United States Government Accountability Office GAO Report to the Chairman, Committee on Finance, U.S. Senate HIGHER EDUCATION December 2012 A Small Percentage of Families Save in 529 Plans GAO-13-64 December 2012 HIGHER EDUCATION A Small Percentage of Families Save in 529 Plans Highlights of GAO-13-64, a report to the Chairman, Committee on Finance, U.S. Senate Why GAO Did This Study What GAO Found Paying for college is becoming more A small percentage of U.S. families saved in 529 plans in 2010, and those who challenging, partly because of rising did tended to be wealthier than others. According to the Survey of Consumer tuition rates. A college savings plan Finances (SCF), less than 3 percent of families saved in a 529 plan or Coverdell can be an option to help meet these Education Savings Account (Coverdell)—a similar but less often used college costs. To encourage families to save savings vehicle also included in the SCF. While the economic downturn may for college, earnings from 529 plans— have reduced income available for education savings, even among those families named after section 529 of the Internal who considered saving for education a priority, fewer than 1 in 10 had a 529 plan Revenue Code—grow tax-deferred (or Coverdell). Families with these accounts had about 25 times the median and are exempt from federal income financial assets of those without. They also had about 3 times the median income tax when they are used for qualified and the percentage who had college degrees was about twice as high as for higher education expenses. In fiscal year 2011, the Department of the families without 529 plans (or Coverdells). Treasury estimated these plans States offer consumers a variety of 529 plan features that, along with several represented $1.6 billion in forgone other factors, can affect participation. Some of the most important features federal revenue. Managed by states, families consider when choosing a 529 plan are tax benefits, fees, and over one hundred 529 plan options investment options, according to experts and state officials GAO interviewed. were available to families nationwide These features can vary across the state plans. For example, in July 2012, total as of July 2012. The number of 529 annual asset-based fees ranged from 0 to 2.78 percent depending on the type of plan accounts and the amount invested plan. 529 plan officials and experts GAO interviewed said participation is also in them has grown during the past affected by families’ ability to save, their awareness of 529 plans as a savings decade. GAO was asked to describe (1) the percentage and characteristics option, and the difficulty in choosing a plan given the amount of variation of families enrolling in 529 plans, (2) between plans (see fig. 1). Selected states, however, have taken steps to plan features and other factors that address these barriers. For example, to address families’ ability to save, affect participation in 529 plans, and particularly for low-income families, some states have adopted plans that include (3) the extent to which savings in 529 less risky investments, have low minimum contributions, and match families’ plans affect financial aid awards. GAO contributions. analyzed government data, including Factors that Affect 529 Plan Participation the SCF. This survey’s 529 plan data are combined with Coverdells, so the SCF estimates used in the report include both 529 and Coverdell data. GAO also analyzed National Postsecondary Student Aid Study data; conducted interviews with federal and state officials, industry and academic experts, and state and institutional higher education officials; reviewed 529 plan and Department of Education documents; conducted a literature review; and reviewed relevant federal Savings in 529 plans affect financial aid similarly to a family’s other assets. For laws, regulations, and guidance. federal aid, a family’s assets affect how much it is expected to contribute to the cost of college. If the amount of those assets exceeds a certain threshold, then a What GAO Recommends percentage is expected to be used for college costs. For example, for students GAO is not making any who are dependent on their parents, the percentage of parental assets, including recommendations in this report. savings in 529 plans, that the family may be expected to contribute ranges from 2.64 to 5.64 percent. Many states and selected institutions also treat 529 plan View GAO-13-64. For more information, savings the same as other family assets. However, a few states provide them contact Michelle Sager, (202) 512-6806, email@example.com with special treatment, such as exempting those funds from their financial aid calculation. United States Government Accountability Office Contents Letter 1 Background 5 Few Families Have 529 Plans and Those Who Do Tend to Be Wealthier 14 States’ 529 Plan Features and Other Factors Can Affect Participation 21 Savings in 529 Plans Affect Financial Aid the Same as Other Assets 34 Concluding Observations 41 Agency Comments 42 Appendix I Objectives, Scope, and Methodology 43 Appendix II Free Application for Federal Student Aid, 2012-2013 52 Appendix III GAO Contacts and Staff Acknowledgments 62 Tables Table 1: Estimates of Selected Characteristics of College Students’ Families (from NPSAS) and Families with and without 529 Plans (from SCF) 20 Table 2: Fee Ranges (Percent) for College Savings Plans, Direct- Sold and Advisor-Sold 24 Figures Figure 1: 529 Plan Timeline and Assets Invested 7 Figure 2: 529 Plan Account Growth 8 Figure 3: FAFSA Questions on Student and Parent Assets, Including 529 Plans 13 Figure 4: Distribution of Income for Families with and without 529 Plans or Coverdells (from SCF) 17 Figure 5: Factors that Affect 529 Plan Participation 26 Figure 6: Expected Family Contribution from Parents’ Assets for Dependent Student 36 Figure 7: FAFSA Questions on Student Untaxed Income 38 Page i GAO-13-64 529 Plans Abbreviations Coverdells Coverdell Education Savings Accounts CSPN College Savings Plan Network Education Department of Education EFC expected family contribution EGTRRA Economic Growth and Tax Relief Reconciliation Act of 2001 FAFSA Free Application for Federal Student Aid FDIC Federal Deposit Insurance Corporation Federal Reserve Board of Governors of the Federal Reserve System IRS Internal Revenue Service MSRB Municipal Securities Rulemaking Board NBER National Bureau of Economic Research NPSAS National Postsecondary Student Aid Study SEC Securities and Exchange Commission SCF Survey of Consumer Finances SOI Statistics of Income Treasury Department of the Treasury 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. Page ii GAO-13-64 529 Plans United States Government Accountability Office Washington, DC 20548 December 12, 2012 The Honorable Max Baucus Chairman Committee on Finance United States Senate Dear Mr. Chairman: While median family income decreased between 2005 and 2011, college tuition and fees increased at an average annual rate of 6 percent, more than double the rate of inflation. As families look for ways to prepare for these rising costs, their options include saving in advance; paying for expenses at the time they occur out-of-pocket and/or through grants and scholarships; or financing the cost through educational loans that they repay at a later date. Loans have become an increasingly popular option to cover these costs. 1 However, with college loan balances now estimated at $914 billion, more than total credit card debt, 2 financing a large portion of college costs potentially creates a future financial burden for students and their families. To encourage people to save for college expenses in advance, Congress, in 1996, added section 529 to the Internal Revenue Code, which authorized state qualified tuition programs, otherwise known as 529 plans. 3 These plans are available in two forms— prepaid tuition plans and college savings plans. In prepaid tuition plans, participants buy tuition credits or certificates on behalf of a designated 1 The College Board reported that the relative stability of the ratio of grant to loan amounts over time indicates that loans have not replaced grants. However, grant aid often fails to increase rapidly enough to fill the growing gap between the costs of attending college and the ability of students and families to pay those costs. College Board Advocacy and Policy Center, Trends in Student Aid 2011 (New York, NY: 2011). 2 Federal Reserve Bank of New York, Research and Statistics Group, Microeconomic Studies. Quarterly Report on Household Debt and Credit. (New York, NY: August 2012). Debt amount is as of second quarter 2012. 3 Small Business Job Protection Act of 1996, Pub. L. No. 104 -188 § 1806, 110 Stat. 1755, 1895, codified at 26 U.S.C. § 529. Page 1 GAO-13-64 529 Plans beneficiary, who can use them for qualified higher education expenses. 4 College savings plans allow for contributions to an account for the purpose of paying a designated beneficiary’s qualified higher education expenses. As of July 2012, more than one hundred 529 plan options have been developed and are managed by states and the District of Columbia. In 2011, there were more than 11 million accounts with assets of $167 billion (in 2012 dollars). 5 While there is no federal tax deduction for payments or contributions to 529 plans, which are made with after-tax dollars, plan earnings accumulate tax-deferred and earnings included in withdrawals used to pay for qualified higher education expenses are not taxed. This tax expenditure cost the federal government an estimated $1.6 billion in forgone revenue in fiscal year 2011. 6 We reported, in 2012, that higher income households are more likely to have 529 plans and tend to benefit more from these tax preferences because of their higher marginal tax 4 Qualified higher education expenses include tuition, books, supplies, and equipment. Costs for room and board are considered qualified education expenses if students are enrolled at least half-time. Qualified higher education expenses also included expenses paid or incurred in 2009 and 2010 for computer technology, equipment, or Internet access and related services. Eligible educational institutions include colleges, universities, vocational schools, or other postsecondary institutions eligible to participate in a federal student aid program authorized by Title IV of the Higher Education Act of 1965, as amended. This includes virtually all accredited, public, nonprofit, and for-profit postsecondary institutions in the U.S. 5 Data provided by the Financial Research Corporation. Account numbers and assets are as of year-end 2011. 6 Tax expenditures are reductions in a federal taxpayer’s tax liability that result from special credits, deductions, exemptions and exclusions, deferrals of tax liability, and preferential tax rates. The federal revenue forgone may be viewed as spending channeled through the tax code. This estimate by the Department of the Treasury (Treasury) includes both college savings plans and prepaid plans. For fiscal year 2011, the Joint Committee on Taxation reported revenue loss estimates of $500 million for college savings plans and less than $50 million for prepaid plans. Revenue loss estimates do not incorporate any behavioral responses and, thus, do not necessarily represent the exact amount of revenue that would be gained if a specific tax expenditure was repealed. Page 2 GAO-13-64 529 Plans rate. 7 We have recommended in the past that tax expenditures be periodically reviewed to determine their effectiveness in achieving specific policy goals, particularly given the nation’s long-term fiscal imbalance. 8 In response to your request to examine the extent to which 529 plans help families of all income levels save for college and how these plans affect student financial aid, we reviewed (1) the percentage and characteristics of families enrolling in 529 plans, (2) the plan features and other factors that affect participation in 529 plans, and (3) the extent to which savings in 529 plans affect financial aid awards. To answer these research objectives, we analyzed government and industry data; conducted interviews with federal and state officials, industry representatives and academic experts, as well as state and institutional higher education officials; reviewed 529 plan and Department of Education (Education) documents; conducted a literature review; and reviewed relevant federal laws, regulations, and guidance. We assessed the reliability of the data we used by reviewing documentation, interviewing knowledgeable officials, and conducting electronic testing on relevant data fields. We found the data we reviewed reliable for the purposes of our analyses. To determine the percentage and characteristics of families enrolling in 529 plans, we reviewed data from the 2010 Survey of Consumer Finances (SCF); the 2007-2008 National Postsecondary Student Aid Study (NPSAS), which was the most recent available and which we adjusted to 2010 dollars; and 2007-2010 Statistics of Income (SOI) federal tax data. Our analysis of data from SCF, NPSAS, and SOI are 7 GAO, Higher Education: Improved Tax Information Could Help Families Pay for College, GAO-12-560 (Washington, D.C.: May 18, 2012). The analysis used in this report also included Coverdell Education Savings Accounts (Coverdells). Similar to 529 plans, Coverdell accounts allow families to save for education expenses. Account earnings accumulate tax-deferred and earnings included in withdrawals used to pay for qualified education expenses are not subject to federal tax. Unlike 529 plans, (1) Coverdell contributors must generally have a modified adjusted gross income of less than $110,000 per year ($220,000 in the case of a joint return), (2) Coverdells have annual contribution limits of $2,000 that must generally stop when a beneficiary reaches 18 years of age, and (3) Coverdells can be used for qualified elementary, secondary, or postsecondary expenses typically for individuals under age 30. 8 GAO, Government Performance and Accountability: Tax Expenditures Represent a Substantial Federal Commitment and Need to Be Reexamined, GAO-05-690 (Washington, D.C.: Sept. 23, 2005). Page 3 GAO-13-64 529 Plans subject to sampling errors because these data sets are based on samples. Unless otherwise noted, all percentage estimates based on SCF, NPSAS, and SOI have 95 percent confidence intervals that are within 5 percentage points of the estimate itself, and all numerical estimates other than percentages have 95 percent confidence intervals that are within 5 percent of the estimate itself. Using SCF data, we generated estimates of the characteristics of households—including wealth (financial assets), income, education, and race/ethnicity—enrolled in 529 plans or Coverdell Education Savings Accounts (Coverdells) 9 and compared those to characteristics of families not enrolled in these plans. Our estimates are based on analyses from the restricted SCF dataset and were conducted with assistance from officials at the Board of Governors of the Federal Reserve System (Federal Reserve). We also developed a profile of the general college population using NPSAS data. In addition, we analyzed taxpayer data from SOI and used a microsimulation model from the National Bureau of Economic Research to understand the extent to which taxpayers use withdrawals (known as distributions) from 529 plans for qualified educational expenses and how the tax savings from these plans are distributed across income levels. To understand the features and factors that affect participation, we interviewed officials from five state 529 plans (Louisiana, Michigan, Pennsylvania, Utah, and Virginia), conducted a literature review, and spoke with academic researchers, industry regulators, financial services companies, financial experts, and consumer interest groups. While we selected the five state plans to represent variety in the types of plans offered, benefits, and geographic location, our findings cannot be generalized to all 529 plans. We also analyzed College Savings Plan Network (CSPN) data to provide a national overview of plan features and benefits. 