Menu
Tax Notes logo

GAO Recommends Form 5500 Revision for Retirement Plan Loan Data

MAR. 28, 2019

GAO-19-179

DATED MAR. 28, 2019
DOCUMENT ATTRIBUTES
Citations: GAO-19-179

RETIREMENT SAVINGS

Additional Data and Analysis Could Provide Insight into Early Withdrawals

Report to the Special Committee on Aging, U.S. Senate

March 2019

Highlights of GAO-19-179

Why GAO Did This Study

Federal law encourages individuals to save for retirement through tax incentives for 401(k) plans and IRAs — the predominant forms of retirement savings in the United States. In 2017, U.S. plans and IRAs reportedly held investments worth nearly $17 trillion dollars. Federal law also allows individuals to withdraw assets from these accounts under certain circumstances. DOL and IRS oversee 401(k) plans, and collect annual plan data — including financial information — on the Form 5500. For both IRAs and 401(k) plans, GAO was asked to examine: (1) the incidence and amount of early withdrawals; (2) factors that might lead individuals to access retirement savings early; and (3) policies and strategies that might reduce the incidence and amounts of early withdrawals.

To answer these questions, GAO analyzed data from IRS, the Census Bureau, and DOL from 2013 (the most recent complete data available); and interviewed a diverse range of stakeholders identified in the literature, including representatives of companies sponsoring 401(k) plans, plan administrators, subject matter experts, industry representatives, and participant advocates.

What GAO Recommends

GAO recommends that, as part of revising the Form 5500, DOL and IRS require plan sponsors to report the incidence and amount of all 401(k) plan loans that are not repaid. DOL and IRS neither agreed nor disagreed with our recommendation.

What GAO Found

In 2013 individuals in their prime working years (ages 25 to 55) removed at least $69 billion (+/- $3.5 billion) of their retirement savings early, according to GAO's analysis of 2013 Internal Revenue Service (IRS) and Department of Labor (DOL) data. Withdrawals from individual retirement accounts (IRA) — $39.5 billion (+/- $2.1 billion) — accounted for much of the money removed early, were equivalent to 3 percent (+/- 0.15 percent) of the age group's total IRA assets, and exceeded their IRA contributions in 2013. Participants in employer-sponsored plans, like 401(k) plans, withdrew at least $29.2 billion (+/- $2.8 billion) early as hardship withdrawals, lump sum payments made at job separation (known as cashouts), and loan balances that borrowers did not repay. Hardship withdrawals in 2013 were equivalent to about 0.5 percent (+/- 0.06 percent) of the age group's total plan assets and about 8 percent (+/- 0.9 percent) of their contributions. However, the incidence and amount of certain unrepaid plan loans cannot be determined because the Form 5500 — the federal government's primary source of information on employee benefit plans — does not capture these data.

Stakeholders GAO interviewed identified flexibilities in plan rules and individuals' pressing financial needs, such as out-of-pocket medical costs, as factors affecting early withdrawals of retirement savings. Stakeholders said that certain plan rules, such as setting high minimum loan thresholds, may cause individuals to take out more of their savings than they need. Stakeholders also identified several elements of the job separation process affecting early withdrawals, such as difficulties transferring account balances to a new plan and plans requiring the immediate repayment of outstanding loans, as relevant factors.

Stakeholders GAO interviewed suggested strategies they believed could balance early access to accounts with the need to build long-term retirement savings. For example, plan sponsors said allowing individuals to continue to repay plan loans after job separation, restricting participant access to plan sponsor contributions, allowing partial distributions at job separation, and building emergency savings features into plan designs, could help preserve retirement savings (see figure). However, they noted, each strategy involves tradeoffs, and the strategies' broader implications require further study.

Example of an Emergency Savings Option with a 401(k) Plan That Could Better Preserve Retirement Savings, According to Stakeholders


Contents

Letter

Background

At Least $69 Billion Dollars in 2013 Left Retirement Accounts Early, Mostly from Individual Retirement Accounts

Plan Rule Flexibilities and Use of Retirement Assets for Pressing Financial Needs Said to Result in Early Withdrawals

Stakeholders Suggested Strategies to Balance Access to Early Withdrawals with the Need to Build Long-term Retirement Savings

Conclusions

Recommendation for Executive Action

Agency Comments and Our Evaluation

Appendix I Objectives, Scope, and Methodology

Appendix II Selected Provisions Related to Early Withdrawals from 401(k) Plans and Individual Retirement Accounts (IRAs)

Appendix III Estimated Incidence of Certain Early Withdrawals of Retirement Savings

Appendix IV Comments from the Department of Labor

Appendix V Comments from the Internal Revenue Service

Appendix VI GAO Contact and Staff Acknowledgment

Related GAO Reports

Tables

Table 1: Options for Separating Employees to Maintain the Tax-Preferred Status of Their 401(k) Account Balance

Table 2: Selected Provisions Related to 401(k) Plans and the Types of Early Withdrawals Affected

Table 3: Selected Provisions Related to Traditional Individual Retirement Accounts (IRA)

Table 4: Selected Provisions Related to Roth Individual Retirement Accounts (IRA)

Table 5: Estimates of the Incidence of Early Withdrawals from Individual Retirement Accounts and 401(k) Plans, by Individual Characteristics

Figures

Figure 1: Some Plan Loan Rules May Increase the Risk of Loan Defaults

Figure 2: Selected Plan Sponsor Policies May Reduce 401(k) Loan Defaults on Participants

Figure 3: Emergency Savings Option within a 401(k) Plan Could Better Preserve Retirement Savings, According to Stakeholders

Abbreviations

BLS

Bureau of Labor Statistics

DB

defined benefit

DC

defined contribution

DOL

Department of Labor

EBSA

Employee Benefits Security Administration

EGTRRA

Economic Growth and Tax Relief Reconciliation Act of 2001

ERISA

Employee Retirement Income Security Act of 1974

ICI

Investment Company Institute

IRA

individual retirement account

IRC

Internal Revenue Code

IRS

Internal Revenue Service

SSA

Social Security Administration

SIPP

Survey of Income and Program Participation

TSP

Thrift Savings Plan

This is a work of the U.S. government and is not subject to copyright protection in the United States. The published product may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately.


March 28, 2019

The Honorable Susan M. Collins
Chairman
The Honorable Robert P. Casey, Jr.
Ranking Member
Special Committee on Aging
United States Senate

Federal law encourages U.S. workers to save for retirement by authorizing tax incentives for the predominant forms of retirement savings in the United States — employer-sponsored defined contribution (DC) plans like 401(k) plans and individual retirement accounts (IRA). In 2017, DC plans and IRAs reportedly held investments worth nearly $17 trillion dollars.1 Federal law also allows workers to access assets in employer-sponsored 401(k) plans before retirement under certain circumstances, such as financial hardship. Early access to retirement savings has been shown to benefit participants by encouraging plan participation, increasing participant contributions, and providing participants with a means of addressing their financial needs. Additionally, IRA owners can take a distribution from their IRA at any time for any reason. While such withdrawals occur for various reasons and can help workers facing financial difficulties, they can also affect a worker's long-term retirement security by reducing account assets and subjecting withdrawn amounts to additional taxation.2

You asked us to examine various aspects of early withdrawals of retirement savings from both IRAs and 401(k) plans. This report examines: (1) the incidence and amount of retirement savings being withdrawn early; (2) what is known about the factors that might lead individuals to access their retirement savings early; and (3) what strategies or policies, if any, might reduce the incidence and amount of early withdrawals of retirement savings.

To examine the incidence and amount of early withdrawals from IRAs and 401(k) plans, we analyzed the most recent nationally representative data available from three relevant federal sources focusing on individuals ages 25 to 55, when possible.3 First, to examine recent incidence and amounts of IRA withdrawals and tax consequences of withdrawal, we analyzed published statistics from the Internal Revenue Service (IRS) for 2013. Second, to examine the incidence and amount of early withdrawals from 401(k) plans, we analyzed the most recently available data on individuals collected in 2013 for the 2014 panel of the U.S. Census Bureau's Survey of Income and Program Participation (SIPP), along with its Social Security Administration (SSA) supplement.4 Third, to examine the incidence and amount of participant loans not paid back to a plan (i.e., plan loans), we analyzed annual plan data from 2013 reported to the Department of Labor (DOL) on the Form 5500.5 We determined that the data from these three sources were sufficiently reliable for the purposes of our report. To learn what is known about factors that lead to early withdrawals, and possible strategies or policies that could reduce early withdrawals, we conducted site visits in four metropolitan areas — Boston, Chicago, San Francisco, and Seattle — and interviewed a diverse sample of stakeholders identified in the literature, including human resource professionals at companies sponsoring 401(k) plans, administrators of large and small plans, subject matter experts; industry representatives; and participant advocates. For more information on the methodology used in developing this report, see appendix I.

We conducted this performance audit from October 2017 to March 2019 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives.

Background

Over the last 3 decades employers have shifted away from sponsoring defined benefit (DB) plans and toward DC plans. This shift also transfers certain types of risk — such as investment risk — from employers to employee participants. DB plans generally offer a fixed level of monthly annuitized retirement income based upon a formula specified in the plan, which usually takes into account factors such as a participant's salary, years of service, and age at retirement, regardless of how the plan's investments perform. In contrast, benefit levels in DC plans — such as 401(k) plans — depend on the contributions made to the plan and the performance of the investments in individual accounts, which may fluctuate in value.6 As we have previously reported, some experts have suggested that the portability of DC plans make them better-suited for a mobile workforce, and that such portability may lead to early withdrawals of retirement savings.7 DOL reported there were 656,241 DC and 46,300 DB plans in the United States in 2016.8

Tax incentives are in place to encourage employers to sponsor retirement plans and employees to participate in plans.9 Under the Employee Retirement Income Security Act of 1974 (ERISA), employers may sponsor DC retirement plans, including 401(k) plans — the predominant type of DC plan, in which benefits are based on contributions to and the performance of the investments in participants' individual accounts. To save in 401(k) plans, participants contribute a portion of their income into an investment account, and in traditional 401(k) plans taxes are deferred on these contributions and associated earnings, which can be withdrawn without penalty after age 59½ (if permitted by plan terms).10 As plan sponsors, employers may decide the amount of employer contributions (if any) and how long participants must work before having a non-forfeitable (i.e., vested) interest in their plan benefit, within limits established by federal law. Plan sponsors often contract with service providers to administer their plans and provide services such as record keeping (e.g., tracking and reporting individual account contributions); investment management (i.e., selecting and managing the securities included in a mutual fund); and custodial or trustee services for plan assets (e.g., holding the plan assets in a bank).11

Individuals also receive tax incentives to save for retirement outside of an employer-sponsored plan. For example, traditional IRAs provide certain individuals with a way to save pre-tax money for retirement, with withdrawals made in retirement taxed as income. In addition, Roth IRAs allow certain individuals to save after-tax money for retirement with withdrawals in retirement generally tax-free.12 IRAs were established under ERISA, in part, to (1) provide a way for individuals not covered by a pension plan to save for retirement; and (2) give retiring workers or individuals changing jobs a way to preserve assets from 401(k) plans by transferring their plan balances into IRAs. The Investment Company Institute (ICI) reported that 34.8 percent of households in the United States owned an IRA in 2017, a percentage that has generally remained stable since 2000.13 In 2017, IRA assets accounted for almost 33 percent (estimated at $9.2 trillion) of total U.S. retirement assets, followed by DC plans, which accounted for 27 percent ($7.7 trillion).14 Further, according to ICI, over 94 percent of funds flowing into traditional IRAs from 2000 to 2015 came from rollovers — primarily from 401(k) plans.15

Oversight of IRAs and 401(k) Plans

IRS, within the Department of the Treasury, is responsible for enforcing IRA tax laws, while IRS and DOL share responsibility for overseeing prohibited transactions relating to IRAs. IRS also works with DOL's Employee Benefits Security Administration (EBSA) to enforce laws governing 401(k) plans. IRS is primarily responsible for interpreting and enforcing provisions of the Internal Revenue Code (IRC) that apply to tax-preferred retirement savings. EBSA enforces ERISA's reporting and disclosure and fiduciary responsibility provisions, which, among other things, include requirements related to the type and extent of information that a plan sponsor must provide to plan participants.

Employers sponsoring employee benefit plans subject to ERISA, such as a 401(k) plans, generally must file detailed information about their plan each year. The Form 5500 serves as the primary source of information collected by the federal government regarding the operation, funding, expenses, and investments of employee benefit plans. The Form 5500 includes information about the financial condition and operation of their plans, among other things. EBSA uses the Form 5500 to monitor and enforce plan administrators and other fiduciaries, and service providers' responsibilities under Title I of ERISA. IRS uses the form to enforce standards that relate to, among other things, how employees become eligible to participate in benefit plans, and how they become eligible to earn rights to benefits.16

Permitted Early Withdrawals of Retirement Savings

In certain instances, sponsors of 401(k) plans may allow participants to access their tax-preferred retirement savings prior to retirement.17 Plan sponsors have flexibility under federal law and regulations to choose whether to allow plan participants access to their retirement savings prior to retirement and what forms of access to allow. Typically, plans allow participants to access their savings in one or more of the following forms:

  • Loans: Plans may allow participants to take loans and limit the number of loans allowed. If the plan provides for loans, the maximum amount that the plan can permit as a loan generally cannot exceed the lesser of (1) the greater of 50 percent of the vested account balance, or $10,000 or (2) $50,000 less the excess of the highest outstanding balance of loans during the 1-year period ending on the day before the day on which a new loan is made over the outstanding balance of loans on the day the new loan is made. Plan loans are generally not treated as early withdrawals unless they are not repaid within the terms specified under the plan.

