Social Security Reform
In Brief: The Effects of Investing the Social Security Trust Funds in GNMA Mortgage-Backed Securities
Research Report
John Gist, AARP Public Policy Institute
February 2006
Learn More
- Report Home
- Full Report (PDF)
- In Brief (HTML)
- In Brief (PDF)
Table of Contents: Introduction | Findings | Conclusions
Introduction
Although one of the most popular and successful government programs in U.S. history, Social Security faces a long-term financing problem, and the Social Security Trustees project that the trust funds will be depleted in 2041. The 1994–96 Advisory Council on Social Security recommended that trust fund investment income be increased as part of the solution to the long-term financing problem. Proposals have been made to achieve this increase by investing part of the trust funds in the stock market or by establishing private Social Security accounts and having workers invest directly in the market. Stock market returns have, over the long run, been higher than returns on U.S. government Treasury bonds (the current investment vehicle for the trust funds), so market investment could in theory increase income to the trust funds. However, equity or corporate bond investments are not a legal option for trust fund assets, and congressional action would be required to permit them.
One alternative investment that is permitted by law is mortgage-backed securities (MBSs) issued by the Government National Mortgage Association (GNMA, or Ginnie Mae). MBSs are debt obligations that represent an ownership claim on cash flows (principal and interest) from a pool of mortgage loans. Ginnie Mae MBSs have typically provided a higher rate of return than Treasury bonds and are guaranteed in principal and interest by the United States government. In the 1960s and 1970s, trust fund investments included securities issued by GNMA and the Federal National Mortgage Association (FNMA, or Fannie Mae). A change in the investments of the trust fund could increase income to the trust funds, thus postponing the insolvency date.
Findings
In this issue paper, Tom Hungerford, a private economic consultant who was formerly with the Social Security Administration and with the Levy Economics Institute, modeled the effect of Ginnie Mae MBS investments on the trust funds, using Social Security Administration information and projections. The model assumed that each year, beginning in 2006, excess tax revenues and trust funds monies due to be rolled over would be invested in GNMA MBSs. The buying of GNMA MBSs would then continue until a certain percentage (either 30% or 50%) of the trust funds is invested in MBSs. Mortgage-backed securities are bought or sold to keep the investment percentage constant. Investments in GNMA MBSs would continue until (1) the trust fund exhaustion date is reached or (2) the trust fund ratio drops below 200.1 Three interest rate scenarios were modeled: The return on GNMA MBSs was assumed to be either 100, 50, or 10 basis points higher than the interest rate of special issues assumed by the Social Security Trustees. The second component of the analysis estimated the effect of Social Security surpluses on the budget of the rest of the federal government by regressing the federal funds surplus on the trust funds surplus and other control variables.
If 50 percent of the Social Security Trust Funds is invested in GNMA MBSs and a return that is 100 basis points higher than special issues is earned, the insolvency date could be postponed by two years to 2043. If the returns were only 50 basis points higher, the insolvency date is postponed by one year. The insolvency date is unchanged if the return is only 10 basis points higher. The results are broadly similar if 30 percent of the trust funds are invested in MBSs. The increase in trust fund investment income varies from $28.2 billion (in present value terms) for the 30 percent trust fund investment and the lowest rate of return to $523.8 billion for the 50 percent investment and highest rate of return.2 These numbers represent 0.01 percent to 0.23 percent of taxable payroll (that is, the present value of 75-year taxable payroll).3
If some of the trust fund surplus were diverted to GNMA MBS investments, the federal government would be forced to borrow from the public by selling U.S. Treasury securities. The question is, how much will the yield on Treasuries need to increase in order to induce the public to purchase these Treasury securities? With 30 percent of trust fund investments in MBSs, the investments could eventually amount to slightly less than 7 percent of gross domestic product (GDP). If there are no changes in the non-Social Security government budget surplus and if the non-Social Security government budget continues to run deficits, federal debt held by the public could eventually rise by 7 percent of GDP and interest rates could rise by 21 basis points. However, the rise in interest rates will be gradual, because it takes 13 years for the trust fund’s investments in MBSs to reach their peak. Furthermore, both the strength of the economy and the Federal Reserve have a greater impact on interest rates.
Conclusions
Clearly, this investment policy will not solve the long-term financing problem of Social Security, but the change could certainly help close the long-term financing gap. Investing in GNMA MBSs does entail some risk compared with investing in special issues. With special issues, the maturity dates are set for the benefit of the trust funds, and special issues can always be redeemed at par. GNMA MBSs carry prepayment risk and the possibility of incurring capital losses. The trust funds have had the experience of selling marketable securities before maturity at a loss. However, with prudent management, the benefits of investing in GNMA MBSs could outweigh the costs.
Footnotes
1 The trust fund ratio is the ratio of end of year trust fund assets to the projected costs in the coming year, expressed as a percentage.
2 The present discounted value shows the amount of money that would have to be invested today in U.S. Treasury securities to yield the nominal dollar amounts.
3 These numbers should be compared to the 75-year actuarial deficit of Social Security, which amounts to 1.9 percent of taxable payroll (present value).
Written by John Gist, AARP Public Policy Institute
January 2006
©2006 AARP
All rights are reserved and content may be reproduced, downloaded, disseminated, or transferred, for single use, or by nonprofit organizations for educational purposes, if correct attribution is made to AARP.
Public Policy Institute, AARP, 601 E Street, NW, Washington, DC 20049
Pub ID: INB116