Choice Architecture and the Locus of Fiduciary Obligation in Defined Contribution Plans

“On January 1, 2011, the oldest Baby Boomers [turned] 65. Every day for the next 19 years, about 10,000 more will cross that threshold. By 2030, when all Baby Boomers will have turned 65, fully 18% of the nation’s population will be at least that age.”1

I.     Introduction

Long-term wealth creation and retirement security for the much-discussed “99 percent”2 depends in large part on employer-sponsored plans that enable employees to save for their retirement. For many employees, their retirement-related savings accounts are their single largest asset—or their second largest asset after their home.3 Currently, Americans hold more than $3.4 trillion4 in their 401(k) plans.5 For perspective, that is the equivalent of 28% of the domestic equity market capitalization of the New York Stock Exchange.6

Despite the trillions of dollars held in these accounts, problems with 401(k) plans are apparent. Research in behavioral economics explains cognitive biases that lead to flawed decision-making.7 Choice architecture builds on that research by recognizing that carefully developed default settings can lead to better outcomes. In light of this insight, statutory reforms and some employers’ voluntary changes have led to the increased use of default settings in 401(k) plans.8 Yet, large numbers of workers still do not have access to 401(k) plans,9 and those who do often make less than optimal investment decisions.10

The severity of the issues with the current system and the contributions of choice architecture have not gone unnoticed. There have been many thoughtful and creative proposals for reform of the employer-based retirement security system. Some have focused primarily on tax incentives.11 Others have discussed ways of salvaging the traditional pension plans that increasingly have been replaced or supplemented by 401(k) plans.12 Other approaches favor increased government intervention and paternalism. For example, one commentator has proposed the creation of a system of Guaranteed Retirement Accounts (“GRAs”), including mandatory contributions for all employees with a government-appointed group of trustees determining the investment of the assets.13 Another somewhat similar proposal would eliminate the 401(k) system, provide government-matching contributions to accounts for low- and middle-income wage earners, and delegate investment authority to a government-selected fund manager.14 A third proposal from Senator Tom Harkin advocates a system that would require all employers to contribute to a plan for employees, who also might contribute, with private-sector funds managing the assets on a conservative basis.15

In this Article, I advocate incremental reform of the current 401(k) system with a continued emphasis on voluntary employer sponsorship and employee choice. This proposal is unique in that it builds upon the contributions that choice architecture theory has made to our understanding of 401(k) plan structure and the use of default settings. Unlike other major reform proposals, it retains the ideological differentiation between the private-employer-based pension system and Social Security. The proposal also reflects this Article’s analysis that the locus of fiduciary responsibility in 401(k) plans has become disconnected from its trust law origins. The reform proposal is counter-intuitive, though on its face not entirely novel16: I argue that portions of the fiduciary responsibility currently shouldered by employers that sponsor 401(k) plans should be shifted to financial services providers. Further, I maintain that small employers should have the ability to entirely avoid fiduciary responsibility for 401(k) investment selection and plan administration. Adoption of this proposal would encourage more employers to sponsor 401(k) plans.17 In addition, more assets would be held in low-cost, appropriately diversified investment vehicles.

This Article proceeds as follows. In Part II, I explore the lessons of choice architecture and behavioral economics for the allocation of decision-making in 401(k) plans, beginning with some background on the economic theory. I then discuss the distribution of decision-making in the current 401(k) plan regime between employers and employees, starting with the plan sponsorship decision, followed by employee contributions, and then investment selection. When viewed through a purely regulatory lens, the employee contributions and investment decisions are entirely in the hands of employees. Behavioral economics research, however, shows that employer decisions on plan terms may significantly affect employee decision-making. As a result of that research, some employers have adopted plan default settings intended to “nudge” preferred employee behavior.18

Part III provides a brief description of Australia’s approach to retirement wealth creation and its financial services-based trust model. Australia’s regulatory reform of default investments offers lessons for the U.S. The expert panel that developed the reform package explicitly based its approach to defaults on choice architecture principles.

Part IV addresses the intersection of default settings and fiduciary obligation in the U.S. It explains that the current allocation of fiduciary responsibility is attributable to the type of retirement plan that was popular when Congress passed the Employee Retirement Income Security Act (“ERISA”)19 in 1974 and its trust-based regulatory structure was established. The Part discusses application of the U.S. employer-centric trust model in the context of today’s 401(k) plans, including default investment products. It concludes that although both Australian and U.S. policymakers have facilitated the use of defaults, the United States’s employer-centric model produces a very different result for default investments than Australia’s financial services-centric approach. Failings of the U.S. approach are evident in the Department of Labor’s (“DOL”) post-financial crisis regulatory initiative on 401(k) plan investment defaults. The analysis shows that continued reliance on an employer-based trust model has compromised U.S. regulatory efforts that were intended to improve the use of default investment products.

Part V considers other 401(k) reform proposals that take a government-centric approach. Although those proposals offer thoughtful ideas to increase retirement security, each of them relies on employer mandates. In Part VI, I offer an alternative proposal that would retain most of the features of the current 401(k) system, including voluntary plan sponsorship, while reallocating certain fiduciary responsibilities and creating a new default investment product—Safe Harbor Automated Retirement Products (“SHARPs”).20 The proposal addresses the current fiduciary misalignment present in the U.S. employer-based model and leverages choice architecture insights to encourage increased levels of plan sponsorship, particularly by small employers, and more appropriate risk allocation within default investment products.

II.     Choice Architecture and Allocation of 401(k) Decisions and Responsibility

This Part begins by describing choice architecture and its intersection with behavioral economics. It goes on to explain the distribution of decision-making in the current 401(k) plan regime between employers and employees. That discussion considers ways choice architecture may be used to affect decision-making. The last Subpart contains significant analysis and discussion of the relevant literature regarding investment decision-making, which supports my ultimate recommendation for the creation of a new investment vehicle for use in 401(k) plans.

A.     Choice Architecture: An Overview

Professors Richard Thaler and Cass Sunstein coined the term “choice architecture” in their influential 2008 book describing how nudges can change decision-making.21 A choice architect “has the responsibility for organizing the context in which people make decisions.”22 Choice architecture describes the organization of that context (such as the structure of a 401(k) plan) and how that organization affects decisions (such as employees’ investment decisions). In that way, it is similar to how the architecture of a building affects the way the building is used.23

Choice architecture relies on behavioral economics. As a field, behavioral economics draws from psychology and economics to explain why human behavior sometimes departs in “persistent and consistent”24 ways from that predicted by traditional utility-maximizing economic theory.25 Researchers in behavioral economics have identified a number of heuristics and biases that help to explain these systematic departures from the decision-making predicted by classical economics.26 It is those specific insights from behavioral economics that choice architects may use in structuring a decision-making context in order to nudge a desired outcome.

A significant body of economic literature evaluates how behavioral economics can be used to influence the design of retirement plans.27 This Article does not attempt to either repeat or summarize the entirety of that continually evolving body of work. Instead, the rest of this Subpart focuses on behavioral economists’ findings on employee engagement with 401(k) plans.

Overall, the behavioral economics literature shows that many employees have a general disinterest in managing their retirement plans. Experiments have shown that employees are willing to make only a minimal time commitment to retirement plan management. Participants in a study conducted by Professors Benartzi and Thaler spent on average less than an hour making asset allocation decisions, and few of those participants reviewed any material other than what investment providers supplied.28 Another indicator of employee disinterest in making investment decisions is how rarely employees change the asset allocations in their defined contribution (“DC”) plans. They fail to revisit their initial investment allocation decisions or to rebalance their account portfolios even if their personal circumstances or financial market conditions change substantially.29 One study found that over the lifetime of a group of university employees, the median number of asset allocation changes was zero.30 Another, more recent study, similarly found that nearly half of the employees with accounts did not change their asset allocations during the ten-year study period.31

In addition to studies showing a general disinterest in retirement plan management, research on financial literacy also provides discouraging data for the retirement prospects of many employees. One relevant study considered the before-and-after test results from a group of employees who received financial literacy education.32 The net result of the education was a one-point increase in the employees’ test scores, from fifty-four to fifty-five.33 Purely random answers should have scored fifty because the test consisted of true/false responses.34

Fortunately, strategies exist at the regulatory and employer level to leverage the employee disengagement and passivity towards retirement plan management. As shown in the next Subpart, choice architecture provides evidence that employers’ default and framing decisions affect plan decision-making formally allocated by law to employees. Some employers have used these insights to construct plan terms to increase the likelihood that their 401(k) plans will provide higher levels of benefits to more employees.

B.     Choice Architecture and 401(k) Plans

In this Subpart, I consider the interaction between choice architecture and the regulatory allocation of decision-making. Subpart II.B.1 explains a setting where the decision authority rests solely with the employer. In Subparts II.B.2–3, though, the allocation of decision authority changes depending on whether the authority is viewed through a regulatory lens or a choice architecture lens.

1.     Plan Sponsorship: Employer Decision

The first decision to be made regarding a 401(k) plan35 is whether to offer a plan at all, and that choice is made by an employer.36 The U.S. private-sector retirement plan system has always been one of voluntary sponsorship.37 As a result, an employer may choose to offer a 401(k) plan, another type of plan, or no plan.

Of the more than $10 trillion of wealth Americans hold in tax-favored accounts intended to promote retirement security, $3.4 trillion is held in 401(k) plans.38 Across the world, retirement-type plans are categorized as DC plans or defined benefit (“DB”) plans.39 In DC plans, the investment risk resides on employees, not employers. Upon retirement, employees typically are entitled to whatever amount has accumulated in their DC plan account.40 401(k) plans are a type of DC plan. The defining quality of a 401(k) plan is that each employee who is eligible to take part in a 401(k) plan must have the right to choose to contribute, or not to contribute, pre-tax earnings to that employee’s own plan account.41 That individual employee decision-making power on whether to contribute is one of the reasons that choice architecture plays such a powerful role in the success of 401(k) plans as long-term wealth accumulation mechanisms.42

Before moving on to address the extent to which employees take part in the 401(k) plans their employers offer, some specialized terminology is important to define. Coverage rates measure whether employees have the option to contribute to 401(k) plans or to take part in other types of plans.43 In comparison, the rates at which employees actually decide to contribute or otherwise accumulate savings in the plans are known as participation rates.44 In DB plans, which traditionally did not accept, let alone require, employee contributions, coverage and participation rates are typically equal or close to equal.45 In plans such as 401(k)s, where employee contributions are optional, coverage rates may be significantly higher than participation rates.46

The percentage of employees covered by any type of retirement-style plan depends on the definition of the employee population being analyzed. Professor Alicia Munnell and colleagues found that, as of 2010, private-sector employer-sponsored retirement plans covered approximately 58% of full-time employees between the ages of twenty-five and sixty-four.47 In 1979, coverage for the same population was above 65%.48 The data on which this research was premised does not isolate 401(k) plans. Another data set indicates that, as of 2010, 401(k) plans covered approximately 68% of the employees who had access to a pension plan.49 Another 13% of employees with pension coverage have a 401(k) and another type of plan.50

The lack of access to plans affects particular categories of employees more than others, and access has declined over the forty-year period studied. Small employers are less likely than larger ones to offer retirement plans. At employers with less than one hundred workers, another researcher estimated that only 49% of employees have access to a plan.51

2.     Contributions—Shared Choice

Assuming an employer has chosen to offer a 401(k) plan, one of the identifying factors of that type of plan is that employees have the right to make voluntary contributions. When considered using a regulatory lens, therefore, the entire decision-making authority on voluntary contributions is allocated to employees. Historically, plans provided that contributions would only be withheld from the wages of employees who affirmatively complied with the plan’s procedures for designating a voluntary contribution.52 Using the regulatory lens, the failure of many employees to enroll and contribute to 401(k) plans was attributed to employee decision-making.53

A choice architecture lens, however, shows that employees and employers each play a role in determining whether an employee who is covered by a plan actually accumulates any assets in the account through contributions. One of the insights of choice architecture is that employer decisions about plan default settings can significantly affect whether employees contribute to a 401(k) plan, and, if so, the rate at which they contribute.54 The general concept of default settings is that they may enable 401(k) contributions to be made with no effort on the part of individual employees. The employer determines the default settings as part of the employer’s decision-making on the basic structure of plan terms.55 Employers always choose a participation default setting for 401(k) plans; however, sometimes employers make those decisions implicitly. In the historic approach, discussed above,56 the default setting was “no participation.” Thus, if the employee did nothing, the employee did not contribute to the plan.

Re-setting the default on participation in 401(k) plans from “no participation” to “participation” is an example of the affirmative use of choice architecture.57 In so-called automatic-enrollment plans, the employer establishes plan terms that default employees into plan participation, but employees may still make an express decision to decline participation (to opt-out).58 Although the ultimate decision remains with employees, opting-out requires an action on their part to override the enrollment.

In these automatic-enrollment plans, employees retain the power not to contribute, but studies have found that the structure of the decision-making (whether the default for those who do not affirmatively decide is “no participation” or “participation”) dramatically affects participation rates. One model indicates that prior to the use of automatic enrollment, 66% of eligible workers participated in 401(k) plans.59 Immediately after introduction of automatic enrollment, participation increased to 92%.60 The cohort that automatic enrollment affects the most is the cohort most at risk of retirement income inadequacy—low-income workers.61

An employer that establishes a plan with an automatic-enrollment default must also set a default that determines the employee’s degree of participation. Whether or not the enrollment was automatic, an employee’s participation requires a decision on how much the employee will contribute. A strategy focused on maximizing wealth creation might choose a setting aligned with the maximum pre-tax contribution the Internal Revenue Code permits.62 A strategy of achieving the highest ratio of employer match to employee contribution could be set at the lowest contribution level required to trigger the maximum match.63 For an employee population typically reluctant to participate in such plans, for example workers at the lower end of the pay scale, an employer might select a default setting at a low dollar or fixed percentage that increases over time.64

One criticism of automatic-enrollment features is that some simulations predict that a substantial portion, perhaps up to 40%, of new hires at companies that use automatic enrollments save less in their 401(k) plans than they would have in the absence of automatic enrollment.65 This prediction is based in part on the fact that most plans set the default contribution rate at 3% of salary, whereas employees who affirmatively elect to participate in plans tend to contribute at 5–10%.66 Depending on the assumptions used, some percentage of employees who would otherwise actively enroll and contribute at the higher rates are likely instead to default into plans with automatic enrollment. Again, depending on the assumptions, those employees may save less than they otherwise would have.67

Studies of the overall effect of automatic enrollment in the current 401(k) system, however, indicate that use of default provisions increases savings for the majority of employees who participate in the plans.68 An author of the study cited for the proposition that up to 40% of participating employees save less in automatic-enrollment plans, pointed out in response that the 40% outcome resulted from the most pessimistic set of the sixteen sets of assumptions modeled in the study.69 Furthermore, higher-paid employees may contribute at higher rates outside of automatic enrollment, a right they have even in plans that use a low default setting. Perhaps more importantly, data consistently show that lower-income employees experience the greatest percentage benefit from automatic-enrollment plans because they so significantly increase the likelihood employees will contribute.70 For those low-income employees who would not have contributed to a 401(k) plan that requires an affirmative participation election, regardless of the default contribution level set by the employer, it is larger than the zero rate at which those employees would otherwise have saved.

Debating the effect of automatic-enrollment plans on initial contribution rates ignores another insight of choice architecture for 401(k) plan structure. Plans may adopt a default setting that leverages employee passivity to increase contributions. Plans that use automatic escalation set a low initial default contribution rate but periodically increase employees’ contribution rates unless employees opt otherwise.71 The plan may even synchronize rate increases with employee raises,72 so employees do not experience a decrease in take-home pay. As one would expect, it appears that automatic escalation significantly increases employee wealth in 401(k) plans, particularly for lower-paid employees.73 A 2010 survey indicated that approximately twenty-eight percent of the 401(k) plans of large employers sponsor utilize automatic escalation features.74

In sum, once an employer has unilaterally decided to sponsor a 401(k) plan, the decision on whether an employee voluntarily contributes and, if so, the amount of those contributions is often thought about as a decision that is delegated to employees. However, choice architecture shows that employer decisions on plan default settings affect participation and contribution rates. Given the passive behavior of individual investors, it has been amply demonstrated that two strategies that successfully increase the numbers of employees who contribute and the amount they contribute to 401(k) plans are automatic enrollment and automatic escalation. Those strategies turn investors’ passivity into a retirement wealth accumulation advantage. Although the default setting for contribution levels is a key factor, and may be influenced by factors outside the scope of this Article, without employer sponsorship and employee participation, contribution levels are a non-issue. Contributions alone, however, do not a comfortable retirement make. The next Subpart considers the investment of those contributions.