  • Hardship withdrawals: Plans may allow participants facing a hardship to take a withdrawal on account of an immediate and heavy financial need, and if the withdrawal is necessary to satisfy the financial need.18 Though plan sponsors can decide whether to offer hardship withdrawals and approve applications for hardship withdrawals, IRS regulations provide “safe harbor” criteria regarding circumstances when a withdrawal is deemed to be on account of an immediate heavy financial need.19 IRS regulations allow certain expenses to qualify under the safe harbor including: (1) certain medical expenses; (2) costs directly relating to the purchase of a principal residence; (3) tuition and related educational fees and expenses for the participant, and their spouse, children, dependents or beneficiary; (4) payments necessary to prevent eviction from, or foreclosure on, a principal residence; (5) certain burial or funeral expenses; and (6) certain expenses for the repair of damage to the employee's principal residence.20 Plans that provide for hardship withdrawals generally specify what information participants must provide to the plan sponsor to demonstrate a hardship meets the definition of an immediate and heavy financial need.

Early withdrawals of retirement savings may have short-term and long-term impacts on participants' ability to accumulate retirement savings. In the short term, IRA owners and participants in 401(k) plans who received a withdrawal before reaching age 59½ generally pay an additional 10 percent tax for early distributions in addition to income taxes on the taxable portion of the distribution amount. The IRC exempts certain distributions from the additional tax, but the exceptions vary among 401(k) plans and IRAs.21 Early withdrawals of any type can result in the permanent removal of assets from retirement accounts thereby reducing the amounts participants can accumulate before retirement, including the loss of compounded interest or other earnings on the amounts over the participant's career.

Disposition of Account Balances at Job Separation

According to DOL's Bureau of Labor Statistics (BLS), U.S. workers are likely to have multiple jobs in their careers as average employee tenure has decreased. In 2017, BLS reported that from 1978 to 2014, workers held an average of 12 jobs between the ages of 18 and 50.22 BLS also reported in 2016 that the median job tenure for a worker was just over 4 years.23

Employees who separate from a job bear responsibility for deciding what to do with their accumulated assets in their former employer's plan. Recent research estimated that 10 million people with a retirement plan change jobs each year, many of whom faced a decision on how to treat their account balance at job separation.24 Plan administrators must provide a tax notice detailing participants' options for handling the balance of their accounts.25 When plan participants separate from their employers, they generally have one of three options:

1. They may leave the balance in the plan,

2. They may ask their employer to roll the money directly into a new qualified employer plan or IRA (known as a direct rollover), or

3. They may request a distribution. Once the participant receives the distribution he or she can (1) within 60 days, roll the distribution into a new qualified employer plan or IRA (in which case the money would remain tax-preferred); or (2) keep the distributed amount, and pay any income taxes or additional taxes associated with the distribution (known as a cashout).26

Sponsors of 401(k) plans may cash out or transfer separating participant accounts if an account balance falls below a certain threshold. The Economic Growth and Tax Relief Reconciliation Act of 2001 (EGTRRA) amended the IRC to provide certain protections for separating participants with account balances between $1,000 and $5,000 by requiring, in the absence of participant direction, plan sponsors to either keep the account in the plan or to transfer the account balance to an IRA to preserve its tax-preferred status.27 Plan sponsors may not distribute accounts with balances of more than $5,000 without participant direction, but have discretion to distribute account balances of $1,000 or less.28

Additional Tax Consequences for Early Withdrawals

The IRC imposes an additional 10 percent tax (in addition to ordinary income tax) on certain early withdrawals from qualified retirement plans, which includes IRAs and 401(k) plans in an effort to discourage the use of plan funds for purposes other than retirement and ensure the favorable tax treatment for plan funds is used to provide retirement income.29 Employers are required to withhold 20 percent of the amount cashed out to cover anticipated income taxes unless the participant pursues a direct rollover into another qualified plan or IRA.30

Employee Financial Literacy and Financial Wellness

Research has found that many employees are concerned about their level of savings and ability to manage their retirement accounts, and some employers provide educational services to improve employees' financial wellness and financial literacy and encourage them to save for retirement.31 A 2017 survey on employee financial wellness in the workplace found more than one-half of workers experienced financial stress and that insufficient emergency savings was a top concern for employees.32 Research has also found that limited financial literacy is widespread among Americans over age 50, and those who lack financial knowledge are less likely to successfully plan for retirement.33 In 2018, the Federal Reserve reported that three-fifths of non-retirees with participant-directed retirement accounts had little to no comfort managing their own investments.34 As we have previously reported, some employers have developed comprehensive programs aimed at overall improvement in employees' financial health. These programs, often called financial wellness programs, may help employees with budgeting, emergency savings, and credit management, in addition to the traditional information and assistance provided for retirement and health benefits.35

At Least $69 Billion Dollars in 2013 Left Retirement Accounts Early, Mostly from Individual Retirement Accounts

In 2013, individuals ages 25 to 55 withdrew at least $68.7 billion early from their retirement accounts.36 Of this amount, IRA owners in this age group withdrew the largest share (about 57 percent) and 401(k) plan participants in this age group withdrew the rest (about 43 percent) However, a total amount withdrawn from 401(k) plans cannot be determined due to data limitations.

Nearly $40 Billion Withdrawn Early from IRAs in 2013

IRA withdrawals were the largest source of early withdrawals of retirement savings, accounting for an estimated $39.5 billion of the total $68.7 billion in early withdrawals made by individuals ages 25 to 55 in 2013.37 According to IRS estimates, 12 percent of IRA owners in this age group withdrew money early that year from their IRAs in 2013. The amount they withdrew early comprised a small percentage of their total IRA assets. Specifically, in 2013, the amount of early withdrawals was equivalent to 3 percent of the cohort's total IRA assets and, according to IRS estimates, the total amount withdrawn by this cohort exceeded their total contributions to IRAs in that year.38

At Least $29 Billion Withdrawn Early from 401(k) Plans in 2013

At least $29.2 billion left 401(k) plans in 2013 in the form of hardship withdrawals, cashouts at job separation, and unrepaid plan loans, according to our analysis of 2013 SIPP data and data from DOL's Form 5500.39 Specifically, we found that:

  • Hardship withdrawals were the largest source of early withdrawals from 401(k) plans with an estimated 4 percent (+/- 0.25) of plan participants ages 25 to 55 withdrawing an aggregate $18.5 billion in 2013.40 The amount of hardship withdrawals was equivalent to 0.5 percent (+/- 0.06) of the cohort's total plan assets and 8 percent (+/-0.9) of the cohort's plan contributions made in 2013.

  • Cashouts of account balances of $1,000 or more at job separation were the second largest source of early withdrawals from 401(k) plans. In 2013, an estimated 1.1 percent (+/- 0.11) of plan participants ages 25 to 55 withdrew an aggregate $9.8 billion from their plans that they did not roll into another qualified plan or IRA.41 Additionally, 86 percent (+/- 2.9) of these participants taking a cashout of $1,000 or more did not roll over the amount in 2013. The amounts cashed out and not rolled over were equivalent to 0.3 percent (+/- 0.05) of the cohort's total plan assets and 4 percent (+/- 0.75) of the cohort's total contributions made in 2013.

  • Loan defaults accounted for at least $800 million withdrawn from 401(k) plans in 2013; however, the amount of distributions of unpaid plan loans is likely larger as DOL data cannot be used to quantify plan loan offsets that are deducted from participants' account balances after they leave a plan. As a result, the amount of loan offsets among terminating participants ages 25 to 55 cannot be determined with certainty.42 Specifically, DOL's Form 5500 instructions require plan sponsors to report unpaid loan balances in two separate places on the Form 5500, depending on whether the loan holder is an active or a terminated participant.43 For active participants, plan sponsors report loan defaults as a single line item on the Form 5500 (i.e., the $800 million in 2013 listed above). For terminated participants, plan sponsors report unrepaid plan loan balances as benefits paid directly to participants — a category that also includes rollovers to employer plans and IRAs.44 According to a DOL official, as a result of this commingling of benefits on this line item, isolating the amount of loan offsets for terminated participants using the Form 5500 data is not possible.45 Without better data of the amount of unrepaid plan loans, the amount of loan offsets and the characteristics of plan participants who did not repay their plan loans at job separation cannot be determined.

Additional Tax Consequences of Early Withdrawals Also Contributed to Reductions in Overall Savings

IRA owners and plan participants taking early withdrawals paid $6.2 billion as a result of the additional 10 percent tax for early distributions in 2013, according to IRS estimates.46 Although the taxes are generally treated separately from the amounts withdrawn, IRA owners and plan participants are expected to pay any applicable taxes resulting from the additional 10 percent tax when filing their income taxes for the tax year in which the withdrawal occurred.47

Certain Characteristics Were Associated With Higher Incidence of Early Withdrawals

Individuals with certain demographic and economic characteristics that we analyzed had higher incidence of early withdrawals of retirement savings, according to our analysis of SIPP data. The characteristics described below reflect statistically significant differences between comparison groups (a full listing of all demographic groups can be found in appendix III).

  • Age. The incidence of IRA withdrawals was higher among individuals ages 45 to 54 (8 percent) than individuals ages 25 to 34 and 35 to 44.

  • Education. Individuals with a high school education or less had higher incidence of cashouts (97 percent) and hardship withdrawals (7 percent) than individuals with some college or some graduate school education.

  • Family size. Individuals in families of seven or more (8 percent) or in families of five to six (7 percent) had higher incidence of hardship withdrawals than individuals in smaller family groups we analyzed. Individuals living alone had higher incidence of IRA withdrawals than individuals living in the larger family groups.

  • Marital status. Widowed, divorced, or separated individuals had higher incidence of IRA withdrawals (11 percent) and hardship withdrawals (7 percent) than married or never married individuals.

  • Race. The incidence of hardship withdrawals among African American (10 percent) and Hispanic individuals (6 percent) was higher than among individuals who were White, Asian, or Other.

  • Residence. The incidence of IRA withdrawals and hardship withdrawals was higher among individuals living in nonmetropolitan areas (7 percent and 6 percent, respectively) than among individuals living in metropolitan areas.

Similarly, individuals with certain economic characteristics that we analyzed had higher incidence of early withdrawals of retirement savings, according to our analysis of SIPP data. The characteristics described below reflect statistically significant differences between comparison groups (a full listing of all demographic groups can be found in appendix III).

  • Employer size. Individuals working for employers with fewer than 25 employees had higher incidence of IRA withdrawals (9 percent) than individuals working for employers with higher number of employees.

  • Employment. Individuals working fewer than 35 hours per week had higher incidence of IRA withdrawals (7 percent) than employees working 35 hours or more.

  • Household debt. Individuals with household debt of $5,000 up to $20,000 had higher incidence of IRA withdrawals (14 percent) than individuals with other debt amounts.

  • Household income. Individuals with household income of less than $25,000 or $25,000 up to $50,000 had higher incidence of IRA withdrawals (12 percent and 9 percent, respectively) and hardship withdrawals (9 percent and 7 percent, respectively) than individuals with higher income amounts.

  • Personal cash reserves. Individuals with personal cash reserves of less than $1,000 had higher incidence of IRA withdrawals (10 percent) and hardship withdrawals (6 percent) than individuals with larger reserves.

  • Retirement assets. Individuals with combined IRA and 401(k) plan assets valued at less than $5,000 had higher incidence of hardship withdrawals (7 percent) than individuals with higher valued assets.

  • Tenure in retirement plan. Individuals with fewer than 3 years in their retirement plan had higher incidence of hardship withdrawals (6 percent) than individuals with longer tenures.

Plan Rule Flexibilities and Use of Retirement Assets for Pressing Financial Needs Said to Result in Early Withdrawals

Stakeholders Said Plan Rules Governing Early Withdrawals May Lead to Reduced Savings for Some Participants

Stakeholders we interviewed said that plan rules related to the disposition of account balances at job separation can lead participants to remove more than they need, up to and including their entire balance. We previously reported U.S. workers are likely to change jobs multiple times in a career.48 Plan sponsors may cash out balances of $1,000 or less at job separation, although they are not required to do so. As a result, plan participants with such balances, including younger employees and others with short job tenures, risk having their account balances distributed in full each time they change jobs.49 As shown in table 1, a separating employee must take multiple steps to ensure that an account balance remains tax-preferred.50

Table 1: Options for Separating Employees to Maintain the Tax-Preferred Status of Their 401(k) Account Balance

Option

Subject to withholding for tax purposes?

Action required of employee to preserve retirement savings?

Subject to income taxes or an additional 10 percent tax?

Extent to which tax-preferred status of retirement savings is preserved?

Leave balance in plan. (No distribution occurs.)

No

No

No

100 percent. Full balance remains in plan.

Direct rollover to another qualified plan or individual retirement account (IRA).

No

Asks plan sponsor to facilitate a direct rollover.a

No

100 percent. Full balance can be transferred to qualified plan or IRA.

Take a distribution from the plan with the intent of rolling balance into another qualified plan or IRA.

