Why a Portfolio Can Outperform with Beta Below One
Summary
The document explains why a portfolio can beat a benchmark even when its estimated CAPM beta is below one. In a time-series regression of excess portfolio returns on excess benchmark returns, beta measures sensitivity to benchmark movements; it does not by itself determine the portfolio's average return relative to the benchmark. Positive alpha or exposures to risks absent from the benchmark can support higher average returns despite lower market sensitivity.
The answers also give an intuitive directional case: a portfolio with beta below one tends to lose less than the market in a declining market and may therefore outperform over that period. In a steadily rising market, the same lower sensitivity can leave it behind. These directional claims assume a simplified portfolio with no idiosyncratic risk; real portfolios have additional risk, so outcomes are tendencies rather than guarantees. The discussion is conceptual and provides no empirical portfolio results or guidance on estimating beta reliably from a short sample.
Key ideas
- A beta below one describes lower sensitivity to benchmark returns, not necessarily lower average performance.
- Positive alpha or exposures outside the benchmark can allow a portfolio to outperform.
- Lower beta can help relative performance in a declining market because losses may be smaller.
- In a consistently rising market, lower market sensitivity can hinder relative performance.
- Idiosyncratic risk and changing conditions make these directional relationships uncertain in practice.
Tags
Full text
# Outperform the market with a Beta lower than 1, is it possible?
# Outperform the market with a Beta lower than 1, is it possible?
I have a portfolio which seems to have outperformed the benchmark for the past 2 years however the risk system I use advises that using recent past data (lookback period of 2 years) the Beta to this benchmark is less than 1. I find this counter intuitive, how is this possible?
## Answer by Stéphane (score 2)
https://quant.stackexchange.com/a/51386
Just to be entirely clear, when you say Beta, you really mean that you estimate a CAPM-style time series equation of the sort: \begin{equation} R_t(portfolio) - r_f = \alpha + \beta*(R_t(benchmark) - r_f) + \epsilon_t \end{equation} by OLS, right? If so, you can get an estimate below unity and still outperform the becnhmark. If you capitalized on either pricing mistakes or risk exposures that aren't found in the benchmark portfolio, you would get higher returns on average.
## Answer by Haphy_Paphy (score 2)
https://quant.stackexchange.com/a/51394
It is possible to outperform the market with a beta lower than 1.
Imagine a portfolio that has a beta = 0.5 and is only exposed to systematic risk (no idiosyncratic risk), then the following three statements are true.
If the market prices are uniformly decreasing, then a portfolio with a beta < 1 will always outperform the market.
If the market prices are uniformly increasing, then a portfolio with a beta < 1 will never outperform the market.
If market prices are not uniformly increasing or decreasing, then it is not immediately clear how a portfolio with a beta < 1 will perform in comparison to the market.
In a realistic scenario the portfolio will also be exposed to idiosyncratic risk, and then the words "always" and "never" must be replaced with "tends to".
So to sum this up, it all depends on the market performance, but basically in a bear market a portfolio with a beta < 1 should outperform the market because the portfolio losses will be lower than the market losses.Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.