A Critique of SIMPLE--Yet Another Tax-Favored Retirement Plan

Richard J. Kovach*

  Introduction and Technical Background

Since January 1, 1997, qualifying employers have been able to implement a Savings Incentive Match Plan to promote retirement income security for their employees.⁽¹⁾ "Simple retirement accounts" (SIMPLE plans) serve employers having no more than 100 employees⁽²⁾ and follow the popular cash or deferred arrangement elective deferral format used in retirement plans qualified under § 401(k) of the Internal Revenue Code.⁽³⁾ SIMPLE permits employers and employees to take advantage of the usual federal tax benefits associated with elective deferral retirement plans.⁽⁴⁾ Unlike employers adopting other tax-favored retirement plans, employers who adopt a SIMPLE arrangement cannot simultaneously maintain another tax-favored retirement plan.⁽⁵⁾ The introduction of SIMPLE plans corresponded with the demise of a similar tax-favored retirement plan known as Salary Reduction Simplified Employee Pensions (SARSEP).⁽⁶⁾ Since SIMPLE plans in effect replace SARSEPs, the technical differences between the two plans invite a brief review.

Both SIMPLE and SARSEP plans use employee elective deferrals to provide funding and Individual Retirement Accounts (IRAs) to hold and invest plan assets.⁽⁷⁾ Congress devised both retirement plans to encourage employee retirement savings while freeing the sponsoring employer from inordinate regulatory complexity.⁽⁸⁾ SARSEPs cannot have a sponsoring employer with more than twenty-five employees.⁽⁹⁾ They operate under a peculiar and limiting set of participation criteria,⁽¹⁰⁾ cannot be maintained unless at least fifty percent of eligible employees elect deferrals in a given year,⁽¹¹⁾ permit elective deferrals of up to $10,000 per year without requiring an employer matching contribution,⁽¹²⁾ and contain a nondiscrimination stricture that permits highly compensated employees to elect maximum deferrals only if rank and file employees choose deferrals at a sufficiently high rate of contribution.⁽¹³⁾

By contrast, SIMPLE plans accommodate sponsoring employers having four times as many employees,⁽¹⁴⁾ primarily base participation on whether an employee earns $5000 or more per year,⁽¹⁵⁾ do not require a particular level of elective participation by rank and file employees, restrict elective deferrals to only $6000 per year,⁽¹⁶⁾ do not contain nondiscrimination requirements,⁽¹⁷⁾ but do require employers to make matching contributions of up to three percent of the compensation of employees who make elective deferrals.⁽¹⁸⁾ Economically, a $4000 reduction in the maximum amount an employee can elect to defer per year and an uncertain amount of mandatory employer matching contributions become the tradeoffs for a somewhat less complex retirement plan that operates without potential reductions in elective deferrals made by highly compensated employees.⁽¹⁹⁾

The SARSEP/SIMPLE substitution deserves praise for eliminating some complexity in the implementation of retirement income security goals, but will not meaningfully increase access to tax-favored retirement savings for American workers. SIMPLE's primary structural defect results from the continued ambivalence of Congress in letting employers avoid funding private retirement plans altogether if they wish,⁽²⁰⁾ while compelling broad pension coverage and funding once employers do venture into retirement plan sponsorship.⁽²¹⁾ Mandated pension coverage and funding assign primary responsibility for retirement savings to employers and characterize tax-favored retirement savings for highly compensated employees as not intrinsically worthy.⁽²²⁾ SIMPLE reaffirms these principles at a time when the popular media repeatedly stresses the importance of national savings, the tenuous nature of Social Security,⁽²³⁾ and the failure of most Americans to adequately prepare for their retirement, let alone for their next possible job loss. Public awareness of the significance of individual retirement savings grows, yet SIMPLE only continues a tradition of pension regulation that prevents individual workers from exercising the most fundamental decisions affecting their retirement income security.

  Attempts to Achieve Retirement Savings Access and Simplicity

The Employee Retirement Income Security Act of 1974 (ERISA)⁽²⁴⁾ introduced a long and detailed set of rules governing the qualification of standard, employer-sponsored retirement plans.⁽²⁵⁾ A large portion of the complexity of this qualification scheme results from a policy that prohibits or restricts economic discrimination tending to favor highly compensated employees over rank and file employees in the implementation of a retirement plan's participation, vesting, and benefit or contribution allocation provisions.⁽²⁶⁾ The nondiscrimination rules presume that the revenue-depleting tax benefits of qualified plans should not be granted to employers and (selected) employees if a plan is structured primarily to provide tax relief to high-income participants.⁽²⁷⁾ Highly-paid persons can take advantage of retirement plan tax benefits, but their employers must ostensibly provide substantial commensurate benefits for a sufficient number of non-highly compensated employees.⁽²⁸⁾

An employer desiring to promote tax-favored retirement income security must pay a price in costs and complexity that discourages plan sponsorship to a substantial degree.⁽²⁹⁾ If an employer decides not to sponsor a qualified retirement plan, its employees will have limited access to tax-favored retirement savings. Even employers who do adopt a qualified plan frequently limit full employee access by exercising options available to restrict participation or reduce funding, despite the general thrust of the nondiscrimination rules.⁽³⁰⁾

ERISA addressed the issues of access and simplification by authorizing do-it-yourself IRAs.⁽³¹⁾ Originally, workers could establish and fund their own IRAs with up to $1500 per year in deductible contributions.⁽³²⁾ Today, the contribution limit is only $2000,⁽³³⁾ although IRAs can receive considerably greater sums if they are used as "rollover" vehicles⁽³⁴⁾ or constitute receptacles for contributions to a SIMPLE plan or a Simplified Employee Pension (SEP).⁽³⁵⁾

The SEP appeared just four years after the IRA, when Congress added subsection (k) to § 408 of the Internal Revenue Code in 1978.⁽³⁶⁾ Like SIMPLE plans, SEPs encourage employers to adopt a retirement income security device that avoids the bulk of regulation associated with qualified plans under I.R.C. § 401(a). Unlike SIMPLE plans, the SEP, as it survives after 1996 without its SARSEP feature,⁽³⁷⁾ requires substantial employer contributions in addition to the direct compensation paid to participating employees.⁽³⁸⁾ The requirement for additional employer funding restrains sponsorship of the SEP. SIMPLE plans, by contrast, involve the less costly elective deferral format favored by employers because much of the plan's funding comes from compensation otherwise paid directly to the participating employees.⁽³⁹⁾

The growth of cash or deferred arrangements under I.R.C. § 401(k) suggests that the primary problem with employee access to qualified retirement plans is not complexity per se, but funding costs mandated against employers under the nondiscrimination strictures of the qualification scheme.⁽⁴⁰⁾ That is, many employers do not seem to mind sponsoring a retirement plan as long as the employees, rather than the employer, fund the plan's benefits. True, even if all benefits come from employee contributions, the employer still faces a certain amount of administrative cost and risk in the implementation of a qualified retirement plan.⁽⁴¹⁾ Employer funding costs, however, enter § 401(k) plans in the form of matching contributions, frequently necessary to avoid violating the § 401(k) deferral percentage nondiscrimination rules, which can prevent highly paid employees from making meaningful elective deferrals.⁽⁴²⁾ To the extent the employer must make matching employer contributions to sustain plan viability, it will hesitate to implement or continue a § 401(k) plan. Many employers simply decide that the blank check obligation placed on them by the annual uncertainty about employee elective deferral decisions justifies their not sponsoring a plan.⁽⁴³⁾

Section 401(k) plans, and their kissing cousins,⁽⁴⁴⁾ certainly increase retirement income security for millions of American workers, but the employers of many millions of other workers refuse to sponsor one. In some cases, employees have adequate access to retirement income security because their employers sponsor a sufficiently generous defined benefit pension plan, profit sharing plan, stock bonus plan, money purchase pension plan, or employee stock ownership plan, to name the more prominent qualified retirement plans other than cash or deferred arrangements. Still, too many employees have no tax-favored option to supplement Social Security income beyond a standard do-it-yourself IRA, which sets a relatively parsimonious contribution rate of no more than $2000 per year and restricts contribution deductions that avoid funding with already-taxed dollars.⁽⁴⁵⁾

To be sure, among workers having no employer-sponsored plan, a significant number will consider themselves at least temporarily unable to afford putting aside even $2000 annually toward building a private pension. This group, however, will contain a subgroup who will approach retirement with a savings wake-up call. These workers will want to find a way to put aside much larger sums than $2000 annually in the few years just prior to their retirements. Their motivation to save may result from looming future needs or perhaps even a new-found desire to retire a bit earlier than formerly anticipated. Nonetheless, these late bloomers must forego the tax benefits normally associated with pension savings, except for the modest $2000 IRA annual contribution.⁽⁴⁶⁾

The last few years just prior to retirement frequently produce economic pressures and incentives largely ignored under current pension arrangements. Some workers deliberately seek to enhance their pre-retirement income during these years by initiating additional employment, working longer hours at their regular job, or working harder and smarter to earn bonuses, raises, or one final promotion. These efforts can involve a worker's spouse as well. Even if the spouse has previously done little or no work outside the home, and even if the spouse is significantly younger than the retiring worker, spousal employment in anticipation of a retirement can become a distinct preparatory prospect. Unfortunately, the $2000 IRA annual limit does no more for a family with dual workers than for a married worker carrying alone the burden of retirement preparation.⁽⁴⁷⁾

Neither SIMPLE, nor other tax-favored plans, adequately address the problem of access to retirement income security. All American workers should have an opportunity to accumulate substantial tax-favored retirement savings, yet SIMPLE and other retirement plans preclude individual participation without employer sponsorship. An employer can reject plan sponsorship because of a plan's funding costs, even if minimal, or simply because of the time, attention, and risks involved in the implementation and maintenance of a plan.⁽⁴⁸⁾ SIMPLE requires some degree of employer funding, with a modest, but discernible, blank check consequence.⁽⁴⁹⁾ Many employers, particularly those competing in low profit industries, balk at the idea of voluntarily incurring any additional employee costs, especially in view of what appears to be an inexorable legislative trend toward mandated employment benefits.⁽⁵⁰⁾ Moreover, even if an employer adopts a SIMPLE plan, the basic $6000 elective deferral limit is not high enough to address the late bloomer problem. Indeed, the $6000 limit does not permit relatively adequate benefit accruals for employees who have the means and economic discipline to begin voluntary retirement savings early in their careers.⁽⁵¹⁾

  SIMPLE Lacks Ultimate Simplicity

The legislation adding SIMPLE to the list of tax-favored retirement devices became law on August 20, 1996.⁽⁵²⁾ Within weeks, and well before the January 1, 1997 effective date of SIMPLE, the Internal Revenue Service (I.R.S.) began issuing a series of interpretive authorities to resolve various questions that otherwise would have hindered the practical implementation of SIMPLE plans by taxpayers. In order to start a SIMPLE plan by January 1, 1997, employers had to contend with the new legislation's sixty-day notice-to-employees requirement, which meant that plans had to be in place by November 2, 1996 under the literal provisions enacted.⁽⁵³⁾ The I.R.S. solved this problem by interpreting the sixty-day notice period as beginning as late as January 1, 1997 for SIMPLE arrangements adopted for the 1997 calendar year.⁽⁵⁴⁾