Yes, employer withholds 20 percent of distribution.

Employee requests a distribution from the plan.a

To preserve tax-preferred status of the account, the employee must roll over the amount of the distribution within 60 days of receipt of the distribution.

To complete a rollover of the full account balance, the employee must make up the 20 percent mandatory withholding with funding from another source.

Any funds not rolled over are treated as a distribution and subject to income taxes and, possibly, an additional 10 percent tax for early distributions.

100 percent. If the 20 percent withheld is replaced.

80 percent. If the 20 percent withheld is not replaced, it is reported as taxable income and subject to income taxes and, possibly, an additional 10 percent tax for early distributions.

0 percent. If the distribution is not rolled over within 60 days after receipt, the entire amount is reported as taxable income and subject to income taxes and, possibly, an additional 10 percent tax for early distributions.

Source: GAO review of federal law and regulations. | GAO-19-179

a Participant direction is not always required for a plan administrator to initiate a direct rollover. For example, a plan administrator may establish a default procedure where a rollover will occur if a plan participant does not make an affirmative election.

Participants who take a distribution from a plan with the intent of rolling it into another qualified plan or IRA must acquire additional funds to complete the rollover and avoid adverse tax consequences. Plan sponsors are required to withhold 20 percent of the account balance to pay anticipated taxes on the distribution. As a result, the sponsor then sends 80 percent of the account balance to the participant, who must acquire outside funds to compensate for the 20 percent withheld or forgo the preferential tax treatment of that portion of their account balance. For example, a participant seeking to roll over a retirement account with a $10,000 balance would receive an $8,000 distribution after tax withholding, requiring them to locate an additional $2,000 to complete the rollover within the 60-day period to avoid a taxable distribution of the withheld amount.51 If participants can replace the 20 percent withheld and complete the rollover within the 60-day period, they do not owe taxes on the distribution.

Stakeholders said that the complexity of rolling a 401(k) account balance from one employer to another may encourage participants to take the relatively simpler route of rolling their balance into an IRA or cashing out altogether. They noted that separating participants had many questions when evaluating their options and had difficulty understanding the notice provided. For example, participants may not fully understand how the decisions made at job separation can have a significant impact on their current tax situation and eventual retirement security. One plan sponsor, describing concerns about giving investment advice, said she watched participants make what she judged to be poor choices with their account balances and felt helpless to intervene. Stakeholders also noted that the lack of a standardized rollover process sometimes bred mistrust among employers and complicated separating participants' ability to successfully facilitate a rollover between plans. For example, one stakeholder told us that some plans were hesitant to accept funds from other employer plans fearing that the funds might come from plans that have failed to comply with plan qualification requirements and could create problems for the receiving plan later on.52 Another stakeholder suggested that the requirement for plan sponsors to provide a notice to separating participants likely caused more participants to take the distribution.

Stakeholders described loans as a useful source of funds in times of need and a way to avoid more expensive options, such as high-interest credit cards. They also noted that certain plan loan policies could lead to early withdrawals of retirement savings.53 (See fig. 1.)

  • Loan repayment at job separation: Stakeholders said loan repayment policies can increase the incidence of defaults on outstanding loans.54 When participants do not repay their loan after separating from a job, the outstanding balance is treated as a distribution, which may subject it to income tax liability and, possibly, an additional 10 percent tax for early distributions. According to stakeholders, the process of changing jobs can inadvertently lead to a distribution of a participant's outstanding loan balance, when the participant could have otherwise repaid the loan.55

  • Extended loan repayment periods: Some plan sponsors allow participants to take loans to purchase a home. Stakeholders told us that the amounts of these home loans tended to be larger than general purpose loans and had longer repayment periods that these extended from 15 to 30 years. A stakeholder further noted that these loans could make it more likely that participants would have larger balances to repay if they lost or changed jobs.

  • Multiple loans: While some plan sponsors noted that their plans limited the number of loans participants can take from their retirement plan, others do not. Some plan sponsors limited participants to between one and three simultaneous loans, and one plan administrator indicated that 92 percent of their plan-sponsor clients allowed no more than two simultaneous loans.56 Other plan sponsors placed no limit on the number of participant loans or limited loans to one or two per calendar year, in which case a participant could take out a new loan at the start of a calendar year regardless of whether or not outstanding loans had been repaid. Stakeholders described some participants as “serial” borrowers, who take out multiple loans and have less disposable income as a result of ongoing loan payments. One plan administrator stated that repeat borrowing from 401(k) plans was common, and some participants took out new loans to pay off old loans.57

  • Other loan restrictions: Allowing no loans or one total outstanding loan can cause participants facing economic shocks to take a hardship withdrawal, resulting in the permanent removal of their savings and subjecting them to income tax liability and, possibly, an additional 10 percent tax for early distributions and a suspension on contributions.58

  • Minimum loan amounts: Minimum loan amounts may result in participants borrowing more than they need to cover planned expenses. For example, a participant may have a $500 expense for which they seek a loan, but may have to borrow $1,000 due to plan loan minimums.

Figure 1: Some Plan Loan Rules May Increase the Risk of Loan Defaults

Stakeholders Said Participants Take Early Withdrawals for Pressing Financial Needs

Stakeholders said that plan participants take plan loans and hardship withdrawals for pressing financial needs. Many plan sponsors we interviewed said they used the IRS safe harbor exclusively as criteria when reviewing a participant's application for a hardship withdrawal.59 Stakeholders said the top two reasons participants took hardship withdrawals were to prevent imminent eviction or foreclosure and to cover out-of-pocket medical costs not covered by health insurance.60 Participants generally took loans to reduce debt, for emergencies, or to purchase a primary residence.61 Stakeholders also said that participants who experienced economic shocks stemming from job loss made early withdrawals.62 They said retirement plans often served as a form of insurance for those between jobs or facing a sudden economic shock and participants accessed their retirement accounts because, for many, they were the only source of savings.63 They cited personal debt, health care costs, and education as significant factors that affected employees across all income levels.

Stakeholders said some participants also used their retirement savings to pay for anticipated expenses. Two plan administrators said education expenses were one of the reasons participants took hardship withdrawals. They said that participants accessed their retirement savings to address the cost of higher education, including paying off their own student loan debt or financing the college costs for family members. For example, plan administrators told us that some participants saved with the expectation of taking a hardship withdrawal to pay for college tuition. Other participants utilized hardship withdrawals to purchase a primary residence.

Reasons for IRA Withdrawals Are Not Reported to IRS

IRA owners generally may take withdrawals at any time and IRS does not analyze the limited information it receives on the reasons for IRA withdrawals. IRA owners can withdraw any amount up to their entire account balance at any time.64 In addition, IRAs have certain exceptions from the additional 10 percent tax for early distributions. For example, IRA withdrawals taken for qualified higher education expenses, certain health insurance premiums, and qualified “first-time” home purchases (up to $10,000) are excepted from the additional 10 percent tax. IRA owners who make an IRA distribution receive a Form1099-R or similar statement from their provider. On the Form 1099-R, IRA providers generally identify whether the withdrawal, among other things, can be categorized as a normal distribution, an early distribution, or a direct distribution to a qualified plan or IRA. For an early distribution, the IRA provider may identify whether a known exception to the additional 10 percent tax applies.65 For their part, IRA owners are required to report early withdrawals on their income tax returns, as well as the reason for any exception from the additional 10 percent tax for a limited number of items.66 In written responses to questions, an IRS official indicated that IRS collected data on the exemption reason codes, but did not use them.67

Stakeholders Suggested Strategies to Balance Access to Early Withdrawals with the Need to Build Long-term Retirement Savings

Some Plan Sponsors Have Implemented Policies to Preserve the Benefits of Early Withdrawals While Reducing Their Long-term Effects

Preserving 401(k) Account Balances at Job Separation

Some plan sponsors we interviewed had policies in place that may reduce the long-term impact of early withdrawals of retirement savings taken at job separation.68 Policies suggested by plan sponsors included:

  • Providing a periodic installment distribution option: Although some plan sponsors may require participants wanting a distribution to take their full account balance at job separation, other plan sponsors provided participants with an option of receiving their account balance in periodic installments.69 For example, one plan sponsor gives separating participants an option to receive periodic installment distributions at intervals determined by the participants. This plan sponsor said separating participants could select distributions on a monthly, quarterly, semi-annual or annual basis. These participants could also elect to stop distributions at any time, preserving the remaining balance in the employer's plan. The plan sponsor said the plan adopted this option to help separating participants address any current financial needs, while preserving some of the account balance for retirement. Another plan sponsor adopted a similar policy to address the cyclical nature of the employer's business, which can result in participants being terminated and rehired within one year.

  • Offering partial distributions: One plan sponsor provided separated participants with the option of receiving a one-time, partial distribution.70 If a participant opted for partial distribution, the plan sponsor issued the distribution for the requested sum and preserved the remainder of the account balance in the plan. The plan sponsor adopted the partial distribution policy to provide separating participants with choices for preserving account balances, while simultaneously providing access to address any immediate financial needs.

  • Providing plan loan repayment options for separated participants: Some plan sponsors allowed former participants to continue making loan repayments after job separation.71 Loan repayments after job separation reduce the loan default risk and associated tax implications for participants.72 Some plan sponsors said that separating participants who have the option to continue repaying an outstanding loan balance generally have three options: (1) to continue repaying the outstanding loan, (2) to repay the entire balance of the loan at separation within a set repayment period, or (3) not to repay the loan. Those participants who continue repaying their loans after separation generally have the option to set up automatic debit payments to facilitate the repayment. Those separated participants who do not set up loan repayment terms within established timeframes, or do not make a payment after the loan repayment plan has been established, default on their loan and face the associated tax consequences, including, possibly, an additional 10 percent tax for early distributions.

Setting Limits on Plan Loans

Some plan sponsors we spoke with placed certain limits on participant loan activity, which may reduce the incidence of loan defaults (see fig. 2).

Figure 2: Selected Plan Sponsor Policies May Reduce 401(k) Loan Defaults on Participants

  • Limiting loan amounts to participant contributions: Some plan sponsors said they limited plan loans to participant contributions and any investment earnings from those contributions to reduce early withdrawals of retirement savings.73 For example, one plan sponsor's policy limited the amount a participant could borrow from their plan to 50 percent of participant contributions and earnings, compared to 50 percent of the total account balance.74

  • Implementing a waiting period after loan repayment before a participant can access a new loan: Some plan sponsors said they had implemented a waiting period between plan loans, in which a participant, having fully paid off the previous loan, was temporarily ineligible to apply for another.75 Among plan sponsors who implemented a waiting period, the length varied from 21 days to 30 days.76

  • Reducing the number of outstanding loans: Some plan sponsors we spoke with limited the number of outstanding plan loans to either one or two loans.77 One plan sponsor had previously allowed one new loan each calendar year, but subsequently revised plan policy to allow participants to have a total of two outstanding loans. The plan sponsor said the rationale was to balance limiting participant loan behavior with the ability of participants to access their account balance.

Reducing Impact of Economic Shocks

Some plan sponsors said they had expanded the definition of immediate and heavy financial need beyond the IRS safe harbor to better align with the economic needs of their participants.78 For example, one plan sponsor approved a hardship withdrawal to help a participant pay expenses related to a divorce settlement. Another plan sponsor developed an expanded list of qualifying hardships, including past-due car, mortgage, or rent payments; and payday loan obligations.79

Some plan sponsors implemented loan programs outside their plan, contracting with third-party vendors to provide short-term loans to employees. For example, one plan sponsor instituted a loan program that allowed employees to borrow up to $5,000 from a third-party vendor that would be repaid through payroll deduction. This plan sponsor said the loan program featured an 8 to 12 percent interest rate, and approval was not based on a participant's credit history. The plan sponsor also observed that they had fewer 401(k) loan applications since the third-party loan program was implemented. A second plan sponsor instituted a similar loan program that allowed employees to borrow up to $500 interest free from a third-party vendor. According to this sponsor, to qualify for a loan, an employee must demonstrate financial hardship and have no outstanding plan loans, and is required to attend a financial counseling course if their loans are approved.

Improving Participants' Financial Wellness

Some plan sponsors said they have provided workplace-based financial wellness resources for their participants to improve their financial literacy. Some implemented optional financial wellness programs that covered topics such as investment education, how plan loans work, and the importance of saving for emergencies. These plan sponsors told us they offered on-site financial counseling with representatives of the plan administrator to help provide guidance on financial decision-making; however, other plan sponsors said that — despite their investment in participant-specific financial education — participation in these programs was low.

Stakeholders Suggested Strategies That Could Preserve the Benefits of Early Withdrawals While Reducing Their Long-term Effects

Stakeholders suggested strategies that they believed could help mitigate the long-term effects of early withdrawals of retirement savings on IRA owners and plan participants. They noted that any of these proposed strategies, if implemented, could (1) increase the costs of administering IRAs and plans, (2) require changes to federal law or regulations, and (3) involve tradeoffs between providing access to retirement savings and preserving savings for retirement.

