Active managers have a mixed track record; some have done very well, others significantly less well, and some have simply slipped off the records, unable to generate or retain capital in the wake of poor returns. One thing is reliable, however; if a manager’ performance is consistently better, they are probably doing something consistent. And if they’re doing something consistent, then it should be possible to emulate them.
Explaining managers’ performance is too big a job for one blog; for now I will simply note that a number of people are doing it with some success1
. Instead, I want to introduce a number of purely mechanistic allocation strategies that have worked historically. History suggests that these strategies, or “styles”, can be used to construct portfolios with significantly higher risk-adjusted returns than simple capital-weighted benchmarks. The list below is in no way comprehensive, it is just an introduction to some of the simpler styles that have been successfully implemented in a variety of markets. Three of the most significant styles are:
The principle is very simple - buy cheap stocks and sell expensive ones.
By looking at company fundamentals, you can create a measure of whether a stock is undervalued, for example P/E and EV/EBITDA.
The implementation is at least as old as 1992, when it appears in a Fama-French paper.
Again, the principle is simple - buy less risky stocks with lower betas to the market, and sell riskier stocks with higher market betas.
The idea is that many investors overlook the less glamorous stocks in boom times, and that these stocks are more resistant to price bubbles.
Historical studies suggest low volatility stocks have higher risk-adjusted returns (for example, Haugen and Heins, “On the Evidence Supporting the Existence of Risk Premiums in the Capital Market”).
The principle is to buy recent winners and sell recent losers, and do it quickly.
The idea is that information does not impact the market instantaneously, and there are lags between causes and effects.
Historical studies suggest momentum investing offers outperformance (albeit with a significant skew) - for example, Jegadeesh and Titman (1993).
We have run some analysis, using a variety of data sources (in particular Deutsche Bank’s indices and very long-term data from Ken French), which shows outperformance of all these strategies. Moreover, we find that they all show reasonably low correlations to both each other and the benchmark, meaning that they could offer meaningful diversification in an investor’s portfolio. As an illustration, the correlations for different styles in equities are shown below.
Figure 1: Correlations of different strategies
Source: Deutsche Bank, Calculations: Redington
These strategies are not magic bullets. The numbers we’ve shown do not account for transaction and balance sheet costs, which could be significant; there is a real risk of survivorship bias; and there are substantial difficulties in converting these broad ideas into a sensible framework for investment decisions.
Despite those provisos, these (and other) strategies have displayed the repeated and explicable success. Therefore, the most sensible conclusion is to think of these strategies as separate and rewarded risk types- as forms of beta in their own right
. Looking at risks from this more detailed perspective could allow risk managers a more accurate picture of their portfolio’s positioning.
 And often along the lines outlined here, e.g.:
Please note that all opinions expressed in this blog are the author’s own and do not constitute financial legal or investment advice. Click here for full disclaimer