10 To provide information on the percentage of students who receive financial aid and the median award amount, we analyzed NPSAS data. To assess efforts by states to disclose 529 plan information to consumers, we reviewed disclosure documents from selected states as well as CSPN disclosure principles. 9 The SCF codes responses for 529 plans and Coverdells together. Survey officials said respondents did not always distinguish between the two account types; therefore, we were unable to separate these responses because of data reliability concerns. 10 CSPN is an affiliate to the National Association of State Treasurers and serves as a clearinghouse for information on 529 plans. Plan data are as of July 2012. Page 4 GAO-13-64 529 Plans We assessed the extent to which savings in 529 plans affect federal financial aid awards by reviewing relevant federal laws, regulations, and guidance, and interviewing Education officials. In addition, we analyzed NPSAS data to determine the extent to which a family’s total assets affected the federal expected family contribution to college expenses. To understand how savings in 529 plans affect state and institutional financial aid, we interviewed six state financial aid offices and six institutions of higher education. We selected states that indicated on a National Association of State Student Grant and Aid Programs survey that they used a financial aid formula other than the federal methodology in their primary needs analysis and/or provided special treatment for state 529 plans. We selected an institution in each of the six states that was reported as using the College Board’s PROFILE or another institutional form, in addition to the Free Application for Federal Student Aid, to gather information used in making decisions regarding institutional financial aid. 11 To obtain an institutional perspective beyond the six schools, we interviewed officials from national organizations representing community colleges, private nonprofit and public colleges and universities, and for- profit schools. A more detailed explanation of our methodology is available in appendix I. We conducted this performance audit from November 2011 to December 2012 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 History of 529 Plans 529 plans are a college savings vehicle that originated in the states. In 1986, Michigan created the Michigan Education Trust to operate what is 11 The College Board is a not-for-profit organization that provides college students with financial aid support and scholarships. It also conducts research and advocacy on behalf of students, educators, schools, and colleges. The PROFILE is a form developed by the College Board to help institutions gather information used to award nonfederal student aid funds. Page 5 GAO-13-64 529 Plans generally considered the first state prepaid tuition plan. In 1996, Congress enacted Section 529 of the Internal Revenue Code, setting out requirements that state 529 plans must meet to be exempt from federal tax. The Economic Growth and Tax Relief Reconciliation Act of 2001 included a provision making earnings included in distributions from 529 plan accounts entirely tax-exempt as long as they are used to pay for qualified higher education expenses. 12 For other key legislative actions and the value of assets invested in these plans, see fig. 1. 12 Pub. L. No. 107-16, § 402, 115 Stat. 38, 60. Previously, earnings had been allowed to grow on a tax-deferred basis, but were taxed upon distribution to pay for qualified education expenses. Page 6 GAO-13-64 529 Plans Figure 1: 529 Plan Timeline and Assets Invested Notes: a Michigan v. United States, 40 F.3d 817 (6th Cir. 1994). b We chose to present account data starting in 2001 because the federal tax-exempt status for distributions was granted in 2001 and data were not available for all previous years. c Pension Protection Act of 2006, Pub. L. No. 109-280, tit. XIII § 1304, 120 Stat. 789, 1109. The number of 529 plan accounts has also increased since the plans were granted expanded federal tax advantages in 2001 (see fig. 2). Page 7 GAO-13-64 529 Plans Figure 2: 529 Plan Account Growth a We did not report prepaid tuition account data for 2010 and 2011 because the method of collecting these data changed in 2009. Federal Requirements for 529 plans are a state-sponsored investment or savings vehicle whose 529 Plans purpose is to encourage people to save for college. 13 Contributions to 529 plans are made with after-tax dollars and are not deductible for federal tax purposes. Annual contributions in excess of $13,000 are generally subject to federal gift taxes. 14 Total contributions may not exceed the amount necessary to provide for the qualified education expenses of the 13 One additional prepaid plan is not run by a state. The Private College 529 Plan provides a prepaid tuition plan at more than 270 participating private colleges and universities. The Economic Growth and Tax Relief Reconciliation Act of 2001 amended section 529 to provide that one or more eligible educational institutions could establish and maintain a prepaid tuition plan. Prior to that amendment, qualified tuition programs were defined as those established and maintained by a state or state agency or instrumentality. 14 Contributors may elect to treat contributions between $13,000 and $65,000 as having been made over a 5 calendar-year period without being subject to federal gift taxes. Page 8 GAO-13-64 529 Plans beneficiary, which is determined by each state; however, individuals may open 529 plan accounts in multiple states. 15 Earnings on contributions grow tax-deferred. When a distribution is made from a 529 plan, the earnings portion is tax-exempt as long as it is used to pay for qualified education expenses. 16 Taxpayers must report to the Internal Revenue Service (IRS) whether the distribution was for qualified higher education expenses. Distributions not used for qualified higher education expenses can be made to either the account owner or beneficiary, but the portion of the nonqualified distribution consisting of investment earnings is taxable and subject to an additional 10 percent penalty. 17 The federal penalty does not apply in some circumstances, for example if the distribution was considered nonqualified because the beneficiary died or received a scholarship. While section 529 provides that account owners and beneficiaries may not directly or indirectly control how contributions or earnings are invested, in 2001, IRS issued a notice setting out a rule permitting a change in investment strategy once per year and upon a change in the designated beneficiary of the account. 18 For 2009 only, this was increased to twice per year. 19 There are few federal restrictions on 529 plan participation. For example, there are no income limits and almost anyone can initially be named as a beneficiary—an individual may open a 529 plan account for a child, grandchild, friend, spouse, or for themselves. Further, the 529 plan account owner may change the beneficiary at any time, though the 15 According to CSPN data, overall contribution limits ranged from $235,000 to $400,000 as of July 2012. However, there is no limit on the number of accounts for which an individual can be named the beneficiary. According to a 2009 Treasury report, An Analysis of Section 529 College Savings and Prepaid Tuition Plans, a beneficiary can have accounts in as many as 44 states, with effectively no limit on the overall 529 account balance. 16 An American Opportunity or Lifetime Learning Tax Credit can be claimed in the same year the beneficiary takes a tax-exempt distribution from a 529 plan as long as the same expenses are not used for both benefits. 17 Qualified education expenses must be reduced by any tax-free educational assistance, such as the tax-free part of scholarships and fellowships, veterans’ education assistance, Pell grants, and employer-provided education assistance. 18 IRS Notice 2001-55. 19 IRS Notice 2009-1. Page 9 GAO-13-64 529 Plans subsequent beneficiary must be a member of the family of the original beneficiary in order for this change to be tax-exempt. 20 529 Plan Operation and Because 529 plans are state-sponsored investments, states determine Features whether and what type of plans to offer (i.e., prepaid tuition or savings) as well as the eligibility criteria (for example, at the time of application prepaid tuition plans may require either the account owner or beneficiary to be a resident of the state administering the plan whereas residents and nonresidents can invest in most states’ college savings plans); administrative and investment fees; and associated state tax benefits. Almost all states offer a college savings plan. 21 In these plans, individuals purchase interests or shares in a trust established by the state. In most cases, the trust assets are invested in mutual funds. The shares in college savings plans can be sold directly by the state or through an external program manager hired by the state (direct-sold) as well as through a financial advisor or broker (advisor-sold). College savings plans may offer a number of investment options, which often include stock mutual funds, bond mutual funds, and money market funds. These investment options can vary in terms of risk and return, ranging, for example, from investments that are insured by the Federal Deposit Insurance Corporation (FDIC) to options that are almost completely invested in aggressive-growth funds. Many plans offer age-based portfolios that shift automatically into more conservative investments as the beneficiary approaches college age. Fifteen states also offer prepaid tuition plans to their state residents. 22 To help run their plans, states may employ marketing staff, advisors, financial consultants, or other experts. 20 A member of the beneficiary’s family is defined as including a father, mother, brother, sister, child, grandchild, niece/nephew, son/daughter-in-law, spouse of any individual listed above, and a first cousin, among others. 21 Wyoming no longer operates its own plan and has entered into an agreement with Colorado’s college savings plan whereby Wyoming residents participate under the same terms as Colorado residents. Washington only offers a prepaid tuition plan. 22 As of July 2012, of the 15 states with prepaid tuition plans, 3 are no longer accepting new enrollees. Kentucky and South Carolina’s prepaid tuition plans are currently closed to new enrollments. Tennessee also closed its prepaid tuition plan to new enrollees and contributions in 2010 due to the cost of tuition increasing faster than investment returns. However, it launched a new college savings plan in September 2012. Page 10 GAO-13-64 529 Plans Some states offer a variety of tax advantages that can include a state deduction or non-refundable credit for plan contributions and tax-deferred earnings. 23 These benefits may apply only to residents who make contributions to their own state’s plan or, in a few states, may include contributions made to other states’ plans. 24 Plan Disclosures Although most 529 college savings plans have been modeled after mutual funds, 529 plans are regulated differently than mutual funds under the federal securities laws because they are regulated as municipal securities. 25 As municipal securities, 529 plans are exempt from the registration and reporting requirements of the federal securities laws. 26 However, broker-dealers selling 529 plans (advisor-sold plans) must comply with the rules of the Municipal Securities Rulemaking Board (MSRB). 27 Specifically, MSRB requires broker-dealers who sell 529 plans to follow certain guidelines, such as having reasonable grounds to believe that the recommended product is suitable for the customer; disclosing certain information, such as plan fees and state tax implications; following certain requirements when advertising; and posting disclosure documents on its Electronic Municipal Market Access Website. 28 However, MSRB rules do not apply to state issuers when they market their 529 plans 23 Tax credits reduce tax filers’ income tax liability on a dollar-for-dollar basis. A nonrefundable credit can reduce the amount of taxes owed (tax liability) to zero but not below. A refundable credit can reduce the tax liability to zero and the remaining credit is paid to the taxpayer as a refund. Tax deductions permit tax filers to subtract the applicable contribution amount from income that would otherwise be taxable. Therefore, deductions reduce filers’ tax liability less than credits for any given amount claimed. 24 Arizona, Kansas, Maine, Missouri, and Pennsylvania extend the same state tax benefits to residents who invest in either their own state’s 529 plan or another state’s plan. 25 Municipal securities are issued by states and political subdivisions or agencies of states. Prepaid tuition plans are generally not considered municipal securities. 26 See Section 3(a)(2) of the Securities Act of 1933 and section 3(a)(29) of the Securities Exchange Act of 1934. Disclosures provided in connection with the sale of 529 plans are, however, subject to the antifraud provisions of the federal securities laws. 27 MSRB establishes standards of fair practice, disclosure, and suitability and professional qualifications for broker-dealers. 28 While MSRB has the authority, subject to Securities and Exchange Commission (SEC) approval, to establish such standards, it does not enforce its own rules. MSRB rules are enforced against broker-dealers by the Financial Industry Regulatory Authority, a self- regulatory organization that is subject to SEC oversight, and by SEC. Page 11 GAO-13-64 529 Plans directly to the investor without the assistance of a broker-dealer (direct- sold plans). In 2004, in response to concerns that 529 plan disclosures were inadequate, CSPN, after working with the Securities and Exchange Commission, the MSRB, and the National Association of Securities Dealers, developed voluntary disclosure principles to be adopted by state issuers on plan performance, fees, and state tax information, among other things. These principles were designed to enhance investors’ ability to compare information across plans. Since 2004, the principles have been updated several times with the most recent update in May 2011. Student Financial Aid As authorized under Title IV of the Higher Education Act of 1965, as amended, the Department of Education provides assistance to help millions of students and families meet the costs of higher education through grants, work-study, and loans. A substantial portion of this federal financial aid is awarded based on the amount of a student’s financial need, which is generally the difference between a student’s cost of attendance and an estimate of his family’s ability to pay these costs, known as the expected family contribution (EFC). In addition to the student’s income and assets, parents’ income and assets are also used to determine the student’s EFC unless the student is classified as independent. Independent students have their income and assets included in the EFC and their spouses’ income and assets, if applicable. Several criteria are used to determine if a student is independent, such as the student’s age, and if he or she is married or separated, enrolled in a master’s or doctoral degree program, or serving on active duty in the military, among other things. To apply for federal financial aid, students and, in the case of dependent students, parents submit information on income, assets, and the number of children enrolled in college through the Free Application for Federal Student Aid (FAFSA). This information is then used to determine the student’s eligibility for federal student aid by calculating the EFC through a process known as federal methodology, which is set out in statute. 