4-3 Optimising a Portfolio
A page using the efficient frontier to think about how to compare weights when holding the same products
The previous pages established that diversification, for shares and funds alike, is judged by contents rather than count. The next question that tends to surface, once your candidate products are settled, is what proportion should I hold each at. Holding the same products, different weights change both the return you can expect and the movement you get. Here we look at the efficient frontier as a handle for comparing weights.
- Optimisation is a tool for comparing which weights make more sense, given the same products.
- Mean-variance optimisation is the basic idea behind drawing an efficient frontier.
- What an individual investor should look at first is not the model's detail but
where on the frontier do I sit. - Further left limits movement more; further right reaches for higher return.
1. What optimisation asks is which weights are better for the same products
When building a portfolio, the question that always follows choosing products is what percentage of each do I hold. Even holding both an equity fund and a bond fund, holding them 50/50 versus 70/30 changes both the return you can expect and the movement.
The curve in the diagram is the efficient frontier — the line joining the combinations with the highest expected return achievable at each level of risk. If your current allocation sits below and to the right of that line, there is room to move closer to a more efficient shape by rearranging the weights, even with the same products.
2. Pick the point on the frontier that matches your risk tolerance
What an individual investor really wants to work out here is where on the frontier could I keep holding. Further left tends to mean smaller movement but a more modest expected return; further right tends to mean larger movement but a higher return within reach. So there is no single answer — where you pick changes with your risk tolerance.
As the start of Chapter 4 set out, what matters is not deciding mechanically by age but how much movement can I absorb. Could you keep going through a 20% fall? Would a 30% swing either way not bother you? Or would you rather avoid large moves altogether? That feel decides whether you pick the left of the frontier, the middle, or the right.
3. Use models as a first draft, and read the finer theory in the column
The best-known basic idea for drawing an efficient frontier is mean-variance optimisation, the starting point of modern portfolio theory. But such models are sensitive to the assumptions you feed them, and a slight change of outlook can produce extreme weights. Black-Litterman, widely used in practice, is positioned as a development that makes such extremes easier to contain.
If you are interested in the detail of the models, see the column What are portfolio optimisation models?. This page prioritises grasping two ideas about weights: even with the same products, some weights are more efficient, and from among those, pick the point that matches your risk tolerance.
4-2 The Three Axes of Diversification
A page starting from the rough guide of around ten holdings, then checking diversification across product type, sector, and country/region
4-4 The Basics of Rebalancing and Review
A page on how to tidy up when an allocation drifts, and the order to review things in