When it comes to corporate defaults, the current recession is, as of now, child's play in comparison not only to the Great Depression, but another little-recalled period: the railroad crisis of 1873-1875.
New research from Stanford shows that over the past 150 years, the U.S. corporate bond market has repeatedly suffered clustered default events. In a surprise finding, the study reveals that default episodes are only weakly related to bad business downturns. Rather, they seem to be a function of permissive bankruptcy laws.
Ilya Strebulaev, associate professor of finance at Stanford GSB and Spence Faculty Scholar for 2010-2011, and three other researchers have developed the first set of data on bond defaults and debt prices reaching back to the 19th century. Their information, much of it hand-collected from historical financial records, has allowed for the first analysis of the relationship between various macroeconomic conditions — business cycles — and financial cycles, such as banking crises.
"Although our findings go back a century and a half, they allow us to reflect upon various theoretical models that are quite relevant for today's environment," says Strebulaev, who conducted the study with Kay Gieseke, assistant professor of management science and engineering at Stanford, Stephen Schaefer of the London Business School, and Francis Longstaff of UCLA's Anderson School of Management.
The study shows that the recent economic downturn in some senses is more similar to the railroad crisis of the 1870s than it is to the Great Depression. "The 19th century debacle involved the collapse of the entire economy, whose engine was the transportation industry — railways and related businesses such as steel production. The railroad boom was akin to the telecom and internet booms," explains Strebulaev.
Yet, from the perspective of bond investors, the 19th century corporate bond default rate was much higher and more volatile than that of today. Default rates peaked at 17% annually in the 1870s, an extremely high rate. In contrast, during the depression of the 1930s, the highest bond default rate was 8%. "The reason it wasn't higher," says Strebulaev, "is that, although a lot of firms defaulted, they weren't the largest ones."
In the current recession, bond default rates have been only 3% annually — still high but lower than those of previous years.
Interestingly, the investigators found that between the 1930s and the 1980s there were almost no corporate defaults. "Then, in the 1980s, you again see a wave of defaults, although less dramatic than in the past," Strebulaev says.
The investigators provide an explanation for the low default rate during that 50-year period. "Prior to the 1930s, there really was no bankruptcy code in the United States and instead courts concocted a procedure to deal with defaulted firms called equity receivership. As a result, what happened in the case of defaults looked similar to what happens with Chapter 11 today — debtors were treated in a more friendly way, while creditors were treated in a less friendly way," Strebulaev says.
After the Great Depression, however, bankruptcy codes were reformed to treat debtors more harshly and creditors more favorably. Then, in 1978, a new bankruptcy code in the United States again reversed the trend, making it easier for corporations to default.
The study shows that the period from the 1930s to the 1980s was thus a time in which it became extremely unpalatable for corporations to default. After the 1978 bankruptcy law change, numerous risky companies began issuing the famous "junk bonds," and the stage was once again set for a wave of defaults.
"Our data should be useful to policy makers, who are involved in a substantial debate about what bankruptcy law should look like," says Strebulaev.
The paper also demonstrates that, contrary to popular assumptions, corporate bonds are not a recent phenomenon. In the 19th and 20th centuries, the corporate bond market was very liquid — sometimes even more so than the equity market. Indeed, many corporations financed themselves primarily with bonds. "The total number of bonds traded then is similar to the number issued as recently as 10 years ago," says Strebulaev. "So, in fact, bonds were even more important then, than now."
The study also finds that default cycles are only weakly related to business cycles — including recessions.
"This is important because there has been considerable debate in macroeconomic research on the influence of financial constraints on macroeconomic factors, such as growth," says Strebulaev. "We show that if you measure financial constraints as the incidence of corporate defaults, there is not much of a link to the economy as a whole. Such defaults are more closely tied to the laws and institutional environment of the day."
The paper also contributes to the role of the government in the economy. "In the great debate about how active the government should be in the economy, some argue that it should remain passive. One consequence of this that should be taken into account, however, is undesirable higher volatility, and corporate default rates of the 19th century provide a good example," Strebulaev concludes.