The study text reads:
MVO often identifies efficient portfolios that are highly concentrated in a subset of asset classes, with zero allocation to others. In other words, lowest standard deviation is not the same as practical diversification.
I understood the theory, but can anyone tell me how exactly this plays out in the real world? What kind of concentration happens with an MVO, what asset classes ultimately get favoured?
Secondly, does this sentence talk about concentration amongst asset classes, or concentration within asset classes (example - does it talk about a concentrated allocation to US equity or does it talk about a concentrated allocation to fintech stocks within the US equity allocation?)
Can you share the exact book and page so we can read better the context of that explanation?
In advance, I can tell that MVO, as a pure mathematical procedure, sometimes will find that the most efficient portfolio is a combination of just a subset of asset classes from the all available to invest and not a combiation of all asset classes (each with some participation). This results come from the risk-return relationship some asset classes show. I mean, some asset classes would be just superior than others for a specified level of risk and therefore MVO tells you to concentrate most of your portfolio on those asset classes to get the most efficient portfolio (highest sharpe). This is an “inconvenience” because it pass beyond the logic of diversification, so one method to “adjust better” the recommendation of the model is to apply constraints.
In the real world, most portfolio managers apply a wide variety of constraints to their MVO models, so it is a sound methodology.