This is the accessible text file for GAO report number GAO-14-535 
entitled 'Debt Management: Floating Rate Notes Can Help Treasury Meet 
Borrowing Goals, but Additional Actions Are Needed to Help Manage 
Risk' which was released on June 23, 2014. 

This text file was formatted by the U.S. Government Accountability 
Office (GAO) to be accessible to users with visual impairments, as 
part of a longer term project to improve GAO products' accessibility. 
Every attempt has been made to maintain the structural and data 
integrity of the original printed product. Accessibility features, 
such as text descriptions of tables, consecutively numbered footnotes 
placed at the end of the file, and the text of agency comment letters, 
are provided but may not exactly duplicate the presentation or format 
of the printed version. The portable document format (PDF) file is an 
exact electronic replica of the printed version. We welcome your 
feedback. Please E-mail your comments regarding the contents or 
accessibility features of this document to Webmaster@gao.gov. 

This is a work of the U.S. government and is not subject to copyright 
protection in the United States. It may be reproduced and distributed 
in its entirety without further permission from GAO. 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. 

United States Government Accountability Office: 
GAO: 

Report to the Chairman, Committee on Ways and Means, House of 
Representatives: 

June 2014: 

Debt Management: 

Floating Rate Notes Can Help Treasury Meet Borrowing Goals, but 
Additional Actions Are Needed to Help Manage Risk: 

GAO-14-535: 

GAO Highlights: 

Highlights of GAO-14-535, a report to the Chairman, Committee on Ways 
and Means, House of Representatives. 

Why GAO Did This Study: 

To continue meeting its goal of financing the federal government's 
borrowing needs at the lowest cost over time, Treasury began issuing a 
new type of security-—a 2-year floating rate note (FRN)-—in January 
2014. The FRN pays interest at a rate that resets periodically based 
on changes in the rate of the 13-week Treasury bill (to which the FRN 
is indexed). GAO was asked to review Treasury debt management, 
including this product and other debt management issues. 

This report (1) evaluates Treasury's rationale for introducing FRNs 
and (2) identifies the demand for Treasury securities from a broad 
range of investors to assess whether changes would help Treasury meet 
its goals. To address these objectives, GAO used Treasury auction data 
from 1980-2014 to simulate the costs of Treasury FRNs, reviewed 
Treasury documents, surveyed a non-generalizable sample of 82 large 
domestic institutional investors across sectors, and interviewed 
market participants and academic experts. (For the survey and results, 
see [hyperlink, http://www.gao.gov/products/GAO-14-562SP].) 

What GAO Found: 

Issuing floating rate notes (FRN) is likely to help the Department of 
the Treasury (Treasury) meet its goals to borrow at the lowest cost 
over time, extend the average maturity of the debt portfolio, and 
increase demand for Treasury securities, but it also presents risks 
related to changes in interest rates. 

GAO simulated the costs of 2-year Treasury FRNs using historical 
Treasury auction data and found that interest costs of the FRNs were 
generally less than costs of fixed-rate 2-year notes, but could be 
either more or less than costs of 13-week bills, depending on 
assumptions about how investors price the FRNs. GAO also found that in 
rising interest rate environments, the FRNs may be more costly than 
these alternatives. 

Multiple components contribute to achieving lowest cost financing over 
time: issuing FRNs is part of Treasury's approach to achieving this 
goal. GAO analysis identified a number of design elements that may 
affect how FRNs contribute to that goal. Treasury officials believe it 
is prudent for Treasury to extend the average maturity of its debt 
portfolio because the debt level is already high and is expected to 
grow. Relative to issuing shorter-term debt, 2-year FRNs will help 
Treasury extend the average maturity of the debt portfolio and thereby 
reduce the risk inherent in going to market. Because the interest rate 
on a FRN can change during the life of the security, FRNs expose 
Treasury to the risk of rising interest rates whereas fixed-rate 
securities of the same maturity do not. These shifts in risk are 
likely to be small because currently FRNs are expected to constitute a 
small proportion of Treasury debt. Although managing interest rate 
risk is an important aspect of Treasury's goal to borrow at the lowest 
cost over time, Treasury does not track and report a measure of the 
average maturity of the portfolio that captures the additional 
interest rate risk of FRNs. 

One element of the design of the 2-year FRN-—the difference between 
the term of its index rate (13 weeks) and the length of its effective 
reset period (one week)—-is not typical for floating rate notes and 
creates tradeoffs in interest rate risks but also may result in 
additional demand for the product. The risks could affect the pricing 
of FRNs and raise Treasury's borrowing costs in environments of high 
and volatile interest rates. Treasury officials told us they examined 
design elements, including this difference, before issuing the 2-year 
FRN. However, Treasury had not analyzed how the difference may affect 
FRN pricing. 

FRNs give Treasury debt managers additional flexibility by increasing 
demand for Treasury securities and by adding a new security that meets 
the high demand for short-term securities. Results from GAO's survey 
of a broad range of investors and interviews with market participants 
found that market participants likely will purchase Treasury FRNs 
primarily as a substitute for other Treasury securities, but they will 
also purchase the FRNs as a substitute for non-Treasury securities, 
bringing new and potentially growing demand to Treasury. To provide 
the lowest cost of financing the government over time, Treasury must 
consider investor demand for new and existing products. Survey 
respondents indicated an interest in FRNs of additional maturities and 
in other new Treasury products. Treasury currently offers many ways 
for market participants to provide input, but GAO's survey identified 
opportunities for Treasury to enhance input from some sectors—-
including state and local government retirement fund managers. 

What GAO Recommends: 

GAO recommends that Treasury (1) track and report a measure of 
interest rate risk in its debt portfolio, (2) analyze the price 
effects of the difference between the term of the index rate and the 
reset period, (3) examine opportunities for additional new types of 
securities, such as FRNs of other maturities, and (4) expand outreach 
to certain market participants. Treasury agreed with the 
recommendations and said that they had already taken steps to begin 
implementing them. 

View [hyperlink, http://www.gao.gov/products/GAO-14-535]. To view the 
e-supplement, click on [hyperlink, 
http://www.gao.gov/products/GAO-14-562SP]. For more information, 
contact Susan J. Irving at (202) 512-6806 or irvings@gao.gov. 

[End of section] 

Contents: 

Letter: 

Background: 

Floating Rate Notes Are Likely to Help Treasury Borrow at the Lowest 
Cost over Time, Extend the Average Maturity, and Increase Demand, but 
They Also Present Certain Risks: 

Market Participants Identified Opportunities for Treasury to Enhance 
Investor Input and Expand Product Offerings: 

Conclusions: 

Recommendations for Executive Action: 

Agency Comments and Our Evaluation: 

Appendix I: Simulations of Floating Rate Note Costs: 

Appendix II: Survey Scope and Methodology: 

Appendix III: GAO Contacts and Staff Acknowledgments: 

Tables: 

Table 1: Effect of Floating Rate Note Design Features on 
Attractiveness to Investors: 

Table 2: Definitions of Interest Rate Environments: 

Table 3: Survey Responses by Recipient Type: 

Figures: 

Figure 1: Marketable Securities Offered By Treasury: 

Figure 2: Design Elements of the Treasury 2-Year Floating Rate Note: 

Figure 3: Marketable Interest-Bearing Securities by Year of Maturity, 
as of December 31, 2013 (Total Outstanding: $11.9 trillion): 

Figure 4: Actual and Projected Net Interest Outlays and Implied 
Average Treasury Interest Rates, 2004-2024: 

Figure 5: Three Different Approaches to Investing in Treasury 
Securities over 2 Years: 

Figure 6: Interest Savings or Added Costs from 2-year Floating Rate 
Notes Compared to 2-Year Notes and 13-Week Bills, by Rate Environment: 

Figure 7: Percent of Cases Where 2-Year Floating Rate Notes (FRN) Save 
or Add to Interest Costs Compared to 13-Week Bills and 2-Year Notes, 
by Rate Environment: 

Figure 8: Weighted Average Maturity of Treasury's Marketable Debt 
Outstanding: 

Figure 9: Extent to Which Treasury Floating Rate Note Purchases Will 
Substitute for Other Investments, According to Survey Respondents: 

Figure 10: Extent to Which Market Participants Said They Have 
Opportunities to Provide Input to Treasury: 

Figure 11: Market Participants' Reported Estimated Purchases of 
Floating Rate Notes with Different Maturities: 

Figure 12: Treasury Products that Could Enhance Investor Demand, 
According to Survey Respondents: 

Figure 13: Treasury Debt Management Practices That Could Enhance 
Investor Demand, According to Survey Respondents: 

Figure 14: Interest Savings or Added Costs from 2-year Floating Rate 
Notes Compared to 13-Week Bills and 2-Year Notes, by Volatility in 
Rates: 

Figure 15: Percent of Cases Where 2-Year Floating Rate Notes Save or 
Add to Interest Costs Compared to 13-Week Bills and 2-Year Notes, by 
Volatility in Rates: 

Abbreviations: 

CMT: Constant Maturity Treasuries: 

FRBNY: Federal Reserve Bank of New York: 

FRN: floating rate note: 

GSE: government-sponsored enterprise: 

RMSE: root mean square error: 

TBAC: Treasury Borrowing Advisory Committee: 

TIPS: Treasury Inflation-Protected Securities: 

Treasury: Department of the Treasury: 

WAM: weighted average maturity: 

[End of section] 

United States Government Accountability Office: 
GAO:
441 G St. N.W. 
Washington, DC 20548: 

June 16, 2014: 

The Honorable Dave Camp: 
Chairman, Committee on Ways and Means: 
House of Representatives: 

Dear Mr. Chairman: 

The U.S. Treasury market is the deepest and most liquid government 
debt market in the world, and throughout history the Department of the 
Treasury (Treasury) has been able to borrow the money it needs to 
finance the federal government. In recent years federal debt held by 
the public has risen dramatically, more than tripling in 10 years, 
from $3.9 trillion at the end of fiscal year 2003 to $12 trillion at 
the end of fiscal year 2013.[Footnote 1] The economic slowdown and 
global financial crisis that contributed to this recent, rapid run-up 
in debt also brought about regulatory changes and a "flight to 
quality": this increased the demand for Treasury securities and helped 
keep the interest rates that Treasury paid to its investors near 
historic lows. Low rates meant that, while the debt more than tripled, 
the interest expense on Federal debt held by the public increased by 
only 58 percent from fiscal year 2003 through 2013 to $248 billion. 
But with debt projected to continue to grow and interest rates 
expected to rise, this cost will increase. 

To continue to meet its goal of financing the federal government's 
borrowing needs at the lowest cost over time, in 2013 Treasury 
announced that for the first time in more than 15 years, it would 
begin issuing a new type of security--a Treasury floating rate note 
(FRN). FRNs differ from Treasury's traditional fixed-rate notes in 
that the FRNs pay interest at a rate that can rise and fall during the 
life of the security. According to Treasury officials, FRNs will help 
Treasury to (1) borrow at the lowest cost over time, (2) increase the 
maturity profile of the debt portfolio, and (3) expand Treasury's 
investor base by attracting new buyers. 

Side bar: 

Floating Rate Note (FRN); An FRN pays interest at a rate that is 
indexed to another rate, which can rise and fall during the life of 
the FRN. Treasury's 2-year FRN is indexed to the rate from Treasury's 
most-recent 13-week bill auction. 

[End of side bar] 

You asked us to review Treasury debt management in the context of its 
growing debt portfolio, including the introduction of FRNs. The 
objectives of this report are to (1) evaluate Treasury's rationale for 
introducing FRNs and (2) identify the demand for Treasury securities 
from a broad range of investors to determine whether changes would 
help Treasury meet its goals. To address these objectives, we did the 
following: 

* Simulated the costs of 2-year FRNs, based on Treasury auction data 
from 1980 to 2014 using two models, each with different assumptions 
about the interest rates that Treasury would pay on the FRNs. We 
compared those costs to Treasury's actual costs of funding with 13-
week bills and 2-year notes. We also analyzed how those costs varied 
over different interest rate environments. For more information on our 
cost simulations, see appendix I. 

* Reviewed the analysis Treasury conducted in developing the 2-year 
FRN and the input Treasury received from market participants. 

* Administered an online survey to 82 (62 completed the survey) of the 
largest domestic institutional holders of Treasury securities in the 
following sectors: money market mutual fund managers, mutual and 
exchange-traded fund managers, state and local government retirement 
fund managers, retail and commercial banks, life insurance providers, 
property-casualty insurance providers, and securities broker-dealers. 
Respondents were selected to achieve 50 percent of the total amount of 
Treasury holdings for each sector. The survey results are not 
generalizable to all investors in Treasury securities. For more 
information on our survey methodology, see appendix II. For aggregate 
survey results reproduced as an e-supplement, see GAO-14-562SP. 

