplied by years of service. The years of service might include all years that an employee worked for the employer, even those before 1994. Section 415(b)(5), however, does place some limit on this, since the defined benefit dollar is phased in over 10 years of plan participation. IRC § 415(b)(5). 46. The regulations create a safe harbor generally permitting a plan to award five years of past-service credit. Prop. Regs. § 1.401(a)(4)-5(a)(3). 47. The subsidy has been justified as an inducement to employers to set up plans by providing substantial tax benefits to upper-income employees who participate. The antidiscrimination rules of § 40 1(a)(4) and coverage rules of § 410 then require that benefits be provided to a substantial percentage of the non-highly compensated portion of the employer's workforce, and § 401(a)(4) requires that the benefits provided to rank-and-file employees be comparable to those provided to highly compensated employees. See generally Michael J. Graetz, The Troubled Marriage of Retirement Security and Tax Policies, 135 U. Pa. L. Rev. 851 (1987); Bruce Wolk, Discrimination Rules for Qualified Retirement Plans: Good Intentions Confront Economic Reality. 70 Va. L. Rev. 419 (1984). Both Professors Graetz and Wolk question the effectiveness of the qualified plan rules in directing benefits to rank-and-file employees. Taking a more extreme view, Professor Joseph Bankman argues that the nondiscrimination mechanism may harm the welfare of rank-and-file employees. Joseph Bankman, Tax Policy and Retirement Income: Are Pension Plan Anti-Discrimination Provisions Desirable?, 55 U. Chi. L. Rev. 790, 805-14 (1988). 19941 Florida Tax Review III. THE PURPOSE OF SECTION 415 AND THE ARGUMENTS FOR ITS ELIMINATION A. Purpose of Section 415 Prior to the enactment of ERISA, the Code placed no direct limita- tions on contributions or benefits for individual employees. 48 It was thus possible for an employer to make exceedingly large contributions or to fund exceedingly large benefits for an employee. Congress, in enacting ERISA, believed that the law's failure to place limits on contributions and benefits represented a shortcoming.49 Senator Long expressed this view in comments before the Senate: [Section 415] makes the tax laws regarding pension plans fairer by limiting the amount of the contributions or benefits that can be provided to any individual under such a plan. The fact that present law does not provide such specific limitations has made it possible for extremely large contribu- tions and benefits to be made under qualified plans for some highly paid individuals. While there is, of course, no objec- tion to large retirement benefits in themselves, it is not appropriate to finance extremely large benefits in part at public expense through the use of special tax treatment.50 The purpose of Section 415 can be understood as limiting contribu- tions and benefits under qualified plans to levels that are proportionate "to the reasonable needs of individuals for a dignified level of retirement income."' Benefits and contributions in excess of these levels are not eligible for the tax deferral available under a qualified plan. One can argue about how well section 415 actually effects this purpose. First, the limits under section 415 are generous, perhaps to a fault. Take the case of a twenty-year-old who receives the maximum contributions to a defined contribution plan for forty-five years. If we assume no inflation 48. See Tax-Qualified Retirement Plans After TEFRA, supra note 3, § 37.02. The pre-ERISA Internal Revenue Code, however, did include rules prohibiting discrimination in favor of shareholders, officers, and other highly compensated employees. See IRC § 401 (a)(4) (1973). In addition, § 162(a) requires that compensation be reasonable. IRC § 162(a)(1). This placed an indirect limit on deferred compensation. 49. See Tax-Qualified Retirement Plans After TEFRA, supra note 3, § 37.01. 50. 120 Cong. Rec. S. 29,946 (1974) (statement of Sen. Long). 51. H.R. Rep. No. 807, 93d Cong., 2d Sess. 112 (1974), reprinted in 1974 U.S.C.C.A.N. 4670, 4777. [Vol. 2:2 Simplification and IRC § 415 and that the plan realizes an annual rate of return of 4% the employee would accumulate a defined contribution account balance of approximately $3,630,000 by the time the employee reaches age sixty-five. 2 In addition, the employer could fund 25% of the maximum defined benefit permitted under section 415(b).13 It is difficult to regard this level of benefit as no more than reasonable. A second problem with the limits is that they apply on a per-employer basis. Thus, an employee who works for more than one employer theoretically can accumulate greater benefits than those just described.-4 Notwithstanding these criticisms, section 415 does at least put an outer limit on benefits and contributions. Moreover, there probably are not very many young people who are credited with $30,000 in annual additions to defined contribution plans or employees whose changes of employer significantly increase their total benefits because of section 415's separate applicability to each employer." Thus, even though section 415 does not work perfectly, it does serve the purpose of placing at least some limits on the amount of tax deferral available to participants in qualified plans. In so doing, section 415 controls the tax expenditures for qualified plans. B. The Argument for Eliminating Section 415 Those who argue that we should repeal section 415 generally do not take issue with its basic policy goals.5 6 Rather, their argument is that section 415's contribution to these goals does not justify the costs imposed by its intractable complexity.' In the next part of this article, however, it will be shown that section 415 is less impervious to simplification efforts than generally has been acknowledged. The argument that section 415 should be 52. As noted in the text, this benefit figure assumes no inflation and a true rate of return of 4%. Assuming an annual inflation rate of 4% (which would increase the (c) limits) and a rate of return on investments of 8%, the employee will accumulate an account balance in excess of $19,500,000 by the time the employee attains age 65. By way of contrast, the present value of a $120,000 retirement annuity at age 65 is approximately $1.6 million. assuming a 4% interest rate and a 20-year-life expectancy. 