A high-yield credit spread looks simple: you get paid extra yield over a Treasury bond to cover the risk of default. That is the story most people hear. But the full story is much richer. A spread is like an onion, and we need to peel back layers of liquidity risk, tax effects, and the scary math of default correlation.
| Layer / Component | Brief Description | Typical Driver |
|---|---|---|
| 1. Expected Default Loss | The actuarial cost of the issuer simply not paying you back. | Company leverage, cash flow |
| 2. Default Correlation Premium | Extra charge for the risk that many issuers default together in a crisis. | Macro environment, system leverage |
| 3. Liquidity Premium | Compensation for not being able to sell the bond quickly at a fair price. | Dealer inventory, market depth |
| 4. Tax / Regulatory Distortion | Spread adjustment caused by different tax treatments between corporate and Treasury bonds. | Tax code, institutional mandates |
You cannot just look at the total spread and call it "risk." Some of it is just market friction. Some of it is a bet on the weather of the whole economy. Let us walk through these layers one by one.
Layer 1: The Expected Default Loss
This is the easiest part to understand. If a bond has a 5% chance of defaulting and you lose 60% of your money when it does, your expected loss is 3% (5% times 60%). The spread should at minimum cover that 3%.
A pizza shop owner offers you a bond with a 20% yield. You discover the shop has a 50% chance of going bust in two years. Even though the yield is high, your expected return might still be negative after you factor in the loss. The high spread is just math, not magic.
However, the actual high-yield spread is almost always wider than just this expected loss. For decades, investors in high-yield bonds have earned an "excess return" above Treasury bonds even after accounting for actual defaults.
| Period | Average Annual Default Loss Rate | Average Spread over Treasuries | Excess Spread "Cushion" |
|---|---|---|---|
| Long-Term Average (1980-2023) | ~2.0% | ~4.5% | ~2.5% |
| Stress Year (2008) | ~6.0% | ~17.0% | ~11.0% |
| Calm Year (2017) | ~0.5% | ~3.5% | ~3.0% |
That extra cushion is not free money. It is the price of sitting through volatility and the risk that the default losses come all at once, right when your stocks are crashing too.
The raw probability of a single bond defaulting explains only a small part of the spread. The market charges a large premium for the timing and clustering of these defaults.
Layer 2: Default Correlation & "The Avalanche"
Here is where it gets really hard. Default correlation means that when one company fails, others often fail at the same time. Think of a snowstorm in a village. If houses are far apart (low correlation), one roof caving in does not affect the others. But in a tightly packed city row house (high correlation), the collapse of one roof triggers a chain reaction.
In 2009, General Motors and Chrysler both filed for bankruptcy within a month. If you owned bonds in both, your diversification was an illusion. Auto parts suppliers also defaulted. The correlation shot up to near 1.0. You owned different names, but the same economic disaster hit them all.
When correlation spikes, the value of a diversified high-yield portfolio drops like a stone. Suddenly, you are not looking at isolated defaults. You are looking at a systemic meltdown. The spread must compensate you for that tiny chance of a massive loss.
| Scenario | Default Correlation | Average Default Rate | Chance of Losing >15% of Principal |
|---|---|---|---|
| Independent Islands | 0.0 | 4% | ~1% |
| Normal Market | 0.2 | 4% | ~5% |
| Systemic Crisis | 0.7 | 4% | ~30% |
You can see the problem. The average default rate does not change, but the tail risk explodes. High-yield spreads blow out in a crisis mainly because everyone starts pricing that systemic correlation, not because they suddenly realized one company has bad management.
This is why high-yield bonds often behave like a shadow of the stock market in a sell-off. The correlation premium turns debt into equity-like risk.
A basket of risky bonds can look safe in normal times. But the moment the economy turns, correlations break the diversification model. A 50-bond portfolio can suddenly act like a single bet on the economy.
Layer 3: The Liquidity Trap
If you cannot sell the bond, do you really have a spread? The high-yield market is notorious for drying up. When the phones stop ringing, you are forced to accept whatever price a dealer gives you, just to get out.
A friend urgently needed to sell a rare vintage chair. The "market value" was $1,000. But in a rush, the antique shop gave him only $400. The $600 gap is the liquidity discount. In bonds, a credit spread often hides a similar liquidity tax, especially for smaller issues under $250 million.