* Interviewed market participants--including six primary dealers and 
four asset managers--regarding the market for FRNs, the structure of 
FRNs, other actions Treasury could consider to expand demand for 
Treasury securities, and opportunities for investors to provide input 
to Treasury.[Footnote 2] We selected primary dealers and asset 
managers based on their holdings of Treasury securities and their 
public comments on Treasury's FRN proposals. The views expressed in 
these interviews are not generalizable to all market participants. 
Finally, we also interviewed five academics recognized as having 
expertise in the Treasury market, regarding the market for FRNs, the 
structure of the FRN, and models for pricing the FRN. 

To assess the reliability of the data used in this study, including 
Treasury auction data and information on the largest holders of 
Treasury securities, we reviewed related documentation, conducted 
testing for missing data, outliers, obvious errors, and traced data 
from source documents, where possible and appropriate. To the extent 
possible, we also corroborated the results of our data analyses and 
interviews with other sources. Based on our assessment we believe that 
the data are reliable for the purposes of this report. 

We conducted this performance audit from April 2013 to June 2014 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: 

To achieve its primary debt management objective of financing the 
federal government's borrowing needs at the lowest cost over time, 
Treasury issues debt through a regular and predictable schedule of 
auctions across a wide range of securities. Most of the securities 
that are issued to the public are marketable, meaning that once the 
government issues them they can be resold by whoever owns them. 
[Footnote 3] Marketable debt consists of bills, notes, bonds, Treasury 
Inflation-Protected Securities (TIPS), and, since January 2014, FRNs 
(see figure 1). Currently, Treasury issues bills with maturities 
ranging from a few days to 52 weeks; notes with maturities of 2, 3, 5, 
7, and 10 years; bonds that mature in 30 years; TIPS with maturities 
of 5, 10, and 30 years; and FRNs that mature in 2 years. 

Figure 1: Marketable Securities Offered By Treasury: 

[Refer to PDF for image: illustration] 

Nominal securities: 

Bills: 
1 year or less. 

Notes: 
More than 1 to 10 year. 

Bonds: 
More than 10 to 30 year. 

Inflation-adjusted securities: 

Treasury inflation-protected securities (TIPS): 
5 to 30 year. 

Variable-rate securities: 

Floating rate notes: 
1 to 10 year. 

Bills: 
Typically sold at a discount from their face value. A $1,000 bill 
might sell at auction for $980. At maturity, the investor receives the 
face value—-in this case $1,000. The difference ($20) equals the 
interest earned. 

Notes: 
Sold below, at, or above face value. They pay interest every six 
months until they mature, at which time the investor is paid the face 
value. 

Bonds: 
Sold below, at, or above face value. They pay interest every six 
months until they mature, at which time the investor is paid the face 
value. 

Treasury inflation-protected securities (TIPS): 
Principal increases with inflation and interest is applied to the 
adjusted principal, so interest payments rise with inflation. When 
TIPS mature, an investor is paid the inflation-adjusted principal. 
Should there be deflation, the principal decreases but not below the 
original face amount. 
				
Floating rate notes: 
Sold below, at, or above face value. They pay interest quarterly at a 
rate that varies with changes in the 13-week Treasury bill rate. At 
maturity, the investor is paid the face value. 

Source: GAO analysis of Treasury information. 

[End of figure] 

Treasury issues securities in a wide range of maturities to diversify 
its portfolio and to appeal to the broadest range of investors. 
Issuing at multiple points across a wide range of maturity lengths 
both ensures high demand for Treasury securities and improves the 
general functioning of the financial markets by providing price 
transparency. Investors in Treasury securities include: 

* domestic private investors, such as individuals, banks, and pension 
funds; 

* investment funds and asset managers; 

* state and local governments; and: 

* foreign investors, including both private investors and foreign 
official institutions, such as central banks and national government-
owned investment funds. 

Side bar: 

Total Federal Debt; Total federal debt consists of two components: 
debt held by the public and debt held by government accounts. 

1. Debt held by the public: 
Debt held by the public represents federal debt issued by Treasury and 
held by investors outside of the federal government, including 
individuals, corporations, state or local governments, the Federal 
Reserve, and foreign governments. 

2. Debt held by government accounts: 
Debt held by government accounts represents balances in the federal 
government's accounts (primarily trust funds, including the Social 
Security and Medicare trust funds) that accumulate surpluses. 

[End of side bar] 

Treasury auctions securities on a regular and predictable schedule to 
reduce market uncertainty, facilitate investor planning, and enhance 
the liquidity of Treasury securities, thereby helping Treasury to 
borrow at the lowest cost over time.[Footnote 4] Regular and 
predictable issuance means that Treasury debt managers choose the 
times and maturities of series of securities--rather than of 
individual issues--as part of their strategy for debt management. 
While this means that Treasury does not "time the market," issuing on 
a regular and predictable schedule does not preclude Treasury debt 
managers from considering fiscal conditions and the contemporaneous 
costs and benefits of shorter-term versus longer-term financing. 
Treasury also announces significant changes long before they are 
implemented and facilitates investor planning by giving market 
participants advance notice of its issuance plans through regular 
quarterly refunding statements. For example, Treasury published a 
notice and request for comments on its potential issuance of FRNs in 
March 2012, almost two years before the first Treasury FRN auction. 

At least in the near term, FRNs will constitute a small proportion of 
Treasury's debt portfolio. Treasury held its first FRN auction on Jan. 
29, 2014, and it plans to issue the 2-year FRN every 3 months with a 
reopening auction in each of the 2 intervening months.[Footnote 5] 
Treasury auctioned $15 billion at the first FRN auction in January 
2014 and $13 billion at each of the reopening auctions in February and 
March. Assuming Treasury FRN issuance amounts remain constant, at the 
end of calendar year 2015 outstanding Treasury FRNs will total $328 
billion, which is less than 3 percent of the current amount of debt 
held by the public.[Footnote 6] The first FRNs have a 2-year maturity. 
Although Treasury regulations allow it to issue FRNs with maturities 
ranging from 1 to 10 years, Treasury has not announced plans to issue 
FRNs with other maturities.[Footnote 7] The interest rate on 
Treasury's 2-year FRN is indexed to the 13-week Treasury bill and 
resets daily. Although the interest rate resets daily, in practice it 
will change only weekly because the 13-week bill is currently 
auctioned weekly. The 2-year FRN pays interest quarterly and accrues 
interest daily at a rate that equals the index rate from the most 
recent 13-week Treasury bill auction plus the FRN spread, which is 
determined competitively at the FRN auction (see figure 2).[Footnote 8] 

Figure 2: Design Elements of the Treasury 2-Year Floating Rate Note: 

[Refer to PDF for image: illustration] 

2-year floating rate note interest rate: 
FRN interest rate: Index rate and FRN spread. 

* The FRN accrues interest daily at a rate that consists of two parts: 
(1) the index rate, which is the rate from the most recent 13-week 
bill auction, and (2) the FRN spread, which is determined 
competitively at the FRN auction and remains fixed throughout the life 
of the FRN. 
* In practice, the FRN interest rate changes only weekly because the 
13-week bill is auctioned weekly. 
* The daily interest accrual cannot be negative. It is floored at zero. 

2-year floating rate note (FRN): 
103 interest rate resets in a 2-year period; 
Interest paid on a quarterly basis; 
Payment of the final interest payment plus the principal of the 
security. 

Source: GAO representation of Treasury information. 

Note: Although the interest rate on the FRN is reset daily, the 
interest rate paid on the FRN will only change weekly because its 
index rate, the 13-week bill rate, is currently auctioned weekly. 

[End of figure] 

The mix of outstanding Treasury securities can significantly influence 
both rollover risk and the federal government's borrowing cost. 
Rollover risk includes two types of risk: (1) interest rate risk--the 
risk that Treasury will have to refinance its debt at less favorable 
interest rates, and (2) market access risk--the risks inherent in 
coming back to the market to refinance the debt.[Footnote 9] As shown 
in figure 3, as of December 31, 2013, $7.9 trillion (or 67 percent) of 
outstanding marketable Treasury securities mature by 2019 and will 
need to be rolled over (i.e., refinanced)--potentially at higher 
interest rates. 

Figure 3: Marketable Interest-Bearing Securities by Year of Maturity, 
as of December 31, 2013 (Total Outstanding: $11.9 trillion): 

[Refer to PDF for image: vertical bar graph] 

Calendar year: 2014; 
Amount: $3.040 trillion; 
Percentage of total outstanding that matures during timeframe: 26%. 

Calendar year: 2015 through 2019; 
Amount: $4.898 trillion; 
Percentage of total outstanding that matures during timeframe: 41%. 

Calendar year: 2020 through 2024; 
Amount: $2.537 trillion; 
Percentage of total outstanding that matures during timeframe: 21%. 

Calendar year: 2025 through 2034; 
Amount: $351 billion; 
Percentage of total outstanding that matures during timeframe: 3%. 

Calendar year: 2035 and later; 
Amount: $1.029 trillion; 
Percentage of total outstanding that matures during timeframe: 9%. 

Source: GAO analysis of Treasury data. 

Note: This includes only debt outstanding as of December 31, 2013. As 
Treasury refinances maturing debt, the dollar amount that will mature 
in subsequent times will increase. Included in this data is a small 
amount of marketable debt held by government accounts. This data does 
not include $15 billion of long-term marketable securities issued by 
the Federal Financing Bank and held by government accounts that are 
not currently traded in the market. 

[End of figure] 

One measure Treasury uses to manage rollover risk is the weighted 
average maturity (WAM) of outstanding marketable Treasury securities. 
[Footnote 10] A high WAM could indicate that Treasury is paying higher 
rates on its debt because longer-term securities generally have 
interest rates higher than shorter-term securities. On the other hand, 
when they need to be refinanced, shorter-term securities expose 
Treasury to the risk of rising interest rates and market access risk. 
The introduction of the new Treasury FRN comes at a time when both 
interest rates and interest costs are expected to rise (see figure 4). 

Figure 4: Actual and Projected Net Interest Outlays and Implied 
Average Treasury Interest Rates, 2004-2024: 

[Refer to PDF for image: multiple line graph] 

Fiscal year: 2004; 
Net interest: $160 billion; 
Implied average interest rate: 4.1%. 

Fiscal year: 2005; 
Net interest: $184 billion; 
Implied average interest rate: 4.3%. 

Fiscal year: 2006; 
Net interest: $227 billion; 
Implied average interest rate: 4.9%. 

Fiscal year: 2007; 
Net interest: $237 billion; 
Implied average interest rate: 4.9%. 

Fiscal year: 2008; 
Net interest: $253 billion; 
Implied average interest rate: 5%. 

Fiscal year: 2009; 
Net interest: $187 billion; 
Implied average interest rate: 3.2%. 

Fiscal year: 2010; 
Net interest: $196 billion; 
Implied average interest rate: 2.6%. 

Fiscal year: 2011; 
Net interest: $230 billion; 
Implied average interest rate: 2.5%. 

Fiscal year: 2012; 
Net interest: $220 billion; 
Implied average interest rate: 2.2%. 

Fiscal year: 2013; 
Net interest: $221 billion; 
Implied average interest rate: 2%. 

Fiscal year: 2014; 
Net interest: $233 billion; 
Implied average interest rate: 1.9%. 

Fiscal year: 2015; 
Net interest: $268 billion; 
Implied average interest rate: 2.1%. 

Fiscal year: 2016; 
Net interest: $326 billion; 
Implied average interest rate: 2.5%. 

Fiscal year: 2017; 
Net interest: $404 billion; 
Implied average interest rate: 2.9%. 

Fiscal year: 2018; 
Net interest: $493 billion; 
Implied average interest rate: 3.4%. 

Fiscal year: 2019; 
Net interest: $569 billion; 
Implied average interest rate: 3.7%. 

Fiscal year: 2020; 
Net interest: $635 billion; 
Implied average interest rate: 4%. 

Fiscal year: 2021; 
Net interest: $694 billion; 
Implied average interest rate: 4.1%. 

Fiscal year: 2022; 
Net interest: $755 billion; 
Implied average interest rate: 4.2%. 

Fiscal year: 2023; 
Net interest: $819 billion; 
Implied average interest rate: 4.3%. 

Fiscal year: 2024; 
Net interest: $880 billion; 
Implied average interest rate: 4.4%. 

Source: GAO analysis of Congressional Budget Office data, including 
February 2014 baseline net interest payment projections. 

Note: Actual values through the end of fiscal year 2013; projected 
values 2014-2024. Net interest outlays are interest expenses on 
Treasury debt held by the public and exclude other interest costs, 
including interest received by other federal entities, such as trust 
funds, and interest on debt issued by agencies other than Treasury 
(primarily the Tennessee Valley Authority). 

