THE MARKET VALUE OF DEBT, MARKET VERSUS BOOK VALUE OF DEBT, AND RETURNS TO ASSETS

This paper measures the market value of long-term debt and assesses ho w using book values of debt as proxies for market values can have seri ous effects in empirical work. Market values of debt are estimated fro m the Lehman Brothers Fixed Income Data Base, which has become availab le only recently. This database contains dealer quotes for end-of-mont h bid prices, and generally these quotes are of acceptable quality. In a significant number of cases, we have quotes on 90% or more of a fir m's outstanding debt issues. The role of debt in corporate decisions a nd performance is a major concern for firm management, investors, and financial scholars. Estimates of debt are used to answer important que stions on the optimal capital structure for a firm, what determines th e actual capital structures observed, and what the firm's overall cost of capital is, a measure that is typically backed out by using data o n firm capital structure. Despite the substantial interest in such iss ues, empirical research usually relies on book rather than market valu e of debt. This reliance arises primarily because of the difficulty of obtaining quality estimates of the market value of firm debt. Because of this difficulty, there has been little work on the nature, extent, and seriousness of the errors introduced by using book rather than ma rket value of debt. This paper begins to close the gap left by empiric al studies and allows investors and corporate decision makers to make sensible assessments of the data to apply the book value of debt. Book values sometimes seriously mismeasure market values of debt; not surp risingly, this mismeasurement is associated with changes in bond-marke t yields. We focus on three key empirical issues to explore how the us e of book values of debt affects empirical results. First, mismeasurem ent can influence cross-sectional studies of capital structure, though we find evidence that the errors introduced may not be important. Sec ond, mismeasurement can influence time series studies of capital struc ture: this influence can be quite important. Third, calculation of a f irm's overall cost of capital requires estimates of the value of debt: we find that mismeasurement can have important effects on these estim ates. We show how some empirical results are sensitive to using market values of bonds, but others are not, depending on the time period and questions considered. As an example, we compare estimates of capital structure that use hook versus market values of debt over the period 1 978-1991. We present two estimates of aggregate debt-to-value ratios f or large firms that on average issue investment-grade debt. One estima te uses our measure of market value of debt, the other book value; bot h use market values of equity. We also provide summary statistics that describe the behavior of capital structure over time for 15 industrie s. In the early part of our study period, estimated long-term-debt-to- value ratios based on book and market values of longterm debt diverge substantially. as expected, differences in book and market capital str ucture are associated with changes in the level of interest rates. In studies that explain capital structure differences across firms at a p oint in time, use of book rather than market value of debt introduces measurement error, but the associated problems may not be severe. Acro ss 15 industries, correlating debt-to-value ratios using book and mark et values for bonds is almost perfect at any point in time. However, w ork with data across industries and time simultaneously can contain su bstantial measurement error, because the time variation of the book va lue of debt often deviates substantially from the time variation of th e market value of debt. Deviations of the two measures arise in major part because changes in yields affect the market value of debt. The ri sk of equity or debt depends on the mixture of the two used in financi ng a firm's assets. In this sense, the risk is likely to be important in understanding the combined risk the firm faces, rather than looking separatory at its debt and equity. For example, the expected rate of return and volatility of a firm's assets are more likely to be time-co nstant than those of either equity or debt alone, because equity and d ebt's expected rates of return and volatilities vary not only with tho se of the firm's assets, but also with changes in the firm's debt-to-v alue ratio. Further, the majority of studies that consider equities an d bonds together use book rather than market values of bonds, another possible source of error. One illustration of these problems involves estimated costs of capital for 15 industries, found in a CAPM-based ap proach using estimated market-model betas. Differences in results acro ss estimation methods can be large. In practice, bond prices are often not available. Bond betas are guessed rather than estimated, which ca n lead to even greater discrepancies. If the capital structure changes a lot over time, it is probably better to first estimate the firm's o verall cost of capital rather than starting with separate estimates of equity and debt cost of capital.