Strategies for IRAs

Stakeholders suggested several strategies that, if implemented, could help reduce early withdrawals from IRAs. These strategies centered on modifying existing rules to reduce early withdrawals from IRAs (and subsequently the amount paid as a result of the additional 10 percent tax for early distributions). Specifically, stakeholders suggested:

  • Raising the age at which the additional 10 percent tax applies: Some stakeholders noted that raising the age at which the additional 10 percent tax for early distributions applies from 59½ to 62 would align it with the earliest age of eligibility to claim Social Security and may encourage individuals to consider a more comprehensive retirement distribution strategy.80 However, other stakeholders cautioned that it could have drawbacks for employees in certain situations. For example, individuals who lose a job late in their careers could face additional tax consequences for accessing an IRA before reaching the age 62.81 In addition, one stakeholder said some individuals may shift to a part-time work schedule later in their careers as they transition to retirement and plan on taking IRA withdrawals to compensate for their lower wages.82

  • Allowing individuals to roll existing plan loans into an IRA: Some stakeholders said that allowing individuals to include an existing plan loan as part of a rollover into an IRA, although currently not allowed, would likely reduce plan loan defaults by giving individuals a way to continue repaying the loan balance. One stakeholder suggested that rolling an existing plan loan into an IRA could be administratively challenging for IRA providers, but doing so to repay the loan may ultimately preserve retirement savings.

  • Allowing IRA loans: While currently a prohibited transaction that could lead to the cessation of an IRA, some stakeholders suggested that IRA loans could theoretically reduce the amounts being permanently removed from the retirement system through early IRA withdrawals.83 One stakeholder said an IRA loan would present a good alternative to an early withdrawal from an IRA account because it would give the account holder access to the balance, defer any tax implications, and improve the likelihood the loaned amount would ultimately be repaid. However, another stakeholder said that allowing IRA loans could increase early withdrawals, given the limited oversight of IRAs, as well as additional administrative costs and challenges for IRA providers.

Strategies for 401(k) Plans

Stakeholders suggested several strategies that, if implemented, could reduce the effect of cashouts at job separation from 401(k) plans.

  • Simplifying the rollover process: Stakeholders proposed two modifications to the current rollover process that they believe could make the process more seamless and reduce the incidence of cashouts. First, stakeholders suggested that a third-party entity tasked with facilitating rollovers between employer plans for a separating participant would likely reduce the incidence of cashouts at job separation. Such an entity could automatically route a participant's account balance from the former plan to a new one.84 One stakeholder said having a third-party entity facilitate the rollover would eliminate the need for a plan participant to negotiate the process.85 Such a service, however, would likely come at cost that may likely be passed onto participants.86 Stakeholders also suggested direct rollovers of account balances between plans could further reduce the incidence of cashouts. One stakeholder, however, cautioned that direct rollovers could have downsides for some participants. For example, participants who prefer to keep their balance in their former employer's plan but provide no direction to the plan sponsor may inadvertently find their account balance rolled into a new employer's plan.

  • Restricting cashouts to participant contributions only: Some stakeholders suggested limiting the assets a participant may access at job separation. For example, some stakeholders said that participants should not be allowed to cash out vested plan sponsor contributions, thus preserving those contributions and their earnings for retirement. However, this strategy could result in participants overseeing and monitoring several retirement accounts.87

Stakeholders suggested several strategies that, if implemented, could limit the adverse effect of hardship withdrawals on retirement savings.

  • Narrowing the IRS safe harbor: Although some plan sponsors are expanding the reasons for a hardship to align with perceived employee needs, some stakeholders said narrowing the IRS safe harbor would likely reduce the incidence of early withdrawals. For example, some stakeholders suggested narrowing the definition of a hardship to exclude the purchase of a primary residence or for postsecondary education costs.88 In addition, one stakeholder said alternatives exist to finance home purchases (mortgages) and postsecondary education (student loans).89 Stakeholders noted that eliminating the purchase of a primary residence and postsecondary education costs from the IRS safe harbor would make hardship withdrawals a tool more strictly used to avoid sudden and unforeseen economic shocks. In combination with the two exclusions, one stakeholder suggested consideration be given to either reducing or eliminating the additional 10 percent tax for early distributions that may apply to hardship withdrawals.

  • Replacing hardship withdrawals with hardship loans: Stakeholders said replacing a hardship withdrawal, which permanently removes money from the retirement system, with a no-interest hardship loan, which would be repaid to the account, would reduce early withdrawals. Under this suggestion, if the loan were not repaid within this predetermined time frame, the remaining loan balance could be considered a deemed distribution and treated as income (similar to the way a hardship withdrawal is treated now).

  • Incorporating emergency savings features into 401(k) plans: Stakeholders said incorporating an emergency savings account into the 401(k) plan structure may help participants absorb economic shocks and better prepare for both short-term financial needs and long-term retirement planning.90 (See fig. 3.)

Figure 3: Emergency Savings Option within a 401(k) Plan Could Better Preserve Retirement Savings, According to Stakeholders

In addition, stakeholders said participants with emergency savings accounts could be better prepared to avoid high interest rate credit options, such as credit cards or payday loans, in the event of an economic shock.91 Stakeholders had several ideas for implementing emergency savings accounts.92 For example, one stakeholder suggested that, were it allowed, plan sponsors could revise automatic account features to include automatic contributions to an emergency savings account.93 Some stakeholders also said emergency savings accounts could be funded with after-tax participant contributions to eliminate the tax implications when withdrawing money from the account.94 However, another stakeholder said emergency savings contributions could reduce contributions to a 401(k) plan.

Conclusions

In the United States, the amount of aggregate savings in retirement accounts continues to grow, with nearly $17 trillion invested in 401(k) plans and IRAs. Early access to retirement savings in these plans may incentivize plan participation, increase participant contributions, and provide participants with a way to address their financial needs. However, billions of dollars continue to leave the retirement system early. Although these withdrawals represent a small percentage of overall assets in these accounts, they can erode or even deplete an individual's retirement savings, especially if the retirement account represents their sole source of savings.

Employers have implemented plan policies that seek to balance the short-term benefits of providing participants early access to their accounts with the long-term need to build retirement savings. However, the way plan sponsors treat outstanding loans after a participant separates from employment has the potential to adversely affect retirement savings. In the event of unexpected job loss or separation, plan loans can leave participants liable for additional taxes. Currently, the incidence and amount of loan offsets in 401(k) plans cannot be determined due to the way DOL collects data from plan sponsors. Additional information on loan offsets would provide insight into how plan loan features might affect long-term retirement savings. Without clear data on the incidence of these loan offsets, which plan sponsors are generally required to include, (but not itemize) on the Form 5500, the overall extent of unrepaid plan loans in 401(k) plans cannot be known.

Recommendation for Executive Action

To better identify the incidence and amount of loan offsets in 401(k) plans nationwide, we recommend that the Secretary of Labor direct the Assistant Secretary for EBSA, in coordination with IRS, to revise the Form 5500 to require plan sponsors to report qualified plan loan offsets as a separate line item distinct from other types of distributions. (Recommendation 1)

Agency Comments and Our Evaluation

We provided a draft of this product to the Department of Labor, the Department of the Treasury, and the Internal Revenue Service for review and comment. In its written comments, reproduced in appendixes IV and V, respectively, DOL and IRS generally agreed with our findings, but neither agreed nor disagreed with our recommendation. DOL said it would consider our recommendation as part of its overall evaluation of the Form 5500, and IRS said it would work with DOL as it responds to our recommendation. The Department of Treasury provided no formal written comments. In addition, DOL, IRS, Treasury and two third-party subject matter experts provided technical comments, which we incorporated in the report, as appropriate

As agreed with your staff, unless you publicly announce the contents of this report earlier, we plan no further distribution until 30 days from the report date. At that time, we will send copies to the Secretary of Labor, Secretary of the Treasury, Commissioner of Internal Revenue, and other interested parties. In addition, the report will be available at no charge on the GAO website at http://www.gao.gov.

If you or your staff have any questions about this report, please contact me at (202) 512-7215 or jeszeckc@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff making key contributions to this report are listed in appendix VI.

Charles A. Jeszeck, Director
Education, Workforce, and Income Security Issues


Appendix I: Objectives, Scope, and Methodology

The objectives of this study were to determine: (1) what are the incidence and amount of retirement savings being withdrawn early; (2) what is known about the factors that might lead individuals to access their retirement savings early; and (3) what strategies or policies, if any, might reduce the incidence and amount of early withdrawals of retirement savings.

Data Analysis

To examine the incidence and amount of early withdrawals from individual retirement accounts (IRA) and 401(k) plans, we analyzed the most recent nationally representative data available in three relevant federal data sources, focusing our analysis on individuals in their prime working years (ages 25 to 55), when possible. For consistency, we analyzed data from 2013 from each data source because it was the most recent year that data were available for all types of early withdrawals we examined. We adjusted all dollar-value estimates derived from each data source for inflation and reported them in constant 2017 dollars. We determined that the data from these sources were sufficiently reliable for the purposes of our report.

  • First, to examine recent incidence and amount of early withdrawals from IRAs and the associated tax consequences for individuals ages 25 to 55, we analyzed IRS estimates based on tax returns as filed by taxpayers before enforcement activity published by the Internal Revenue Service's (IRS) Statistics of Income Division for tax year 2013. Specifically, we analyzed the number of taxpayers reporting early withdrawals from their IRAs in 2013 and the aggregate amount of these withdrawals. To provide additional context on the scope of these early withdrawals, we analyzed the age cohort's total IRA contributions and the end-of-year fair market value of the IRAs, and compared these amounts to the aggregate amount withdrawn. To examine the incidence and amount of taxes paid as a result of the additional 10 percent tax for early distributions, we analyzed estimates on the additional 10 percent tax paid on qualified retirement plans in 2013. Although IRS did not delineate these data by age, we used these data as proxy because IRS assesses the additional 10 percent tax on distributions to taxpayers who have not reached age 59½. Given the delay between a withdrawal date and the date of the tax filing, it is possible that some of the taxes were paid in the year following the withdrawal. We reviewed technical documentation and developed the 95 percent confidence intervals that correspond to these estimates.

  • Second, to examine the incidence and amount of early withdrawals from 401(k) plans, we analyzed data included in the 2014 panel of the U.S. Census Bureau's Survey of Income and Program Participation (SIPP) — a nationally representative survey of household income, finances, and use of federal social safety net programs — along with retirement account contribution and withdrawal data included in the SIPP's Social Security Administration (SSA) Supplement on Retirement, Pensions, and Related Content.1 Specifically, we developed percentage and dollar-value estimates of the incidence and amount of lump sum payments received and hardship withdrawals taken by participants in 401(k) plans in 2013. Because the SIPP is based upon a complex probability sample, we used Balanced Repeated Replication methods with a Fay adjustment to derive all percentage, dollar-total, and dollar-ratio estimates and their 95 percent confidence intervals.2 To better understand the characteristics of individuals who received a lump sum and/or took a hardship withdrawal in 2013, we analyzed a range of selected individual and household demographic variables and identified characteristics associated with a higher incidence of withdrawals. We applied domain estimation methods to make estimates for these subpopulations. (For a list of variables used and the results of our analysis, please see appendix III.) We attempted to develop a multiple regression model to estimate the unique association between each characteristic and withdrawals, but determined that the SIPP did not measure key variables in enough detail to develop persuasive causal explanations. The sample size of respondents receiving lump sums was too small to precisely estimate the partial correlations of many demographic variables at once. Even with adequate sample sizes, associations between broad demographic variables, such as age and income, likely reflected underlying causes, such as retirement and financial planning strategies, which SIPP did not measure in detail.

  • Third, to examine the incidence and amount of unrepaid plan loans from 401(k) plans, we analyzed the latest filing of annual plan data that plan sponsors reported on the Form 5500 to the Department of Labor (DOL) for the 2013 plan year.3 We looked at unrepaid plan loans reported by sponsors of large plans (Schedule H) and small plans (Schedule I). For each schedule, we analyzed two variables related to unrepaid plan loans: (1) deemed distributions of participant loans (which captures the amount of loan defaults by active participants) and (2) benefits distributed directly to participants (which includes plan loan offsets for a variety of reasons, including plan loans that remain unpaid after a participant separates from a plan). Because plan sponsors report data in aggregate and do not differentiate by participant age, we calculated and reported the aggregate of loan defaults identified as deemed distributions in both schedules. We could not determine the amount of plan loan offsets based on the way that plan sponsors are required to report them. Specifically, plan sponsors are required to treat unrepaid loans occurring after a participant separates from a plan as reductions or offsets in plan assets, and are required to report them as part of a larger commingled category of offsets that also includes large-dollar items like rollovers of account balances to another qualified plan or IRA. As a result, we were unable to isolate and report the amount of this category of unrepaid plan loans.

Literature Search

To identify what is known about the factors that might lead individuals to access their 401(k) plans and IRAs and what strategies or policies might reduce the early withdrawal of retirement savings, we performed a literature search using multiple databases to locate documents regarding early withdrawals of retirement savings published since 2008 and to identify experts for interviews. The search yielded a wide variety of scholarly articles, published articles from various think tank organizations, congressional testimonies, and news reports. We reviewed these studies and identified factors that lead individuals to withdraw retirement savings early, as well as potential strategies or policies that might reduce this behavior. The search also helped us identify additional potential interviewees.

Interviews

To answer our second and third objectives, we visited four metropolitan areas and conducted 51 interviews with a wide range of stakeholders that we identified in the literature. In some cases, to accommodate stakeholder schedules, we conducted phone interviews or accepted written responses. Specifically, we interviewed human resource professionals from 22 private-sector companies (including 4 written responses), representatives from 8 plan administrators, 13 retirement research experts (including 1 written response), representatives from 4 industry associations, representatives from 2 participant advocacy organizations, and representatives from 2 financial technology companies.