[End of figure] 

Floating Rate Notes Are Likely to Help Treasury Borrow at the Lowest 
Cost over Time, Extend the Average Maturity, and Increase Demand, but 
They Also Present Certain Risks: 

2-Year FRNs Are Likely to Cost Less than 2-Year Fixed-Rate Notes but 
Could Cost More or Less Than Bills, and the Cost Will Vary by the 
Interest Rate Environment: 

Side bar: 

Floating Rate Note (FRN) Spread; The difference (i.e., spread) between 
the index rate and the interest rate for the FRN. The FRN spread is 
determined at the FRN auction and is fixed over the life of the FRN. 

[End of side bar] 

We analyzed the potential cost to Treasury of issuing 2-year FRNs and 
found they are likely to have interest costs lower than 2-year fixed-
rate notes and not substantially different than 13-week bills. As a 
result, FRNs will likely result in savings over the long run, helping 
Treasury achieve its goal of borrowing at the lowest cost over time. 
Our simulations found interest costs and savings varied depending on 
the security to which the FRN is compared, how the FRN is treated by 
investors, and the interest rate environment. We found that the cost 
of 2-year FRNs was generally less than that of fixed-rate 2-year 
notes, but that it could be either more or less than the cost of 13-
week bills depending on assumptions regarding investor treatment of 
the FRN. In addition, in all cases and in all environments, savings 
tended to be greater--or added costs lower--under a model that sets 
the FRN spread based on its weekly reset than under an alternative 
model where the FRN spread is influenced by its final maturity of 2 
years. We also found that, while issuing 2-year FRNs generally results 
in cost savings, they may be more costly than other alternatives in 
certain rate environments, such as rising rate environments. 

Prior to issuance of the first FRN, Treasury conducted its own 
analysis of the potential cost of FRNs. Treasury's analysis found that 
from 1982 to 2010, issuance of 2-year FRNs would have led to cost 
savings compared to fixed-rate notes. Treasury's analysis, however: 
(1) compared the cost of 2-year FRNs only to 2-year notes and not to 
other alternatives, and (2) assumed a fixed spread of 15 basis points 
(or 0.15 percentage points). 

Side bar: 

Floating Rate Note (FRN) Index Rate; The rate to which the interest 
rate of an FRN is indexed. Treasury's 2-year FRN is indexed to the 
rate from Treasury's most-recent 13-week bill auction. 

[End of side bar] 

To estimate the potential cost of FRNs to Treasury, we compared the 
cost of hypothetical 2-year FRNs both to the cost of 2-year fixed-rate 
notes and to series of rolling 13-week bills, using historical auction 
data from January 1980 to March 2014 (see figure 5 below). We made 
these comparisons using two models, each with different assumptions 
about the spread over the index rate that Treasury would pay. We also 
compared the cost of FRNs in the various interest rate environments. 

Although it is uncertain what Treasury would issue in the absence of 
FRNs, Treasury has indicated that, at least initially, the FRNs would 
be a substitute for Treasury bill issuance. Both in interviews and in 
our survey of large holders of Treasury securities, market 
participants also indicated that they see the FRNs as a substitute for 
bills. However, Treasury has also indicated that it intends to reduce 
the share of debt funded by bills in order to increase its WAM. 
Without the 2-year FRN, Treasury might have increased the WAM by the 
same amount by instead increasing its issuance of 2-year fixed-rate 
notes, making them an appropriate benchmark with which to compare the 
costs of the FRNs. 

Figure 5: Three Different Approaches to Investing in Treasury 
Securities over 2 Years: 

[Refer to PDF for image: illustration] 

2-year floating rate note: 
Maturity: 2 years; 
Interest rate: Set at purchase and reset every week (103 interest 
resets in a 2-year period). 

Series of 13-week Treasury bills: 
Maturity: 13 weeks; 
Interest rate: Set at purchase every 13 weeks (8 purchases in a 2-year 
period). 

2-year fixed-rate note: 
Maturity: 2 years; 
Interest rate: Set at purchase and fixed until maturity. 

Source: GAO presentation of Treasury information. 

Note: Although the interest rate on the FRN is reset daily, the 
interest rate paid on the FRN will likely only change weekly because 
its index rate, the 13-week bill rate, is currently auctioned weekly. 

[End of figure] 

Our analysis used two models for how the FRN spread--the spread 
between the index rate and the interest rate for the FRN--may vary 
over time. The FRN spread is set at auction and is expected to vary in 
response to changes in the level and volatility of interest rates. 
Because there is uncertainty about how market participants will price 
the FRN relative to other products, we considered two different models 
of the response of spreads to changes in different interest rates: 

* A "maturity-based" model, where the spread estimate is influenced by 
the 2-year term of the FRN. 

* A "reset-based" model, where the spread estimate is derived from the 
weekly reset term, which determines the nature of the interest rate 
risk faced by investors in FRNs. 

These two models are designed to approximate the range of potential 
spreads at which the 2-year FRN would have been expected to have been 
auctioned in historical interest rate environments. For more details 
on our models for FRN cost, including other models we considered, see 
appendix I. 

Because interest rate environments vary substantially over time, we 
also compared how the cost of FRNs may vary based on changes in the 
level and volatility of interest rates.[Footnote 11] Although these 
views are not generalizable, market participants and experts we 
interviewed expect the demand for FRNs to vary based on the interest 
rate environment. In addition, 58 of 62 respondents to our survey 
indicated that FRNs would be more attractive when interest rates are 
expected to rise; 49 of 62 indicated that FRNs would be less 
attractive when interest rates are expected to fall. 

Costs of 2-Year FRNs Vary Based on What Treasury Would Have Issued 
Instead, Investor Treatment of FRNs, and Interest Rate Environment: 

We found that compared to 2-year fixed-rate notes, FRNs are likely to 
result in interest savings to Treasury regardless of how the FRN is 
treated by market participants; however, compared to 13-week bills, 
they could result in either savings or additional costs (see figure 
6). Compared to 2-year fixed-rate notes, 2-year FRNs historically 
would have saved between $8.1 million in interest costs annually per 
billion in issuance under our maturity-based model, and $13.6 million 
under our reset-based model. Compared to 13-week bills, the FRN would 
have resulted in annual savings of $2.4 million per billion of 
issuance under our reset-based model but additional annual costs of 
$3.1 million per billion of issuance under our maturity-based model. 

Figure 6: Interest Savings or Added Costs from 2-Year Floating Rate 
Notes Compared to 2-Year Notes and 13-Week Bills, by Rate Environment: 

[Refer to PDF for image: 2 illustrated graphs] 

Graphs depicts the following, in millions: 

Actual results: 
Actual 2-year fixed-rate note; 
Actual 13-week bills. 

FRN modeled results: 
Maturity-based FRN model; 
Reset-based FRN model. 

Interest savings or added costs per year from floating rate notes 
(FRNs) compared to 2-year fixed-rate notes per $1 billion of
issuance: 2-year fixed-rate note: 

Rate environment: 
Overall (n=387): 
Rising (n=89): 
Steady (n=83): 
Falling (n=143): 
Trough (n=25): 
Peak (n=47): 

Interest savings or added costs per year from floating rate notes 
(FRNs) compared to 13-week bills per $1 billion of issuance: 13-week 
bills: 

Rate environment: 
Overall (n=387): 
Rising (n=89): 
Steady (n=83): 
Falling (n=143): 
Trough (n=25): 
Peak (n=47): 

Source: GAO simulation based on Treasury and Federal Reserve data. 

[End of figure] 

In addition to examining estimates of the relative savings and costs 
from issuing 2-year FRNs, we also analyzed the share of cases in our 
simulations where FRNs save or add to interest costs across different 
interest rate environments (see figure 7). We found that compared to 2-
year fixed-rate notes, the 2-year FRN would have resulted in savings 
in 82 percent of cases under our reset-based model and in 72 percent 
of cases under our maturity-based model. Compared to 13-week bills, 2-
year FRNs would have resulted in savings in 85 percent of cases under 
our reset-based model but added to costs in 81 percent of cases under 
our maturity-based model. 

Figure 7: Percent of Cases Where 2-Year Floating Rate Notes (FRN) Save 
or Add to Interest Costs Compared to 13-Week Bills and 2-Year Notes, 
by Rate Environment: 

[Refer to PDF for image: 4 horizontal bar graphs] 

Maturity-based FRN model compared to 2-year fixed-rate notes: 
Rate environment: 
Overall (n=387): 
Percentage of cases with interest savings: 72%; 
Percentage of cases with added interest costs: 28%; 
Rising (n=89): 
Percentage of cases with interest savings: 24%; 
Percentage of cases with added interest costs: 76%; 
Steady (n=83): 
Percentage of cases with interest savings: 95%; 
Percentage of cases with added interest costs: 5%; 
Falling (n=143): 
Percentage of cases with interest savings: 100%; 
Percentage of cases with added interest costs: 0%; 
Trough (n=25): 
Percentage of cases with interest savings: 100%; 
Percentage of cases with added interest costs: 0%; 
Peak (n=47): 
Percentage of cases with interest savings: 23%; 
Percentage of cases with added interest costs: 77%. 

Reset-based FRN model compared to 2-year fixed-rate notes: 
Rate environment: 
Overall (n=387): 
Percentage of cases with interest savings: 82%; 
Percentage of cases with added interest costs: 18%; 
Rising (n=89): 
Percentage of cases with interest savings: 48%; 
Percentage of cases with added interest costs: 52%; 
Steady (n=83): 
Percentage of cases with interest savings: 96%; 
Percentage of cases with added interest costs: 4%; 
Falling (n=143): 
Percentage of cases with interest savings: 100%; 
Percentage of cases with added interest costs: 0%; 
Trough (n=25): 
Percentage of cases with interest savings: 100%; 
Percentage of cases with added interest costs: 0%; 
Peak (n=47): 
Percentage of cases with interest savings: 60%; 
Percentage of cases with added interest costs: 40%. 

Maturity-based FRN model compared to 13-week bills: 
Rate environment: 
Overall (n=387): 
Percentage of cases with interest savings: 19%; 
Percentage of cases with added interest costs: 81%; 
Rising (n=89): 
Percentage of cases with interest savings: 1%; 
Percentage of cases with added interest costs: 99%; 
Steady (n=83): 
Percentage of cases with interest savings: 4%; 
Percentage of cases with added interest costs: 96%; 
Falling (n=143): 
Percentage of cases with interest savings: 42%; 
Percentage of cases with added interest costs: 58%; 
Trough (n=25): 
Percentage of cases with interest savings: 20%; 
Percentage of cases with added interest costs: 80%; 
Peak (n=47): 
Percentage of cases with interest savings: 11%; 
Percentage of cases with added interest costs: 89%. 

Reset-based FRN model compared to 13-week bills: 
Rate environment: 
Overall (n=387): 
Percentage of cases with interest savings: 85%; 
Percentage of cases with added interest costs: 15%; 
Rising (n=89): 
Percentage of cases with interest savings: 58%; 
Percentage of cases with added interest costs: 42%; 
Steady (n=83): 
Percentage of cases with interest savings: 88%; 
Percentage of cases with added interest costs: 12%; 
Falling (n=143): 
Percentage of cases with interest savings: 97%; 
Percentage of cases with added interest costs: 3%; 
Trough (n=25): 
Percentage of cases with interest savings: 84%; 
Percentage of cases with added interest costs: 16%; 
Peak (n=47): 
Percentage of cases with interest savings: 94%; 
Percentage of cases with added interest costs: 6%. 

Source: GAO simulation based on Treasury and Federal Reserve data. 

[End of figure] 

We also found that the interest savings or added costs from 2-year 
FRNs varied with the interest rate environment regardless of how the 
FRN is treated or whether it is being compared to 2-year fixed-rate 
notes or 13-week bills. Relative to 2-year fixed-rate notes, FRNs 
tended to be more costly in rising rate environments compared to other 
environments. Compared to 13-week bills, FRNs tended to be more costly 
(in the case of our maturity-based model) or to produce less savings 
(in the case of our reset-based model). The extra cost or reduced 
savings in rising rate environments, however, tended to be less than 
the savings in steady and falling rate environments. As shown in 
figures 6 and 7 above, under our maturity-based model: 

* in rising rate environments, 2-year FRNs were less costly than 2-
year fixed-rate notes in only 24 percent of cases and, on average, 
increased Treasury interest costs by 0.48 percentage points, resulting 
in $4.8 million in annual interest costs per billion in issuance; and: 

* in falling rate environments, 2-year FRNs were less costly than 2-
year fixed-rate notes in all cases and, on average, reduced interest 
costs by 2.07 percentage points, resulting in $20.7 million in annual 
interest savings per billion in issuance. 

We also analyzed the potential costs and savings from FRNs in 
environments with different levels of rate volatility and found that, 
at all levels of volatility, there was little variation between our 
two models. In periods of low, moderate, and high volatility, 2-year 
FRNs tended to produce savings compared to 2-year fixed-rate notes, 
but compared to 13-week bills, could produce either costs or savings, 
depending on which model is used. In periods of extreme (i.e., higher 
than "high") volatility, FRNs produced savings under both models. For 
more information on the results of this analysis, see appendix I. 