3.     Investments—Shared Choice

Investment selection at the employee account level75 in 401(k) plans has, like contributions, often been regarded as an employee decision.76 In fact, in plans that meet specified regulatory criteria, employers are relieved of the fiduciary liability associated with account-level investment selection.77 Not surprisingly, most plans comply with those criteria. Because in U.S. pension parlance the employees and their beneficiaries who participate in benefit plans are known as participants, those plans are known as participant-directed plans.78

Again, as with contribution decisions, the behavioral-economics literature provides the basis for a more sophisticated understanding of account-level investment decisions, requiring acknowledgement that the structural decisions employers make about plans affect employees’ investment decisions. The insights of choice architecture have led to the development of default mechanisms to counteract the negative effects of employer decisions on plan investment menus and how those menus are presented.79 At the same time, the default mechanisms leave ultimate power over account-level investment decisions with those employees who affirmatively choose to exercise it.

Every 401(k) plan that uses automatic enrollment must set a third default in addition to the positive contribution default and the default specifying the contribution amount. That third default is the investment product that will hold the contributions in the employee’s 401(k) plan account.80 Employees who affirmatively exercise their right to designate their account investments typically may affirmatively elect the default product.81

Regulation and employer choice of investment default settings have changed significantly since 2007. Before evaluating that regulation and its impact, the next subparts explain the findings of the behavioral economics literature on plan investment-related terms, decisions regarding the number of investment options, and employer-matching contributions made in employer stock.

a.     Number of Investment Options

401(k) plans vary significantly in terms of the investment options they offer.82 This Subpart considers the non-intuitive problem an overabundance of plan investment options creates, and contrasts it with behavioral patterns found in plans with a small number of options.

Current law provides an incentive for an employer to offer at least three investment options in its 401(k) plan. In participant-directed plans, employers not only shift the investment risk to their employees,83 they also avoid fiduciary liability for employees’ account-level investment decisions. In order to qualify as participant-directed, among other requirements, a plan must offer at least three investment options that have sufficient variety in their “risk and return characteristics” to permit employees to select a portfolio appropriate for their needs.84

Professors Thaler and Benartzi conducted one of the early behavioral economics studies of employee decision-making in benefit plans, and the study involved a choice between two investment options.85 They asked two groups of university employees, faculty and staff, to allocate their retirement accounts. Each study participant chose from one of three menus of investment options: (1) a stock fund and a bond fund; (2) a stock fund and a balanced fund that was invested half in stock and half in bonds; and (3) a bond fund and a balanced fund.86 The study determined that employees’ allocation decisions depended heavily upon the menu from which the employee selected investments.87 That is to say that the decision was not based on objective merits of the investment, rather it was influenced by the combination of investment options. The group that selected between the stock and the balanced fund allocated the largest percentage of assets to stock, followed by the group with the stock and bond fund.88 The group offered the bond and balanced fund allocated the lowest percentage of assets to stock.89 The experiment illustrates what is known as the “1/n heuristic,” which describes the tendency of investors to vary asset allocations, often evenly, among available basic investment alternatives.90

As an example of the 1/n heuristic in action, Thaler and Sunstein quote Dr. Harry Markowitz, who co-won the Nobel Prize in Economic Sciences for his work on modern portfolio theory.91 Speaking about his own retirement account, Dr. Markowitz said: “I should have computed the historic covariances of the asset classes and drawn an efficient frontier. Instead . . . I split my contributions fifty-fifty between bonds and equities.”92 Rather than making a decision, or receiving guidance, even the most financially sophisticated investors may resort to unsophisticated schemes for allocating their retirement dollars.

Problematic decision-making is not limited to plans with a small number of investment options and most plans offer substantially more than three options. One recent study found that the average number of options offered was eighteen.93 At the seventy-fifth percentile, firms offered twenty-one options.94 This is a situation, though, where more is not necessarily better.

Research indicates that when the range of choices becomes too large for employees to make investment decisions using a simplified heuristic, such as the 1/n heuristic,95 some employees become immobilized and tend not to make a decision.96 In fact, the same study showed that as the number of investment options increases, employees’ participation in plans without automatic enrollment decreases.97 Plans with fifty-nine options had participation rates of approximately 60% whereas plans with two investment options had average participation rates of 75%.98 That delta of fifteen percentage points represents an increase of twenty-five percentage points causally associated with the reduced complexity of the investment options.

In sum, there may be a Goldilocks-like effect with plan investment options. A few options will result in less than optimal employee investment allocations because many employees will use the 1/n heuristic. On the other hand, a large number of options will decrease the likelihood employees will participate in a plan without automatic enrollment. The closest to a “just right” choice arguably is for employers to designate a default investment vehicle. This Article discusses those vehicles in some detail in Subpart IV.A.3.

b.     Employer Stock

In another group of studies, behavioral economists explored the use of employer stock in 401(k) plans. Professors Thaler and Sunstein summarized the consistent results of studies on the behavior of employees who automatically receive some stock of their employer in their 401(k) plans. These employees, when presented with the option to direct stock purchases, invest a higher percentage of their contributions in employer stock than do similarly situated employees in plans where employees do not receive automatic allocations of employer stock.99 Behavioral economists speculate that the explanation for this behavior is that employees view the automatic allocations of employer stock as implicit advice that the stock is a good investment.100 As a result, employees rely on this tacit guidance in making their investment decisions.

Thaler and Sunstein underscore the risk associated with this overinvestment in employer stock, citing the example of an Enron employee who retired in 2000 with $1.3 million worth of Enron stock. The value of that stock fell to zero the following year in the Enron bankruptcy.101 Though extreme, this is just one individual at one company among the many individuals at many companies who have wagered their retirement security on the financial success of their employer. In the late 1990s estimates indicated that 30–40% of the assets employees held in 401(k) plans that permitted employees to invest in employer stock were held in that stock, and lower-paid employees held the highest level of concentrations.102 Employers now are less likely to match contributions using employer stock or to provide it as an investment option to employees, perhaps because of the potential fiduciary risk.103 The phenomena of reliance on the tacit guidance of employers, however, will be relevant below in the discussion of default investment products.104 First, though, the next Part looks at the Australian approach to private-sector retirement plans and its reliance on choice architecture to reform default investment product regulation.

III.     Long-Term Retirement Wealth Accumulation Down Under: The Australian Approach

A.     Australia’s “Mandatory” System of Workplace Retirement Savings

As is the case with most retirement systems, including that of the United States, Australia’s pension system consists of three components105: (1) a government-administered program funded through general revenues, the Age Pension, which pays monthly benefits to retirees who need those benefits as determined by income and asset tests;106 (2) some tax incentives for individual savings, and (3) an employment-based system.107 The focus in this Article is primarily on the employment-based component, which is known as the Superannuation Guarantee (“SG System”).108 The SG System developed in the early 1990s through the Australian approach to setting wages and benefits by using industry awards.109 At the outset, the SG System required110 employers to contribute 3% of most earnings for most employees to an individual DC account.111 The rate of contribution has increased over time to the current level of 9% and is scheduled to grow to 12% by 2020.112 Members113 may begin withdrawing funds from their SG System accounts when they reach the “preservation age,” which is between age fifty-five and sixty, depending on date of birth.114

The SG System’s mandatory contribution requirement does provide for a few exceptions. Contributions are not mandatory for employees who earn below a stipulated amount per month, individuals below age eighteen or over age seventy, and the self-employed.115 On the other hand, although the 9% rate is the minimum level for contributions for most Australians, it does not act as a ceiling. Employees can save additional monies in the SG System through one of two approaches. First, employees may elect to set aside a portion of their future salary, an election that is known as a “salary sacrifice.” In that case, the employer forwards the contribution to the employee’s SG System account.116 Alternatively, some employers and employees may enter into enterprise agreements or individual employee contractual arrangements to contribute at a rate above the 9% minimum.117 By 2007, the ability of individuals to voluntarily contribute to SG System accounts expanded to the point that any unretired Australian of at least age fifteen could contribute to her own SG System account or receive contributions in her account from her spouse’s after-tax income.118

A number of studies and surveys have demonstrated the success of the Australian approach to retirement plan management. Coverage estimates from a 2007 survey indicated that 94% of Australian employees were members of the SG System.119 Another indicator of the success of the Australian SG System has been its ability to grow pension assets. According to a study of thirteen countries with significant pension systems, during the ten-year period ending in December 2008 Australia’s pension assets grew at the fastest rate among those countries and at more than triple the rate of growth in the U.S. system.120 SG System assets totaled $1.301 trillion as of the end of 2011.121

Between the early 1990s and mid-2005, nearly all SG System assets were held in industry funds, company-sponsored funds, or public-sector funds. Industry funds were established on an industry-by-industry basis and governance of the funds was divided between employers and employees.122 In mid-2005 members began to receive the right to choose both the “fund” and the investment product within a fund to receive the SG System contributions made on their behalf.123 Fund choice gave a boost to for-profit funds, known as “retail funds,” which are not affiliated with a particular industry or employer but instead parallel U.S. mutual funds.124

From a broad perspective, default settings are used in fewer contexts in Australia than in the U.S. because of the structure of the SG System. Because contributions to the system are mandatory for nearly all Australian workers, there is no need for a default setting on participation. The statutory minimum of 9% nullifies the need for a default contribution setting. However, once SG System members received the right to choose both the fund and the investment product that would receive their contributions, investment default settings became important because some employees fail to make an explicit choice.

Various mechanisms are used to determine the default investment product. In some instances the default fund and product are negotiated through what are called enterprise agreements or modern awards, which replace the system of industry awards.125 In other situations, the employer typically selects a fund and product to receive contributions made on behalf of the employees.126 Industry funds currently hold more assets than any other fund category.127 Default investment products have proven to be popular in the SG System. In recent years, 68% of the assets held in industry funds were held in the default product of the particular fund.128

B.     Choice Architecture in Default Investment Reforms

Partly in reaction to the global financial crisis and its negative impact on retirement savings, in May 2009 the Australian Minister for Superannuation and Corporation Law announced a review of the entire SG System.129 A full-time chairperson, Jeremy Cooper, assisted by five part-time members, undertook the review.130 In December 2011, the panel issued a two-part report, the “Cooper Report,” on its findings and recommendations.131 The Cooper Report contained ten packages of recommendations.132 The Australian government supported most of them133 and later issued the key design elements of the overall reform, which it has named “Stronger Super.”134 As of this writing, the Stronger Super reforms are at the stage of draft legislation.135 Because it appears nearly certain that the basic reform package will be enacted, the applicable agency is in the process of issuing guidance on the new default investment products.136

The Cooper Panel explicitly relied on choice architecture to shape Australian reform to serve employees with various levels of interest in engaging with their SG System accounts, including those who prefer minimal engagement.137 Surveys and data show that Australians follow a pattern similar to that of U.S. 401(k) participants discussed earlier in this Article.138 Many Australians are passive with respect to their investments, do not make active plan choices, and have limited financial literacy.139 Others, however, are actively engaged in the management of their plan account.140 The Cooper Panel determined that Australia’s SG System should be structured to maximize the long-term wealth accumulation of employees, regardless of their individual locus on that engagement spectrum.141

Recognizing that choice architecture highlights the importance of defaults for disinterested employees, the Cooper Panel developed a framework for a new type of default investment product, “MySuper.”142 Once the reforms are implemented, MySuper products will be the only permitted type of default investment product. In addition, employees who wish to make explicit investment decisions may designate a MySuper product to receive their contributions.143

The framework is relatively simple. In general, SG System funds will each be permitted to offer a single default MySuper investment product. MySuper products will only be allowed to provide a limited menu of services and are expected to have relatively low fees. Those fees will be reported in such a way as to make them comparable across MySuper products. The Australian Prudential Regulatory Authority (“APRA”) will gather and publish data on MySuper product performance and fees to facilitate competition among the offerings.144 The Cooper Panel also addressed fiduciary responsibility and the operating standards for MySuper products. Its approach is discussed in the next Part.

IV.     Fiduciary Obligations, Trust Models, and Choice Architecture

This Part begins by describing the United States’ employer-centric approach to assigning fiduciary responsibility for private-sector pension plans. It analyzes that approach as applied to investment of 401(k) account assets, including the use of default investment products. Next, this Part contrasts Australia’s financial services-based fiduciary model. The Part ends by comparing the reaction of the two countries to the performance of default investments during the global financial crisis.

A.     Employer-Centric Trust Model

1.     Regulatory Framework

ERISA includes a set of fiduciary provisions that are based in traditional trust law. ERISA’s counterpart to the trust law duty of loyalty is found in its requirement that fiduciaries act “solely in the interest of the participants and beneficiaries and . . . for the exclusive purpose of . . . providing benefits to participants and their beneficiaries.”145 The statute sets the fiduciary standard of care as that of a prudent person familiar with the benefit plan matters at issue.146 ERISA’s other substantive fiduciary standards require benefit plan fiduciaries to minimize the risk of large losses by diversifying plan investments147 and to act in accordance with plan documents.148

ERISA modifies traditional trust concepts to fit the benefit plan paradigm.149 These modifications are particularly evident in ERISA’s definition of who bears fiduciary responsibilities and the scope of those responsibilities. Whereas traditional donative trusts typically designated a single or limited number of trustees designated to hold the trust property,150 ERISA fiduciary status may arise either through designation or by an action that is defined as giving rise to fiduciary status.151

In ERISA’s employer-centric trust model, employer actions in creating, amending, and terminating a benefit plan, such as a 401(k) plan, are treated as being similar to the actions taken by the creator of a traditional trust. Thus, they are known as “settlor” functions.152 In contrast, ERISA fiduciary actions are those involving discretionary plan administration, asset or plan management, or investment advice.153

In this model, all employers that sponsor 401(k) plans are fiduciaries with respect to the plan for at least some actions.154 This is because employers necessarily appoint and monitor the plan fiduciaries that engage in plan administration, management, or provide any available investment advice; and the appointment and monitoring functions are fiduciary acts.155 A typical 401(k) plan may have numerous fiduciaries, including a plan committee.156 But, in the employer-centric trust model, ultimately the power of all fiduciaries derives from the employer’s direct or indirect delegation of authority.

2.     Fiduciary Status for Plan Investments

The selection of and risk associated with plan investments has shifted significantly since Congress enacted ERISA’s fiduciary provisions in 1974. Senate investigations prior to ERISA uncovered repeated misuse and diversions of pension plan assets by the entities charged with responsibility for those assets.157 By the time the legislation that was to become ERISA reached the conference committee, both the House and Senate versions imposed duties of care and loyalty on plan fiduciaries.158

At that time, DB plans were the primary type of retirement plan in the U.S. and 401(k) plans did not even exist.159 Employers promised, through DB plans, to pay monthly benefits for a retiree’s lifetime. Those benefits typically were based on a formula that took into account employee salary and years of work with the employer.160 Prior to the enactment of ERISA, if a DB plan held insufficient funds to pay the promised benefits, retirees and employees would lose some or all of those expected benefits.161 In conjunction with minimum DB plan funding rules, avoiding that outcome was one of the primary motivators of ERISA’s fiduciary provisions.162

In the DB system, where employers voluntarily created pension plans, established the formulae for benefit payments, funded the plans, and made the investment decisions, the parallel with a traditional trust settlor was reasonably strong. Because employers bore the investment risk in such plans,163 they had incentives to develop expertise regarding investment of plan assets. For employees, the plan funding and fiduciary regulations had the potential to ensure that they received their promised plan benefits without having to develop investment expertise or to engage in extensive monitoring. This division of responsibility and authority in DB plans aligned with the classic understanding of fiduciary relationships as being efficient when a party acting on behalf of another possesses superior expertise and a significant degree of control over the subject matter of the relationship.164

The decline of DB plans has been widely analyzed and frequently bemoaned.165 But, for better or worse, DC plans currently constitute the primary wealth-accumulation vehicles—other than possibly their homes—for the retirement security of many Americans.166 In terms of assets, private-sector employer-sponsored DB plans held an estimated $2.5 trillion as of March 2012.167 In comparison, DC plans held approximately $4.8 trillion in assets and of that $3.4 trillion was held in 401(k) plans.168

In terms of the trust model and related fiduciary provisions, 401(k) plans differ on the dimensions of control and required expertise for investment decision-making as well as on risk allocation. As with DB plans, it is employers that establish 401(k) plans.169 The requirement that assets be held in trust also applies to 401(k) plans. But, as discussed above, employees typically bear the investment risk and make the account-level investment decisions for their plan account.170 An employer who establishes a 401(k) plan may avoid fiduciary duties associated with the account-level investment decisions by ensuring that the plan meets regulatory safe-harbor criteria.171 Thus, 401(k) plans do not align with the traditional defining factors of a fiduciary relationship between employees and employers for the DB plans that were the primary pension vehicle at the time the statutory fiduciary model was established.

