Corporate Credit Risk Modeling

(1. Structural Models)

www.peaks2tails.com
www.peaks2tails.com Satyapriya Ojha
Corporate Credit Risk Modelling Structural Models

Structural Models - Introduction

Structural Models derive Credit Risk of a firm based on its Capital Structure.
The key idea is if the firm value falls below the safety level, that leads to default.
Key Modelling components include
• Stochastic process for the Asset
• Default Barrier Model
• Time to default

www.peaks2tails.com Satyapriya Ojha 2

Corporate Credit Risk Modelling Structural Models

Merton Model - Assumptions

The Assumptions under Merton Model are
• Firm has two classes of security ; Equity 𝐸𝑡 and Debt 𝐷𝑡
• No taxes, transaction costs, bankruptcy costs
• Debt is a zero-coupon bond with a face value of 𝐵 payable at maturity 𝑇
• Default can occur only at 𝑇 , if the firm’s Asset falls below B , i.e., 𝑉𝑇 ≤ 𝐵

𝑉𝑇 (no default)

𝐵
𝑉𝑇 (default)

𝑇
www.peaks2tails.com Satyapriya Ojha 3
Corporate Credit Risk Modelling Structural Models

Merton Model – Asset Dynamics

Asset follows lognormal SDE

𝑑𝑉𝑡
= 𝑟 − 𝛿 𝑑𝑡 + 𝜎𝑑𝑊𝑡
𝑉𝑡

𝑟 := risk free rate, 𝛿:= proportional dividend payout

The solution is given as

1
𝑉𝑡 = 𝑉0 𝑒 𝑋𝑡 𝑋𝑡 = 𝜃𝑡 + 𝜎𝑊𝑡 𝜃 = 𝑟 − 𝛿 + 𝜎2
2

www.peaks2tails.com Satyapriya Ojha 4

Corporate Credit Risk Modelling Structural Models

Merton Model – Debt & Equity

At T, two things can happen
• Firm defaults 𝑉𝑇 ≤ 𝐵 : Debt holders get 𝑉𝑇 and Equity holders get 0
• Firm survives 𝑉𝑇 > 𝐵 : Debt holders get 𝐵 and Equity holders get 𝑉𝑇 − 𝐵

𝐷𝑇 = 𝐵𝟏𝑉𝑇 >𝐵 + 𝑉𝑇 𝟏𝑉𝑇 ≤𝐵 = min 𝐵, 𝑉𝑇 = 𝐵 − max 𝐵 − 𝑉𝑇 , 0

𝐸𝑇 = 𝑉𝑇 − 𝐵 𝟏𝑉𝑇 >𝐵 = max 𝑉𝑇 −𝐵, 0

𝑉
ln 𝐵0 + 𝜃 + 𝜎 2 𝑇
𝐸0 = 𝔼 𝑒 −𝑟𝑇 𝐸𝑇 = 𝑉0 𝑒 −𝛿𝑇 Φ 𝑑1 − 𝐵𝑒 −𝑟𝑇 Φ 𝑑2 𝑑1 = 𝑑2 = 𝑑1 − 𝜎 𝑇
𝜎 𝑇

𝐷0 = 𝔼 𝑒 −𝑟𝑇 𝐷𝑇 = 𝐵𝑒 −𝑟𝑇 − Put 0 = 𝐵𝑒 −𝑟𝑇 Φ 𝑑2 + 𝑉0 𝑒 −𝛿𝑇 Φ −𝑑1

www.peaks2tails.com Satyapriya Ojha 5

Corporate Credit Risk Modelling Structural Models

Merton Model – Default Probability

The risk-neutral probability of default is given by

ℙ 𝜏 = 𝑇 = ℙ 𝑉𝑇 ≤ 𝐵 = Φ −𝑑2

Note that in the Merton’s model, the firm can only default at T, so ℙ 𝜏 ≤ 𝑇 is not defined.

