A New Method for Estimating Within-Industry Corporate Default Correlation

• Activity Number: 519 - SPEED: Methodological Advances in Time Series: BandE Speed Session, Part 2
• Type: Contributed
• Date/Time: Wednesday, July 31, 2019 : 10:30 AM to 11:15 AM
• Sponsor: Business and Economic Statistics Section
• Author(s): Gary Witt* and Marcus Sobel
• Companies: Temple University and Temple University
• Keywords: Default Correlation; Corporate Default Distribution; Correlated Binary Data; Clustered Binary Data

Abstract

The default probability of corporate debt (bonds and loans) can be estimated from credit ratings, credit spreads, prices or related market values. The default distribution of portfolios of corporate debt depends on the default probability of each debt instrument but also on the correlation between them. Debt from corporations in the same industry may be more correlated than debt from different industries. Using market variables to estimate this within-industry default correlation is non-trivial.

This paper describes a new method for direct estimation of the within-industry default correlation from historical default data. A distribution previously used for modeling correlated, clustered binary data from toxicity studies is applied to default data from Moody’s Default and Recovery Database for seven overlapping ten-year periods beginning with 1995-2004 until 2007-2016. The distribution assumes exchangeability and is applied to data from a one rating category at a time. For each time period, maximum likelihood estimation is used to estimate the two parameters, default probability and default correlation, for each of three each rating categories: Baa, Ba and B.