Bringing quant investing to non-quants is tough but worthwhile. Wes Gray is founder and CEO of Alpha Architect, an asset management firm that focuses on high conviction Value and Momentum factor strategies. He has published multiple academic papers and four books, including three on investing. After serving as a captain in the United States Marine Corps, Wes earned an MBA and a PhD in Finance from the University of Chicago. We asked him about his views on quant investing and the importance of knowledge sharing.
Founder and CEO of Alpha Architect
“I started off as a stock picker and was always a fan of Ben Graham and Warren Buffett. I was also into special situations, such as liquidations, trading closed-end fund discount/premiums, and esoteric microcap companies. I was essentially the antithesis of quant. I really got involved in ‘99 at the height of the tech bubble when I was in college. I was lucky to have a heavy value bias at the time, so I felt bad for a while as the QQQ went flying higher and higher, but eventually the tech bubble crashed. I had a great run for five or five years and continued investing while I was on the PhD program at the University of Chicago.”
“While I was ‘stock picking’ on the PhD program, I was simultaneously learning a lot about quantitative techniques, factor investing, and behavioral finance. I quickly recognized the folly of my ways when I compared my stock picking performance to the generic small-cap value factor portfolio from 1999 to around 2004. Long story short, I had some ‘alpha’ as a stock picker, but I could have saved myself a lot of brain damage (and eliminated a lot of stress) by simply buying a portfolio of cheap small-caps. This experience opened my eyes to the power of quantitative techniques.”
“Investing is a personal thing and one should understand their own strengths and weaknesses. For me, I am a fan of automating 100% of the stock selection process. The risk of behavioral bias is simply too high relative to the potential of my ability to add value via a judgement call. For others, this trade-off might be more favorable.”
“There is an argument that a ‘quantamental’ approach, which combines the benefits of quantitative techniques with human judgement, might work out for an asset manager. I’ve written a long piece1 on the subject and I believe the evidence is stacked against quantamental, but reasonable people can disagree on this topic.”
“I’d also like to point out that while we are 100% systematic, I am a fan of having humans review the data and sanity check the outputs from a model. We won’t override a model’s selection, but if there are data errors that cannot be reconciled, we will move past that selection to minimize ‘garbage in, garbage out’ type problems.”
“No, this is not easy, but we think it is worthwhile. And our goal is not to convince everyone that ‘systematic’ is the answer. We are really set up to help those who would like to be helped. The reality of educational efforts is they require the student to have an interest in learning. We can’t force-feed someone knowledge and expect them to gain any benefits. Getting people to understand that less is more can be quite challenging. Our whole lives we are taught that when you try harder, you often get more. Investing is a unique arena where beating your head against the wall doesn’t make your skull thicker but often gets you injured.”
“We think value and momentum are the kings of the factor world, but buried inside our systems are elements of size and quality. We believe the characteristics of cheapness and momentum proxy for elements of risk and mispricing that are tough to exploit. Quantifying how much is associated with risk and how much is attached to mispricing would give too much precision to an inherently noisy process. But if I had to rank order them, I’d say for value you have more fundamental risk and less mispricing, and for momentum you have more mispricing and less fundamental risk.”
“I don’t think there is much to be gleamed from hammering on the data at this point. Many of the ‘new findings’ inevitably end up being noisy proxies for what was already known, especially when you redo the results and stress test the findings in your own laboratory. I have been hacking on data for almost 20 years now and there doesn’t seem to be much that is new under the sun. Buy cheap, buy strong, and hold ‘em long. Wish someone had told me that 20 years ago – I would have saved myself a lot of time.”
“That said, portfolio construction is still an important area that can be used to differentiate yourself from the crowd. You can take the same exact signal and vary the dimensions of turnover and number of holdings to create wide dispersion of expected outcomes. There are also elements of portfolio optimization that can be important; for example, how to minimize trading costs or taxes. Plenty of areas where an asset manager can earn their keep via craftsmanship decisions.”
“We’ve written 30 or 40 blog posts on this topic. I think it is interesting. In the end, from a long-only perspective, I believe that a cheap/quality portfolio will act similarly to a low vol/beta portfolio most of the time. So setting aside minutia arguments, I think it is probably close to six of one, half a dozen of the other. However, I fundamentally believe in the risk/behavior arguments underlying ‘value’ over the arguments underlying the low risk effect. The good news is I am not the arbiter of these debates. The best strategy is the strategy you can stick with through thick and thin. That will differ for all investors.”
1Gray, W., 2014, ‘Behavioral Finance and Investing: Are you Trying Too Hard?
This article is an excerpt of a longer interview.