Black-Litterman Model
The Black Litterman model starts from market implied returns and blends in an investor's views with stated confidence, producing stable and intuitive portfolios.
Mean variance optimisation promises the best portfolio for given inputs, but in practice it often recommends extreme, unstable positions because expected return estimates are so noisy. The Black Litterman model, developed by Fischer Black and Robert Litterman at Goldman Sachs and published in 1990 and 1992, offers a solution. It starts from the returns implied by market prices, which by construction produce a sensible, diversified portfolio, and then adjusts them toward the investor's own views in proportion to how confident the investor is.
The two ingredients#
| Ingredient | What it is |
|---|---|
| Equilibrium (implied) returns | The expected returns that would make the current market cap weighted portfolio optimal |
| Investor views | Opinions such as "European stocks will beat US stocks by 2%", each with a confidence level |
The model combines them using Bayesian statistics: the market equilibrium is the prior, the views are new evidence, and the result is a posterior set of expected returns. See Bayesian Statistics.
Reverse optimisation#
Instead of guessing expected returns, Black Litterman works backwards from market weights:
Implied excess returns = Risk aversion × Covariance matrix × Market weightsAdding views#
Views can be absolute or relative:
| Type | Example |
|---|---|
| Absolute | "US government bonds will return 4% over the next year" |
| Relative | "Emerging market stocks will outperform developed market stocks by 3%" |
Each view has a confidence. A high confidence view moves expected returns, and therefore weights, strongly; a low confidence view barely moves them. Assets not mentioned in any view stay close to their market weights, except where correlations link them to assets that are.
Why it produces better portfolios#
| Problem with plain optimisation | Black Litterman response |
|---|---|
| Extreme weights from noisy inputs | Starts from market weights, a diversified anchor |
| Requires a return forecast for every asset | Only requires views where you have them |
| Results swing with small input changes | Blending stabilises outputs |
| Hard to explain | Departures from market weights trace directly to stated views |
Practical issues#
- Choosing risk aversion and the uncertainty scaling parameter (often called tau) affects results.
- Specifying view confidence is subjective; methods exist to express it as a percentage.
- The covariance matrix still needs careful estimation. See Covariance and Correlation.
- Market weights for some asset classes, such as global bonds or private assets, are hard to measure.
Who uses it#
Institutional investors, multi asset managers and robo advisers use Black Litterman or similar Bayesian approaches for strategic and tactical asset allocation. It is also a useful mental model for individuals: start from a neutral, diversified mix and tilt only where you have real conviction. See Asset Allocation and Portfolio Construction.
Frequently asked questions#
What is the Black Litterman model?#
A portfolio allocation method that blends market implied expected returns with an investor's views, weighted by confidence, to produce stable portfolios.
Why use implied returns?#
They are the returns that make the market portfolio optimal, giving a neutral, diversified starting point instead of noisy forecasts.
How does Black Litterman differ from mean variance optimisation?#
It uses the same optimiser but with blended expected returns, avoiding the extreme, unstable weights that come from raw return forecasts.
Next, learn how factors explain returns in Factor Models.
3 quick questions on this lesson. Get them all right to finish it.
Turn on JavaScript to take the quiz.