We conducted in-person interviews at four sites to collect information from three different groups: (1) human resource officials in private-sector companies, (2) top 20 plan administrators or recordkeepers, and (3) retirement research experts. We selected site visit locations in four metropolitan locations that were home to representatives of each group. To select companies for potential interviews, we reached out to a broad sample of Fortune 500 companies that offered a 401(k) plan to employees and varied by geographic location, industry, and number of employees. We selected plan administrators based on Pensions and Investments rankings for assets under management and number of individual accounts. We selected retirement research experts who had published research on early withdrawals from retirement savings, as well as experts that we had interviewed in our prior work. Based on these criteria, we conducted site visits in Boston, Massachusetts; Chicago, Illinois; the San Francisco Bay Area, California; and Seattle, Washington. We held interviews with parties in each category who responded affirmatively to our request. In each interview, we solicited names of additional stakeholders to interview. We also interviewed representatives of organizations, such as financial technology companies, participant advocacy organizations, industry associations, and plan administrators focused on small businesses, whose work we deemed relevant to our study.

We developed a common question set for each stakeholder category that we interviewed. We based our interview questions on our literature review, research objectives, and the kind of information we were soliciting from each stakeholder category. In each interview, we asked follow-up questions based on the specific responses provided by interviewees.

  • In our company interviews, we asked how companies administered retirement benefits for employees; company policies and procedures regarding separating employees and the disposition of their retirement accounts; company policies regarding plan loans, hardship withdrawals, and rollovers from other 401(k) plans; and company strategies to reduce early withdrawals from retirement savings.

  • In our interviews with plan administrators, we asked about factors that led individuals to access their retirement savings early, how plan providers interacted with companies and separating employees, available data on loans and hardship withdrawals from client retirement plans, and potential strategies to reduce the incidence and amount of early withdrawals.

  • In our interviews with retirement research experts, financial technology companies, participant advocacy organizations, and industry associations we asked about factors that led individuals to make early withdrawals from their retirement savings and any potential strategies that may reduce the incidence and amount of early withdrawals.

In our interviews with plan administrators and retirement research experts, we also provided a supplementary table outlining 37 potential strategies to reduce early withdrawals from retirement savings. We asked interviewees to comment on the strengths and weaknesses of each strategy in terms of its potential to reduce early withdrawals, and gave them opportunity to provide other potential strategies not listed in the tables. We developed the list of strategies based on the results of our literature review.

Some interviewees also provided us with additional data and documents to assist our research. For example, some companies and plan administrators we interviewed provided quantitative data on the number of plan participants, the average cashout or rollover amounts, the percentage of participants who took loans or hardship withdrawals from their retirement accounts, and known reasons for these withdrawals. Some research experts also provided us with documentation, including published articles and white papers that supplemented our interviews and literature review. All data collected through these methods are nongeneralizable and reflect the views and experiences of the respondents and not the entire population of their respective constituent groups.

Analysis of Interview Responses

To answer our second and third objectives, we analyzed the content of our stakeholder interview responses and corroborated our analysis with information obtained from our literature review and quantitative information provided by our interviewees. To examine what is known about the factors leading individuals to access retirement savings early, we catalogued common factors that stakeholders identified as contributing to early withdrawals from retirement savings. We also collected information on plan rules governing early participant withdrawals of retirement savings.

To identify potential strategies or policies that might reduce the incidence and amount of early withdrawals, we analyzed interview responses and catalogued (1) company practices that employers identified as having an effect in reducing early withdrawals and (2) strategies that stakeholders suggested that could achieve a similar outcome. GAO is not endorsing or recommending any strategy in this report, and has not evaluated these strategies for their behavioral or other effects on retirement savings or on tax revenues.


Appendix II: Selected Provisions Related to Early Withdrawals from 401(k) Plans and Individual Retirement Accounts (IRAs)

Table 2: Selected Provisions Related to 401(k) Plans and the Types of Early Withdrawals Affected

Selected provision

Requirements

Cashouts

Hardship withdrawals

Loan defaults

26 U.S.C. § 72(p)(2)

Sets the maximum amount that the plan can permit as a loan to a participant as generally (1) the greater of $10,000 or 50 percent of a participant's vested account balance or (2) $50,000, whichever is less.

 

 

X

26 U.S.C. § 72(t)(1)

Imposes an additional 10 percent tax for early distributions from qualified retirement plans

X

X

X

26 U.S.C. § 72(t)(2)

Provides exceptions for paying the additional 10 percent tax on early distributions from qualified retirement plans in certain instances, including those involving death, disability, or severance from service after age 55..

X

 

 

26 U.S.C. § 401(a)(31)(B)

Requires automatic rollover of certain mandatory (under plan terms) distributions unless a participant opts out and caps involuntary cashouts at job separation at $1,000.

X

 

 

26 U.S.C. § 401(k)

Provides for a cash or deferred arrangement under section 401(k) of the Internal Revenue Code.

X

X

X

26 U.S.C. § 401(k)(2)(B)

Provides that a participant's elective contributions to a 401(k) plan generally may not be distributed prior to the occurrence of certain events, such as severance from employment or a hardship.

X

X

 

26 U.S.C. § 401(k)(14)(A)

Allows for certain amounts to be distributed from qualified plans upon hardship, including qualified nonelective contributions and qualified matching contributions.

 

X

 

26 U.S.C. § 401(k)(14)(B)

Allows employees in qualified plans to take hardship withdrawals without seeking any available loans under the plan.

 

X

 

26 U.S.C. § 402(c)(4)(C)

Prohibits a hardship withdrawal from being rolled over into an IRA or other qualified plan.

 

X

 

26 U.S.C. § 411(a)(11)(A)

Provides that plan administrators generally may not cash out an account balance that exceeds $5,000 without the consent of the participant.

X

 

 

29 U.S.C. § 1022

Requires plan administrators to furnish participants with a summary plan description to ensure that participants and beneficiaries in participant-directed individual account plans, among others, have the information relating to their benefits and rights under their plans.

X

X

X

26 C.F.R. § 1.401(k)-1(d)(3)(iii)

Provides a safe harbor for certain hardship withdrawals, which includes distributions for certain medical, tuition, and funeral expenses.

 

X

 

Source: GAO review of federal laws and regulations, and interviews with stakeholders. | GAO-19-179

Note: We identified many of these provisions in GAO-09-715 and included additional provisions based on interviews with stakeholders. This table is not intended to be an exhaustive list of the provisions relating to early withdrawals of retirement savings from 401(k) plans.

Table 3: Selected Provisions Related to Traditional Individual Retirement Accounts (IRA)

Selected provision

Requirements

26 U.S.C. § 72(t)(2)(E) and (F)

Provides an exception for distributions for qualified higher education expenses and for qualified “first-time” home purchases made before age 59½ from the additional 10 percent tax for early distributions

26 U.S.C. § 72(t)(8)

Defines “qualified first-time homebuyer distribution” and “first-time homebuyer,” and prescribes the lifetime dollar limit on such distributions, among other things.

26 U.S.C. § 408(a)

Allows eligible individuals to make tax-deductible contributions to individual retirement accounts, subject to limits based, for example, on income and pension coverage.

26 U.S.C. § 408(e)(2)

Provides for the loss of exemption for an IRA if the IRA owner engages in a prohibited transaction, which results in the IRA being treated as distributing all of its assets to the IRA owner at the fair market value on the first day of the year in which the transaction occurred.

26 U.S.C. § 4975(c)(1)(B)

Defines a prohibited transaction to include the lending of money or other extension of credit between a plan and a disqualified person.

Source: GAO review of federal laws and regulations, and interviews with stakeholders. | GAO-19-179

Note: We identified many of these provisions through interviews with stakeholders. This table is not intended to be an exhaustive list of the provisions relating to early withdrawals of retirement savings from traditional IRAs.

Table 4: Selected Provisions Related to Roth Individual Retirement Accounts (IRA)

Selected provision

Requirements

26 U.S.C. § 408A

Allows eligible individuals to make contributions to a Roth IRA that are not tax-deductible. Distributions from the account can generally be treated as a qualified distribution if a distribution is made on or after the Roth IRA owner reaches age 59½ and the distributions is made after the 5-taxable year period beginning when the account was initially opened.

26 U.S.C. § 4975(c)(1)(B)

Defines a prohibited transaction to include the lending of money or other extension of credit between a plan and a disqualified person.

Source: GAO review of federal laws and regulations, and interviews with stakeholders. | GAO-19-179

Note: We identified many of these provisions based on interviews with stakeholders. This table is not intended to be an exhaustive list of the provisions relating to early withdrawals of retirement savings from Roth IRAs.


Appendix III: Estimated Incidence of Certain Early Withdrawals of Retirement Savings

Table 5: Estimates of the Incidence of Early Withdrawals from Individual Retirement Accounts and 401(k) Plans, by Individual Characteristics

Table 5 (Continued)

Table 5 (Continued)


Appendix IV: Comments from the Department of Labor

Charles A. Jeszeck
Director, Education, Workforce and Income Security
Government Accountability Office
Washington D.C. 20548

Dear Mr. Jeszeck:

Thank you for the opportunity to review the Government Accountability Office draft report entitled “Retirement Savings Additional Data and Analysis Could Provide Insight into Early Withdrawals.” The draft report contains one recommendation for the Department of Labor (Department or DOL). Specifically, you recommend that the Department, in coordination with the Internal Revenue Service (IRS), revise the Form 5500 Annual Report to require employee benefit plans to report loan defaults that occur after a participate leaves a 401(k)-type plan as a separate line item distinct from other types of distributions.

As we have noted in connection with other GAO recommendations on Form 5500 reporting, the Department, together with the IRS and the Pension Benefit Guaranty Corporation (collectively “Agencies”), published in July 2016 a joint Notice of Proposed Forms Revisions and a related DOL Notice of Proposed Rulemaking as part of an overall re-examination of the Form 5500. Public comments are posted on EBSA's website. The stated goal of the project is to modernize and improve the Form 5500 and enhance the Agencies' ability to collect employee benefit plan data that best meets the needs of changing compliance projects, programs, and activities. This regulatory project continues to be listed as a long term action on the Department's semi-annual agenda of regulatory and deregulatory actions, but the Department has not decided whether to pursue changes to the forms or DOL regulations nor decided on a timeline for any changes the DOL ultimately might decide to propose.

The Form 5500 reporting change that GAO is now recommending would itself require notice and comment rulemaking. The Department agrees with GAO that 401(k)-type plans should already be keeping records that differentiate loan offsets from other benefit distributions, but the Department does not believe that it would be appropriate, or an efficient allocation of resources, for the Department to pursue GAO's recommended reporting change on loan offsets in isolation. Rather, the Department will consider GAO's recommendation as part of its overall evaluation of the Form 5500 modernization project.

Thank you again for sharing your draft report and recommendation.

Sincerely,

Preston Rutledge
Assistant Secretary


Appendix V: Comments from the Internal Revenue Service

February 28, 2019

Mr. Charles A. Jeszeck
Director, Education, Workforce, and Income Security Issues
United States Government Accountability Office
441 G Street, NW
Washington, DC 20548

Dear Mr. Jeszeck:

Thank you for the opportunity to review the draft report of the Government Accountability Office entitled "Retirement Savings: Additional Data and Analysis Could Provide Insight into Early Withdrawals" (GAO-19-179, Job Code 102365). We appreciate you highlighting this important issue.

The draft report contains one recommendation for the Department of Labor (DOL). Specifically, it recommends that DOL, in coordination with the Internal Revenue Service (IRS), revise the Form 5500, Annual Return/Report of Employee Benefit Plan, to require employee benefit plans to report loan defaults that occur after a participant leaves a 401(k)-type plan as a separate line item distinct from other types of distributions.

The IRS will work with DOL as it responds to this GAO recommendation. We note that the IRS already gathers information regarding participant loans that are not repaid under Form 1099-R, Distributions from Pensions, Annuities, Retirement or Profit-Sharing Plans, IRAs, Insurance Contracts, etc. In particular, Code L of Box 7 identifies “Loans treated as distributions,” and Code M of Box 7 identifies “Qualified plan loan offset” amounts. Thus, information is already available with respect to participant loans that are treated as deemed distributions due to, for example, plan participants' failure to repay participant loans. Information is also available with respect to plan loan offset amounts that are distributed from a qualified employer plan to a participant or beneficiary solely by reason of: (1) the termination of a qualified employer plan or (2) the failure to meet the repayment terms of the loan from such a plan because of the severance from employment of the participant.

We expect that decisions regarding additional reporting on the Form 5500 will take into account the information that is already required on Form 1099-R. We also expect that these decisions will comport with the Paperwork Reduction Act, 5 C.F.R. § 1320.9, which requires that agencies certify that information collected, among other things, is “necessary for the proper performance of the functions of the agency, including that the information to be collected will have practical utility” and is not “unnecessarily duplicative of information otherwise reasonably accessible to the agency.”

We appreciate GAO's interest in the important topic of asset withdrawal from the tax-favored retirement system and the helpful information on this topic contained in the draft report. If you have questions, please contact me, or a member of your staff may contact Catherine L. Jones, Acting Director, Employee Plans, at 202-317-8700.