29 In terms of assets, figure 3 shows the information required 30 by the FAFSA 29 The methodology for determining the federal EFC is found in Part F of Title IV of the Higher Education Act of 1965, as amended, codified at 20 U.S.C. §§ 1087kk – 1087vv. 30 Assets that are not required to be reported on the FAFSA include principal place of residence, a family farm, family-owned small businesses, retirement plans, and whole life insurance. Page 12 GAO-13-64 529 Plans regarding the net worth of students’ and parents’ investments, which includes savings in 529 plans along with other investments such as Coverdells, money market funds, stocks, and mutual funds (for a full copy of the FAFSA see app. II). 31 Figure 3: FAFSA Questions on Student and Parent Assets, Including 529 Plans States and institutions may also offer financial aid. To determine the amount of such aid, some states and institutions choose to gather information in addition to what is required by the FAFSA. One form used by some institutions is the College Board’s PROFILE form. The PROFILE 31 Prepaid plans are worth the refund value of the credits or certificates. Page 13 GAO-13-64 529 Plans asks for information not included on the FAFSA, such as home equity and medical expenses, as well as more detail about information that is included on the FAFSA. The institutions may then use an individualized institutional methodology to determine the student’s EFC for institutional financial aid. 32 Few Families Have 529 Plans and Those Who Do Tend to Be Wealthier A Small Percentage of According to the 2010 Survey of Consumer Finances (SCF), less than 3 Families Have 529 Plans percent of U.S. families had 529 plans 33 or Coverdells, a similar but less often used education savings account. 34 Even among families who acknowledged upcoming education expenses, 529 plans were not widely used. Of the approximately 25 percent of families who said they expected major education expenses in 5-10 years, about 7 percent of them had 529 plans or Coverdells. Similarly, of the approximately 18 percent of families who reported saving for education was a priority, only about 9 percent had 529 plans or Coverdells. 529 plans are also less commonly 32 Institutional methodology is a College Board formula developed by financial aid professionals, in consultation with economists, to measure a family’s ability to pay for college. A basic principle of institutional methodology is the idea that a family’s capacity to pay is a function of income and assets. 33 The SCF asks about survey respondents and members of their households (sometimes referred to as primary economic units), which we refer to as families. About 88 percent of families with 529 plans or Coverdells had children 25 years of age or younger living with them. Among the approximately 39 percent of all families that had children under 25 living with them, only about 6 percent had 529 plans or Coverdells. 34 The SCF combines responses for 529 plans and Coverdells. Because officials from the Federal Reserve, the federal agency that sponsors the SCF, said respondents did not necessarily distinguish between Coverdells and 529 plans, we did not separate the two account types. However, the officials indicated that a larger share of SCF respondents reported having 529 plans than Coverdells. Further, using SOI data, we estimate that in 2010 approximately 85 percent of tax filers who took a distribution from either a 529 plan or a Coverdell reported distributions from a 529 plan while 14 percent reported distributions from a Coverdell and 1 percent reported distributions from both. Page 14 GAO-13-64 529 Plans used than other savings vehicles among those saving for college. 35 For example, a 2010 Sallie Mae survey found that most parents saved for college in general savings accounts or certificates of deposit and, of those who did invest, more used general investment vehicles than 529 plans. 36 Based on our analysis of SCF data, the median amount in 529 plan or Coverdell accounts was about $14,700. 37 Families with 529 Plans Families with 529 plans or Coverdells typically had much more wealth Generally Have More than families without these accounts, according to our analysis of SCF Wealth and Education than data. Based on our analysis of the 2010 SCF, we estimate that the median financial asset 38 value for families with 529 plans or Coverdells Those without 529 Plans was about $413,500, which is about twenty-five times the median financial asset value for families without 529 plans or Coverdells (about $15,400). 39 For example, families with 529 plans or Coverdells had more retirement assets than other families. Of families with 529 plans or 35 There are other vehicles with tax benefits that can be used for education savings. Such vehicles include Coverdells, funds under the Uniform Gift to Minors Act and Uniform Transfer to Minors Act, individual retirement accounts, and savings bonds. 36 Information is reported for parents of children younger than 18 who believed their child is likely to attend college and who were saving for this purpose. The survey also found that nearly a quarter of these parents saved for college in retirement savings accounts such as a 401(k) or individual retirement account. Sallie Mae, How America Saves for College: Sallie Mae’s National Study of Parents with Children under 18 Conducted by Gallup (Reston, VA: 2010). 37 We are 95 percent confident that the median amount in these accounts was between $9,300 and $20,100. Because Coverdells had annual contribution limits of $2,000 in 2010, this estimate may understate the median amount in 529 plans. However, other data sources have similar findings specific to amounts saved in 529 plan accounts. For instance, in 2011 the College Savings Plan Network reported that the average amount in 529 savings plans was $15,492 in 2010. Unless otherwise noted, all SCF percentage estimates have 95 percent confidence intervals within 5 percentage points of the estimate itself. 38 Financial assets include, among other things, resources in 529 plans, checking and savings accounts, stocks, bonds, and retirement accounts. See appendix I for more detail. 39 We are 95 percent confident that the median financial asset value for families with 529 plans or Coverdells was between $247,400 and $579,600. For families without 529 plans or Coverdells, the 95 percent confidence interval is between $13,700 and $17,100. Page 15 GAO-13-64 529 Plans Coverdells, about 94 percent had retirement assets, 40 such as those in 401(k) accounts or traditional pensions. 41 In contrast, approximately 49 percent of families without 529 plans or Coverdells had these retirement assets. Further, the median value of retirement assets was much greater for those with 529 plans or Coverdells. Specifically, the median value in retirement accounts was about $213,600 for families with 529 plans or Coverdells, while the median value for families without 529 plans or Coverdells was about $40,300. 42 A larger share of families with 529 plans or Coverdells (27 percent) also believed they will have more than enough retirement income from pensions and Social Security to maintain current living standards than the share of families without 529 plans or Coverdells (16 percent), which may put them in a better position to save for college. 43 Further, the median income of families with 529 plans or Coverdells was about three times the median income of families without these accounts. 44 Specifically, families with 529 plans or Coverdells had median incomes of about $142,400 per year compared to $45,100 for other families. 45 Moreover, we estimate that 47 percent of families with 529 plans or Coverdells had incomes over $150,000, compared to 8 percent for 40 For our purposes, retirement assets include those in defined contribution plans (e.g. retirement savings plans such as a 401(k)), individual retirement accounts, and defined benefit plans from which the participant has the option to borrow or withdraw for both the SCF respondent and his/her spouse. 41 We use the term ‘traditional pensions’ to refer to defined benefit plans. 42 We are 95 percent confident that the amount of retirement assets for families with 529 plans or Coverdell accounts was between $149,700 and $277,500, while the amount for families without these accounts was between $37,400 and $43,200. 43 We are 95 percent confident that the percentage of families with 529 plans or Coverdell accounts who believed they will have more than enough retirement income was between 19.9 and 33.9 percent. For families without 529 plans or Coverdells, the 95 percent confidence interval was between 15.1 and 16.9 percent. 44 In 2012, we reported similar findings using data from the 2007 SCF. GAO-12-560. 45 Because Coverdell contributors must generally have had a modified adjusted gross income of less than $110,000 per year ($220,000 in the case of a joint return) in 2010, this estimate may understate the median income of families with 529 plans. We are 95 percent confident that the median income for families with 529 plans or Coverdells was between $125,400 and $159,400. Page 16 GAO-13-64 529 Plans families without these accounts (see fig. 4). 46 In 2009, Treasury reported that participation in 529 plans is likely to increase with income, in part, because the tax benefits and overall savings rates increase with income. 47 (For more information see text box on 529 plan savings, below). Figure 4: Distribution of Income for Families with and without 529 Plans or Coverdells (from SCF) Notes: Numbers may not add to 100 percent due to rounding. Error bars in the graph display the 95 percent confidence intervals for these estimates. We are 95 percent confident that the percentage of families with 529 plans or Coverdells who have incomes of $1 to $100,000 was between 21.9 and 37.1 percent, that those who have incomes of $100,001 to $150,000 was between 17.4 and 29.8 percent, and that those who have incomes over $150,000 was between 38.8 and 54.4 percent. We are 95 percent confident that the percentage of families without 529 plans or Coverdells who have incomes of $1 to $100,000 was between 81.5 and 83.3 percent, that those who have incomes of $100,001 to $150,000 was between 8.3 and 9.7 percent, and that those who have incomes over $150,000 was between 7.6 and 8.6 percent. 46 We are 95 percent confident that that the percentage of families with 529 plans or Coverdells who have incomes over $150,000 was between 38.8 and 54.4. For families without 529 plans or Coverdells, the 95 percent confidence interval was between 7.6 and 8.6 percent. 47 The Department of the Treasury, An Analysis of Section 529 College Savings and Prepaid Tuition Plans: A Report Prepared by the Department of Treasury for the White House Task Force on Middle Class Working Families (Washington, D.C.: September 9, 2009). Page 17 GAO-13-64 529 Plans 529 Plan Federal Tax Savings Using IRS taxpayer data and a microsimulation model from the National Bureau of Economic Research (NBER), we found that taxpayers with higher incomes both took larger distributions from 529 plans and received a larger tax benefit from those distributions (see table below). As we reported in 2012, 529 plan tax incentives are more beneficial to families with higher tax liabilities, in part, because these families have a higher marginal tax rate (see GAO-12-560). Distributions and Estimated Federal Tax Savings of 529 Plans in 2010 by Income Total income Median distribution Median tax savings $1 to $100,000 $7,491 $561 $100,001 to $150,000 $13,394 $1,958 Over $150,000 $18,039 $3,132 Source: GAO analysis of IRS SOI 2010 data using NBER’s TAXSIM Model, a microsimulation model. Notes: We are 95 percent confident that in 2010: the median distribution was between $6,261 and $8,280 and the median tax savings was between $493 and $664 for taxpayers whose total income was between $1 and $100,000; the median distribution was between $10,327 and $16,110 and the median tax savings was between $1,587 and $2,456 for taxpayers whose total income was between $100,001 and $150,000; and the median distribution was between $16,280 and $20,107 and the median tax savings was between $2,858 and $3,415 for taxpayers whose total income was more than $150,000. As higher-income families tend to have higher levels of education, it is logical that the wealthier families who save in 529 plans or Coverdells also have higher educational attainment than other families. Specifically, about 91 percent of SCF respondents with 529 plans or Coverdells indicated that either they or their spouse/partner had at least a college degree, compared to about 44 percent of families without these accounts. Further, because families with 529 plans have higher education levels, children in these families may be more likely to attend college in the first place as research shows that when parents have obtained a college degree, their children are more likely to attend college. 48 Also, a larger proportion of respondents in families who had 529 plans or Coverdells were non-minorities, according to our analysis of SCF data. 49 Specifically, about 84 percent of respondents whose families had 529 48 For example, see OECD (2012), Education at a Glance 2012: OECD Indicators, OECD Publishing. Accessed on September 11, 2012. http://dx.doi.org/10.1787/eag-2012-en. 49 Research shows that racial or ethnic minorities are more likely to be low-income, which is a risk factor that may be associated with less positive educational outcomes and can affect educational achievement. Page 18 GAO-13-64 529 Plans plans or Coverdells identified as white (non-Hispanic). 50 In contrast, about 69 percent of respondents in families without these accounts said they were white (non-Hispanic). Moreover, only 5 percent of respondents in families with 529 plans or Coverdells identified as Black/African- American, compared to 14 percent without these accounts. While SCF does not provide insight on why more white families use 529 plans than others, a 2010 Sallie Mae study found that a larger percentage of white parents saved for college in general when compared to African-American and Hispanic parents. Further, the study found that white families save more for college even when controlling for income. To determine how the characteristics of families with 529 plans relate to families who already have students in college, we reviewed characteristics of college students’ families using data from the 2007- 2008 NPSAS. 