Content area
Full text
This paper is the first in a series examining the credit quality of an evolving sample of seasoned, medium to high-grade, long-term railroad bonds during the late 19th and early 20th Centuries. The paper develops a simple model capable ofpricing these bonds and assembles the requisite financial statement and market data needed to calibrate the model. The exercise demonstrates that these bonds were priced according to the financial strength of their issuer and their priority of claim (and modified by other factors, many of which are interesting in themselves). These findings suggest that these bonds can be ratedfrom information available in real-time before Moody's innovation of bond ratings in 1909.
Keywords: bond rating, bond quality, priority of claim, regional interest rate differentials
INTRODUCTION
The history of interest rates in the United States is complicated by periods during which bonds of unquestioned credit quality weren't actively traded. For the period between the Civil War and WWI, we have mainly relied on Frederick Macaulay's (1938) Basic Yields. These yields are averages of an evolving sample of actively-traded, long-term,high-grade railroad bonds. As helpful as these yields have been, they are deficient. As Thies (2005) showed, Macaulay's sample includes bonds of heterogenous quality (within the range of "high quality"), and this heterogeneity masks the effect of the Gold Clause on bond yields during the period of Silver Agitation. Specifically, by being more careful in selecting currency bonds and gold bonds, Thies showed that the gold bonds were priced at a premium relative to currency bonds during Silver Agitation.
Similar to Macaulay's Basic Yields are David Durand's (1942) Basic Yields. See also Durand (1958) and Durand and Winn (1947). Durand's yields are monotonic envelopes of the yields of all actively-traded bonds arranged by term to maturity. These yields give an approximate idea of the slope of the yield curve. But, to suppose the Durand Basic Yields are precise measurements enabling, e.g., the inference of forward interest rates, would be a mistake (Durand, 1958, pp. 349-353).
In particular, Durand over-smoothed the yield curve, emasculating any humped yield curves during his study period. Despite Durand's warning, researchers have relied on his yields. Reuben Kessel (1971, p. 366) said, "Before the 1930s, judging by Durand's data, liquidity premiums were much smaller or...