To: ild who wrote (24586 ) 2/28/2005 2:29:07 PM From: mishedlo Read Replies (1) | Respond to of 116555 Why the Long Bond Isn't coming back [This seems like a good explanation - mish] U.S.: The Name is Bond -- "Long" Bond David Greenlaw (New York) On October 31, 2001, former US Treasury Undersecretary Peter Fisher announced that issuance of the 30-year bond would be suspended. At that time, bonds with more than 10 years remaining maturity accounted for nearly 18% of publicly held Treasury debt outstanding. Today the long end (excluding TIPS) represents less than 12% of the market. The narrowing of the 10’s-30’s spread over the past few months has led to renewed speculation that the Treasury Department may soon resume issuance of 30-year bonds. From mid 2002 through late 2004, the 10’s-30’s spread held within a range of 75 bp to 110 bp. However, in the past four months the spread has steadily narrowed to around 40 bp. Why the big move? Even though Fed Chairman Greenspan omitted pension fund demand as a possible explanation for the low-yield “conundrum,” this factor certainly appears to have helped support the long end of the bond market of late. My colleagues Richard Berner and Trevor Harris addressed the pension topic in great detail in a piece entitled “Pension Missiles: Is the Cure Worse than the Disease” (March 25, 2004) and extended the analysis following the release of the latest Bush administration pension reform proposal (see “Financial Market Implications of Pension Reform” in the January 18 edition of the Global Economic Forum). They concluded that pension reform could prompt a significant reallocation out of equities into bonds – to the tune of $650 billion. Moreover, the portfolio adjustment “could also entail a major increase in duration.” Of course, enactment of meaningful pension reform is still highly uncertain – a similar legislative initiative on the part of the Bush administration back in 2001 went nowhere. Still, even in the absence of major regulatory change, there appears to be a need to alter the portfolio composition of the nearly $4 trillion of assets that resided in public and private defined-benefit pension plans at the end of 2004. The market’s focus on the potential for massive flows into long-dated Treasuries – as opposed to the onset of actual buying by pension funds – appears to have played an important role in the recent flattening of the 10’s-30’s spread. Will the Treasury respond by resuming issuance of the 30-year bond? In order to address this question, it is important to understand the motivation for the elimination of the long bond. According to Treasury Department officials, the decision was made in accordance with the primary objective of debt management – to achieve the lowest-cost financing over time. In the view of Fisher and his staff, continued issuance of 30-year bonds was costly for two main reasons. First, it limited the Treasury’s flexibility. It was becoming increasing evident that the federal government’s finances were volatile and very difficult to predict. The budget swung from deficit to surplus much more rapidly than anyone had anticipated during the late-1990s. The buyback operations conducted during the surplus years were an expensive form of cash management. The Treasury determined that it could achieve greater flexibility by reducing the average maturity of the outstanding debt, and continued issuance of long bonds made it difficult to move in this direction. The second justification for the suspension of the bond was more directly related to a determination that long bond issuance was costly. Treasury officials observed that the yield curve was upward sloping over time, on average. The federal government is very different from a private corporation that might find it advantageous to pay up a little bit for the safety of locking in fixed funding for a long period of time. The Treasury is a frequent issuer with zero credit risk dealing in an extremely large and liquid market with a broad range of investor classes. Thus, unless the government attempts to engage in market timing – a practice that a long history of Treasury debt managers from both Republican and Democratic administrations has deemed to be extremely unwise –it makes no sense to pay the term premium embodied in issuing long-term debt. Of course, this line of reasoning raises the question of why the Treasury doesn’t confine its issuance to the very front end of the curve – say, by rolling over $3 trillion or so of short-dated bills every week. This is an interesting counterargument although it is obviously a bit extreme. The Treasury can maintain large, liquid issues out to 10 years and still achieve the target average maturity (somewhere around 3 years) that provides the desired flexibility Although former Treasury Undersecretary Peter Fisher, who appeared to be the chief proponent of eliminating bond issuance, has moved on to the private sector, the current debt management team has consistently backed this decision. Indeed, Tim Bitsberger, the current head of debt management, has offered strong public support for the suspension of bond issuance on numerous occasions. For example, in presentations delivered to investor groups in Philadelphia and New York in late 2003, Bitsberger concluded that issuance of long bonds “does not meet our criteria for achieving the lowest borrowing cost over time.” Indeed, as recently as this past Wednesday – the day that France auctioned a 50-year OAT – Bitsberger made the following statement during a brief press interview: “We view our current issuance calendar as being very flexible in being able to handle the variety of fiscal outcome. We’re very confident that we don’t need to make any changes to our issuance calendar at this point.” From our standpoint, two things need to happen to change the Treasury Department’s position. First, the administration has to be ready to concede that there will be a need to finance fairly sizable budget gaps for an extended period of time. This would reduce the need for short-run flexibility. While such a declaration would certainly appear to be realistic at this point, it may not be politically palatable. Second, the 10’s-30’s portion of the yield curve must invert and appear likely to stay that way for an extended period. In such an environment, the resumption of bond issuance would be perfectly consistent with the objective of lowest cost financing. In the past couple of weeks, portfolio managers have been publicly urging the Treasury to bring back the bond in order to meet pension-related demand (for example, see the Current Yield column in the February issue of Barron’s and the Wall Street Journal Credit Markets column of February 24). But, there is little evidence at this point of any abnormal compression of the long end of the curve related to pension fund demand. Indeed, the 10’s-30’s spread is somewhat wider than might be expected from a historical standpoint. Using monthly average yield data published in the Fed’s H15 report from 1977 (when the Treasury first started offering a wide range of maturities on a regular basis) through the end of 2001 (when the Treasury ceased issuance of 30-year bonds), we ran a simple regression of the difference between the yield on 30-year bonds and 10-year notes and the difference between the yield on 10-year and 2-year notes. In other words, the 10’s-30’s spread versus the 10’s-2’s spread. The equation showed that 10’s-30’s is generally 28% of the 10’s-2’s spread (the intercept term is close to zero). As of today, the 10’s-2’s spread is 75 bp, which would imply an expected 10’s-30’s spread of 21 bp. Instead, the actual 10’s-30’s spread is around 40 bp – or more than one full standard error wider than the historic norm. The focus on potential pension fund flows has simply brought the 10’s-30’s spread from an extremely abnormal level to something that is closer to historic norms. If meaningful pension reform is enacted and/or fund managers are otherwise incentivized to switch into long-duration Treasuries, the 10’s-30’s portion of the curve could conceivably invert. But, until that happens, we wouldn’t count on a return visit from Mr. “Long” Bond.morganstanley.com