Employers, however, cannot avoid all fiduciary responsibility associated with 401(k) plan investments—they must select investment options in accordance with ERISA’s fiduciary standards of prudence and loyalty.172 As discussed below, this concept is important in the context of default investment selection because it means that employers currently have fiduciary responsibility for those choices.173 Although there is a dearth of case law on the predominant type of default investments, the principle of fiduciary liability for selection of investment menu options has arisen with some frequency in two contexts. The first is the use of employer stock as an investment alternative.174 The second context comprises claims that employers did not act with the proper degree of care or loyalty when establishing plan menus that contain investment products with unreasonably high fees.175

The cases alleging inappropriately costly investment products are more useful for thinking about fiduciary liability associated with default products than the employer stock cases because ERISA contains specific provisions permitting, and arguably encouraging, the use of employer stock.176 The General Accounting Office, in a study of 401(k) plan fees and disclosure, summarized one aspect of the fees problem: “[I]t is hard for [employees] to make comparisons across investment options because they have to piece together the fees that they pay, and assessing fees across investment options can be difficult . . . .”177 That plans use a variety of mechanisms to charge for plan administration, investment management, and other services exacerbates the problem.178 Because of the complexity, comparing fees across plans and investment alternatives is difficult for even sophisticated investors. Studies finding significant variations in fees across 401(k) account holders confirmed the anticompetitive effect of the lack of transparency.179 Perhaps it is not surprising, then, that employees have brought a number of lawsuits alleging that employers acting as 401(k) plan sponsors breached their fiduciary duties by choosing investment options with unreasonably high fees. The results of the litigation have been mixed,180 generating costs for employers and limited compensation for payments of high fees for employees. The DOL’s response has been to increase mandatory disclosures that plan service providers must issue to plans181 and, in turn, the disclosures that plans must provide employees.182

The oddity from a fiduciary perspective is that some of those service providers recommend, and effectively determine, plan investment menu options, including the allegedly high fee options, but have no ERISA fiduciary obligations in the current employer-centric fiduciary model. Fiduciary responsibility frequently does not extend to the consultants and financial services entities that advise employees on selection and monitoring of plan investments, or to the investment managers that make the fee-related and investment decisions for the mutual funds and similar products that constitute typical plan-account investment products.183 This is an artifact of the DB plan system. Shortly after Congress passed ERISA, the DOL defined through regulation a narrower set of criteria for when investment advisers become fiduciaries than provided for by the statutory definition.184 By conforming its actions to the regulatory definition, a financial adviser can avoid fiduciary responsibility for advice provided on the selection of a 401(k) plan’s investment menu.185 Recognizing that the plan paradigm has shifted substantially since the regulation was issued, the DOL issued a proposed regulation to harmonize the scope of the regulatory definition with that of the broader statutory provision in 2010.186 The proposal drew a firestorm of opposition and in September 2011, the DOL withdrew the proposed regulation.187 As of the writing of this Article, it appears the DOL is in the process of revising the proposed regulation and plans to re-propose it.188

In sum, ERISA established trust-based fiduciary duties that govern employer actions as well as those of other plan fiduciaries. The ultimate fiduciary responsibility, though, frequently lies with the employer, who is responsible for appointing and monitoring other fiduciaries. In the DB era, the employer-centric fiduciary approach aligned with the traditional concept of assigning fiduciary obligation when one party had superior expertise and control. The transition to DC plans means, though, that employers no longer make direct decisions on the investment of plan assets. Nor do they bear any investment risk for those decisions. Instead, the employer’s role, vis-à-vis investments, is to select and monitor the menu of investment products available to employees. The DOL’s recognition of limitations on the effectiveness of the employer-related model in the 401(k) paradigm led it to attempt to impose fiduciary obligations on the advisory community. A discussion of the implications of those efforts is beyond the scope of this Article. But, the DOL’s concern highlights the general problem of fiduciary liability for selection of 401(k) plan investments. The next Subpart discusses the employer-centric model in the specific setting of default investments.

3.     Fiduciary Status and Investment Defaults

In 1998189 and 2000,190 the IRS authorized the use of auto-enrollment defaults in 401(k) plans, yet a number of concerns slowed employer implementation. Potential liability claims associated with the employer’s choice of the default investment selection led the list of concerns, which also included potential conflicts with state statutes regulating wage garnishments.191 In addition, there was no incentive for employers to take on these risks.192

Employer adoption of default settings began to change after the enactment of the Pension Protection Act of 2006 (“PPA”),193 which provided an incentive for plans to use defaults.194 To implement the PPA, in 2007, the DOL issued final regulations that partially eliminated fiduciary liability for employers who select “qualified default investment alternatives” (“QDIAs”)195 as the investment default for their plan.196

The 2007 final regulations essentially provide the same protection to the employer as the employer would have for any investment explicitly designated by an employee in a participant-directed plan.197 As one commentator explained: “[P]articipants who do not direct the investment of their accounts will be treated as if they had, if the fiduciaries invest their account in a [QDIA].”198 The protection for employers, however, is only partial. The final regulation makes clear that: “Nothing in this [regulation] shall relieve a fiduciary from his or her duties under . . . ERISA to prudently select and monitor any [QDIA] under the plan or from any liability that results from a failure to satisfy these duties, including liability for any resulting losses.”199 Even if the employer appoints an individual or a group of individuals to make the actual QDIA selection, the employer remains obligated as a fiduciary to monitor those decision-makers.200

Compared to the fiduciary responsibility of employers, the fiduciary responsibility of experts who provide advice on QDIA selection to the ERISA fiduciaries is more complex. Because of the narrow definition established in the 1975 regulations,201 the professionals that provide advice on QDIA selection can quite easily avoid becoming ERISA fiduciaries.202 Those professionals may be subject to other federal laws depending on their status and the scope of advice they provide. As a general matter, those who are compensated for providing advice related to investments in securities are subject to the Investment Advisers Act of 1940 (“Advisers Act”).203 As with ERISA, however, a number of exceptions exist from Advisers Act regulation.204 The Securities and Exchange Commission (“SEC”) has considered the extension of fiduciary obligations to a wider variety of advice providers, but whether, and if so when, that will happen remains uncertain.205

The net result of U.S. default regulation, then, is that employers are responsible as fiduciaries for the selection and monitoring of default (and all other) investment products offered in their plans. Investment professionals, including expert advisers, may become involved in providing information and advice regarding the selection of QDIAs for plans. Those experts, however, frequently do not have fiduciary obligations to the employer, the 401(k) plan, or to the employees whose retirement assets are invested.206

Finally, most QDIAs being used by plans are target date funds (“TDFs”), which typically set the portfolio’s risk level based on length of time to retirement and are usually organized as mutual funds.207 Thus, the entity that holds the TDF’s assets is almost certainly an investment company subject to the Investment Company Act of 1940.208 This inserts another layer of regulation, but not much additional protection for employees or employers. Professor Donald C. Langevoort has described the relationship between mutual funds and their investment advisers, highlighting the conflict of interest: “The typical mutual fund is organized by a sponsor who expects to profit by providing advisory and other services to the fund, with returns growing as the fund grows in size.”209 Further, ERISA explicitly provides an exemption from fiduciary status for mutual funds and their investment advisers.210

B.     Financial Services-Based Model

Trust law and fiduciary obligation are important concepts in Australia’s SG System. The fiduciary duties of care and loyalty, although country-specific in the details, parallel the U.S. approach in their general application.211 However, Australia uses a financial services-based trust model rather than the U.S.’s employer-based model. Whether organized as an industry fund, an employer-sponsored fund, or a retail fund, from the beginning of the mandatory SG System the investment vehicles that hold account assets have been governed by entity trustees (sometimes referred to as “corporate trustees”).212 Trust and fiduciary principles apply to the relationship between that trustee and the employees who have invested SG Account assets in the fund.213 Legislation requires that trust documents of SG System funds include covenants on the basic trust law duties of loyalty and care to reinforce the trust and fiduciary principles.214

The locus of the loyalty and care obligations with the trustee of the relevant investment funds is as consistent within the context of the development of the SG System as the U.S. employer-based model was during the era of the DB system. The U.S. system recognized that employers not only established DB plans, but also funded them, controlled their investments, and established benefit levels. Locating the “buck stops here” fiduciary exposure with employers was consistent with their expertise and control.215 In Australia, though, the SG System is mandatory, minimum contribution levels are mandatory, and DC plans have always been the primary type of SG System plan.216 Funds established and managed by a single employer hold less than 5% of SG System assets.217 Instead of establishing a separate fund, employers typically remit contributions to either an industry or retail fund governed by an entity trustee.218 Once employees received the statutory right to choose among any qualified investment fund, the role of employers was limited, at most, to input on the selection of the default investment fund, and it appears that the employer owes no particular level of care in making this selection.219 Given the comparative lack of employer expertise and involvement with administrative and investment matters, the Australian decision to locate fiduciary-like obligations of care and loyalty with investment fund trustees was logical.

Consistent with that history, the regulatory approach to MySuper default products imposes an enhanced set of duties on MySuper fund trustees and on the boards that govern the trustees.220 Employers play no significant role and have no significant liability in this system. The enhanced obligations of MySuper trustees will essentially operate as an additional layer of duties on top of the basic set of requirements that applies to all entity trustees of funds that hold SG System assets. The enhanced duties required of MySuper entity trustees are to “promote the financial interests of MySuper members, in particular net returns; annually assess sufficiency of scale; and include in their investment strategy an investment return target and level of risk for MySuper members.”221 To be clear, although determination of the investment strategy is within the scope of the trustee’s obligations, the choice of strategy is constrained. One of the requirements for MySuper products is that the fund trustee would have to “formulate and give effect to a single, diversified investment strategy at an overall cost aimed at optimising fund members’ financial best interests, as reflected in the net investment return over the longer term.”222 In addition, trustees must be licensed and meet specific standards with respect to the operation of a MySuper product.223

A board of directors governs the entity trustees that hold SG System assets.224 The trustee–directors of any fund that offers a MySuper product also will be subject to an enhanced set of duties. Each trustee–director will have an obligation to ensure that the fund’s corporate trustee fulfills its obligations, including the duties specific to MySuper products.225

The imposition of enhanced standards on the trustees and directors of MySuper products is consistent with the Cooper Panel’s finding that Australia’s earlier decision to grant employees the right to choose investment funds and products failed to achieve a competitive fund market and optimal investment decision-making. The Panel observed that the failure of many employees to affirmatively make a fund election contributed to the lack of an efficient market for SG System funds, but other factors also play a role.226 According to the Panel, employees lack awareness of the performance and fees associated with their retirement investments.227 In part, this is because they do not actively make payments into their accounts and, in many cases, do not expect to access the funds for many years.228 In addition, fund performance and fees can often be difficult to compare, and switching funds takes effort and time.229

Mr. Cooper summarized the goals of MySuper this way:

MySuper would have a number of features designed solely with the [employee] in mind: specific trustee duties designed to deliver lower cost outcomes for [employees]; increased transparency leading to better comparability, especially of costs and long-term net performance; provision of intra-fund advice; simpler communications; and an embedded retirement product. It has been designed to sit within existing superannuation structures and is based on existing widely-offered and well-understood default investment options.230

The astounding fact to any reader familiar with the U.S. employer-based fiduciary approach is that nowhere in Mr. Cooper’s summary of MySuper is there a single reference to employers.231 Instead of the employer-based model used by the U.S., Australia’s approach to fiduciary obligation within its SG System has always been, and continues with MySuper to be, a financial services-based model.

C.     A Choice Architecture-Based Comparison of the Models

As noted above, one of the motivating factors for the Australian SG System reforms, including the attention to default investments, was the effect of the global financial crisis on employees and retirees.232 Similarly, the financial crisis motivated the DOL to examine whether TDFs held as QDIAs performed effectively in 401(k) accounts during that time period.233 This Subpart briefly describes the effect of the financial crisis on U.S. and Australian retirement savings accounts before comparing the DOL’s response on default investments with Australia’s reform efforts.

1.     The Global Financial Crisis and Retirement Savings Accounts

Not surprisingly, the global financial crisis resulted in substantial loss of wealth in 401(k) plans, particularly in the early years of the crisis. In the U.S., the S&P 500 Index,234 the Dow Jones Industrial Index,235 and the Dow Jones Wilshire 5000236 each lost between 34% and 37% in 2008. Economic modeling indicates that the shift from DB to DC plans is also a factor in increasing the percentage of individuals over age sixty who remain in the workforce.237

The Employee Benefit Research Institute (“EBRI”) has been a leader in modeling 401(k) data using a proprietary database, which includes more than twenty million plan participants.238 Even prior to the economic downturn, EBRI projected that 401(k) accounts would likely generate income-replacement rates239 that varied from a range of 21–26% at the low end to 51–69% at the high end, depending upon assumptions for worker participation.240 In the early period of the crisis, EBRI estimated that the effect of the financial downturn on the size of 401(k) accounts depended—in addition to the extent to which accounts were invested in equities—on account size, age, and job tenure.241 In smaller accounts, new contributions had a larger effect than investment losses.242 As a result, net account balances in small accounts actually were estimated to have increased in 2008.243 However, estimates indicate that accounts with balances of over $200,000 averaged losses of more than 25%.244

A typical investment pattern driven by professional guidance would change as the investor approached retirement, reducing the proportion invested in volatile equities to a more “age-appropriate” level. Significant numbers of employees, however, retain equity-rich investment blends into their late 50s and beyond.245 The investor passivity observed in behavioral-economics research should lead policy makers to conclude that at least some of these employees did not regularly assess their 401(k) account investments and decide to engage in such a high-risk strategy. Instead, many of them are likely to have selected an investment mix at the time they entered the plan and never changed it.

Estimates of losses in account value due to high levels of equity investment illustrate the potential problem. Professor Munnell and colleagues used Vanguard data, which showed that the age fifty-five to sixty-four cohort went into the financial crisis holding an average of 67% of their account assets in equities.246 Using U.S. stock market performance data based on the Wilshire 5000 for the one year period from October 8, 2007 (the peak of the U.S. markets) to October 9, 2008, they found that the average account value for that cohort dropped by 42%.247 Similarly, when researchers at the Urban Institute modeled the effects of the market crash they determined that most DC account holders born prior to 1945 will experience an 18.75% decrease in retirement account income even if the stock markets make a full recovery after ten years.248

The mandatory contribution of 9% of earnings may have helped to slightly mitigate the effect of the global financial crisis on the Australian superannuation system. Although the Australian S&P/ASX200 index dropped slightly more than the comparative U.S. equity indices, at 38.45% as compared to 37%,249 a study estimated the decline in total retirement assets in the U.S. at 18.9% as compared to a 16.2% decline in Australia.250 But, the short-term impact of the decline on Australian employees was a shock, just as it was in the U.S.