The Department of Labor addressed an additional timing problem that arose when SIMPLE came into law. The legislation requires an employer to turn over its employees' salary reduction contributions to the financial institution that serves as trustee for the designated IRAs no later than the thirtieth day of the month following the month in which the contributions would otherwise have been paid to the employees in cash.⁽⁵⁵⁾ Because Title I of ERISA, administered by the Department of Labor, covers SIMPLE arrangements⁽⁵⁶⁾ and authorizes a Labor Department regulation that creates a fifteen-day rule for depositing such contributions,⁽⁵⁷⁾ an immediate conflict existed. This clash between a tax rule and a labor rule resulted in an acquiescence by the Department of Labor, which agreed to amend its fifteen-day rule consistently with the thirty-day rule of the Internal Revenue Code.⁽⁵⁸⁾

Timing issues, however, were only a warm up for more substantial interpretations that followed. The I.R.S. published a new Form 5305-SIMPLE on October 31, 1996,⁽⁵⁹⁾ just in time for employers who desired to implement a SIMPLE plan for the first full year of eligibility, calendar year 1997. Like other model plan forms published to provide guidance to taxpayers adopting more complicated retirement plans, the Form 5305-SIMPLE attempts to organize the salient features of the plan's enabling legislation in a comprehensible manner with specific instructions.⁽⁶⁰⁾ Form 5305-SIMPLE did not resolve several issues that required further elaboration. For example, questions arose regarding employer aggregation possibilities, leased employees, the definition of eligible employees, employee election mechanics, interrelationships between SIMPLE contributions and employment taxes, tax-free rollover rules, application of the § 401(a)(17) compensation limitation, designations for contribution deposits, and plan establishment procedures.⁽⁶¹⁾

Not only has SIMPLE launched its own series of interpretive authorities, but because § 401(k) plans can now supplant their nondiscrimination rules with a set of contribution mechanics that parrot the SIMPLE scheme, the I.R.S. must issue additional interpretations to address the hybridization of SIMPLE and § 401(k) concepts.⁽⁶²⁾ Indeed, the incorporation of the SIMPLE scheme as an elective mechanism into § 401(k) plans complicates retirement plan selection for employers choosing an elective deferral plan and causes employers with existing § 401(k) plans to second-guess their present arrangements.

Employers must try to evaluate whether the annual deferral percentage rules in force at the time the SIMPLE legislation was enacted⁽⁶³⁾ will allow sufficient levels of future elective deferrals for highly compensated employees, compared to the maximum $6000 elective deferrals allowed for all participating employees under a SIMPLE arrangement.⁽⁶⁴⁾ The pre-existing deferral percentage rules can result in limiting the elective deferrals of highly compensated employees to a much lower amount than $6000 annually if rank and file employees elect only modest deferrals.⁽⁶⁵⁾ If elective deferrals by the rank and file are sufficiently high, highly compensated employees are permitted to defer as much as $4000 more than the SIMPLE limitation on individual elective deferrals.⁽⁶⁶⁾

In many instances, employers must induce their rank and file employees to make sufficiently high elective deferrals by offering matching employer contributions. Other employers have rank and file employees who have a keen interest in self-funding their retirement income security needs with elective deferrals and thus contribute at levels that automatically permit highly compensated employees to elect maximum annual deferrals without the necessity for any matching employer contributions.⁽⁶⁷⁾ If a sponsoring employer does not mind limiting maximum elective deferrals to no more than $6000, it can adopt the SIMPLE scheme as part of its § 401(k) plan,⁽⁶⁸⁾ but the cost of this election is compliance with the mandated matching contributions that are an integral part of the SIMPLE concept under I.R.C. § 408(p)(2).⁽⁶⁹⁾

The mandated matching contributions required under SIMPLE appear relatively modest when compared to the large matching contributions indirectly necessitated in some plans under the standard deferral percentage rules.⁽⁷⁰⁾ Since other § 401(k) plans, however, might not practically need employer matching contributions to maximize individual elective deferrals, the question of whether and to what extent an employer should incur the expense of matching contributions requires a complicated determination for a large number of possible work force situations. Introducing the SIMPLE election into this determination process only adds additional complexity, since the basic SIMPLE format represents a fixed tradeoff involving a precisely lower maximum individual deferral limitation ($6000) and a somewhat fixed amount of matching contributions required of the employer.⁽⁷¹⁾

By declining the SIMPLE option, the employer possibly stretches the $6000 limitation upward by as much as $4000, but might have to exceed the SIMPLE matching contribution requirement as a practical consequence of the standard § 401(k) deferral percentage rules, which encourage, but do not actually require, matching contributions for rank and file employees.⁽⁷²⁾ The SIMPLE election eliminates the nondiscrimination testing requirement completely, but could result both in the employer's incurring unnecessary costs for required contributions and artificially reducing the annual elective deferral limitation for employees who desire to maximize their savings for retirement income security.⁽⁷³⁾

Offering the SIMPLE mechanism as an alternative to § 401(k) plans does not simplify planning prospects for plan sponsors, even though using the SIMPLE election eliminates the annual compliance difficulties encountered under the deferral percentage tests otherwise applicable. Likewise, the basic pension planning decision as to which kind of tax-favored arrangement to select in the first place becomes more difficult because of the independent existence of SIMPLE plans, aside from the potential use of the SIMPLE concept in § 401(k) plans. Once adopted, a SIMPLE plan indeed eliminates compliance complexity associated with plans subject to nondiscrimination testing.⁽⁷⁴⁾

Having many choices can produce a bane as easily a boon. As the number of kinds of tax-favored retirement plans grows,⁽⁷⁵⁾ employers face increasingly difficult planning choices. Respecting SIMPLE, the problem of choice is exacerbated by the requirement that a sponsoring employer cannot simultaneously maintain another tax-favored savings arrangement.⁽⁷⁶⁾ In view of the economic magic resulting from the compounding of investment returns in any tax-exempt savings trust, limiting annual inputs of each employee to just $6000 per year (plus a modest employer matching contribution), implies very serious long-term financial consequences. As much as $30,000 per year per employee can flow into other defined contribution retirement savings devices.⁽⁷⁷⁾ Many employers will prefer to let their advisers tinker with the complicated, yet somewhat manageable, nondiscrimination burdens afflicting less simple plans rather than adopt a plan that offers so much less economic promise.⁽⁷⁸⁾

Just as under the SIMPLE feature analysis for § 401(k) plans, however, cases will arise in which employer costs necessitated by the general nondiscrimination strictures of § 401(a) will exceed the relative loss of economic benefit implicit in the fixed contribution limitations imposed upon SIMPLE plans.⁽⁷⁹⁾ The current array of retirement planning vehicles available makes guessing exactly when the SIMPLE choice offers the best result a very difficult task, especially in view of the great work force fluctuations that characterize the modern American economy. Inertia no doubt infects pension planning decisions as detrimentally as any other set of economic choices. Once an employer implements a particular plan, it will tend to maintain that plan even after the work force (or the law governing retirement plans) has changed in a way that suggests a different kind of plan would be more economically productive.⁽⁸⁰⁾ SIMPLE gives smaller employers more to think about when choosing a retirement plan. The relatively stingy SIMPLE contribution limitation guarantees a difficult economic analysis in many instances.⁽⁸¹⁾

  How Tax-Favored Retirement Plans Became Complicated

Tax-favored⁽⁸²⁾ describes a set of income tax characteristics that primarily benefit the participating employees of a retirement plan. The sponsoring employer gets a deduction under I.R.C. § 404 for contributions made, subject to limitations that vary in complexity depending on the kind of plan implemented.⁽⁸³⁾ Nevertheless, if the deferred compensation otherwise contributed to a retirement plan's trust were paid directly to employees, the employer would, in most instances, obtain a deduction directly under I.R.C. § 162 for such payments.⁽⁸⁴⁾ The participating employees benefit by not having to recognize income on plan contributions until distributions are actually made from the plan's trust (often many years subsequent to the date the employer makes contributions).⁽⁸⁵⁾ Employers similarly defer income recognition on earnings accumulated in the trust fund that receives plan contributions.⁽⁸⁶⁾ Participants avoid employment taxes on plan contributions in many instances,⁽⁸⁷⁾ and employees or their beneficiaries frequently receive distributions under more favorable tax conditions than otherwise would exist if plan contributions were paid as direct compensation.⁽⁸⁸⁾

Some treasury officials might view this package of tax benefits as a considerable revenue subsidy that favors increased retirement income security as a national goal in much the same way as the home mortgage interest expense deduction of I.R.C. § 163(h) favors increased home ownership for Americans.⁽⁸⁹⁾ As revenue detractors, tax-favored retirement plans invite attention to tax savings that might inordinately flow to wealthier users of such plans. Since the owners of businesses can easily become employees of their enterprises, the blessings of tax-favored retirement plans do not inure only to common employees. Accordingly, Congress created the technical concept of the highly compensated employee⁽⁹⁰⁾ and authorized a complex set of rules designed to predicate use of tax-favored retirement plans by affluent employees upon an adequate level of contributions or benefits for a sufficient number of rank and file participants.⁽⁹¹⁾

These rules allow highly compensated employees to take full advantage of tax-favored retirement plans only if the sponsoring employer (they implicitly control) takes care of the retirement income security needs of common workers in the enterprise. Indeed, this parental view of the employer extends beyond mere coverage and contribution/benefit impositions to embrace other social concerns peripheral to retirement income security. Thus, the qualification scheme of § 401(a) also includes rules that implement spendthrift protection for retirement plan assets,⁽⁹²⁾ protect the surviving spouses of deceased participants,⁽⁹³⁾ and ensure that plan assets are not diverted to the benefit of the employer or other outsiders.⁽⁹⁴⁾

Because violation of these rules can result in reversal of the tax benefits otherwise conferred as a reward for compliance,⁽⁹⁵⁾ the complex qualification rules create a multitude of possibilities for plan administrators to adversely affect the tax postures of large numbers of employees. Consequently, administrative costs resulting from reasonable attempts to comply with such abstruse rules weigh against the substantial tax savings at stake. Administrative costs, however, frequently constitute only a small portion of total costs associated with the implementation and maintenance of tax-favored retirement plans. Contributory costs necessitated by the coverage and nondiscrimination strictures of I.R.C. § 410 and § 401(a)(4) account for the bulk of plan expenses.⁽⁹⁶⁾

Employers do not uniformly incur these costs and risks. Contributory expenses vary greatly, even for work forces similarly situated, due to a number of technical devices that give plan sponsors a wide degree of discretion to reduce contributory costs depending upon the individual design of particular plans. A plan's design can take advantage of both statutory coverage exclusions and employer-created exclusions to greatly reduce the number of employees eligible for plan coverage within a work force.⁽⁹⁷⁾ Even before taking advantage of coverage exclusions, the employer can implement technical devices that eliminate employees per se from the work force in question. These include use of independent providers of services,⁽⁹⁸⁾ leased employees,⁽⁹⁹⁾ affiliated entities,⁽¹⁰⁰⁾ and "separate lines of business" to reduce plan contribution costs quite significantly.⁽¹⁰¹⁾

Once such technical mechanisms become operative, the employer can look to yet another set of technical rules that further reduce contribution costs respecting rank and file employees, while preserving relatively generous contributions for highly compensated employees. These devices include Social Security integration,⁽¹⁰²⁾ age-weighted allocations,⁽¹⁰³⁾ contribution and benefit allocations based on proportionate current compensation,⁽¹⁰⁴⁾ and vesting schedules that result in forfeitures for prematurely terminating employees.⁽¹⁰⁵⁾ Of course, implementation and monitoring of these cost-saving devices result in further complexity and administrative costs.

