Credit Trading: Single Name and Portfolio Strategies
This is a retelling of Chapter 7 (Part 2) from “Trading the Fixed Income, Inflation and Credit Markets: A Relative Value Guide” by Neil C. Schofield and Troy Bowler (Wiley, 2011, ISBN: 978-0-470-74229-7).
In the first half of Chapter 7, we covered the CDS basis, credit indices, forward CDS spreads, and credit term structure trading. Now we get to the real question at the heart of relative value: given a view on a specific credit (or a basket of credits), what’s the best way to express it?
Credit-Linked Notes
A credit-linked note (CLN) is a synthetic bond that gives you credit exposure to a specific reference entity. The investor pays par and receives LIBOR plus an enhanced spread until either the note matures or the reference entity defaults.
Here’s how it’s built. The CLN issuer takes your money and parks it in a money-market deposit earning LIBOR. Simultaneously, the issuer sells CDS protection on the reference entity. The premium from the CDS gets passed through to you as the yield enhancement above LIBOR.
Why buy a CLN instead of the actual bond?
- You get a higher return than the bond’s asset swap spread.
- The actual bonds might not exist in the maturity or currency you want.
- The issuer can customize maturity, currency, and coupon.
- If the reference entity has a positive CDS basis (CDS spread higher than bond spread) and the note issuer has a high funding spread, the CLN return can beat both the bond and the CDS.
The risks are real, though. You’re exposed to two types of default. First, if the reference entity defaults, you get back your investment minus the CDS auction recovery rate. No more coupons. Maximum loss is 100% of what you put in. Second, if the CLN issuer defaults, the note terminates regardless of the reference entity’s health.
Key differences from owning the actual bond: restructuring triggers a CLN payout (bondholders just get restructured flows); a debt buyback doesn’t trigger the CDS, so the CLN keeps running; secondary market liquidity is basically zero.
There’s also a correlation wrinkle in the pricing. If the issuer’s funding costs and the reference entity’s default risk are positively correlated, both sides of the CLN could go wrong at once. The issuer wants to pay you less for this risk. You want more because it makes the note riskier. The book’s example shows a CLN paying 325bp vs 300bp on the asset swap and 250bp on the CDS. The extra yield compensates for these added risks.
Choosing the Right Instrument for a Single Name
The book walks through a real-world scenario. You’re managing a portfolio and want credit exposure to “Big Bank plc,” an A-rated UK financial services company. Available instruments include:
- 5-year secured term loan at LIBOR + 125bp
- 5-year asset swap at LIBOR + 155bp
- 5-year CDS at 155bp
- 6-year EUR floating-rate note at EURIBOR + 175bp
- 5-year GBP credit-linked note at LIBOR + 165bp
- 3-month ATM receiver option into 5-year CDS at 86bp
Which one? There’s no single correct answer. The term loan is secured but illiquid. The CDS is unfunded, so your own borrowing costs matter. The FRN needs FX hedging. The CLN gives extra yield but carries issuer default risk.
The point is that relative value means comparing the full picture: liquidity, funding costs, credit events, FX exposure, curve shape between tenors, and whether you’re isolating spread risk or default risk.
Credit Swaption Strategies
Beyond basic directional plays, credit swaptions have some specifically credit-flavored applications.
Expressing views on credit quality. Buy out-of-the-money receivers on a high-yield index and sell out-of-the-money payers on investment grade names, structured for zero premium. This is a “compression” trade. You’re betting that in a spread-widening environment, high-yield spreads won’t widen as much as investment grade spreads. All spreads move up but they bunch together.
Regional relative value. If you think CDX Investment Grade will outperform iTraxx Main, sell receiver swaptions on iTraxx and use the proceeds to buy receivers on CDX. If spreads fall, the iTraxx leg loses money but the CDX leg makes money.
Cross-asset class views. If you think equities will outperform credit in a rally, sell receivers on a credit index and buy calls on an equity index. Credit receivers lose money when spreads tighten (which happens in a rally), but the equity calls should more than compensate.
Buying a payer swaption is often expensive. The premium can exceed the cost of just buying CDS protection outright. So a payer makes more sense when you think the credit will probably improve (option expires worthless, you only lost the premium) but you need insurance in case you’re wrong.
