Momentum's Run and Stumbles
Bridgeway's John Montgomery discusses momentum's long history of outperformance and troubles during market inflection points.
Shannon Zimmerman: For Morningstar, I am Shannon Zimmerman, here today with John Montgomery of Bridgeway Capital Management, which is based in the great state of Texas and the city of Houston.
John, thank you very much for joining us today.
John Montgomery: Very glad to be here, Shannon.
Zimmerman: John is in for a panel discussion that we're going to have tomorrow about the persistence of momentum, which has a decades-long track record of outperformance. I want to begin first, though, with a big picture take on Bridgeway and how you manage your funds. I believe you have 13 funds in the lineup now; all of them are quantitatively managed. What is the case for quantitative management? And then talk a little bit, because you have a interesting background as well, what was the genesis of the approach that you use at Bridgeway?
Montgomery: I think the centerpiece number one with quantitative investing, and we also say evidence-based investing, so able to look backward, there are some big advantages around risk management because returns, as we know, go up and down and with all the caveats of past performance doesn't guarantee future returns--risk tends to be more persistent, however, and the quantitative style really allows you to measure that and manage that well.
But beyond that, taking a emotion out of the process, to me, if you have a disciplined process, now there are individuals and portfolio managers that may have a quant-based process, but if at end of the process they overlay a fundamental or come up with ideas about something that might happen with the stock market or the economy, then you're really significantly getting away from what we call evidence-based or quantitative investment.
Zimmerman: That would be more tactical, lurching into and out of the market, for instance, not based on signals from your models, but just because you're making a market call, essentially.
Montgomery: That's right.
Zimmerman: So, as a segue into talking a little bit more about momentum, and that's obviously not the only model that you have at Bridgeway--I think you have 17 models--although they're not deployed across all…
Montgomery: 17 stock-picking models.
Zimmerman: Right, one of which is momentum, and again that one's not used at all of the funds, but there is one fund that you've launched, I believe, last year, the Bridgeway Small-Cap Momentum Fund, and as we were speaking earlier, there is a risk-control aspect to that, but in terms of models, it's the only single-model fund at Bridgeway, and it's based purely on price momentum, if I understand correctly?
Montgomery: That's correct. So the momentum feature of that is purely simple priced-based momentum. It's looking back 12 months, the relative strength, the companies that have done the best over the last year, looking to hold those at least three months into the future and we rebalance on a roughly quarterly basis on that. The thing that's unusual about Bridgeway is that we overlay a risk analysis feature to that, so there are really two things going on with that one model, the momentum and then the risk part.
The risk part solves a major problem in the design of the fund that exists within momentum styles generally, which is they do great in a steadily up-moving market--that's what momentum is. They tend to do poorly at the infection points, and consequently the volatility of returns in a momentum style tends to be quite high. Our flavor of that actually ratchets that back down, generally speaking, to the market level, and we do much better at those infection points.
Zimmerman: At the market level as defined by the S&P, you're trying to have a beta of one relative to the S&P, is that…
Montgomery: Well this is a small-cap fund, so the Russell 2000 is our performance benchmark and how we think of the market in that area is.
Zimmerman: And the goal of the risk optimization is to have a neutral beta relative to the Russell 2000?
Montgomery: The beta actually can vary over time, but it's specifically in those infection points. So think of 2008, which wasn't too bad, but actually 2009 was much harder. It was one of the kind of blowups--simple price momentum lagged on the order of 10% in the small-cap space, and that's very unusual. That only happens every decade, maybe less. But the risk-adjustment part allowed us to have a slightly…
Zimmerman: Smoother ride?
Montgomery: ... I have to caveat that, of course, the performance of our fund is only a year old, so this is a model, and I need to disclose that with all of the normal things of the past doesn't equal the future. When you're talking about a back-tested return, there is an extra level of caveat there. We did just celebrate the one-year anniversary of the fund, and it's been an unusual period for momentum, both price momentum and other kinds of momentum.
Zimmerman: When you mentioned the infection point vulnerability, and a lot of funds that use momentum either as a sole model or as a part of an overall quantitative approach to management suffered quite a bit because momentum is vulnerable at infection points, which stands to reason--you're investing primarily based on the things that have done the best over the last 12 months, that was the most thoroughly documented form of price momentum. Then when things turn on the dime, as they did in late 2008, well then you're really stuck holding the wrong stocks for the most part.