One study that modeled the effect of the financial crisis on older Australian employees estimated that those employees would need to raise their contribution levels significantly in order to retire with the account balances they would have accumulated in the absence of the crisis.251 For example, the modeling showed that an employee who was fifty-one years old at the end of the crisis would have to save 19% of earnings for the next nine years to offset the effect of the crisis.252 Assuming accounts returned long-term average results after the crisis, employees between the ages of thirty and forty at the end of the crisis would need to save an additional 1–2% per year until retirement to offset the impact of the crisis.253

2.     Default Investments: Responses to the Global Financial Crisis

In the wake of the financial crisis, U.S. regulatory concern about default investments focused on TDFs. The DOL and the SEC held joint hearings in 2009 to consider issues regarding TDFs that had surfaced during the economic downturn.254 The concerns expressed during those hearings included a perceived lack of understanding by investors, including 401(k) account holders, of the risks of TDFs.255 Similarly, SEC Chairman Mary L. Shapiro argued that TDF marketing materials might have caused investors to misunderstand the strategies undertaken by those funds.256

The Senate Special Committee on Aging (“Aging Committee”) issued a White Paper in late 2009 addressing TDF issues that went beyond disclosure problems.257 One might expect generic TDFs with equivalent target dates to have reasonably similar asset allocations. Yet, the Aging Committee reported that “the allocation of assets among stocks, bonds, cash-equivalents varied greatly among TDFs with the same target retirement date, with select firms’ 2010 TDFs’ equity holdings ranging anywhere from 24–68%.”258 As noted above, large allocations in equities performed poorly during the economic downturn, so not surprisingly, given the heterogeneity in asset allocations, the performance of TDFs during 2008 varied substantially.259 One explanation given for the differing levels of equity holdings is that the goals of funds with respect to their target date may range from achieving a maximum asset accumulation as of that date to managing to achieve a low or zero balance on that date because the date was defined as an expected mortality date.260 Recent research indicates that another cause of the variation in investment allocations is due to the increased risk appetite of new TDFs as compared to longer established TDFs.261

The second substantive weakness that the Aging Committee identified with TDFs was a variance in expense ratios, with a minimum of 0.19% and a maximum of 1.82%.262 Because of the long-term nature of 401(k) accounts, even small differences in fees can have an important effect on wealth accumulation.263 Some commentators believed that one explanation for higher fees in some TDFs was due to the layering of fees that occurs when TDFs invest in other funds or even in funds of funds.264 This is one front on which TDFs have made progress since the Aging Committee’s report. A 2012 study found that TDF fees have declined, perhaps because of economies of scale and competition.265

The regulatory response to the congressional investigation and agency hearings has been to increase education and disclosure efforts. First, in mid-2010 the DOL and SEC published a joint investor bulletin intended to educate investors about TDFs.266 In addition, the DOL reevaluated the disclosures provided when 401(k) plans use TDFs, particularly when they serve as QDIAs.267 The proposed regulatory revisions rely entirely on enhanced disclosure obligations. Specifically, the DOL’s proposed regulations, developed in collaboration with the SEC, would require that participants in TDFs, including participants defaulted into those funds designated as QDIAs, be provided with specific information about the TDF.268 The mandatory disclosure would have to discuss asset allocation and the way in which the allocation changes over time.269 It also would be obligated to address the fees and costs and include a warning that losses are possible in TDFs.270

Increasing the disclosure about TDFs, while perhaps the only tool available to the DOL without statutory amendment, fails to respond to the problems of varying asset allocations and high fees in a way that is consistent with the insights of choice architecture and behavioral economics. First, the investor passivity that is the basis for encouraging both automatic enrollment and default investments with appropriate levels of diversification means that increased disclosures and education will be ineffective in changing the behavior of many employees who are defaulted into those investments. One of the primary purposes of default settings is to recognize investor passivity and use defaults as nudges to achieve better results. An expectation that 401(k) account holders who are defaulted into TDFs will read disclosures, understand them, and act upon them ignores the behavioral economics research. In fact, the research on employee behavior regarding employer stock indicates that, to the extent that employees are not disengaged from investment decision-making, they may rely on their employer’s choice of a default investment as tacit guidance that the vehicle is a good investment.271 Second, a significant body of research indicates that “mandated disclosure as a remedy . . . is often ineffective.”272 Studies in behavioral branches of psychology, economics, and ethics as well as cognitive science indicate that disclosure often fails to enable the person receiving the disclosure to act rationally.273 And, in fact, mandatory disclosure may result in worse substantive behavior by the person providing the disclosure.274

Compare Australia’s very different approach to the regulation of default investment products in the post-global financial crisis era. The Australian reform explicitly relies on the principles of choice architecture to leverage investor passivity to increase wealth accumulation. It does this by imposing a combination of regulatory requirements and fiduciary-based obligations on MySuper products to ensure they are reasonably appropriate for the SG System account holders who are defaulted into those investments.

Although Australian employers will be obligated to choose a MySuper product that is registered as such with ASIC, those employers have no fiduciary obligation in the choice or oversight of the MySuper product. Instead, the trustee and board of the MySuper product bear not just the standard fiduciary obligations of fund trustees and board members, but also the enhanced responsibility to ensure that the investments and fees are appropriate for the employees whose retirement savings are invested in their MySuper product. Thus, in Australia, the responsibility for default investment vehicles coincides with both the locus of investment expertise and the responsibility for investment strategy. Or, to state it slightly differently, consistent with the traditional allocation of fiduciary responsibility, the investment-related obligations owed to those whose retirement assets are invested with MySuper funds will reside with those who have the expertise and power to manage the funds. In contrast, the U.S. approach has been to increase disclosure to employees, which is inconsistent with the point of QDIAs and the behavioral economics findings on investor passivity.

V.     Proposals for 401(k) Reform: Mandates

Commentators have long worried that the 401(k) system would fail to enable large numbers of Americans to accumulate sufficient assets to enjoy a secure retirement. This Part briefly summarizes some of the relevant data on that point. It then turns to a brief review of some of the alternative retirement savings systems other commentators have suggested. The common thread among those proposals is reliance on a government mandate that employers provide access to retirement savings vehicles, although the scope of the mandate varies.

A.     Inadequacy of the Current System

Although some 401(k) investors, either through prudent investing or through the luck of the draw, have created sufficient wealth for a secure retirement, the problem of too little wealth accumulation for too many is extensive and long-term. One problem is that due to inadequate and consistent savings patterns, employees simply do not accumulate enough assets in these plans.275 The implications for U.S. workers are best understood in terms of the overall lack of retirement readiness. An analysis by the Center for Retirement Research at Boston College concluded that households “at risk” of having retirement income at age sixty-five that was inadequate increased from 43% in 2004, to 44% in 2007, and then to 51% in 2009.276

The view of researchers looking from the outside in on retirement readiness is confirmed by the perspective of those considering their own prospects for retirement. The 2011 Retirement Confidence Survey reported that the confidence levels of Americans in their prospects for a comfortable retirement are at an all-time low.277 The number of workers who do not believe they will have enough money in retirement increased by five percentage points to 27% between 2010 and 2011.278 They expect this will dramatically affect their retirement lifestyle as 74% of current workers expect to work for pay during their retirement.279

B.     Mandatory Systems

1.     Guaranteed Retirement Accounts

Perhaps the most carefully developed and argued alternative to 401(k) plans is the proposal developed by Professor Teresa Ghilarducci. The system she advocates, Guaranteed Retirement Accounts (“GRAs”),280 would replace all DC plans, including 401(k) plans. An employer with a DB plan that meets specified minimum criteria would be permitted to substitute that plan for a GRA.281 Both employers and employees would make mandatory contributions of 2.5% of salary to fund GRAs (for a total of 5%).282 Every individual, regardless of income, who contributes to a GRA would receive a refundable federal tax credit of a flat amount, estimated at $600 annually.283 Individuals would have government-administered accounts, and investment decisions would be made by the Thrift Savings Plan (“TSP”), which administers and invests the DC accounts of federal employees.284

The investment vehicle associated with GRAs would differ significantly from the current TSP. Federal employees have the right in the TSP to allocate investments among a six-fund menu.285 Other than the fund comprised of U.S. Treasury securities, a private-sector financial services firm manages the funds.286 These are structured as typical mutual funds and vary in risk, and the accounts of federal employees reflect the investment gains or losses of the funds.287 It is unclear whether Professor Ghilarducci contemplates direct management of GRA investments by the Board of the TSP or by professional managers selected and monitored by that Board. Either way, she advocates a guaranteed annual investment return of 3% for the GRAs of individual workers.288 Unlike the existing funds in the TSP, the GRA assets would thus be protected and even continue to grow in periods of financial market downturns, but would experience limited returns when markets are thriving. At retirement, individuals could elect to receive up to 10% of their account balances as a lump sum, but the rest of the account balance would be paid as a life annuity.289

2.     Low- and Middle-Income Mandates

Professor Jeff Schwartz recently proposed a similar government-run system that also would replace 401(k) plans.290 In the Schwartz system, employers would be obligated to transmit employee contributions to an employee-designated investment account.291 Employers would be permitted, but not required, to contribute to the accounts.292 Because that is the extent of their involvement, employers would play a reduced role in this system as compared to 401(k)s.

Government, however, would provide a larger role because it would match the contributions of low- and middle-income workers293 and it would designate a private-sector fund manager to invest the assets.294 The level of the government match would equal the amount of the current tax subsidies.295 Left unspecified are: (1) any future adjustments to the government match; (2) whether the match would replace only 401(k) tax incentives, all private-sector employer-sponsored pension-related tax incentives or also the incentives that currently exist for Individual Retirement Accounts (“IRAs”); (3) the rate of match calculated as a factor of salary; and (4) the levels of earnings at which workers would be entitled to the match.

The Schwartz system relies on nudges for employee contributions and the investment designation. The low- and middle-income workers who would be entitled to a government match would be automatically enrolled in the system at a contribution rate that would entitle them to the maximum government match for their income level.296 Those employees would be permitted to opt-out of the system.297 High-income workers would be permitted to establish accounts298 and presumably would be entitled to elect to invest in the government-selected default fund. Other than the right to invest in that fund, it is unclear what incentive high-income individuals would have to participate in this system. Because the matches for low- and middle-income earners are to be funded using the entire amount of current tax incentives, the Schwartz system does not appear to contemplate any tax incentives to encourage high-income workers to save within the system.

The default investment product would consist of a portfolio made up of a U.S. equity index fund and treasury-inflation protected securities (“TIPS”).299 While not formally promising a guaranteed minimum-investment return, the use of TIPS is intended to provide a “guarantee[d] return of principal in real terms at retirement.”300 The allocation between equities and TIPS, and thus the effective guarantee, would vary according to employee age.301 The federal government would assume from employers the obligation to appoint the manager of the default fund.302 To reinforce the investment nudge, an employee who elects other than the government-selected default fund might lose some or all of the government matching contribution.303

In sum, Professor Schwartz advocates replacing the 401(k) system with one that would impose limited requirements on employers—to transmit contributions from employees who either elect to contribute to the new system or are defaulted into that system. The federal government would assume responsibility for funding matching contributions for low- and middle-income workers. A federal agency would be charged with designating and, although not discussed, presumably also with overseeing a single private-sector investment manager to run what would quickly become a multi-trillion dollar fund.304

3.     Universal, Secure, and Adaptable Retirement Funds

The most recent of the major reform proposals is that of Senator Tom Harkin, the Chairman of the Senate Health, Education, Labor & Pensions Committee. Unlike Professor Ghilarducci’s GRAs and Professor Schwartz’s system, Senator Harkin’s plan would exist within the current 401(k) system rather than replace it.305 Employers would be permitted to continue their current 401(k) or other retirement plans. An employer that does not offer a plan meeting the minimum criteria would be required to default employees into a newly created type of private-sector pension plan, a Universal, Secure, and Adaptable (“USA”) Retirement Fund.306 Under the default system, employees would be permitted to opt-out.307 Employers would have to make “modest contributions”308 on behalf of employees and low-wage workers would be entitled to government contributions.309 It is not clear whether employees who opt-out of the plan would receive the employer or government contributions. It also is not clear whether employees would have any choice among USA Retirement Funds; the only reference to choice in Senator Harkin’s plan is that employers choose the default fund.310

The innovative structure of USA Retirement Funds merges the administrative responsibilities of accepting contributions, calculating, and reporting benefit entitlements, investing assets, and paying benefits.311 All benefits would be paid as life annuities.312 There are indications that employees would have individual accounts because the proposal states: “The amount of a person’s monthly benefit would be determined based on the total amount of contributions made by, or on behalf of, the participant and investment performance over time.”313 But, the proposal also contemplates risk sharing, the type and amount of which is ambiguous. That risk sharing delegates to the trustees of each fund the flexibility to gradually increase or decrease benefits depending on investment performance.314 This obviously is incompatible with a system that calculates individual benefits based purely on account balances. The proposal contemplates that benefits would be “entirely portable”315 across USA Retirement Funds, but it is not clear how that portability would work in the context of risk pooling.

The fiduciary responsibility for USA Retirement Funds would lie with the fund trustees charged with plan management.316 Trustees would represent various constituencies: “employee[s], retiree[s], and employer representatives.”317 USA Retirement Funds would be licensed by an unspecified entity.318 Employers would not have any fiduciary liability for the selection of a USA Retirement Fund for their employees and, in fact, would be permitted to “use the ‘default’ fund identified for the region, industry, or through collective bargaining.”319 Presumably a federal agency would determine the default fund for various regions and industries.

In sum, the reform proposals set forth by Professors Ghilarducci and Schwartz and Senator Harkin differ in their details, but each would confront basic problems in the current 401(k) system. All three proposals would make some level of employer participation mandatory. All three proposals include at least default coverage for all workers of low- and middle-income. Senator Harkin’s plan would extend the default to all employees. Professor Ghilarducci’s plan would mandate contributions by and on behalf of all employees.

Each proposal also provides for an investment vehicle. In the systems suggested by Professors Ghilarducci and Schwartz, the government would be heavily involved in the selection and oversight of a single investment product. This alone is enough to be of concern to those familiar with the investment policy struggles of the Pension Benefit Guaranty Corporation (“PBGC”).320 Professor Ghilarducci would not permit any employee choice regarding the investment. Senator Harkin’s proposal provides for a variety of private-sector investment vehicles and, like Professor Ghilarducci’s, would require risk sharing across employees. But it does not appear that employees would have any choice of investment vehicle in USA Funds; instead, the fund designation is made by employers.

VI.     SHARPs: Nudges and Realignment of Fiduciary Responsibility

This Part begins by engaging with the ideology underlying the three reform proposals outlined in Part V. After arguing that it is useful to consider a less government-centric approach, I offer an alternative proposal.

A.     Government-Centric Retirement Security Provision in the United States

Each of the proposals outlined above relies on a government mandate. Mandates are consistent with the Australian SG System, which this Article looks to for default investment product principles that are consistent with choice architecture. But, to understand the ideology underlying mandates and universal coverage in the SG System, it is necessary to put the private-sector system in the context of Australia’s Age Pension. The public pension program in Australia, the Age Pension, is a welfare type of safety net system. Only retirees who pass both asset and income tests are entitled to receive Age Pension benefits.321 The universal system of earned retirement income is the SG System, which relieves the pressure on the Age Pension by ensuring that the vast majority of workers have retirement savings accounts.

In the U.S., it is Social Security that provides a basic retirement benefit for nearly all workers. 322 In that sense, Social Security is more like Australia’s SG System than its Age Pension. In fact, in some ways as a universal system Social Security is superior to the SG System. Social Security benefits are based on an earnings-related formula, which, unlike the SG System structure, protects pensioners against financial market fluctuations.323 Perhaps the most important factor in the consistent overall support that Social Security has had among the American populace is what one expert referred to as the “characteriz[ation of Social Security benefits] as an ‘earned right.’”324 Unlike the Australian Age Pension, ideologically Social Security is not a welfare plan, although it does have a redistributive aspect.325

The ideology underlying the three reform proposals discussed in Part V better aligns with the ideology of the Social Security system than with the private-sector employer-based system. Professor Ghilarducci’s proposal to mandate employer and employee contributions on behalf of all employees could be achieved through higher Social Security contribution rates. In addition, she suggests that the Social Security Administration administer her proposed GRAs. Again, like Social Security, her approach includes a benefit “guarantee.” Similarly, the guaranteed benefit and effort to enhance retirement income streams for low- to middle-income Americans that Professor Schwartz suggests could be met through increased Social Security contribution requirements for those individuals, perhaps even with an opt-out, or more redistributive calculations. Senator Harkin’s mandate that all employees be defaulted into a savings plan is the least easily wrapped into the Social Security system since he calls for individual accounts invested in a competitive environment of multiple private-sector investment vehicles. But, the conservative investment focus, risk sharing, and annuity requirements of Senator Harkin’s plan share aspects of the Social Security system.

These three proposed plans are purportedly intended to operate alongside Social Security. In fact, in addition to their private-sector reforms, Senator Harkin326 and Professor Ghilarducci327 both advocate strengthening Social Security. Professor Schwartz points to the political risk inherent in the government-funded nature of Social Security as a rationale in support of his proposal.328 Ironically, though, the overlap in ideology between these plans and Social Security may further imperil Social Security and the long-term viability of any of the three reform proposals that is adopted. The explicit or implicit federal guarantee on the investment of assets in these reform proposals creates substantial political risk. Even if the government does not explicitly back those guarantees, implicit expectations will exist based on the government’s regulatory role in establishing the guarantees and its involvement in investment decisions.329 Furthermore, the use of individual accounts in conjunction with mandates may provide fodder for those who believe Social Security should facilitate individual investment accounts.330 And, if every employee is defaulted into a savings type of account or, under Professor Ghilarducci’s plan is mandated to participate in such a plan, opponents of Social Security may argue that its universal coverage is much less important than in the current system of voluntary plan sponsorship. In short, any of these proposals may be viewed as “the” basic retirement system for American workers, replacing the role Social Security has played since its enactment.