53. Sponsors of certain top-heavy plans, however, would not be able to provide the defined benefit. See supra notes 22-24 and accompanying text. 54. See Bruce Wolk, The New Excise and Estate Taxes on Excess Retirement Plan Distributions and Accumulations, 39 U. Fla. L. Rev. 987, 989 (1987). 55. Moreover, an employee who works for two related employers within the meaning of §§ 414(b), (c), (m), (n), or (o), is subject to a single, aggregate § 415 limit. See IRC § 415(g). 56. None of the sources cited supra note 5 argue against the basic policy goal of placing limits on the use of tax deferral provided by qualified plans. 57. See sources cited supra note 5. 1994] Florida Tax Review repealed is weakened considerably if it is correct that the section can be simplified in a manner that substantially reduces its administrative costs. This still leaves the question of whether section 415 has anything to add to other sections of the Code that also serve the purpose of limiting tax deferral. If section 415 serves little or no purpose, there is no reason to retain even a simplified version of it. Proponents of eliminating section 415 argue that the policy goals effected by section 415 can be served adequately by the section 4980A excise tax on excess distributions, which is less complex than the section 415 rules. 8 Section 4980A imposes a 15% excise tax on "excess distributions" from qualified plans.59 An excess distribution is the amount of distributions from all qualified plans, individual retirement accounts, and section 403(b) annuities in a given year to the extent that such distributions exceed a threshold amount.6° The threshold amount is currently $150,000, except for taxpayers who elected a grandfather provision.6' Example: A is an individual who in a single year receives a $90,000 annual annuity payment from the defined benefit plan of Employer X and a $40,000 payment from a defined contribution plan of the same employer. In addition, A receives another $50,000 annuity payment from a defined benefit plan of Employer Y and also withdraws $30,000 from an individual retirement account. In all, A receives $210,000 in distributions in plans whose distributions are subject to section 4980A in the year in question. The distributions exceed the threshold amount by $60,000. Thus, A would pay $9,000 in tax (15% x $60,000) under section 4980A.62 58. See generally sources cited supra note 5. Note that even those who argue that § 415 should be repealed do not seem to regard repeal as likely. Dianne Bennett, who makes the case against § 415 articulately, notes that consideration of her argument might lead to simplification of § 415 even if it does not lead to repeal. Bennett, supra note 5, at 40. 59. IRC § 4980A(a). For a discussion of the operation of the excise tax and opportunities to plan around it, see Dianne Bennett et al., Taxation of Distributions from Qualified Plans ch. 13 (1991). 60. IRC § 4980A(c)(1), (e). 61. IRC § 4980A(c)(1), (f). The grandfather provisions are found in § 4980A(f). Dianne Bennett proposed deleting the grandfather provisions from the Code. See Bennett, supra note 5, at 39. 62. Section 4980A includes special rules reducing the tax in a year in which the employee receives a lump sum distribution. IRC § 4980A(c)(4). The provision is elective and may be made only once. IRC §§ 4980A(c)(4), 402(d)(4)(B). [Vo61 2:2 Simplification and IRC § 415 An analogous tax provision increases the estate tax of decedents who die with excessive accumulations in plans whose distributions are subject to section 4980A. 63 In a nutshell, then, section 4980A is an additional tax on unreason- ably large benefit distributions from qualified plans.' In theory, the section recoups a portion of the tax deferral attributable to the excessive portion of the benefits. The effect of the section should be limiting benefit size, which is, of course, also the effect of section 415. But where section 415 operates directly by limiting what goes into a plan,6 5 section 4980A operates indirect- ly by imposing a tax on what comes out of the plan. Section 415 works as a prophylactic, section 4980A as a corrective. Section 4980A, however, would not be an adequate substitute for section 415.66 The primary problem with section 4980A is that it is a flat tax that does not take into account many of the variables that contribute to the extent of the tax deferral an employee will realize from participation in a qualified plan. 67 A 15% tax, for example, will do more than recover the tax savings embedded in assets that have been in the plan for a short time but recover only a small percentage of the tax savings reflected in assets that have been accumulating over an extended time period. Example: Assume a taxpayer with a 39.6% marginal tax rate and assume that both the taxpayer and the qualified plan will earn a 6%, pre-tax, annual rate of return on investments. The taxpayer is paid $1,000 in taxable compensation in 1991, yielding $604 after tax. The $604 will grow to $1,757 with a pre-tax 6% rate of return after thirty years. Assume now that instead of being paid $1,000 in taxable compensation, the taxpayer's employer contributed $1,000 to a qualified plan, which earns a 6% rate of return. The $1,000 will grow to $5,743 in thirty years. If this amount is then distributed to the taxpayer in a lump sum, the taxpayer will retain $2,608 after paying a 39.6% income tax and a 15% section 4980A tax on the distribution. This is approximately 148% of the amount that the taxpayer could have accumulat- ed outside the plan. 63. IRC § 4980A(d). 64. Tax Reform, supra note 4, at 350-53. 65. The (c) limit directly limits what goes into a defined contribution plan; the (b) limit, by providing a maximum benefit, indirectly limits the contributions that are made to a defined benefit plan. 66. See Wolk, supra note 54, for an excellent critique of the § 4980A excise tax. 67. Id. at 1022-24. 