Magazine article Business Credit

Using State-of-the-Art Statistics to Manage Global Corporate Credit Risk

Magazine article Business Credit

Using State-of-the-Art Statistics to Manage Global Corporate Credit Risk

Article excerpt

The Kamakura Troubled Company Index, first published in 2003, measures the percentage of public firms worldwide that have annualized one-month default probabilities of more than 1%. Kamakura considers any firm with a short-term default probability of more than 1% to be "troubled" since they fall in the bottom 13% of firms in credit quality over the period from 1990 to the present. The trends in the index since 1990 track the major global credit downturns very precisely (see Figure 1).

Modern statistical methods and "big data" information technology infrastructure allowed Kamakura to leapfrog the traditional practices of the legacy rating agencies very quickly. In a sense, the index is a big international credit portfolio. The next few paragraphs explain why this is so and how a modern credit portfolio manager can use the same insights on a daily basis.

At the time the index was launched, the common mindset of traditional credit analysts was focused on a few key aspects of credit assessment: annual time periods, fundamental analysis and a small number of credit grades of great stability, with qualitative aspects of credit analysis largely outweighing quantitative considerations. Professors Gary King of Harvard and Samir Soneji of Dartmouth in their article about the Social Security System (1) in the United States provided a comment that could just as well have described the traditional approach to credit analysis: "... informal forecasts may be intuitively appealing, ... but they suffer from humans' well-known poor abilities to judge and weight information informally.... Indeed, a large literature covering diverse fields extending over 50 years has shown that formal statistical procedures regularly outperform informal intuition-based approaches of even the wisest and most well-trained experts...."

Finding a Better and Faster Way to Assess Credit Risk

Kamakura took quick action to take full advantage of modern statistics, information technology and financial economics with their default probability service. They implemented 10,000 "ratings," running from 0.00% to 100% in one-basis point increments. The default probability term structure varies by maturity, unlike rating agencies that haven't linked a maturity to ratings in the last 150 years. Instead of putting out a press release each time a legacy rating changes, Kamakura updates the default probabilities each day, using the most recent information available.

Geography is also important. Today, ratings are dominated by the American origins of the rating agencies. Of the 2,582 firms rated by one agency, 1,219 of those firms are based in the United States. Kamakura covers 38,602 firms in 67 countries. U.S. firms are less than 1/7 of the total, at 5,490 firms. The modeling effort recognizes the links between public firms, banks that are subsidiaries of public firms, sovereigns and nonpublic firms. With an amazing insight called Stein's paradox, the company knows with certainty that modeling of firms internationally in a carefully integrated way produces more accurate default probability estimates than either modeling each country separately or using the same model for all firms worldwide. The same is true when firms are grouped by industry. A carefully specified model that analyzes all industries together but with some differences in coefficients and explanatory variables has been proven mathematically to produce the most accurate results. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.