Convertible bond (CB) arbitrage is a relative value strategy used by hedge funds to exploit mispricing between a company’s convertible bond and its underlying common stock. By constructing a delta-neutral portfolio, arbitrageurs aim to isolate the returns attributable to volatility and credit quality, while minimizing exposure to the underlying stock’s price movements.
Modeling this strategy requires the decomposition of the CB’s price into its fundamental risk components: equity sensitivity, credit risk, and volatility.
1. The Delta-Neutral Strategy (The Core Hedge)
The convertible bond is a hybrid instrument that can be mathematically decomposed into two primary parts: a straight corporate bond (debt component) and an American call option on the issuer’s stock (equity component).
A. The Arbitrage Setup
A typical convertible arbitrage trade involves a long-short position:
- Long the Convertible Bond: Captures the bond floor (downside protection) and the value of the embedded option.
- Short the Underlying Equity: Hedges the market risk (equity price movements).
B. Achieving Delta Neutrality
Delta ($\Delta$) measures the convertible bond’s price sensitivity to a $\$1$ change in the underlying stock price.
$$\Delta = \frac{\partial \text{CB Value}}{\partial \text{Stock Price}}$$
To create a delta-neutral portfolio, the arbitrageur matches the sensitivity of the long position (CB) with the sensitivity of the short position (stock).
$$\text{Short Shares} = \Delta \times \frac{\text{Face Value of CBs}}{\text{Stock Price}}$$
This neutralizes the portfolio’s directional risk. The position’s P&L is then primarily driven by changes in volatility ($\mathbf{Vega}$), the rate of change in delta ($\mathbf{Gamma}$), and credit quality ($\mathbf{Omicron}$ or credit spread risk).
C. Gamma Trading and Rebalancing
Crucially, delta is not constant; it changes as the stock price moves. Gamma ($\Gamma$) measures the rate of change of delta.
- Positive Gamma: Convertible bonds typically exhibit positive gamma, meaning the delta increases as the stock price rises and decreases as the stock price falls.
- Dynamic Hedge: Arbitrageurs must dynamically rebalance the short stock position (i.e., gamma trading) to maintain neutrality as the stock price fluctuates. This process of buying or selling the underlying stock creates an additional source of profit derived from volatility.
2. Modeling Credit Spread Decomposition
Accurate CB valuation must account for the issuer’s credit risk, as the bond floor value is discounted using a risky rate.
A. The Hybrid Valuation Challenge
The main challenge is that the value of the bond component is negatively impacted by credit risk, while the value of the option component is often positively correlated with credit spread widening (higher volatility often accompanies credit deterioration).
A fundamental approach for modeling credit risk involves decomposing the corporate bond yield ($y$) into two parts:
$$y = r_f + \text{Credit Spread}$$
Where $r_f$ is the risk-free rate. The credit spread compensates investors for the expected loss from default and the uncertainty of that loss.
B. Integrating Credit Risk Models
Sophisticated models integrate credit risk in one of two main ways:
- Structural Models (Merton-Style): Treat default as a consequence of the firm’s asset value falling below its debt liability. The model links the firm’s equity volatility (observable) to its asset volatility (unobservable) to derive the probability of default, which, in turn, informs the credit spread used to discount the bond component.
- Intensity Models (Reduced-Form): Model default as an unpredictable, instantaneous event driven by a hazard rate (default intensity). The risky cash flows are discounted at a risky rate that explicitly incorporates the hazard rate.
C. The Credit-Equity Correlation
A key subtlety is the negative correlation between the underlying stock price and the issuer’s credit spread. As the stock price falls, the credit risk typically increases (wider spread), reducing the bond floor value. The model must consistently capture this joint default-equity dynamic.
3. Capturing Volatility and Risk Components
The arbitrage strategy is essentially a long volatility position. Modeling must accurately capture the value of the embedded option.
A. Volatility Components
- Implied Volatility (IV): The volatility priced into the CB’s market price. This is what the arbitrageur seeks to exploit. If the CB’s implied volatility is lower than the expected realized volatility or the IV of listed options, the CB is considered underpriced (long volatility trade).
- Realized Volatility: The actual volatility of the underlying stock during the holding period, which drives the gamma profits from dynamic rebalancing.
B. Option Pricing Models
Convertible bond valuation typically requires adapting option pricing models (like Black-Scholes or binomial/trinomial trees) to account for:
- American Exercise Feature: The holder can convert at any time, requiring a dynamic programming approach (like a binomial tree) to value the early conversion option.
- Issuer Call/Put Options: Most CBs are callable by the issuer and sometimes puttable by the investor, adding complexity that the model must evaluate at every time step to determine the optimal exercise strategy for both parties.
The ultimate goal of the convertible arbitrage model is to accurately calculate the theoretical fair value of the CB. Any difference between this model price and the market price represents the mispricing, which the delta-neutral strategy attempts to capture while remaining immunized against broader market movements.
Do you have an inquiry? Schedule a free initial consultation