1994] Florida Tax Review The differential between the value of after-tax savings and the qualified plan increases as tax or interest rates increase.68 Assuming the applicability of the section 4980A tax, the point at which a contribution to a qualified plan becomes more valuable than immediately taxable compensation is approxi- mately twelve years, given the assumptions about interest and tax rates in the example.69 In other words, given these assumptions and all other things being equal, a taxpayer whose retirement distributions will be subject to the section 4980A excise tax should choose deferred compensation if the deferral period is greater than twelve years.7" Thus, section 4980A is not a very effective tax to limit accumulations in qualified plans during much of a taxpayer's career. Despite the existence of section 4980A, eliminating section 415 would give some taxpayers a reasonably free hand to decide how much compensation to defer on a tax-advantaged basis." One could argue that section 4980A could be restructured to target more precisely the embedded tax advantage in a qualified plan distribution, or that the section 4980A tax rate could be increased.72 A more precise section 4980A, however, would make section 415 appear a model of simplicity. An excise tax that perfectly compensates for previous tax deferral attributable to "excess distributions" would have to be based on the difference between (1) the plan's pre-tax return on all contributions multiplied by the taxpayer's marginal tax rate at the time of distribution, and (2) a theoretical after-tax rate of return that the taxpayer would have realized had she paid immediate tax on each contribution and invested it in a taxable investment format. Such calculations would require that the taxpayer identify the amount 68. Id. at 1020-21. The benefit of the qualified plan is enhanced in states with an income tax. 69. The figures were derived on a spreadsheet, which compared the after-tax accumulation of an after-tax payment to the employee in the hypothetical with the after-tax distribution of a contribution to a plan. The figures assume that the entire plan distribution was subject to a 39.6% marginal tax rate, plus the 15% excise tax. An earlier version of this article assumed a 31% marginal tax rate and an 8% return. These assumptions yielded an approximately 10-year break-even point. 70. Different assumptions would either lengthen or shorten the period. In addition, there are planning opportunities to reduce the effect of the excise tax. For example, § 4980A(c)(4) provides for special treatment of a lump sum distribution qualifying under § 402(d)(4)(B) and § 4980A(d)(5) for an effective marital deduction for an electing spouse who elects to treat post-mortem payments as a retirement distribution to the spouse. IRC § 4980A(c)(4), (d)(5). See Bennett, supra note 59, 13.4, .6[7][c][ii]; see also Wolk, supra note 54, at 996-97, 1006-10. Professor Wolk also suggests planning possibilities in community property states. Id. at 999-1002. 71. It is possible that some employers ignore the § 415 limits because of their complexity. Part of the issue, then, might be which is preferable: § 4980A with compliance or § 415 without? 72. See Bennett, supra note 5, at 38-39; Irish Memo, supra note 5, at 7-8. [Vol. 2:2 Simplification and IRC § 415 contributed to the plan in each year, the plan's rate of return for each year, the taxpayer's marginal tax rate for each year, and the taxpayer's theoretical rate of return on nonqualified plan investments for each year.'" No one could seriously suggest that such an approach is workable. The record- keeping requirements would be staggering and determining alternate rates of return virtually impossible without resort to across-the-board, rough-justice assumptions. What's worse still, the complexity would be visited at the individual taxpayer level. At least with section 415, the complexity occurs at the plan level, where it is likely to be less bedeviling.74 Increasing the section 4980A tax also is an unappealing idea. If the increase is high enough to discourage excessive early accumulations of assets, it will impose a substantial tax penalty on individuals whose pension wealth is attributable either to strong investment performance or late plan contribu- tions." Moreover, assuming annual returns on investment of 6% and an employee with a constant 39.6% marginal tax rate, the excise tax would have to be approximately 27.2% to prevent tax deferral advantages for benefits resulting from level contributions over a forty-year career.76 Finally, changes in income tax rates would virtually mandate complicated adjustments of the excise tax.77 73. See Wolk, supra note 54, at 1020-21. 74. Interestingly, the Department of the Treasury, when it proposed the excise tax as a replacement for § 415(e), argued that complexity at the participant level was preferable to complexity at the plan level because it did not impose an administrative burden on the plan. Tax Reform, supra note 4, at 352-53. 75. See Gridlock, supra note 1, at 33 (noting that the tax "acts as a penalty on investment success"); Bennett, supra note 5, at 39 (noting that "the excise tax penalizes good investment experience rather than simply addressing overaccumulation from contributions," but also that even "good investment experience [has] the advantage of tax-free build-up in a plan"). Professor Wolk, however, notes that the tax acts particularly unfairly in cases of unrealized appreciation. Wolk, supra note 54, at 1022. In such cases, the qualified plan offers no tax benefit since unrealized gain is not taxed outside the plan until disposition. Id. Subjecting what would be unrealized appreciation outside the plan to an excise tax when the appreciated assets are distributed is difficult to defend. 76. The 27.2% tax rate reflects a further assumption: that plan benefits were paid out in a lump sum at retirement age. For benefits paid out as annuities, the tax would have to be still higher to reflect the fact that tax deferral continues while assets are held by the plan. In an earlier draft of this article, which assumed an 8% rate of return and a 31% marginal income tax rate, the excise tax would have needed to increase to 33.3%. If the marginal tax rate were 31% and the rate of return were 6%, the excise tax would need to be 26%. 