The dependent and independent variables were collected for each month between July 1988 and January 1999 (May 1991 - Jan 1999 for AAA and BBB 10-yr spreads). All changes described below are month-to-month. The following tables describe the variables used in the analysis. Please note that, in the second table, the expected correlation is to the credit spread, not the price of credit-risk securities. In other words, an increase in a positively correlated variable would cause an increase in the credit spread, or a drop in credit-risk security prices.
Dependent Variable |
Source |
Form |
Explanation |
Credit spread - aggregate index |
Datastream |
Spread (yield) change |
Value-weighted index of the middle yield on U.S. corporate bonds less the yield on the current 10-year Treasury. The index includes all maturities and investment grade credit ratings. |
Credit spread - AAA, 10-year |
Bloomberg |
Spread (yield) change |
Value-weighted index of the middle yield on 10-year corporate bonds rated A3 and above less the yield on the current 10-year Treasury. |
Credit spread - BBB, 10-year |
Bloomberg |
Spread (yield) change |
Value-weighted index of the middle yield on 10-year corporate bonds rated BBB less the yield on the current 10-year Treasury. |
Independent Variable |
Source |
Form |
Explanation |
Expected Correlation |
Lagged credit spreads |
See above |
Yield change |
Is the autocorrelation between current and lagged credit spreads predictive? |
Unknown |
S&P 500 Composite Index |
Datastream |
Percent change (monthly return) |
The equity market could be relevant to corporate bond credit spreads in several ways: as an alternate investment, as a measure of capital markets investment levels, as a yield-producing security. |
Negative. The equity market as an alternate to fixed income investment is the most plausible in today's environment. |
S&P 500 volatility |
Datastream |
6-month percent change (calc'd from percent change of Index levels) |
See above. |
Positive. Increases in equity price volatility would likely cause risk averse investors to change to the relative safety of fixed income securities. |
S&P 500 dividend yield |
Datastream |
Yield change |
See above. |
Positive. Increases in the dividend yield makes the equity market more attractive as a yield-bearing investment. |
Interest rates - yield on the current three month T-bill |
Datastream |
Yield |
Are there any specific effects of the interest rate level on credit spreads? |
Unknown |
Bond volatility |
Datastream |
6-month yield change (calc'd from the month-to-month yield change on current Treasury) |
Volatility is used as a measure of risk. As actual prices are not applicable from on-the-run to when-issued Treasuries, yield was used in the calculation. |
Negative. The volatility of Treasury prices can be interpreted as a measure of the risk for the entire fixed income market. As volatility increases, the risk averse investor will withdraw his exposure. |
Yield curve changes (6mo- 3mo, 1yr-6mo, 2-1, 3-2, 5-3, 10-5, 30-10) |
Datastream |
Spread (yield) change |
Many economists believe that an upward-sloping yield curve (higher rates for longer maturities) indicates higher interest rates in the future. |
Positive. The expectation of higher interest rates in the future would suggest lower prices (higher spreads) on fixed income securities. The degree to which yield curve changes are captured in Treasury yields versus credit spreads is unknown. |
Swap spreads |
Datastream |
Spread (yield) change |
The swap rate is a proxy for the AA credit rate. As the swap market is significantly deeper and more liquid than that for corporate bonds, swap rates may provide a forward indication of the direction of credit spreads. |
Positive. As swap spreads increase, so should credit spreads. |
TED (Treasury - Eurodollar) spreads |
Datastream |
Spread (yield) change |
TED spreads capture the credit rating of the United States economy. |
Negative. The TED spreads used in this analysis are always negative (T - ED). A positive change in the spread is a tightening of the spread. If TED spreads tighten, the U.S. economy is attractive to foreign investment. Credit spreads should tighten. |
Consumer confidence, Consumer expectations, Consumer confidence - present situation |
Datastream |
Percent change |
Consumer confidence is a forward indicator of consumer spending and, therefore, GDP growth. |
Negative. Increases in consumer confidence should indicate that there is more investment capital available. Thus, credit spreads should tighten. |
Industrial production |
Datastream/ Bloomberg |
Percent change |
Industrial production is a forward indicator of GDP growth. |
Unknown. Increases in this measure could mean a healthy U.S. economy or be an inflationary warning. The effect on credit spreads would vary. |
See the model.
Return to Top of Page
Return to Main Page Contents
Return to the BA453 Homepage
Go to Campbell Harvey's Homepage
Comments? Send a message to the webmaster.