Technical Demand May Increase Savings to Treasury, While a Liquidity 
Premium May Increase Costs: 

Factors other than interest rates may affect demand for FRNs, and 
Treasury could realize additional savings from FRNs due to these 
elements of technical demand. Both of the models we used to estimate 
the cost of FRNs assume the FRN spread is based solely on the relative 
value of FRNs compared to other Treasury securities. However, both our 
interviews with market participants and our survey responses indicate 
that demand for FRNs is also likely to be affected by technical 
factors, such as investment guidelines or regulatory requirements to 
hold certain types of investments. For example, Treasury officials and 
market participants told us that Treasury structured the FRNs in a way 
that makes them especially attractive to money market investors. 
[Footnote 12] To meet investment guidelines and regulatory 
requirements, these funds tend to hold mostly short-term securities 
like Treasury bills and, because their interest rate resets 
frequently, FRNs.[Footnote 13] These factors would create some 
technical demand for Treasury FRNs that is less sensitive to the 
relative value of the FRN. This generally would lower Treasury's costs 
since some investors would be willing to accept a lower interest rate 
at auction. 

Side bar: 

Technical Demand: 
Technical demand is driven by factors such as investment guidelines or 
regulatory requirements and is less sensitive to the relative value of 
the security. 

[End of side bar] 

Our survey results confirm that technical factors affect the 
attractiveness of FRNs for at least some investors. Twenty-seven of 
the sixty-two survey respondents said that FRNs' consistency with 
client or fund investment guidelines make them attractive to a great 
or very great extent. Results of our survey also show that 2-year FRNs 
are more attractive because they conform to regulatory requirements 
for certain sectors. Six of the seven money market mutual fund 
managers that responded to our survey indicated that conformance with 
limits on their holdings make the FRNs attractive to a great or very 
great extent. Similarly, five of the nine retail and commercial banks 
that responded to our survey indicated that conformance with new 
capital requirements made the FRNs attractive to a great or very great 
extent. 

Treasury's costs could be increased if Treasury FRNs have a higher 
liquidity premium than other Treasury securities. Debt issuers, 
including Treasury, generally have to pay a liquidity premium on less 
liquid products--products that cannot be easily bought and sold in 
large volumes without meaningfully affecting the price--to compensate 
investors for the possibility that they might not be able to sell the 
security as readily as a more liquid product. A liquidity premium on 
FRNs that is greater than the premium on other Treasury securities 
could increase costs compared to our estimates. Although Treasury 
securities are generally considered very liquid and have very low 
liquidity premiums, market participants we interviewed said that FRNs 
might be less liquid than bills--Treasury's most liquid product--but 
more liquid than TIPS--its least liquid product.[Footnote 14] FRNs are 
expected to be less liquid than bills because (1) investors are more 
likely to buy and hold rather than to trade FRNs, and (2) FRNs are 
expected to have a smaller relative market size. Several market 
participants said that liquidity is likely to be lower initially and 
to improve as Treasury issues more FRNs. 

Results of Initial FRN Auctions Were Within the Range Estimated by Our 
Models: 

The results of Treasury's first three FRN auctions were within the 
range estimated by our models. At the first FRN auction in January 
2014, FRNs were auctioned with an FRN spread of 0.045 percentage 
points. At the February and March 2014 auctions, FRNs were auctioned 
with discount margins of 0.064 and 0.069 percentage points, 
respectively.[Footnote 15] The actual auction results appear linked to 
the spreads predicted by our reset-based model. In each of the three 
auctions, the actual auction results equaled the spread predicted by 
our reset-based model plus a small and consistent premium. 

Mismatch Between the FRN's Index Rate Maturity and Reset Frequency 
Poses Risk That Treasury Has Not Fully Analyzed: 

One element of the design of the Treasury 2-year FRN is that it is 
what the market refers to as a "mismatched floater." The difference 
(i.e., the mismatch) between the term of its index rate (13 weeks) and 
the length of its reset period (stated as daily, but effectively 
weekly) may introduce the risk of price instability on the reset date 
that is not typical of most floating rate securities. This is 
particularly the case if market participants treat the FRNs more like 
series of rolling 1-week bills. This might affect demand for the 
product in certain interest rate environments and, if so, could raise 
Treasury's borrowing costs. 

For most floating rate securities, the maturity of the index rate and 
the frequency of the interest rate reset match. For example, a 
floating rate note indexed to the 3-month LIBOR--an interbank lending 
rate that is the most common index for non-Treasury floating rate 
notes--would typically reset every 3 months. Absent a change in the 
credit risk of an issuer, the value of a typical floating rate 
security returns to par--the value at maturity--at each reset. This 
leads to a higher level of price stability in floating rate securities 
compared to fixed-rate securities of the same maturity. This price 
stability is highly desirable to some investors. 

Side bar: 

Yield Curve Risk: 
The risk that interest rates at different maturity points--for example 
the rates for a one-week bill and a 13-week bill--will change relative 
to one another. 

[End of side bar] 

The Treasury 2-year FRN is different from a typical floating rate 
security in that it will reset every week to a 13-week rate. This 
mismatch introduces a tradeoff between yield curve risk and interest 
rate risk. Unlike a typical FRN, the price of the Treasury 2-year FRN 
will not reliably return precisely to par at each reset date before 
its 2 year maturity. This is because investors factor in changes 
between the 1-week bill rate and the 13-week bill rate. However, the 
price of the Treasury 2-year FRN should return close to par weekly, 
which is more frequent than if it had a 13-week reset. Treasury 
officials told us they believe that the frequent resets provide 
increased price stability for the FRN. They said that they expect 
investors to price the 2-year FRN in a way that reflects the 
expectation that the yield curve risk for Treasury's 2-year FRN is 
likely to be small relative to its reduced interest rate risk. 
However, if the difference between the 1-week rate and the 13-week 
rate changes substantially over the two year term, either in fact or 
in expectations, then the yield curve risk that the investor faces 
would be more substantial. It is possible that in higher and changing 
interest rate environments, the tradeoff between yield curve risk and 
interest rate risk may not be favorable to investors. This could be 
reflected in the spread, as investors bid for FRNs at auction in a way 
that compensates them for this additional risk, which could raise 
Treasury's borrowing costs. 

Side bar: 

Money Market Fund: 
A money market fund is a type of investment fund that is required by 
law to invest in low-risk securities. These funds have relatively low 
risks compared to other mutual funds and pay dividends that generally 
reflect short-term interest rates. Money market funds typically invest 
in government securities (including Treasury bills and notes), 
certificates of deposit, commercial paper of companies, or other 
highly liquid and low-risk securities. 

[End of side bar] 

The mismatch between the index rate maturity and the frequency of the 
interest rate reset could have adverse effects on the costs of FRNs to 
Treasury. Treasury officials told us they discussed the design of the 
2-year FRN both internally and with market participants and structured 
the 2-year FRN in this way for two reasons. First, as both those who 
commented on Treasury's proposal and Treasury have noted, the 13-week 
bill market is a large, liquid, and transparent market. Second, 
Treasury designed the 2-year FRN to meet high demand for short-term 
securities, and both Treasury officials and the market participants we 
spoke with cited the 2-year FRN's frequent reset as a reason for 
greater demand from money market funds. These funds face constraints 
on the average maturity of their holdings, which the weekly reset of 
the Treasury 2-year FRN helps address. This additional demand would 
likely result in lower costs and helps establish the new product for 
Treasury, which may outweigh the potential cost of the mismatch. 
Results of our survey show that overall, the FRN's index rate and the 
frequency of its interest rate reset chosen by Treasury--as well as 
the difference between the two--made the FRN more attractive to 
investors (see table 1). 

Table 1: Effect of Floating Rate Note Design Features on 
Attractiveness to Investors: 

Feature: Index rate maturity; 
Effect on attractiveness of FRN: 
More attractive: 24; 
Neither more nor less attractive: 28; 
Less attractive: 6; 
No basis to judge/no opinion: 4. 

Feature: Interest rate reset frequency; 
Effect on attractiveness of FRN: 
More attractive: 34; 
Neither more nor less attractive: 12; 
Less attractive: 12; 
No basis to judge/no opinion: 4. 

Feature: Difference between index rate maturity and interest rate 
reset period; 
Effect on attractiveness of FRN: 
More attractive: 16; 
Neither more nor less attractive: 34; 
Less attractive: 7; 
No basis to judge/no opinion: 5. 

Source: GAO analysis of survey data. 

[End of table] 

Although Treasury officials told us they discussed the potential 
benefits and risks of the mismatch, Treasury had not analyzed how the 
mismatch could affect pricing. After we briefed Treasury officials on 
the issue in April 2014, Treasury began taking steps to study the 
mismatch to more fully understand its potential pricing risks. While 
its practice of regular and predicable issuance means Treasury issues 
all products in all environments, it is important that the risks of 
different securities are considered when making decisions about the 
mix of securities to issue. Treasury did analyze and consider how 
other design elements would affect pricing of the 2-year FRN and 
incorporated the results of that analysis into their final design. For 
example, Treasury analyzed how setting a minimum spread for the FRN 
would affect pricing. This analysis led Treasury officials to conclude 
that a minimum spread would unnecessarily complicate pricing, and it 
was excluded from the final structure of the FRN. 

FRNs Can Help to Extend the Maturity of the Debt Portfolio, but They 
Make Treasury's Weighted Average Maturity an Incomplete Measure of 
Rollover Risk: 

Consistent with its goal of borrowing at the lowest cost over time, 
Treasury tracks and manages the average maturity of the debt 
portfolio. Relative to issuing bills, 2-year FRNs will help Treasury 
extend the average maturity of the debt portfolio, but only slightly. 
The effect that the FRNs will have on the average maturity is likely 
to be small because at least at the onset, FRNs are expected to 
constitute a small proportion of overall Treasury debt. Even so, 
Treasury officials said it is prudent to reduce Treasury's rollover 
risk by extending the average maturity. Treasury officials told us 
that FRNs will help Treasury continue to increase the maturity profile 
of the debt portfolio while meeting high demand for high-quality, 
short-term securities. Treasury could extend the average maturity of 
the portfolio by replacing issuance of shorter term notes and bills 
with longer term fixed-rate notes and bonds, rather than issue FRNs. 
In deciding what to issue, however, Treasury is confronted with making 
prudent decisions about investor demand by product. If Treasury issues 
the wrong mix of products, its overall cost of funding would increase, 
as investors would express their preferences in prices bid at auction. 

Side bar: 

Weighted Average Maturity (WAM): 
The WAM of outstanding marketable Treasury securities is calculated by 
averaging the remaining maturity of all outstanding marketable 
Treasury securities, weighted by the dollar value of the securities. 

Rollover Risk: 
Rollover risk includes two types of risk: 
1. Interest rate risk; For a borrower, such as Treasury, interest rate 
risk is the risk of having to refinance its debt at less favorable 
interest rates and, for floating rate debt, of interest rates rising 
during the life of the security; 
2. Market access risk; The risk associated with coming back to the 
market to refinance the debt. In times of federal budget deficits, 
maturing federal debt must be rolled over into new issuance. 

Marketable Debt: 
Marketable securities can be resold by whoever owns them. In addition 
to marketable securities, Treasury issues nonmarketable securities 
that cannot be resold, such as U.S. savings bonds and special 
securities for state and local governments. 

[End of side bar] 

Treasury tracks the WAM of outstanding marketable securities and 
publicly releases WAM data quarterly. Treasury debt managers do not 
have a WAM target, but over the past 30 years they have generally kept 
the WAM between 50 and 70 months (see figure 8). As of February 28, 
2014, the WAM of the Treasury's outstanding marketable debt was 67 
months, well above the historical average of 58.6 months. As of 
January 2014, Treasury continued to increase the WAM in a way that 
Treasury officials stated is consistent with their long-term 
objectives of financing the government at the lowest cost over time 
and ensuring regular and predictable management of the debt portfolio. 

Figure 8: Weighted Average Maturity of Treasury's Marketable Debt 
Outstanding: 

[Refer to PDF for image: line graph] 

Values shown are for January of each fiscal year. 

Fiscal year: 1980; 
Weighted average maturity (WAM): 44.2 months; 
Average WAM: 58.6 months. 

Fiscal year: 1981; 
Weighted average maturity (WAM): 43.9 months; 
Average WAM: 58.6 months. 

Fiscal year: 1982; 
Weighted average maturity (WAM): 45.4 months; 
Average WAM: 58.6 months. 

Fiscal year: 1983; 
Weighted average maturity (WAM): 43.3 months; 
Average WAM: 58.6 months. 

Fiscal year: 1984; 
Weighted average maturity (WAM): 46.4 months; 
Average WAM: 58.6 months. 

Fiscal year: 1985; 
Weighted average maturity (WAM): 49.7 months; 
Average WAM: 58.6 months. 

Fiscal year: 1986; 
Weighted average maturity (WAM): 54.6 months; 
Average WAM: 58.6 months. 