51 While families with 529 plans are not directly comparable to families of college students, 52 we developed profiles of these two groups to understand the broad trends and found several differences. When compared to families with students already in college, families with 529 plans or Coverdells still tended to have higher income and more education (see table 1). 53 This tendency persisted even when examining the median incomes of dependent and independent students separately. Specifically, median parental income for dependent students was slightly more than half the median income of families with 529 plans or Coverdells, and independent students had slightly more than one-quarter of the median income of those with 529 plans or Coverdells. 50 The family may be multi-racial or mixed race, which is not captured in these data. For example, the respondent’s spouse/partner may have a different race or ethnicity. We are 95 percent confident that the percentage of respondents whose families had 529 plans or Coverdells who identified as white (non-Hispanic) was between 78.2 and 88.8 percent. 51 These were the most recent NPSAS data available. 52 While our estimates using SCF are for all families (including families with children in college, with children not in college, and with no children), our estimates for families using NPSAS are exclusively for families with a current college student. In NPSAS, families include the student and the student’s parents (if the student is dependent) or the student’s spouse (if the student is independent). Further, the population of NPSAS students’ families is not the same as the population of families with a child in college because a family may have more than one student in college in a given year. Consequently students’ family characteristics derived from NPSAS are not directly comparable to family characteristics based on the SCF. 53 Data were not available on the financial assets held by families of college students. Page 19 GAO-13-64 529 Plans Table 1: Estimates of Selected Characteristics of College Students’ Families (from NPSAS) and Families with and without 529 Plans (from SCF) All Families (from SCF) College students’ Families with Families without Characteristic families (from NPSAS) 529 plans or Coverdells 529 plans or Coverdells a Median income $47,747 $142,400 $45,100 b Percent with at least a college degree 42% 91% 44% c Percent white (non-Hispanic) 62% 84% 69% Source: GAO analysis of NPSAS 2007-2008 and SCF 2010 data. Notes: All amounts are in 2010 dollars. Unless otherwise noted, percentage estimates have 95 percent confidence intervals within 5 percentage points of the estimate, and median income estimates have 95 percent confidence intervals within 5 percent of the estimate itself. a For college students’ families, this is the students’ parents’ income (for dependent students) or the students’ income including their spouses’ if married (for independent students). For “all families”, this is the total household income. We are 95 percent confident that the median income for families with 529 plans or Coverdells was between $125,400 and $159,400. b For college students’ families, this is the percentage of students who had at least one parent with a college degree. For “all families”, this is the percentage of families in which the respondent or his/her spouse or partner had at least a college degree. c For college students’ families, this is the percentage of students with this race/ethnicity. For “all families”, this is the percentage of families for which the respondent had this race/ethnicity. We are 95 percent confident that the percentage of respondents whose families had 529 plans or Coverdells who identified as white (non-Hispanic) was between 78.2 and 88.8 percent. Page 20 GAO-13-64 529 Plans For additional information on 529 plan distributions, see text box below. 529 Plan Distributions In 2010, more than 1 million taxpayers reported taking approximately $23.8 billion in distributions from 529 plans, which they generally reported using for qualified education expenses. Our review of IRS data found that about 5 percent of taxpayers who took distributions from a 529 plan reported that a portion of their 529 plan distributions were nonqualified and subject to a penalty. See table below for additional data. 2007 2008 2009 2010 Number of taxpayers taking distributions from 529 plans (in 0.82 0.97 1.07 1.08 millions) Total amount of distributions from $8.5 $19.1 $20.5 $23.8 529 plans (in billions of 2010 dollars) Percent of taxpayers for which some portion of the 529 plan distribution was reported as nonqualified and 7.8% 6.7% 6% 5.3% subject to a penalty Source: GAO analysis of IRS SOI 2007-2010 data. Notes: We are 95 percent confident that the number of taxpayers who reported taking distributions from 529 plans was between 772,000 and 867,000 in 2007, between 916,000 and 1,019,000 in 2008, between 1,012,000 and 1,127,000 in 2009, and between 1,023,000 and 1,135,000 in 2010. We are 95 percent confident that the total amount of 529 plan distributions (in 2010 dollars) was between $7.8 and $9.2 billion in 2007, between $17.7 and $20.5 billion in 2008, between $18.9 and $22.2 billion in 2009, and between $21.2 and $26.4 billion in 2010. States’ 529 Plan Features and Other Factors Can Affect Participation Tax Benefits, Fees, and Officials in every state and most experts and representatives we Investment Options Vary interviewed identified tax benefits, fees, and investment options as some Across State 529 Plans of the most important features consumers consider when choosing Page 21 GAO-13-64 529 Plans whether or not to participate in a 529 plan and, if so, which plan to choose. These features vary by state and plan. 54 All states offer at least one plan and many offer a combination of college savings (either direct- sold, advisor-sold, or both) and prepaid plans. For example, 14 states offer a direct-sold plan only, 22 states offer both direct-sold and advisor- sold plans, and 6 states offer all three plan types: direct-sold, advisor- sold, and prepaid. The popularity of direct-sold college savings plans has grown over time and in 2011 total assets were essentially evenly split between those and advisor-sold plans. State Tax Benefits States offer a range of tax benefits for 529 plans, and these benefits are a primary incentive to investing in a 529 plan, according to many state officials we interviewed. In addition to earnings growing tax-deferred, our analysis of CSPN data shows the majority of states with an income tax offer some form of benefits: 33 offer a tax deduction and 3 offer a nonrefundable tax credit to residents who participate in their state’s plan. 55 Five states also extend benefits to residents who participate in any state’s plan. Almost all states limit tax benefits to the account owners, but one state extends those benefits to grandparents, aunts, and uncles who contribute to the plan. Officials in some states we interviewed said they provide additional tax benefits, for example, one state offers an exemption from the state inheritance tax. Others allow contribution amounts that exceed the annual deduction limit to be carried over to the following year’s return. 56 Fees Various fees and expenses may be associated with 529 plans, including administrative and investment fees. Administrative fees, which are charged by the state and/or the program manager hired by the state, 54 Throughout the report, the term states’ plans also includes the plan offered by the District of Columbia. 55 As of July 2012, 5 states allow residents to deduct all 529 plan contributions in a given tax year. Overall, 28 states provide a tax deduction on contributions, with a variety of limits. Among these states, for example, 18 states allow deductions that range from $500 to $10,000 for a single filer and $1,000 to $20,000 for married couples filing jointly. Additionally, 9 other states also allow deductions—ranging from $250 to $13,000—per beneficiary. 56 We previously reported tax filers may have difficulty figuring out how to maximize federal tax benefits given interactions with state tax codes. To maximize their combined federal and state tax benefit, tax filers may also need to take into account the state treatment of federal higher education tax expenditures. See GAO-12-560. Page 22 GAO-13-64 529 Plans cover administration of the 529 program, including customer service and marketing. Investment fees are charged by the investment company to manage the funds. The aggregate of these administrative and investment fees is often referred to as “annual asset-based fees,” which are expressed as a percentage of the fund’s average net assets. In addition, advisor-sold plans may also charge a “sales load”—that is, a fee paid to the selling broker when the fund is purchased or redeemed—and direct- sold and advisor-sold plans may also charge participants additional fees for services such as enrolling or changing the account owner. Fees among 529 plans vary widely; total annual asset-based fees among plans nationwide ranged from 0 percent to 1.97 percent for direct-sold plans and 0 percent to 2.78 percent for advisor-sold plans, as of July 2012. As seen in table 2, there is variation among states in both administrative and investment fees. Such variation occurred even among states with similar administrative structures. For example, among three of the states we reviewed where most administrative functions were conducted in-house, one state charged administrative fees of between 0.44 percent and 0.46 percent of the balance annually, another charged between 0.15 percent and 0.20 percent, and a third charged no administrative fees, instead covering operational costs and salaries through an annual state appropriation. 57 Investment fees also varied: for example, underlying mutual fund fees ranged from 0 percent to 1.82 percent of the balance annually, depending on the type of investment option a participant chooses. For advisor-sold funds, sales loads also varied, ranging from 0 percent to 5.75 percent, 58 in part based on the fund class. 59 In addition, among the five states we reviewed, four did not charge an enrollment or application fee, while one charged $25, although the fee may be waived through promotions to encourage participation. 57 A state’s administrative fees may vary within a range based on the investment option chosen by the participant. 58 Initial sales charges are paid when the shares are purchased and deferred sales charges are paid when the shares are redeemed. 59 Some mutual funds offer investors different types of shares, known as classes. Each class will invest in the same investment portfolio but will have different shareholder services and/or distribution arrangements with different fees and expenses. Page 23 GAO-13-64 529 Plans Table 2: Fee Ranges (Percent) for College Savings Plans, Direct-Sold and Advisor-Sold Range for Direct-Sold Range for Advisor-sold Fee Description Plans (in percent) Plans (in percent) Administrative Fees State Fee Fees charged by the state for operational costs. 0 to 0.75 0 to 0.15 Program Manager Fee Fee charged by program manager for administering 0 to 0.8 0 to 1.15 the plan. Investment Fees Underlying Fund Expense Expenses or fees charged by an investment firm for 0 to 1.82 0 to 1.82 managing the funds in the plans. Distribution Fee Fee charged in advisor-sold plans provided to — 0 to 1.0 brokers who sold the fund shares. Miscellaneous Fees 0 to 0.77 0 to 0.5 Source: GAO analysis of CSPN data as of July 2012. Note: In addition to the fees listed above, plans may also charge application, cancellation, change in beneficiary, change in investment options, or other fees. 529 plan fees remain higher than fees for similar mutual funds an investor might purchase outside of a 529 plan. According to a 2011 study by Morningstar, 529 plan mutual funds charged, on average, an additional 0.31 percent of the account balance annually in investment fees compared with their respective mutual fund categories in the open market. 60 The administrative fees charged by most 529 funds raise the cost even higher. However, Morningstar does note that fees for 529 plans have declined in recent years and officials at the majority of state plans we interviewed told us they have taken steps to reduce fees – for example, by renegotiating program manager contracts, using competitive bidding for program management, or consolidating functions in-house rather than using a program manager. As we have previously reported, fees are one of many factors participants should consider when investing because even a small fee increase can significantly decrease savings over time. 61 60 Morningstar, 2011 529 College Savings Plans Research Paper and Industry Survey (Chicago, IL: October 2011) 61 GAO, Private Pensions: Changes Needed to Provide 401(k) Plan Participants and the Department of Labor Better Information on Fees, GAO-07-21 (Washington, D.C.: November 16, 2006) and 401(k) Plans: Increased Educational Outreach and Broader Oversight May Help Reduce Plan Fees, GAO-12-325 (Washington, D.C.: April 24, 2012). Page 24 GAO-13-64 529 Plans Investment Options State plans offer a variety of investment options to 529 college savings plan participants. 62 Plans in the states we reviewed, for example, include up to 17 different investment options, including age-based, static, and customized portfolios, to cater to participants’ various levels of risk tolerance and investment sophistication. Age-based options were generally the most popular and, according to state officials we interviewed, may appeal to investors who might have more limited investment experience or a lower risk tolerance. One state plan we reviewed also offers a customized option for participants who seek more control over their investments, which allows them to designate their own allocations in funds such as stocks and bonds. For more risk-averse participants, some states also offer a FDIC-insured investment option or one that in some other way guarantees the investment’s principal. 63 To help investors determine which plan best meets their needs, officials we interviewed in two states said their states provide risk assessment information through customer call centers. 64 One state developed a risk tolerance questionnaire to explain investment scenarios, while the other had a representative ask informal questions to help potential investors assess their own risk level. Families can also choose to invest in prepaid plans, which were offered in three of the five states we reviewed. These plans also vary in fees, payment options, and cost. Two plans, for example, charged an annual administrative fee of just under 0.50 percent and the third charged no annual fee. In terms of payment options and costs, two states we interviewed offered prepaid plans by academic periods or units that can be used to pay for future tuition costs with the option of paying in lump sum or through a monthly payment program. According to state officials, the cost of these prepaid plans is generally determined by forecasting future tuition and fees at different types of schools (4-year, community college, etc.), given a number of actuarial assumptions on tuition inflation and anticipated investment return. One state, for example, offered a contract to cover four years of college costs for a child currently under 62 In states we reviewed, assets in prepaid plans are invested by the state. 63 According to CSPN data, 45 states offer age-based investment options and 20 states offer a guaranteed investment option. 