77. This is so because a change in tax rates would change the size of the tax subsidy, sometimes reducing it, sometimes increasing it. Moreover, changes in tax rates could affect people differently. Assume a taxpayer accumulated pension wealth while her marginal tax rate was 40%. When the taxpayer retires, Congress increases the tax rates such that the taxpayer's marginal tax rate is now 50%. One could argue that the excise tax should be 19941 Florida Tax Review Even if section 4980A were substantially increased or amended to precisely, or at least more precisely, target the tax deferral of "excess distributions," elimination of section 415 might still be a source of potential havoc for revenue collection over a transition period, because the revenue loss would occur annually as contributions were made to qualified plans while revenue gains would be delayed until benefits are distributed. (This assumes that current distributions subject to section 4980A are relatively small compared to the contributions that will ultimately result in excess distribu- tions in the future.) Moreover, it is possible that the day of revenue payback will never arrive for some taxpayers. Change in the tax law is as much a certainty as death and taxes themselves. Congress may in the future respond to pressures about the complexity and unfairness of the section 4980A tax by repealing section 4980A. Of course, it is also plausible that Congress may respond to pressures to raise more revenue by increasing the tax or lowering the taxability threshold. Thus, an argument for repeal of section 415 based on the similar policy goals of section 4980A is not persuasive.7" Despite this, a repeal of section 415 probably would not result in employees of large employers receiving significantly larger benefits from qualified plans than they receive today. The rules prohibiting discrimination against non-highly compensated employees would be a substantial economic disincentive to providing excessively large benefits for the highly paid because similar levels of benefits would have to be provided for all employees.79 Large employers lowered to reflect the fact that the distributions will be taxed at a 50% rate. However, an active employee whose marginal rate jumped from 40% to 50% would now realize greater deferral from qualified plan savings. One could thus also argue that the excise tax should be increased. Increasing the rate for some taxpayers, while reducing it for others, would breed complexity. 78. Without § 415, the only direct limit on contributions to a defined contribution plan would be the § 404 rule limiting contributions to a profit sharing or stock bonus plan to 15% of compensation and contributions to two or more plans to the greater of 25% of compensation or the minimum contribution to a defined benefit plan required under § 412. See infra note 81 for further discussion of the § 404 limits on deductions. Whether § 4980A should be retained as an adjunct to § 415 is a difficult question. Professor Wolk views the tax as fundamentally flawed. Wolk, supra note 54, at 1020-26. The APPWP has argued that the tax should be repealed. Gridlock, supra note 1, at 32-34. But see Bennett, supra note 5 (expressing the view that § 4980A does not present difficult issues). Given the fact that § 415 permits wealthy taxpayers to accumulate rather large tax-favored pools of money, one could argue that § 4980A, despite its problems, should be left in the Code. 79. See supra note 39 and accompanying text. [Vol. 2:2 Simplification and IRC § 415 normally have too many non-highly compensated employees to make such strategies appealing.'0 Repeal of section 415 would, however, provide tax planning opportu- nities for small businesses seeking to benefit principal employees. A vivid illustration of this is provided by a consideration of a successful self-em- ployed professional with no employees. Assume the professional earns $500,000 annually, which will be allocated between current compensation and qualified deferred compensation. Under current law, the professional could contribute $30,000 each year to a defined contribution plan. Were section 415 repealed, the taxpayer could contribute perhaps $100,000 or more to the plan in 1994, and perhaps could contribute as much compensation as the taxpayer desired, depending on how limitations on deductions are interpreted.8 1 Moreover, section 404 places no direct limit on the size of a defined benefit that can be funded with deductible contributions; without section 415, the only limit on such contributions would be the general restriction under section 162 limiting compensation to reasonable amounts. It might be argued that allowing a class of affluent taxpayers to defer large amounts of income is a small price to pay for banishing section 415 from the Code. But the revenue costs of effectively unfettering many profes- sionals and small business owners from the limits of section 415 are likely to be high, particularly if income tax rates continue to rise. Moreover, to allow even a small group of taxpayers to decide how much tax they should 80. Employers currently may provide for benefits in excess of § 415 limitations through nonqualified "excess benefit" plans. The floor debates on ERISA indicated an awareness of the difficulty that large employers would have in providing extraordinary benefits through qualified plans. See Subcomm. on Labor of the Senate Comm. on Labor and Pub. Welfare. 94th Cong.. 2d Sess., Legislative History of the Employee Retirement Income Security Act of 1974 1740 (Comm. Print 1976) (statement of Sen. Long noting that discrimination rules and threat of shareholder litigation would prevent loading up benefits for executives from qualified pension plans). 81. The sole constraint on this would be the limitations on deductions under § 404. While deductions are limited to 15% of compensation for contributions to a stock bonus or profit-sharing plan, and to the greater of 25% of compensation or the minimum contribution to a defined benefit plan under § 412 if the employer maintains a defined contribution and defined benefit plan, § 44(a)(7), there is no limit on deductions for an employer who maintains only a money-purchase pension plan. Of course, one could argue that were § 415 repealed, the 25% limitation should apply (or be legislatively amended to apply) to employers who maintain only money-purchase plans. In addition, § 401(a)(17) would limit the compensa- tion to $150,000 in 1994. If the 25% limit on deductions applied, the taxpayer's contribution would be limited to $37,500 which would be 25% of the S 150.000 of § 401(a)(17) compensa- tion. However, the argument that the 25% limit should apply is difficult to make. for two plans are not involved. Moreover, § 404(j), which reduces deductions by the amount that annual additions exceed the § 415 limits, suggests that the § 404 25% limit need not apply to single-plan situations. Thus, textual statements seem correct under the current version of § 404. 