Viewed from afar, the forces that cause complexity in tax-favored retirement plans seem quite bizarre. Legislation providing tax relief for retirement savings arrangements demands that sponsoring employers incur the burden of providing socially protected benefits for a seemingly large number of rank and file employees. Whether out of fear that the social burden will appear too great to induce many employers to implement plans, or as a result of legislative pressures that outright favor employers, Congress permits employers to minimize contribution costs at their discretion in various ways. The complexity resulting from both the cost-reduction methods and the nonwaivable social protection aspects of the legislative scheme has been manageable enough to allow large numbers of employers to sponsor tax-favored plans. A similar kind of complexity found its way into the Internal Revenue Code as a result of the inconsistent existence of high marginal tax rates and discretionarily implemented tax shelters prior to the Tax Reform Act of 1986.⁽¹⁰⁶⁾ For pension plans, complexity results from the inconsistent existence of rules promoting extensive employee coverage and funding with contrary rules that permit some employers, especially smaller ones, to find ways to exclude rank and file employees or otherwise minimize funding costs for them, while inordinately benefiting highly compensated employees.

This conceptual inconsistency, a grand but confusing legislative experiment, continues to the detriment of millions of uncovered or undercovered workers who lack access to substantial tax-favored retirement savings. The elective deferral trend led by § 401(k) plans has not sufficiently closed this pension access gap. The power to create a substantial retirement savings arrangement still remains with employers, and too many employers believe that exposure to complexity, possible litigation, and open-ended contribution costs under these plans does not justify the potential benefit to their work forces.

SIMPLE, with its employer contribution costs, 100 employee limitation, one-plan requirement, relatively weak contribution allowance, and employer sponsorship necessity will surely not result in full retirement savings access for all American workers.⁽¹⁰⁷⁾ Small employers eligible for SIMPLE still have many attractive options available with other kinds of plans that offer greater economic benefits for highly compensated employees while holding down contributions costs for rank and file employees. SIMPLE competes poorly among such high-powered choices.

  A Simple Way to Close the Retirement Savings Access Gap

The current complex system of tax-favored retirement savings plans for small employers resembles a cat-and-mouse game in which revenue enforcers attempt to lure employers into making substantial contributions on behalf of rank and file employees, while employers attempt to avoid tax disqualification despite restricting contributions as much as possible to key personnel.⁽¹⁰⁸⁾ Even if certain employers fail to use all significant means to reduce plan coverage and contributions for rank and file employees, a subgroup of such employers can nevertheless offset contribution costs by reassigning other components in their employees' compensation packages, such as future wage increases and fringe benefits.⁽¹⁰⁹⁾ Compensation offsets serve to redistribute contribution costs back to employees. They produce an economic effect similar to that of elective deferral arrangements, but without volunteerism. De facto self-funding for rank and file employees through offsets can occur at the same time substantial tax-favored deferred compensation enhances the compensation packages of highly paid workers. Offsets can make the legislative attempt to force employers to pay additional compensation to rank and file workers largely ineffective, if not outright futile.⁽¹¹⁰⁾

More importantly, attempts to broaden retirement savings access for ordinary workers by forcing their employers to adhere to minimum coverage and contribution standards fail not just because of discretionary loophole devices, but also because employers have complete discretion to decide whether and to what extent their work forces will attain access to tax-favored retirement plans. Rank and file workers themselves do not determine, beyond collective bargaining representation⁽¹¹¹⁾ and the meager possibility for direct IRA contributions, the extent of their individual access to tax-favored retirement savings.⁽¹¹²⁾ SIMPLE does not alter this basic fact of life for the majority of retirement-minded employees. If Congress closed the current loopholes permitting small employers to adopt qualified plans while severely restricting benefits for rank and file employees, many employers would shun retirement plans, resulting in even less access for common employees than currently exists. Furthermore, attempts to mandate anything close to universal employer-sponsored pension coverage would no doubt fail politically in view of the employer contributions already required under the Social Security system and widespread resistance to employer mandates resulting from many years of progressively imposed impositions outside the realm of retirement income security concerns.⁽¹¹³⁾

Policy makers should now consider the implications of removing the burden of the pension access choice from employers and placing it directly on employees. Shifting the burden of plan sponsorship would eliminate substantial complexity, since uncomplicated, self-funded IRAs already exist. The annual $2000 contribution limitation for self-funded IRAs must be raised to some greater figure set perhaps somewhere between the $10,000 limitation for elective deferrals under § 401(k) plans⁽¹¹⁴⁾ and the $30,000 annual addition limitation imposed by I.R.C. § 415 on other tax-favored defined contribution plans.⁽¹¹⁵⁾ Correspondingly, Congress could abolish employer sponsorship of most tax-favored retirement plans.⁽¹¹⁶⁾ In order to accommodate employees who start their retirement planning late in their careers, an amended IRA scheme could also include a provision allowing make-up contributions to some degree, similar to a contribution expansion feature now associated with I.R.C. § 403(b) plans.⁽¹¹⁷⁾

Likely, a proposal to eliminate employer sponsorship of retirement plans would immediately inspire a number of objections reflecting the flawed, but traditional, thinking about tax-favored retirement plans. The foremost objection might arise from the concern that every highly compensated employee would gain immediate access to a substantial annual tax benefit even if no rank and file colleagues decide to contribute to a personal IRA. Next, revenue considerations might lead to the suggestion that such a proposal would be too costly to implement. Additionally, critics would lament the threatened loss of future employer-provided contributions as would continue under existing plans, theorizing that such contributions are extra compensation otherwise not inuring to the benefit of the general work force.

The first of these objections assumes that retirement plan tax benefits should not be granted to higher income workers without requiring employers to fund benefits for lower income workers. This assumption suggests that no substantial reason exists, beyond legislative leverage to impose employer benefit mandates, to promote retirement savings among more affluent employees. To the contrary, encouraging retirement savings among better-paid workers creates salutary economic effects that ultimately benefit all workers and capital markets as well. Allowing higher-paid workers to prepare adequately for retirement permits them to leave the work force earlier and thus promotes upward mobility into managerial and professional ranks for younger workers who would otherwise remain nonhighly compensated workers for a longer time.⁽¹¹⁸⁾

Even if particular highly compensated workers want to remain working into advanced age, their potential economic independence resulting from significant retirement savings lets these workers give up particular positions in order to take advantage of opportunities to start a new business, enter more flexible consulting arrangements, or take a less secure (but perhaps more interesting) job with a different employer. This enhanced mobility benefits both older, higher-paid workers and the economy in general, since an efficient coupling of talent transferability, personal interest, and fluctuating market demand for services promotes economic growth.⁽¹¹⁹⁾ Additionally, enhanced retirement savings for any group of workers promotes more stable capital markets, since retirement savers less frequently make indiscriminate conversions of savings into consumption over the short term.⁽¹²⁰⁾

Another beneficial effect of allowing highly compensated employees to accumulate retirement savings in a tax-favored manner comes from the instructive influence their activity potentially exerts on the savings habits of rank and file employees. A universal system of totally elective retirement savings outside the Social Security system would not uniformly attract all rank and file workers. If a universal system not embracing mandated employer contributions produced less rank and file participation than exists under the current patchwork system of retirement plans, only the virtue of decreased complexity would compensate for reduced coverage. Decreased complexity alone, however, would not likely justify a change.

No doubt many employees, given a choice between a larger take-home pay with no retirement savings and a somewhat smaller take-home pay with substantial tax savings and enhanced retirement security would still choose to forego retirement savings. On the other hand, many rank and file employees whose employers have exercised their right not to sponsor a tax-favored retirement plan would greatly appreciate the opportunity to expand their retirement savings beyond the $2000 IRA annual limitation now in effect.⁽¹²¹⁾ Expanding the group of employees who wish to take advantage of tax-favored retirement savings opportunities presents a practical problem. Limiting the ability of highly compensated employees to demonstrate thrift in amassing retirement assets lessens the role model effect that encourages less affluent employees to take advantage of an elective savings program.⁽¹²²⁾

Promoting retirement income security as a voluntary national goal to benefit all Americans makes as much sense as promoting a national goal toward home ownership. Both goals merit support under the Internal Revenue Code.⁽¹²³⁾ For decades, tax benefits from home ownership have flowed to the affluent and those of modest means alike. As a result, approximately two-thirds of Americans own their own residence. Although allowance of tax benefits to encourage home ownership has generated some criticism,⁽¹²⁴⁾ no credible critic has yet suggested that the tax benefits of affluent homeowners should in any way be predicated upon the level of home ownership attained by less affluent citizens in a particular geographic area. This sort of linkage, however, burdens private retirement plans. Additionally, the tax and other regulatory⁽¹²⁵⁾ complexity of retirement plans continues to increase, while employee coverage remains less than satisfactory, even considering the popularity of elective deferral arrangements.

The attempt to promote rank and file employee coverage via employer mandates should be abandoned in favor of other mechanisms. The Internal Revenue Code could encourage participation by allowing cumulative contributions that favor rank and file employees in particular.⁽¹²⁶⁾ Lower-paid, young employees struggling to raise children, pay a home mortgage, and keep the family car running will often reluctantly forego any amount of take-home pay in favor of the seemingly quite distant prospect of retirement. The virtue of deferred gratification has its practical limitations. Frequently, however, a young worker becomes an older worker with children who have left home,⁽¹²⁷⁾ a bit higher income, a nearly-paid mortgage, and a more reliable automobile. This older worker also experiences more fatigue and looks forward to the day when the disciplined routine of employment will no longer be necessary. Surely, such a person would benefit greatly from a retirement savings feature that allows a reasonable opportunity to amass substantial retirement assets in a tax-favored manner within a relatively short number of years just prior to an intended retirement date. If, for example, this worker decided to get serious about retirement savings just seven years prior to retirement, the current $2000 IRA contribution limit would, of course, severely restrict retirement planning compared with the $6000 SIMPLE limitation or the $10,000 limitation applicable to a § 401(k) plan.⁽¹²⁸⁾

The worker would not have the opportunity to save at the higher levels permitted under SIMPLE or § 401(k) if the worker's employer, like thousands of employers, decided not to adopt such a plan. With direct access to higher savings limitations, the worker's plight would greatly improve and improve even more if the law allowed catch-up contributions. Possibly, a revised IRA scheme could allow double contributions for up to ten years prior to retirement if annual contributions were not made during an equivalent earlier period.⁽¹²⁹⁾ If the new IRA program generously permitted any worker to save the § 415 defined contribution plan dollar limitation of $30,000 annually, as many high earning self-employed professionals astute enough to maintain a § 401(a) retirement plan currently do, the retirement savings potential approaches miraculous dimensions by comparison.⁽¹³⁰⁾

Congress would have to set the precise contribution limit under a universal IRA program, as well as the extent of a cumulative contributions feature, consistently with politically acceptable revenue considerations. Potential revenue savings resulting from the discontinuance of various qualified plans now offered would no doubt affect the contribution limit of a universal IRA program. Critics of an expanded IRA substitution for the current menagerie of qualified plans would stress that highly compensated employees would get unfettered access to tax-favored retirement savings. In response, the designers of a universal IRA program could consider some form of means tested limitation. For example, the catch-up feature might phase out for highly compensated employees on the assumption that they already had sufficient opportunities to accumulate substantial retirement assets.⁽¹³¹⁾ Another approach could phase out either the catch-up feature or the basic annual contribution limitation depending on how large a total tax-favored accumulation the taxpayer owns.⁽¹³²⁾ These limitation mechanisms would not force employers to make contributions for rank and file employees, but the charge that the program might inordinately transfer tax dollars to the affluent would carry less weight.⁽¹³³⁾