B.     SHARPs: A Fiduciary Model Built on Choice Architecture

In this Subpart, I offer an alternative approach that is ideologically consistent with the traditional U.S. system of voluntary employer sponsorship. In fact, it would operate within the existing 401(k) system. Two characteristics of Australia’s approach to default investments inform this proposal. First, Australia has incorporated into the SG System the recognition that many people who opt, implicitly or explicitly, into the default investment products do not want to be actively involved in monitoring the investments in their accounts. Second, Australian reforms are structured to address the reality that employers also may not have the expertise or the inclination to become experts in investment product selection and monitoring. The administrative and investment products that I recommend, SHARPs, leverage both of those aspects of the Australian approach while retaining the philosophical goals of avoiding employer mandates and enabling some employee investment choice. This Subpart sketches the proposed regulatory framework for SHARPs, discusses how they would be integrated into the current 401(k) system, and explains how they would: (1) increase 401(k) plan sponsorship by decreasing barriers to employer plan sponsorship; (2) introduce greater integrity and appropriate diversification into default investment products; and (3) increase participation through the use of defaults.

1.     Decreased Barriers to Plan Sponsorship

SHARPs would replace QDIAs and, as an incentive for employers to use default settings in 401(k) plans, would provide employers with safe-harbor protection from fiduciary liability for the selection and monitoring of SHARPs. With SHARPs, employers would be relieved of the onerous task of qualifying and monitoring the myriad products they should consider as the default investment product for their plan. As well, they could be assured that they would not face the potential litigation costs or liability of the sort sought in the 401(k) employer stock and fee lawsuits.331

In order to further incent 401(k) plan sponsorship by the most reluctant group, small employers,332 those employers should be permitted to offer participant-directed 401(k) plans that offer only two SHARP products—one as the default product and a second to provide employees with an alternative. Most employers elect to sponsor 401(k) plans as participant-directed plans because, although the employer retains fiduciary liability for the selection and monitoring of the plan’s investment options, the employer has no liability for actual investment decisions an employee makes.333 As discussed above,334 to qualify as a participant-directed plan currently a 401(k) plan must offer a minimum of three investment options that together and separately meet a variety of requirements for the plan. Those standards would remain unchanged for medium and large employers.

The use of SHARPs to relieve employers of the costs and risks of fiduciary obligation in the choice of default investment products will remove that roadblock to plan sponsorships. SHARPs would provide a second incentive for small employers. In addition to investment management, each SHARP would be required to provide, as a fiduciary and at an employer’s option, all necessary administrative and reporting services. Small employers could rely on the two SHARP products for those services, freeing small employers of all ongoing administrative obligations and liability. Medium and large employers may find it more appropriate to retain a single plan administrator and only use a SHARP’s investment management services.

 2.     Greater Integrity and Appropriate Diversification of Investment Products

Elimination of employer fiduciary obligations for SHARPs will not result in a loss of protection for employees. In lieu of employer fiduciary obligation for SHARPs, I propose a two-part mechanism consisting of: (1) assigning fiduciary responsibility to the investment managers and fund directors that determine and implement a SHARP’s investment strategy; and (2) licensing by and reporting to a federal regulatory agency. The disclosure requirements would promote the ability to make competitive comparisons among SHARPs.

In addition to the reporting requirements, SHARPs would be restricted to only a limited set of features, including administrative features. As with the “financial services-centric” allocation of fiduciary duty, this limitation is modeled after the Australian reform. The notion underlying SHARPs is that they are default funds, with their primary market being employees who do not prefer to be deeply engaged in their 401(k) plan investment decision-making. Thus, there is no need to permit frequent transfers into and out of SHARPs, daily access to account balances, and other features that add costs and increase volatility. Requiring each SHARP to offer a uniform set of administrative features will provide economies of scale for the small employers who choose to rely on SHARPs to provide those services and promote competition through comparability.

The investment strategy of SHARPs is critical to employees’ wealth accumulation. SHARPs would be permitted to use any investment strategy that would currently meet the QDIA requirements.335 To drive investor-focused performance and low fees, the investment managers of SHARPs would have fiduciary liability to act in the best interest of the participants, including determination, disclosure, and implementation of an appropriate asset-allocation strategy. As a final check, the board members of a SHARP would be responsible for its compliance with regulatory standards and its disclosed strategy.

Admittedly, the restriction of investment strategies to those currently permitted of QDIAs will somewhat limit innovation in SHARPs. Gains in this tradeoff, however, are that the investment focus is on appropriate diversification and regulators will be better able to vet compliance with licensing requirements. Furthermore, the restriction on investment strategies will only limit the investment choices of employees of small employers who choose to rely on the participant-directed safe harbor of offering two SHARPs within their 401(k). Employees who are covered by other 401(k) plans will continue to have the opportunity to choose from the array of investments their plan offers, in addition to any SHARPs on their plan’s menu. Once participants enter the retirement-age distribution phase, they would be required to roll their assets into a non-SHARP investment vehicle. This will incentivize investment companies and annuity providers to develop innovative, competitive, and appropriate products for retirees.

In its efforts to protect individual investors, the current regulation of default investment products relies primarily on disclosure to employees as a supplement to employer fiduciary obligation in the selection and monitoring of default investment vehicles.336 In contrast, the SHARPs regime would rely on allocation of fiduciary responsibility to the investment managers that manage those products and on appropriate regulation and disclosure. In assessing the viability and importance of such a shift, consider the lessons realized from the use of default investment products during the financial crisis. Some TDFs incurred significant losses because they maintained substantial equity allocations, even for investors with near-term target retirement dates.337 The DOL’s short-term response, coordinated with the SEC, was to issue an investor bulletin explaining the risks of investing in TDFs.338 The bulletin contained three pages of potentially useful information in an easy-to-read format combining charts, questions, and answers.339 In the longer term, the DOL has been drafting enhanced disclosure guidelines that would require plans for employees with 401(k) assets invested in TDFs to include more information about those funds.340 Ultimately, that guidance and the required disclosures are likely to include valuable information for the plan sponsors and participants that read and understand them.

Addressing issues with default investment products through education and disclosure, however, is entirely inconsistent with the principles of a default regime. Behavioral economics and choice architecture show that no retirement system can or should rely on all individuals in the system acquiring and exercising the expertise required to make appropriate investment decisions. Congress at least implicitly recognized the contribution that choice architecture could make to wealth accumulation in the 401(k) system when it enacted, through the PPA, incentives for plans to implement automatic enrollment. There is nothing in the U.S. system of 401(k) and similar accounts that ensures that participants will read investor bulletins, disclosures delivered by their employers, or any other investment-related materials, let alone that they will understand that material or take action based on it. Research indicates that many participants do none of those things.341

The success of a defaults system, especially defaults into investment products, depends on the existence of appropriate default settings. It is inconsistent to, on one hand, argue that default settings are important because an overwhelming array of research shows that individuals are subject to biases, lack interest in becoming investment experts, etc., and then, when addressing potential issues with default settings, respond by providing information to those same individuals so they can determine whether the default settings are appropriate or not. By definition, the appropriate locus of decision-making in default settings is not the individual plan participant, and disclosures directed to those participants are likely to have limited effect. SHARPs address this by allocating fiduciary responsibility to the experts involved in investment decision-making and by establishing a regime of appropriate regulatory oversight.

3.     Increased Employee Participation in 401(k) Plans

SHARPS will increase the numbers of employees who participate in plans both because more employees, particularly at small employers, would have access to 401(k) plans and because employers that sponsor plans will be more likely to use automatic enrollment settings in their plans. More plans in existence will mean that more employees have the opportunity to contribute to 401(k) plans. Increased use of automatic enrollment will result in employees participating by default.

Although some plans had previously adopted automatic-enrollment provisions, the increased partial protection from fiduciary liability associated with QDIAs that resulted from the 2006 enactment of the PPA appears to be responsible for increasing the adoption of automatic enrollment.342 One survey found that in 2010, 41.8% of 401(k) plans used automatic enrollment.343 That is an increase from 38.4% in 2009 and a dramatic shift from 17% in 2005, just prior to enactment of the PPA.344 Obviously, though, adoption of automatic-enrollment features has been far from universal.

If the moderate levels of protection against fiduciary liability that PPA provided for protection had such a significant effect on employers’ adoption of automatic enrollment, it is reasonable to believe that the greater protection of SHARPs would also have a positive impact. This is especially true since any small employer adopting a 401(k) as a result of the SHARPs incentive for small employers would also be utilizing automatic enrollment. Empirically, we know that employee participation in 401(k) plans increases dramatically when plans adopt automatic enrollment.345

4.     The Benefits of SHARPs

In sum, SHARPs would not achieve the nearly 100% coverage that Australia has or that would exist under the three proposed reforms discussed in Part V. That is exactly the point—SHARPs are intended to work within the ideology and structure of the existing U.S. voluntary system of plan sponsorship. As an incremental change to the 401(k) regulatory framework, legislative implementation of SHARPs should be politically possible in the near term.

SHARPs would appeal to the key decision-makers in the 401(k) debate as SHARPs would provide benefits to the three major constituencies associated with 401(k) plans. For employers, SHARPs would provide a total safe harbor from liability associated with default investment products. For employees, the benefits from SHARPs are potentially three-fold. First, given increased protection from fiduciary liability for investment selection, a greater number of employers, particularly small employers, should be willing to sponsor 401(k) plans, providing more employees with access to those plans.346 Second, employers with 401(k) plans should be more likely to use automatic-enrollment settings because data clearly shows a dramatic increase in the rates at which employees contribute to 401(k) plans.347 Third, rather than investing their plan assets in an undiversified manner, which tends to result from a series of risk-inducing factors, including a lack of financial expertise, lack of interest, and a variety of investment biases and errors,348 employees who do not wish to be active in managing their retirement accounts will be invested in an appropriately diversified retirement product. The third constituency, a powerful voice in any debate over reform, is the product providers. Unlike in Professors Ghilarducci’s and Schwartz’s proposals for government-run investment programs, in this system the financial services industry would be free to innovate and create products that would spur wealth creation for workers and efficient capital allocation, which would be subject to appropriate regulation and fiduciary obligations to employee-investors. Nor would investment approaches be arbitrarily limited, as in Senator Harkin’s proposal, to conservative investments.

VII.     Conclusion

The creation of long-term wealth for the majority of U.S. employees is dependent in large part on the system of private-sector employer-sponsored DC plans, particularly 401(k) plans. In the current system, too few employees have access to 401(k) plans and the assets of too many employees who participate in 401(k) plans are not optimally invested. To address these problems, I propose that the locus of fiduciary responsibility for default investments be reallocated from employers to the financial services firms that offer those investment vehicles. This would involve the creation and regulation of a new type of investment product: SHARPs. Behavioral economics research and principles of choice architecture provide the theoretical foundation for SHARPs.

Every day, approximately 10,000 Americans turn sixty-five.349 For many of them, that date or some date soon will represent the end of their time as wage earners. Two-thirds of them worry about not having enough money for retirement.350 A quarter of workers, in one study, admitted to not even opening their 401(k) statement for fear of receiving bad news.351 Of those who did open their statements, almost three quarters spent less than three minutes reviewing them.352

Given the uncertainty of the financial and job markets, the limited availability of retirement plans, and the lack of engagement by many employees with their 401(k) accounts, a shift in approach is needed. Tweaks to the system cannot remedy the extensive gap; but, replacement of the 401(k) system with mandates and government-run investment vehicles is not ideologically consistent with the U.S. reliance on Social Security as the mandatory government-run pension system.

The proposal made here would provide the means to implement a solution that benefits not only employees, but also their employers and the financial professionals and investment companies that service 401(k) plans. With modifications to the reform in Australia tailored to the unique American environment and ideology, we can make significant progress in driving wealth creation, preservation, and growth. American employees work too hard to see their retirements in peril.

  1. [1]. D’Vera Cohn & Paul Taylor, Baby Boomers Approach Age 65 – Glumly, Pew Res. Soc. & Demographic Trends (Dec. 20, 2010),

  2. [2]. See, e.g., Peter J. Henning, Making Sure “The Buck Stops Here”: Barring Executives for Corporate Violations, 2012 U. Chi. Legal F. 91, 91 (internal quotation marks omitted) (“The movement called ‘Occupy Wall Street’ has sought to take over locations in New York City and elsewhere to protest what it sees as corporate greed and corruption that have led to a growing inequality between powerful moneyed interests and ‘the other 99 percent.’”).

  3. [3]. See Alan Lavine, New Opportunities with 401(k)s, Fin. Advisor Mag. (Nov. 1, 2010),; Dan Lewerenz, The Man Behind the 401(k), ABC News (Jan. 18, 2012),

  4. [4]. Retirement Assets Total $18.9 Trillion in First Quarter 2012, ICI Global (June 28, 2012),

  5. [5]. 401(k) plans are a type of defined contribution (“DC”) plan. For an explanation of DC plans and how they differ from defined benefit (“DB”) plans, see infra text accompanying notes 39–42, 165–68. Many of the concepts discussed in this Article could be extended to other types of defined contribution plans, including those sponsored by governmental entities. For purposes of scope, I focus the discussion here on 401(k) plans.

  6. [6]. World Fed’n of Exchs., 2011 WFE Market Highlights 6 (2012), available at
    %20Highlights.pdf (reporting New York Stock Exchange capitalization of almost $12 billion).

  7. [7]. See infra text accompanying notes 28–34.

  8. [8]. See infra text accompanying notes 54–64.

  9. [9]. See infra text accompanying notes 47–51.

  10. [10]. See infra text accompanying notes 84–100.

  11. [11]. See, e.g., Colleen E. Medill, Targeted Pension Reform, 27 J. Legis. 1, 3 (2001) (proposing closure of loopholes in the tax system that result in benefits being lower than they otherwise would be for lower wage workers); Michael W. Melton, Making the Nondiscrimination Rules of Tax-Qualified Retirement Plans More Effective, 71 B.U. L. Rev. 47, 50 (1991) (arguing that tax incentives are not sufficient to induce low-income workers to save for retirement); see also Paul M. Secunda, 401K Follies: A Proposal to Reinvigorate the United States Annuity Market, 30 ABA Sec. Tax’n News Q. 13, 14–15 (2010) (arguing for tax law changes to require 401(k) plans to offer annuitized distribution options).

  12. [12]. See, e.g., Michael J. Collins, Reviving Defined Benefit Plans: Analysis and Suggestions for Reform, 20 Va. Tax Rev. 599, 602 (2001) (advocating simplification of the tax laws governing defined benefit plans in order to encourage plan sponsorship); Barry Kozak, The Cash Balance Plan: An Integral Component of the Defined Benefit Plan Renaissance, 37 J. Marshall L. Rev. 753, 800–04 (2004) (arguing that cash balance plans could ensure reliable lifetime income for retirees and reduce employer risk).

  13. [13]. See, e.g., Teresa Ghilarducci, When I’m Sixty-Four: The Plot Against Pensions and the Plan to Save Them 260–74 (2008).

  14. [14]. Jeff Schwartz, Rethinking 401(k)s, 49 Harv. J. on Legis. 53, 74–78 (2012).

  15. [15]. Tom Harkin, U.S. Senate Comm. on Health Educ. Labor & Pensions, The Retirement Crisis and a Plan to Solve It 5–7 (2012), available at

  16. [16]. See id. (proposing to relieve employers of fiduciary obligation if they use the new fund structure). Differences between my proposal and Senator Harkin’s plan are discussed infra throughout Part VI.

  17. [17]. See infra Part VI.B.4.

  18. [18]. Richard H. Thaler & Cass R. Sunstein, Nudge: Improving Decisions About Health, Wealth, and Happiness 6 (2008). A “nudge,” according to the authors, is “any aspect of the choice architecture that alters people’s behavior in a predictable way without forbidding any options or significantly changing their economic incentives.” Id.

  19. [19]. Employee Retirement Income Security Act (ERISA) of 1974, Pub. L. No. 93-406, 88 Stat. 829 (codified as amended at 29 U.S.C. §§ 1001–1461 (2006)).

