To view or print the PDF content on this page, download the free Adobe® Acrobat® Reader®.
September 15, 2008
Treasury Office of Debt Management Director Karthik Ramanathan
New York City - Good Morning. Thank you for giving me this opportunity to share my thoughts on Treasury debt management. I'm going to start off by explaining the role that our office and Treasury securities play in the capital markets, then describe some of the challenges posed by recent market conditions, and finally, address Treasury Inflation Protected Securities.
In my role as the Director of the Office of Debt Management, I provide recommendations on matters related to the Treasury's debt management policy, the issuance of Treasury securities, and the state of financial markets. By actively engaging with reserve managers, institutional investors, and market participants like you, we remain well informed regarding financial market conditions, liquidity in the fixed income markets, and Treasury-market-specific issues.
Our mission is to issue debt in a manner that provides the U.S. government – and ultimately the taxpayer – with the lowest cost of financing over time. With over $4 trillion of annual marketable Treasury issuance, more than 200 Treasury auctions each year, and nearly $9.5 trillion in total federal debt, the numbers are large, so discussing these issues is quite relevant.
U.S. Treasuries play an important role in the global capital markets. They are actively used by portfolio managers, investors, and traders to hedge existing positions, to serve as the risk-free pricing benchmark, and to provide the ultimate source of liquidity. In addition, the Federal Reserve uses Treasury securities to affect the supply of reserves in the banking system, and Treasuries provide foreign central banks with a highly liquid investment vehicle. The central roles that Treasuries play contribute to a lower overall cost of capital.
However, despite this prominent role, we do not take our position in the debt markets for granted. We constantly strive to enhance Treasury's status as the preeminent sovereign debt market by adhering to our clear mission. This gives us confidence that Treasury will remain the borrower of choice in global capital markets.
Now, before I delve into more detail on TIPS, I'd like to first talk about Treasury's debt management objectives, then financial market challenges over the past year, and finally, how we as debt managers responded.
Debt Management Objectives and Operating Principles
The Treasury Department's primary goal in debt management is to finance the government's borrowing needs at the lowest cost over time. In meeting this objective, we face numerous constraints and risks.
Perhaps the most prevalent of these constraints is uncertainty. Uncertainty arises from many different sources including changes in economic conditions, unexpected legislative initiatives, and fluctuations in non-marketable debt issuance. In a given year, these factors could easily shift borrowing by a significant amount with little advance warning.
Another major source of uncertainty stems from deficit forecast errors with various estimates off by more than $100 billion on average. Collectively, the private sector as well as policy makers have difficulty in consistently projecting deficits in the coming twelve months – let alone further into the future.
Given the extensive uncertainty Treasury faces, debt management requires considerable flexibility. The recent fiscal stimulus package, in which rebates were literally place into the hands of Americans just 20 weeks after enactment, shows such a need for adapting to rapidly changing conditions. At the same time, our large size makes behaving opportunistically impractical. Moreover, it would not necessarily lower borrowing costs. Financial market participants would likely model our interest rate objectives and anticipate our debt issuance behavior, limiting any potential gains we might hope to achieve.
In addition, opportunistic behavior would increase investor uncertainty and could limit demand for our securities. Therefore, in order to ensure ready market access, we issue debt regularly and in predictable amounts. Our set of instruments consists of 8 nominal issues and 3 inflation indexed issues – a very simple, liquid set of benchmarks which investors can tailor to their needs.
Under these conditions, we must not create additional constraints based on externalities that result from a particular debt management strategy. For example, the U.S. Treasury often receives requests for debt tailored to particular interests such as GDP-linked debt, annuitizing debt instruments, and callable debt. However, it is the Treasury's policy not to issue debt targeted to any particular constituency. As taxpayers, we are better off with the Treasury market being deep and liquid. We consider the market's appetite for certain securities or debt management practices; however, debt management policy decisions cannot be held captive solely to the market's preferences.
For example, when Treasury discontinued the 52-week bill in 2001, we did so despite positive externalities associated with the security including its wide use as a benchmark point for pricing derivatives (such as adjustable rate mortgages and interest rate swaps), and its use for setting rates for statutorily required credit programs (which required Congress and Treasury to make legislative changes). We simply did not need it at the time, and acted in the best interest of the taxpayer. Similarly, the 30-year bond and 30-year TIPS were also discontinued in 2001 due to reduced borrowing needs.
Decisions to adopt or suspend particular debt management practices are similarly made by always keeping in mind our goal of the lowest borrowing cost over time. For instance, despite the benefits of multiple-price auctions to certain more sophisticated segments of the financial markets, Treasury moved to uniform-price auctions. Internal studies and empirical analysis indicated that such auctions broaden the distribution of auction awards, promote efficiency in the markets, and lead to more aggressive bidding – all factors which collectively reduce the cost of financing the federal debt.