Encouraging sufficient participation by rank and file workers should, however, take precedence over preventing or limiting the retirement savings of affluent workers. Even if overall participation increased as a result of providing meaningful access to retirement savings for workers not now covered under an employer-sponsored plan, policy makers would certainly desire further participation enhancement. Potential means to encourage participation abound. Government leaflets, instructions, and explanatory materials published in connection with administration of Social Security benefits and federal income tax collections could stress the ease and importance of securing tax and retirement benefits under an expanded IRA program.⁽¹³⁴⁾ Of course, banks, investment firms, and other private financial services institutions would have a direct interest in promoting the virtues of a universal retirement system that would gather huge amounts of capital for investment.⁽¹³⁵⁾ Implementation of rules that would allow certain employers to require their employees to make specified levels of IRA contributions as a term and condition of employment would also support widespread participation.⁽¹³⁶⁾

Yet, Congress should tread lightly on the turf of mandatory employee contributions. Most American employees already labor under a mandatory pension system that sets aside 12.4% of their compensations.⁽¹³⁷⁾ In tribute to the concept of individual choice, a program of elective retirement contributions beyond the Social Security system ought to permit each worker to decide personally whether and when to effect retirement income security beyond basic needs. Not everyone wishes to retire early, or even retire at some commonly viewed normal retirement age.⁽¹³⁸⁾ Age discrimination rules now allow employees to remain on the job longer, even if the progressive diseases of old age tend to interfere with employment effectiveness.⁽¹³⁹⁾ Some workers wish to semi-retire by continuing part-time remunerative activity after ceasing full-time employment.⁽¹⁴⁰⁾ Others plan on consuming assets accumulated outside formal pension plans during their working years, such as investment portfolios, whole life insurance policies, and home equities available through reverse mortgages that allow homeowners to supplement their income without disposing of their homes while alive.⁽¹⁴¹⁾

Many workers, including some not particularly affluent, do not need tax-favored supplemental retirement savings for a variety of reasons that include working until death, reliance on a spouse or other family members, reduced needs in retirement, and inheritances. Consequently, an expanded IRA program need not have universal participation as long as American workers have universal access to the system. Although arrangements like § 401(k) plans embrace an elective concept, all the most important choices of pension planning belong to employers and not those who directly benefit from tax-deferred retirement savings.⁽¹⁴²⁾ For this reason, current deficiencies in participation relate both to personal preferences and circumstances beyond the control of average workers. An expanded IRA program would link participation levels solely to the free choices of those who have a direct and personal interest in retirement planning.

  Technical Considerations Under a Universal Retirement Savings System

A potential exception to the elective feature of a universal access retirement system could protect the working poor, who currently benefit under the earned income tax credit.⁽¹⁴³⁾ Since the earned income tax credit is a refundable credit that effectively returns all or a portion of the dollars paid by low income workers as Social Security contributions, some of the dollars refunded could be used to fund an IRA on behalf of these workers. The dollars refunded come from general revenues paid by all taxed workers. Mandatory transfer of a portion of the credit to provide additional deferred compensation for workers who will receive the lowest amounts of Social Security benefits would not seriously violate the freedom of choice concept, since earned income tax credit recipients effectively get double credit for their Social Security payments and thus have not actually earned the dollars refunded.⁽¹⁴⁴⁾

Naturally, low income workers have many potential uses for the earned income tax credits they receive, and shunting some of these dollars into a deferred compensation arrangement might not serve them well in all cases. To address this consideration, the portion of a worker's earned income tax credit that might be mandatorily diverted to retirement savings could be adjusted on a sliding scale from zero to (near) 100% depending on a few basic variables, such as the worker's age and status regarding dependents.⁽¹⁴⁵⁾

A universal retirement savings system that promotes substantial thrift for both affluent workers and the working poor might also generate substantial interest from workers in the middle. To further emphasize the importance of retirement income security, the drafters of a revamped IRA system should retain restrictions on early withdrawals of accumulations. Indeed, Congress might consider tightening these restrictions, as they did for SIMPLE plans.⁽¹⁴⁶⁾ Young workers, however, might resist making retirement contributions that cannot practically (without a severe penalty) serve them until many years have passed. Implementation of the early withdrawal penalty on a sliding scale according to age could help remedy this problem. The penalty could be smaller--tolerable, but still not de minimus--for younger workers, while greater for older workers who more greatly jeopardize their retirement income security by converting retirement assets into current consumption shortly before retirement.⁽¹⁴⁷⁾

Another approach to remedy the reluctance of young workers to save for retirement would be to permit IRA holders to borrow from their accounts, as now allowed under many employer-sponsored plans.⁽¹⁴⁸⁾ Younger participants often do not mind deferring compensation if they can achieve substantial tax benefits while having immediate access to their money.⁽¹⁴⁹⁾ Allowing participant loans for IRAs, unfortunately, would inject a great deal of complexity into what otherwise serves as a uniquely simple retirement savings arrangement.⁽¹⁵⁰⁾ Perhaps rule makers could mitigate the complexity associated with plan loans by altering the current system of regulation to restrict plan credit to borrowings that relate only to the most fundamental economic events that face workers.⁽¹⁵¹⁾

Not applying § 415(c) would also encourage IRA contributions under a universal system. Section 415 sets the maximum annual addition to defined contribution plans at the lesser of $30,000 or twenty-five percent of a participant's yearly compensation.⁽¹⁵²⁾ A universal IRA system must, like SIMPLE, set a maximum on a participant's annual contribution. The dollar limit under a universal system would probably exceed the $10,000 limit now permitted under § 401(k) plans, and certainly exceed the $6000 SIMPLE limit, but, due to revenue considerations, not exceed the $30,000 allowed under § 415(c) for other tax-favored defined contribution plans. If, for example, the limit were set at $15,000, application of the twenty-five percent limitation under § 415 would mean that workers earning less than $60,000 per year could not contribute the full $15,000 per year. Furthermore, the twenty-five percent restriction would conflict with a catch-up feature designed to allow late starters to accumulate sufficient retirement savings in the last few years prior to retirement.⁽¹⁵³⁾ More importantly, the twenty-five percent restriction would interfere with the economic decisions of workers who might arrange their finances, if only temporarily, to divert substantially more than one-quarter of their income toward the important goal of retirement income security.

To illustrate, two spouses earning $15,000 and $45,000 annually might at some point decide that each could devote a full $15,000 contribution to retirement savings, for a total annual contribution of $30,000, or one-half of their combined incomes. Under the twenty-five percent restriction of § 415, the contribution of the lower-earning spouse could not exceed $3750 and even the contribution of the higher-earning spouse could not exceed $11,250.⁽¹⁵⁴⁾ Not applying the § 415 limitation in the SIMPLE scheme encourages voluntary retirement savings,⁽¹⁵⁵⁾ and workers under a universal IRA program would similarly benefit if Congress were to choose to keep § 415 from applying to self-funded IRAs.

Enhanced voluntary retirement savings opportunities, nevertheless, might result in inordinate tax-favored accumulations for some workers, depending on whether an expanded IRA system would replace both defined contribution and defined benefit retirement plans.⁽¹⁵⁶⁾ Retention of defined benefit plans would allow some persons to participate in a double system of private tax-favored retirement savings. Interestingly, the legislation that created SIMPLE also contained a provision that favors such private plan double dipping by repealing former § 415(e).⁽¹⁵⁷⁾ Currently, § 415(e) applies a complex limitation on accrued benefits when a worker participates in both a defined benefit and defined contribution plan sponsored by the same employer.⁽¹⁵⁸⁾ The separate limitations of § 415 applicable to defined benefit or defined contribution plans will apply individually as though the employer sponsors only the one kind of plan instead of both.⁽¹⁵⁹⁾ Under the present system of multiple tax-favored retirement plans, an employer can choose multiple benefit accruals that result in accumulations far in excess of those available by choice of the employee under a cash or deferred arrangement.⁽¹⁶⁰⁾ A comparison of the fortunate employees who participate in multiple benefit accrual arrangements with the hapless employees of employers that refuse to implement any retirement plan, makes the current $2000 per year self-funded IRA contribution appear most inadequate.

If potential revenue loss causes major objections to an expanded IRA scheme, surely pension policy analysts could revisit the § 415(e) issue. A maximum IRA contribution of $15,000 annually could be reduced to $10,000, $8000, or $6000 (to borrow the dollar limitations applicable in 1998 to existing elective deferral arrangements)⁽¹⁶¹⁾ if a participant also receives a meaningful benefit accrual under a defined benefit pension plan.⁽¹⁶²⁾ Similarly, if at the effective date of an expanded IRA scheme certain workers have already attained substantial benefit accruals under tax-favored retirement plans of whatever stripe, including putatively defunct defined contribution plans, the universal IRA scheme could reduce maximum contributions for these workers. Alternatively, the system could refrain from tiered limitations altogether and take its pound of flesh later via imposition of excise taxes on excess accumulations as formerly imposed under I.R.C. § 4980A.⁽¹⁶³⁾

  Conclusion

The implementation in 1997 of yet another tax-favored retirement plan specifically labeled SIMPLE, created false implications. In effect, Congress told employers that substantial reductions from prevailing funding possibilities must accompany a plan that truly avoids the complexity associated with standard retirement plans. Correspondingly, law makers told employees that simplicity alone would persuade large numbers of employers to broaden coverage opportunities otherwise unavailable to their work forces. Yet, employers remain free to deny access to tax-favored retirement savings altogether or engage in selective sponsorship of complex individually designed plans that simultaneously deny coverage to large numbers of workers, while permitting benefit accruals for some greatly in excess of the $6000 annual limitation available under SIMPLE. Either way, SIMPLE appears wanting. It has less power, in a tax sense, than competing plans and only adds to the overall complexity resulting from the many choices available for employers to choose selective benefit accruals while restricting coverage.⁽¹⁶⁴⁾

Offering ease of administration while forcing broadened coverage and greatly reduced contribution limitations, SIMPLE compares to a teacher who offers to reduce a student's rote homework in exchange for the student's tutoring other students, while accepting an automatically lower grade. The student will view the rote homework as a punishment rather than as a prerequisite for learning, those who need tutoring will not ultimately get it, and the teacher's awarding of grades will not correlate to achievement. Complexity in the administration of retirement plans should not be viewed as a punishment that is withheld only if an employer endures sacrifices, yet SIMPLE suggests as much.⁽¹⁶⁵⁾

The primary flaw in the current retirement savings system stems from the combined effect of trying to compel plan sponsoring employers to pay additional deferred compensation, while giving them near exclusive control over employee access to tax-favored retirement savings.⁽¹⁶⁶⁾ To the extent employers do allocate additional compensation to some workers as a result of plan sponsorship, economic distortions occur if such allocations do not reflect either enhanced productivity or the exigencies of labor markets.⁽¹⁶⁷⁾ If labor market realities support an economic theory that total compensation (direct and deferred) available to a work force cannot artificially expand as a result of government mandates or incentives, then letting employees, and not employers, determine the amount of tax-favored deferred compensation employees receive will not diminish the overall compensatory benefit paid to a work force.⁽¹⁶⁸⁾