Notwithstanding that infection point vulnerability, momentum has a remarkable track record of generating outperformance over decades, the CRSP Data goes back to 1929, I think, and over all the subsequent decades, it has generated alpha, and tomorrow we'll be on a panel with Tom Hancock of GMO who's done some number crunching around that. One flavor of momentum three annualized percentage points of outperformance relative to the broader market, a slightly tweaked version of price momentum, where you discount the last month's returns…
Montgomery: ...Short-term reversal effect.
Zimmerman: Exactly has done even better. It's such a simple strategy, so easily replicated, why does it persist as being an alpha-generating tactic?
Montgomery: Well so there is a whole body of research that's trying to answer that question, and honestly, I don't think we have a definitive answer to that. Value investing is kind of been around longer and more accepted, and it's a little easier to get your head around value, when something is very cheap relative to either prices or sales or book value. Something that creates value into the future, it kind of makes sense that reversion to mean prices would go up in the future.
So, price momentum is a little bit different. There are three theories about why that might be, and I think they are interesting. One is that there is a delay in absorbing news that comes out about a company. So, for example, an earnings report comes out and it's a good report; it's better than what Wall Street might expect. So you'd expect the price to have some adjustment with that and especially to the degree that that persists.
So, there is some theory that people discount that. It's like, things haven't been too good over the last few years. All of a sudden you have a good quarter and people are like, "I'm going to sit back and really see if this continues to happen." So, there is some theory that there could be some delay. Interesting of that one theory, that's the one that I think arguably could have declined over the last decade versus prior history.
Zimmerman: As it's become more well-known.
Montgomery: Well simply because we have more information, and computers are easily available to process it. I don't know whether that's true or not, but that's one theory.
Another theory that I really hold more sway in is that shareholders tend to sell their gains early, they lock in their gains and they let their -- interestingly they let their…
Zimmerman: Losers run.
Montgomery: ...Their losers run, which means there is some artificial pricing that would tend to have a momentum effect with it. So, that's a theory, and I've actually seen that first-hand at Bridgeway. In 1998 we had a very volatile ultra-small company fund that went way down with the Asian flu. So, small-value stocks went way down, and our shareholders stuck with us. What was interesting, when we got all the way back whole a year later, which was amazing that we got back that quickly, we actually did have some redemptions, and it was clearly the people that had gotten back to their previous peak they thought "oh, that was more volatility than I could handle."
So, that would be an example of real-world in that. And then the biggest one, though, is simple kind of herd following. Joe got this great stock pick last year, made a lot of money. I'm a little late, but maybe I should get in now or the Internet bubble, obviously.
Zimmerman: People like winners in sports and in stocks, as well.
Montgomery: That's right.
Zimmerman: You were referring earlier to the value premium, and there's also the small-cap effect, that over time less-expensive stocks tend to outperform more-expensive stocks, and smaller cap stocks outperform their larger counterparts.
At an earlier Morningstar Investment Conference, Jack Bogle gave a fantastic speech called "The Telltale Chart," where he looked into those things, and of course, he a great believer in the mean reversion. He looked into the time series, where in the aggregate those things are true, and showed that relatively small slivers of the time series accounted for these great premiums that people believe are going to continue into perpetuity.
Zimmerman: Is it possible that that's the case with momentum, too? You look at the outperformance over a 75-year track record. How much of that 75 years really gave momentum its effect?
Montgomery: Well that's interesting, I mentioned earlier that Bridgeway's small-cap momentum fund is the only fund we have based on a single model, why is that? Normally, we like to put models together that diversify the risk of that model's buy within a fund. The reason we were willing to do that with this particular risk-adjusted momentum is the extreme consistency of outperformance.
Zimmerman: Okay, interesting.
Montgomery: So this is our single most consistent model at Bridgeway, and so we were willing to put it out there all by itself. Another reason we would do that is it's a very strong diversifier of the value style.
Zimmerman: That's right.
Montgomery: So, if you are overly dependent on value managers, which by the way after the last decade which was strongly valued, many peoples portfolios have moved that direction. Momentum style tends to be actually a much better diversifier of value than it is just straightforward growth investing.
Zimmerman: Right and a lot of the academic literature confirms exactly what you are saying. There is a way in which you can sort of use a momentum sleeve in your portfolio as a proxy or a substitute for growth, which over the same very lengthy time series, value has outperformed, growth has of course underperformed. So it's an alpha detractor in the broader sense.
John Montgomery, thank you very much for being with us today.
Montgomery: Okay. Thank you very much. I enjoyed it.
Zimmerman: For Morningstar, I'm Shannon Zimmerman.
Shannon Zimmerman does not own (actual or beneficial) shares in any of the securities mentioned above. Find out about Morningstar’s editorial policies.