1994] Florida Tax Review pay in a particular year sends a symbolically appalling message to other taxpayers, which may be unwise for a system dependent on voluntary compli- ance. There is another potential disadvantage to repeal of section 415: a probable increase in the use of defined benefit plans, which will make it more likely that benefits for non-highly compensated employees will be reduced. We have already seen that defined benefit plans favor older employees, and that as a result some small employers wishing to benefit a particular older employee will switch from a defined contribution plan to a defined benefit plan. This not only maximizes contributions under section 415, but also minimizes the cost of providing benefits for their younger employees. Notwithstanding these facts, some such employers do not swap plan types because the size of the permissible benefit for the favored employee under section 415 may not be great enough to warrant the costs and bother of administering a defined benefit plan. Such costs, however, might be seen as trivial if section 415 ceased to limit benefits under defined benefit plans. Employers currently shunning such plans because of a cost/benefit analysis might well begin adopting such plans. As just noted, adoption of such plans sometimes will result in disproportionately low benefits for non-highly compensated employees, an outcome that is at odds with one of the purposes of the qualified-plan tax subsidy: providing approximately proportional benefits (as a percentage of compensation) for all participants. C. Consideration of the Arguments for Eliminating Section 415(e) Most of the complexity of section 415 is attributable to section 415(e), which provides the limitation for combinations of defined benefit and defined contribution plans. A case can thus be made that section 415 should be retained but that subsection (e) should be repealed.82 Repeal of section 415(e), however, might substantially reduce revenues. An employer would be free both to fund a maximum defined benefit and to make maximum contributions to a defined contribution plan. In all, this would permit some employees to accumulate by retirement age a defined benefit and defined contribution account with an aggregate value of approximately five million dollars, given current tax rates and relatively low interest assumptions.83 The value could be still higher depending on the 82. The Department of the Treasury initiated the idea of substituting an excise tax on distributions for § 415(e). See Tax Reform, supra note 4, at 352-53. 83. This assumes a 4% investment return and no inflation (after the dollar limit is adjusted to $120,000). The accumulation in the defined contribution plan would be approxi- mately $3,600,000, and the value of a $120,000 annuity would be approximately $1,600,000, assuming a life expectancy of 20 years at retirement. See supra note 52 and accompanying [ Vol 2:2 Simplification and IRC § 415 investment return in the defined contribution plan and on the actuarial assumptions used to value the defined benefit. A substantial portion of the benefit's value would reflect the value of the tax deferral extended to qualified plans.84 One possible means of reducing the revenue loss that would result from repealing section 415(e) would be a lowering of the (b) and (c) limits for employees of employers who maintain both types of plans. However, this approach, if it were to work, would reintroduce the very complexity it seeks to eliminate, for the lowered limits would have to apply not only to employ- ers who simultaneously sponsor defined benefit and defined contribution plans, but also to employers who sponsor only one type of plan for a number of years and then either switch to, or add, the other type of plan in later years. Rules to apply the reduced limits for employers who adopt plans seriatim would look very much like the current section 415(e) limitation. Another approach to reducing the revenue loss would be to lower the (b) and (c) limits for all plans, regardless of whether the employer sponsors more than one type of plan. There are serious policy problems with this approach. The reduction in the (b) and (c) limits might have to be substantial in order to control the tax benefits of sponsoring both plan types. But if the reduction were substantial, it would reduce the size of benefits that an employer who sponsored only one type of plan could provide for its employ- ees. Employers who wished to provide reasonably high levels of benefits thus would be forced to sponsor both types of plans, saddling some employers who would prefer sponsoring only a defined contribution plan with the added expense and complexity of defined benefit plan sponsorship."s There is a third possible approach for recovering the revenue loss, which merits consideration: reduce the defined contribution limits for younger employees. This might be done with respect only to the dollar limits, or with respect to both the dollar and compensation limits. For example, the dollar limit might be cut in half for employees under the age of thirty, and then gradually rise until the limit reaches its maximum level when an employee text. The amount of the benefit would be considerably higher if we assumed even moderate rates of inflation. However, as banks in advertising individual retirement accounts have learned, stating the benefit in terms of future nominal dollars tends to inflate the value of the tax deferral. The repeal would be valuable especially to participants in those top-heavy plans whose denominator for the dollar limit is I rather than 1.25. See supra text accompanying notes 22-24. 84. The number of employers who might adopt defined benefit plans because of the repeal of § 415(e) is of course a matter of conjecture. The complexity and cost of defined benefit plans would continue to deter at least some employers from sponsoring such plans. 85. Under current law, employers who want to maximize benefits must sponsor both types of plans, but high levels of benefits can be provided through either plan format alone. 