Back on Nov. 21 I said in a post that was refining some research that I had done on the performance of high yield bonds (as an asset class).
As with other assets, high yield bonds offer the chance to reduce the risk of a portfolio (the fluctuation in annual returns) because their annual returns are imperfectly correlated with those of other asset classes. I'll note up front that the comparisons between high yield bonds and other asset classes will vary depending upon the time period studied and the index used. There are several indexes of high yield bond performance (and also of small cap stock performance), and none are as generally accepted as the S&P 500 for large cap stocks. But I'm sure that essentially the same relationships would appear regardless of the indexes that are used to measure them.
A few years ago, T.Rowe Price presented, in its newsletter, the following data on correlation coefficients between high yield bonds and other asset classes. The data was based on just 5 years ending September 30, 1996. The correlation of high yield bonds with "high grade" bonds was R=0.46; with "large cap" stocks, R=0.28; and with "small cap" stocks, R=0.37. Although the data set was pretty limited, it was sufficient to illustrate that the correlation between high yield bonds and other bonds was only modest, and that the correlation with stocks was low (although higher with small caps than with large caps).
It occurred to me, though, that there might be a higher correlation between high yield bonds and a mix of other bonds and stocks, since high yield bonds are subject to both interest rate risk (like bonds) and solvency risk (like stocks). And sure enough! Here's what I found, based on 18 years of data from 1985 through 2002. I used the Russell 2000 as the index of small cap stocks, and the Lehman High Yield Index as the index of high yield bonds (which is used by Vanguard as the benchmark for its High Yield Corporate Fund).
Through trial and error, I found that the strongest correlation occurs between high yield bonds and a portfolio consisting of 67% small cap stocks and 33% long term corporate bonds (as measured by Ibbotson Associates). The correlation in the annual returns is R=0.86, which is quite high.
Furthermore, the risk/return characteristics over the period were very similar. The small cap/long term corporate bond mix had an average annual return of 9.8% with a standard deviation of 13.1%, while the high yield bonds had an average return of 11.1% with a standard deviation of 13.6%. (Using a different index for small cap stocks would indicate a slight advantage for the stock/bond mix.)
The general conclusion here is that, over time, market efficiency does its thing and prevents high yield bonds from performing much differently than a mix of assets that are subject to similar economic risks.
Including high yield bonds in a portfolio can still be expected to have the benefit of reducing its risk slightly. But in deciding whether to include high yield bonds in a portfolio, tax considerations become more important, in that most of the gain from high yield bonds is taxable interest, while most of the gain from a mix of small cap stocks and corporate bonds would be capital gains.
As with other assets, high yield bonds offer the chance to reduce the risk of a portfolio (the fluctuation in annual returns) because their annual returns are imperfectly correlated with those of other asset classes. I'll note up front that the comparisons between high yield bonds and other asset classes will vary depending upon the time period studied and the index used. There are several indexes of high yield bond performance (and also of small cap stock performance), and none are as generally accepted as the S&P 500 for large cap stocks. But I'm sure that essentially the same relationships would appear regardless of the indexes that are used to measure them.
A few years ago, T.Rowe Price presented, in its newsletter, the following data on correlation coefficients between high yield bonds and other asset classes. The data was based on just 5 years ending September 30, 1996. The correlation of high yield bonds with "high grade" bonds was R=0.46; with "large cap" stocks, R=0.28; and with "small cap" stocks, R=0.37. Although the data set was pretty limited, it was sufficient to illustrate that the correlation between high yield bonds and other bonds was only modest, and that the correlation with stocks was low (although higher with small caps than with large caps).
It occurred to me, though, that there might be a higher correlation between high yield bonds and a mix of other bonds and stocks, since high yield bonds are subject to both interest rate risk (like bonds) and solvency risk (like stocks). And sure enough! Here's what I found, based on 18 years of data from 1985 through 2002. I used the Russell 2000 as the index of small cap stocks, and the Lehman High Yield Index as the index of high yield bonds (which is used by Vanguard as the benchmark for its High Yield Corporate Fund).
Through trial and error, I found that the strongest correlation occurs between high yield bonds and a portfolio consisting of 67% small cap stocks and 33% long term corporate bonds (as measured by Ibbotson Associates). The correlation in the annual returns is R=0.86, which is quite high.
Furthermore, the risk/return characteristics over the period were very similar. The small cap/long term corporate bond mix had an average annual return of 9.8% with a standard deviation of 13.1%, while the high yield bonds had an average return of 11.1% with a standard deviation of 13.6%. (Using a different index for small cap stocks would indicate a slight advantage for the stock/bond mix.)
The general conclusion here is that, over time, market efficiency does its thing and prevents high yield bonds from performing much differently than a mix of assets that are subject to similar economic risks.
Including high yield bonds in a portfolio can still be expected to have the benefit of reducing its risk slightly. But in deciding whether to include high yield bonds in a portfolio, tax considerations become more important, in that most of the gain from high yield bonds is taxable interest, while most of the gain from a mix of small cap stocks and corporate bonds would be capital gains.