Distress costs come in two sizes, and the small one gets all the attention.
Direct vs indirect costs of financial distress
Direct costs are the fees: insolvency professionals, lawyers, valuers, the resolution process itself. Published estimates for large firms cluster in the low single digits as a percentage of asset value, and they are the part everybody can see, because somebody invoices for them.
Indirect costs are the business decaying while the balance sheet is argued about, and they are far larger. Customers stop signing multi-year contracts with a supplier who may not be there to honour them. Suppliers move from 60-day terms to cash in advance. Capable people leave first, because they have the most options. The capital budget is cancelled and competitors price against a rival who cannot respond. None of it appears on an invoice, and all of it shows up in the cash flow.
How do you put a number on the cost of financial distress?
Published work on the total is a range, not a point. Take the total cost of distress, if it happens, as a stated fraction of unlevered value, defend the fraction, and run the answer at two other fractions to show what it does. The worked example uses 40% and reports the optimum at 25% and 55% as well.
Advertisement
The probability side comes from the credit rating, and the rating comes from interest coverage through the synthetic-rating lookup the valuation phase already built. That creates a small circularity worth naming: the spread determines the interest, the interest determines the coverage, and the coverage determines the spread. It is solved by iteration, starting from the narrowest spread and working down until the coverage a spread implies sits inside the band that spread belongs to.
The iteration has a wrinkle. Because the rating table is a step function, more than one band can be self-consistent at high leverage. In the worked example at three turns of EBITDA, starting from the top converges on B+ at a 3.00% spread and coverage of 2.632, while starting from the bottom converges on B at a 3.70% spread and coverage of 2.451. Both satisfy their own band. The convention used throughout, stated because a convention that is not stated is a hidden assumption, is to iterate from the best rating downward and take the first self-consistent band, which selects the higher rating. State which convention you used and the second answer stops being a surprise.