64 While these assessments are designed to help investors decide which plan best meets their risk tolerance and investment goals, state officials told us they do not provide investment advice. Page 25 GAO-13-64 529 Plans age five at a lump sum of $56,600 and another state offered a similar contract for just under $66,500. A third state we reviewed does not offer units or contracts, but allows participants to contribute any amount to the plan. When the participant withdraws the funds for qualified educational expenses, they will receive the amount they contributed adjusted by a tuition inflation value. Participation is Affected by Families encounter a number of barriers as they consider saving for Ability to Save and Other college: they may struggle with making saving a priority, and for those Factors, but Some States who do plan to save, many do not know 529 plans exist as a savings option. Additionally, once families decide to invest in a 529 plan they may Have Adopted Strategies to have trouble understanding how it works and the variation across plans Address Barriers may affect their ability to select one that best meets their needs (see fig. 5). Figure 5: Factors that Affect 529 Plan Participation Ability to Save Families may encounter a variety of barriers saving for college, such as insufficient income, underestimating the cost of college, and misconceptions about financial aid availability, but selected states are taking steps to help address these barriers. A 2010 national survey published by Sallie Mae found that while nearly nine out of ten parents Page 26 GAO-13-64 529 Plans expected their child would attend some form of higher education, only three out of five parents of college-bound children have saved or invested for their oldest child’s education. 65 First, many families may not save because they lack adequate income or have competing financial priorities. The same Sallie Mae survey reported that 68 percent of those who are not saving cited a lack of money as a major reason. 66 State officials and some experts we interviewed also cited this as a challenge: for example, one state official said that the economic downturn has affected some families who are reluctant to make deposits or participate in a 529 plan because they may need to choose between paying their mortgage and saving for college. In terms of competing priorities, two industry representatives we interviewed stressed that retirement should be a higher priority than saving for college. Officials from one state added when a family’s budget shrinks or the economy is uncertain, families reduce college rather than retirement savings. Furthermore, a few industry representatives said families should consider using other tax-deferred savings vehicles where funds could be used for multiple purposes, such as retirement and education. The states we selected to review have adopted strategies to expand participation among lower income families who may have limited resources to allocate towards savings, including offering matching programs, low minimum initial contributions, and less risky investment options. • Matching Programs: While a limited number of states offer such options, matching programs to expand low-income families’ participation and increase contributions by incentivizing saving have increased this group’s participation and college access, according to some state officials we interviewed. According to CSPN data, 14 states offer some form of matching program and two of the three states we reviewed specifically use matching programs to target low-income families. To qualify in one state, for example, a family must earn less than 200 percent of the federal poverty level 67 and deposit a minimum of $100 during the participating calendar year. The state then matches contributions, 65 Sallie Mae, August 2010. 66 Sallie Mae, August 2010. 67 For example, in 2011 a family of 4 earning less than $44,700 would qualify, according to plan documentation. Page 27 GAO-13-64 529 Plans dollar for dollar, up to $400 annually per beneficiary for up to 4 years. In addition to increasing participation, officials from one state plan noted that the matching program can also help minimize student loans and reduce the amount students will have to work while in school. An ongoing experiment conducted by the Center for Social Development also found a positive impact on the number of 529 plan accounts for families who were automatically enrolled in a state-owned 529 account with a matching program in one state. 68 In addition to participating in the automatically opened account, families in the treatment group were offered an additional $100 to open a private account. These families opened private 529 accounts at a higher rate (17 percent of families with a match compared to 2 percent of those in the control group without the incentives), and deposited more into those accounts. 69 While matching programs may have positive results, two states we reviewed reported challenges with funding and awareness. One state’s program had not been authorized since 2008 and officials in another state said their enrollment remained low despite being open to all participants because the state’s 529 marketing budget was eliminated. • Low or No Minimum Initial Contributions: Low or no minimum initial contributions and fee waivers may also help increase participation among low-income families, according to state officials and others we interviewed. Nationally, minimum initial 68 Center for Social Development, George Warren Brown School of Social Work, Washington University in St. Louis, The SEED for Oklahoma Kids Experiment: Initial Account Opening and Savings, (St. Louis, MO: 2010). The study was conducted in partnership with the State of Oklahoma and RTI International. SEED for Oklahoma first automatically enrolled treatment participants who received $1,000 in an Oklahoma 529 account. (Among the 1,361 treatment participants, one declined opening the state-owned account.) These participants were also provided a time-limited incentive of $100 to open their own private accounts, savings matches, and information on Oklahoma 529 accounts. In contrast, control participants were offered no SEED for Oklahoma financial incentives or information about Oklahoma 529, although they could open their own 529 plan accounts, just as any non-study participant. Additional data on participants are expected to be collected over the next few years. 69 Rates only include additional private accounts opened by participants, i.e. accounts that were separate from the state-owned accounts that were opened automatically for the treatment group. The average deposit amount was $61 for treatment participants and $40 for the control group (not statistically significant). However for the subcategory of private accounts owned by parents or guardians, treatment participants deposited an average of $47 versus $13 for the control group (p<.05). Page 28 GAO-13-64 529 Plans contributions range from $10 to $5,000, according to our analysis of CSPN data; however, the majority of states require an initial contribution of $25 or less. Two states allow participants to open an account with any amount. Officials in one state reported that keeping the initial deposit amounts low can also help facilitate one of their main goals: to help spur the mental commitment and habit to save. • Less Risky Investment Options: Officials from many states we reviewed said they offer investment options that pose less risk to the investor, which can appeal to low- to moderate-income families. One state, for example, partnered with two local banks to provide a FDIC-insured option to target families who might otherwise save in the bank’s savings account. According to an official from the plan’s banking partner, clients with more assets often use financial planners and are aware of 529 plans, while the FDIC-insured option was designed for those without financial planners and who use the bank’s more traditional products. According to state officials, most of the states we reviewed are not tracking participant’s demographic information such as income, however, making the success of these efforts for low-income families difficult to assess. Second, in addition to insufficient income, some families may not save because they procrastinate or underestimate the true cost of college, according to officials from most of the states we reviewed. Some parents may not budget money to save for college due to a lack of understanding about what college really costs or they become overwhelmed and do nothing, officials at one state 529 plan said. To address these challenges, selected state 529 plans have adopted financial literacy programs and marketing strategies emphasizing the importance of saving even a small Page 29 GAO-13-64 529 Plans amount early and often. 70 To target families with younger children, two states provide materials to parents of newborns through the hospital or direct mail and two states work with elementary schools to distribute materials on the states’ 529 plans. Some states also establish contribution deadlines linked to certain benefits, such as discounted enrollment, or provide incentives to families who contribute during certain times of the year. To prevent families from feeling overwhelmed about college costs, one state has focused its marketing on saving a small amount each month, $25, to help reduce the student’s future debt, instead of focusing on the total cost of college. Third, parents may not save for college because they have misconceptions about financial aid availability, according to state officials and some higher education experts. Some noted that families may not understand that most students receive aid in the form of loans that will need to be repaid, rather than receiving grants or scholarships. According to a 2010 national study, 84 percent of non-saving parents expect their child to qualify for enough scholarships or financial aid to cover the costs of college with 49 percent citing it as a major reason for not saving. 71 However, according to our analysis of NPSAS data for all postsecondary students in the 2007-2008 school year, only about half received grants with a median amount of $3,400 (in 2010 dollars). About 40 percent of students took out loans, with a median amount of about $6,800 (in 2010 dollars). Financial aid including grants and loans only covered 37 percent of a student’s cost of attendance, with the median out-of-pocket expenses totaling about $7,000 per student (in 2010 dollars). Further, while many families hope to fund college through financial aid, one recent study found that 2011 levels of scholarships were unsustainable as colleges felt the 70 The Department of Education has emphasized the importance of financial literacy related to college savings. For example, the GEAR UP program, where grantees design projects that promote participating students’ secondary school completion and enrollment in postsecondary education, also includes promotion of financial literacy and economic literacy education or counseling. In 2012, Education announced a plan to introduce in fiscal year 2013 a new demonstration project, which includes an evaluation component, to determine the effectiveness of pairing federally-supported college savings accounts with GEAR UP activities as part of an overall college access and success strategy. Education has also recently sponsored research assessing interventions that provide families with information on 529 college savings plans and incentives for them to invest in such accounts. The study will gather data on the effects of these interventions on college savings behavior and college outcomes. 71 Sallie Mae, August 2010. Page 30 GAO-13-64 529 Plans impact of a difficult economic climate, constraints on endowments, and tighter budgets. 72 In response, many of the states we reviewed have attempted to address misconceptions about the financial aid process. For example, one state lists common myths about 529 plans on its website, explaining that approximately 60 percent of federal financial aid comes in the form of loans, a debt the family must repay. The site encourages families to save even in small amounts to offset the amount of debt the family will incur. Awareness of 529 Plans as a For those who are saving for college, awareness that 529 plans exist as a Savings Option savings option is a challenge to participation, according to officials in most of the states we reviewed. In addition, among parents who are saving for college, one study found that almost half are unfamiliar with 529 plans. An additional 4 percent volunteered that they had never heard of the plan or did not know what it was. 73 Families also learn about 529 plans through financial planners, according to many state officials we interviewed; therefore, awareness may be a particular challenge for low-income families who generally do not have access to such resources. Further, some state officials and industry representatives we interviewed encountered families with misperceptions about how 529 plans work, such as not understanding they can invest in plans outside their home state or use savings at any college or university. For example, officials from two states reported families mistakenly believe prepaid plans can only be used at an in-state institution. 74 State Variation, Tax While the high number of plans and variation in investment options and Implications, and a Limit on cost can offer consumers choice, families who ultimately decide to save Investment Option Changes with a 529 plan can find it difficult to compare plans, according to many state officials and experts. For example, the wide range of fee types can be difficult to understand and compare across plans because they are not consistent and it is difficult to compare asset-based fees based on percentages with flat fees such as an annual fee in dollar amounts, according to one financial expert. Plan complexity can also make 72 Sallie Mae, How America Pays for College 2012 (Newark, DE: 2012). 73 Sallie Mae, August 2010. 74 If a beneficiary of a prepaid tuition plan elects to attend an out-of-state college, the state 529 plan will typically pay the student’s chosen institution the tuition and fees it would have paid at an in-state public college, which may be less than the tuition at the chosen institution. Page 31 GAO-13-64 529 Plans marketing and communication difficult, according to officials we interviewed in two states. Marketing officials from one state told us consumers requested more information on 529 plans, but it was difficult to communicate information about the complex plans in a simple, consumer- friendly way. Officials in another state noted that plan complexity and a lack of clear information can discourage families from researching and enrolling in a plan. Many state officials and some academic experts and industry representatives reported that simplifying the information available to consumers might keep families from feeling overwhelmed. Because MSRB rules do not apply to state issuers when they market their direct- sold 529 plans, CSPN developed a set of disclosure principles to help states provide consistent information. The voluntary principles contain recommendations to help consumers understand plans and compare various features, such as fees, tax issues, and risk. While disclosures have been helpful, according to one expert who consults with a number of states, there is room for improvement: disclosures could be more rigorous in ensuring that consumers are informed of less-costly options within a state if they exist and should cover information on prepaid plans, which is currently not standardized. When comparing direct-sold disclosure documents across states we reviewed, we found that the five states generally adhered to the CSPN disclosure principles and contained consistent information a consumer could compare. Three states, however, were missing some information that could be helpful to consumers, such as information on the risk of state tax law changes and a statement that 529 plans should only be used to save for qualified higher education expenses. The structure of federal and state tax benefits, a primary incentive for some 529 plan investors, can also affect participation as they may not be as helpful to low-income families, according to some academic experts, industry representatives, and state officials we interviewed. 