Economic Modelling · Simple models for credit risk
Reduced Form Models and Hazard Rates in Credit Risk
Updated 11 October 2026 · Fact-checked
A reduced form model treats default as a random event that occurs at a rate called the hazard rate (intensity) λ. With constant λ, survival to time t has probability e^(−λt). A defaultable zero-coupon bond is priced by discounting its payoff, including recovery, using risk-neutral default probabilities.
Understand Reduced Form Models and Hazard Rates
A reduced form model does not explain why a firm defaults. It does not model the firm's assets or debt. It simply says default arrives at a random time, and it models how likely that arrival is.
The key quantity is the hazard rate (also called default intensity) λ(t). Over a short interval dt, the probability of default, given no default so far, is approximately λ(t) dt. If λ is constant, the time to default τ is exponential. Survival probability is P(τ > t) = e^(−λt). If λ varies with time, survival is e^(−∫λ(s)ds) from 0 to t.
There are two sets of default probabilities. Real-world (physical) probabilities come from historical default data and describe what actually happens. Risk-neutral probabilities are implied by market prices of bonds. They are used to price. Because investors demand compensation for default risk, the risk-neutral hazard rate is usually higher than the real-world one. Using the real-world rate to price a bond is wrong.
Pricing is then simple. A defaultable zero-coupon bond pays 1 at time T if there is no default. If default occurs, it pays a recovery amount. With risk-free force of interest r and a constant risk-neutral hazard rate λ, the credit spread is roughly (1 − R)λ, where R is the recovery rate. Under the simplest assumption that recovery R is paid at maturity, the price is e^(−rT) × [R + (1 − R) e^(−λT)].
Key rules to remember
- Survival probability (constant hazard)
- P(τ > t) = e^(−λt)
- τ is the default time. Time to default is exponential with mean 1/λ.
- Survival probability (time-varying hazard)
- P(τ > t) = exp(−∫₀ᵗ λ(s) ds)
- For piecewise constant λ, add λ × length over each period inside the exponent.
- Default probability by time t
- P(τ ≤ t) = 1 − e^(−λt)
- Use the risk-neutral λ for pricing and the real-world λ for real-world risk measurement.
- Hazard rate definition
- λ(t) = f(t) ÷ S(t)
- f is the density of τ and S is the survival function. This is the same idea as force of mortality.
- Defaultable zero-coupon bond, recovery of face value at maturity
- Price = e^(−rT) × [R + (1 − R) e^(−λT)]
- Assumes constant r and λ, independent, with recovery R of face value paid at T if default occurs before T.
- Defaultable zero-coupon bond, zero recovery
- Price = e^(−(r + λ)T)
- The spread equals λ when R = 0.
- Approximate credit spread
- s ≈ (1 − R) × λ
- Approximation for small λ. Rearrange to get λ ≈ s ÷ (1 − R).
How to solve Reduced Form Models and Hazard Rates questions
Use this method for most questions on hazard rates, risk-neutral default probabilities and defaultable bonds.
- 1Identify what is given: hazard rate λ (real-world or risk-neutral), risk-free rate r, recovery rate R, maturity T, and whether recovery is paid at default or at maturity.
- 2State your assumptions: constant or piecewise hazard, constant interest, independence of default and interest rates, and the recovery convention.
- 3Find the survival probability using e^(−∫λ) and the default probability as one minus it.
- 4Check which probability measure the question needs. Use risk-neutral λ for pricing. Use real-world λ for expected loss or real-world probabilities.
- 5Write the payoff in each state: no default pays face value, default pays the recovery amount.
- 6Discount the expected payoff at the risk-free rate under the risk-neutral measure, to get the price.
- 7If asked for the spread or the implied hazard rate, compare with the risk-free bond or use s ≈ (1 − R)λ, and show it is an approximation.
- 8Check reasonableness: the price should be below the risk-free price, and the spread should be positive.
Quickest way: Price by splitting into survival and default parts
When to use it: Use for multiple-choice questions or short written parts with constant λ, constant r and recovery of face value at maturity.
- Compute the risk-free discount factor e^(−rT).
- Compute the survival probability q = e^(−λT).
- Compute expected payoff = q + (1 − q) × R.
- Multiply expected payoff by the discount factor.
- For an implied λ from a market price, divide the price by e^(−rT), then solve q = (payoff − R) ÷ (1 − R) and λ = −ln(q) ÷ T.
Common mistakes in Reduced Form Models and Hazard Rates
Using the real-world default probability to price a bond.
Students see a historical default rate given in the question and assume it is the pricing input.
Fix: Price only with risk-neutral probabilities. Use real-world ones only when the question asks about actual expected losses.
Writing default probability as λT for any T.
The approximation λ dt works for very small intervals and gets over-applied.
