bankruptcies in the United States.16 In our case, these EDFs are calculated monthly
and measure the probability that a firm will default on its debt obligations over the
next 12 months. We used EDFs as of the last month of the quarter when merging
MKMV data to the quarterly Compustat balance sheet variables.
It should be noted that MKMV does not disclose how the mapping between the
distance to default and the EDF is computed. However, these timely, forward-looking
measures of default risk are widely used by financial market participants when assess-
ing credit risk. One clear advantage of EDFs over the traditional measures of default
risk based, for example, on credit ratings stems from the fact that the dynamics of
EDFs are driven primarily by the movements in equity values. As a result, EDF-
based measures of credit risk have the ability to react more rapidly to deterioration
in the firm’s credit quality as well as to reflect more promptly changes in aggregate
economic conditions.
3.2 Descriptive Statistics
Table 1 contains several summary statistics for our panel. Despite our focus on
firms that have both equity and a portion of their debt traded in open markets, firm
size—measured by sales or market capitalization—varies widely in our sample. Not
surprisingly, though, most of the firms in our dataset are quite large. The median
firm has more than $670 million in sales and a market capitalization of about $2.1
billion. About one-half of observations are associated with leverage ratios greater
than one. The relatively high leverage in our sample is due in part to the steep fall
in equity prices that started in the spring of 2000, which significantly reduced the
market capitalization of firms, thereby driving up their leverage ratios.
16 The MKMV’s mapping of distances to default to EDFs restricts the probability estimates to the
range between 0.02 percent and 20 percent because of sparse data beyond these points.
15