S&P 500 Median Distance to Default (Merton)

## What this measures

The distance to default of the typical non-financial S&P 500 company: half are further from their default point, half closer. Merton (1974) model, solved the KMV way: each company's equity is treated as a call option on its assets, struck at a default point of short-term debt plus half of long-term debt (finance leases included, operating leases not) from its latest 10-Q or 10-K that was public that day. Equity value = the day's close × the current SEC share count; equity volatility = the last 252 daily returns, annualized; risk-free rate = the 1-year Treasury yield; horizon one year. The two model equations are solved for asset value and asset volatility, and distance to default = (ln(V/F) + (r − σ²/2)) / σ: the number of standard deviations the company's asset value sits above its default point. Published only on days at least 85% of non-financial members have a reading.

## Why it matters

Credit spreads for most companies are licensed data. Distance to default reads the same risk out of the stock market and public filings: it falls when stock prices drop, volatility rises or debt grows, which is what credit markets price.

## How to read it

Higher is safer. Watch direction: a falling median means equity cushions are shrinking across the market. Compare with fed_gz_spread and the CMDI.

## What it does not say

It is a model reading, not a rating or a forecast of default. The Merton probability (N(−DD)) is risk-neutral and badly calibrated in the tails; use the distance for ranking and change over time. Share counts are today's, so earlier days ignore buybacks and issuance. Financial companies are excluded (deposits are not debt in this sense). Companies that report no debt, or tag it only with their own XBRL extensions, have no default point and are left out (29 of 425 when this was built). Today's index membership is applied to past days. Volatility is an equal-weighted year of daily returns, so a volatile week lowers the reading for a year and then drops out at once, which moves the line in steps (April 2025 in, April 2026 out).

## Source

Palanor calculation from SEC EDGAR XBRL filings (debt, public domain), licensed daily closes and the U.S. Treasury 1-year yield.