Fix: Use 1 − e^(−λT). Use λT only if the question says the period is small and approximation is acceptable.
Treating the hazard rate as the unconditional probability of default in a year.
The word rate hides that it is conditional on survival so far.
Fix: Remember λ(t) is the instantaneous rate given no default before t. The one-year default probability is 1 − e^(−λ) for constant λ.
Forgetting the recovery term, or applying recovery to the wrong amount.
Students rush and use e^(−(r+λ)T), which is correct only for zero recovery.
Fix: Write the payoff in both states first. Check whether recovery is a fraction of face value and when it is paid.
Using s = λ when recovery is positive.
The zero-recovery result is remembered and applied everywhere.
Fix: Use s ≈ (1 − R)λ. Only with R = 0 does the spread equal λ.
Adding hazard rates of different periods without multiplying by period length.
Piecewise hazard questions look like a list of rates.
Fix: Integrate: add λ × years for each period inside the exponent, then exponentiate once.
Worked examples
Example 1
A firm has a constant risk-neutral hazard rate of 3% per year. The risk-free force of interest is 4% per year. A 2-year zero-coupon bond has face value ₹1,00,000. If default occurs before maturity, 40% of face value is recovered at maturity. Find the price of the bond.
Show the solution
- Survival probability: q = e^(−0.03 × 2) = e^(−0.06) = 0.94176.
- Default probability = 1 − 0.94176 = 0.05824.
- Expected payoff at maturity = 0.94176 × 1,00,000 + 0.05824 × 40,000 = 94,176 + 2,330 = ₹96,506.
- Discount factor: e^(−0.04 × 2) = e^(−0.08) = 0.92312.
- Price = 96,506 × 0.92312 = ₹89,086 approximately.
Answer: The price is about ₹89,086, compared with a risk-free price of about ₹92,312.
Example 2
A 1-year zero-coupon bond with face value 100 trades at 93.00. The risk-free force of interest is 3%. Recovery is 30% of face value paid at maturity on default. Find the implied risk-neutral hazard rate λ, assuming it is constant.
Show the solution
- Risk-free discount factor: e^(−0.03) = 0.970446.
- Price ÷ discount factor = 93.00 ÷ 0.970446 = 95.8325, the expected payoff.
- Expected payoff = 100 × [R + (1 − R) q] = 100 × [0.3 + 0.7q].
- So 0.3 + 0.7q = 0.958325, giving q = 0.658325 ÷ 0.7 = 0.940464.
- λ = −ln(0.940464) ÷ 1 = 0.06139.
Answer: The implied risk-neutral hazard rate is about 6.14% per year.
Exam tips
- Always state the recovery convention (fraction of face value, paid at maturity or at default). Marks are lost when this is left implicit.
- Say clearly which measure you use. Write 'risk-neutral' next to any probability used for pricing.
- Show the form e^(−λt) before substituting numbers. Written papers award method marks for the formula.
- Use the approximation s ≈ (1 − R)λ for quick checks, but do the exact calculation when the question asks for a price.
- Be ready to explain in words why risk-neutral default probabilities usually exceed real-world ones.
Practice questions from Simple models for credit risk
- A one-year zero-coupon corporate bond with face value Rs 1,000 trades at Rs 920. The one-year risk-free zero-coupon bond trades at Rs 950 pe…
- A continuous-time Markov model for credit ratings uses a generator matrix with off-diagonal entries equal to transition intensities. A bond …
- A one-year zero-coupon bond from an Indian corporate has face value Rs 100. The risk-free one-year rate is 6% p.a. effective. The probabilit…
- Under a reduced form model with zero recovery, a risky zero-coupon bond has continuously compounded yield 7.5% and the equivalent risk-free …
- In a rating-based credit risk model using a transition matrix, which statement is correct?
Reduced Form Models and Hazard Rates in other exams
The same ground in other exams, if you are preparing for more than one or want another angle on it.
Reduced Form Models and Hazard Rates: frequently asked questions
What is the difference between real-world and risk-neutral default probabilities?
Real-world probabilities estimate how often default actually happens, using historical data. Risk-neutral probabilities are backed out of market prices and include compensation for bearing default risk. They are used for pricing and are usually higher than real-world ones.
How is a hazard rate related to force of mortality?
They are the same idea. Both are the instantaneous rate of the event, given survival so far. Survival is exp(−∫λ) in both cases, so your CS2 survival models knowledge transfers directly.
How do I price a defaultable zero-coupon bond with recovery?
Find the survival probability from the hazard rate. Write the payoff as face value if no default and the recovery amount if default. Take the expected payoff under risk-neutral probabilities and discount at the risk-free rate.
How is a reduced form model different from the Merton model?
The Merton model is structural. It links default to the firm's asset value falling below its debt. A reduced form model treats default as an unexplained random event with an intensity, which is easier to calibrate to market prices.