Making a living in the bond market used to be pretty simple. Figure out the next Fed move. Keep an eye on default rates. Don’t buy too many bonds denominated in Danish kroner or any other secondary currency.
U.S. national debt reduction changed all that. As budget deficits turned into budget surpluses, the U.S. government began cutting back on the size and frequency of new bond offerings, dramatically altering the fundamental pricing and investing dynamics of the bond market. The bellwether 30-year Treasury bond began moving in response to its own supply and demand factors, not broad economic fundamentals. Spreads between Treasuries and corporate bonds, hostage to the same supply squeeze, ballooned to levels once thought impossible. And long-term interest rates, which typically jumped when the U.S. Federal Reserve Board lifted short-term rates, now sometimes mysteriously stayed put.
These unintended consequences of deficit reduction are deeply ironic. The Clinton administration’s early economic strategy, after all, focused on keeping interest rates low by reducing the deficit. And chief aides, such as former policy and strategy adviser George Stephanopoulos, tracked the rise and fall of the 30-year bond daily. Interest rates have stayed low, but the benchmark died in the process.
Traders, underwriters and investors have been forced to make dramatic adjustments. Such major issuers as DaimlerChrysler Corp. and Kreditanstalt für Wiederaufbau started pricing new offerings in relation to corporate and agency credits rather than traditional benchmark Treasuries. Investors increased their scrutiny of corporate balance sheets rather than relying on the conventional reading of credit spreads. Traders began factoring in much greater expected volatility on corporate and agency bond positions. And the entire industry began debating whether Fannie Mae or Freddie Mac paper might replace the Treasury bond as a reliable benchmark. “We recognize that the 30-year is an aberrant security,” says Ian McKinnon, head of fixed income at Vanguard Group, which manages $140 billion in bonds.
Larger questions loom. If the Treasury market continues to shrink, the changing risk profile of the bond market could have a dramatic, still largely unexamined impact on the management of money. The bond market is supposed to be a relative safe haven for assets in uncertain economic times. But this haven—as defined by the overall market bond portfolio—just got a whole lot riskier.
Bond managers, like most investors, measure themselves against accepted industry benchmarks. The most important benchmark for the fixed-income market, the Lehman Brothers aggregate bond index, is designed to reflect the outstanding levels of bonds in the overall market. As a result of the declining supply of Treasuries, the index’s Treasury weighting has plunged from a peak of 47 percent in 1994 to 31.9 percent as of this April. Since 1997 alone the government bond weighting has shed more than 10 percentage points. (Between 1997 and this May, Treasuries’ share of the index fell from 42.9 percent to 30.7 percent. Agencies’ share has jumped from 6.6 percent to 9.8 percent in the same period, while that of corporate bonds has increased from 19.3 to 22.4 percent.)
Those changes mean that the many bond investors who track or shadow the Lehman index are these days holding much riskier portfolios. And they are increasingly focused on what most investors would consider murkier areas of the fixedincome market. Mortgages, for example, which have grown from a 30.2 percent weighting in 1997 to 34.3 percent in May, now account for a greater share of the Lehman index than Treasuries. A safe haven mortgages are not.
Bond managers face technical challenges from these larger mortgage positions—the negative convexity of mortgages means they prepay in a falling rate environment, reducing their average life just when the overall bond market is rallying and long-dated bonds should benefit. “Mortgages are going to dominate the index,” predicts Patricia Cook, portfolio manager at New York-based bond manager Fischer, Francis, Trees & Watts. “The negative convexity of mortgages makes them a less than optimal asset class. Exactly when plan sponsors want duration, the mortgages will prepay.”
Bad markets in mortgages are even more problematic. The mortgage-backed-securities sector, with its illiquidity and unreliable prepayment models, has been a killing field for a long list of money managers.
Bond investors, of course, will continue to face the greatest challenges. And they will have to make adjustments and shifts under already trying conditions. After the collapse of LongTerm Capital Management and the other hedge fund meltdowns of 1998, dealers pulled back their risk capital lines severely, making it difficult and expensive for institutional investors to move out of positions. Bond hedge funds, whose highly leveraged trading provided a stream of rapid-fire buying and selling, have disappeared or been forced to scale back positions, further reducing risk capital in large sections of the fixed-income market. “The great catastrophe of 1998 damaged liquidity terribly,” says Tad Rivelle, a portfolio manager at Metropolitan West Asset Management. “You’ve got risk-seeking capital in the equity market, and the fixed-income market is getting starved of risk-seeking capital.”