Expanding the self-funded IRA scheme while restricting or abolishing other defined contribution plans would place no hardship on employees even if employer contributions otherwise paid under conventional employer-sponsored retirement plans ceased. Employees whose high productivity or unique skills would let them command direct compensations equal to their former combined direct and deferred compensations could replicate on their own, with substantial IRA contributions, their former employer contributions to qualified retirement plans. As a bonus, these employees would gain complete economic freedom over their retirement income security. Employees who would not earn direct compensation equal to their former direct and deferred compensation absent conventional plan contributions still would have the opportunity to carve out for themselves some measure of tax-favored retirement savings. Their employers would correspondingly acquire the economic freedom to redeploy any presumed compensation differential in some manner that benefits both employers and the work force overall. For example, the differential could remanifest as another benefit in the compensation package (like health care), as additional compensation awarded to more productive segments of the work force, or even as capital investment that ultimately results in the hiring of additional employees.⁽¹⁶⁹⁾

The recent popularity of cash or deferred arrangements points to an overall trend toward employee self-reliance. Many American workers now understand that all employees are entrepreneurs in a sense and must count on skill development and job change possibilities to build a lasting career. Few employees now expect a particular employer to retain them until retirement. Despite these notions, retirement plan funding still operates under the substantial control of employers. Young employees especially, who face the harsh new realities of career development in a global economy, ought to bristle against the effective parentalism (employers and government have become co-parents) that still serves as the basic premise for the American retirement plan system. Many employees apprehensively wonder whether full Social Security benefits will exist when their turn to retire arrives. Perhaps the time has come to empower all employees to make retirement savings decisions for themselves under a universal, expanded IRA system--a truly simple and effective approach to retirement income security.

* Professor of Law, The University of Akron School of Law; A.B., 1970, Oberlin College; J.D., 1974, Harvard Law School.

1. See Small Business Job Protection Act of 1996, Pub. L. No. 104-188, § 1421, 110 Stat. 1755, 1792-95 (codified at I.R.C. § 408(p) (West Supp. 1997)).

2. I.R.C. § 408(p)(2)(C)(i)(I) (West Supp. 1997).

3. See id. § 408(p)(2)(A)(i); see also I.R.C. § 401(k)(1)-(2) (1994).

4. For a description of the lucrative benefits resulting from tax-favored retirement plans, see infra text accompanying notes 82-88.

5. See I.R.C. § 408(p)(2)(D) (West Supp. 1997).

6. See I.R.C. § 408(k) (1994). This section allows SARSEPs existing on December 31, 1996, to continue as tax-favored retirement plans, but prevents the creation of new SARSEPs after that date. See I.R.C. § 408(k)(6)(H) (West Supp. 1997).

7. Compare I.R.C. § 408(k)(1), (6) (1994) with I.R.C. § 408(p)(1)-(2) (West Supp. 1997). IRAs have multiple functions in the tax-favored retirement plan scheme. In addition to serving as funding vehicles for SIMPLE, SARSEP, and (non-salary reduction) Simplified Employee Pension (SEP) arrangements, they serve as repositories for direct individual retirement savings of up to $2000 per year, see I.R.C. § 408(a)(1) (1994), and "rollover[s]" of funds from other tax-favored retirement plans without a dollar limitation in order to preserve income recognition deferral following distributions. Id.; see also I.R.C. § 408(d)(3) (1994). Their existing widespread use, familiarity, and simplicity makes them an ideal choice for a universal retirement plan replacement system as discussed further in this Article. See infra Parts V-VI.

8. See Small Business Job Protection Act of 1996, H.R. Conf. Rep. No. 104-737, at 235-36 (1996), reprinted in 1996 U.S.S.C.A.N. 1677, 1727-28.

9. See I.R.C. § 408(k)(6)(B) (1994).

10. See id. § 408(k)(2). These participation requirements allow sponsoring employers to exclude employees for five independent reasons that relate to minimum age attainment, minimum service performance, minimum compensation earned, inclusion under a collective bargaining agreement, and status as a nonresident alien receiving no income from the employer within the United States. See id.; see also I.R.C. § 410(b)(3) (1994). In some situations, these exclusions allow an employer to avoid contribution costs for the majority of its employees.

11. See I.R.C. § 408(k)(6)(A)(ii) (1994).

12. See id. § 408(k)(6)(A)(iv). This provision incorporates by reference I.R.C. § 401(a)(30) (1994), which incorporates by reference I.R.C. § 402(g)(1) (1994), which sets a $7000 limit for nonrecognized elective deferrals. See id. The amount was increased to $10,000 in 1998 by a cost-of-living adjustment authorized in § 402(g)(5). See I.R.S. Notice 97-58, 1997-45 I.R.B. 7. Section 408(k)(6) does not require employers to make matching contributions.

13. See I.R.C. § 408(k)(6)(A)(iii) (1994). The annual deferral percentage of each highly compensated employee eligible to participate cannot exceed 1.25 times the average deferral percentage of rank and file employees eligible to participate. See id. Thus, employers sometimes make matching contributions to encourage rank and file employees to elect deferrals sufficient to permit highly compensated employees to maximize their elective deferrals under this rule.

14. See I.R.C. § 408(p)(2)(C)(i)(I) (West Supp. 1997); see also supra note 2 and accompanying text.

15. See I.R.C. § 408(p)(4)(A)(i) (West Supp. 1997).

16. See id. § 408(p)(2)(A)(ii).

17. Section 408(k)(3) subjects SEPs to the general requirement that contributions may not discriminate in favor of highly compensated employees. This nondiscrimination concept originates in § 401(a)(4) and applies to standard tax-favored deferred compensation arrangements that take a variety of forms, such as profit-sharing plans, stock bonus plans, defined benefit plans, and others qualified under § 401(a). Section 408(k)(1)(B) also requires SEPs to meet the top-heavy rule of I.R.C. § 416(c)(2) (1994), itself a type of economic nondiscrimination stricture that mandates a contribution of at least three percent of compensation for "non-key employee[s]" when accrued benefits under a plan inordinately favor "key employee[s]." I.R.C. § 416(c)(2)(A)-(B) (1994). These complex nondiscrimination rules do not appear in § 408(p) as part of the SIMPLE scheme, nor do SIMPLE plans involve deferral percentage strictures that reduce elective deferrals for highly compensated employees. See supra note 13.

18. See I.R.C. § 408(p)(2)(A)(ii)-(iii), (2)(C)(ii)(I) (West Supp. 1997). Two somewhat complex exceptions in § 408(p)(2)(B) and (2)(C)(ii)(II) allow an employer to substitute two percent nonelective contributions or reduce the three percent limitation on employer matching contributions to just one percent for selected years. Deciding when and how to make these elections complicates planning choices for employers who might reasonably not expect such variations from the simplicity theme that supposedly characterizes SIMPLE arrangements.

19. In addition, a subtle tradeoff results from the difference in participation requirements between SIMPLE and SARSEP plans. A comparison of § 408(k)(2) and § 408(p)(4) (West Supp. 1997), given a single workforce, would likely produce two different numbers of employees eligible to participate. For plans involving matching employer contributions, this difference represents a cost differential to the employer.

20. Mandatory plan sponsorship exists only under the Social Security system, collective bargaining agreements involving union representation, and private employment contracts. Otherwise, the decision to sponsor a tax-favored retirement plan under either § 401 or § 408 belongs solely to the employer. Individual employees who lack bargaining leverage to determine key terms and conditions of their employment have no more control over their employers in obtaining tax-favored deferred compensation than in the determination of their direct compensation.

21. Once an employer decides to adopt a tax-favored retirement plan, participation and nondiscrimination rules, like those mentioned previously, tend to mandate coverage and funding for a broad segment of the workforce. See supra notes 13, 17, 19 and accompanying text. Unless employers can control coverage and contribution costs within these kinds of rules, they have little incentive to implement a retirement plan.

22. See I.R.C. § 414(q)(1) (West Supp. 1997) (defining a "`highly compensated employee'" as either a person who owns more than five percent of the employer sponsoring a plan or an employee earning annual compensation in excess of $80,000). Employers have many options for compensating such employees, including tax-favored stock awards, see I.R.C. §§ 421-24 (1994), and nonqualified deferred compensation. See Rev. Rul. 60-31, 1960-1 C.B. 174. Rewarding highly compensated employees through participation in a tax-favored retirement plan, however, results in potential coverage and nondiscrimination compliance costs that can cause the employer to reject formal retirement arrangements in favor of more flexible and less costly devices.

23. One commentator likens Social Security to a pyramid scheme that generously pays older recipients more than they contributed by taking the difference from younger recipients, contrary to the original conception of Social Security as minimal insurance against destitution (which leaves to individual responsibility the task of providing a complete and comfortable pension). See Ronald H. Marcks, Social Security's Most Basic Infirmity, Wall St. J., Mar. 12, 1997, at A18.

24. Employee Retirement Income Security Act of 1974, Pub. L. No. 93-406, 88 Stat. 829 (codified as amended in scattered sections of 26 U.S.C.).

25. See I.R.C. § 401(a) (1994).

26. See id. § 401(a)(3)-(4), (7).

27. See supra note 22.

28. An employer sponsoring a § 401(a) plan has wide discretion in choosing how many employees to exclude from participation if the employer relies on the various possible categories of exclusion under I.R.C. § 410(a)(1), (b)(1) (1994).

29. See Elizabeth MacDonald, Pension War Breaks Out at Accounting Giant, Wall St. J., Sept. 23, 1997, at B1, for a fascinating example of how an employer's adoption of a retirement plan can cause inordinate difficulties. Deloitte & Touche, one of the world's top pension consulting firms, sponsored a "`cash balance'" pension plan for its own employees, but learned too late that the plan had unintended adverse consequences for some of the employees covered, who could have fared much better under a predecessor plan. Id. at B1, B10. Even pension experts lament the complexity of qualified pension arrangements. See id. That many smaller employers shun sponsoring tax-favored retirement plans should come as no surprise.

30. Aside from the participation exclusions available in § 410, see supra note 28, the plan qualification scheme permits employers to implement plans that do not require specific contribution levels, such as profit sharing or stock bonus plans. Additionally, employers can reduce benefits or contributions with permitted disparities that integrate a plan with the Social Security system under § 401(a)(5)(C). Relative to contributions made on behalf of older employees, employers can reduce contributions for younger employees in plans that use actuarial funding concepts. See infra note 103.

31. See I.R.C. § 408(a)-(e) (1994).

32. See I.R.C. § 408(a)(1) (Supp. IV 1974) (current version at I.R.C. § 408(a)(1) (1994)). Many taxpayers cannot take deductions for contributions to self-funded IRAs. See infra note 45.

33. See I.R.C. § 408(a)(1) (1994).

34. Id. § 402(c)(1), (c)(8)(B)(i).

35. See id. § 408(j).

36. See Revenue Act of 1978, Pub. L. No. 95-600, § 152, 92 Stat. 2763, 2797-800 (codified as amended at I.R.C. § 408(k) (1994)).

