Explore how credit default swaps (CDS) are priced and the key factors—such as default probability, market conditions, and recovery rates—that drive CDS spreads.
Enhance Your Learning:
Anyone who has peeked into the CDS world tends to focus first on this fundamental question: “How on earth do we actually price one of these contracts?” I still remember my first real CDS trade—my heart was pounding, partly because I was worried about messing up the math, but also because I knew how crucial these instruments can be for hedging credit risk. So, let’s break it down.
At its core, a credit default swap (CDS) can be thought of as an insurance policy on an underlying reference entity’s debt. The buyer pays a periodic fee—also called the premium or spread—and in return, the seller agrees to compensate the buyer if that reference entity defaults. The fair spread on a CDS is found by equating:
• The present value of the premium leg (the buyer’s expected payments).
• The present value of the protection leg (the buyer’s expected payout in the event of default).
You can think of the premium leg as a series of payments made over the life of the CDS, typically quoted in basis points (bps) of the notional amount. If no default occurs, the buyer continues to pay this annualized spread until maturity (or until the contract is terminated).
The protection leg, on the other hand, represents the seller’s obligation if the reference entity defaults. If default happens, the buyer generally receives (1 – recovery rate) × notional, either via physical settlement (delivering the defaulted bond in exchange for par) or cash settlement. The exact formula for the “fair” CDS spread ensures that, at inception, the expected present value of these two legs is equal.
Here’s a more “mathy” representation that might look intimidating, but don’t worry—we’ll keep it conceptual:
Where:
• \(R\) is the recovery rate.
• \(\lambda(t)\) is the hazard rate (a.k.a. default intensity).
• \(r(u)\) is the risk-free or benchmark discounting rate.
• \(T\) is the CDS maturity.
Intuitively, the numerator captures the expected loss portion (that is, how likely default is, multiplied by how much you lose), and the denominator captures a measure of how long you expect to pay spreads (discounted to present value).
Below is a quick mermaid diagram to visualize the flow between the CDS buyer and seller:
flowchart LR
A["CDS Buyer <br/> (Pays CDS Spread)"] --> B["CDS Seller <br/> (Provides Default Protection)"]
B-- "Compensation if <br/> Reference Entity Defaults" -->A
In practice, we often discount future payments at LIBOR/SOFR-based rates. The discount curve should reflect the risk-free or near-risk-free environment, or sometimes the swap curve for the corresponding maturities. This discount rate assumption is key: an underestimation can inflate the present value of the premium leg, while an overestimation might distort the cost of protection.
Ever notice how, during a market scare, CDS spreads across a whole industry suddenly widen (and sometimes it feels like everything in your portfolio goes red)? That’s because CDS spreads respond massively to systemic or idiosyncratic shocks:
• Default Risk Perception: If investors believe that the chance of default is rising, they’re willing to pay more for protection, pushing up the CDS spread.
• Credit Quality Deterioration: A downgrade or negative news—like a looming debt maturity or questionable earnings—often triggers an almost immediate reaction in spreads.
• Systemic Factors: During economic downturns, or periods of high volatility (think major political shocks), entire sectors can see their average spread spike. Investors just get more defensive.
One personal anecdote: back in 2008, I recall seeing normally stable industrial names with CDS spreads that soared in tandem with finance companies. It demonstrated how quickly contagion and panic can ripple across the market, even when fundamentals might not have changed all that much.
Though I joked earlier about the math, advanced pricing models use a hazard rate (or default intensity) to describe how likely an entity is to default over a small slice of time. This approach is typical in reduced-form credit models.
Under the hazard rate method, we assume that the probability of surviving (i.e., not defaulting) up to time \(t\) is:
where \(\lambda(\tau)\) is the default intensity at time \(\tau\). Intuitively, we integrate the hazard rate to find the cumulative chance of making it each step of the way.
We usually calibrate \(\lambda(t)\) by matching observed CDS spreads at various maturities. If we see that a 5-year CDS is trading at 300 bps, we back into the hazard rate that explains this spread. Of course, it’s never quite that straightforward in real life—liquidity premiums, small sample sizes, or unusual capital structures can cause calibration headaches.
The portion of par the lender (or protection buyer) expects to recover upon default is crucial to the CDS pricing process. A lower recovery rate means a larger expected loss on the underlying bond, and therefore a higher CDS spread.