  20. [20]. Although the name bears a resemblance, there is no relationship intended with the well-known “Sharpe ratio” used in analyzing investments. See Houman B. Shadab, The Law and Economics of Hedge Funds: Financial Innovation and Investor Protection, 6 Berkeley Bus. L.J. 240, 264 n.152 (2009) (“The Sharpe ratio is the most common way of measuring risk-adjusted returns.”). I also hope it does not suffer any negative connotations from the pejorative “sharp business practice.” See, e.g., Jeremy A. Rabkin, Recalling the Case for Sovereignty, 5 Chi. J. Int’l L. 435, 440 (2005) (referring to “‘economic coercion’ from sharp business practices”).

  21. [21]. Thaler & Sunstein, supra note 18, at 3.

  22. [22]. Id.

  23. [23]. See id.

  24. [24]. Swee-Hoon Chuah & James Devlin, Behavioural Economics and Financial Services Marketing: A Review, 29 Int’l J. Bank Marketing 456, 457 (2011).

  25. [25]. See generally Sendhil Mullainathan & Richard H. Thaler, Behavioral Economics (MIT Dept. of Econs., Working Paper No. 00-27, 2000), available at
    papers.cfm?abstract_id=245828 (providing an overview of behavioral economics).

  26. [26]. See, e.g., Chuah & Devlin, supra note 24, at 457–58 (listing the factors covered in their review of financial services marketing).

  27. [27]. See, e.g., Thaler & Sunstein, supra note 18, at 108–11; Olivia S. Mitchell & Stephen P. Utkus, Lessons from Behavioral Finance for Retirement Plan Design 1–25 (Pension Research Council, Working Paper No. 2003-6, 2003), available at

  28. [28]. Shlomo Benartzi & Richard H. Thaler, Risk Aversion or Myopia? Choices in Repeated Gambles and Retirement Investments, 45 Mgmt. Sci. 364, 375 (1999).

  29. [29]. See Susan J. Stabile, Freedom to Choose Unwisely: Congress’ Misguided Decision to Leave 401(k) Plan Participants to Their Own Devices, 11 Cornell J.L. & Pub. Pol’y 361, 376 (2002).

  30. [30]. See Richard H. Thaler & Shlomo Benartzi, Save More Tomorrow™: Using Behavioral Economics to Increase Employee Saving, 112 J. Pol. Econ. S164, S168 (2004).

  31. [31]. Id.

  32. [32]. Thaler & Sunstein, supra note 18, at 112.

  33. [33]. Id.

  34. [34]. Id.

  35. [35]. See infra text accompanying notes 167–68 (discussing the prevalence of 401(k) plans).

  36. [36]. The standard 401(k) plans discussed in this Article must be sponsored by employers, who then nearly always act as the plan sponsors. See 29 U.S.C. § 1002(1) (2006) (requiring plans to be sponsored by an employer or an employee organization). Therefore, the terms employer and plan sponsor are used interchangeably.

  37. [37]. See Dana M. Muir, From Yuppies to Guppies: Unfunded Mandates and Benefit Plan Regulation, 34 Ga. L. Rev. 195, 209–11 (1999) (discussing the history of voluntary plan sponsorship in the U.S.).

  38. [38]. Retirement Assets Total $18.9 Trillion in First Quarter 2012, supra note 4 (reporting as of Mar. 31, 2012).

  39. [39]. See Dana M. Muir & John A. Turner, Constructing the Ideal Pension System: The Visions of Ten Country Experts, in Imagining the Ideal Pension System: International Perspectives 1, 4–10 (Dana M. Muir & John A. Turner eds., 2011) (discussing the pension systems in a number of countries as defined benefit (“DB”) or defined contribution plans); see also infra text accompanying notes 159–67 (describing DB plans).

  40. [40]. Dana M. Muir, Plant Closings and ERISA’s Noninterference Provision, 36 B.C. L. Rev. 201, 205 (1995).

  41. [41]. Employee Benefits Law 255 (Dana M. Muir ed., 2d ed. Supp. 2010).

  42. [42]. See infra Part II.B.2.

  43. [43]. Muir & Turner, supra note 39, at 24.

  44. [44]. Id.

  45. [45]. Kathryn L. Moore, An Overview of the U.S. Retirement Income Security System and the Principles and Values It Reflects, 33 Comp. Lab. L. & Pol’y J. 5, 18 n.104 (2011).

  46. [46]. See infra text accompanying notes 59–61..

  47. [47]. Alicia H. Munnell et al., Ctr. for Ret. Research at Bos. Coll., The Pension Coverage Problem in the Private Sector 1 (2012), available at

  48. [48]. See id. at 2 fig.1.

  49. [49]. Alicia H. Munnell, Ctr. for Ret. Research at Bos. Coll., 401(k) Plans in 2010: An Update from the SCF 3 (2012), available at

  50. [50]. Id.

  51. [51]. William J. Wiatrowski, Changing Landscape of Employment-Based Retirement Benefits, U.S. Bureau Lab. Stat. (Sept. 29, 2011),

  52. [52]. See Richard L. Kaplan, Enron, Pension Policy, and Social Security Privatization, 46 Ariz. L. Rev. 53, 66 (2004) (“Most 401(k) plans require affirmative enrollment by employees . . . .”).

  53. [53]. See, e.g., Karen C. Burke & Grayson M.P. McCouch, Social Security Reform: Lessons from Private Pensions, 92 Cornell L. Rev. 297, 308 (2007) (“In making decisions about participation . . . more than a quarter of all eligible employees do not do so at all.”); James M. Poterba, Individual Decision Making and Risk in Defined Contribution Plans, 13 Elder L.J. 285, 307 (2005) (“[S]ome 401(k) plan participants make decisions that . . . fail to take full advantage of the opportunities for 401(k) plans to contribute to their retirement income security.”).

  54. [54]. See infra text accompanying notes 59–61, 66.

  55. [55]. See Vada Waters Lindsey, Encouraging Savings Under the Earned Income Tax Credit: A Nudge in the Right Direction, 44 U. Mich. J.L. Reform 83, 113 (2010) (explaining the availability of sample provisions to employers adopting or amending a 401(k) plan).

  56. [56]. See supra text accompanying note 52.

  57. [57]. Default settings are not the only way that the configuration of plan terms may be used to affect employee participation. Other basic plan terms, such as whether the employer “matches” the contributions made by employees or otherwise contributes to the plan are affected by complex rules intended to ensure that 401(k) plans are fairly available and used across a broad spectrum of employees, not just by those who are highly compensated. Susan J. Stabile, Is It Time to Admit the Failure of an Employer-Based Pension System?, 11 Lewis & Clark L. Rev. 305, 318 n.60 (2007).

  58. [58]. Moore, supra note 45, at 21.

  59. [59]. Sarah Holden & Jack VanDerhei, The Influence of Automatic Enrollment, Catch-Up, and IRA Contributions on 401(k) Accumulations at Retirement, Inv. Co. Inst. Persp., July 2005, at 1, 4, available at

  60. [60]. Id.

  61. [61]. Jack VanDerhei, What Do You Call a Glass That is 60–85% Full?, Emp. Benefit Res. Inst. (July 7, 2011),

  62. [62]. For a discussion of the Internal Revenue Code rules governing maximum pre-tax deferrals, see Employee Benefits Law, supra note 41, at 256–59. Though opinions will vary, and what one individual considers necessary another might not, the fact is that few employees are contributing at the maximum rate allowed by tax law. In fact, only 8.4% of those who participate in 401(k)s do so at the maximum amount. Alicia H. Munnell & Annika Sunden, Coming Up Short: The Challenge of 401(k) Plans 62 tbls.3 & 4 (2004). Failure to participate at the maximum, by itself, does not necessarily indicate inadequate contributions, but it does point to a failure to take advantage of the full tax advantages available.

  63. [63]. Poterba, supra note 53, at 290 (discussing employer matching contributions).

  64. [64]. See, e.g., Thaler & Benartzi, supra note 30, at S171–79 (evaluating three approaches to automatically increasing employee contributions).

  65. [65]. Anne Tergesen, 401(k) Law Suppresses Saving for Retirement, Wall St. J. (July 7, 2011),; see also Thaler & Benartzi, supra note 30, at S169 (discussing research indicating that automatic enrollment may result in decreased savings rates).

  66. [66]. Tergesen, supra note 65.

  67. [67]. Thaler & Benartzi, supra note 30, at S169.

  68. [68]. VanDerhei, supra note 61.

  69. [69]. Id.

  70. [70]. Id.

  71. [71]. Steven D. Cohen, Autoenrollment and Annuitization: Enabling the 401(k) “DB-ation,5 N.Y.U. J.L. & Bus. 281, 303 (2009) (explaining automatic escalation).

  72. [72]. See Thaler & Benartzi, supra note 30, at S170 (advocating and testing a slightly different approach where employees elect in advance to contribute portions of future pay raises).

  73. [73]. Jack L. VanDerhei, The Expected Impact of Automatic Escalation of 401(k) Contributions on Retirement Income, EBRI Notes, Sept. 2007, at 2, 6, available at

  74. [74]. S. Kathi Brown, Automatic 401(k) Plans: Employer Views on Enrolling New and Existing Employees, AARP (June 2010),

  75. [75]. Investment of account assets in specific investment vehicles occurs at the account level. This is distinguished from the choices made by employers at the plan level regarding what investment vehicles are available to receive investments. The plan-level choices are discussed infra at text accompanying notes 170–78.

  76. [76]. Burke & McCouch, supra note 53, at 308 (“401(k) participants often make objectively bad investment decisions . . . .”).

  77. [77]. Debra A. Davis, How Much Is Enough? Giving Fiduciaries and Participants Adequate Information About Plan Expenses, 41 J. Marshall L. Rev. 1005, 1008 (2008).

  78. [78]. Id. (explaining that approximately 89% of 401(k) plans are participant-directed at least in part).

  79. [79]. See infra Part IV.A.3.

  80. [80]. This investment default may include only the “employee’s” contribution or may also include the amount matched or otherwise contributed by the employer.

  81. [81]. See Cohen, supra note 71, at 310 (noting that a life-cycle fund may be both a default and an investment option).

  82. [82]. See infra text accompanying notes 93-94.

  83. [83]. See supra text accompanying note 78 (defining participant-directed plans).

  84. [84]. Dana M. Muir, The Dichotomy Between Investment Advice and Investment Education: Is No Advice Really the Best Advice?, 23 Berkeley J. Emp. & Lab. L. 1, 10 (2002).

  85. [85]. Shlomo Benartzi & Richard H. Thaler, Naive Diversification Strategies in Defined Contribution Saving Plans, 91 Am. Econ. Rev. 79, 81 (2001).

  86. [86]. Id. at 82.

  87. [87]. Id.

  88. [88]. Id.

  89. [89]. Id.

  90. [90]. Id. at 79, 82. This tendency is reportedly traceable at least to the Talmud, which advises that an investor should divide assets “a third into land, a third into merchandise, and . . . a third at hand.” Id. at 79.

  91. [91]. Dr. Markowitz co-won the Nobel Prize in Economic Science with Drs. Merton Miller and William Sharpe in 1990 for their work in balancing risk and return through diversification. See Hal Varian, A Portfolio of Nobel Laureates: Markowitz, Miller and Sharpe, 7 J. Econ. Persp. 159, 159 (1993).

  92. [92]. Thaler & Sunstein, supra 18, at 123.

  93. [93]. Deloitte et al., Annual 401(k) Benchmarking Survey: Plan Sponsors and Providers Work at Closing the Retirement Readiness Gap While Getting Ready for New Fee Disclosure Regulations 20 (2011),

  94. [94]. Id. at 21.

  95. [95]. See supra text accompanying note 90.

  96. [96]. Sheena S. Iyengar et al., How Much Choice Is Too Much?: Contributions to 401(k) Retirement Plans 9-10 (Pension Research Council, Working Paper No. 2003-10, 2003), available at

  97. [97]. Id. at 9.

  98. [98]. Id.

  99. [99]. Thaler & Sunstein, supra note 18, at 127.

  100. [100]. Id. The over-allocation to employer stock in this situation also may reflect confirmation bias, as investors convince themselves that the employer stock their employer requires them to hold is a good investment. Gary Belsky & Thomas Gilovich, Why Smart People Make Big Money Mistakes and How to Correct Them: Lessons from the New Science of Behavioral Economics 129-35 (1999) (discussing confirmation bias). Confirmation bias theory predicts that once people develop a belief or opinion, they will interpret new data to best support the prior belief or opinion. Richard A. Posner, An Economic Approach to the Law of Evidence, 51 Stan. L. Rev. 1477, 1495 (1999).

  101. [101]. Thaler & Sunstein, supra 18, at 126.

  102. [102]. Susan J. Stabile, Another Look at 401(k) Plan Investments in Employer Securities, 35 J. Marshall L. Rev. 539, 542-43 (2002). Over-investment in employer stock in 401(k) plans is not a thing of the past. Reportedly, 17% of Best Buy 401(k) assets in 2012 were invested in employer stock although it was not a successful year for Best Buy. See Edward Siedle, Best Buy’s 401(k) Meltdown, Forbes (Nov. 23, 2012, 12:13 PM),

  103. [103]. See infra text accompanying notes 169-75.

  104. [104]. See infra text accompanying notes 268–70.

  105. [105]. In 1994, the World Bank proposed a model pension system. It came to be known as the “three pillar” model and relied on three sources of pension income, which together would provide sustainable and sufficient benefits. Those sources are: (1) a state-run defined benefit system; (2) an occupational-based system; and (3) personal savings. The World Bank, Averting the Old Age Crisis 238-39, 291 (1994).

  106. [106]. Dana M. Muir, Building Value in the Australian Defined Contribution System: A Values Perspective, 33 Comp. Lab. L. & Pol’y J. 93, 109 (2011).

  107. [107]. For more extensive discussion of all three components of the Australian pension system, see id. at 97-114.

  108. [108]. See A Recent History of Superannuation in Australia, APRA Insight, no.2, 2007, at 3–4, available at Australians typically use the term “superannuation” to refer to retirement. One explanation is that pensions are thought to be annuity streams and that lump sums historically were more common in Australia than annuities. Int’l Soc. Sec. Ass’n et al., Complementary and Private Pensions Throughout the World 2008, at 502 (2008).

  109. [109]. A more detailed history can be found at Muir, supra note 106, at 97-100. Industry awards are industry-wide sets of employment terms determined through Australia’s negotiation and arbitration process. See Shingo Takahashi, How Multi-Tasking Job Designs Affect Productivity: Evidence from the Australian Coal Mining Industry, 64 Indus. & Lab. Rel. Rev. 841, 843 (2011).

  110. [110]. Technically, contributions to the SG System are not mandatory. Instead, Australian law requires an employer who fails to make the minimum contribution to pay a charge (tax) to the government that is higher than the minimum contribution. As a result, there is a clear incentive to make the minimum contribution and commentators typically refer to the system as one of mandatory contributions. See A Recent History of Superannuation in Australia, supra note 108, at 4.

  111. [111]. Hazel Bateman et al., Forced Saving: Mandating Private Retirement Incomes 190 (2001).

  112. [112]. Austl. Gov’t, Superannuation—Increasing the Superannuation Guarantee Rate to 12 Per Cent 1 (2011), available at
    downloads/Fact_sheet_SG_rate_increase.pdf; see also Super Sys. Review, Final Report—Part Two: Recommendation Packages 297 (2010) [hereinafter Cooper Report, Part II], available at

  113. [113]. “Members” is the term used in Australia for those who have SG System accounts. It is generally synonymous with the U.S. term “participant.” For ease of reading here, the text typically refers to participants and members as “employees.”

  114. [114]. See Terry Carney, The Future of Welfare Law in a Changing World: Lessons from Australia and Singapore, 32 Sydney L. Rev. 193, 203 (2010).

  115. [115]. Int’l Soc. Sec. Ass’n et al., supra note 108, at 502-04. On July 1, 2013, Australia increased the age limit to age seventy-five. Austl. Gov’t, Superannuation—Raising the Superannuation Guarantee Age Limit from 70 to 75, at 1 (2011), available at http://

  116. [116]. Int’l Soc. Sec. Ass’n et al., supra note 108, at 503. Salary sacrifice is encouraged through favorable tax treatment. Hazel Bateman, Retirement Incomes in Australia in the Wake of the Global Financial Crisis, in Protecting Pension Rights in Times of Economic Turmoil 63, 68 (Yves Stevens ed., 2011).

