Merton Model: Your Company's Equity Is Just an Option
Welcome to Part Four of Wilmott’s book: Credit Risk. Up until now, every product we priced assumed that all cashflows are guaranteed. Coupons get paid. Bonds get redeemed. Nobody goes bankrupt. That was a comfortable world to live in, but it is not reality.
Companies go bust. Countries fail to pay their debts. Over-the-counter derivatives carry counterparty risk. Chapter 39 tackles this head on by asking: how do you value debt when the borrower might not pay you back?
And the answer is surprisingly elegant. Your company’s equity is just a call option on its assets.
The Setup
There are two main approaches to modeling default risk. One models default as something that happens inside the firm, based on the firm’s actual financial health. The other treats default as a random external event, like lightning striking. Chapter 39 focuses on the first approach.
The star of this chapter is the Merton model, published in 1974 by Robert Merton (yes, the same Merton from Black-Scholes-Merton). It is beautiful in its simplicity.
The Merton Model
Imagine a company with assets worth $A$. The company has issued debt of amount $D$ that must be repaid at time $T$. The assets follow a random walk:
$$dA = \mu A , dt + \sigma A , dX$$
Just like a stock price. Random, with drift $\mu$ and volatility $\sigma$.
Now, what happens at time $T$?
If the assets $A$ are worth more than the debt $D$, the company pays off the debt and the shareholders keep the rest. The equity value is $A - D$.
If the assets are worth less than $D$, the company defaults. It hands over whatever assets remain to the creditors. Shareholders get nothing. The equity value is zero.
So the equity value at maturity is:
$$S = \max(A - D, 0)$$
Wait. That is the payoff of a call option. The underlying is the asset value $A$. The strike price is the debt $D$. The equity of a company IS a call option on the firm’s assets.
This is not an analogy. It is exact. The math is identical. You can apply Black-Scholes directly.
What About the Debt?
Since assets equal equity plus debt ($A = S + V$), and we know how to value the equity (it is a call), the debt value is simply:
$$V = A - S$$
This means the debt is equivalent to owning the assets minus a call option. Or equivalently, it is like a risk-free bond minus a put option on the assets. The put option represents the possibility of default, the chance that the assets will not be enough to cover the debt.
The riskier the company (higher $\sigma$), the more valuable the “equity option” and the less valuable the debt. Makes intuitive sense. High volatility is great if you are a shareholder (unlimited upside, limited downside at zero). High volatility is terrible if you are a bondholder (you never get more than $D$, but you might get less).
Default Before Maturity
The basic Merton model only allows default at the maturity date $T$. But in reality, creditors can force liquidation before maturity if the asset value falls below some critical level $K(t)$.
This transforms the problem into a barrier option. The equity becomes a down-and-out call option. If the assets hit the barrier $K(t)$, the firm is liquidated and equity holders lose everything.
On the barrier, the debt value equals whatever assets remain:
$$V(K(t), t) = K(t)$$
The probability of default before maturity $P(A, t)$ becomes equivalent to the probability of hitting the barrier, a well-studied problem in the barrier options world.
Adding Stochastic Interest Rates
Making the model more realistic means adding a random interest rate. Now the debt value depends on three things: the asset level, the interest rate, and time. $V(A, r, t)$.
To hedge, you need both the company’s stock (to hedge asset risk) and a riskless zero-coupon bond (to hedge interest rate risk). The standard two-factor pricing equation follows.
The correlation between asset values and interest rates adds another dimension to the problem but does not change the fundamental logic.
The Big Criticism
The Merton model is intellectually elegant but practically frustrating. The main problem? You cannot easily observe or measure the inputs.
What exactly is the “asset value” of a company? It is not just the stock price. It includes all the physical assets, intellectual property, brand value, future earnings potential. And the volatility of that unobservable quantity? Even harder to pin down.
Wilmott acknowledges this directly. The model works well for understanding the relative value of different types of debt from the same issuer. But as an absolute pricing tool, the parameter estimation problem is a serious obstacle.
A Better Approach: Measurable Parameters
Wilmott then describes an alternative model that uses quantities you can actually observe: earnings and cash in the bank.
Instead of modeling the mysterious “asset value,” model the gross earnings $E$ of the company:
$$dE = \mu_E E , dt + \sigma_E E , dX$$
The company has fixed costs of $E^$ per year and floating costs of $kE$. The profit $(1-k)E - E^$ goes into a bank account earning interest rate $r$. The cash balance $C$ satisfies:
$$dC = rC , dt + (1-k)E , dt - E^* , dt$$
The randomness comes from earnings. Cash is determined by the accumulated earnings, costs, and interest. Both $E$ and $C$ are observable from the company’s financial statements.
Valuing the Debt
The company owes $D$ at time $T$. The repayment is:
$$\text{Repayment} = \min(\max(C, 0), D)$$
If cash $C > D$: full repayment. If $0 < C < D$: partial repayment. If $C < 0$: nothing.
The present value of this expected repayment is the debt value $V(E, C, t)$. It satisfies a two-dimensional PDE that you can solve numerically.
Wilmott shows results for a specific example: $100,000 debt due in two years, 5% risk-free rate, $30,000 fixed costs, 7% variable costs, 10% earnings drift, 25% earnings volatility. When both earnings and cash are high, the debt value approaches a risk-free zero-coupon bond. When either is low, the value drops, reflecting the higher chance of default.
The credit spread, how much extra yield the debt carries over the risk-free rate, falls out naturally from the model.
Valuing the Firm Itself
A small tweak to the boundary conditions turns this into a firm valuation model. Instead of asking “what is the debt worth?”, ask “what is the whole business worth?”
Take the value of the business as the present value of expected cash at some future horizon $T_0$:
$$V(E, C, T_0) = \max(C, 0)$$
for a limited liability company (shareholders walk away from negative cash), or:
$$V(E, C, T_0) = C$$
for a partnership (owners are personally liable for debts).
Optimal Close-Down
Here is a genuinely useful application. If the model says your company is worth $3,000,000 but you have $5,000,000 in the bank, the model is telling you that bad times are coming. The future expected earnings do not justify continuing operations. You would be better off closing down and keeping the cash.
This optimal close-down decision looks exactly like an American option problem. The constraint is:
$$V(E, C, t) \geq C$$
At any point, the value of the firm must be at least as much as the cash in the bank. If it is not, you close down. The free boundary between “keep operating” and “close down” can be found with the usual American option techniques.
Why This Matters
The Merton model may seem abstract, but the core insight has massive practical implications:
Shareholders love risk. Their position is a call option. More volatility means more upside potential, and the downside is capped at zero. This explains why shareholders sometimes push for risky strategies that terrify bondholders.
Bondholders hate risk. Their position is like selling a put. More volatility means more chance of getting less than promised.
Capital structure matters. The split between debt and equity is not just an accounting detail. It fundamentally changes the risk profile and incentives of different stakeholders.
The measurable-parameter model shows that you do not need mythical “asset values” to think about firm valuation and credit risk. Earnings and cash are enough. And the math is the same PDE framework we have been using throughout the book.
Next chapter, we leave the world of modeling the firm’s internals and look at credit risk from the outside, treating default as something random and exogenous.
This is part of a series covering “Paul Wilmott on Quantitative Finance”. Next up: Credit Risk: Modeling the Chance of Default.