Hull-White Model

What Is the Hull-White Model?

The Hull-White model is a mathematical model that describes how short rate interest rates move over time. It is widely used by banks and financial institutions to price interest rate derivatives and manage interest rate risk. Economists John Hull and Alan White introduced it around 1990 as an extension of the earlier Vasicek model.

The model's key feature is that it can be set up to match the current yield curve exactly. The Vasicek model assumed that rates revert to one fixed long-run average, so it generally could not reproduce the bond prices observed in the market. Hull and White allowed the long-run level to vary over time, so the model fits today's curve while still describing how rates may move in the future.

How the Model Works

The one-factor version is usually written as:

dr = [θ(t) − a·r] dt + σ dW

Here r is the short rate, a is the speed of mean reversion (how quickly rates pull back toward their average level), σ is the volatility of rate changes, dW represents random market movements, and θ(t) is a time-dependent term chosen so the model reproduces the starting yield curve. In practice, a and σ are estimated through calibration to market prices of instruments such as swaptions, which are options on interest rate swaps.

Because the model uses a normal (Gaussian) distribution for rate changes, it can produce negative interest rates, a useful feature in markets where rates have fallen below zero.

Why It Matters and Limitations

For investors, the model is mainly a tool for professionals who value bonds with embedded options, such as callable bonds, and other complex interest rate products. Individual investors rarely use it directly, but it helps explain how institutions price the interest rate risk that affects bond values.

The model has limitations. As a one-factor model, it assumes all rates move together in a single pattern, which may not capture cases where short-term and long-term rates move differently. Results depend heavily on the chosen parameters, and a model that fits current prices well may give less reliable results when market conditions change.