The uncertainty and illiquidity in the bond industry are not likely to lift anytime soon. U.S. Treasury Secretary Lawrence Summers has predicted zero Treasury debt by 2013, which would translate into zero Treasuries in the industy’s benchmarks. Politics or a recession could change hat schedule, with presidential candidate George W. Bush already promising to slash taxes instead of paying down more of the debt. But the 30-year bond continues to lose relevance, with the Wall Street journal recently announcing that it would use the ten-year Treasury bond as its benchmark, rather than the traditional 30-year. Meanwhile, the market debate about what might become a reliable benchmark substitute for government bonds remains unresolved, and the ongoing sniping against Fannie Mae’s special semi-government-backed status by some congressional critics, not, to mention the Treasury Department, makes a quick resolution unlikely. “If we take Treasury projections at face value, the surplus will change dramatically the way we and other people manage fixed income,” says Lee Crabbe, a portfolio manager at Credit Suisse Asset Management. "[But] what we have learned over 20 years is that you can’t be very certain about the Treasury deficit.”
Bond fund managers have little choice but to experiment with new techniques of risk management. But in making many of these important decisions, from benchmarking portfolios to hedging, they still must resort to guesswork. “We are seeing a fundamental shift,” says Greg Boal, head of institutional fixed income at Scudder Kemper Investments. “We are struggling with what to use as our benchmark. But do we understand entirely what it means? Absolutely not.”
THE NATIONAL DEBT DAMAGED THE U.S.'S pride but nourished its bond business. Deep, liquid and ever expanding, government bonds became an invaluable touchstone for the highly fragmented fixed-income market. Moving in highly predictable patterns in response to economic stimuli and Federal Reserve actions, Treasuries served as a very reliable benchmark of bond value. Treasuries plus x became the standard way to price new bonds. Government bonds were also an immensely liquid hedging market for swaps, mortgages and corporate bonds.
Then the Treasury market began disappearing at a rate few had predicted. Since 1997 the supply of Treasury bonds has been shrinking by about $100 billion a year, from a peak of $3.6 trillion to about $3.3 trillion now. Though the pile of government bonds remains huge, the Treasury’s decision to hold fewer auctions has already severely limited the supply of socalled on-the-run paper, or new bonds that are commonly used for hedging mortgages and corporate bonds.
The phenomenon of debt reduction is global. Australia, Belgium, Canada, Denmark, Ireland, the Netherlands, Norway, Spain and the U.K. are all either paying down their debt, reducing their high-coupon debt or spending less than they are borrowing.
As the U.S. government reduced its debt, Treasury bonds quickly began losing some benchmarking and hedging value. Government bonds started trading in relation to expectations about supply and demand of particular issues rather than on the basis of predictions about long-term rates. At the beginning of February, for example, the Treasury said that tax receipts were far higher than expected. For the first time the government announced plans to buy back bonds instead of just issuing fewer of them. The result was a powerful rally at the long end of the yield curve, as investors realized that 30-year Treasury bonds were likely to become increasingly rare. Long-dated corporate spreads widened dramatically, which would normally be taken as a signal of emerging credit problems. In this case the move was more likely a reflection of concerns about a supply squeeze on the 30-year bond.
Decades of bond data may no longer be relevant. Spreads are now wider than they were even during the tumultuous autumn of 1998, and many historical models now seem illogical. Ten-year swap spreads, now the commonly acknowledged bellwether of corporate credit, recently came out to an unprecedented 142 basis points—a good 30 basis points above their 1998 levels. Add to that the enormous volatility that has become commonplace in the market, and judging value becomes especially treacherous. “In the old days a two-thirds-basis-point widening was a big move,” explains Fares Noujaim, co-head of debt capital markets at Bear, Stearns & Co. “Now there are lots of days when there is a 10-, 15-, 20-basis-point widening or tightening in a day.”
Trading and investing have become more troublesome and dangerous. “Risk is much more difficult to manage,” says Sykes Wilford, a portfolio manager at CDC Asset Management. “Right now spreads don’t make any sense. The kind of spreads we see for high-yield bonds in this economy make no sense.” The situation is unlikely to improve in the near term. Bond managers now face the possibility that the government might reduce not just its 30-year debt but also its heavily traded ten-year debt, as it moves to slash its interest rate costs. Ten-year bonds have traditionally been the interest rate hedging vehicle of choice for mortgage lenders, because mortgages have an average life of about ten years. But after 30-year debt, ten-year debt is the next most expensive to service. Longer-dated bonds tend to have higher coupons, because investors normally ask for more interest when they lend for longer dates.
The shifting internal bond market structure has helped wreak havoc with bond managers this year. As some bond managers began hoarding long-dated government bonds, long-term interest rates remained sticky despite three interest rate hikes by the Federal Reserve. That helped to make government bonds the best-performing fixed-income sector this year. Of course, bond managers who were rebalancing to match the Lehman index were lightening up on governments and loading up on corporates and other bonds that underperformed the government bonds they were fleeing. To complicate matters, investors who switched some assets from Treasuries to agencies got hit by a political crosscurrent when everyone from Fed chairman Alan Greenspan to Representative Richard Baker of Louisiana attacked so-called government-sponsored enterprises for their implied subsidies. Agency bonds, for example, sold off massively after Gary Gensler, Treasury undersecretary for domestic finance, lent his weight to proposed legislation that would weaken the agencies’ government backing. So much for the simple bond market.