As another example, we continue to have the authority to conduct buybacks of Treasury securities. However, Treasury suspended this practice when they were no longer practical despite the benefits to some investors of having a regular buyer for relatively illiquid securities.
While not limiting potential responses to ever changing financial market conditions, we make any changes to Treasury debt issuance and debt management practices in a transparent manner, in consultation with market participants, and based on analyses of how to best meet our goals.
Financial Market and Debt Management Challenges
The financial markets have faced challenging conditions over the past year, many of which have impacted the broader economy. Tighter credit standards and pressure on interest rate spreads have made it more difficult to obtain credit in many sectors. In the face of these challenges, Treasury has responded aggressively.
To provide confidence and stability to financial markets, Treasury financed an economic stimulus package and moved to support the housing government sponsored enterprises Fannie Mae and Freddie Mac. A cyclical correction is underway, particularly in the housing sector, and this process of re-pricing of risk and deleveraging across asset classes has created additional challenges for market participants.
Borrowing requirements have risen swiftly in response to the economic stimulus legislation as well as actions take by the Federal Reserve related to its liquidity initiatives. To sterilize the monetary effects of these initiatives, the Federal Reserve redeemed Treasury securities held in its portfolio and also sold securities outright which resulted in nearly $300 billion in additional Treasury issuance. We responded successfully to these challenges by increasing bill auction sizes, reintroducing the 52-week bill, and issuing longer dated cash management bills.
Looking ahead, a wide variety of factors could potentially impact Treasury's marketable borrowing including a less robust economy, the possibility of additional legislative initiatives enacted by Congress, and further pressures on the financial sector.
In this rapidly evolving economic and financial market environment, Treasury has responded to changes in marketable borrowing needs in its traditional manner by first reviewing the size of our existing securities, then addressing the frequency of issuance, and finally, making adjustments to the auction calendar as necessary.
In addition, through our meetings with major investors domestically and abroad, many ideas have been suggested to better position Treasury to meet these challenges. Some recommendations include increasing the frequency of the 10-year note and 30-year bond, reintroducing other securities including the 3-year note and 7-year note, and reintroducing a Treasury buyback program to better manage our debt maturity profile.
Other ideas included issuing additional longer dated inflation linked securities versus shorter dated securities to potentially better capture any inflation premium. We appreciate all of these suggestions, and take them all under consideration in achieving the lowest cost of financing over time.
Treasury Inflation-Protected Securities
Now, turning to inflation linked securities, Treasury has been issuing TIPS for over 10 years and is the largest issuer of inflation linked bonds globally. We have held 60 TIPS auctions since the inception of the program and have over half a trillion dollars of such debt outstanding. Average daily trading volume of $9 billion also makes the TIPS market the most liquid of any sovereign inflation-linked debt market. With 27 issues outstanding, the TIPS curve is well established out to 10 years, and well on its way to being complete out to 20 year.
Let me give you some other figures about the size and liquidity if the TIPS market. In fiscal year 2007, TIPS issuance totaled $57 billion, or 10 percent of total Treasury coupon issuance, and represented about 30 percent of total global inflation linked debt issuance. In comparison, such issuance was $29 billion in the United Kingdom, $27 billion in Japan, and $26 billion in France. Although the growth rate of TIPS has slowed, it still outpaces nominal coupons.
In terms of secondary market liquidity, which often receives much attention, the TIPS market is much more liquid than any other sovereign inflation linked market. Prior to this most recent period of stress in credit markets, the typical bid-ask spread on the benchmark 10-year TIPS on a trading size of $50 million was about 1 basis point. In contrast, for similar benchmark issues in the United Kingdom, France and Japan, the bid–ask spreads ranged between 2.5 and 5 basis points.
So, while TIPS will likely remain less liquid than nominal coupons due to their unique qualities, from an investor perspective, the depth of the TIPS market is unrivaled. Calls to increase liquidity through much larger issuance need to be carefully evaluated.
Taken together, these market statistics illustrate our preeminent stance in the inflation-indexed market in terms of size, depth, and liquidity, even in an environment without regulatory mandates and in the absence of a high rate of indexation to inflation compared to other sovereigns.
While the TIPS auction calendar has seen several changes since the inception of the program, we have repeatedly communicated to market participants our commitment to the program. Let me again reassure you that inflation linked securities are an important part of our portfolio.
As I mentioned earlier, determining the proper mix of our portfolio in pursuit of the lowest cost of borrowing is an ongoing effort. From our perspective, TIPS offer potential benefits including a more diversified portfolio, a potentially broader investor base, and a liability that theoretically tracks tax receipts.
However, you may be aware that recently the Treasury's Borrowing Advisory Committee of the Securities Industry and Financial Markets Association, or TBAC, prepared a presentation on TIPS. The presenting Committee member concluded that the cost of the program, compared to nominal debt issued at a similar time, was estimated at close to $30 billion. This cost was attributed to two factors: the low level of breakeven inflation and the reduced level of liquidity of TIPS compared to nominal securities.