19941 Florida Tax Review reaches age fifty-five. Such an approach might constructively contribute to tax and retirement policy. As already noted, if an employer annually contrib- utes $30,000 to an employee's account in a defined contribution plan beginning at age twenty-one, the total accumulation will be approximately $3,600,000 at age sixty-five, assuming an annual rate of return of 4%. This is an unusually large accumulation, worth approximately twice the present value of the maximum defined benefit at age sixty-five. A large portion of the accumulation is attributable to the tax deferral associated with the earliest contributions. By reducing the early contributions and the concomitant tax- deferred investment growth, the value of career-long participation in the defined contribution plan, would be brought more in line with the value of a maximum defined benefit. Arguably this is a desirable result in a world with or without section 415(e). Under this approach, the cost of repealing section 415(e) would be born by affluent employees who participate in defined contribution plans while they are young. There are some reasonable objections one might make to this approach. First, the approach goes beyond simplification: it creates a class of people that will pay for repeal of section 415(e) that is broader than the class that would benefit from the repeal.8 6 Second, it is by no means clear that the tax savings under this approach would equal the tax costs of repealing section 415(e). Third, some employers may stop sponsorship of defined contribution plans if the benefit they can provide to favored young employees is substan- tially reduced. 7 Section 415(e)'s repeal would create problems in addition to revenue loss. A repeal of section 415(e) might encourage more employers to adopt defined benefit plans when the employees it wishes to favor have reached mid-to-late middle age, resulting in diminished contributions on behalf of younger non-highly paid employees. Moreover, as we have seen, Congress has substantially tightened the section 415(e) limits for certain top-heavy 86. While this is true, note that under current law, a 30-year-old employee who receives an annual $30,000 contribution to a defined contribution plan and obtains a 4% rate of return, will accumulate a defined contribution account in excess of $3 million dollars by age 65. This large benefit also may be supplemented by a benefit from a defined benefit plan. Cutting the limits along the lines hypothesized still would produce a very large benefit. 87. This certainly would be true to some extent, but a considerable amount of conjecture would be required to determine to what extent plan sponsorship actually would be affected. Many small employers-who are the ones most likely to stop plan sponsorship-have already substituted § 401(k) plans for employer-funded defined contribution plans, despite the maximum employee contributions permissible under such plans. [Vol 2:2 Simplification and IRC § 415 plans.8 Repeal of section 415(e) would also repeal the stricter limits on the benefits for key employees in such plans." IV. APPROACHES TO SIMPLIFICATION OF SECTION 415(e) The case for repeal of section 415 loses appeal if the section can be simplified- We have already seen that much of the complexity derives from subsection (e). In this part, I will suggest two approaches toward simplifica- tion of subsection (e). The approaches stand alone, although they also could be combined. 90 A. Lengthening the Dollar Linit Phase-hI Under Section 415(b)(5), with Annual Allocation of a Section 415(e) Percentage The major cause of complexity in section 415(e) is that the (c) limits on additions to defined contribution plans are annual, while the (b) limits on a defined benefit are aimed at limiting a benefit accrued over a participant's career with an employer. Section 415(e) attempts to cope with this dissimilar- ity of limits by forcing the defined contribution fraction into a career mold, reflecting defined contribution utilization for all of an employee's years of service with the employer. It is this fraction that is at the root of the complex- ity in section 415(e), for it is this fraction that requires a backward look at all previous additions to a participant's account in light of both the dollar and compensation limits; results in annually changing fractions; and results in different fractions for each employee. Because there does not appear to be any simple means of equating the (c) limit to a career benefit, simplification of the defined contribution fraction, and thus subsection 415(e), has not seemed possible. A change in focus, however, yields a possibility for simplification. Instead of attempting to force the defined contribution fraction to mimic a career benefit, it would be possible to do the reverse, (i.e., equate defined benefit accruals to annual additions). This could be accomplished by extending the current ten-year 88. See supra text accompanying notes 22-24. 89. Any proposals to repeal § 415(e) thus should be inapplicable to top-heavy plans, although it should be noted that repeal of the rules applicable to such plans is another target of simplification-minded reformers. See Gridlock, supra note 1. at 34. It should be noted that the proposal of the ABA Tax Section to eliminate § 415(e) would apply to all plans, including super top-heavy plans. See ABA Report, supra note 6. at 12. 90. There are a number of relatively small, and relatively noncontroversial, changes that would simplify § 415. See supra notes 12 & 21. 19941 Florida Tax Review phase-in for the dollar limits, perhaps to thirty-five or so years.9' If this were done, employers could simply allocate a section 415(e) percentage between the defined benefit and defined contribution plans for each year in which an employer maintains two plans. This approach dispenses with the need to recompute annually the defined contribution limitation, the primary source of section 415(e)'s complexity. If the employer does not alter the allocations of the section 415(e) percentage from year to year, the plan would be able to determine an employee's defined benefit limit by multiplying a participant's years of participation by the portion of the overall percentage allocated to the defined benefit limit. Even if the employer did choose to vary the allocations, the section would not present a high degree of complexity or significantly increase the amount of recordkeeping for the plan. 92 To illustrate, assume that Congress amended subsection (e) to permit a 120% combined limit of the sort suggested. Each employer that sponsors both types