Fiscal year: 1987; 
Weighted average maturity (WAM): 58.2 months; 
Average WAM: 58.6 months. 

Fiscal year: 1988; 
Weighted average maturity (WAM): 61.9 months; 
Average WAM: 58.6 months. 

Fiscal year: 1989; 
Weighted average maturity (WAM): 62.3 months; 
Average WAM: 58.6 months. 

Fiscal year: 1990; 
Weighted average maturity (WAM): 64.9 months; 
Average WAM: 58.6 months. 

Fiscal year: 1991; 
Weighted average maturity (WAM): 64.4 months; 
Average WAM: 58.6 months. 

Fiscal year: 1992; 
Weighted average maturity (WAM): 65.1 months; 
Average WAM: 58.6 months. 

Fiscal year: 1993; 
Weighted average maturity (WAM): 64.6 months; 
Average WAM: 58.6 months. 

Fiscal year: 1994; 
Weighted average maturity (WAM): 62.4 months; 
Average WAM: 58.6 months. 

Fiscal year: 1995; 
Weighted average maturity (WAM): 60.0 months; 
Average WAM: 58.6 months. 

Fiscal year: 1996; 
Weighted average maturity (WAM): 58.2 months; 
Average WAM: 58.6 months. 

Fiscal year: 1997; 
Weighted average maturity (WAM): 59.3 months; 
Average WAM: 58.6 months. 

Fiscal year: 1998; 
Weighted average maturity (WAM): 61.5 months; 
Average WAM: 58.6 months. 

Fiscal year: 1999; 
Weighted average maturity (WAM): 65.8 months; 
Average WAM: 58.6 months. 

Fiscal year: 2000; 
Weighted average maturity (WAM): 68.1 months; 
Average WAM: 58.6 months. 

Fiscal year: 2001; 
Weighted average maturity (WAM): 68.6 months; 
Average WAM: 58.6 months. 

Fiscal year: 2002; 
Weighted average maturity (WAM): 66.1 months; 
Average WAM: 58.6 months. 

Fiscal year: 2003; 
Weighted average maturity (WAM): 60.4 months; 
Average WAM: 58.6 months. 

Fiscal year: 2004; 
Weighted average maturity (WAM): 56.6 months; 
Average WAM: 58.6 months. 

Fiscal year: 2005; 
Weighted average maturity (WAM): 54.7 months; 
Average WAM: 58.6 months. 

Fiscal year: 2006; 
Weighted average maturity (WAM): 53.3 months; 
Average WAM: 58.6 months. 

Fiscal year: 2007; 
Weighted average maturity (WAM): 54.7 months; 
Average WAM: 58.6 months. 

Fiscal year: 2008; 
Weighted average maturity (WAM): 55.8 months; 
Average WAM: 58.6 months. 

Fiscal year: 2009; 
Weighted average maturity (WAM): 50.3 months; 
Average WAM: 58.6 months. 

Fiscal year: 2010; 
Weighted average maturity (WAM): 56.4 months; 
Average WAM: 58.6 months. 

Fiscal year: 2011; 
Weighted average maturity (WAM): 59.2 months; 
Average WAM: 58.6 months. 

Fiscal year: 2012; 
Weighted average maturity (WAM): 62.9 months; 
Average WAM: 58.6 months. 

Fiscal year: 2013; 
Weighted average maturity (WAM): 65.0 months; 
Average WAM: 58.6 months. 

Fiscal year: 2014; 
Weighted average maturity (WAM): 67.2 months; 
Average WAM: 58.6 months. 

Source: GAO analysis of Treasury data. 

[End of figure] 

Before Treasury began issuing FRNs, its WAM metric captured both 
components of rollover risk--interest rate risk and market access 
risk. Increasing the WAM by issuing 2-year FRNs instead of bills does 
not have the same effect on Treasury's risk profile that issuing 2-
year fixed-rate notes would have. Since both 2-year securities have 
the same final maturity, they carry the same market access risk. The 
FRNs, however, carry a larger interest rate risk. The weekly reset 
means the interest risk on the FRN is similar to that of a 1-week 
bill, making it the shortest term product in Treasury's regularly 
issued portfolio from an interest rate risk perspective.[Footnote 16] 
As noted above, it is uncertain what Treasury would issue in the 
absence of FRNs and whether 2-year FRNs will substitute primarily for 
2-year fixed-rate notes, other fixed-rate notes, or bills. Relative to 
issuance of regularly issued bills, 2-year FRNs reduce Treasury's 
market access risk by locking in that funding for a longer period but 
increase interest rate risk due to the weekly reset. 

As FRN issuance grows, Treasury's single WAM metric will remain a 
meaningful measure of market access risk; however it will no longer be 
an accurate measure of interest rate risk because floating rate 
securities carry higher interest rate risk than fixed-rate securities 
of the same term. Treasury debt management officials said that because 
Treasury issues in all interest rate environments and does not time 
the market, they primarily use the WAM metric as a proxy for market 
access risk in the portfolio. However, because market access risk and 
interest rate risk are related, Treasury understands the importance of 
both parts of the measure. Managing interest rate risk is an important 
aspect of Treasury's goal to borrow at the lowest cost over time; 
guidelines for sovereign debt management also emphasize the importance 
of managing the exposure of the debt portfolio to interest rate risk, 
as well as to market access risk.[Footnote 17] Market participants 
that hold both floating rate and fixed rate debt use more than one 
measure to assess the length of their portfolios. For example, certain 
money market mutual funds are required to measure the average term of 
the fund in two ways--one based on the final maturity dates of the 
securities in the fund and the other based on the interest rate reset 
dates of the securities in the fund.[Footnote 18] When we raised this 
issue with them, Treasury officials told us that they could easily 
publish another measure of interest rate risk. 

FRNs Provide Treasury with Additional Flexibility in Its Debt Issuance: 

FRNs provide Treasury with additional flexibility in its debt issuance 
by adding a new type of security to Treasury's debt portfolio and by 
increasing overall demand for Treasury securities. If a new security 
brings incremental demand for Treasury securities, Treasury can grow 
its debt portfolio without increasing by as much as it might otherwise 
have had to the amount needed to finance the debt through existing 
securities. Our interviews and survey results found that although 
market participants will likely primarily purchase Treasury FRNs as a 
substitute for other Treasury securities (especially bills), market 
participants will also purchase Treasury FRNs as a substitute for 
other investment options, including FRNs from other issuers and 
repurchase agreements (see figure 9). 

Figure 9: Extent to Which Treasury Floating Rate Note Purchases Will 
Substitute for Other Investments, According to Survey Respondents: 

[Refer to PDF for image: illustrated table] 

Treasury bills: 
Very great extent: 8; 
Great extent: 11; 
Moderate extent: 10; 
Some extent: 14; 
Little or no extent: 18; 
No opinion or no basis to judge: 1. 

Treasury notes: 
Very great extent: 2; 
Great extent: 2; 
Moderate extent: 10; 
Some extent: 18; 
Little or no extent: 27; 
No opinion or no basis to judge: 2. 

Floating rate notes from other issuers: 
Very great extent: 2; 
Great extent: 2; 
Moderate extent: 9; 
Some extent: 21; 
Little or no extent: 26; 
No opinion or no basis to judge: 2. 

Repurchase agreements: 
Very great extent: 1; 
Great extent: 3; 
Moderate extent: 10; 
Some extent: 9; 
Little or no extent: 32; 
No opinion or no basis to judge: 7. 

Commercial paper: 
Very great extent: 3; 
Great extent: 0; 
Moderate extent: 6; 
Some extent: 16; 
Little or no extent: 29; 
No opinion or no basis to judge: 8. 

Bank deposits: 
Very great extent: 1; 
Great extent: 1; 
Moderate extent: 5; 
Some extent: 9; 
Little or no extent: 36; 
No opinion or no basis to judge: 10. 

Money market fund shares: 
Very great extent: 1; 
Great extent: 2; 
Moderate extent: 6; 
Some extent: 7; 
Little or no extent: 34; 
No opinion or no basis to judge: 12. 

Derivatives: 
Very great extent: 1; 
Great extent: 0; 
Moderate extent: 0; 
Some extent: 5; 
Little or no extent: 40; 
No opinion or no basis to judge: 13. 

Source: GAO analysis of survey data. 

[End of figure] 

The results of the early FRN auctions indicate high demand for the 
security. The bid-to-cover ratios at the first three FRN auctions were 
higher than bid-to-cover ratios for other Treasury securities 
auctioned around that time. The first auction on January 29, 2014 had 
a bid-to-cover ratio of 5.67 and the reopening auctions in February 
and March had bid-to-cover ratios of 5.29 and 4.67, respectively. This 
compares to the bid-to-cover ratios in 13-week bill and fixed-rate 2-
year note auctions averaging 4.53 and 3.37, respectively, over the 
same January to March period. According to Federal Reserve Bank of New 
York (FRBNY) officials, trading for the FRNs in the when-issued market 
was limited ahead of the three FRN auctions; nevertheless, the rates 
quoted in the when-issued market were very close to the auction 
results, an indicator that the auctions came very close to market 
expectations. This suggests that the price discovery mechanism of the 
market was functioning well for FRNs and that the market embraces and 
understands the security, which in turn indicates strong current and 
continuing demand that helps Treasury borrow at lower cost over time. 

Side bar: 

Bid-to-Cover Ratio: 
In a Treasury auction the bid-to-cover ratio is the dollar value of 
all bids received in the auction, divided by the dollar value of the 
securities auctioned. 

Reopening Auction: 
An auction of additional amounts of a previously issued security. 

When-Issued Market: 
When-issued trades are contracts for the purchase and sale of a new 
security before the security has been auctioned. When-issued trades 
settle on the issue date of the new security, when the security is 
first available for delivery. 

[End of side bar] 

Our survey results suggest demand for Treasury FRNs is likely to grow. 
Eighteen out of 61 survey respondents participated in the first 
Treasury FRN auction, but more said they plan to purchase Treasury 
FRNs this year. About half of all respondents (32 of 62) said their 
organizations definitely or probably will purchase Treasury FRNs in 
2014.[Footnote 19] Survey respondents anticipate that money market 
mutual funds, corporate treasuries, and foreign central banks are 
likely to have the most demand for 2-year FRNs. Survey respondents 
noted a number of reasons why Treasury FRNs are an attractive 
investment option, including the interest rate risk protection they 
provide the purchaser, their price stability, their use as a cash 
management tool, their consistency with investment guidelines and 
regulatory requirements, and the liquidity of the securities. The 
successful launch of a new type of security relies both on the 
readiness of investors and on Treasury's own operational readiness. 
Overall, market participants felt prepared for the introduction of a 
new security. According to almost all of the market participants we 
surveyed, Treasury provided sufficient information regarding its plans 
to issue FRNs (53 out of 62 respondents noted that Treasury provided 
sufficient information and the remaining 9 noted that they had "no 
opinion or no basis to judge.") In addition, of the 48 survey 
respondents that said that they would need to make systems changes to 
purchase FRNs, 36 said that Treasury or the Federal Reserve had 
provided adequate assistance or information to make the necessary 
changes. Some respondents noted that as of March 2014, they had not 
yet completed systems changes that will be needed to purchase FRNs. 
Demand for FRNs may increase as additional investors complete systems 
changes. 

Although issuance of FRNs brings incremental demand for Treasury 
securities and demand in the initial auctions was high and is likely 
to grow, one design feature of the 2-year FRN may constrain Treasury's 
flexibility in the issuance of 13-week bills. Treasury officials and 
market participants both told us that because the FRN is indexed to 
the 13-week Treasury bill rate, Treasury will have to be more 
judicious in adjusting the size or timing of Treasury auctions of 13-
week bills. As some comments on the proposed rule noted, there is some 
risk in indexing a floating rate note to a product from the same 
issuer. However, given that the 13-week bill is one of Treasury's 
largest and most liquid markets, its selection as the index rate 
minimizes this risk. 

Market Participants Identified Opportunities for Treasury to Enhance 
Investor Input and Expand Product Offerings: 

Overall Treasury's Communication with Investors Is Strong, but Certain 
Sectors Said It Could Be Improved: 

As our prior work has found, communication with investors is essential 
as Treasury faces the need to finance historically large deficits 
expected in the medium and long term.[Footnote 20] Overall, survey 
respondents said that Treasury provides sufficient information to 
investors on its debt management plans. Forty-three out of the 62 
survey respondents said communication from Treasury occurred to a 
great or very great extent; no respondents said communication occurred 
to little or no extent (one had no basis to judge). In addition, most 
survey respondents said that they were able to provide sufficient 
input to Treasury, but respondents from some sectors reported lower 
levels of opportunity to provide input. The 26 respondents who 
reported opportunities existed to some or little to no extent included 
10 state or local government retirement fund managers, 4 money market 
mutual fund managers, and 3 life insurance providers (see figure 10 
below). 

