Default Probability and Recovery Rate Basics

When a bond defaults, you don't lose everything. Senior secured bondholders historically recover 50-60 cents on the dollar; subordinated debt holders recover 20-30 cents. That gap (the seniority stack) determines whether default means a manageable haircut or a wipeout.

Credit analysis boils down to two questions: How likely is this issuer to default? And if it does, how much will I get back? Master these metrics and you'll understand why a BB-rated bond pays 150-250 bps more than investment grade, and whether that spread actually compensates for the risk.

Default Rates: The Rating System's Report Card

Rating agencies earn their keep by predicting defaults. Here's how accurate they've been (Moody's and S&P data, 1981-2024):

One-Year Default Rates by Rating:

• AAA/AA: 0.00-0.02% (essentially zero)
• A: 0.08%
• BBB: 0.26%
• BB: 0.97%
• B: 4.93%
• CCC/C: 27.98%

Notice the cliff between investment grade and speculative grade. Drop from BBB to BB and your one-year default probability quadruples. Drop to B and it jumps to nearly 5%. Hit CCC territory and you're looking at one-in-four odds of default within twelve months.

Five-Year Cumulative Default Rates:

• AAA: 0.39%
• BBB: 2.53%
• BB: 9.51%
• B: 22.30%
• CCC: 48.05%

The five-year numbers tell the real story for hold-to-maturity investors. A B-rated bond has better than one-in-five odds of defaulting before year five. A CCC bond is a coin flip.

Recovery Rates: The Seniority Stack

When defaults happen, creditors don't all get treated equally. The seniority hierarchy determines who gets paid first (and how much):

Historical Recovery Rates by Seniority (Moody's, 1987-2022):

The math is straightforward: senior secured bondholders recover 124% of what senior unsecured holders get. Subordinated holders recover only 51% of senior unsecured. That's the price of being lower in the capital structure.

(Industry matters too. Utilities with hard assets recover better than service companies with intangible value. Economic conditions matter: 2009 recoveries were notably lower than 2019 recoveries on similar debt.)

The Expected Loss Formula

Here's the calculation that drives credit analysis:

Expected Loss = Default Probability x Loss Given Default

Where Loss Given Default (LGD) = 1 - Recovery Rate

Worked Example:

You own $100,000 of a BB-rated senior unsecured bond:

• Default probability: 2% (roughly BB one-year rate)
• Recovery rate: 47% (senior unsecured average)
• LGD: 53% (1 - 0.47)
• Expected loss: $1,060 ($100,000 x 0.02 x 0.53)

That $1,060 expected loss translates to roughly 106 bps of credit spread compensation needed annually.

Now compare the same $100,000 in a senior secured position:

• Default probability: 2%
• Recovery rate: 61%
• LGD: 39%
• Expected loss: $780 ($100,000 x 0.02 x 0.39)

The senior secured position has 26% lower expected loss from the same issuer. That's why seniority commands tighter spreads.

Reading Spreads Through the Default Lens

Credit professionals use this approximation:

Credit Spread (approximately) = Default Probability x LGD

Rearranging: Implied Default Probability = Spread / (1 - Recovery Rate)

Practical Application:

A bond trades at 200 bps over Treasuries with an assumed 40% recovery rate:

• LGD = 60%
• Implied default probability = 0.02 / 0.60 = 3.33%

Is 3.33% reasonable for this credit? If the rating is B (historical one-year default rate of 4.93%), the market might be underpricing risk. If it's BB (historical rate of 0.97%), the market is pricing in significant concern.

(The real-world relationship is messier. Liquidity premiums, risk appetite, and technical factors all affect spreads beyond pure default math. But this framework gives you a starting point.)

The Credit Quality Checklist

Before buying any corporate bond, work through this assessment:

Tier 1: Default Probability

• Current credit rating and trend (stable, positive, negative outlook)
• Rating agency default probability tables for this rating
• Company-specific factors that might differ from rating averages

Tier 2: Recovery Analysis

• Seniority of the specific bond (not just the issuer rating)
• Secured vs. unsecured status
• Asset coverage and tangibility
• Position in capital structure relative to other debt

Tier 3: Expected Loss Calculation

• Calculate PD x LGD for the specific security
• Compare expected loss to credit spread received
• Assess whether spread adequately compensates for risk

What Actually Drives Recovery Rates

Recovery isn't random. Moody's research identifies the key determinants:

Seniority and security explain the largest variation. A secured creditor with collateral recovers more than an unsecured creditor (obvious, but quantifiably true at 124% relative recovery).

Debt cushion matters: more junior debt below you means better recovery. If subordinated debt absorbs losses first, senior holders recover more.

Asset tangibility affects outcomes. Manufacturing companies with physical plants recover better than professional services firms. Real estate recovers better than intellectual property.

Industry defaults cluster in recessions, when buyers for distressed assets are scarce. The 2024 default environment saw 5.1% speculative-grade default rates in the US, with distressed exchanges accounting for 63% of all defaults (the highest share on record).

Putting It Together: A Decision Framework

When evaluating a credit position, run this sequence:

Step 1: Identify the rating and look up historical default rates. A BB bond has roughly 1% one-year default risk and 9.5% five-year cumulative risk.

Step 2: Identify the seniority. Senior secured bonds recover roughly 61%; subordinated bonds recover 28%. The difference is enormous.

Step 3: Calculate expected loss. A BB senior secured bond with 1% default probability and 39% LGD has expected loss of 0.39% annually. A BB subordinated bond with the same default probability but 72% LGD has expected loss of 0.72% (nearly double).

Step 4: Compare to spread. If the senior secured bond pays 80 bps and the subordinated pays 150 bps, is that 70 bps spread difference worth the 0.33% higher expected loss? Probably yes (you're getting compensated 70 bps for 33 bps of extra risk).

This framework won't capture every nuance of credit analysis. But it will keep you from buying subordinated debt at spreads that don't compensate for the seniority risk, or from overpaying for senior secured bonds when spreads are too tight.

Source: Moody's Annual Default Study (1987-2024), S&P Global Default, Transition, and Recovery studies