75 Low-income families with low or no tax liability see less benefit from federal tax benefits and may see no benefit from nonrefundable state tax credits provided to 529 plan investors. According to a 2009 Treasury report, families saving in 529 plans may need to carefully consider whether their 75 Prior GAO work has also found education savings accounts, such as 529 plans, are more advantageous to families with higher incomes and tax liabilities. See GAO-12-560. Page 32 GAO-13-64 529 Plans child will go to college because the penalty incurred if the funds are not used for qualified education expenses may outweigh the tax benefits for low-income families. 76 In addition, Treasury noted most states do not extend tax benefits to residents investing in out-of-state plans, limiting competition. As a result, families have a strong incentive to choose their home state plan even if other plans offer preferable investment choices. In 2009, Treasury recommended states eliminate this “home-state bias” to provide more investment options to consumers, more intense competition between plans, and potentially lower fees. 77 According to an annual report from one state that extends its tax benefits to residents who invest in other states’ plans, doing so, when other states do not, puts the home state plan at a competitive disadvantage. State officials explained that this policy results in other plans marketing their products in the state. Residents, therefore, may be unaware of their home state plan’s benefits, according to the report. Finally, the fact that account holders may change their investment option only one time per year may affect participation in 529 plans, according to state officials and some industry representatives we interviewed. Officials from one financial services company advocated removing any limits on changing investment choices beyond those imposed by the financial services company sponsoring the fund, as is the case with 401(k) plans and individual retirement accounts. Another industry representative observed some 529 plan participants changing their account beneficiary solely because it would allow them to change their investment options. However, the representative cautioned that participants should not change their investment options too frequently; many experts advocate that investors are best served by sticking with a long-term investment plan. 76 Treasury, 2009. 77 Treasury, 2009. In addition, Treasury also recommended increasing the provision of age-based indexed funds; making contribution limits more effective by making them per- beneficiary limits rather than per-beneficiary per-state limits; improving industry reporting of plans’ historical returns, plan participation by income, and how plans are invested at the account level; and improving government monitoring of 529 plan accounts and their disbursements. Page 33 GAO-13-64 529 Plans Savings in 529 Plans Affect Financial Aid the Same as Other Assets Savings in 529 Plans Are The extent to which savings in 529 plans, or other investments, affect Treated Similarly to Other how much a family is expected to contribute to the cost of college—the Assets That Are Included federal expected family contribution (EFC)—generally depends on the family’s amount of assets. Education incorporates the amount of specific When Calculating the types of assets into various calculations to determine the EFC. 78 Expected Family However, in two calculations, families who meet certain criteria are either Contribution not expected to contribute to the cost of college (automatic zero EFC) or they qualify for a simplified calculation. 79 In both cases, assets, including savings in 529 plans, are not included in the calculation of the EFC. According to the 2007-2008 NPSAS, about a quarter of families who filed FAFSAs met these criteria. In other calculations, assets, including savings in 529 plans, may affect the EFC to different extents depending on whether students are dependent on their parents or are independent with dependents of their own. For dependent students, between 2.64 percent and 5.64 percent of parental assets may be included in the EFC as described below: • First, the parents report the net worth (current value minus debt) of their investments (see fig. 6 #1), 80 but before the total contribution from assets is calculated, an amount known as the “education savings and asset protection allowance” is subtracted 78 Assets that are not required to be reported on the FAFSA include principal place of residence, a family farm, family-owned small businesses, retirement plans, and whole life insurance. 79 Students can qualify for an automatic zero EFC or a simplified EFC based on their parents’ income (dependent students) or their income (independent students) and any of the following: (1) receipt of federal benefits, such as Social Security Supplemental Security Income or food stamps, (2) no requirement to file an income tax return or may file an IRS 1040A or 1040EZ, or (3) a parent (dependent student) or student/spouse (independent student) is a dislocated worker. 80 Prepaid plans are worth the refund value of the tuition credits or certificates. Page 34 GAO-13-64 529 Plans (see fig. 6, #2). This allowance is designed to help protect a portion of the parents’ assets. 81 • Second, 12 percent of any parental asset amount that exceeds the education savings and asset protection allowance is used to determine the contribution from assets that will be considered in the final EFC calculation (see fig. 6 #3). • Third, this contribution from assets is added to the parents’ available income to determine their adjusted available income (see fig. 6, #4). • Fourth, a marginal rate, from 22 percent up to a maximum of 47 percent, is applied to the sum of the parents’ available income and contributions from assets (known collectively as the adjusted available income) to determine their EFC (see fig. 6, #5). As a result, the amount of net parental assets, including savings in 529 plans, that can be included in the EFC ranges from 2.64 percent to 5.64 percent. 82 81 The education savings and asset protection allowance increases with the age of the older parent. For example, in the 2012-2013 school year, the allowance for two parents, the older of which was 45, was $41,300 whereas the allowance for two parents, the older of which was 55, was $53,400. According to Education’s Federal Student Aid Handbook, the allowances approximate the present cost of an annuity, which, when combined with Social Security benefits, would provide at age 65 a moderate level of living for a retired couple or single person. 82 The 2.64 percent is the result of assessing 22 percent of the 12 percent of assets that are included in the EFC (.22 X .12 = .0264). Similarly, 5.64 percent results from assessing 47 percent of 12 percent (.47 X .12 = .0564). Page 35 GAO-13-64 529 Plans Figure 6: Expected Family Contribution from Parents’ Assets for Dependent Student Even if the dependent student is the 529 plan account owner, the savings are still assessed at the parents’ asset rate. The EFC includes 20 percent of the value of a dependent student’s assets; however, savings in a 529 plan where the student is the account owner are still considered assets of the dependent student’s parent(s). Page 36 GAO-13-64 529 Plans Independent students can have up to 20 percent of their 529 plans savings and other assets included in their EFC. However, assets of independent students with dependents (other than a spouse) may be assessed at a lower rate. After an asset protection allowance is subtracted from their net assets, any remainder is multiplied by 7 percent. Then, a marginal rate, from 22 percent to 47 percent, is applied to the sum of their available income and contribution from assets—similar to the process used for parental assets of dependents students. Therefore, from 1.54 percent to 3.29 percent of assets may be included in their EFC. In contrast, independent students who do not have dependents (other than a spouse) will have 20 percent of any assets exceeding the asset protection allowance included in their EFC. Education officials said independent students without children to support are generally expected to contribute a higher percentage of their assets because their primary focus should be on paying for their education. Distributions from 529 plans owned by parents and/or the student will not be considered as income in the EFC calculation in future years if they are used for qualified education expenses. However, if a student receives 529 plan distributions from an account owned by someone other than himself or the custodial parent, those funds count as student income and could affect the EFC in subsequent years. 83 For example, if a student received funds from a 529 plan owned by a grandparent, he would have to report those funds as untaxed student income on the next year’s FAFSA, according to Education officials (see fig. 7). 83 Since the EFC is based on the previous year’s income, distributions from a 529 plan not owned by the parent or student would not affect the EFC for the student’s last year in college if the student received the funds after his final FAFSA had been filed. Page 37 GAO-13-64 529 Plans Figure 7: FAFSA Questions on Student Untaxed Income A small percentage of families who applied for federal financial aid in 2007-2008 had enough assets to be included in the determination of their EFC. In our analysis of families who filled out the FAFSA, we found that 13 percent of all students—24 percent of dependent students and 4 percent of independent students—had enough assets to be included in their EFC. 84 In other words, the net worth of their (and possibly their parents’ or spouses’) assets exceeded the savings and asset protection allowance and was included in the EFC at some percentage. Education officials said that because the asset protection allowance is high, federal student aid decisions do not heavily rely on assets, such as savings in 529 plans. Officials told us that while home equity was removed from the list of assets used to calculate the EFC in 1992, the asset protection allowance remained the same. Since then, they said, assets have been a less relevant factor in calculating the EFC. 84 In 2007-2008, 61 percent of dependent and 52 percent of independent students filled out the FAFSA. Page 38 GAO-13-64 529 Plans Many States and Selected Institutions Also Treat 529 Plan Savings As Assets States Most state financial aid offices also consider savings in 529 plans as assets. 85 According to the 2009-2010 National Association of State Student Grant and Aid Programs survey, 86 35 states reported that they used the federal methodology for determining the EFC for state aid. 87 However, some states that reported using federal methodology for their primary student needs analysis also indicated they provide special treatment for state 529 college savings or prepaid plans when determining student eligibility for aid. Specifically, seven states that used federal methodology to award their state aid excluded the state’s 529 college savings plan and three excluded the state’s prepaid plan from their calculation for state aid. Of the officials in the six state financial aid offices we interviewed, none said they considered assets to a greater extent than the FAFSA and a few said their state took specific steps to exempt savings in these plans from consideration. Specifically, officials in two states said there is language in their 529 plan authorizing legislation that exempts plan savings when determining a student’s eligibility for state financial aid. Officials in another state said their state issued a regulation stating that savings in a 529 plan would not affect state grant eligibility for residents attending nonprofit higher education institutions. An official in a fourth state said the legislature changed its higher education authorization language so that students would still be eligible to receive a state scholarship even if they enrolled in the state’s prepaid plan. Institutions Institutional financial aid practices vary with regard to assets, but those with more aid to award may gather additional information about a family’s financial status, according to some representatives of national financial 85 According to the 2007-2008 NPSAS, 15 percent of all students received state financial aid. 86 This survey is a central repository for information on state support of students and families paying for postsecondary education. 87 The District of Columbia and Puerto Rico also reported that they use federal methodology. Page 39 GAO-13-64 529 Plans aid organizations and institutional officials we interviewed. 