Figure 10: Extent to Which Market Participants Said They Have 
Opportunities to Provide Input to Treasury: 

[Refer to PDF for image: illustrated table] 

Money market mutual fund managers: 
Very great extent: 0; 
Great extent: 1; 
Moderate extent: 2; 
Some extent: 3; 
Little or no extent: 1; 
No opinion or no basis to judge: 0. 

Mutual fund and exchange-traded fund managers: 
Very great extent: 1; 
Great extent: 1; 
Moderate extent: 3; 
Some extent: 2; 
Little or no extent: 0; 
No opinion or no basis to judge: 1. 

State and local government retirement fund managers: 
Very great extent: 0; 
Great extent: 1; 
Moderate extent: 5; 
Some extent: 8; 
Little or no extent: 2; 
No opinion or no basis to judge: 3. 

Retail and commercial banks: 
Very great extent: 1; 
Great extent: 2; 
Moderate extent: 2; 
Some extent: 4; 
Little or no extent: 0; 
No opinion or no basis to judge: 0. 

Life insurance providers: 
Very great extent: 0; 
Great extent: 1; 
Moderate extent: 1; 
Some extent: 2; 
Little or no extent: 1; 
No opinion or no basis to judge: 0. 

Property-casualty insurance providers: 
Very great extent: 1; 
Great extent: 0; 
Moderate extent: 3; 
Some extent: 3; 
Little or no extent: 0; 
No opinion or no basis to judge: 0. 

Securities broker-dealers: 
Very great extent: 1; 
Great extent: 6; 
Moderate extent: 0; 
Some extent: 0; 
Little or no extent: 0; 
No opinion or no basis to judge: 0. 

Total responses: 
Very great extent: 4; 
Great extent: 12; 
Moderate extent: 16; 
Some extent: 22; 
Little or no extent: 4; 
No opinion or no basis to judge: 4. 

Source: GAO analysis of survey data. 

[End of figure] 

To manage risks associated with borrowing, Treasury monitors market 
activity and, if necessary, responds with appropriate changes in debt 
issuance based on analysis and consultation with market participants. 
Treasury offers a number of ways for market participants to give 
input, such as providing comments on regulations solicited through the 
Federal Register and through the email box on the Treasury website. 
The Treasury Borrowing Advisory Committee (TBAC) is comprised of 
senior representatives from investment funds and banks and holds 
quarterly meetings to provide insights to Treasury on the overall 
strength of the U.S. economy and recommendations on debt management 
issues. In addition, FRBNY administers the network of primary dealers 
that also provide market information and analysis to Treasury. 
However, Treasury's Office of Debt Management does not meet regularly 
with all sectors, such as state and local government retirement fund 
managers. Survey respondent suggestions for improving communication 
with Treasury included administering surveys, holding regular meetings 
or calls with investors outside of the TBAC, polling investors on new 
product ideas, and providing a mechanism for submitting annual 
recommendations to Treasury from large investors. Without targeted 
outreach to all major sectors of investors in Treasury securities, 
Treasury could miss important insights to improve its debt management 
plans. 

Survey Respondents Reported Interest Both in FRNs with Different 
Maturities and Other New Types of Treasury Securities: 

Responses from our survey of market participants indicate an interest 
in FRNs of both shorter-and medium-term maturities, but respondents 
expressed more limited interest in 7-and 10-year FRNs than in shorter-
term FRNs (see figure 11). Survey respondents expressed the most 
interest in the introduction of a 1-year FRN. Interest in the 1-year 
FRN varied by sector, with mutual funds (including money market funds) 
expressing substantial interest in this maturity, while retail and 
commercial banks had little interest. Securities broker-dealers and 
state and local retirement fund managers expressed the most interest 
in FRNs with maturities other than 2 years, but other sectors--such as 
banks and property-casualty insurance providers--also showed some 
interest in these other securities. Treasury officials said they might 
consider issuing FRNs with longer maturities once both they and the 
market gain some experience with the 2-year Treasury FRN. Over the 
long run, Treasury FRNs with maturities other than 2 years are likely 
to provide a cost savings to Treasury relative to issuance of fixed-
rate securities with the same maturity. 

Figure 11: Market Participants' Reported Estimated Purchases of 
Floating Rate Notes with Different Maturities: 

[Refer to PDF for image: illustrated table] 

1-year floating: rate note: 
$0: 11; 
More than $0 to $1 billion: 25; 
More than $1 billion: 16; 
No basis to judge: 10. 

3-year floating rate note: 
$0: 19; 
More than $0 to $1 billion: 25; 
More than $1 billion: 3; 
No basis to judge: 13. 

5-year floating rate note: 
$0: 22; 
More than $0 to $1 billion: 24; 
More than $1 billion: 2; 
No basis to judge: 14. 

7-year floating rate note: 
$0: 29; 
More than $0 to $1 billion: 15; 
More than $1 billion: 2; 
No basis to judge: 16. 

10-year floating rate note: 
$0: 29; 
More than $0 to $1 billion: 13; 
More than $1 billion: 3; 
No basis to judge: 17. 

Source: GAO analysis of survey data. 

[End of figure] 

Survey respondents expressed their views on certain design features of 
FRNs with maturities other than 2 years. For instance, if Treasury 
were to issue FRNs with different maturities, almost all survey 
respondents (57 out of 62) thought those FRNs should also be indexed 
to the 13-week Treasury bill. More respondents said they would prefer 
daily interest rate resets to any other reset period for FRNs with 
maturities other than 2 years. Of the respondents who wanted new FRNs 
to be indexed to the 13-week Treasury bill rate, 13 would also prefer 
daily resets for all hypothetical maturities, including 4 state and 
local government retirement fund managers and 5 securities broker-
dealers. Although this suggests that these respondents would prefer a 
"mismatched floater," as discussed earlier in this report, the 
mismatch feature may raise risks that result in higher costs to 
Treasury in certain interest rate environments. Additionally, 
respondents generally preferred quarterly interest payments for FRNs 
with other maturities and monthly auctions for 1-year and 3-year FRNs 
and quarterly auctions for FRNs with other maturities. 

Survey respondents also expressed an interest in possible new Treasury 
securities. Suggestions were ultra-long bonds, callable securities, 
FRNs indexed to inflation, and zero-coupon notes or bonds (see figure 
12).[Footnote 21] 

Figure 12: Treasury Products that Could Enhance Investor Demand, 
According to Survey Respondents: 

[Refer to PDF for image: illustrated table] 

Money market mutual fund managers: 
Ultra-long bonds: 3; 
Callable securities: 4; 
Floating rate notes indexed to inflation: 5; 
Zero-coupon notes or bonds: 5. 

Mutual fund and exchange-traded fund managers: 
Ultra-long bonds: 4; 
Callable securities: 0; 
Floating rate notes indexed to inflation: 4; 
Zero-coupon notes or bonds: 0. 

State and local government retirement fund managers: 
Ultra-long bonds: 14; 
Callable securities: 10; 
Floating rate notes indexed to inflation: 14; 
Zero-coupon notes or bonds: 13. 

Retail and commercial banks: 
Ultra-long bonds: 5; 
Callable securities: 5; 
Floating rate notes indexed to inflation: 4; 
Zero-coupon notes or bonds: 3. 

Life insurance providers: 
Ultra-long bonds: 5; 
Callable securities: 1; 
Floating rate notes indexed to inflation: 3; 
Zero-coupon notes or bonds: 4. 

Property-casualty insurance providers: 
Ultra-long bonds: 6; 
Callable securities: 2; 
Floating rate notes indexed to inflation: 6; 
Zero-coupon notes or bonds: 4. 

Securities broker-dealers: 
Ultra-long bonds: 4; 
Callable securities: 3; 
Floating rate notes indexed to inflation: 1; 
Zero-coupon notes or bonds: 1. 

Total responses: 
Ultra-long bonds: 41; 
Callable securities: 25; 
Floating rate notes indexed to inflation: 37; 
Zero-coupon notes or bonds: 30. 

Source: GAO analysis of survey data. 

[End of figure] 

In addition, respondents suggested that certain debt management 
practices, specifically buybacks and reverse inquiry window, would 
enhance demand for Treasury securities.[Footnote 22] However, 
respondents said that in general, changes to Treasury's current debt 
management practices--such as frequency of initial and reopening 
auctions, issuance sizes, and non-competitive award limits--would not 
enhance demand (see figure 13). 

Figure 13: Treasury Debt Management Practices That Could Enhance 
Investor Demand, According to Survey Respondents: 

[Refer to PDF for image: illustrated table] 

Money market mutual fund managers: 
Buybacks: 2; 
Use of reverse inquiry window: 4; 
More frequent initial auctions: 1; 
More frequent reopenings: 0; 
Increase issuance sizes: 4; 
Increase limit on non-competitive awards in auctions: 1. 

Mutual fund and exchange-traded fund managers: 
Buybacks: 1; 
Use of reverse inquiry window: 1; 
More frequent initial auctions: 0; 
More frequent reopenings: 0; 
Increase issuance sizes: 1; 
Increase limit on non-competitive awards in auctions: 1. 

State and local government retirement fund managers: 
Buybacks: 9; 
Use of reverse inquiry window: 9; 
More frequent initial auctions: 1; 
More frequent reopenings: 1; 
Increase issuance sizes: 0; 
Increase limit on non-competitive awards in auctions: 1. 

Retail and commercial banks: 
Buybacks: 4; 
Use of reverse inquiry window: 4; 
More frequent initial auctions: 1; 
More frequent reopenings: 1; 
Increase issuance sizes: 1; 
Increase limit on non-competitive awards in auctions: 2. 

Life insurance providers: 
Buybacks: 1; 
Use of reverse inquiry window: 1; 
More frequent initial auctions: 0; 
More frequent reopenings: 0; 
Increase issuance sizes: 0; 
Increase limit on non-competitive awards in auctions: 0. 

Property-casualty insurance providers: 
Buybacks: 3; 
Use of reverse inquiry window: 3; 
More frequent initial auctions: 0; 
More frequent reopenings: 0; 
Increase issuance sizes: 2; 
Increase limit on non-competitive awards in auctions: 1. 

Securities broker-dealers: 
Buybacks: 3; 
Use of reverse inquiry window: 1; 
More frequent initial auctions: 0; 
More frequent reopenings: 0; 
Increase issuance sizes: 0; 
Increase limit on non-competitive awards in auctions: 0. 

Total responses: 
Buybacks: 23; 
Use of reverse inquiry window: 23; 
More frequent initial auctions: 3; 
More frequent reopenings: 2; 
Increase issuance sizes: 8; 
Increase limit on non-competitive awards in auctions: 6. 

Source: GAO analysis of survey data. 

[End of figure] 

To achieve the lowest cost of financing the government over time, it 
is important that Treasury spread debt across maturities and take into 
account investor demand for new and existing products. The medium and 
long term fiscal outlook make evaluating the demand for Treasury 
securities, including new securities, increasingly important. 
Currently, Treasury feels unable to conduct a broad survey of market 
participants. For this reason, the insights on potential demand for 
new products from our survey can provide Treasury with a starting 
point so that it does not miss opportunities. 

Conclusions: 

The U.S. Treasury market is the deepest and most liquid government 
debt market in the world. Nevertheless, Treasury faces challenges in 
managing the debt at a time when debt levels are high and projected to 
increase and when interest rates are also expected to rise. Given the 
market uncertainties and the federal government's fiscal challenges, 
increasing Treasury's flexibility to respond to changing market 
conditions in ways that minimize costs is prudent. FRNs are a tool 
that can help meet these goals. Over the long term, FRNs can reduce 
Treasury interest costs relative to fixed-rate securities that lock in 
funding for the same term. FRNs can also help enhance Treasury 
flexibility by marginally increasing demand for Treasury securities. 

The design and implementation of FRNs has implications for Treasury's 
ability to minimize borrowing costs over time and for the balance of 
risks in Treasury's debt portfolio. Our cost analysis finds that in 
comparison to issuance of 2-year fixed-rate notes, Treasury is taking 
on additional interest rate risk but is likely to achieve interest 
cost savings while not increasing market access risk. The mismatch 
feature of Treasury's first FRN presents a tradeoff between different 
risks for both investors and Treasury that could raise Treasury's 
borrowing costs when interest rates are high and the yield curve is 
volatile. However, the mismatch also helps Treasury tap into the 
current high demand for high-quality short-term securities. Without 
analyzing how the mismatch between the frequency of the reset period 
and the maturity of the index could affect pricing, however, Treasury 
is unable to judge either (1) the risks (and therefore the ultimate 
cost) of FRNs in a different interest environment, or (2) whether the 
additional demand from money market funds due to the mismatch feature 
outweighs the potential costs it creates. A better understanding of 
these tradeoffs will be important when Treasury considers issuing FRNs 
with maturities other than 2 years. Furthermore, with the addition of 
FRNs to Treasury's debt portfolio, the weighted average maturity 
length of securities in the portfolio (i.e., the WAM) is now an 
incomplete measure of rollover risk because it does not accurately 
measure interest rate risk. Tracking and reporting an additional 
measure of the length of the debt portfolio that captures interest 
rate risk could help Treasury debt managers understand and weigh risks 
in the portfolio, and publicly reporting that measure would facilitate 
transparency and market understanding of Treasury debt management 
decisions. 

