As a speaker at this summer’s Risk Management Conference in Muskoka, ON (August 12 to 14), Andrew Ang, author, Asset Management: A Systematic Approach to Factor Investing will lead a session on factor investing. In advance of the conference, we asked Andrew a few questions about the nature of factor premiums and how to exploit them.
How is factor investing different from more traditional value and small-cap tilts?
Factors are drivers of returns over the long run and investors holding those factors are rewarded with risk premiums. The traditional value and small cap tilts are just two examples of factor risk premiums. There’s good academic theory that the value and size premiums will, over the long run, outperform the market. But there are more than just size and value risk premiums. There’s a whole library of them, so why limit yourself to just those two?
How many are there?
Cam Harvey at Duke has a paper that documents 300+ factors, but many of those I think are ephemeral, many of them are not investable, many of them don’t survive transaction costs.
I think there are around five to 10 risk factors. I don’t want to put an exact number on them because it really depends on the type of investment. How often you can make decisions, for example? Your ability to trade. And some of them require varying amounts of skill to harvest.
You say that what matters is not asset class labels but bundles of overlapping risk factors. How do you optimize factors?
They are two separate questions. First, what are the factors driving asset classes or individual securities? Assets are bundles of factor risks. Credit risk manifests in bonds, equities, structured credit, and derivatives. One of the problems that investors encountered in 2008-2009 was precisely that they didn’t take into account these factor risks properly and double-counted many of these risks.
Second, in answering how we optimally combine factors we must ask what factor exposures are right for me? Then given the factor exposures, how do I express those factor exposures in a menu of different asset classes, or funds, or investment vehicles? I am developing tools to help investors optimally combine these factors. Different types of factors are going to be appropriate for different types of investors.
Ultimately, it depends on the type of investor you are. Not everyone can be a value investor. For every value investor, there has to be a growth investor. You should be a value investor if you go against the trend or can stomach the losses when value does badly—like during the financial crisis or the late 1990s. Not everyone can reap a volatility risk premium. For some people, it’s better to buy protection when volatile times come. For every person who sells volatility, there has to be a person on the other side.
When we’re talking about factors, inevitably there’s a time horizon involved. How do we match the time horizon over which the factor expresses its outperformance to investor risk tolerance.
This is a general problem and it is valid beyond style risk factors. Even for plain-vanilla equities, people too often overestimate their risk tolerance during good times. They get scared and disinvest right after a crash—right at the wrong time. Equities, during the 2000s, had negative returns relative to bonds. You have to stay the course.
We see the same issue for general factors. Factors vary over time, just like the equity risk premium does. Value goes through periods of underperformance for years, like during the late 1990s; and momentum had a huge draw down during the financial crisis. We have to hold for the long run. We have to stay the course. The worst thing an investor new to factor investing can do is to pile on after good performance and give up after bad performance. Recently, factor risk premiums have done well, but the bad times will come. That’s the reason there are factor risk premiums: they are a long-term reward for the possibility for suffering losses in the short term.
To learn more about the Risk Management Conference, please visit the conferences section of the CIR website. If you are interested in attending this event, please email Alison Webb to be considered, as limited space available.