Sincerely,

Kirsten B. Wielobob
Deputy Commissioner for Services and Enforcement


Appendix VI: GAO Contact and Staff Acknowledgment

GAO Contact:

Charles Jeszeck, (202) 512-7215 or jeszeckc@gao.gov

Staff Acknowledgment:

In addition to the contact named above, Dave Lehrer (Assistant Director); Jonathan S. McMurray (Analyst-in-Charge); Gustavo O. Fernandez; Sean Miskell; Jeff Tessin; and Adam Wendel made key contributions to this report. James Bennett, Holly Dye, Sara Edmondson, Sarah Gilliland, Sheila R. McCoy, Ed Nannenhorn, Katya Rodriguez, MaryLynn Sergent, Linda Siegel, Rachel Stoiko, Frank Todisco, and Sonya Vartivarian also provided support.

Related GAO Reports

The Nation's Fiscal Health: Action Is Needed to Address the Federal Government's Future. GAO-18-299SP. Washington, D.C.: June 21, 2018.

The Nation's Retirement System: A Comprehensive Re-evaluation is Needed to Better Promote Future Retirement Security. GAO-18-111SP. Washington, D.C.: October 18, 2017.

Retirement Security: Improved Guidance Could Help Account Owners Understand the Risks of Investing in Unconventional Assets. GAO-17-102. Washington, D.C.: December 8, 2016.

401K Plans: Effects of Eligibility and Vesting Policies on Workers' Retirement Savings. GAO-17-69. Washington, D.C.: October 21, 2016.

Retirement Security: Low Defined Contribution Savings May Pose Challenges. GAO-16-408. Washington, D.C.: May 5, 2016.

Retirement Security: Shorter Life Expectancy Reduces Projected Lifetime Benefits for Lower Earners. GAO-16-354. Washington, D.C.: March 25, 2016.

Social Security's Future: Answers to Key Questions. GAO-16-75SP. Washington, D.C.: October 27, 2015.

Retirement Security: Federal Action Could Help State Efforts to Expand Private Sector Coverage. GAO-15-556. Washington, D.C.: September 10, 2015.

Highlights of a Forum: Financial Literacy: The Role of the Workplace. GAO-15-639SP. Washington, D.C.: July 7, 2015.

401(K) Plans: Greater Protections Needed for Forced Transfers and Inactive Accounts. GAO-15-73. Washington, D.C.: November 21, 2014.

Older Americans: Inability to Repay Student Loans May Affect Financial Security of a Small Percentage of Retirees. GAO-14-866T. Washington, D.C.: September 10, 2014.

Financial Literacy: Overview of Federal Activities, Programs, and Challenges. GAO-14-556T. Washington, D.C.: April 30, 2014.

Retirement Security: Trends in Marriage and Work Patterns May Increase Economic Vulnerability for Some Retirees. GAO-14-33. Washington, D.C.: January 15, 2014.

401(K) Plans: Labor and IRS Could Improve the Rollover Process for Participants. GAO-13-30. Washington, D.C.: March 7, 2013.

Retirement Security: Women Still Face Challenges. GAO-12-699. Washington, D.C.: July 19, 2012.

401(K) Plans: Policy Changes Could Reduce the Long-term Effects of Leakage on Workers' Retirement Savings. GAO-09-715. Washington, D.C: August 28, 2009.

GAO's Mission

The Government Accountability Office, the audit, evaluation, and investigative arm of Congress, exists to support Congress in meeting its constitutional responsibilities and to help improve the performance and accountability of the federal government for the American people. GAO examines the use of public funds; evaluates federal programs and policies; and provides analyses, recommendations, and other assistance to help Congress make informed oversight, policy, and funding decisions. GAO's commitment to good government is reflected in its core values of accountability, integrity, and reliability.

Obtaining Copies of GAO Reports and Testimony

The fastest and easiest way to obtain copies of GAO documents at no cost is through GAO's website (https://www.gao.gov). Each weekday afternoon, GAO posts on its website newly released reports, testimony, and correspondence. To have GAO e-mail you a list of newly posted products, go to https://www.gao.gov and select “E-mail Updates.”

Order by Phone

The price of each GAO publication reflects GAO's actual cost of production and distribution and depends on the number of pages in the publication and whether the publication is printed in color or black and white. Pricing and ordering information is posted on GAO's website, https://www.gao.gov/ordering.htm.

Place orders by calling (202) 512-6000, toll free (866) 801-7077, or TDD (202) 512-2537.

Orders may be paid for using American Express, Discover Card, MasterCard, Visa, check, or money order. Call for additional information.

Connect with GAO

Connect with GAO on Facebook, Flickr, Twitter, and YouTube.

Subscribe to our RSS Feeds or E-mail Updates. Listen to our Podcasts.

Visit GAO on the web at https://www.gao.gov.

To Report Fraud, Waste, and Abuse in Federal Programs

Contact:

Website: https://www.gao.gov/fraudnet/fraudnet.htm

Automated answering system: (800) 424-5454 or (202) 512-7700

Congressional Relations

Orice Williams Brown, Managing Director, WilliamsO@gao.gov, (202) 512-4400, U.S. Government Accountability Office, 441 G Street NW, Room 7125, Washington, DC 20548

Public Affairs

Chuck Young, Managing Director, youngc1@gao.gov, (202) 512-4800, U.S. Government Accountability Office, 441 G Street NW, Room 7149, Washington, DC 20548

Strategic Planning and External Liaison

James-Christian Blockwood, Managing Director, spel@gao.gov, (202) 512-4707, U.S. Government Accountability Office, 441 G Street NW, Room 7814, Washington, DC 20548

FOOTNOTES

1Investment Company Institute (ICI), 2018 Investment Company Fact Book, 58th edition (2018).

2In general, there are three main sources of early withdrawals from 401(k) plans: cashouts of account balances at job separation, hardship withdrawals, and plan loans that are not repaid. We use the term “cashout” to refer to any lump-sum distribution received by an employee at job separation that is not subsequently rolled over into a qualified retirement account or an IRA. Plan loans that are repaid are generally not included as a form of early withdrawal. In addition, IRA owners can generally take withdrawals for any reason at any age, although certain distributions taken by an IRA owner before reaching age 59½ can be subject to additional tax.

3We selected the 25 to 55 age cohort to capture the population of working-age adults. DOL data on the amount of loan defaults do not include information on the ages of participants who default on their plan loans.

4The SSA Supplement to the 2014 SIPP collected data on respondents who owned a 401(k), 403(b), 503(b), or a Thrift Savings Plan (TSP) account. In this report, for ease of reference we generally refer to these plans as 401(k) plans.

5Form 5500 is the primary source of information collected by the federal government regarding the operation, funding, assets, and investments of private pension plans and other employee benefit plans. DOL, IRS, and the Pension Benefit Guaranty Corporation (PBGC) jointly developed the Form 5500 so employee benefit plans could satisfy annual reporting requirements under the Employee Retirement Income Security Act of 1974 (ERISA) and the Internal Revenue Code (IRC). Approximately 700,000 private pension plans must file Form 5500 annually.

6We previously reported that nearly half of the private-sector workforce did not participate in a workplace retirement savings program in 2012, mainly due to lack of access. Depending on how participation is measured, between 45 and 54 percent of all private-sector workers were participating in a workplace retirement savings program. See GAO, Retirement Security: Federal Action Could Help State Efforts to Expand Private Coverage, GAO-15-556 (Washington, D.C: Sept. 10, 2015) and GAO, The Nation's Retirement System: A Comprehensive Re-evaluation Is Needed to Better Promote Future Retirement Security, GAO-18-111SP (Washington, D.C.: Oct. 18, 2017).

7See GAO-18-111SP and GAO, 401(k) Plans: Policy Changes Could Reduce the Long-term Effects of Leakage on Workers' Retirement Savings, GAO-09-715 (Washington, D.C.: Aug. 28, 2009).

8Number of Pension Plans by Type of Plan, 1975-2016 in Employee Benefits Security Administration, United States Department of Labor, Private Pension Plan Bulletin Historical Tables and Graphs, 1975-2016 (Washington, D.C., Dec. 2018) table E1.

9Subject to certain limits employer contributions to qualified plans are a tax-deductible business expense and, in general, contributions and investment earnings on those contributions are not taxed as income until the employee withdraws them from the plan.

10Additionally, some plan sponsors offer Roth 401(k) plans that allow plan participants to make elective after-tax contributions through payroll deduction. Employee contributions can be a set dollar amount or a percentage of pay, within the contribution limits adjusted annually for the cost of living. In 2019, participants can contribute up to $19,000 per year. Catch-up contributions, permitted for employees age 50 and over who participate in 401(k) plans, allowed up to an additional $6,000 per year in 2019. See 26 U.S.C. § 402(g)(1)(A)-(C) & (4). Employer contributions are held to other limits. In 2019, the combined total annual additions to a participant's account by the participant and employer cannot exceed the lesser of 100 percent of the participant's compensation, or $56,000 ($62,000 including catch-up contributions). See 26 U.S.C. § 415(c)-(d). The Internal Revenue Code excepts certain early withdrawals from qualified plans from the additional tax for early distributions if the distributions are made to a beneficiary or estate on or after the death of the employee; attributable to certain disability of the employee; part of a series of substantially equal periodic payments over the life or life expectancy of the employee; made due to an IRS levy of the plan under 26 U.S.C. § 6331;or made to a participant after separation from service after attainment of age 55; among other things.

11For more information on 401(k) plan service provider arrangements and fees see GAO, 401(k) Plans: Increased Educational Outreach and Broader Oversight May Help Reduce Plan Fees, GAO-12-325 (Washington, D.C.: Apr. 24, 2012).

12Distributions from Roth IRAs generally not included in gross income are those (1) made on or after the date the IRA owner age 59½, (2) made to a beneficiary on or after the death of the IRA owner, (3) attributable to certain disability of the IRA owner, or (4) made for qualified “first-time” home purchases. In addition, 5 years must have elapsed since the individual's first Roth IRA contribution for a withdrawal to be tax-free. See 26 U.S.C. § 408A(d).

13ICI, “Appendix: Additional Data on IRA Ownership in 2017,” ICI Research Perspective, vol. 23, no.10A (Dec. 2017).

14The remaining assets in 2017 consisted of private DB plans ($3.1 trillion or 11 percent), state and local government DB plans ($4.3 trillion or 15 percent), federal DB plans ($1.7 trillion or 6 percent), and annuity reserves ($2.2 trillion or 8 percent). See ICI, “The US Retirement Market, Second Quarter 2018,” (Sept. 27, 2018). (Percentages calculated by GAO).

15Ibid. ICI's data distinguishes between traditional IRAs and Roth IRAs. (Percentages calculated by GAO.)

16PBGC also uses the form as a tool to inform its efforts to insure the benefits of participants in most private-sector DB pension plans.

17For a list of selected provisions potentially related to early withdrawals from 401(k) plans and IRAs, see appendix II.

18See 26 C.F.R. § 1.401(k)-1(d)(3)(i).

19See 26 C.F.R. § 1.401(k)-1(d)(3)(iii)(B). Plan sponsors may allow hardship withdrawals for reasons other than those enumerated in the safe harbor; however, under IRS regulations, the determination of whether a participant has an immediate and heavy financial need is to be determined based on all the relevant facts and circumstances.

20In November 2018, IRS issued proposed regulations that would allow certain distributions relating to expenses and losses incurred on account of a disaster to be deemed to be on account of an immediate and heavy financial need. See Hardship Distributions of Elective Contributions, Qualified Matching Contributions, Qualified Nonelective Contributions, and Earnings, 83 Fed. Reg. 56,763, 56,767 (Nov. 14, 2018).

21For a full list of available exceptions to the additional tax on early withdrawals for IRAs and 401(k) plans, respectively, see IRS, “Retirement Topics — Exceptions to Tax on Early Distributions” (Dec. 29, 2017).

22BLS, Economic News Release, “Table 1. Number of Jobs Held by Individuals From Age 18 to Age 50 in 1978-2014 by Educational Attainment, Sex, Race, Hispanic or Latino Ethnicity, and Age” (Aug. 24, 2017).

23BLS, “28 Percent of Workers Age 55 and Over Have Been with Their Current Employer 20 Years or More,” The Economics Daily (Sept. 27, 2016).

24Alicia H. Munnell, Anek Belbase, and Geoffrey T. Sanzenbacher, “An Analysis of Retirement Models to Improve Portability and Coverage,” Center for Retirement Research at Boston College (March 2018).

2526 U.S.C. §402(f)(1).

26For more information on the effects of cashouts see GAO-09-715. Not all plans accept rollovers from other plans. For detailed information on rollover eligibility for different plans, see IRS, “Rollovers of Retirement Plan and IRA Distributions” (May 30, 2018).

27As we previously reported, a forced transfer may occur when a plan participant separates from an employer, leaving vested savings of more than $1,000 but not over $5,000 in the employer's 401(k) plan and provides no direction regarding its disposition. In those instances, a plan sponsor may transfer the account balance to an IRA, although the plan sponsor may decide not to force the account from the plan. See GAO, 401(k) Plans: Greater Protections Needed for Forced Transfers and Inactive Accounts, GAO-15-73 (Washington, D.C.: Nov. 21, 2014). Prior to EGTRRA, plans could, in the absence of participant instruction, distribute balances of not more than $5,000 by disbursing them directly to the participant.