Introducing FRNs at this time--when demand is high--can help Treasury 
and market participants become more familiar with the new security so 
that Treasury can expand to FRNs with different maturities if Treasury 
determines that doing so would enhance its flexibility and advance its 
debt management goals. It will also be important for Treasury to gauge 
market demand for FRNs and other products by soliciting input from all 
sectors of Treasury investors, specifically state and local government 
retirement fund managers. Such input can help inform Treasury 
decisions about changes to Treasury issuance or debt management 
practices that could enhance overall demand for Treasury securities. 
When deciding what to issue, Treasury must make prudent decisions 
about investor demand by product. If Treasury issues the wrong mix of 
products, its overall cost of funding will increase, as investors 
express their preferences in prices bid at auction. 

Recommendations for Executive Action: 

To help minimize Treasury borrowing costs over time by better 
understanding and managing the risks posed by Treasury floating rate 
notes and by enhancing demand for Treasury securities, we recommend 
that the Secretary of the Treasury take the following four actions: 

1. Analyze the price effects of the mismatch between the term of the 
index rate and the reset period; 

2. Track and report an additional measure of the length of the 
portfolio that captures the interest rate reset frequency of 
securities in the portfolio; 

3. Expand outreach to state and local government retirement fund 
managers; and: 

4. Examine opportunities for additional new security types, such as 
FRNs with maturities other than 2 years or ultra-long bonds. 

Agency Comments and Our Evaluation: 

We provided a draft of this product and the accompanying e-supplement 
(GAO-14-562SP) to Treasury for comment. On May 23, 2014 the Assistant 
Secretary for Financial Markets told us that Treasury thought it was 
an excellent report, that they agreed with the recommendations, and 
that they had already taken steps to begin implementing them. For 
example, he told us that Treasury's new Office of State and Local 
Finance will bolster outreach to investors in the state and local 
sectors. Treasury also provided technical comments that were 
incorporated as appropriate. Further, Treasury told us they had no 
comments on the e-supplement. 

As agreed with your office, unless you publicly announce the contents 
of this report earlier, we plan no further distribution until 7 days 
from the report date. At that time, we will send copies to the 
Secretary of the Treasury, the appropriate congressional committees, 
and other interested parties. In addition, the report will be 
available at no charge on the GAO website at [hyperlink, 
http://www.gao.gov]. 

Should you or your staff have any questions about this report, please 
contact me at (202) 512-6806 or irvings@gao.gov. Contact points for 
our Offices of Congressional Relations and Public Affairs may be found 
on the last page of this report. GAO staff who made major 
contributions to this report are listed in appendix III. 

Sincerely yours, 

Signed by: 

Susan J. Irving: 
Director for Federal Budget Analysis: 
Strategic Issues: 

[End of section] 

Appendix I: Simulations of Floating Rate Note Costs: 

To estimate the potential cost of floating rate notes (FRN) to 
Treasury, we simulated the costs of 2-year FRNs based on Department of 
the Treasury (Treasury) auction data from January 1980 to March 2014 
using two models, each with different assumptions about the spread 
over the index rate that Treasury would pay. We compared those costs 
to Treasury's actual costs of funding with 13-week bills and 2-year 
notes. We also analyzed how those costs varied over different interest 
rate environments. 

Estimating Spreads: 

To estimate the range of potential costs from FRNs, we used two models 
of the costs of FRNs to Treasury: 

1. A "maturity-based" model where the spread estimate is influenced by 
the 2-year term of the FRN. 

In the maturity-based model, the FRN spread--the difference between 
the index rate and the interest rate on the FRN--split the difference 
between the 13-week bill and 2-year note yields on the date of the FRN 
auction: 

Formula: 

SpreadFRN = 1/2 (Yield2yfixed - Yieldindex). 

This model was suggested to us by a market participant as one way to 
estimate the likely spread for the Treasury FRN, and we found it to be 
reasonable. 

2. A "reset-based" model where the spread estimate is derived from the 
weekly reset term, which determines the nature of most of the interest 
rate risk faced by investors in FRNs. 

In the reset-based model, the FRN spread adjusted the yield of the FRN 
from the 13-week index rate to a yield equivalent to what a 1-week 
bill would provide. We imputed a 1-week bill rate based on a straight-
line extension of the 13-week bill and 26-week bill rates on the date 
of the FRN auction. Where possible, we used the Constant Maturity 
Treasuries (CMT) series issued by the Board of Governors of the 
Federal Reserve System, which calculates rates for 13-and 26-week 
bills on a daily basis using secondary market data. However, the CMT 
series only begins in 1982, so for our simulations in 1980 and 1981, 
we used the weekly Treasury auction data. Once we imputed the 1-week 
bill rate, we subtracted the current index rate to compute the 
predicted spread: 

Formula: 

SpreadFRN = 12/13 (Yield26w - Yield13w) + Yield13w - Yieldindex) 

This frequently results in a negative FRN spread, meaning that, under 
this model, the FRN generally has a yield lower than a 13-week bill. 
We allowed for negative spreads under this model because Treasury 
regulations allow the FRN to auction with a negative spread and, in 
very low interest rate environments, short term bills on the secondary 
market have sometimes traded with a negative yield. 

While we considered other models for determining the cost of FRNs, 
these two models are designed to approximate the range of potential 
spreads Treasury's 2-year FRN would be expected to have been auctioned 
at in historical interest rate environments. We also considered models 
based on: 

* FRNs from government-sponsored enterprises (GSEs). Several market 
participants we spoke with indicated that FRNs issued by Fannie Mae 
and Freddie Mac would be the closest comparison for Treasury FRNs. 
However, we determined that GSE FRNs were not sufficiently comparable 
for our purposes due to the issuance practices and FRN structures used 
by Fannie Mae and Freddie Mac. 

* Swap prices. Several market participants also suggested interest 
rate and asset swaps could be used to estimate spreads on Treasury 
FRNs. We reviewed results of simulations of FRN spreads published by 
one market participant, and found the estimates from this model 
usually to be within our own estimates for the FRN spread. 

* Theoretically derived formula. We explored modifying the formulas 
used in Don Smith's "Negative Duration: The Odd Case of GMAC's 
Floating-Rate Note" to derive a theoretically correct spread price. 
[Footnote 23] This approach predicted FRN spreads comparatively very 
close to zero and which generally lie within the costs predicted by 
the maturity-and reset-based models. This pricing model did not 
incorporate the pricing consequences of the mismatch between the reset 
rate and the maturity of the index, and so does not fully capture the 
pricing risks faced by the FRN. 

Defining Interest Rate Environments: 

Because interest rate environments vary substantially over time, we 
compared the relative costs of the FRNs in various interest rate 
environments. The different environments, as used in our analysis and 
discussed in our report, are described below (see table 2). 

Table 2: Definitions of Interest Rate Environments: 

Change in Rates: 

Rising; 
Cases where the trend of the 13-week bill rate increases by 1 
percentage point or more over 2 years. 

Steady; 
Cases where the trend of the 13-week bill rate changes by less than ±1 
percentage point over 2 years. 

Falling; 
Cases where the trend of the 13-week bill rate decreases by 1 
percentage point or more over 2 years. 

Trough; 
Cases where the trend in interest rates initially falls, but then 
rises, over 2 years and where the trend in interest rates changes at 
least 0.5 percentage points in each direction. 

Peak; 
Cases where the trend in interest rates initially rises, but then 
falls, over 2 years and where the trend in interest rates changes at 
least 0.5 percentage points in each direction. 

Volatility in Rates: 

Low; 
Cases where the root mean square error (RMSE) of 13-week bill rates in 
the sample is less than 0.112. 

Moderate; 
Cases where the RMSE of 13-week bill rates in the sample is between 
0.112 and 0.219. 

High; 
Cases where the RMSE of 13-week bill rates in the sample is between 
0.235 and 0.500. 

Extreme; 
Cases where the RMSE of 13-week bill rates in the sample is above 
0.521. 

Source: GAO definitions using Treasury auction data. 

[End of table] 

To determine the trend of 13-week yields over a two year period, we 
estimated a linear time trend on the first difference of weekly yields 
(where t is an index of the number of weeks since the start of the two-
year window): 

Formula: 
Yieldt+1 - Yieldt = a + b ± t. 

This is essentially equivalent to fitting a second degree polynomial 
to the yields, allowing us to capture changes in direction of the 
interest rate trend (i.e., peaks and troughs) as well as the slope of 
a linear trend. The estimated curves were used in classifying the 
interest rate environments. 

The cut-offs for assigning an interest rate trend to a category of 
rising or falling--versus steady--were based on our professional 
judgment. Other approaches--such as using traditional statistical 
significance tests--conflate volatility with assessment of the 
presence of a trend and therefore are not appropriate for this 
determination. 

We were able to use a data-derived approach to assign 2-year periods 
to our volatility categories. We use the RMSE statistic as an 
aggregate measure of the weekly yields' total deviation from the 
trend. We then used a k-mean cluster analysis to divide the sample 
into four volatility groups: low, moderate, high, and extreme. 

Estimating Interest Costs: 

Using the maturity-and reset-based models, we estimated what the 
spread would be for FRNs auctioned on the same day as actual 2-year 
fixed-rate notes from January 1980 to March 2012, resulting in 387 
simulated FRNs. We then applied these estimated spreads to the actual 
weekly 13-week bill auctions from January 1980 to March 2014, and 
calculated what the total interest cost would have been for each 
simulated FRN during this period. Like the actual 2-year FRN, we used 
a floor of zero for the daily interest accrual of our simulated FRNs. 

To determine the relative interest cost of the FRN, we compared the 
estimated costs of the simulated FRNs to the costs of the actual 2-
year fixed-rate notes and a rolling series of 13-week bills for each 2-
year period. We estimated the average interest costs relative to 2-
year notes and 13-week bills as well as the percent of cases where 
FRNs generate savings or additional costs compared to bills or notes. 

Results of Simulations of FRN Costs by Volatility in Rates: 

In addition to the results presented in the body of our report, we 
estimated the cost of 2-year FRNs by volatility of the rate 
environment. As shown in figures 14 and 15 below, we found that, at 
all levels of volatility, there was little variation between our two 
models. In periods of low, moderate, and high volatility, 2-year FRNs 
tended to produce savings compared to 2-year fixed-rate notes, but 
could produce either costs or savings compared to 13-week bills 
depending on which model is used. In periods of extreme volatility, 
FRNs produced savings under both models. 

Figure 14: Interest Savings or Added Costs from 2-year Floating Rate 
Notes Compared to 13-Week Bills and 2-Year Notes, by Volatility in 
Rates: 

[Refer to PDF for image: 2 illustrated graphs] 

Graphs depicts the following, in millions: 

Actual results: 
Actual 2-year fixed-rate note; 
Actual 13-week bills. 

FRN modeled results: 
Maturity-based FRN model; 
Reset-based FRN model. 

Interest savings or added costs per year from floating rate notes 
(FRNs) compared to 2-year fixed-rate notes per $1 billion of
issuance: 2-year fixed-rate note: 

Rate environment: 
Overall (n=387): 
Low (n=194): 
Moderate (n=113): 
High (n=58): 
Extreme (n=22): 

Interest savings or added costs per year from floating rate notes 
(FRNs) compared to 13-week bills per $1 billion of issuance: 13-week 
bills: 

Rate environment: 
Overall (n=387): 
Low (n=194): 
Moderate (n=113): 
High (n=58): 
Extreme (n=22): 

Source: GAO simulation based on Treasury and Federal Reserve data. 

[End of figure] 

Figure 15: Percent of Cases Where 2-Year Floating Rate Notes Save or 
Add to Interest Costs Compared to 13-Week Bills and 2-Year Notes, by 
Volatility in Rates: 

[Refer to PDF for image: 4 horizontal bar graphs] 

Maturity-based FRN model compared to 2-year fixed-rate notes: 
Rate environment: 
Overall (n=387): 
Percentage of cases with interest savings: 72%; 
Percentage of cases with added interest costs: 28%; 
Low (n=194): 
Percentage of cases with interest savings: 72%; 
Percentage of cases with added interest costs: 28%; 
Moderate (n=113): 
Percentage of cases with interest savings: 76%; 
Percentage of cases with added interest costs: 24%; 
High (n=58): 
Percentage of cases with interest savings: 67%; 
Percentage of cases with added interest costs: 33%; 
Extreme (n=22): 
Percentage of cases with interest savings: 64%; 
Percentage of cases with added interest costs: 36%. 