It was also noted that private issuers have been less willing to issue inflation linked securities because they view them as costly and because of unfavorable accounting treatment. In fact, over the past ten years, there have been only a handful of corporate issuers of inflation indexed debt, thereby limiting the growth of the inflation derivatives markets. The large majority of corporate issuance is immediately hedged with TIPS, so it does not really create new inflation-linked supply. Instead, the Treasury continues to be the only significant payer of inflation in the United States.
This situation naturally raises many questions. Why are corporations not issuing such debt more generally? Are we selling insurance on inflation protection for too little? Should unknown future liabilities resulting from inflation accretion concern debt issuers? Certainly in a corporation or financial institution, these issues would be taken into consideration. Are we as sovereign debt issuers doing the same?
Not surprisingly, the financial press has been discussing the deliberations of the TBAC's presentation. Some have agreed with its findings, while others have disputed them. As debt managers, we greatly appreciate all viewpoints and encourage further constructive dialogue. The growth in the significance of the inflation linked debt market over the past decade by sovereign issuers has made having an accurate understanding of the costs involved in their issuance more important than ever.
Sovereign debt issuers have issued inflation indexed debt with the belief that such issuance would diversify their portfolios and better track receipts. In addition, there was a belief that borrowing costs would be lower due to the willingness of some investors to pay a premium in return for inflation compensation. Other nations have recently joined this trend, issuing inflation linked debt with the same intentions and with little empirical or analytical studies of costs versus benefits. Unfortunately, though, there have been only a few published studies of the costs of issuing TIPS, and these have offered conflicting conclusions.
In May of 2004 then Federal Reserve economists Brian Sack and Robert Elsasser estimated that TIPS issuance since the inception of the program had been expensive relative to comparable nominal securities, primarily due to the difficulties in launching a new asset class and the flight to quality earlier in the decade. Sack and Elsasser used realized costs - an "ex post" approach - in their estimates. They estimated that the cost of the program as of June 2003 was $2.8 billion, and estimated a projected total cost of $12.3 billion.
An update to this analysis in 2007 by Sack showed that the 10-year TIPS that matured in January 2007 saved the Treasury $1.1 billion.
In October 2007, also using ex-post calculations, Federal Reserve Board economist Jennifer Roush concluded that TIPS issuance was relatively costly due to illiquidity in the early years of the program. She estimated the cost of the program through early 2007 at around $5 billion to $8 billion.
Roush's analysis, however, also suggested that beginning with issuance in 2004, TIPS have actually saved Treasury a small amount of money, and will save Treasury from $1 billion to $4 billion over the entire life of these securities. Moreover, she finds that if the illiquidity effects of the early years of the TIPS program are excluded, the TIPS program would have saved the Treasury a substantial amount – from $14 billion to $17 billion through early 2007. The "liquidity premium" effectively cost Treasury between $10 billion and $15 billion.
On the other hand, some economists have suggested that ex-post analyses are too simplistic and that the relevant question is whether the Treasury obtained the financing it needed at a lower "ex-ante cost;" that is, using expected inflation at the time of issuance in determining the cost. However, one of the difficulties of an ex-ante approach is obtaining an accurate estimate of investors' inflation expectations. Studies using an ex-ante approach have shown neither a benefit nor a cost from issuing TIPS relative to nominal securities.
As can be seen from these studies, the results to date have been conflicting and, at times, inconclusive. Assumptions which are used greatly vary. Perhaps most disturbing, few rigorous, analytical approaches have been undertaken to fully understand the efficiency of the program.
Treasury has a duty to ensure that taxpayers attain the lowest cost of borrowing over time. In that vein, we must continuously study our models, develop alternative perspectives, and institute changes if warranted. We need to focus on our mission and less on positive externalities. From our perspective as debt issuers, we have a wealth of information to examine. We know the details of every competitive bid made at each auction. We know the concentration of bids by particular investors at given auctions. And we know our alternative funding choices. Moreover, we have a large set of observations. While the growth of inflation-indexed securities remains robust, and the importance investors place on them continues to grow, taking a step back, evaluating our practices, and examining the costs and benefits of any program in a deliberative manner is only prudent. .
Treasury, like other major sovereign issuers of debt, needs to attract capital from the market, but we need to do so in a thoughtful manner. So I want to make the earnest request to all sovereign debt issuers and members of the financial market community actively develop an appropriate framework for assessing the cost of issuing inflation linked debt versus nominal debt.
Such an undertaking will benefit all who participate in this market, most importantly the taxpayer. As we undertake these deliberations in concert with financial market participants, Treasury will continue to issue TIPS in a regular and predictable manner and continue to maintain these securities as a significant portion of our overall debt portfolio.