  117. [117]. See, e.g., Super in Enterprise Agreements, Austl. Council Trade Unions, http://www. (last visited Sept. 16, 2013) (“Through union collective bargaining, 23% of the workforce have now achieved superannuation that is above the minimum 9% contribution.”).

  118. [118]. Trends in Superannuation Coverage, Austl. Soc. Trends, March 2009, at 39, 43, available at

  119. [119]. Id. at 41.

  120. [120]. Towers Watson, Global Pension Assets Study 2012, at 12 (2012), available at
    Global-Pensions-Asset-Study-2012. The other countries included in the study were: Brazil, Canada, France, Germany, Hong Kong, Ireland, Japan, the Netherlands, South Africa, Switzerland, the United Kingdom, and the United States. Id. at 3.

  121. [121]. Id. at 7.

  122. [122]. See Muir, supra note 106, at 99.

  123. [123]. Id. at 100.

  124. [124]. Id. at 101.

  125. [125]. Enterprise agreements are labor agreements between a specific company and its employees as opposed to industry-wide labor agreements. See Breen Creighton & Pam Nuttall, Good Faith Bargaining Down Under, 33 Comp. Lab. L. & Pol’y J. 257, 265 (2012).

  126. [126]. See Josh Fear & Geraldine Pace, Choosing Not to Choose: Making Superannuation Work by Default 10 (The Austl. Inst. Indus. Super Network, Discussion Paper No. 103, 2008), available at

  127. [127]. Austl. Prudential Regulation Auth., Statistics: Annual Superannuation Bulletin 44 tbl.18 (2011) (revised Jan. 9, 2013), available at
    Super/Publications/Documents/June%202011%20Revised%20Annual%20Superannuation%20Bulletin.pdf; see also Muir, supra note 106, at 100-01, 115.

  128. [128]. Austl. Prudential Regulation Auth., supra note 127, at 43 tbl.17.

  129. [129]. Super Sys. Review, Final Report—Part One: Overview and Recommendations v (2010) [hereinafter Cooper Report, Part I], available at http://www.supersystemreview.

  130. [130]. Id. at vi. For Jeremy Cooper’s credentials and a list of the review panel members and other contributors, see id. at 64-66.

  131. [131]. Id. at v-vi.

  132. [132]. Id. at 10.

  133. [133]. See generally Commonwealth of Austl., Stronger Super (2010), available at
    Stronger_Super.pdf. In parallel, Australia is in the process of reforming its regulation of financial advisers, including those who provide advice to SG members. See ASIC’s Plans for FoFA Reforms, Austr. Sec. & Inv. Commission (Dec. 13, 2011),

  134. [134]. See Austl. Gov’t, Stronger Super: Information Pack (2011), available at http://

  135. [135]. Superannuation Legislation Amendment (Trustee Obligations and Prudential Standards) Act 2012 (Cth) (Austl.) [hereinafter Proposed Trustee and Prudential Standards], available at http://; Superannuation Legislation Amendment (MySuper Core Provisions) Bill 2011 (Cth) (Austl.) [hereinafter Proposed MySuper Core Provisions], available at

  136. [136]. See, e.g., Austl. Prudential Regulation Auth., Instruction Guide: Application Form—Authority to Offer a MySuper Product (2012), available at

  137. [137]. Cooper Report, Part I, supra note 129, at 8-9.

  138. [138]. See supra text accompanying notes 28-31.

  139. [139]. Cooper Report, Part I, supra note 129, at 8-9.

  140. [140]. Id. at 9.

  141. [141]. Id. at 10; see also M. Scott Donald, What’s in a Name? Examining the Consequences of Inter-Legality in Australia’s Superannuation System, 33 Sydney L. Rev. 295, 301 (2011).

  142. [142]. Cooper Report, Part I, supra note 129, at 10.

  143. [143]. Id.

  144. [144]. See Austl. Gov’t, supra note 134, at v-vi.

  145. [145]. 29 U.S.C. § 1104(a)(1)(A)(i) (2006 & Supp. V 2011); cf. Daniel Fischel & John H. Langbein, ERISA’s Fundamental Contradiction: The Exclusive Benefit Rule, 55 U. Chi. L. Rev. 1105, 1108 (1988) (“ERISA’s exclusive benefit rule . . . imports into pension fiduciary law one of the most fundamental and distinctive principles of trust law, the duty of loyalty.”).

  146. [146]. See 29 U.S.C. § 1104(a)(1)(B); see also 29 C.F.R. § 2550.404a-1 (1997) (explaining the application of the prudence standard to investment duties).

  147. [147]. See 29 U.S.C. § 1104(a)(1)(C).

  148. [148]. See id. § 1104(a)(1)(D).

  149. [149]. Mass. Mut. Life Ins. Co. v. Russell, 473 U.S. 134, 152 n.6 (1985) (Brennan, J., concurring) (“The principles of fiduciary conduct are adopted from existing trust law, but with modifications appropriate for employee benefit plans.” (quoting H.R. Rep. No. 93-533, at 13 (1973) (internal quotation marks omitted))).

  150. [150]. Restatement (Second) of Trusts § 3(3) (1959) (“The person holding property in the trust is the trustee.”).

  151. [151]. Actions defined as fiduciary acts include: “render[ing] investment advice for a fee or other compensation, direct or indirect, with respect to any moneys or other property of [a] plan, or has any authority or responsibility to do so . . . .” 29 U.S.C. § 1002(21)(A)(ii) (2006). Fiduciary acts also include discretionary authority over plan administration, or management of the plan or plan assets. Id. § 1002(21)(A)(i), (iii).

  152. [152]. Muir, supra note 84, at 20.

  153. [153]. 29 U.S.C. § 1002(21)(A)(i)-(iii).

  154. [154]. See, e.g., Andrew S. Hartley, Making the Case for Mandatory Removal of Imprudent Investment Vehicles: Inside Information Can Make Employer Securities a Bad 401(k) Option, 5 Appalachian J.L. 99, 108 (2006) (“An employer/trustee is always subject to the duty to monitor . . . .”).

  155. [155]. Id.

  156. [156]. Dana M. Muir & Cindy A. Schipani, Fiduciary Constraints: Correlating Obligation with Liability, 42 Wake Forest L. Rev. 697, 702 (2007) (“Plans may designate individuals or entities, such as the corporation or a plan committee, as the named fiduciary.”).

  157. [157]. James A. Wooten, The Employee Retirement Income Security Act of 1974: A Political History 118 (2004).

  158. [158]. Id. at 257.

  159. [159]. The Treasury Department first issued regulations under I.R.C. § 401(k) in late 1981 that established the parameters of what have come to be known as “401(k) plans.” See Employee Benefits Law, supra note 41, at 254–55 (discussing the term “401(k) plan”); see also, e.g., 26 C.F.R. § 1.401(k)–1 (2011) (setting forth the basic requirements for 401(k) plans).

  160. [160]. See Muir, supra note 40, at 205-06.

  161. [161]. See Wooten, supra note 157, at 51 (discussing the failure of Studebaker’s pension plan).

  162. [162]. See id. at 122 (discussing the trust basis of ERISA’s fiduciary standards).

  163. [163]. Muir, supra note 84, at 4–5.

  164. [164]. See Tamar Frankel, Fiduciary Law in the Twenty-First Century, 91 B.U. L. Rev. 1289, 1293 (2011).

  165. [165]. See, e.g., Ghilarducci, supra note 13, at 58–115; Thomas J. Mackell, Jr., When the Good Pensions Go Away: Why America Needs a New Deal for Pension and Health Care Reform 47–59 (2008); Burke & McCouch, supra note 53, at 302–04; Jack VanDerhei & Craig Copeland, The Changing Face of Private Retirement Plans, EBRI Issue Brief, Apr. 2001, at 1, 5–6, available at

  166. [166]. Wiatrowski, supra note 51.

  167. [167]. Retirement Assets Total $18.9 Trillion in First Quarter 2012, supra note 4 (reporting as of Mar. 31, 2012).

  168. [168]. Id. Additionally, Americans had $5.2 trillion invested in Individual Retirement Accounts (“IRAs”). Id. (reporting as of Mar. 31, 2012).

  169. [169]. See supra text accompanying note 36.

  170. [170]. See supra text accompanying notes 75–77.

  171. [171]. See supra text accompanying note 77.

  172. [172]. See, e.g., Davis, supra note 77, at 1011–15.

  173. [173]. See infra text accompanying notes 195–200.

  174. [174]. See, e.g., Susan J. Stabile, I Believed My Employer and Didn’t Sell My Company Stock: Is There an ERISA (or ’34 Act) Remedy for Me?, 36 Conn. L. Rev. 385, 386–87 (2004); see also Dana M. Muir & Cindy A. Schipani, New Standards of Director Loyalty and Care in the Post-Enron Era: Are Some Shareholders More Equal than Others?, 8 N.Y.U. J. Legis. & Pub. Pol’y 279, 331–43 (2005).

  175. [175]. See infra text accompanying notes 177–80.

  176. [176]. See José Martin Jara, What Is the Correct Standard of Prudence in Employer Stock Cases?, 45 J. Marshall L. Rev. 541, 545 (2012) (referring to the “congressional objective of encouraging employee ownership”); Janice Kay McClendon, The Death Knell of Traditional Defined Benefit Plans: Avoiding a Race to the 401(k) Bottom, 80 Temp. L. Rev. 809, 814 n.36 (2007) (discussing Employee Stock Ownership Plans).

  177. [177]. U.S. Gov’t Accountability Office, Private Pensions: Information That Sponsors and Participants Need to Understand 401(k) Plan Fees 15 n.20 (2007), available at

  178. [178]. See Deloitte Consulting LLP, Inside the Structure of Defined Contribution/401(k) Plan Fees: A Study Assessing the Mechanics of the ‘All-In’ Fee 5–6 (2011), available at

  179. [179]. See id. at 22 (finding that the median annual fees are 0.78% of plan assets, but the fees at the 10th percentile were 0.28% and 1.38% at the 90th percentile).

  180. [180]. See, e.g., Emily Adams, Protecting America’s Financial Future: Why Courts Should Enforce ERISA’s Duties of Prudence and Disclosure, 26 A.B.A. J. Lab. & Emp. L. 345, 354–57 (2011) (discussing the cases alleging employer fiduciary breaches due to unreasonably high fees and nondisclosure of fees such as revenue sharing); Jill E. Fisch, Rethinking the Regulation of Securities Intermediaries, 158 U. Pa. L. Rev. 1961, 2008–10 (2010) (same). One case that received press recently was the settlement of 401(k) fees involving Wal-Mart’s plan for $13.5 million. Barry B. Burr, Wal-Mart, Merrill Lynch Settle 401(k) Fee Suit, Pensions & Invs. (Dec. 12, 2011), Reportedly, Wal-Mart agreed to pay $3.5 million and Merrill Lynch agreed to pay $10 million of the settlement. Id.

  181. [181]. Reasonable Contract or Arrangement Under Section 408(b)(2)—Fee Disclosure, 77 Fed. Reg. 5632, 5655–57 (Feb. 3, 2012) (to be codified at 29 C.F.R. pt. 2550) (requiring service providers who are fiduciaries or registered investment advisers providing a broad array of services and expecting to receive at least $1000 in compensation to disclose that compensation to plan fiduciaries).

  182. [182]. Fiduciary Requirements for Disclosure in Participant-Directed Individual Account Plans, 75 Fed. Reg. 64,910, 64,937–39 (Oct. 20, 2010) (to be codified at 29 C.F.R. pt. 2550).

  183. [183]. See Definition of the Term “Fiduciary,” 75 Fed. Reg. 65,263, 65,264–65 (proposed Oct. 22, 2010) (to be codified at 29 C.F.R. pt. 2510).

  184. [184]. The summary version is that an investment adviser is not a fiduciary when giving advice regarding benefit plan assets or an Individual Retirement Account (“IRA”) unless: (1) the adviser advises on securities valuation or makes recommendations on the purchase or sale of securities; (2) on a regular basis; (3) according to a mutual agreement with the plan or a plan fiduciary; (4) provides that the advice will serve as the primary basis for decisions on investments; and (5) the advice is individualized to the plan’s needs. Id.

  185. [185]. John A. Turner & Dana M. Muir, The Market for Financial Advisers, in The Market for Retirement Financial Advice 13, 34 (Olivia S. Mitchell & Kent Smetters eds., (forthcoming 2013)) (on file with authors).

  186. [186]. See Definition of the Term “Fiduciary,” 75 Fed. Reg. at 65,277–78.

  187. [187]. US Labor Department’s EBSA to Re-propose Rule on Definition of a Fiduciary, U.S. Dep’t Labor (Sept. 19, 2011),

  188. [188]. Hazel Bradford, Retirement Savings Tax Incentives in Danger Post Election, Pensions & Invs. (Nov. 12, 2012), (“Ms. Borzi has also made her agency’s fiduciary rule a priority.”).

  189. [189]. Rev. Rul. 98-30, 1998-1 C.B. 1273.

  190. [190]. Rev. Rul. 2000-8, 2000-1 C.B. 617.

  191. [191]. Steven D. Cohen, supra note 71, at 300–01.

  192. [192]. Id.

  193. [193]. Pension Protection Act of 2006, Pub. L. No. 109-280, 120 Stat. 780.

  194. [194]. If an employer includes specific provisions in the automatic-enrollment plan, the plan will automatically meet ERISA and the IRC’s nondiscrimination requirements, which otherwise require significant analysis. See Cass R. Sunstein, Empirically Informed Regulation, 78 U. Chi. L. Rev. 1349, 1393–94 (2011).

  195. [195]. Four types of investment products qualify as QDIAs. First, a short-term, capital preservation product, which may be used for the first 120 days of an employee’s plan participation, is the only conservative product. See 29 C.F.R. § 2550.404c-5(e)(4) (2012). Second, three categories of long-term products meet the requirements to be QDIAs. See id. Third, two must be appropriate to the individual characteristics of the specific employee. See id. Specifically, “targeted-retirement-date” funds, more commonly known as target date funds (“TDFs”), may qualify. See id. Fourth, a QDIA may consist of a product that contains investments tailored to account for characteristics of the plan participants as a group. See id. Approximately 60% of the 401(k) plans with a default investment feature now designate TDFs as a QDIA. US Department of Labor: Spring Regulatory Agenda 2010, U.S. Dep’t Labor,
    regulations/factsheets/ebsa-fs-QDIA-TDF.htm (last visited Sept. 17, 2013). TDFs are not easily defined, at least not in all of their permutations. For a significant discussion of TDFs, see Jonathan Miller et al., Target Date Funds: Can One Just Glide into Retirement?, 10 J. Int’l Bus. & L. 349 (2011). One industry professional explained that investment strategies targeted to a retirement date “rebalance on an ongoing basis and adjust allocations as a[n employee] ages.” Stacy L. Schaus, Designing Successful Target-Date Strategies for Defined Contribution Plans: Putting Participants on the Optimal Glide Path 14 (2010).

  196. [196]. Fiduciary Relief for Investments in Qualified Default Investment Alternatives, 29 C.F.R. § 2550.404c-5; see also Davis, supra note 77, at 1031–32 (discussing QDIA regulation).

  197. [197]. See 29 C.F.R. § 2550.404c-5(b)(1).

  198. [198]. Davis, supra note 77, at 1031.

  199. [199]. 29 C.F.R. § 2550.404c-5(b)(2).

  200. [200]. See Muir & Schipani, supra note 174, at 336–37.

  201. [201]. See supra text accompanying note 184.

  202. [202]. See Turner & Muir, supra note 185, at 34; see also Definition of the Term “Fiduciary,” 75 Fed. Reg. 65,263, 65,264–65 (proposed Oct. 22, 2010) (to be codified at 29 C.F.R. pt. 2510) (proposing revisions expanding the regulation’s definition of “fiduciary” due to difficulties regulating plan service providers).

  203. [203]. 15 U.S.C. §§ 80b-1 to -21 (2012).

  204. [204]. See, e.g., Arthur B. Laby, Selling Advice and Creating Expectations: Why Brokers Should Be Fiduciaries, 87 Wash. L. Rev. 707, 723–24 (2012) (describing the exception for broker–dealers who meet specific standards).

  205. [205]. See Staff of the Div. of Inv. Mgmt., U.S. Sec. & Exch. Comm’n, Study on Enhancing Investment Adviser Examinations (2011), available at
    2011/914studyfinal.pdf (recommending an extension of fiduciary obligations); see also Jake Zamansky, SEC Struggles with Investor-Protection Rules, Forbes (Jan. 24, 2012, 1:20 PM), (indicating that the SEC’s regulatory effort has been delayed indefinitely).