Liquidity varies massively between a big "fallen angel" (a former investment grade company) and a small "zombie" issuer. The bid-ask spread can swallow weeks of accrued interest.
| Bond Category | Issue Size | Typical Bid-Ask Spread | Liquidity Score |
|---|---|---|---|
| Large "Fallen Angel" | >$2 Billion | 0.5% | High |
| Standard High-Yield | $500M - $1B | 1.5% | Moderate |
| Small Cap / Distressed | <$300M | 4.0%+ | Very Low |
A smart investor checks not just the yield but the round-trip cost. If you buy a small bond and the market stumbles, you might lose 10% just on the bid-ask gap before a single default even happens. That liquidity risk is baked into the spread.
Layer 4: Tax and Regulatory Noise
There is a technical layer that few people think about. In the U.S., Treasury bond interest is exempt from state and local taxes. Corporate bond interest is not. Because of this, the spread you see is partly just a tax shield adjustment.
Assume the state tax rate is 5%. A Treasury bond paying 5% is actually worth more than a corporate bond paying 5.5% to a high-net-worth individual in a high-tax state. The extra 0.5% covers the tax bill, not the credit risk. If you ignore taxes, you misread the spread.
Banks and insurance companies also have capital requirements that make holding low-rated bonds expensive. They demand a higher spread just to offset the regulatory burden, even if the default risk is low.
| Factor | Effect on Spread | True Risk Transfer? |
|---|---|---|
| State Tax Exemption of UST | Widens corp. spread by ~20-50 bps | No (Tax policy) |
| Insurance Capital Charges | Widens spread for low-rated (NAIC-3,4) | No (Regulation) |
| Benchmarking to On-the-Run UST | Artificial yield difference (financing specials) | No (Technical) |
Once you strip out these technical layers, the "pure credit spread" is often much smaller than it looks on the screen. The headline spread is an amalgamation of fear, friction, and fiscal policy.
Some of the spread just pays for taxes. Some of it pays for the cost of compliance. Do not treat the entire gap to Treasuries as a default risk premium.
Putting It All Together: The Real Decomposition
When you buy a high-yield bond with a 7% yield and the Treasury is at 3%, you have a 400 basis point spread. The market might tell you it is all about default. But the truth is more delicate.
Imagine the spread is a suitcase. You open it expecting to find only cash for default risk. Instead, you find a smaller stack of money for expected loss; a big, volatile stack for systemic correlation risk; a small, sticky note for liquidity costs; and a receipt for state taxes. The total looks the same, but the contents are very different.
Focusing only on the total spread number is what causes investors to panic. If the spread widens from 400 to 800, did expected defaults double? Probably not. Most likely, correlation expectations and liquidity premiums exploded.
Understanding the pieces helps you stay calm. It tells you that a wide spread might not predict a depression. It might just predict that dealers are scared and tax laws are punitive.
| Component | Estimated Contribution (bps) | Stable or Jumpy? |
|---|---|---|
| Expected Actuarial Loss | 120 bps | Stable |
| Default Correlation Premium | 150 bps | Very Jumpy |
| Liquidity Discount | 80 bps | Jumpy |
| Tax / Technical Distortion | 50 bps | Stable |
| Total Observed Spread | 400 bps | - |
The jumpy part is what makes markets risky. The stable part is what you can actually plan around. A good high-yield investor tries to harvest the stable parts while insuring against the jumpy parts.
Key Takeaways
| Key Point | What It Means | Action Item |
|---|---|---|
| Spreads are multi-layered | The yield gap includes default, correlation, liquidity, and tax factors. | Never treat a corporate spread like a pure default indicator. |
| Correlation drives tail risk | A portfolio suffers most when defaults happen simultaneously. | Stress-test your holdings for 2008-style correlation spikes. |
| Liquidity is a hidden cost | Small high-yield bonds carry a high round-trip transaction cost. | Prefer larger issues and fallen angels over small orphans. |
| Tax laws distort yields | State taxes make Treasuries artificially attractive to some. | Adjust spreads for your specific tax bracket before comparing. |