88 Some schools require students to provide information in addition to the FAFSA, such as filling out the College Board’s PROFILE form or submitting tax returns. One official said the PROFILE provides more detailed information on a family’s assets, such as home equity and retirement account balances, which helps the university prioritize the students with the most need. Institutional officials we interviewed said their schools considered savings in 529 plans as assets, even if they used different methodologies to calculate their financial aid or included the assets at different percentages. Officials at two institutions said they did not consider savings in 529 plans beyond how they are already reported by the family on the FAFSA. An official at a third institution said the school does not collect any additional information on savings in 529 plans beyond what is requested on the FAFSA even though the school requires families to fill out the PROFILE form and uses an institutional methodology to award its financial aid. The remaining institutional officials said they collect additional family financial information when calculating student aid, but consider savings in 529 plans similarly to the family’s other assets. Specifically, one institutional official said her school uses the PROFILE form to gather more detailed information about a family’s financial situation. Even so, 529 plan savings do not affect a student’s need any differently than other assets, she said, which are assessed by the institution at about five percent of their value. Additionally, she said 529 plan assets are considered parental assets even if they are reported as student assets because the school assesses parental assets at a lower percentage. An official at another institution said her school assesses assets at around 20 percent of their value when calculating the EFC for institutional aid. Officials’ opinions varied on whether savings in 529 plans should affect financial aid, but many said families’ concerns that these savings will have an adverse effect are common. One state financial aid official said it would be helpful if 529 plan savings were excluded entirely from the calculation because including them can be a deterrent to saving. She said her office often encounters families who feel penalized for saving 88 According to the 2007-2008 NPSAS, 22 percent of all students received institutional financial aid. Page 40 GAO-13-64 529 Plans because they believe the students without savings receive financial aid. Likewise, a 529 plan official said regardless of whether the student’s financial aid will be reduced by savings in a 529 plan, there is the perception that it will. In contrast, one institutional financial aid official said savings in 529 plans should not be treated any differently than other assets because the need analysis is meant to determine the family’s fair share of college expenses and excluding 529 plans would be counter to this aim. One researcher we interviewed found that the issue may be most important for those families who are on the margin of receiving federal financial aid. Regardless of the perceived effect 529 plan savings may have on financial aid, some of the officials we interviewed said they encourage families to save for college because much of the aid they may be offered could be in the form of loans, so saving will generally be in the student’s long-term financial interest. As currently designed, 529 college savings plans benefit a small Concluding percentage of U.S. families. In general these families tend to be wealthier Observations than others. It is not clear whether the $1.6 billion in federal tax expenditures that these plans represent strategically targets limited federal resources. Although 529 plans do help some families save for college, families with less income and who are uncertain about whether their children will attend college may have less incentive to invest resources in 529 plans than in other forms of savings. In addition, the tax benefits attractive to a higher-income family do not offer as much benefit to a family with lower tax liability. Questions about who benefits from this tax expenditure occur in an environment of long-term fiscal challenges and difficult choices about how the federal government allocates limited resources. Reviewing 529 plans in conjunction with the other billions of dollars in federal educational assistance provided through tax expenditures, credits, and deductions could help Congress determine whether this program is meeting its goals. Similar to GAO’s prior work on higher-education related tax expenditures, our analysis of 529 college savings plans was not able to address all questions that could inform future policy choices regarding 529 plans. For example, what is the purpose of the federal tax benefits provided through 529 plans? Are the goals and objectives clearly defined and measurable? Who is the target population for 529 plans and does the current structure provide appropriate incentives for that population? How do the 529 plan federal tax benefits interact with other programs, such as federal financial aid and other higher education tax benefits and savings vehicles? Page 41 GAO-13-64 529 Plans Consideration of these questions could facilitate continued congressional oversight of this tax expenditure. We provided a draft of this report to Education, Treasury, and IRS for Agency Comments comment. The agencies provided technical comments that were incorporated, as appropriate. We are sending copies of this report to the Secretary of Education, Secretary of the Treasury, Commissioner of Internal Revenue, relevant congressional committees, and other interested parties. In addition, the report will be available at no charge on the GAO Web site at http://www.gao.gov. If you or your staff have any questions about this report, please contact me at firstname.lastname@example.org or 202-512-6806. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff who made key contributions to this report are listed in appendix III. Sincerely yours, Michelle Sager, Acting Director Education, Workforce, and Income Security Issues Page 42 GAO-13-64 529 Plans Appendix I: Objectives, Scope, and Appendix I: Objectives, Scope, and Methodology Methodology Our review examined: (1) the percentage and characteristics of families enrolling in 529 plans, (2) the plan features and other factors that affect participation in 529 plans, and (3) the extent to which savings in 529 plans affect financial aid awards. To answer these research objectives, we analyzed government data; interviewed state 529 plan officials from select states as well as industry representatives and academic experts; reviewed plan documents and analyzed industry data; conducted a literature review; interviewed federal, state, and institutional financial aid officials; and reviewed Department of Education (Education) and Internal Revenue Service (IRS) documents as well as relevant federal laws, regulations and guidance. We assessed the reliability of the data we used by reviewing documentation, interviewing knowledgeable officials, and conducting electronic testing on relevant data fields. We found the data we reviewed reliable for the purposes of our analyses. We conducted this performance audit from November 2011 to December 2012 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. Analysis of Government To determine the percentage and characteristics of families enrolling in Data 529 plans, we reviewed data from the 2010 Survey of Consumer Finances (SCF); the 2007-2008 National Postsecondary Student Aid Study (NPSAS); and 2007-2010 Statistics of Income (SOI) federal tax data. The 2010 SCF, 2007-2008 NPSAS, and 2010 SOI were the most recent data available at the time of our engagement, so to ensure consistency in reporting we adjusted all dollar amounts from previous years’ data to 2010 dollars. Each of these three data sources (SCF, NPSAS, and SOI) are based on probability samples and estimates are formed using the appropriate estimation weights provided with each survey’s data. Because each of these samples follows a probability procedure based on random selections, they represent only one of a large number of samples that could have been drawn. Since each sample could have provided different estimates, we express our confidence in the precision of our particular sample’s results as a 95 percent confidence interval (e.g., plus or minus 2.5 percentage points). This is the interval that would contain the actual Page 43 GAO-13-64 529 Plans Appendix I: Objectives, Scope, and Methodology population value for 95 percent of the samples we could have drawn. Unless otherwise noted, all percentage estimates based on the SCF, NPSAS, and SOI have 95 percent confidence intervals that are within 5 percentage points of the estimate itself, and all numerical estimates other than percentages have 95 percent confidence intervals that are within 5 percent of the estimate itself. For our analysis of the percentage and characteristics of families who held 529 plans, we relied primarily on restricted data from the 2010 SCF. SCF is a triennial survey sponsored by the Board of Governors of the Federal Reserve System (Federal Reserve) to provide detailed information on the finances of U.S. households. The SCF sample of 6,492 households represented approximately 118 million households in 2010. It collects detailed financial characteristics on an economically dominant single individual or couple (married or living as partners) in a household, which we refer to as a family for the purposes of this report. 1 For our analysis, we aggregated financial information so that, unless otherwise noted, all SCF estimates are for the family rather than the individual survey respondent. We did not restrict our analysis to families with children, in part, because 529 plans can be used for nearly anyone, including one’s child, grandchild, niece, nephew, and oneself. 2 However, about 88 percent of families with 529 plans or Coverdell Education Savings Accounts (Coverdells), 3 a similar education savings vehicle, had 1 The unit of analysis for the SCF includes this economically dominant individual (or couple) along with economically interdependent individuals (such as minor children) also living in the household. 2 Additionally, because of the limited number of families in the SCF sample that had 529 plans or Coverdell Education Savings Accounts, we were unable to produce reliable estimates for subgroups of families with these accounts, such as characteristics by age group or by whether families had children living with them. We were, however, able to generate reliable estimates for all families with 529 plans or Coverdell Education Savings Accounts. 3 Similar to 529 plans, Coverdell accounts allow families to save for education expenses. Account earnings accumulate tax-deferred and earnings included in withdrawals used to pay for qualified education expenses are not subject to federal tax. Unlike 529 plans, (1) Coverdell contributors must generally have a modified adjusted gross income of less than $110,000 per year ($220,000 in the case of a joint return), (2) Coverdells have annual contribution limits of $2,000 that must generally stop when a beneficiary reaches 18 years of age, and (3) Coverdells can be used for qualified elementary, secondary, or postsecondary expenses typically for individuals under age 30. These Coverdell features will be affected if the changes in the Economic Growth and Tax Relief Reconciliation Act of 2001 are not extended. Page 44 GAO-13-64 529 Plans Appendix I: Objectives, Scope, and Methodology children 25 years of age or younger living with them. Our estimates for 529 plans included Coverdells because Federal Reserve officials said respondents did not always distinguish between the two account types; therefore, we did not separate these responses because of data reliability concerns. However, the officials indicated that a larger share of the SCF respondents reported having 529 plans than Coverdells. Further, using SOI data, we estimate that in 2010 approximately 85 percent of tax filers who took a distribution from either a 529 plan or a Coverdell reported distributions from a 529 plan while 14 percent reported distributions from a Coverdell and 1 percent reported distributions from both. We wrote an analysis program that the Federal Reserve ran using their restricted SCF dataset to separate information on Medical Savings Accounts and Health Savings Accounts, which had been included in the 2010 public dataset with 529 plans and Coverdells. Federal Reserve officials modified some resulting information to protect the privacy of survey respondents, for example by rounding dollar amounts. Using SCF, we generated estimates on the percentage and characteristics of families enrolled in 529 plans or Coverdells and of families not enrolled in these plans. We examined family characteristics such as wealth (financial assets), income, education, and race or ethnicity. To calculate financial assets, we used the methodology the Federal Reserve uses to produce variables for its published Bulletin articles. This methodology included assets held in checking, savings, and brokerage accounts, certificates of deposit, mutual funds, stocks, bonds, life insurance, retirement accounts, and other vehicles such as 529 plans. Assets held in retirement accounts included those in defined contribution plans (e.g. a 401(k), individual retirement account, or thrift savings plan) as well as in traditional pensions or defined benefit plans. To calculate income, we used the family’s self-reported total income. To report the family’s highest educational attainment, we reviewed the education of each respondent and his or her partner or spouse and included whichever was higher. We reported information on the respondent’s race or ethnicity, which does not necessarily indicate the race or ethnicity of other family members. 4 We used 2007-2008 NPSAS data to develop a similar demographic profile for college students and generate other estimates on college costs 4 Further, the respondent is not necessarily the head of the household. Page 45 GAO-13-64 529 Plans Appendix I: Objectives, Scope, and Methodology and financial aid amounts. NPSAS is a comprehensive study by Education that examines how students and their families pay for higher education. It includes nationally representative samples of 113,535 undergraduates, 12,585 graduate students, and 1,581 first-professional students 5 enrolled any time between July 1, 2007 and June 30, 2008. The NPSAS data are based on administrative records and student interviews, and NPSAS includes survey results from both students who received financial aid and those who did not. While we used NPSAS to develop a demographic profile for college students similar to the one we developed for the general population using SCF, families with 529 plans are not directly comparable to families of college students. For example, while our estimates using SCF are for all families (including families with children in college, with children not in college, and with no children), our estimates for families using NPSAS are exclusively for families with a current college student. In NPSAS, families include the student and the student’s parents (if the student is dependent) or the student’s spouse and dependents (if the student is independent). Further, the population of NPSAS students’ families is not the same as the population of families with a child in college because a family may have more than one student in college in a given year. Consequently students’ family characteristics derived from NPSAS are not directly comparable to family characteristics based on the SCF, though for the purposes of our report we use similar terminology to describe them. Similar to our analysis of SCF, we generated estimates of the characteristics of college students’ families— including income, 6 education, and race or ethnicity. To report income, we calculated the total income of (1) the student’s parents (if the student was dependent) and (2) the student and the student’s spouse (if the student was independent). To report the family’s highest educational attainment, we reviewed the education of each student’s mother and father and included whichever was higher. We also reported information on the student’s race or ethnicity, which does not necessarily indicate the race or ethnicity of other family members. We also developed separate estimates for students who are considered either dependent on their parents or independent for financial aid purposes. 