Limitations :
• Default is only considered at maturity
• Capital structure assumed in them model is overly simplistic
• The perfect market assumptions are unrealistic
• Short term credit spreads and default probabilities are significantly underestimated

www.peaks2tails.com Satyapriya Ojha 6

Corporate Credit Risk Modelling Structural Models

Merton Model – Parameter Estimation

Since Equity is a call option on value of the firm

𝐸𝑡 = Call 𝑉𝑡 , 𝑡

𝜕Call
𝑑𝐸𝑡 = … 𝑑𝑡 + 𝜎 𝑉 𝑑𝑊
𝜕𝑉 𝑉 𝑡 𝑡

Comparing with the Equity lognormal dynamics 𝑑𝐸𝑡 = … 𝐸𝑡 𝑑𝑡 + 𝜎𝐸 𝐸𝑡 𝑑𝑊𝑡

𝑉
𝜎𝐸 = 𝜎𝑉 Δ𝐶𝑎𝑙𝑙 (Jones Formula)
𝐸

The Equity volatility can be estimated from observable stock prices

www.peaks2tails.com Satyapriya Ojha 7

Corporate Credit Risk Modelling Structural Models

Vasicek-Merton extension
Vasicek Model is the cornerstone of IRB where we model conditional PD based on systematic factor
return 𝑧 . It’s possible to extend the Merton Model PD to be conditioned on 𝑧 . In Merton Model,
the unconditional PD is given as

𝑉 𝑉
ln 𝐵0 + 𝜃𝑇 ln 𝐵0 + 𝜃𝑇
𝑃𝐷 = ℙ 𝑉𝑇 < 𝐵 = ℙ 𝑊𝑇 < − = Φ −𝑑2 𝑑2 =
𝜎 𝜎 𝑇

Under Vasicek, 𝑊𝑇 = 𝜌𝑊𝑇 (1) + 1 − 𝜌𝑊𝑇 (2) . To obtain conditional PD(z), we need to replace the
systematic return 𝑊𝑇 (1) = 𝑧 𝑇, Putting it in the above equation yields

𝑉
ln 𝐵0 + 𝜃𝑇
+ 𝜌𝑧
𝜎 𝑇 𝑑2 + 𝜌𝑧
𝑃𝐷(𝑧) = ℙ 𝑍𝑇 (2) <− = Φ −𝑑2 𝑧 , 𝑑2 (𝑧) =
1−𝜌 1−𝜌

www.peaks2tails.com Satyapriya Ojha 8

Corporate Credit Risk Modelling Structural Models

Black-Cox Model - Introduction

The main idea of Black-Cox model revolves around premature default i.e., the firm can default any
time 𝑡 ≤ 𝑇, where the firm value drops below a safety level ( exogenous default barrier 𝑉𝐵 (𝑡) ). The
default happens at the first time the firm value crosses the barrier, therefore Black-Cox model is a
first-passage time model.

(no default)
as per Merton

𝑉𝑇

𝑉𝐵

(default) as 𝑇
per Black-Cox

www.peaks2tails.com Satyapriya Ojha 9

Corporate Credit Risk Modelling Structural Models

Black-Cox Model – Default time & Survival Probability

The Default time 𝜏 is given as

𝜏 = inf 𝑡 ≥ 0: 𝑉𝑡 ≤ 𝑉𝐵 (𝑡)
𝐴
ln + 𝐶𝑇
𝐵
Assume a constant Barrier, 𝑉𝐵 𝑡 = 𝑉𝐵 . Define 𝑧 𝐴, 𝐵, 𝐶 = ,𝜂 = 𝜃 2 + 2𝑟𝜎 2
𝜎 𝑇

Survival Probability (running minimum is always above the barrier and the terminal value is above the
face value of the debt) is given as
𝑉𝐵 𝐵
ℙ 𝜏 > 𝑇, 𝑉𝑇 > 𝐵 = ℙ min 𝑉𝑡 > 𝑉𝐵 , 𝑉𝑇 > 𝐵 = ℙ min 𝑋𝑡 > ln , 𝑋𝑇 > ln
0≤𝑡≤𝑇 0≤𝑡≤𝑇 𝑉0 𝑉0
2𝜃
𝑉𝐵 𝜎2
= Φ 𝑧 𝑉0 , 𝐵, 𝜃 − Φ 𝑧 𝑉𝐵 2 , 𝐵𝑉0 , 𝜃
𝑉0
(Merton’s
Survival Prob)

