Uncertainty abounds with regards to future interest rate movements in the long end of the Treasury curve. For obligations that often serve as a proxy for the "risk-free rate" in finance, given the full faith and credit backstop of the U.S. government, the returns of Treasury bonds in recent periods have been increasingly variable. The year-to-date sell-off in the long end has led to some market pundits suggesting the long awaited swap from Treasury bonds to riskier asset classes is finally underway. While the future directionality and amplitude of yield curve movements is uncertain prospectively, we can look back at the variability of Treasury returns historically and gauge how risky this part the thirty-year end of yield curve is relative to other investment classes. The results may surprise.
click to enlarge
The variability of returns of thirty year Treasury bonds have been roughly equivalent to single-B rated corporate bonds over long time periods. Single-B rated corporate bonds are four to six steps below investment grade. Moody's lists the average annual issuer-weighted corporate default rates of bonds with this rating class at 3.42%. The five largest single-B rated issuers in the Merrill Lynch U.S. High Yield Master Index II include: 1) Ally Financial, the former General Motors Acceptance Corporation, which received substantial federal assistance in the downturn and recently sported the worst performance in the Federal Reserve's stress tests; 2) Sprint, the highly leveraged and unprofitable telecommunications provider; 3) CIT Group, the commercial lender which filed for bankruptcy protection in late 2009, but has subsequently re-emerged; 4) HCA, the world's largest private hospital operator, which completed a then record leveraged buyout in 2006, and is still trying to work down its high financial leverage from this transaction, and 5) MGM, the second largest gaming company in the world, which came under tremendous stress in 2009 as gaming revenues dropped amidst the company's debt-fueled Las Vegas CityCenter project.
By listing some of the larger single-B issuers, I have attempted to illustrate the credit risk inherent in this portion of the bond universe to investors less familiar with speculative grade credit. The interest rate risk inherent in long U.S. Treasuries, which are universally recognized as the flight to quality instrument in times of credit stress, must be roughly equivalent to the credit risk inherent in these highly leveraged issuers. Below are summary statistics for the indices used in this analysis.
Duration has a dual meaning in fixed income parlance. It is the weighted average timing of bond cash flows (future coupon payments and the return of principal), but also the interest sensitivity of a bond for a parallel shift in the yield curve. If the yield on the thirty year Treasury increased 100 basis points, the long U.S. Treasuries would see approximately a negative 15.9% return. High yield corporate bonds typically have much shorter durations, so contain less interest rate sensitivity, but do have higher levels of credit risk, which is reflected in the option adjusted spread.
From our historical return series, we see that the greater credit risk in single-Bs has been roughly equivalent to the greater interest rate sensitivity in 30-yr Treasuries. Investors should demand equivalent compensation for equivalent levels of risk, but currently holders of 30-yr Treasuries are getting paid nearly four percent less than holders of single-B bonds as seen through their relative yields. Investors must be assuming that future interest rate fluctuations for the thirty-year Treasury will be more muted than credit spread swings in single-B rated bonds.
As the economy slowly recovers, high yield defaults should remain relatively muted. Despite the economic uncertainty emanating out of Europe, corporate defaults hit a four year low in 2011. In this environment, high yield corporate bonds are likely to continue to outperform. Treasury bonds are like to underperform if continued economic growth drives inflation higher. Currently, Treasury bond yields through the ten year maturity are less than or equal to estimates of domestic inflation, which indicates that investors nominal coupon return on Treasury notes is completely offset by the depreciation in the purchasing power of their dollar. It makes little sense that investors would accept zero real returns for an investment in Treasury securities that appears likely to provide negative and more variable future nominal returns.
Long U.S. Treasury bonds have generated tremendous returns through the credit crisis, and generally through the past generation, as U.S. government bond yields have fallen. Surprisingly, risk-adjusting these returns to take into account their variability would have still made long U.S. Treasury bonds an inferior asset than BB and B-rated corporate bonds. Currently, U.S. Treasury bond yields do not adequately compensate investors for the variability of these returns and the likelihood of higher rates. Investors should shorten the duration of their bond portfolios, and look to go down the quality spectrum to earn higher risk-adjusted returns.