International bond managers have faced a slightly different set of issues. A shrinking worldwide government bond supply has forced many of these investors to load up on Japanese government bonds. Unfortunately, with the exchange rate now standing at just over ¥100 to the dollar, most now consider the Japanese currency overvalued. With the 20-year bond yielding only 2.16 percent, these positions may carry very little upside and lots of risk.
Any type of market, of course, will inevitably produce winners. Seven of the top-ten-performing fixed-income mutual funds in the first quarter were Treasury bond funds, according to Morningstar. Bonds that held discounted zero-coupon Treasury bonds did especially well, even though these funds would normally have suffered in a rising rare environment because of their long duration. This year the shortage of Treasuries overcame those effects, and so far dumb luck has contributed more to performance than has good judgment. After all, government bonds have underperformed the other constituents of the Lehman index for years. “It is true that some of our funds benefited from the notion that Treasury supply will be reduced in coming years,” concedes David Schroeder, a portfolio manager whose American Century target-maturity 2015, 2020 and 2025 funds occupy three of the top ten slots, says Morningstar.
Bond managers are trying to make some reasonable predictions about the size of the U.S. Treasury market going forward, but forecasting future supply will not be easy. Texas Governor George W. Bush has already said that if he becomes president, he will reduce the surplus by cutting taxes, which would presumably slow the disappearance of Treasuries. Many believe that predictions of the market’s complete disappearance are far fetched. “The circumstances are very unusual. To project this forward is somewhat silly,” argues CDC’s Wilford. “They’re getting more receipts, a higher percentage tax take, than in any year since World War II.”
But difficult markers foster creativity. And bond dealers and investors have come up with some innovative stopgap measures to deal with the marker’s shifting fundamentals.
On the corporate finance side, confusion has disrupted the new-issue market. Corporate bond issuance is down roughly 25 percent year on year. One reason for the fall is that new issues were unusually high last year when corporations borrowed heavily to head off Y2K worries. But market professionals mostly blame market turmoil. “The major driver has been volatility,” says Bear Stearns’ Noujaim, “although we have not seen as much M&A activity that’s debt-financed.” A few major issuers have attempted to overcome the problem by marketing new offerings priced off similar corporate bonds rather than Treasuries, whose gyrations have been responsible for much of the uncertainty. DaimlerChrysler, for example, started a trend last August, when it priced its $4.5 billion offering off Ford Motor Co.'s global benchmark bonds (Institutional Investor, September 1999).
Whether another benchmark can permanently replace Treasuries in the hearts and road shows of bond underwriters remains to be seen. Swaps, agencies and corporates have all been used as new-issue benchmarks since last fall. And along with those Treasury bonds that remain unaffected by supply constraints, they will probably all be the benchmarks of the future. All of them have good claims to become the bond manager’s main yardstick of value. The swaps market is now considered to be as liquid as the Treasury market and more liquid than the agencies market. Swap spreads tend to follow corporate spreads quite closely. But agencies trade far closer to the government’s risk-free borrowing rate than swaps. High-grade corporate bonds used to be the bond market’s benchmarks before Treasuries became so prevalent, and deeply liquid issues like Ford’s securities have recently revived their claim to benchmark status.
In the past Treasuries have been the hedge of choice for corporate and mortgage bond positions, but the shortage of new bonds makes them less reliable and more expensive. The result has been a risk management nightmare. “Certainly, from a benchmarking or hedging point of view, people are rethinking it a lot—from a perspective of what the relationship is between corporate bonds and Treasuries, but also swaps and Treasuries,” says Seth Waugh, head of global markets at Deutsche Bank. “People are saying, ‘We were trying to mitigate interest risk, but maybe we were piling it on.’” Increasingly, bond managers are turning to the swaps market as a replacement.
Many bond managers have also beefed up their basic credit analysis as credit spreads become an increasingly unreliable gauge of creditworthiness. With the market no longer functioning as a reliable measure of credit quality, bond buyers must start tearing apart balance sheets and making credit calls themselves.
Lehman Brothers and other dealers are trying to alleviate the pressure on global bond managers by creating global indexes that would include more than just government bonds. Both the newly launched Lehman Brothers global aggregate index and the Lehman Brothers universal index have lower weights of JGBs than traditional international government bond indexes. The JGB weighting of 25.50 percent in Lehman’s global Treasury index, for example, is reduced to 13.58 percent for all yen products in the Lehman global aggregate.
In the long run the money managers will have to grapple with the practical and philosophical meaning of the bond market’s shifts. Risk and return, after all, are the ultimate relative measures. And if the risk profile of the bond business has permanently changed, every fund manager needs to adjust his basic risk-allocation guidelines and rules of thumb. Without such adjustments, many institutional investors will end up in a lockstep march up the risk curve.
As bond professionals can attest, these are not easy questions. “We continue to look at absolute spreads against Treasuries, but we are starting to track day-to-day movements in swaps and agencies,” says Angelo Manioudakis, a principal at Morgan Stanley Dean Witter Investment Management, which manages $123 billion in bonds. “We are still as confused as anybody else.”
