How External Analysts Use Distress Prediction Models
Corporate Financial Distress, Restructuring, and Bankruptcy by Edward I. Altman, Edith Hotchkiss, and Wei Wang (Wiley, ISBN 978-1-119-48180-5)
Previous: Z-Score Default Prediction and Applications | Next: Managing a Financial Turnaround with Z-Score
Chapter 10 listed where distress prediction models get used. Chapter 11 zooms in on the outsiders: lenders, investors, auditors, lawyers, regulators, and consultants who analyze distressed companies from the outside looking in.
Altman wrote this chapter because practitioners kept finding new ways to use his models. Some applications he expected. Others surprised him, like equity analysts logging more Z-Score lookups than bond analysts on Bloomberg and S&P Capital IQ.
Lenders: Pricing Risk, Not Just Approving It
The most obvious use is lending. Basel II and III pushed banks to build internal rating-based (IRB) models with explicit probability of default and loss-given-default estimates. Altman worries that U.S. regulators’ decision not to require most American banks to follow Basel II may have slowed adoption.
Loan pricing is where models earn their keep. Altman’s example: a five-year senior unsecured BBB loan with 0.3% annual expected default, 70% recovery, and various capital charges produces a required rate of 6.94% under economic capital rules vs. 7.69% under flat Basel I capital. Accurate PDs change the price. Getting them wrong means lending too cheap to risky borrowers or pricing yourself out of good ones.
Bond Investors: Quality Junk and Fallen Angels
About 22% of all defaulting issues from 1971-2017 started as investment grade. Nineteen percent of 2007-2017 defaults did too. “Investment grade” on the label does not mean investment grade on the balance sheet.
For distressed bonds trading 1,000+ basis points above Treasuries, the question is whether the company keeps sliding or stabilizes. In 2009, the average return on non-investment-grade bonds was about 60%, driven by distressed bonds that recovered toward par in the benign post-crisis cycle.
Altman’s “quality junk” strategy targets bonds in the upper-left quadrant of a return/risk chart: high yield spreads but strong bond-rating equivalents from the credit model. When the model says B+ but the market prices the bond like CCC, that gap is where the money sits.
Equity Investors: Avoiding Total Wipes
Equity analysts use Z-Scores more than bond analysts, which surprised Altman. The logic is capital preservation. When a company defaults, equity holders usually get wiped out.
Goldman Sachs ran a long/short strategy in 2008: long stocks with Z-Scores above 4.0, short those below 2.0. The long/short basket returned 12.9% from February to December 2008 while the S&P 500 lost 31.2%. STOXX launched “Strong Balance Sheet Indices” in 2014, selecting firms with three-year Z-Score track records above 3.5.
The strategy works best in stressed markets. In strong bull runs, low Z-Score firms that do not default often outperform. High Z-Score firms may have already peaked because market value of equity (X4) captures past stock performance.
Auditors, Regulators, and Lawyers
Auditors face going-concern decisions. In 1974, Altman found that only about 40% of bankrupt firms got a going-concern qualification the year before filing, while Z-Score flagged over 80% as distressed. Arthur Andersen built an “A-Score” model for audit risk. It was apparently not used on Enron, where Z"-Score showed single-B while agencies still rated triple-B.
Regulators need to evaluate bank credit systems under Basel III. Central banks like the Federal Reserve and Banque de France run their own scoring tools to monitor bank portfolios.
Bankruptcy lawyers use distress models for timing filing decisions and for legal arguments like the failing company doctrine (allowing antitrust mergers when one party would fail anyway) and deepening insolvency (claiming that prolonging a dying company enriches advisers at creditors’ expense). Altman argues that a Z-Score below zero combined with a KMV expected default frequency above 20% is a reasonable test for the “zone of insolvency.”
Government and Supply Chain Screening
The U.S. Department of Defense and other agencies screen vendors using Z-Score models. Poor scores trigger closer review or disqualification. Private companies with just-in-time supply chains should run the same screens on their suppliers. The 2004-2009 auto industry meltdown showed why: parts suppliers failed before GM and Chrysler filed.
My Take
This chapter reads like a catalog of every job in finance that touches a distressed company. What ties them together is the same insight: a single number from a transparent model beats gut feel when the downside is bankruptcy.
The equity-analyst surprise is the detail I keep thinking about. Stock pickers care about balance sheet strength more than many bond investors admit. And the Enron/A-Score anecdote is a quiet warning. Building a model is step one. Actually using it when it matters is step two. Andersen had the tool. They skipped it on the one client that destroyed them.
Chapter 12 flips the perspective entirely: what happens when management uses the Z-Score not to watch the company from outside, but to actively steer it away from bankruptcy.
Previous: Z-Score Default Prediction and Applications | Next: Managing a Financial Turnaround with Z-Score