Only puny secrets need protection. Big discoveries are protected by public incredulity.
Random portfolios have the power to improve the practice of asset management in several ways. Here are six.
1) Measure active managers
There is no convincing evidence that more than a handful of funds have consistently outperformed. This should tell every active fund manager on the planet that the present form of performance measurement is inadequate.
Active fund managers presumably believe outperformance is plausible. Perhaps you don’t believe any fund truly outperforms. That’s possible (with a few exceptions), but we are lacking convincing evidence of that as well.
Random portfolios can provide performance evidence for any particular fund. Performance measurement is — or should be — the evaluation of the quality of decisions. Fund decisions can be mimicked randomly to really show whether they look like skill or luck.
2) Improve active managers
The same technology that can test how a fund manager has done can be used by the fund manager to try to improve their processes.
For example, backtests of systematic funds can be calibrated to see how significant the test’s performance really is.
3) Unchain active managers
Fund managers are commonly constrained by the investment mandate to have a maximum tracking error from their benchmark index. The logic is that this allows the fund’s performance to be assessed. Let’s count the ways this is wrong:
- Performance measurement via a benchmark is hopelessly noisy — it takes decades to get a real answer.
- There is little value for the investor to pay for active management to get results that are very similar to those of an index that they can get virtually for free.
- A fund manager that can outperform will, in general, do better when the tracking error constraint is removed
Much better is for the investor to put some money in index funds, and to use random portfolios to measure the performance of active funds to see if they are adding value.
Actually there should be tracking error constraints on funds — minimum tracking error constraints. It is in the investor’s best interest for the active funds they invest in to be as uncorrelated as possible with the indices that they invest in passively. That means a large tracking error.
4) Strengthen risk modeling
Random portfolios provide realistic test cases to put through risk models.
Often the only visible test of a risk model is of how responsive it is to changes in volatility over time. For some applications that is important. For other applications — portfolio optimization, for instance — it is completely irrelevant.
What matters for optimization is how well the model orders the volatility of portfolios. Given two portfolios, how well does the model predict which will have the higher volatility? You pretty much need random portfolios to answer this question.
5) Rationalize constraints
Portfolio constraints are ubiquitous. When we impose constraints, we are buying insurance — we are willing to give up some up-side in order to avoid some down-side.
So what does this insurance market currently look like?
We buy the same insurance our grandfather bought. We have no idea of either the cost of the premiums or the benefit (if any) of the policy.
We’re not going to know the effect of the constraints in the future, but we can see the effect they have had in the past. We can see — over some history — the distribution of returns (or utility) with one set of constraints versus the distribution with a different set of returns.
6) Entertain quants
As far as I can tell, there are two prevailing attitudes towards quants:
- they are the wunderkind of the industry
- they should be locked away from anything that matters
Whichever your attitude, you should give them random portfolios to play with.