37. See supra note 6 and accompanying text.

38. See I.R.C. § 408(k)(2) (1994).

39. See supra note 3 and accompanying text.

40. See I.R.C. § 401(k) (1994). Section 401(k) is one of those rare technical provisions of federal tax law having such widespread impact that the Code citation itself has passed into general parlance. Many persons seeking employment understand the significance of a prospective employer's offering a "401(k)" plan as part of a compensation package. These workers know that such a plan requires employee contributions and thus represents a personal opportunity to save for retirement in a tax-favored manner.

41. Tax law complexity usually assures some expense for technical advice. Risk in the implementation of a tax-favored retirement plan results from a range of possibilities that include daily determined penalties for failure to timely file a plan's I.R.S. Form 5500, loss of the employer's deduction for contributions under the complex strictures of I.R.C. § 404 (1994) when the same payments made as direct compensation would easily be deductible under I.R.C. § 162 (1994), and exposure to claims by employees or their beneficiaries if plan qualification ceases or the plan fiduciaries fail to administer it correctly.

42. See I.R.C. § 401(k)(3)(A)(ii) (1994). This section contains rules similar to those previously mentioned regarding deferral percentage relationships in SARSEP plans. See id.; see also supra note 13.

43. Without sufficient matching contributions to induce a high level of participation by rank and file employees, highly compensated employees often must take back their contributions to a § 401(k) plan under the "excess contributions" rules of § 401(k)(8).

44. See I.R.C. §§ 457, 501 (1994). Employees of § 501(c)(3) organizations or public schools can participate in a special form of cash or deferred plan by virtue of I.R.C. § 403(b) (1994), and government employees have their own version under § 457.

45. See supra note 33 and accompanying text. I.R.C. § 219(g)(1) (1994) phases out the deductibility of IRA contributions for taxpayers who exceed minimum levels of adjusted gross income if an individual or the individual's spouse is an "active participant" in any designated tax-favored retirement plan.

46. See supra note 33 and accompanying text. The employees previously mentioned, see supra note 44, have the advantage of making special "catch-up" contributions if full contributions have not been made in earlier years. Treas. Reg. § 1.457-2(f) (1982); see also Treas. Reg. § 1.403(b)-1(d)(1) (as amended in 1986).

47. In each case, the spouses can make two $2000 contributions per year to their (potentially deductible) IRAs, since § 219(f)(2) permits separate computation of each spouse's deductible amount and only one spouse need have earned compensation for the year under § 219(c)(1).

48. Employers who might sponsor plans qualified under § 401(a) must pay particular attention to these considerations. The instructions accompanying I.R.S. Form 5500, the annual return for such employee benefit plans, state that the estimated time for preparing the form, learning about the law or the form, and keeping necessary records totals nearly 105 hours. Of course, complete administration of standard retirement plans, including investment analysis, claims dispositions, and interpretive matters, involves time commitments that far exceed mere preparation of the annual reporting form.

49. See supra note 18 and accompanying text. An employer can mitigate the blank check consequence by electing to substitute two percent nonelective contributions for matching contributions as authorized under § 408(p)(2)(B) (West Supp. 1997), but this election creates mandatory funding for all participants, even if some make no elective contributions on their own behalf.

50. Employers still scramble to figure out the compliance details of such recent manifestations of this trend as the Family and Medical Leave Act of 1993, Pub. L. No. 103-3, 107 Stat. 6 (codified at 29 U.S.C. §§ 2601-54 (1994)), and the Americans with Disabilities Act of 1990, Pub. L. No. 101-336, 104 Stat. 327 (codified as amended at 42 U.S.C. §§ 12101-213 (1994)).

51. See supra note 16 and accompanying text. Many employer-sponsored defined contribution retirement plans can receive contributions per participant of up to $30,000 per year, the dollar limitation imposed as a qualification requirement by I.R.C. § 415(c)(1)(A) (1994). The percentage limitation of § 415(c)(1)(B) assures that only participants earning at least $120,000 per year can benefit from the $30,000 maximum, but even employees earning far less can receive contribution allocations of as much as 25% of their compensations.

52. See Small Business Job Protection Act of 1996, Pub. L. No. 104-188, § 1421, 110 Stat. 1755, 1792-95 (codified at I.R.C. § 408(p) (West Supp. 1997)).

53. See I.R.C. § 408(p)(5)(C) (West Supp. 1997). Each participant in a SIMPLE arrangement must choose to make or change elective deferrals during the 60-day period before the beginning of a year. See id.

54. See I.R.S. Announcement 96-112, 1996-45 I.R.B. 7.

55. See I.R.C. § 408(p)(5)(A) (West Supp. 1997).

56. Employee Retirement Income Security Act of 1974, Pub. L. No. 93-406, § 4, 88 Stat. 829, 839-40 (codified as amended at 29 U.S.C. § 1003 (1994)).

57. See 29 C.F.R. § 2510.3-102(b) (1997).

58. See Labor Department Relaxes Deposit Time Limit for SIMPLE Plans, CCH Fed. Tax Wkly. (CCH Inc., Chicago, IL.), Nov. 27, 1996, at 565, 565.

59. See I.R.S. News Release IR-96-46 (Oct. 31, 1996).

60. The instructions to Form 5305-SIMPLE estimate that a taxpayer needs 6 hours and 51 minutes to complete the form. As with all such time estimates made by the I.R.S., the precision of the estimate belies a dubious assumption that the planning and compliance tasks underlying a form are uniformly manageable and relatively uncomplicated. Yet, the instructions to Form 5305-SIMPLE contain (within just three pages) numerous references to various provisions of the Internal Revenue Code and I.R.S. publications that would certainly confound any layperson who attempted to read them. Presumably, the 6 hour and 51 minute estimate only applies to persons who already have a professional familiarity with the complex rules incorporated by reference into the form's instructions.

61. See I.R.S. Notice 97-6, 1997-2 I.R.B. 26. This notice addresses these issues, but their inherent complexity suggests that further interpretations will follow as related sub-issues come to the attention of the I.R.S. Indeed, the Taxpayer Relief Act of 1997, Pub. L. No. 105-34, § 1601, 111 Stat. 788, 1086 (to be codified at 26 U.S.C.), contained nine additional technical interpretations affecting SIMPLE arrangements just one year following their legislative creation.

62. See generally Rev. Proc. 97-9, 1997-2 I.R.B. 55.

63. See I.R.C. § 401(k)(3) (1994).

64. See I.R.C. § 408(p)(2)(A)(ii) (West Supp. 1997).

65. For example, if rank and file participants in a particular year make elective deferrals that average only one percent of their compensations, highly compensated employees participating in the same plan cannot defer more than two percent of their compensations. Since § 401(a)(17) sets a limit of $160,000 (as of 1998) for the annual compensation of any employee taken into account under a qualified retirement plan, the elective deferrals of the highly compensated employees could not exceed an average of $3200.

66. See I.R.C. § 408(k)(6)(A)(iv) (1994); see also supra note 12 and accompanying text.

67. Even lower-paid employees seriously consider making elective deferrals once they get sufficiently close to retirement ages. Some work forces produce high levels of elective deferrals by rank and file participants simply because so many workers earn relatively high incomes (sustaining interest in tax-favored savings) without technically becoming highly compensated employees as defined in § 414(q). Many Silicon Valley employers have such work forces. Note that under § 414(q)(1)(B)(ii) (West Supp. 1997), an employer can elect to exclude from highly compensated employee status all employees who earn more than $80,000 per year and who also do not fit into the group consisting of the top 20% of employees when ranked on the basis of compensation paid during a year.

68. See I.R.C. § 401(k)(11)(B)(i) (West Supp. 1997).

69. See supra note 18 and accompanying text.

70. Compare the three percent maximum matching contribution required under the SIMPLE scheme by virtue of § 401(k)(11)(B)(i)(II) (West Supp. 1997), with a dollar-for-dollar match frequently available in a § 401(k) plan. To permit highly compensated employees averaging earnings of $100,000 per year to defer a full $10,000 each, rank and file deferrals would have to average at least eight percent of compensation (ten percent divided by 1.25) as required by § 401(k)(3)(A)(ii)(I). The employer matching contributions for rank and file participants would correspondingly amount to eight percent. Even a fifty-cents-for-a-dollar match would involve substantially more funding by the employer, especially in view of the option to make two percent nonelective contributions in situations where such election would minimize employer funding. See I.R.C. § 408(k)(11)(B)(ii) (West Supp. 1997).

71. See supra notes 16, 18 and accompanying text. Electing the SIMPLE option for a § 401(k) plan leads to an economic analysis to determine whether the employer will spend less by choosing the (up to) three percent matching contribution rule of § 401(k)(11)(B)(i)(II) (West Supp. 1997), or the fixed two percent nonelective contribution rule of § 401(k)(11)(B)(ii) (West Supp. 1997). Electing a SIMPLE plan, rather than a standard § 401(k) plan (without the SIMPLE option), complicates the analysis regarding the two percent nonelective contribution election, since the three percent matching contribution rule described in § 408(p)(2)(C)(ii) (West Supp. 1997) contains a sub-election allowing an employer to cap the matching contribution at only one percent for two out of five years.

72. See supra note 12 and accompanying text.

73. See supra notes 16-18 and accompanying text.

74. See supra note 17 and accompanying text.

75. A quick review of the table of contents for the complicated qualified plan nondiscrimination rules reveals the relatively recent regulatory acceptance of such esoteric retirement devices as "cash balance plans" and "floor-offset arrangements." Treas. Reg. § 1.401(a)(4)-0 (as amended in 1993).

76. See I.R.C. § 408(p)(2)(D) (West Supp. 1997); see also supra note 5 and accompanying text.

77. See supra note 51.

78. Compare SIMPLE plans with § 401(a) plans by making computations based on an assumed interest rate to illustrate the differing lump sums available after twenty or thirty years depending on whether the annual contribution amount is $6000 or $30,000. The difference emphatically quantifies the reward for tackling the tax complexity associated with the § 401(a) plans.

79. The economic difference mentioned in the preceding note will sometimes accompany substantial offsets in the form of funding costs for nonhighly compensated employees necessitated by the participation coverage rules of § 410 (incorporated by reference as a special kind of nondiscrimination rule in § 401(a)(3)). Unfortunately, the precise effect of § 401(a) on funding costs varies greatly depending on the exact composition of an employer's workforce and, most troubling, the skill and risk tolerance of the advisers who interpret and apply the complex facets of § 401(a) for an employer.

80. In some cases, terminating a plan in favor of implementing an alternative arrangement results in disqualification of the plan under a permanency requirement read into the § 401(a) qualification scheme. See Treas. Reg. § 1.401-1(b)(2) (as amended in 1976).

81. See Adopting SIMPLE Retirement Plan May Not Be Simple Decision For Small Businesses, CCH Fed. Tax Wkly. (CCH Inc., Chicago, IL.), Sept. 12, 1996, at 437, 437-38.

82. Many who work with retirement plans use the term qualified to describe plans that meet the requirements of § 401(a) and do not regard arrangements meeting the requirements of § 408 as technically qualified. Because plans under both § 401(a) and § 408 enjoy the same basic taxation features, however, some tax advisers refer to plans qualified under § 401(a) while also referencing plans qualified under § 408. This usage emphasizes that each kind of plan must meet a set of codified requirements (although very different sets of requirements) in order to achieve the same taxation results. This Article uses the phrase tax-favored to avoid the quibble over exactly which plans should be regarded as qualified.

83. See I.R.C. § 404 (1994). Because actuarial principles apply to defined benefit plans and result in highly variable funding levels over time, the deduction limitation for such plans in § 404(a)(1) represents one of the least comprehensible allowances contained in the Internal Revenue Code.