5 First-professional students are students pursuing degrees in fields such as pharmacy, dentistry, medicine, or law. 6 Information was not available on the assets for families of college students. Page 46 GAO-13-64 529 Plans Appendix I: Objectives, Scope, and Methodology We also used NPSAS to generate other estimates related to the cost of college and amount of financial aid awards. First, we estimated the median annual cost of attendance at 4-year public and private non-profit institutions. This included tuition and fees, room and board, transportation, and personal expenses, though the estimate is valid only for students who attended one institution. Second, we estimated the percentage of students who received grants and loans, as well as the median amount of these grants and loans and the percent and amount of college expenses remaining. Third, we generated estimates for the proportion of students who filled out the Free Application for Federal Student Aid (FAFSA) and, for those who did fill out the FAFSA, the proportion who met certain criteria to have assets excluded from the federal expected family contribution (EFC) and the proportion whose assets affected the EFC. Finally, we calculated the percentage of students who received state and institutional financial aid. We also analyzed 2007-2010 taxpayer data from SOI to determine the extent to which taxpayers used distributions from 529 plans for qualified education expenses and how the tax savings from these plans were distributed across income levels. The SOI individual tax return file is a stratified probability sample of income returns filed with the IRS. The SOI sample of 308,583 returns represented approximately 143 million tax returns filed for 2010. We combined data from the SOI individual tax file with information from the Form 1099-Q. A 529 plan must file a Form 1099-Q with the IRS and the account owner or beneficiary each time a taxpayer receives a distribution from a 529 plan account. 7 This form includes information on the amount of the distribution and the earnings (or loss) on the distribution. When taxpayers receive a Form 1099-Q, they must determine if the distribution was used for qualified education expenses. If the distribution, or any portion of it, was nonqualified, the earnings portion is subject to taxes and, in some cases, a penalty. The taxpayer determines the amount of taxes and penalty owed on the nonqualified distribution by completing Form 5329, which is contained in the individual tax return file. By combining information from the 1099-Q with information in the individual tax return file, we identified the percentage of taxpayers who reported nonqualified distributions that were 7 Distributions from a 529 plan include those used for qualified education expenses as well as those used for nonqualified expenses or refunds to the account owner or beneficiary. Page 47 GAO-13-64 529 Plans Appendix I: Objectives, Scope, and Methodology subject to a penalty. 8 We also used SOI data to estimate the tax savings by using the National Bureau of Economic Research’s (NBER) TAXSIM Model, a microsimulation model of U.S. federal and state income tax systems. 9 TAXSIM calculates estimated liabilities under U.S. federal and state income tax laws from actual tax returns that have been prepared for public use by the Statistics of Income Division of the IRS. 10 Our analysis of the tax savings from 529 plans excludes returns with a filing status of married filing separately. Interviews with State 529 To provide information on the factors that affect participation we Plan Representatives, interviewed officials from the following five state 529 plans and their Academic Experts, and industry partners: Louisiana, Michigan, Pennsylvania, Utah, and Virginia. We used College Savings Plan Network (CSPN) data to select states that Industry Representatives represented a variety of plan types (direct-sold, advisor-sold, and pre- paid), offered a number of features (i.e., various state tax benefits, state matching program), and were geographically diverse. We also used suggestions provided by academic experts and industry representatives to inform our selection as well as to provide information on 529 plan participation. We interviewed academic researchers (including the Center for Social Development), industry regulators (the Financial Industry Regulatory Authority and the Municipal Securities Rulemaking Board), financial services companies (American Funds and UPromise), financial experts (such as Financial Research Corporation and Morningstar), College Savings Plan Network, Savingforcollege.com, and consumer interest groups (Investment Company Institute and the American Association of Individual Investors). Analysis of Plan and We analyzed CSPN data on state 529 plans to provide a national Industry Data and overview of plan features, such as fees and state tax benefits. Biennially, Documentation states submit plan data to CSPN through an online system to be posted 8 While earnings on all distributions are subject to a tax if the distribution was not used for qualified education expenses, the penalty is waived under certain circumstances, such as when a beneficiary dies or receives a scholarship. 9 NBER provided GAO with a copy of TAXSIM that we executed within our secure tax computing environment. 10 See Daniel Feenberg and Elisabeth Coutts, “An Introduction to the TAXSIM Model,” Journal of Policy Analysis and Management, vol. 12, no. 1, (1993): 189-194. Page 48 GAO-13-64 529 Plans Appendix I: Objectives, Scope, and Methodology on the CSPN website. CSPN provided us with data on each state as of July 2012. We analyzed the data for every state for both direct-sold and advisor-sold plans on the following features: whether the state offers a matching grant program, whether the state offers tax deductions for contributions and the amount, whether the state offers tax credits for contributions and the amount, types of investment options offered, total contribution limits, and required initial contribution amounts. We also analyzed the following fee categories: program manager fee, state fee, annual account maintenance fee, miscellaneous fee, annual distribution fee, estimated underlying fund expenses, total annual asset-based fees, maximum deferred sales charge, and minimum initial sales charge. Further, we compared CSPN disclosure principles with direct-sold plan disclosure documentation for the five states we interviewed. We reviewed the extent to which the selected states incorporated elements of the CSPN disclosure principles and whether plan documentation was easily comparable across states. Specifically, we compared whether the state documents contained eleven elements outlined in the principles, including: a summary of key features, an assessment of the individual summary features, a statement of any guarantee by the state issuer or the state, information on state tax treatment and other benefits, information that the state offers more than one plan, fee descriptions, and investment risks, among others. These elements were chosen based on discussions with states and experts who identified plan fees, tax benefits, and investment options as some of the most important features consumers consider when choosing whether or not to participate in a 529 plan. In addition to recording whether states have disclosed the information listed above, we assessed whether any information was missing, where the information was located in the document, and any other observations about the ability to find and understand plan information. Literature Review We reviewed studies conducted by academics, researchers, industry representatives, and federal agencies on why families choose to participate in 529 plans and what features might serve as barriers or incentives. We identified literature published since 2006, when Congress passed the Pension Protection Act of 2006, Pub. L. No. 109-250, which made permanent the tax-exemption on 529 plan distributions used for qualified education expenses. Our review included scholarly/peer reviewed material, government reports, hearings and transcripts, trade/industry articles, association/nonprofit/think tank publications, and working papers. We searched information sources such as EconLit, Page 49 GAO-13-64 529 Plans Appendix I: Objectives, Scope, and Methodology ProQuest, ERIC, PolicyFile, WorldCat, ECO, PapersFirst, ArticleFirst, and Academic OneFile. These online sources are nationally recognized databases that index and abstract research literature. We selected search terms to capture literature that specifically addressed 529 plans, college savings plans, qualified state tuition programs, and prepaid tuition. Of the 32 studies we identified, 12 studies met the following criteria: 1) included information on plan features in specific states, 2) addressed the consequences for consumers of choosing one type of 529 plan over another, 3) identified barriers or incentives for consumers to choose 529 plans, 4) included data collected by states on plan participation, and/or, 5) included information on plan disclosures to consumers. All studies cited in the report were reviewed by at least two GAO analysts. Studies that included statistical methods were reviewed by a GAO statistician and social science analyst. All studies were reviewed for methodological soundness and to ensure that any limitations associated with study methodologies were conveyed to readers in our report text, footnotes, or this appendix. Financial Aid Analysis To understand the extent to which savings in 529 plans affect federal financial aid awards, we interviewed Education officials in the Office of Postsecondary Education. We also reviewed relevant statutory provisions, the FAFSA, the Federal Student Aid Handbook, and other Education documents related to calculating the EFC. To understand the extent to which savings in 529 plans are considered in state financial aid calculations, we interviewed officials from state financial aid offices in six states. To select the state financial aid offices, we used information from a 2009-2010 survey by the National Association of State Student Grant and Aid Programs to identify states that indicated they used a financial aid formula other than the federal methodology in their primary needs analysis and/or provided special treatment for state 529 plans. For report consistency, we selected the same states selected for 529 plan site visit locations to the extent possible (i.e., where the data supported the selection based on the criteria). We interviewed representatives in the following state financial aid offices: Louisiana Office of Student Financial Assistance, Michigan Office of Scholarships and Grants, New York Higher Education Services Corporation, Pennsylvania Higher Education Assistance Agency, Utah Higher Education Assistance Authority, and State Council of Higher Education for Virginia. We also selected six institutions from the states whose financial aid offices were selected for interviews. To obtain a national perspective on Page 50 GAO-13-64 529 Plans Appendix I: Objectives, Scope, and Methodology institutional financial aid and determine the best method for selecting the individual institutions, we interviewed representatives at several financial aid organizations including the Association of Private Sector Colleges and Universities, the College Board, the National Association of Student Financial Aid Administrators, the National Association of Independent Colleges and Universities, and the American Association of Community Colleges. In these interviews, some officials said that schools with larger endowments were likely to require families to provide additional information, such as that required on the College Board’s PROFILE application, to award their institutional financial aid. 11 We matched the 2012-2013 College Board’s list of institutions that use the PROFILE application with Education’s 2009-2010 Integrated Postsecondary Education Data System to calculate endowment amounts per student at public and private non-profit four-year institutions. We also reviewed the list of schools that participate in the Private 529 Consortium and selected at least one school that was also part of this group. One state did not have an institution that used the PROFILE application so we reviewed websites of postsecondary schools in that state to identify a school that collected data in addition to the FAFSA. We interviewed representatives at the following institutions: Xavier University of Louisiana, University of Michigan, St. Lawrence University, Swarthmore College, University of Utah, and University of Richmond. 11 The College Board is a not-for-profit membership organization that provides college students with financial aid support and scholarships. It also conducts research and advocacy on behalf of students, educators, schools, and colleges. The PROFILE is an application developed by the College Board to help institutions gather information used to award nonfederal student aid funds. Page 51 GAO-13-64 529 Plans Appendix II: Free Application for Federal Appendix II: Free Application for Federal Student Aid, 2012-2013 Student Aid, 2012-2013 Page 52 GAO-13-64 529 Plans Appendix II: Free Application for Federal Student Aid, 2012-2013 Page 53 GAO-13-64 529 Plans Appendix II: Free Application for Federal Student Aid, 2012-2013 Page 54 GAO-13-64 529 Plans Appendix II: Free Application for Federal Student Aid, 2012-2013 Page 55 GAO-13-64 529 Plans Appendix II: Free Application for Federal Student Aid, 2012-2013 Page 56 GAO-13-64 529 Plans Appendix II: Free Application for Federal Student Aid, 2012-2013 Page 57 GAO-13-64 529 Plans Appendix II: Free Application for Federal Student Aid, 2012-2013 Page 58 GAO-13-64 529 Plans Appendix II: Free Application for Federal Student Aid, 2012-2013 Page 59 GAO-13-64 529 Plans Appendix II: Free Application for Federal Student Aid, 2012-2013 Page 60 GAO-13-64 529 Plans Appendix II: Free Application for Federal Student Aid, 2012-2013 Page 61 GAO-13-64 529 Plans Appendix III: GAO Contacts and Staff Appendix III: GAO Contacts and Staff Acknowledgments Acknowledgments Michelle Sager, Acting Director, Education, Workforce, and Income GAO Contacts Security Issues, 202-512-6806 or email@example.com. In addition to the contact named above, Gretta Goodwin (Assistant Staff Director), Amy Anderson, Rachel Beers, and Laura Henry contributed to Acknowledgments all aspects of this report. Also making key contributions were Carl Barden, James Bennett, Nora Boretti, Jessica Botsford, Jason Bromberg, Alicia Cackley, Melinda Cordero, Patrick Dudley, Shannon Finnegan, Kim Frankena, Mark Glickman, David Lewis, Ashley McCall, John Mingus, Mark Ramage, MaryLynn Sergent, George Scott, Walter Vance, Kathleen van Gelder, and Michelle Loutoo Wilson. (131126) Page 62 GAO-13-64 529 Plans The Government Accountability Office, the audit, evaluation, and GAO’s Mission 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. The fastest and easiest way to obtain copies of GAO documents at no Obtaining Copies of cost is through GAO’s website (http://www.gao.gov). Each weekday GAO Reports and afternoon, GAO posts on its website newly released reports, testimony, and correspondence. To have GAO e-mail you a list of newly posted Testimony products, go to http://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, http://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 on Facebook, Flickr, Twitter, and YouTube. Connect with GAO Subscribe to our RSS Feeds or E-mail Updates. Listen to our Podcasts. Visit GAO on the web at www.gao.gov. Contact: To Report Fraud, Website: http://www.gao.gov/fraudnet/fraudnet.htm Waste, and Abuse in E-mail: firstname.lastname@example.org Federal Programs Automated answering system: (800) 424-5454 or (202) 512-7470 Katherine Siggerud, Managing Director, email@example.com, (202) 512- Congressional 4400, U.S. Government Accountability Office, 441 G Street NW, Room Relations 7125, Washington, DC 20548 Chuck Young, Managing Director, firstname.lastname@example.org, (202) 512-4800 Public Affairs U.S. Government Accountability Office, 441 G Street NW, Room 7149 Washington, DC 20548 Please Print on Recycled Paper.
Higher Education: A Small Percentage of Families Save in 529 Plans
Published by the Government Accountability Office on 2012-12-12.
Below is a raw (and likely hideous) rendition of the original report. (PDF)