of plans would allocate the 120% between the separate limits for each plan type. Assume an employer allocates 100% of the combined limit to the defined benefit plan and 20% to the defined contribution plan. An employee could receive 20% of the maximum annual additions to the defined contribution plan in a particular year and would also be credited with 1/35 of the defined benefit limitation for that year.93 When the employee retired, his defined benefit limit would be the sum of the annual phase-in credits, multiplied by the adjusted defined benefit dollar limit at the time of benefit payment. Thus, an employee with twenty years of credit under such an allocation, could receive a maximum benefit of 20/35, or 57% multiplied by whatever the indexed dollar limit is at retirement. Or assume that the employer allocates 40% of a year's allocation to a defined contribution plan and 80% to a defined benefit plan. In such a year, the employer could contribute 40% of the (c) dollar limit to the defined contribution plan. In addition, the employee's dollar limit for the defined benefit would be 80% of 1/35 of the indexed dollar limit at retirement. An employee who had twenty years of service under a plan that maintained such 91. This article suggests the phase-in for the dollar limit only. The alternative limit restricting benefits to 100% of compensation could continue to apply but would not require a phase-in period based on participation. 92. The plan would have to keep track of the employer's allocation of the § 415(e) percentage for each plan year, but the same allocation would apply to all employees. In addition, transition rules would require that one-time transition data be created and retained. 93. APPWP has suggested that Congress repeal the compensation limit on additions to defined contribution plans because this limitation is unnecessary to control benefits of the highly compensated because of the applicability of the dollar and deduction limitations and can harm lower-paid employees. See Gridlock, supra note 1, at 34. [Vol 2:2 Simplification and IRC § 415 an allocation of the section 415(e) percentage would earn 80% of 20/35, or 46% of the dollar limit as indexed through the date of retirement. One policy objection to this approach toward simplification is the limitations it would place on an employer who has not in the past sponsored any retirement plans and now wishes to sponsor a defined benefit plan that will provide a generous benefit for all employees, even those within ten years of retirement. Under current law, the employer could provide a benefit equal to 100% of the (b) dollar limit; under the approach suggested here, however, the benefit could not exceed 10/35 of the dollar limit. This does not seem a particularly troubling problem. The employer can still fund a very large benefit for the employee over ten years, just not as large as is permissible under current law. This may not be too steep a price to pay for simplicity. In any event, it would be possible to base the phase-in on years of service rather than participation with respect to any prior year in which the employer had not maintained a qualified plan, which would permit a more accelerated phase-in of the (b) dollar limits.9' A formula could also be devised to award prorated years of past service in cases in which less than the maximum annual additions were made to an employee's defined contribu- tion account. This would add a measure of complexity to things, but at least the complexity would be a one-time occasion and affect only a relatively small number of employers. The restructuring of the defined benefit limitation into annual increments suggested here would bring with it an advantage in addition to simplicity. Under current law, the limitations are applied separately for each employer. Thus, an employee who changes employment may accumulate 94. Section 415(b)(5) currently requires that the dollar limit be phased in over 10 years of participation in the plan. IRC § 415(b)(5)(A). The compensation limit, on the other hand, may be phased in over 10 years of service, which means that the phase-in has no effect on most participants in a plan whose benefits are based on past service. The 10-year phase-in of the dollar limit was intended to prevent a small employer from timing the establishment of a defined benefit plan to shortly before retirement of a key employee, who would have enough service to receive a fully funded benefit while other employees might lack sufficient service. After the key employee retires, the employer may decide to terminate the plan. depriving other employees of the opportunity to earn a full benefit. Thus, if this article's proposal permitted past-service credit for purposes of the extended phase-in period for the dollar limits, an additional 10-year participation phase-in still would be necessary to prevent employers from timing establishment of defined benefit plans in the manner described. The current phase-in period for the dollar limits applies not only to establishment of a defined benefit plan, but, to the extent provided in regulations, to benefit improvements in existing plans as well. IRC § 415(b)(5)(D). The Service. however, has decided not to implement this provision because the nondiscrimination regulations substantially eliminated the need for the requirement. Rev. Proc. 92-42, 1992-1 C.B. 872, §§ 2.02, 3. In any event. there is no reason why an extended 35-year phase-in period should apply to benefit improve- ments if the proposal suggested here were adopted. 