84. See I.R.C. § 162 (1994). The basic obstruction for deductibility of direct compensation is the "reasonableness" requirement of § 162(a)(1). Section 404 is much more complex, not just because of its own rules, but because § 404(a) indirectly incorporates § 162 as a preliminary requirement (contributions must "otherwise be deductible") for deducting an employer's retirement plan contributions.

85. See I.R.C. § 402(a) (1994).

86. See id. § 501(a). This section makes the trust fund tax exempt, avoiding entity taxation on earnings ultimately distributed. See id.

87. See id. § 3121(a)(5).

88. In many instances, a retiring distributee has a lower marginal tax rate bracket than while working. Additionally, distributees often realize tax savings from rollovers, see id. § 402(c), income averaging for lump sum distributions, see id. § 402(d), and extended deferral of income recognition respecting "net unrealized appreciation" in employer securities included in a distribution. Id. § 402(e)(4).

89. The viewpoint implicit in the concept of a tax subsidy creates some controversy, primarily because the concept tends to ignore the fundamental source of the money presumed granted as a government subsidy under specific tax provisions. Perhaps less controversy results when analysts regard tax allowances as incentives to mitigate the economic detriment of progressive governmental takings. These allowances thus permit categories of taxpayers some control over their effective rates of taxation.

90. See supra note 22.

91. See generally Treas. Reg. § 1.401(a)(4) (as amended in 1993).

92. See I.R.C. § 401(a)(13) (1994).

93. See id. § 401(a)(11).

94. See id. § 401(a)(1)-(2).

95. See supra note 80. Disqualification of a plan is not the only possible sanction for rules violations. Transgressions of qualification rules that also constitute "prohibited transactions" under I.R.C. § 4975 (1994) subject the involved parties to penalty tax assessments. The possibilities for inadvertent disqualifications and prohibited transactions create a level of risk that discourage employers from implementing tax-favored retirement plans in general, to the great detriment of employees who desire retirement savings.

96. See supra note 79. The comment therein, respecting the compliance implications of § 410, applies equally to the interpretive difficulties faced by sponsoring employers under the recondite and voluminous nondiscrimination rules of Treas. Reg. § 1.401(a)(4) (as amended in 1993).

97. For example, the employer can place numerous part-time employees outside plan coverage by using the 1000 hours "year of service" exclusion of § 410(d)(1)(A)(ii), (d)(3)(A), while simultaneously excluding large numbers of full-time employees on the basis of such factors as job classifications, functional or geographic divisions, and status as salaried or blue collar workers--as permitted under § 410(b)(1).

98. Independent contractors cannot participate in tax-favored retirement plans sponsored by a business for whom they work, since they are not "employees" as clearly required in the preamble language of § 401(a) and similar references in § 408(k) and § 408(p) (West Supp. 1997).

99. See I.R.C. § 414(n) (1994).

100. See id. § 414(b)-(c), (m).

101. Id. § 410(b)(5)(A).

102. See id. § 401(a)(5)(C), (a)(l).

103. Age-weighted defined contribution plans in which employers skew contributions beyond compensation proportionality in favor or older participants (who frequently include many highly compensated employees) receive authorization under the "cross-testing" rules of Treas. Reg. § 1.401(a)(4)-8 (as amended in 1993), which permit testing of such plans for nondiscrimination pursuant to actuarial principles that otherwise produce higher funding for older participants in defined benefit plans.

104. See I.R.C. § 401(a)(5)(B) (1994).

105. See id. § 411. Section 411 permits an employer either to reduce its contributions or enhance the accrued benefits of participants without enhancing contributions by allocating vesting forfeitures respecting participants who terminate service prematurely. Many highly compensated employees stay with their employers long enough to become fully vested and thus can disproportionately benefit from vesting forfeitures when they are used to augment accrued benefits.

106. See Tax Reform Act of 1986, Pub. L. No. 99-514, 100 Stat. 2085 (codified as amended in scattered sections of 26 U.S.C.). Congress specifically killed the tax shelter industry by enacting the passive activity limitations of I.R.C. § 469 (1994). Those who work with § 469 might agree that the death of tax shelters did not entirely mitigate the tax complexity associated with remunerative loss deductions. As set forth ahead, Congress could eliminate the complexity associated with tax-favored plans much more cleanly by expanding the existing scheme for self-funded IRAs. See infra Parts V-VI.

107. See supra notes 2-5, 14-18 and accomanying text.

108. Most owner-employees who start closely-held businesses would love to make substantial and recurring tax-favored retirement contributions on their own behalf, while avoiding substantial additional deferred compensation costs respecting their rank and file employees. Due to the volume and complexity of the § 401(a) rules, two small businesses in the same industry and having the same revenues could easily experience vastly different retirement plan funding costs depending on their respective choices of plans and the skill/aggressiveness of their professional advisers. In any event, such large economic differences often have little to do with the desires of common employees respecting their retirement income security. A plan with high funding for rank and file employees might involve a relatively young workforce that would much rather see more take-home pay than deferred compensation, and a plan with low funding for its common employees might involve an older workforce that desperately desires tax-favored retirement savings.

109. The manner in which an employer creates such offsets must not violate section 510 of the Employee Retirement Income Security Act of 1974, Pub. L. No. 93-406, 88 Stat. 829, 895 (codified at 29 U.S.C. § 1140 (1994)), which prohibits disciplinary, retaliatory, or discriminatory acts against employees who exercise their rights under an employee benefit plan and pension law. Offsets in contemplation of contributions to tax-favored retirement plans otherwise yield only to individual employment contracts, collectively bargained agreements, or market forces.

110. Even if offsets allow some employers to avoid substantial funding costs in the implementation of tax-favored plans, these employers might still resent having to engineer offsets that interfere with their flexibility to effect efficient business decisions respecting resource allocations made for optimum productivity within the organization.

111. The possibility for plan implementation as a result of collective bargaining by union representatives supports a participation exclusion for qualified retirement plans under § 410(b)(3)(A). Thus, many private employers exclude union employees from participation in plans covering non-union employees. Of course, individual union members cannot decide for themselves whether they will participate in a union-only plan, since the union membership at large decides whether to forego retirement benefits in favor of negotiating enhancements in direct compensation, other employee benefits, and working conditions.

112. Without increasing the basic $2000 contribution limitation, the federal legislature tinkered with IRAs in the Taxpayer Relief Act of 1997, Pub. L. No. 105-34, § 302, 111 Stat. 788, 825-28 (to be codified at I.R.C. § 408A). Most notably, Congress created I.R.C. § 408A, which "backloads" tax advantages in a "Roth IRA" by letting taxpayers withdraw investment build-up without income taxation depending on how and when distributions occur.

113. Resistance to employer mandates reached a dramatic level in 1993, when President Clinton's massive health care proposal failed to result in Congressional approval, even though the President's party then had control of Congress.

114. See I.R.C. § 401(k) (1994).

115. See id. § 415(c)(1)(A); see also supra note 51.

116. Perhaps existing plans could be grandfathered and contributions to such plans could serve as an offset against the annual contribution limitation of a universal IRA program. Whether only defined contribution plans should be abolished, leaving employers free to implement defined benefit plans, presents an ancillary issue involving policy and technical considerations not here considered. Indeed, if an effective offset system were constructed, abolishment of qualified plans might not be necessary at all, although many private employers would be tempted to terminate existing plans to avoid administrative burdens fundamentally redundant in light of employee rights to self-fund substantial pension benefits.

117. See I.R.C. § 403(b)(2)(A) (1994); see also supra note 46.

118. The next generation of professional workers would become more capable of initiating sustained retirement savings for themselves, and a cycle or culture of saving and retirement planning would emerge.

119. In other words, financial independence among older professional workers indirectly benefits everybody. When the most talented workers do not get locked into their current employment, their talents can more efficiently serve the economy. Those concerned about pension portability have long addressed this mobility question in a similar context involving primarily younger workers. The best solution for any worker facing multiple job changes suggests transfer of control over the pension-building process from employers to individual employees.

120. See I.R.C. § 72(t) (1994). This section, which penalizes takers of premature retirement plan distributions by imposing a penalty tax beyond standard income taxation, indirectly promotes capital stability and enhances the accumulation of retirement savings until needed following the cessation of remunerative activity. See id.

121. Employees whose employers do not sponsor a retirement plan find the $2000 limitation of § 408(a)(1) particularly burdensome as they approach retirement age and cannot make up contributions foregone in earlier years when retirement planning seemed a much less urgent concern.

122. The role model effect of highly paid employees no doubt varies considerably depending on the employment culture involved. Consider, however, whether retirement plan innovations, like employee stock ownership plans, see generally I.R.C. § 409 (1994), would have attracted sufficient interest by rank and file employees without the instructive example of executive stock ownership in employer securities occasioned by the stock incentive awards and direct entrepeneurial activity.

123. See I.R.C. § 163(h) (1994) (respecting the home mortgage interest expense deduction).

124. A manifestation of such criticism takes the form of a $1,000,000 limit on the size of a home mortgage that yields a deductible interest expense. See I.R.C. § 163(h)(3)(B)(ii) (1994).

125. The Employee Retirement Income Security Act of 1974, Pub. L. No. 93-406, 88 Stat. 829 (codified as amended in scattered sections of 29 U.S.C.), contains four distinct titles, only one of which, Title II, involves amendments to the Internal Revenue Code.

126. See supra note 46 and accompanying text.

127. Notwithstanding widely reported trends that large numbers of adult children return home to live with their parents (or never leave), older workers facing protracted family responsibilities still will want to think seriously about retirement even if circumstances postpone it.

128. The savings ratios under assumed rates of return on investment will frequently exceed two to six or two to ten, given the possibilities for matching employer contributions under both kinds of cash or deferred arrangements. This suggests that the $6000/$10,000 elective deferral limitations under these arrangements would not adequately serve the retirement savings needs of taxpayers under an expanded, universal IRA program designed to grant unfettered retirement savings access to all workers. Allowing employers to continue matching contributions voluntarily would help the situation, but increasing the deferral limitation would permit every employee to achieve substantial accumulations.

129. Likewise, if the worker contributed less than the deferral limitation, the final deferral years' limitation could expand to account for the amount of shortfall from the prior years.

130. If the § 415 dollar limitation were used as the ultimate contribution limitation under a revised IRA program, the percentage limitation (25% of compensation) of § 415(c)(1)(B) would inappropriately prevent many rank and file employees from sufficiently realizing the intended benefits of a catch-up feature. Accordingly, the 25% limitation probably does not belong in a universal IRA scheme.

131. Since many employees rise though the ranks and frequently spend years as nonhighly compensated employees before achieving the highly compensated employee status as defined in § 414(q), a phase out of a catch-up feature should take into account the length of time the worker held the highly compensated employee status.

132. Congress created a precedent for limiting the tax benefits of retirement plans according to the size of a taxpayer's accrued benefit by enacting I.R.C. § 4980A(a) (1994), which imposed an additional 15% tax on plan distributions or accumulations in excess of specified limits. This provision was repealed by the Taxpayer Relief Act of 1997, Pub. L. No. 105-34, § 1073, 111 Stat. 788, 948-49 (to be codified at 26 U.S.C.).