Sometimes the market sets a “standard” recovery rate for an industry (e.g., 40% for a certain senior unsecured bond). But as soon as there’s any whiff of a sector meltdown—maybe shipping or energy—market participants quickly revise these assumptions, which can push spreads higher. That shift can ripple through hazard rate calibrations.
If you want a quick rule of thumb:
• Higher assumed recovery → Lower spread.
• Lower assumed recovery → Higher spread.
Even a small change in your recovery assumption can dramatically alter the final CDS price.
Though the reduced-form approach is more common in day-to-day trading, structural models (like the Merton model) also inform participants about default risk. The Merton model sees a firm’s equity as a call option on its underlying assets—if assets dip below the firm’s liabilities, that’s effectively a default scenario.
In practice, many credit desks blend both approaches: they look at the firm’s option-implied volatility, capital structure details, and then cross-check with simpler hazard rate calibrations. They also incorporate swap curve adjustments (especially after LIBOR’s transition to SOFR), and factor in any liquidity premiums. Let’s face it, a meltdown scenario with no liquidity is going to widen CDS spreads even if the fundamental default risk hasn’t changed.
It’s not enough to just look at a single issuer in a vacuum. When multiple firms are operating in the same sector, or when the entire economy is in freefall, correlation can spike. Market participants might buy protection on a basket of names or an index (like the CDX or iTraxx series), which can push spreads up wholesale, even for firms that remain relatively solid. We saw this dynamic vividly in past crises: the tide goes out, and correlation leaps.
In an exam setting, you might get a meaty scenario: imagine a hypothetical chemical manufacturer rumored to be facing litigation that could severely hamper its financials. The question might ask you to calculate the implied default probability given a certain bid/ask on the 5-year CDS, or how the spread should shift if the recovery rate changes from 40% to 30%. They might also throw in a scenario where the yield curve shifts, and you’ve got to recast the discount factors. Summaries for the exam:
• Carefully identify the contract terms (notional, maturity, recovery assumption).
• Convert annual spreads to the correct payment frequency.
• Adjust for discount factors.
• Solve for or interpret the hazard rate.
• Weigh changes in market condition and how they shift the credit spread.
Also, watch out for trick questions comparing the bond spread vs. the CDS spread. A bond’s yield spread may differ from the CDS spread due to contract differences, deliverable instruments, and the so-called “cheapest-to-deliver” option embedded in physical settlement.
Banks, asset managers, and hedge funds use CDS for both hedging and speculative bets. If a bond manager wants to reduce exposure to a particular credit, they might buy CDS protection. Conversely, a manager might sell CDS protection if they believe the market’s default estimates are overly pessimistic.
Some institutions use the implied default rates from CDS markets to guide capital reserves or to set risk limits. A sharp jump in CDS spreads can signal a forward-looking problem that might not yet show up in fundamental credit analysis (like a rating downgrade).
A trader could buy short-dated protection and sell long-dated protection on the same name, hoping the short end of the curve will widen relative to the long end. These so-called curve trades rely on shapes and shifts in the term structure of credit spreads.
• CDS Spread (Premium): The annualized fee, in basis points of notional, paid by the protection buyer.
• Default Intensity (Hazard Rate): The instantaneous probability of default at any point in time.
• Recovery Rate: The assumed fraction of par value recovered if default happens.
• Loss Given Default (LGD): (1 – Recovery Rate). Represents the portion of par value lost in the event of default.
• Term Structure of Credit Spreads: How CDS spreads vary with different maturities.
• Upfront Payment: Sometimes an up-front adjustment is needed if the on-market spread differs from a standardized “running” spread (e.g., 100 bps).
• Basis Points (bps): One bps = 0.01%.
• Jarrow, R. & Turnbull, S. (1995). “Pricing Derivatives on Financial Securities Subject to Credit Risk.” Journal of Finance.
• Hull, J. (2012). Options, Futures, and Other Derivatives. Pearson. (Credit Derivatives Chapters)
• CFA Institute Level II Curriculum – Fixed Income and Derivatives (Readings on pricing credit risk and CDS).
• Moody’s Analytics. “The Merton Model and Credit Spreads.”
Important Notice: FinancialAnalystGuide.com provides supplemental CFA study materials, including mock exams, sample exam questions, and other practice resources to aid your exam preparation. These resources are not affiliated with or endorsed by the CFA Institute. CFA® and Chartered Financial Analyst® are registered trademarks owned exclusively by CFA Institute. Our content is independent, and we do not guarantee exam success. CFA Institute does not endorse, promote, or warrant the accuracy or quality of our products.