28Plan sponsors may also opt to transfer balances of $1,000 or less into an IRA.

29See 26 U.S.C. § 72(t). The IRC exempts certain early distributions from the additional 10 percent tax, such as those made to a participant after separation from service with an employer after reaching age 55. See 26 U.S.C. § 72(t)(2)(A)(v). These early distributions are also subject to federal income tax withholding and taxed at the marginal income tax rate as ordinary income.

30See 26 U.S.C. § 3405(c)(1)(B).

31We previously reported that life expectancy varies substantially across different groups, with low-income individuals generally living fewer years than those with high incomes. See GAO, Retirement Security: Shorter Life Expectancy Reduces Projected Lifetime Benefits for Lower Earners, GAO-16-354, (Washington, D.C.: Mar. 25, 2016). A similar study found that while Americans are generally living longer than previous generations, those with higher socioeconomic status have substantially longer lifespans than others. See Matthew S. Rutledge, “How Is the Mortality Gap Affecting Social Security Progressivity?” Center for Retirement Research at Boston College (Sept. 2008).

32PricewaterhouseCoopers, “Special Report: Financial Stress and the Bottom Line: Why Employee Financial Wellness Matters to your Organization” (Sept. 2017).

33Annamaria Lusardi and Olivia S. Mitchell, “Financial Literacy and Planning: Implications for Retirement Wellbeing,” National Bureau of Economic Research, Working Paper No. 17078 (May 2011). Researchers have also found that one-third of the population lacks an understanding of compounding interest. See Lusardi and Peter Tufano, “Debt Literacy, Financial Experiences, and Overindebtedness.” National Bureau of Economic Research, Working Paper 14808 (March 2009).

34Board of Governors of the Federal Reserve System, Report on the Economic Well-Being of U.S. Households in 2017 (May 2018).

35In 2015, we convened a select group of leaders and experts who identified a number of employer practices shown to help employees improve their overall financial wellness. Several experts noted that reducing financial stresses can improve employee productivity and aid recruiting and retention. See GAO, Highlights of a Forum: Financial Literacy: The Role of the Workplace, GAO-15-639SP (Washington, D.C.: July 7, 2015). Recent research has also found that financial stress, particularly regarding debt, has a negative effect on worker productivity. See Fidelity Investments, “What is Total Well-being: Illuminating the Connections among Health, Wealth, Work, and Life,” Workplace Solutions Thought Leadership (Spring 2018), and PricewaterhouseCoopers, “Special Report: Financial Stress and the Bottom Line: Why Employee Financial Wellness Matters to your Organization” (Sept. 2017).

36This estimate has a 95 percent margin of error within +/- 5 percent of the estimate itself. All dollar amounts have been adjusted for inflation and are reported in constant 2017 dollars.

37When developing estimates for IRA withdrawals, IRS excludes direct rollovers and conversions to Roth IRAs. However, in written responses to questions, an IRS official indicated that the estimate may include some amount of IRA monies cashed out by an IRA owner and subsequently rolled over into a new IRA or rollovers not identified as such on Form 1099-R, but said these occurrences are likely infrequent.

38The 95 percent margin of error for this estimate was +/- 0.15 percentage points. These proportions apply to all IRA owners in the cohort whether or not they took a withdrawal. The proportion of assets and contributions of IRA owners who withdrew assets would have been higher, but IRS data did not allow such calculations. As we previously reported, the bulk of assets in IRAs do not stem from annual contributions (capped at $6,000 annually for 2019), but rather from rollovers from 401(k) plans into IRAs. See GAO, 401(k) Plans: Labor and IRS Could Improve the Rollover Process for Participants, GAO-13-30 (Washington, D.C.: Mar. 7, 2013).

39This estimate has a 95 percent margin of error within +/- 9.5 percent of the estimate itself.

40This estimate has a 95 percent margin of error within +/- 11.3 percent of the estimate itself.

41Our estimate is derived from withdrawals of $1,000 or more that 401(k) participants received in 2013. We excluded withdrawals less than this amount because of sample size restrictions. This estimate has a 95 percent margin of error within +/- 18.5 percent of the estimate itself. The estimate is lower than the cashout estimate that we reported in 2009. In that report, we estimated the incidence and amount of cashouts using a variable included in the 2006 SIPP panel that asked the population of 401(k) plan participants if they had ever received a lump sum payment from their account and the amount of such payments. The redesigned 2014 SIPP included a similar question, but this question was only asked to a small population of participants who expected to receive benefits from a DB plan. Therefore, we used a variable that asked all respondents whether or not they had received a lump sum payment from 401(k) plan or pension plan in 2013, and the amount.

42The Form 5500 does not collect any demographic data on participants, including their age. A plan's terms will generally specify how the plan handles unrepaid loan balances. A loan that is in default is generally treated as a distribution from the plan of the entire outstanding balance of the loan (a “deemed distribution”). A plan may provide that a loan does not become a deemed distribution until the end of the calendar quarter following the quarter in which the repayment was missed. A distribution of a plan loan offset generally occurs when the accrued benefit of the participant or beneficiary is reduced in order to repay the loan; the amount of the balance that is offset again the loan is an actual distribution. A distribution of a plan loan offset could occur in a variety of circumstances, such as when the terms of the plan require that. However, a plan may be prohibited from making a plan loan offset under certain IRC provisions prohibiting or limiting distributions to active employees.

43According to Form 5500 instructions, active participants are individuals who are currently in employment covered by the plan and who are earning or retaining credited service under the plan. Employers generally treat plan loan defaults by active participants as deemed distributions, and unpaid loan amounts are subject to income tax liability and any additional tax consequences. Employers generally treat unrepaid loan balances of terminated participants — individuals who have left the plan and had their account balance distributed in some way — as actual distributions to participants, and offset the plan's assets accordingly. In tax year 2018 plan sponsors will begin reporting the amount of qualified plan loan offsets using Code M on IRS Form 1099-R sent to individuals receiving a plan distribution.

44IRS Publication 575 defines a “plan loan offset” as the amount that an employer plan account balance is reduced, or offset, to repay a loan from the plan. For a plan loan offset to be a “qualified plan loan offset,” the offset must be caused by either a plan termination or an employee's severance from the employer sponsoring the plan. See Publication 575, Pension and Annuity Income, Department of Treasury, Internal Revenue Service.

45In 2009, we reported on the amount that left the retirement system in the form of loan defaults. See GAO-09-715. A subsequent study in 2017 found this amount included only loan defaults of active plan participants, and did not include the loan offsets of terminated participants who had an outstanding loan balance when they left a plan. This study, which used recordkeeper data from 2004 to 2009 to calculate the amount of offsets caused by unrepaid plan loans, estimated that offsets from terminated participants outnumbered loan defaults among active participants by a 10:1 margin. See Timothy (Jun) Lu, Olivia S. Mitchell, Stephen P. Utkus, and Jean A. Young. “Borrowing From the Future? 401(k) Plan Loans and Loan Defaults,” National Tax Journal, National Tax Association, vol. 70(1), 77-110 (March 2017).

46IRS estimates of amounts paid as a result of the additional 10 percent tax for early distributions are derived directly from tax filings directly reported by taxpayers that have not been subject to IRS audit. The 95 percent confidence interval for this estimate has a margin of error within +/- 4 percent of the estimate itself.

47Exceptions to the additional 10 percent tax for early distributions apply to certain withdrawals made by individuals under age 59½. For example, IRA owners and 401(k) participants under age 59½ are excepted from the additional 10 percent tax in the case of a distribution made to a beneficiary (or estate) on or after the death of the owner or participant, attributable to certain disability of owner or participant, or for certain unreimbursed medical expenses above a certain threshold. See 26 U.S.C. § 72(t)(2)(A)(ii)-(iii) & (B). In addition, the IRC exempts IRA owners from the additional 10 percent tax for distributions for qualified higher education expenses, health insurance premiums paid while unemployed, and qualified “first-time” home purchases up to $10,000. See 26 U.S.C. § 72(t)(2)(D)-(F).

48GAO, Effects of Eligibility and Vesting Policies on Workers' Retirement Savings, GAO-17-69 (Washington, D.C.: Oct. 21, 2016).

49We have previously reported on issues associated with forced transfers of 401(k) plans. See GAO-15-73.

50Plan sponsors are required to provide a notice to separating participants, within a reasonable period of time before making an eligible rollover distribution, describing options available for handling their account balance. See 26 U.S.C. § 402(f)(1). This notice must explain the tax implications of the different distribution options, including the rules applicable to a direct rollover, the rules under which the participant may defer tax on the distribution if it is contributed in a rollover to an eligible retirement plan within 60 days of the distribution, and the mandatory withholding of 20 percent on certain distributions (including those that result in an indirect rollover). IRS regulations generally require plan sponsors to provide the notice to participants no less than 30 days, and no more than 90 days, before the date of distribution. See 26 C.F.R. § 1.402(f)-1, Q&A 2. Pursuant to provisions in the Pension Protection Act of 2006, IRS issued proposed regulations on October 9, 2008, to substitute 180 days for 90 days in the regulations. See Notice to Participants of Consequences of Failing to Defer Receipt of Qualified Retirement Plan Distributions; Expansion of Applicable Election Period and Period for Notices, 73 Fed. Reg. 59,575 (Oct. 9, 2008). Although final regulations effecting this change have not be promulgated, the preamble to the proposed regulation states that “plans may rely on these proposed regulations for notices provided . . . during the period beginning on the first day of the plan year beginning on or after January 1, 2017 and ending on the effective date of final regulations.”

51For a successful indirect rollover, the 20 percent withheld is resolved when the participant files their federal income taxes, very likely after the 60-day rollover window has passed.

52A plan could lose its tax-preferred status if it fails to comply with the plan qualification requirements in the IRC. In general, if a plan accepts a rollover contribution, the contribution will be treated as meeting the qualification requirements if the plan administrator (1) reasonably concludes that the rollover contribution is valid, and (2) distributes any ineligible rollover contribution, with earnings, within a reasonable time of discovering the error. See 26 C.F.R. § 1.401(a)(31)-1, Q&A 14. Treasury expressed its view that a plan administrator receiving a rollover could have reasonably concluded that an incoming contribution was a valid rollover contribution after checking the sending plan's Form 5500 submissions available in DOL's EFAST2 database and finding no indication that the plan was not intended to be a qualified plan. See Rev. Rul. 2014-9, 2014-47 I.R.B. 975.

53Most plan sponsors we interviewed said they allowed loans. Plan administrators we spoke with indicated that between 66 and 95 percent of their plan sponsor clients allowed loans.

54Aon Hewitt, a retirement consulting firm, reported a loan default was most likely to occur when a participant had an outstanding loan at job separation, which resulted in a 69 percent default rate compared with 3 percent for actively employed loan recipients. Aon Hewitt, “Minimizing Defined Contribution Plan Leakage” (Oct. 2013). Lu, Mitchell, Utkus, and Young, Borrowing From the Future (2017) found that 86 percent of workers who changed jobs with an outstanding loan balance defaulted on their loans.

55The December 2017 tax law extended the period in which a participant can roll over a qualified plan loan offset amount until the due date, including extensions, for their income tax return for the year in which the amount is treated as distributed from the plan. See Pub. L. No. 115-97, § 13613, 131 Stat. 2166 (2017) (codified at 26 U.S.C. § 402(c)(3)).

56Industry sources reported that more than 90 percent of plans allowed a maximum of two loans. Vanguard reported that 91 percent of plans limited loans to two or fewer. Vanguard, “How America Saves 2018” (June 2018). Plan Sponsor reported that 91 percent of plans limit loans to two or fewer. Plan Sponsor, “2017 DC Survey: Plan Benchmarking” (Dec. 4, 2017).

57Aon Hewitt has published research on loans finding in 2015 that one-quarter of all participants had a loan outstanding with the principal balance of 20 percent of the account on average.” Aon Hewitt found that 44 percent of participants with outstanding loans had multiple loans when more than one loan was allowed. See Aon Hewitt, 2016 Universe Benchmarks: Employee Savings and Investing Behavior in Defined Contribution Plans (2016). Additional research has found that workers borrow more when a plan permits multiple loans. See Lu, Mitchell, Utkus, and Young, (2017).

58The Bipartisan Budget Act of 2018 directs IRS to remove a provision in the regulations that provided that after taking a hardship withdrawal an employee was prohibited from contributing to the plan for 6 months. In addition, it also amended the IRC to state that a distribution shall not otherwise fail to be treated as a hardship distribution solely because the employee does not take any available loan under the plan. See Pub. L. No. 115-123, § 41113. IRS recently published proposed amendments to the regulations relating to hardship distributions from 401(k) plans to correspond with the changes in statute. 83 FR 56763 (Nov. 14, 2018) However, it remains to be seen whether plan sponsors will amend plan rules to remove the contribution suspension and loan exhaustion provisions.

59Many plan sponsors allow hardship withdrawals. Four plan administrators provided data on their plan sponsor clients showing the majority of their clients (70 percent to 95 percent) allow hardship withdrawals. However, while hardship withdrawals are generally available to retirement plan participants, only a small percentage make hardship withdrawals. According to one plan administrator, about 2.3 percent of participants make a hardship withdrawal annually, and this is consistent year to year. Another plan administrator provided data demonstrating that 81 percent of their plan sponsor clients use the safe harbor guidelines to determine participant eligibility for a hardship withdrawal. These data included 333 clients with a total of 10 million employees.

