What it means
Interest rates do not drift off to infinity. They tend to wander around a typical level, rising when they are low and falling when they are high, and the Vasicek model builds that tendency in through a property called mean reversion.
The model has three main settings. One is the long-run average rate, another is the speed at which the rate is pulled back towards that average, and the third is the volatility, which is the size of the random shocks.
On each small step in time, the expected change in the rate depends on the gap between the current rate and the long-run average. A rate below the average is expected to rise, a rate above it is expected to fall, and a random shock is added to every step.
By simulating many paths, analysts can price bonds, estimate the risk of loan portfolios and test how a business would fare in different rate environments. The model also produces a full yield curve, which is the pattern of interest rates over different maturities.
One well-known limitation is that the model allows interest rates to become negative. That used to be seen as a weakness, but in periods when rates were below zero in some markets, it looked more realistic.
The model is also simple, and a single factor cannot capture every kind of curve movement. For a non-specialist, the practical message is that the model turns a vague worry about rates into a set of numbers.
It shows how likely rates are to rise or fall by a given amount, which makes it easier to decide how much to hedge, how much cash to hold and how much risk the business can bear. More elaborate models extend the idea, and the output should always be tested against market data.
In practice
Real-world examples.
Example
A bank's risk team simulates 10,000 paths for short-term rates using the Vasicek model. They use the results to estimate how much its profit from lending would vary. The risk committee reviews the worst 5% of outcomes and asks whether the bank could absorb them without breaching its limits.
Example
A pension fund prices a long-dated bond portfolio and needs a view on future rates. The actuary calibrates the model to current market yields. She uses it to estimate the value of liabilities under different paths.
Example
A corporate treasurer is considering fixing the rate on a floating-rate loan. The treasury team uses a simple Vasicek simulation to see how often rates would be above the fixed rate. The results help the board decide whether paying a fixed rate is worth the certainty it buys.
Formula
Calculation
Change in rate = a x (b - r) x dt + sigma x square root of dt x z
Here r is the current rate, b the long-run average, a the speed of reversion, sigma the volatility, dt the length of the time step in years, and z a random draw from a standard normal distribution.
Suppose r = 2%, b = 4%, a = 0.5, sigma = 1% and dt = 0.25 (one quarter). The pull towards the average is 0.5 x (0.04 - 0.02) x 0.25 = 0.0025, or 0.25%. If the random draw is z = +1, the shock is 0.01 x 0.5 x 1 = 0.005, or 0.50%. New rate = 2% + 0.25% + 0.50% = 2.75%. If z = -1, the new rate is 2% + 0.25% - 0.50% = 1.75%.Case study
Seen in the real world.
This illustrative story is about a fictional lender, Sandford Home Finance, whose finance team wanted to understand how its profit would behave if rates moved. The team set up a Vasicek model with a long-run rate of 4%, a reversion speed of 0.5 and a volatility of 1%.
After simulating thousands of rate paths, they found that in a noticeable share of the paths, rates would fall below 1.5% within two years, squeezing the margin on its fixed-rate loans. The finance director used this finding to decide how much to hedge with interest rate swaps.
The fictional team also noted that the model's results depended on the chosen settings, so it re-tested them against actual market rates each quarter. The case shows how the model supports planning without claiming to predict the future.
Watch out
Common mistakes.
- Treating the model's output as a forecast. It produces a range of possible paths, not a prediction of what will happen.
- Using settings that are never updated. The long-run average and speed of reversion should be calibrated to current market data.
- Ignoring the possibility of negative rates. The model can produce them, and this should be considered in the analysis.
Questions
People also ask.
What does mean reversion mean?
It is the tendency of a rate to be pulled back towards a long-run average after moving away from it.
Who developed the model?
Oldrich Vasicek, in a paper published in 1977.
How does it differ from the Cox-Ingersoll-Ross model?
The Cox-Ingersoll-Ross model is similar but prevents rates from going negative, while the Vasicek model does not.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%Related