www.peaks2tails.com Satyapriya Ojha 10

Corporate Credit Risk Modelling Structural Models

Black-Cox Model – CDS Curve

The CDS spread in Black-Cox is given as

𝑟 1 − 𝑅𝑅 𝔼 𝑒 −𝑟𝜏 1𝜏≤𝑇
𝑠=
1 − 𝑒 −𝑟𝑇 ℙ 𝜏 > 𝑇 − 𝔼 𝑒 −𝑟𝜏 1𝜏≤𝑇

From the theory of first-passage times of Brownian Motion

𝑉𝐵 𝑉𝐵 − 𝜂𝑇 𝑉𝐵 𝑉𝐵 + 𝜂𝑇
𝜃−𝜂 𝜃+𝜂
𝔼 𝑒 −𝑟𝜏 1𝜏≤𝑇 = 𝑒 𝜎2 Φ + 𝑒 𝜎2 Φ
𝜎 𝑇 𝜎 𝑇

2𝜃
𝑉𝐵 𝜎2
ℙ 𝜏 > 𝑇 = Φ 𝑧 𝑉0 , 𝑉𝐵 , 𝜃 − Φ 𝑧 𝑉𝐵 , 𝑉0 , 𝜃
𝑉0

We can plugin the above in the equation and solve for 𝜎 from observed CDS spreads using the
condition that at default 𝑉𝐵 = 𝑅𝑅. 𝐵

www.peaks2tails.com Satyapriya Ojha 11

Corporate Credit Risk Modelling Structural Models

Vasicek-Black-Cox extension
The Black-Cox model can be extended to Vasicek distribution to obtain a conditional PD. The Black-
Cox unconditional PD can be written as
2𝜃 𝑉 𝑉𝐵 2
ln 𝐵0 + 𝜃𝑇 ln 𝐵𝑉 + 𝜃𝑇
𝑉𝐵 𝜎2 0
𝑃𝐷 = Φ −𝑑2 + Φ 𝑑ҧ2 , 𝑑2 = , 𝑑ҧ2 =
𝑉0 𝜎 𝑇 𝜎 𝑇

The conditional form is simply obtained by making the following transformations

2𝜃
𝑉𝐵 𝜎2 𝑑2 + 𝜌𝑧 𝑑ҧ2 + 𝜌𝑧
𝑃𝐷(𝑧) = Φ −𝑑2 (𝑧) + Φ 𝑑ҧ2 (𝑧) , 𝑑2 (𝑧) = , 𝑑ҧ2 (𝑧) =
𝑉0 1−𝜌 1−𝜌

www.peaks2tails.com Satyapriya Ojha 12

Corporate Credit Risk Modelling Structural Models

Excel Implementation

www.peaks2tails.com Satyapriya Ojha 13

Corporate Credit Risk Modelling Structural Models

Merton Model Basics

www.peaks2tails.com Satyapriya Ojha 14

Corporate Credit Risk Modelling Structural Models

Merton Model – Subordinated Debt valuation

www.peaks2tails.com Satyapriya Ojha 15

Corporate Credit Risk Modelling Structural Models

Merton Model – (with stochastic rates)

www.peaks2tails.com Satyapriya Ojha 16

Corporate Credit Risk Modelling Structural Models

Merton Model – (with jumps)

www.peaks2tails.com Satyapriya Ojha 17

Corporate Credit Risk Modelling Structural Models

Merton Model – Calibration

www.peaks2tails.com Satyapriya Ojha 18

Corporate Credit Risk Modelling Structural Models

Merton Model – Vasicek Extension

www.peaks2tails.com Satyapriya Ojha 19

Corporate Credit Risk Modelling Structural Models

Black-Cox Model – First Passage time basics

www.peaks2tails.com Satyapriya Ojha 20

Corporate Credit Risk Modelling Structural Models

Black-Cox Model – Constant Barrier

www.peaks2tails.com Satyapriya Ojha 21

Corporate Credit Risk Modelling Structural Models

Black-Cox Model – Deterministic Barrier

www.peaks2tails.com Satyapriya Ojha 22