  206. [206]. See supra text accompanying notes 183–88.

  207. [207]. See infra notes 257–65 and accompanying text.

  208. [208]. See 15 U.S.C. §§ 80b-1 to -21.

  209. [209]. Donald C. Langevoort, Private Litigation to Enforce Fiduciary Duties in Mutual Funds: Derivative Suits, Disinterested Directors and the Ideology of Investor Sovereignty, 83 Wash. U. L.Q. 1017, 1019 (2005).

  210. [210]. See 29 U.S.C. § 1002(21)(B) (2006).

  211. [211]. See M. Scott Donald, What Contribution Does Trust Law Make to the Regulatory Scheme Shaping Superannuation in Australia? 4–5 (2010) (unpublished summary of Ph.D. thesis, University of New South Wales), available at

  212. [212]. Mark Blair & Ian Ramsay, Collective Investment Schemes: The Role of the Trustee, Austl. Acct. Rev., no. 3, 1992 at 10, 17 (noting that a trustee is required in order to receive certain tax benefits); see also Jeremy Cooper, Super for Members: A New Paradigm for Australia’s Retirement Income System, Rotman Int’l J. Pension Mgmt., Fall 2010, at 8, 8.

  213. [213]. See Donald, supra note 211, at 8–10.

  214. [214]. Id. at 13–14.

  215. [215]. See supra text accompanying notes 154–56.

  216. [216]. See supra text accompanying notes 108–11.

  217. [217]. Cooper Report, Part I, supra note 129, at 70 tbl.B1.

  218. [218]. Id. (showing that retail and industry funds hold 45.9% of SG System assets).

  219. [219]. Donald, supra note 141, at 307–08.

  220. [220]. See Cooper Report, Part II, supra note 112, at 12–14; Austl. Gov’t, supra note 134, at 19–20.

  221. [221]. See Proposed Trustee and Prudential Standards, supra note 135, at 8.

  222. [222]. Cooper Report, Part I, supra note 129, at 25; see also Donald, supra note 141, at 301 (explaining that the appropriateness of the investment strategy is determined at the collective level of the MySuper fund, not at the level of the individual member).

  223. [223]. See Proposed MySuper Core Provisions, supra note 135,, at 15.

  224. [224]. See Proposed Trustee and Prudential Standards, supra note 135, at 16.

  225. [225]. Id.

  226. [226]. Cooper Report, Part I, supra note 129, at 7–8.

  227. [227]. Id.

  228. [228]. Id.

  229. [229]. Id.

  230. [230]. Cooper, supra note 212, at 12.

  231. [231]. See id.

  232. [232]. See supra text accompanying note 129.

  233. [233]. See infra text accompanying notes 254–55.

  234. [234]. Jack VanDerhei, The Impact of the Recent Financial Crisis on 401(k) Account Balances, EBRI Issue Brief, Feb. 2009, at 1, 3, available at (stating that the S&P 500 Index lost 37%).

  235. [235]. See Timothy E. Lynch, Deeply and Persistently Conflicted: Credit Rating Agencies in the Current Regulatory Environment, 59 Case W. Res. L. Rev. 227, 230 n.8 (2009) (stating that the Dow Jones Industrial Index lost 33.8%).

  236. [236]. Wilshire Index Calculator, Wilshire (search “Broad/Style” for “Wilshire 5000”; then click “1Yr”; then search “As of Date” for “January 1, 2009”; then click “Submit”) (stating that the Dow Jones Wilshire 5000 lost 37.23%).

  237. [237]. Frank W. Heiland & Zhe Li, Changes in Labor Force Participation of Older Americans and Their Pension Structures: A Policy Perspective 6–7 (Ctr. for Ret. Research at Bos. Coll., Working Paper No. 2012-18, 2012), available at

  238. [238]. See Retirement Research Centers, Emp. Benefit Research Inst.,
    research/retirement-research-centers/ (last visited Sept. 17, 2013).

  239. [239]. A replacement rate, sometimes called a replacement ratio, is “the ratio of an individual’s income in retirement to that they received at work.” Bryn Davies, Imagining the Ideal UK Pension System, in Imagining the Ideal Pension System: International Perspectives, supra note 39, at 45, 48. There is no consensus on what constitutes an appropriate minimum replacement rate with commentators generally viewing 70–80% rates as a good goal. Muir & Turner, supra note 39, at 31.

  240. [240]. VanDerhei, supra note 234, at 3.

  241. [241]. Id. at 1.

  242. [242]. Id.

  243. [243]. Id. Account balances of less than $10,000 grew by 40% during 2008. Id.

  244. [244]. Id.

  245. [245]. See, e.g., id. at 11 (“[T]he 2007 asset allocation information . . . shows that almost a quarter (22 percent) of the oldest 401(k) participants (ages 56–65 in 2007) had 90 percent or more of their 401(k) assets in equities. Another 10 percent had 80–90 percent in equities, and 11 percent had 70–80 percent in equities.”).

  246. [246]. Alicia H. Munnell et al., An Update on 401(k) Plans: Insights from the 2007 Survey of Consumer Finances 33 fig.12 (Ctr. for Ret. Research at Bos. Coll., Working Paper No. 2009-26, 2009), available at

  247. [247]. Id.

  248. [248]. Barbara A. Butrica et al., What the 2008 Stock Market Crash Means for Retirement Security 17–18 (The Urban Inst., Discussion Paper No. 09-03, 2009), available at

  249. [249]. Market Volatility and Your Investments, Asgard,
    investors/volatility/faqs.html (last visited Sept. 17, 2013). The S&P/ASX 200 includes about 80% of the capitalization on Australian equity markets. S&P/ASX 200, S&P Dow Jones Indices, (last visited Sept. 17, 2013); see also supra notes 234–36 and accompanying text (discussing the performance of the U.S. markets during that time period).

  250. [250]. Watson Wyatt Worldwide, 2009 Global Pension Assets Study 12 (2009), available at

  251. [251]. Bateman, supra note 116, at 82–84.

  252. [252]. Id.

  253. [253]. Id.

  254. [254]. DOL/SEC Hearing on Target Date Funds and Similar Investment Options, U.S. Dep’t of Labor, (last visited Sept. 17, 2013).

  255. [255]. Press Release, Sec. & Exch. Comm’n, SEC Proposes New Measures to Help Investors in Target Date Funds (June 16, 2010), available at

  256. [256]. Id.

  257. [257]. Majority Staff of S. Special Comm. on Aging, 111th Cong., Target Date Retirement Funds: Lack of Clarity Among Structures and Fees Raises Concerns (2009).

  258. [258]. Id. at 8.

  259. [259]. Id. at 14.

  260. [260]. Schaus, supra note 195, at 78–79.

  261. [261]. Pierluigi Balduzzi & Jonathan Reuter, Heterogeneity in Target-Date Funds and the Pension Protection Act of 2006, at 4–5 (Nat’l Bureau of Econ. Research, Working Paper No. 17886, 2012), available at

  262. [262]. Majority Staff of S. Special Comm. on Aging, supra note 257, at 15.

  263. [263]. Id.

  264. [264]. See Schaus, supra note 195, at 74.

  265. [265]. Josh Charlson & Laura Pavlenko Lutton, Morningstar Fund Research, Target-Date Series Research Paper: 2012 Industry Survey 31 (2012), available at http://corporate.

  266. [266]. Office of Investor Educ. & Advocacy, U.S. Dep’t of Labor, Investor Bulletin: Target Date Retirement Funds (2010), available at
    investorbulletin.pdf. For discussion of potential additional regulation of TDFs, see Miller et al., supra note 195, at 353–55.

  267. [267]. Target Date Disclosure, 75 Fed. Reg. 73,987, 73,988 (proposed Nov. 30, 2010) (to be codified at 29 C.F.R. pt. 2550). The DOL reopened the comment period on these proposed regulations. Target Date Disclosure, 77 Fed. Reg. 30,928, 30,928 (proposed May 24, 2012) (to be codified at 29 C.F.R. pt. 2550).

  268. [268]. Target Date Disclosure, 75 Fed. Reg. at 73,988–89. The DOL reopened the comment period on these proposed regulations. Target Date Disclosure, 77 Fed. Reg. at 30,928.

  269. [269]. Target Date Disclosure, 75 Fed. Reg. at 73,989.

  270. [270]. Id.

  271. [271]. See supra text accompanying note 100.

  272. [272]. Robert A. Prentice, Moral Equilibrium: Stock Brokers and the Limits of Disclosure, 2011 Wis. L. Rev. 1059, 1105.

  273. [273]. Id. at 1060–61.

  274. [274]. See id. at 1094–1105 (explaining that “if [someone] take[s] an ethical action that validates their self-image as a good person, they may well give themselves permission to play fast and loose with the rules for a while”).

  275. [275]. Munnell & Sunden, supra note 62, at 51; see also Stabile, supra note 57, at 314 (“[M]any people, particularly at the lower end of the income scale—those who benefited most from the promise afforded by the traditional defined benefit plan—will retire with inadequate 401(k) plan account balances to see them through their retirement years . . . .”); Karen C. Burke & Grayson M.P. McCouch, Privatizing Social Security: Administration and Implementation, 58 Wash. & Lee L. Rev. 1325, 1339 (2001).

  276. [276]. Alicia H. Munnell et al., Ctr. for Ret. Research at Bos. Coll., The National Retirement Risk Index: After the Crash 6 (2009), available at

  277. [277]. Ruth Helman et al., The 2011 Retirement Confidence Survey: Confidence Drops to Record Lows, Reflecting “the New Normal,EBRI Issue Brief, Mar. 2011, at 1, 1, available at http://www.

  278. [278]. Id. at 7.

  279. [279]. Id. at 31.

  280. [280]. Ghilarducci, supra note 13, at 262.

  281. [281]. See id. at 272.

  282. [282]. Id. at 264.

  283. [283]. Id.

  284. [284]. Id. at 264–65.

  285. [285]. See Fund Management, Thrift Savings Plan,
    fundsoverview/fundManagement.shtml (last visited Sept. 17, 2013).

  286. [286]. See id.

  287. [287]. Id.

  288. [288]. Ghilarducci, supra note 13, at 264–65.

  289. [289]. Id. at 265.

  290. [290]. Schwartz, supra note 14, at 74–79.

  291. [291]. Id. at 74.

  292. [292]. Id. at 76.

  293. [293]. Id. at 74–75, 78–79.

  294. [294]. Id. at 85.

  295. [295]. Id. at 79.

  296. [296]. Id. at 74–75.

  297. [297]. Id. at 74.

  298. [298]. Id.

  299. [299]. Id. at 83.

  300. [300]. Id.

  301. [301]. Id.

  302. [302]. Id. at 85.

  303. [303]. Id. at 85–86.

  304. [304]. Professor Schwartz does not discuss the transition from the current 401(k) plan system to his recommended system. However, it would seem that employers would have little incentive to continue 401(k) plans for existing accounts. Many of those assets would probably be transferred to the new system, resulting in rapid growth in the investment fund.

  305. [305]. Harkin, supra note 15, at 7.

  306. [306]. Id.

  307. [307]. Id.

  308. [308]. Id.

  309. [309]. Id. at 6.

  310. [310]. Id. at 7.

  311. [311]. See id.

  312. [312]. Id. at 6.

  313. [313]. Id.

  314. [314]. Id. at 7.

  315. [315]. Id. at 6.

  316. [316]. Id.

  317. [317]. Id.

  318. [318]. Id.

  319. [319]. Id. at 7.

  320. [320]. U.S. Gov’t Accountability Office, Pension Benefit Guaranty Corporation: Asset Management Needs Better Stewardship 1 (2011), available at
    items/d11271.pdf (“The PBGC’s investment objectives and stated asset allocation targets have changed frequently in the last 8 years, alternating between more conservative and more aggressive approaches to investing.”).

  321. [321]. See supra text accompanying note 106.

  322. [322]. Moore, supra note 45, at 8.

  323. [323]. Id. at 10, 14.

  324. [324]. Id. at 9.

  325. [325]. Id. at 12–13.

  326. [326]. Harkin, supra note 15, at 8.

  327. [327]. See Ghilarducci, supra note 13, at 139–178 (discussing the risks to Social Security and objecting particularly to the possibility of individual savings accounts within the Social Security system).

  328. [328]. Schwartz, supra note 14, at 81–82.

  329. [329]. Commentators have argued that the government is likely to guarantee the obligations of the PBGC even though it technically has no obligation to do so. See, e.g., Joshua Gad-Harf, Note, The Decline of Traditional Pensions, the Impact of the Pension Protection Act of 2006, and the Future of America’s Defined-Benefit Pension System, 83 Chi.-Kent L. Rev. 1409, 1431 (2008) (“The more liability the PBGC inherits, the greater the chances are that the government will have to bail out the PBGC . . . .”); Nicholas J. Brannick, Note, At the Crossroads of Three Codes: How Employers Are Using ERISA, the Tax Code, and Bankruptcy to Evade Their Pension Obligations, 65 Ohio St. L.J. 1577, 1588 n.52 (2004) (“And while the PBGC may not use general revenues to bail out failed pension plans, this does not mean that general revenues may not prove necessary to bail out the PBGC, despite the fact that the government is not technically liable for any of the PBGC’s obligations.”).

  330. [330]. See Elizabeth D. Tedrow, Social Security Privatization in Other Countries—What Lessons Can Be Learned for the United States?, 14 Elder L.J. 35, 51–53 (2006) (discussing Australia’s SG System as a counterpart to a privatized Social Security system).

  331. [331]. See supra text accompanying notes 172–82.

  332. [332]. A threshold of 100 employees would track the Bureau of Labor Statistics reporting. See Colleen E. Medill, Targeted Pension Reform, 27 J. Legis. 1, 5 n.11 (2001).

  333. [333]. See supra text accompanying note 83.

  334. [334]. See supra text accompanying note 84.

  335. [335]. See supra note 195.

  336. [336]. See supra text accompanying notes 266–70.

  337. [337]. See supra text accompanying notes 259–61.

  338. [338]. See supra text accompanying notes 266–70.

  339. [339]. Office of Investor Educ. & Advocacy, supra note 266.

  340. [340]. See supra text accompanying notes 268–70.

  341. [341]. See supra text accompanying notes 28–31.

  342. [342]. See Barbara A. Butrica & Mauricio Soto, Does Autoenrollment Affect Employer Contributions?, Older Ams. Econ. Sec.: Ret. Pol’y Program, Dec. 2009, at 1, 1, available at

  343. [343]. Sarah Holden et al., 401(k) Plan Asset Allocation, Account Balances, and Loan Activity in 2010, ICI Res. Persp., Dec. 2011, at 1, 63 n.36, available at; see also Brown, supra note 74 (finding that 42% of large employer plans utilized automatic enrollment in 2010).

  344. [344]. See Holden et al., supra note 343, at 63 n.36.

  345. [345]. See supra text accompanying notes 59–60.

  346. [346]. This will be particularly true if a system develops to easily enable employers to join with other employers to form multiple-employer 401(k) plans in order to increase administrative and reporting efficiencies.

  347. [347]. A secondary benefit of automatic enrollment features is that they open up the possibility that the plan will then also utilize automatic escalation provisions, which are shown to increase the amounts that employees save in 401(k) plans. See supra text accompanying notes 71–74.

  348. [348]. See supra text accompanying notes 28–31.

  349. [349]. Cohn & Taylor, supra note 1.

  350. [350]. Elizabeth Mendes, Lack of Retirement Funds Is Americans’ Biggest Financial Worry: Concern About Being Able to Maintain Standard of Living at New High, Gallup (June 15, 2011), http://

  351. [351]. Kathleen Koster, Trying to Avoid Bad News and Confusion, Many Participants Leave 401(k) Statements Unopened, Emp. Benefit News (Apr. 15, 2009),

  352. [352]. Id.


Arthur F. Thurnau Professor of Business Law, Stephen M. Ross School of Business at the University of Michigan. J.D., The University of Michigan Law School. 734.763.3091. Copyright © 2013. 

I appreciate the research support provided by Michigan Ross and thank Loretta Tracy for research assistance. I received helpful comments from Mark DeBofsky, Donald Fuerst, Lissa Paris, Paul Secunda, Norman Stein, John Turner, and the participants in the Employee Benefits in an Era of Retrenchment and the 2013 American Economic Association conferences.