Reset-based FRN model compared to 2-year fixed-rate notes: 
Rate environment: 
Overall (n=387): 
Percentage of cases with interest savings: 82%; 
Percentage of cases with added interest costs: 18%; 
Low (n=194): 
Percentage of cases with interest savings: 78%; 
Percentage of cases with added interest costs: 22%; 
Moderate (n=113): 
Percentage of cases with interest savings: 92%; 
Percentage of cases with added interest costs: 8%; 
High (n=58): 
Percentage of cases with interest savings: 84%; 
Percentage of cases with added interest costs: 16%; 
Extreme (n=22): 
Percentage of cases with interest savings: 64%; 
Percentage of cases with added interest costs: 36%. 

Maturity-based FRN model compared to 13-week bills: 
Rate environment: 
Overall (n=387): 
Percentage of cases with interest savings: 19%; 
Percentage of cases with added interest costs: 81%; 
Low (n=194): 
Percentage of cases with interest savings: 5%; 
Percentage of cases with added interest costs: 95%; 
Moderate (n=113): 
Percentage of cases with interest savings: 26%; 
Percentage of cases with added interest costs: 74%; 
High (n=58): 
Percentage of cases with interest savings: 38%; 
Percentage of cases with added interest costs: 62%; 
Extreme (n=22): 
Percentage of cases with interest savings: 64%; 
Percentage of cases with added interest costs: 36%. 

Reset-based FRN model compared to 13-week bills: 
Rate environment: 
Overall (n=387): 
Percentage of cases with interest savings: 85%; 
Percentage of cases with added interest costs: 15%; 
Low (n=194): 
Percentage of cases with interest savings: 80%; 
Percentage of cases with added interest costs: 20%; 
Moderate (n=113): 
Percentage of cases with interest savings: 96%; 
Percentage of cases with added interest costs: 4%; 
High (n=58): 
Percentage of cases with interest savings: 78%; 
Percentage of cases with added interest costs: 22%; 
Extreme (n=22): 
Percentage of cases with interest savings: 91%; 
Percentage of cases with added interest costs: 9%. 

Source: GAO simulation based on Treasury and Federal Reserve data. 

[End of figure] 

[End of section] 

Appendix II: Survey Scope and Methodology: 

To address both of our objectives, we surveyed and interviewed market 
participants regarding (1) the market for FRNs, (2) the structure of 
FRNs, (3) other actions Treasury may consider to expand demand for 
Treasury securities, and (4) communication between Treasury and 
investors. To gather information from a broader range of investors, we 
administered an online survey to 82 of the largest domestic 
institutional holders of Treasury securities in the following sectors: 
money market mutual fund managers, mutual and exchange-traded fund 
managers, state and local government retirement fund managers, retail 
and commercial banks, life insurance providers, property-casualty 
insurance providers, and securities broker-dealers (see table 3). 
Results of the survey are not generalizable. For aggregate survey 
results reproduced as an e-supplement, see GAO-14-562SP. 

Table 3: Survey Responses by Recipient Type: 

Sector: Money market mutual fund managers; 
Total recipients of survey: 9; 
Total completed surveys: 7; 
Response rate (percent): 78%. 

Sector: Mutual fund and exchange-traded fund managers; 
Total recipients of survey: 10; 
Total completed surveys: 8; 
Response rate (percent): 80%. 

Sector: State and local government retirement fund managers; 
Total recipients of survey: 26; 
Total completed surveys: 19[A]; 
Response rate (percent): 73%. 

Sector: Retail and commercial banks; 
Total recipients of survey: 10; 
Total completed surveys: 9; 
Response rate (percent): 90%. 

Sector: Life insurance providers; 
Total recipients of survey: 8; 
Total completed surveys: 5[B]; 
Response rate (percent): 63%. 

Sector: Property-casualty insurance providers; 
Total recipients of survey: 12; 
Total completed surveys: 7; 
Response rate (percent): 58%. 

Sector: Securities broker-dealers; 
Total recipients of survey: 7; 
Total completed surveys: 7; 
Response rate (percent): 100%. 

Sector: Total; 
Total recipients of survey: 82; 
Total completed surveys: 62; 
Response rate (percent): 76%. 

Source: GAO analysis of survey results. 

[A] Three state-local government retirement funds refused to answer 
the survey; two specifically told us they fell outside the scope of 
our review. 

[B] One life insurance provider refused to answer the survey. 

[End of table] 

To identify sectors for our sample, we reviewed data from the Federal 
Reserve's Financial Accounts of the United States, (table L.209, third 
quarter 2013) to identify which sectors have at least $60 billion in 
Treasury holdings. We excluded some sectors due to challenges in 
contacting certain entities, such as foreign monetary authorities, 
other foreign investors, and the household sector. 

To identify the organizations within each sector that would receive 
our web-based survey, we used rankings of the largest organizations in 
each sector based on total assets or an equivalent financial 
indicator, such as assets under management or direct premiums written. 
From these ranked lists, we determined Treasury holdings for each 
organization and selected as many organizations as needed to represent 
at least 50 percent of the total amount of Treasury holdings for that 
sector (based on table L.209 of the Federal Reserve's Financial 
Accounts of the United States) or in the case of mutual funds, 
exchange traded funds, and money market funds, based on information 
from the Investment Company Institute on total assets under management 
in Treasury-and government-focused funds. 

[End of section] 

Appendix III: GAO Contacts and Staff Acknowledgments: 

GAO Contact: 

Susan J. Irving, (202) 512-6806 or irvings@gao.gov: 

Staff Acknowledgments: 

In addition to the contact named above, Tara Carter (Assistant 
Director), Susan E. Murphy, (Analyst-in-Charge), Abigail Brown, Emily 
Gruenwald, Daniel Ramsey, and Albert Sim made key contributions to 
this report. Amy Bowser, Dianne Guensberg, Stuart Kaufman, Risto 
Laboski, Donna Miller, Dawn Simpson, and Stewart W. Small provided 
subject matter assistance. 

[End of section] 

Footnotes: 

[1] Total federal debt consists of two components: debt held by the 
public and debt held by government accounts. Debt held by the public 
is the value of all federal securities held by the public, i.e., 
investors outside of the federal government. Debt held by government 
accounts represents balances in the federal government's accounts--
primarily trust funds, including the Social Security and Medicare 
trust funds--that accumulate surpluses. See GAO, "Fiscal Outlook: 
Understanding the Federal Debt," Key Issues, (Washington, D.C.: 2014), 
accessed Apr. 22, 2014, [hyperlink, 
http://www.gao.gov/fiscal_outlook/understanding_federal_debt/overview] 
and GAO, Financial Audit: Bureau of the Fiscal Service's Fiscal Years 
2013 and 2012 Schedules of Federal Debt, [hyperlink, 
http://www.gao.gov/products/GAO-14-173] (Washington, D.C.: Dec.12, 
2013). 

[2] Twenty-two banks and securities broker-dealers are currently 
designated by the Federal Reserve Bank of New York (FRBNY) as primary 
dealers and are expected to participate meaningfully in every Treasury 
auction by bidding for, at a minimum, an amount of securities 
representing their share of the offered amount (based on the number of 
primary dealers at the time of the auction). Primary dealers also have 
a role in making a secondary market for Treasury securities. (The 
secondary market is the market in which previously issued Treasury 
marketable securities are bought and sold among investors.) 

[3] In addition to marketable securities, Treasury issues 
nonmarketable securities that cannot be resold, such as U.S. savings 
bonds and special securities for state and local governments. 

[4] In a liquid market, trading can be completed at will and the offer 
and purchase prices differ only slightly. Liquidity is important to 
Treasury because investors are willing to pay more for liquid 
securities that can be easily traded, resulting in lower borrowing 
costs for Treasury. 

[5] In a reopening, Treasury auctions additional amounts of a 
previously issued security. Reopened securities have the same maturity 
date and coupon interest rate or spread as the original securities, 
but have a different issue date and usually a different purchase price. 

[6] Consistent with its practice for other securities, Treasury does 
not announce specific FRN offering amounts until shortly before each 
auction. 

[7] 31 C.F.R. § 356.5(b)(3). 

[8] The FRN index rate is the high rate from the 13-week Treasury bill 
auction converted to a simple-interest money market yield on an 
actual/360 basis. 

[9] In times of federal budget deficits, all maturing debt must be 
rolled over into new issuances. Rollover risk also includes 
operational risk, such as the risk that a Treasury auction cannot be 
held due to technical problems with the auction systems. 

[10] The weighted average maturity of outstanding marketable Treasury 
securities is calculated by averaging the remaining maturity of all 
outstanding marketable Treasury securities, weighted by the dollar 
value of the securities. 

[11] Definitions of the interest rate environments we used can be 
found in appendix I. 

[12] A money market fund is a type of investment fund that is required 
by law to invest in low-risk securities. These funds have relatively 
low risks compared to other mutual funds and pay dividends that 
generally reflect short-term interest rates. Money market funds 
typically invest in government securities (including Treasury bills 
and notes), certificates of deposit, commercial paper of companies, or 
other highly liquid and low-risk securities. 

[13] 17 C.F.R. § 270.2a-7(c)(2). 

[14] Previously, we recommended that Treasury take actions to increase 
TIPS liquidity; Treasury has implemented each of our recommendations. 
See GAO, Debt Management: Treasury Inflation Protected Securities 
Should Play a Heightened Role in Addressing Debt Management 
Challenges, [hyperlink, http://www.gao.gov/products/GAO-09-932] 
(Washington, D.C.: Sept. 29, 2009). 

[15] In a Treasury FRN auction, bids are made in terms of a desired 
discount margin. The highest accepted discount margin in the initial 
auction for a given FRN (which we refer to as the FRN spread) becomes 
the spread for that FRN, and bidders pay the full value of the FRN. At 
subsequent reopening auctions of the FRN, the spread is fixed based on 
the results of the initial auction. Bidders at the auction still bid 
on a discount margin basis and may pay more, less, or the same as the 
full value, depending on whether the discount margin is less, more, or 
the same as the initial auction. 

[16] Treasury issues 4-week, 13-week, 26-week, and 52-week bills on a 
regular schedule. Treasury issues cash management bills of varying 
maturities only as Treasury financing needs require. 

[17] International Monetary Fund and the World Bank, Guidelines for 
Public Debt Management, as amended Dec. 9, 2003. 

[18] 17 C.F.R. § 270.2a-7. 

[19] These planned purchases include purchases at Treasury auctions, 
in the secondary market, and in the when-issued market. 

[20] GAO, Debt Management: Treasury Was Able to Fund Economic 
Stabilization and Recovery Expenditures in a Short Period of Time, but 
Debt Management Challenges Remain, [hyperlink, 
http://www.gao.gov/products/GAO-10-498] (Washington, D.C.: May 18, 
2010). 

[21] Ultra-long bonds are issued with a maturity greater than 
Treasury's current offerings, such as 40-, 50-or 100-years. Callable 
bonds are bonds that can be redeemed or paid off by the issuer prior 
to the bond's maturity date. FRNs indexed to inflation are floating 
rate notes that adjust the coupon rate for inflation while the 
principal is held constant. Zero-coupon bonds are bonds that do not 
pay interest during the life of the bonds. Zero-coupon bonds pay a 
lump sum, equal to the initial investment plus imputed interest, at 
maturity. 

[22] Buybacks are the redemption of marketable securities prior to 
their maturity dates. We previously recommended that Treasury build 
the capacity for a buyback program that could be used to respond to 
potential changes in market conditions during times of deficit. This 
recommendation has not been implemented to date. See GAO, Debt 
Management: Buybacks Can Enhance Treasury's Capacity to Manage under 
Changing Market Conditions, [hyperlink, 
http://www.gao.gov/products/GAO-12-314] (Washington, D.C.: Mar. 7, 
2012). A reverse inquiry window is a method of purchasing securities 
where an investor consults with a broker-dealer who proposes to an 
issuer that a specific security be issued to meet the investor's needs. 

[23] D. J. Smith, "Negative Duration: The Odd Case of GMAC's Floating-
Rate Note," Journal of Applied Finance, vol. 16, no. 2 (2006). 

[End of section] 

GAO's Mission: 

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

Obtaining Copies of GAO Reports and Testimony: 

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

Connect with GAO on facebook, flickr, twitter, and YouTube.
Subscribe to our RSS Feeds or E mail Updates. Listen to our Podcasts.
Visit GAO on the web at [hyperlink, http://www.gao.gov]. 

To Report Fraud, Waste, and Abuse in Federal Programs: 

Contact: 
Website: [hyperlink, http://www.gao.gov/fraudnet/fraudnet.htm]; 
E-mail: fraudnet@gao.gov; 
Automated answering system: (800) 424-5454 or (202) 512-7470. 

Congressional Relations: 

Katherine Siggerud, Managing Director, siggerudk@gao.gov: 
(202) 512-4400: 
U.S. Government Accountability Office: 
441 G Street NW, Room 7125: 
Washington, DC 20548. 

Public Affairs: 

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

[End of document]