133. Those who favor the use of taxation systems to effect wealth redistribution might like this idea: Highly compensated employees would be allowed to contribute fully to the tax-favored savings program, but excess accumulations would invoke a separate distributions tax as was imposed under § 4980A, see supra note 132, and the revenue thus collected would directly fund starter IRAs for lower paid workers who have difficulty saving for retirement. No doubt the details for implementing this idea would create an interesting political exercise.

134. Older readers might recall the effective publicity campaign of the federal government in selling treasury bonds during World War II.

135. See I.R.C. § 408(a)(2) (1994) (requiring that an IRA have an institutional trustee).

136. Many statutory pension arrangements involving teachers, police officers, and various other state and local government workers mandate employee contributions. These workers tend to have very good accrued benefits compared to their privately-employed counterparts.

137. See I.R.C. §§ 3101(a), 3111(a) (1994). The 12.4% rate counts both the employee's and employer's contributions to the Social Security pension system, excluding Medicare contributions. See id.

138. Federal tax regulation of qualified retirement plans implicitly recognizes the Social Security retirement age by adjusting the defined benefit plan limitation of § 415(b) downward when private pensions commence prior to the Social Security retirement age. See I.R.C. § 415(b)(2)(C) (1994).

139. Congress recognized the extended working lives of older workers in section 1404(a) of the Small Business Job Protection Act of 1996, Pub. L. No. 104-188, 110 Stat. 1755, 1791 (codified at I.R.C. § 401 (West Supp. 1997)), which amended § 401(a)(9)(C) to extend the required beginning dates for qualified retirement plan distributions beyond age 70½ for most plan participants still employed after attaining that age.

140. Recipients of Social Security benefits who have not attained age 70 have their benefits reduced if they continue to earn income above designated amounts. See 20 C.F.R. § 404.415(a) (as amended in 1986).

141. Not every highly compensated person, given the choice, will direct earnings into a tax-favored retirement plan. Some wish to maximize their after-tax income for reinvestment into unincorporated closely-held businesses or refrain from taking more compensation than necessary for living expenses while maximizing retained earnings within an incorporated business. Their retirement plan is the business itself, which they can later sell at a gain subject to a favorable capital gains rate available under I.R.C. § 1(h) (1994).

142. Not only do employers decide whether their workforces will have access to an employer-sponsored plan, but once an employer does decide to initiate a plan, it can exercise a number of ancillary choices that determine not just basic features like participation, allocation, and vesting criteria, but as well as important details that greatly convenience or inconvenience the plan's participants. These details appear in provisions that determine when and how distributions will be made, whether participants can borrow against their accrued benefits, whether life insurance will constitute a plan benefit, how plan assets will be invested, and whether accrued benefits will be built fundamentally from employer contributions, employee contributions, or some combination of both.

143. See I.R.C. § 32 (1994).

144. The earned income credit of § 32 appears in the Internal Revenue Code under Subpart C, "Refundable Credits," of Part IV, "Credits Against Tax," of I.R.C. Subtitle A, Chapter 1, Subchapter A (1994). As a "refundable credit," even filers who owe no income tax get a tax refund of at least the credit amount, making the earned income credit the equivalent of a "negative income tax," or "transfer payment."

145. A young worker with young dependents would less likely want or need set asides for retirement income security. The earned income tax credit also benefits low income individuals who have no children and who have attained age 25, but not yet attained age 65. See I.R.C. § 32(c)(1)(A)(ii) (1994).

146. See I.R.C. § 72(t)(6) (West Supp. 1997). Congress increased the penalty for early distribution from 10% to 25% during the first two years of participation in a SIMPLE arrangement. See id.

147. The current early distribution penalty contains exceptions for annuity payouts, distributions used to pay for medical care, and other distributions made under special circumstances. See I.R.C. § 72(t)(2)(A)(iv), (B) (1994).

148. See I.R.C. §§ 72(p)(2)(A), 401(a)(13), 4975(d)(1) (1994). Section 72(p)(2)(A) permits participant loans of up to $50,000 without causing distributions taxation. Section 401(a)(13) permits the participant's accrued benefit to serve as security for a plan loan without violating the qualification stricture against assignment or alienation of benefits in favor of creditors. Section 4975(d)(1) exempts participant loans from prohibited transactions taxation if they meet designated conditions.

149. Both matching employer contributions and access to accrued benefits via loan features permit sponsors of § 401(k) plans to increase elective deferrals among rank and file workers, resulting in better compliance with the deferral percentage tests that otherwise restrict the elective deferrals of highly compensated employees under § 401(k)(3)(A)(ii).

150. Various interpretive problems arise under the technical loan rules discussed previously. See supra note 148 and accompanying text. For example, determining a "reasonable rate of interest" as required by § 4975(d)(1)(D) can cause difficulties for a plan administrator. Additionally, the plan administrator takes on many of the practical headaches of commercial lenders who must decide whether and when to foreclose on the security for a loan after guessing whether irregularities in payments will lead to a permanent default.

151. As a precedent, § 401(k)(2)(B)(i)(IV) allows distributions from § 401(k) plans upon "hardship" of the employee, defined in Treas. Reg. § 1.401(k)-1(d)(2)(iv) (as amended in 1994) to include expenses for medical care, purchase of a principal residence, and tuition payments. If some taxable distributions require a showing of hardship, nontaxable loans under similar strictures should not cause inordinate regulatory problems and might make loan provisions much more manageable for the institutional trustees of IRAs in the event the prohibitions against IRA loans of § 408(e) were lifted.

152. I.R.C. § 415(c)(1) (1994). The "annual addition," defined in § 415(c)(2), consists of "employer contributions, the employee contributions, and [vesting] forfeitures." Id. § 415(c)(2)(A)-(C). Section 408(a)(4) prohibits vesting forfeitures in IRAs.

153. For a discussion of catch-up features, see supra notes 126-33 and accompanying text.

154. Of course, contributions need not come directly from current taxable income. Prior savings, nontaxable gifts or other windfalls, and asset conversions can also serve to fund tax-favored retirement plans.

155. Section 415(a)(2), although imposing its limitations as a requirement for an IRA-based SEP under § 408(k), does not similarly reference SIMPLE arrangements under § 408(p).

156. Congress would unlikely abolish employer-sponsored deferred compensation arrangements, although law makers could remove the tax benefits of such plans or at least offset employer contributions against the universal scheme's contribution limitations. Eliminating tax favoritism for all plans except IRAs might not work since many existing plans would continue to hold substantial retirement savings pending future distributions. Additionally, many defined benefit plans for teachers and other public service workers involve mandatory employer and employee contributions under state law and often replace the Social Security system for these workers. Terminating tax benefits for public employee plans would probably lead to significant political consequences.

157. See Small Businesses Job Protection Act of 1996, Pub. L. No. 104-188, § 1452(a), 110 Stat. 1755, 1816 (codified at I.R.C. § 415(e) (West Supp. 1997) (becoming effective for years beginning after December 31, 1999).

158. The § 415(e) limitation reduces the maximum allowable accrued benefit for the defined benefit plan(s), defined contribution plan(s), or combination of plans of an employer adopting both kinds of plans. Plan administrators must make complicated computations in the form of a "defined benefit plan fraction" and a "defined contribution plan fraction." Treas. Reg. § 1.415-7 (1980).

159. See I.R.C. § 415(b)(1), (c)(1) (1994).

160. Compare I.R.C. § 402(g) (1994) (discussing the limitation on elective deferrals) with I.R.C. § 415(b)(1), (c)(1) (1994) (addressing the combined limitations). Note, however, that employers sponsoring combinations of plans still have limited contribution deductions under § 404(a)(7).

161. See I.R.S. Notice 97-58, 1997-45 I.R.B. 7.

162. Perhaps Congress could define meaningful benefit by designating a fixed dollar amount of annual benefit, as in § 415(b)(4), which excuses benefits that do not exceed $10,000 per year from the percentage of compensation limitation of § 415(b)(1)(B).

163. See supra note 133.

164. When comparing SIMPLE to other tax-favored plans, employers must guess whether the relative simplicity of § 408(p) arrangements outweighs the relative disadvantages of SIMPLE, such as its lower elective deferral limitation, see I.R.C. § 408(p)(2)(A)(ii) (West Supp. 1997), "only plan" requirement, Id. § 408(p)(2)(D), and restrictive coverage exclusions. See id. § 408(p)(4). Although employers can analyze the relative disadvantages of SIMPLE by identifying resulting quantitative effects, how do potential plan sponsors meaningfully evaluate the relative advantage of simplicity?

165. The punishment/sacrifice concepts appear most graphically in the § 401(k)(11) (West Supp. 1997) election to substitute the SIMPLE scheme for the deferral percentage tests of § 401(k)(3)(A), since the balance of § 401(k) applies regardless of whether the SIMPLE contribution format is adopted within a particular § 401(k) plan. Two § 401(k) plans similar in all respects except for the § 401(k)(11) election will illustrate the direct tradeoff between avoidance of the complicated and potentially disqualifying deferral percentage tests and a $4000 differential in maximum annual elective deferrals permitted for individual participants.

166. An employer having no obligation at all to implement a tax-favored retirement plan, upon deciding nonetheless to sponsor a plan qualified under § 401, must incur substantial funding and administrative costs determined under vague sets of rules arising under the Internal Revenue Code and its Treasury Regulations. See generally Treas. Reg. § 1.401 (as amended in 1997). SIMPLE arrangements eliminate rule vagueness but not indeterminacy of funding costs, since the employer has no control over employee elective deferrals, matching contributions, or future eligibility to participate under the respective strictures of § 408(p)(2)(A)(i), (iii), (p)(4) (West Supp. 1997).

167. Tax-favored retirement plans substantially contrast with nonqualified deferred compensation arrangements in this regard, since an employer can negotiate many varieties of nonqualified plans on an employee-by-employee basis taking into account only an employee's worth to the employer. Unfortunately, nonqualified plans achieve only income recognition deferral for the employee and not the other tax benefits mentioned previously. See supra text accompanying notes 82-88; see also Rev. Rul. 60-31, 1960-1 C.B. 174.

168. The collective bargaining process illustrates an effective application of employee self-determination in the formulation of overall wage and benefit packages. Unfortunately, the negotiating power resulting from collectivization necessitates a subordination of the desires and needs of employees who might otherwise construct individual benefit arrangements differently from the collectively determined result. A universal IRA replacement for most qualified plans would give all employees individual choice in the retirement income security process.

169. Let the economists, with their usually conflicting political vantage points, argue whether the differential would simply disappear into profits. Some would lament the loss of any governmental mandate to improve employee benefits, even if applied only to those employers who choose to sponsor a retirement plan. Other analysts would emphasize that only free markets can produce meaningful wage or benefit enhancements and that regulatory attempts to do so must produce offsets in the economy that affect overall productivity, to the detriment of all. Adam Smith might delight in observing that the "invisible hand" of a relatively free labor market ensures that the great majority of workers earn more than the mandated minimum wage. Employers cannot help but pay more if they want an efficient workforce, even if they are not forced by law to pay more than minimum wage. Does the "visible hand" of government regulation cause market distortions that can harm both profitability and compensation levels? Free market advocates would likely argue that forcing employers who adopt qualified retirement plans to observe the coverage and contribution edicts of § 401(a) regardless of which workers are thereby benefited results in displacements of compensatory funds that prevent additional hiring of, or higher direct wages for, more productive and skilled workers. These latter workers remain undercompensated or must work elsewhere, while the employer's opportunities to enhance economic efficiency become correspondingly limited.