1994] Florida Tax Review substantially larger retirement savings than an employee who remains with one employer for his or her career. Because the approach suggested here would be applied separately for each year, the benefits for employees who work for multiple employers would be reduced.95 B. Restricting Section 415(e) Calculations to Small Employers We have earlier seen that a repeal of section 415(e) would not have a significant effect on the benefits provided by large employers." The reason for this is that the nondiscrimination rules of section 401(a)(4) will generally require that whatever benefits are provided to highly compensated employees (as a percentage of their compensation, with a $150,000 limit on compensation taken into account) 97 are also provided to other employees. As a result, the costs of providing high benefits to non-highly compensated employees would constrain most large employers from providing too high a level of benefits for highly compensated employees. And if these costs did not constrain employers, there would be a measurable gain in the extent to which the tax subsidy is benefiting all employees, a desirable policy goal. If this analysis is correct, one approach to simplification of section 415(e) would be to define a class of employers for whom its application does not generally serve any purpose and then exempt that class from its require- ments.98 One possible definition of plans that could be exempted from 95. This article does not propose transition rules to the suggested simplification of the (e) limits. There are many directions transition rules could take, however, depending on how one resolves various policy choices. An outline of one approach would be for the defined benefit dollar limit under prior law to be frozen as of the effective date and to assign a fresh-start dollar limit to each employee. Each employee's fresh-start limit would grow annually under the new § 415(e) approach (i.e., a percentage would be added to it each year; the percentage in each year would be the same for each employee) and would be applicable when it exceeded the frozen limit under prior law. The defined contribution fraction could be dispensed with immediately. One possibility for the initial fresh-start limit would be the lesser of (i) the frozen limit and (ii) years of defined benefit plan participation as of the effective date multiplied by 1/35, with reduced amounts for years in which the employee participated in both a defined contribution and a defined benefit plan. Another possibility for the fresh-start limit might be (i) years of service multiplied by (ii) the § 415(e) percentage (which as suggested in the text, would be 120%) less the defined contribution fraction as of the effective date (but in no event to be greater than 100%). 96. See supra text accompanying note 79. 97. IRC § 401(a)(17). Prior to the Omnibus Budget Reconciliation Act of 1993, P.L. No. 103-66, the compensation limit was $200,000. 98. Indeed, the legislative debate over ERISA included early proposals to apply limitations only to "proprietary employees" in relatively small corporations. See S. 1179, 93d Cong., 1st Sess. § 702 (1973) (bill reported out of Senate Finance Committee applied limits [Vol 2:2 Simplification and IRC § 415 section 415(e) is "top-heavy plans," as defined in section 416(g).9 While determining whether a plan is top-heavy is itself a complex undertaking, plans already are required to make the determination. To further limit complexity, the section 415(e) limitation might be applied only to "key employees" under section 416(i)(1) or "highly compensated employees" under section 414(q).'O While this approach to simplification may be sensible, it does go against the grain of an apparent policy in subchapter D that distinctions between large and small employers generally are inappropriate.'"' Neverthe- less, the argument that section 415(e) should be eliminated for the class of employers who would not in any event be in a position to load up contribu- tions in excess of what the (e) limits already provided is a strong one; if the contours of such a class can be outlined, administration of such plans would be simplified. V. CONCLUSION Section 415, as it currently exists, serves a purpose but its complexity extracts a cost. Is the cost too steep a price to pay for the section's contribu- tions? The answer is debatable, but is also beside the point if the section can be simplified. If simplification is possible it should occur, regardless of one's feelings about whether the section in its current form is worth its current costs. In this article I have shown that it is possible to reduce substantially the section's complexity by altering the structure of subsection (e). I have only to plans of "proprietary employees," which were defined as plans in which at least 25% of the benefits were for employees with 2% or more of the corporation's stock and to plans of self-employed persons); see also Subcomm. on Labor of the Senate Comm. on Labor and Pub. Welfare, supra note 80, at 1653 (statement of Sen. Curtis): id. at 1656. 1658-59 (statement of Sen. Nelson). 99. See IRC § 416(a). See also supra notes 23-24 and accompanying text. 100. There is at least one problem with applying § 415(e) to top-heavy plans: how to treat plans that are top-heavy in some years, but not others. For example, assume a defined benefit plan that is not top-heavy and in which all participants have accrued the maximum benefits permissible under § 415(b). Assume further that the employer also has made maximum contributions to a defined contribution plan. In a subsequent year, the plan becomes top-heavy. If the § 415(e) limit now applies, the plan will have to reduce accrued benefits. This scenario is certainly not desirable. But neither is it inevitable: § 415(e) could apply only to benefit accruals that take place in years when the plan is top-heavy. 101. It apparently was this concern that caused Congress to create limits for all plans rather than only for smaller plans with proprietary employees. Subcomm. on Labor of the Senate Comm. on Labor and Pub. Welfare, supra note 80, at 1704 (statement of Sen. Buckley); id. at 1706-07 (statement of Sen. Gravel); id. at 1755 (statement by Sen. Bellmon); id. at 1777-1780 (statement of Sen. Thurmond). see also supra note 94. 1994] 98 Florida Tax Review [VoL 2:2 also suggested that it might be possible to limit the effects of the section's complexity by targeting the section more precisely at employers in a position to exploit the tax planning possibilities its repeal would make possible. Either, or both of these suggestions, would simplify the administration of qualified plans. It is true, of course, that outfight repeal of section 415, or section 415(e), would simplify administration of qualified plans even more, but it is also true that outright repeal would result in a loss of revenue at a time when we can ill afford it. Moreover, repeal might indirectly reduce the benefits of younger non-highly compensated employees by creating fresh incentives for some small employers to substitute defined benefit plans for defined contribu- tion plans. Repeal, then, should be regarded as the remedy of last resort, not the remedy of choice.