60For example, one plan administrator provided data demonstrating that, in 2017, 35 percent of hardship withdrawals were made to prevent eviction or foreclosure, 30 percent were for medical expenses, and 14 percent were for educational expenses. Another plan administrator stated that 40 percent of hardship withdrawals are for medical expenses while 30 percent are to prevent eviction or foreclosure. In addition, Aon Hewitt reported that 50 percent of hardship withdrawals in 2010 were made to prevent eviction or foreclosure while 13 percent were made to address medical costs. See Aon Hewitt, “Leakage of Participants' DC Assets: How Loans, Withdrawals, and Cashouts Are Eroding Retirement Income” (2011).

61See TIAA-CREF Financial Services, “Are Your Employees Borrowing From Their Futures?” (New York, NY: 2014).

62One study on early withdrawals found that job terminations accounted for 22 percent of total retirement savings dollars leaving the retirement system. See Barbara A. Butrica, Sheila R. Zedlewski, and Philip Issa, “Understanding Early Withdrawals from Retirement Accounts” The Urban Institute, Discussion Paper Series 10-02 (May 2010).

63Households that incurred additional 10 percent tax as a result of early distributions more likely to have low levels of nonretirement savings and have experienced an economic shock. See Gene Amromin and Paul Smith, “What Explains Early Withdrawals from Retirement Accounts? Evidence from a Panel of Taxpayers,” National Tax Journal, vol. LVI, no. 3 (Sept. 2003). This study also found that more than half of employees believe it is likely they will need to use their retirement savings for other expenses. Additional research found nearly a quarter of households would use funds in their retirement savings account in a financial emergency. See The Pew Charitable Trusts, “What Resources Do Families Have for Financial Emergencies?” (Nov. 2015).

64See IRS, “IRA FAQs — Distributions (Withdrawals)” (May 30, 2018).

65IRS instructions for Form 1099-R require IRA providers to report early withdrawals as “early distribution, no known exception” even when the distribution is made for a generally excepted reason, such as certain medical expenses, certain health insurance premiums, qualified higher education expenses, a qualified “first-time” home purchase, or a qualified reservist distribution under 26 U.S.C. § 72(t)(2)(B), (D), (E), (F), or (G). IRA providers can only categorize the early withdrawal as exempt from additional 10 percent tax for early distributions under limited circumstances.

66IRA owners who receive an early distribution must file income taxes using Form 1040 and, if claiming an exception from the additional 10 percent tax for early distributions, Form 5329. Instructions to Form 5329 require taxpayers who meet an exception not indicated on Form 1099-R to enter a code detailing the exception they are claiming. For example, the Form 5329 includes codes for IRA distributions used to pay certain health insurance premiums while unemployed, qualified higher education expenses, or qualified “first-time” home purchases, up to $10,000.

67Some research has found households' early withdrawals to respond to financial emergencies include both 401(k) plans and IRAs. The Pew Charitable Trust's Survey of American Family Finances found that about 13 percent of people with retirement accounts including IRAs and 401(k) plans said they had drawn on these savings in the previous year, during which they had experienced a financial shock such as unemployment, a pay cut, or a marital change such as divorce, separation, or the death of a spouse. Further, Pew found households that had experienced more financial shocks were more likely to withdraw from their retirement account. See Pew Charitable Trusts, “Financial Shocks Put Retirement Security at Risk” (Oct. 2017). Other researchers have also found that income loss and changes in marital status are important causes of retirement account withdrawals. See Robert, Argento, Victoria L. Bryant, and John Sabelhaus, “Early Withdrawals from Retirement Accounts During the Great Recession” Contemporary Economic Policy, vol. 33, issue 1 (Jan. 2015).

68For this report, we contacted a nongeneralizable sample of companies. Of the company officials we interviewed, some had implemented specific policies intended to reduce early withdrawals of retirement savings, while others had not. Some companies that had not implemented certain policies told us that they would do so only after receiving federal guidance on which policies would be permissible. GAO is not endorsing or recommending any strategy in this report, and has not evaluated these strategies for their behavioral or other effects on retirement savings or on tax revenues. (See appendix I for a detailed description of the method we used to select companies to contact.)

69In 2018, Callan, an investment consulting firm, reported that 50.5 percent of plan sponsors who participated in a Callan-conducted survey offered installment cashout options, with an additional 14.6 percent of plan sponsors surveyed noting that their company will take steps to offer installment payments in 2018. See Callan, “2018 Defined Contribution Trends.”

70Similarly, Callan reported that 56.3 percent of plan sponsors who participated in a Callan-conducted survey offered partial distributions, with an additional 16.7 percent of plan sponsors surveyed noting that their company will take steps to offer partial distributions in 2018. See Callan (2018).

71In 2017, Alight, a benefits and financial services provider, reported that 65 percent of the employers who participated in their survey offered some form of continued loan payments after the participant had separated from the employer. This number had increased from 44 percent in 2013, and 54 percent in 2015. Fifty-one percent of employers who participated in the survey and did not offer post-separation loan repayment programs indicated that they did not do so because of administrative cost and complexity. Alight, “Trends & Experience in Defined Contribution Plans” (2017). Another survey of employers, conducted by Callan in 2017, found that 38.8 percent of employers surveyed allowed terminated or retired participants to continue paying off loans, and an additional 12.5 percent planned on making this service available in 2018. See Callan (2018).

72One plan administrator told us they encouraged plan sponsors to adopt 401(k) loan repayment programs for separating participants. However, another plan administrator advised its small plan provider clients not to adopt post-separation 401(k) loan repayment programs on the basis that managing such a program could be complex for small plan providers to manage effectively.

73For example, the TSP, the DC plan for federal civil service employees and certain members of the uniformed services, also limits plan loans to employee contributions and any earnings generated from those contributions. A participant who wants to apply for a TSP loan is not able to borrow from agency contributions or any earnings from those contributions.

74In guidance distributed to plan sponsors on 401(k) plan loans TIAA, a financial services company, suggested limiting loans to participant contributions only, rather than allowing participants to take loans from employer contributions as well. TIAA-CREF Financial Services (2014).

75Stakeholders we spoke with suggested that a waiting period may be an effective strategy for reducing participant loans. In 2011, Aon Hewitt recommended that plan sponsors consider requiring a waiting period between participant loan repayment and the initiation of an additional loan. See Aon Hewitt, “Leakage of Participants' DC Assets: How Loans, Withdrawals, and Cashouts Are Eroding Retirement Income” (2011).

76One plan sponsor noted that if a current employee defaults on paying their 401(k) loan the plan sponsor will apply a 6-month waiting period before that employee is eligible to apply for another plan loan. While not current policy, the plan sponsor noted that a 3-month waiting period between plan loans is an option worth further study. One plan administrator said that while waiting periods were not as common as she would prefer, she encouraged plan sponsors to adopt this policy.

77Some plan administrators recommended the number of plan loans should be limited. One plan administrator encouraged plan sponsors to minimize the number of loans offered. Vanguard reported that in 2017 56 percent of plan sponsors that contracted with it as a plan administrator allowed one plan loan, 35 percent allowed two loans, and 9 percent allowed three or more loans. See Vanguard (2018).

78As noted earlier, the IRS safe harbor includes the following six reasons for distributions deemed to be due to an immediate and heavy financial need: (1) costs directly related to the purchase of a principal residence; (2) certain medical expenses; (3) payments of tuition and other related educational expenses; (4) payments necessary to prevent eviction from, or foreclosure on a principal residence; (5) certain burial or funeral expenses; and (6) certain expenses for the repair of damage to the employee's principal residence. See 26 C.F.R. § 1.401(k)-1(d)(3)(iii)(B).

79According to the plan sponsor instituting the supplemental list led to an overall decline in the number of hardship withdrawal applications the plan sponsor had to review. The plan sponsor cited the clarity of the list as a reason for the overall decline in applications.

80For a more detailed discussion of Social Security, see GAO, Social Security's Future: Answers to Key Questions, GAO-16-75SP (Washington, D.C.: Oct. 27, 2015).

81For more information, see GAO, Unemployed Older Workers: Many Experience Challenges Regaining Employment and Face Reduced Retirement Security, GAO-12-445 (Washington, D.C.: Apr. 25, 2012).

82For more information on phased retirement see GAO, Older Workers: Phased Retirement Programs, Although Uncommon, Provide Flexibility for Workers and Employers, GAO-17-536 (Washington, D.C.: July 20, 2017).

83An early withdrawal from an IRA account is subject to income taxes on tax-deferred amounts and, possibly, an additional 10 percent tax for early distributions. In addition, an early withdrawal from an IRA leads to the permanent removal of money from an individual's account. An IRA loan is currently not permitted. According to the IRS, were an IRA loan attempted, it could result in the IRA ceasing to be an IRA. See 26 U.S.C. § 408(e)(2).

84A recent study found that tasking a private third-party to facilitate automatic rollovers would need buy-in from major financial institutions that manage 401(k) plans, as well as government regulators. See Munnell, Belbase, and Sanzenbacher, An Analysis of Retirement Models (2018).

86In November 2018, DOL issued a notice of proposed exemption from certain ERISA prohibited transaction restrictions of ERISA and/or the IRC to a financial technology services organization that works with plan sponsors and plan administrators to streamline retirement account consolidation processes. The exemption would allow the company to apply its technology to help employees who may have multiple job changes over their careers consolidate small accounts held in prior employers' individual account plans and rollover IRAs into their new employers' individual accounts or 401(k) plans. DOL, Notice of Proposed Exemption Involving Retirement Clearinghouse, LLC (RCH or the Applicant) — Located in Charlotte, North Carolina, 83 Fed. Reg. 55,741(Nov. 7, 2018).

88The IRS safe harbor deems distributions for costs directly related to the purchase of a principal residence and tuition and other related educational expenses, among other things, to be on account of an immediate and heavy financial need for purposes of hardship withdrawal from 401(k) plans.

89In addition to student loans, those planning for postsecondary education costs can save through 529 plans. Authorized by section 529 of the IRC, a 529 plan is a tax-advantaged savings plan designed to encourage saving for future education costs. 529 plans are operated by states or educational institutions.

90One stakeholder and one plan sponsor we interviewed said they would consider implementing an emergency savings account program if the federal government clarified how such a program could work. Several papers have studied the emergency savings account proposal. See John Beshears, James J. Choi, J. Mark Iwry, David C. John, David Laibson, and Brigitte C. Madrian, “Building Emergency Savings through Employer-Sponsored Rainy Day Savings Accounts” (Oct. 2017). See also, Joanna Ain, J. Mark Iwry, and David Newville, “Saving for Now & Saving for Later: Rainy Day Savings Accounts to Boost Low-Wage Workers' Financial Security” Prosperity Now (June 2018). See also, David S. Mitchell, and Gracie Lynne, “Driving Retirement Innovation: Can Sidecar Accounts Meet Consumers' Short- and Long-Term Financial Needs?” (Washington, D.C: The Aspen Institute's Financial Security Program, June 27, 2017).

91One plan administrator noted that a workplace-based emergency savings account may be a good option for employees new to the workforce, who may have little in the form of personal savings.

92A 2017 paper examined various ways of designing an emergency savings account, including using after-tax employee contributions as part of a 401(k) plan, a deemed Roth IRA associated with a 401(k) plan, or a separate rainy day savings account at a bank or credit union. The authors suggested that each approach should be further explored through pilot programs and additional experimentation. See Beshears, Choi, Iwry, John, Laibson, and Madrian, Building Emergency Savings (2017). Another paper outlined several forms that emergency savings accounts could take. The paper presented the potential benefits and drawbacks of several design choices that policymakers would need to consider before implementing an emergency savings account vehicle. See, Mitchell and Lynne, Driving Innovation (2017).

93Beshears et al. (2017) discuss the merits of automatic enrollment of employees into emergency savings accounts. The authors found that an emergency savings account with automatic features may be a good way to help individuals accumulate savings in advance of a possible economic shock.

94In a configuration in which emergency savings accounts were funded with after-tax participant contributions any investment earnings could be taxable upon withdrawal, as well as subject to an additional 10 percent tax for early distributions. See Ain, Iwry, and Newville, Saving for Now (2018).

1According to the Census Bureau, the 2014 SIPP removed certain topics from the SIPP survey in order to reduce overall respondent burden since it is not necessary to ask them each wave. However, SSA still needed the data from these topics to analyze the economic and social situation of people with disabilities and people in or approaching retirement. As a result, Census conducted the SSA Supplement on behalf of SSA as a separate survey from the SIPP, even though the sample consisted of households that completed 2014 Wave 1 SIPP interviews. SSA selected topics for the SSA Supplement, which included personal retirement account contributions and withdrawals, from previous SIPP panel topical modules.

2We used replicate weights provided by the Census Bureau and a Fay adjustment of 0.5 for variance estimation purposes. We applied the SIPP person weight to make all point estimates, since our estimates of interest were at the person level.

3We selected the 2013 plan year to allow for consistent reporting across our data sources and to present a comprehensive snapshot of early withdrawals in a single year.

END FOOTNOTES

DOCUMENT ATTRIBUTES
Copy RID