Credit events and options. Single-name CDS options come with knock-in or knock-out features. A knock-out means the option terminates if a credit event happens before expiry. That’s a problem for long payer holders. Right when the trade is most valuable (the reference entity just defaulted), the option disappears. To protect against this, you’d separately buy CDS protection until the option expiry.
Index options don’t have knock-in/knock-out features. The holder decides at expiry whether to exercise based on the full basket economics, including any defaulted names.
Total Return Swaps
Total return swaps (TRS) let you replicate the return of a bond index without actually buying the bonds. The bank pays you the total return (price changes plus coupons) from the index. You pay the bank LIBOR plus or minus a spread.
Key features:
- You pick the index.
- No tracking error since the bank replicates the exact index.
- The tenor is negotiable.
- You’re exposed to counterparty default risk.
- It can be funded (you invest par, like a note) or unfunded (pure swap, no upfront cash).
The bank hedges by buying the physical bonds or entering an offsetting TRS. Their profit comes from the gap between your LIBOR spread and their funding cost.
Basket Default Swaps
These come in first-to-default and nth-to-default flavors. They’re not very liquid but they’re great for understanding default correlation.
A first-to-default basket is a CDS on a group of names (usually five) that pays out when the first name defaults. After that, the trade terminates. No more payments either way.
The seller earns a premium higher than any single name’s CDS spread. In the book’s example, individual spreads range from 25bp to 92bp with an average of 45bp, but the first-to-default basket trades at 188bp. That’s the compensation for taking “first loss” risk across all five names.
The premium has two boundaries:
- It must be higher than the weakest single-name spread. Otherwise, why take on additional names?
- It must be lower than the sum of all single-name spreads. Otherwise, the buyer might as well protect each name individually.
Default correlation is what makes baskets tick. It describes how likely one default is to trigger others.
- Low correlation: You’ll probably see a few losses but unlikely to see either zero or catastrophic losses.
- High correlation: The portfolio acts like a single credit. Either everything survives or everything blows up.
For first-to-default baskets, higher correlation means a lower premium. If all credits move together, the chance of any single default is lower (the basket either all defaults or all survives). For nth-to-default baskets, higher correlation means a higher premium because catastrophic multiple defaults become more likely.
Index Tranches
This is where correlation trading lives. Tranches slice a credit index into layers based on how many defaults you’re willing to absorb.
Picture the iTraxx Main with its 125 names worth 1 billion euros total. Instead of selling protection on the whole thing, you choose your layer:
- Equity tranche (0-3%): You start losing from the very first default. Highest spread (the book cites 2,820bp in one example), highest risk.
- Mezzanine tranche (3-6%): You only start losing after the first 3% of the portfolio has been wiped out. Moderate spread (200bp).
- Senior tranche (6-100%): Only loses money after significant carnage. Lowest spread (10bp).
The spreads across all tranches, when combined, can’t exceed the index spread. Tranches redistribute risk; they don’t create it.
Here’s how losses work in practice. Say a name defaults and the recovery rate is 40%. The loss is 60% of that name’s weight. In iTraxx Main, each name is 0.8% of the index, so the loss is 0.48% of the portfolio. The equity tranche absorbs this. After enough defaults eat through the equity tranche’s 3% cushion, the mezzanine investor starts bleeding.
Correlation drives everything. Low correlation means you’ll likely see a few defaults scattered around. That’s bad for equity tranche sellers (they absorb the first losses) but fine for senior tranche sellers (losses never pile up enough to reach them). High correlation means either nothing defaults or everything does. That’s better for equity sellers (lower chance of any default) but worse for senior sellers (if defaults happen, they’re catastrophic).
The math:
- Equity tranche premium is inversely related to correlation. Selling equity tranche protection = long correlation. You make money if correlation rises.
- Senior tranche premium is directly related to correlation. Selling senior tranche protection = short correlation. You lose money if correlation rises.
- Mezzanine sits somewhere in between and is harder to generalize.
The market convention for “long” and “short” in tranches refers to credit risk exposure, not the direction of the correlation bet. Long tranche = sold protection = long credit risk. Short tranche = bought protection = short credit risk.
That wraps up Chapter 7 and the credit section of the book. We’ve gone from bond-CDS basis trading through single-name instrument selection to portfolio-level strategies like tranche investing. The core message: there are many ways to express the same credit view, and the best choice depends on your specific constraints, funding costs, and the relative pricing across instruments.