Momentum tekee kipeää

Thomas Hancock: Thank you. My pleasure.

Zimmerman: Tom is here at the Morningstar Investment Conference, and we've just finished up what I thought was a terrific panel discussion with John Montgomery of Bridgeway and Doug Ramsey of Leuthold about price momentum, and you've done a great deal of research into this persistent effect over the course of many, many decades. It seems to be the case that a simple strategy of just investing in the stocks that are performing the best over the last 12 months significantly outperforms.

And you've put together a lot of research in the paper called, "A Contrarian Case for Following the Herd," which is available on the GMO website, and I think it's a terrific and very accessible paper, both for folks who know about momentum and those who just want to learn about it. But we can learn something now. Tell us a little bit about why something that is as simple as a price-momentum strategy hasn't been arbitraged away?

Hancock: I think the main reason for that is it's just a very painful strategy to run, and the bulk of assets in this world are managed indirectly--people like me or organizations like ours managing money for clients--and when you're using a price-momentum strategy you're basically buying stocks that go up, and hoping they go up some more, but if they don't, they go down, you sell them and buy the other ones that went up instead.

So, the situation in which you lose is you've just bought stuff that's gone up, then it goes down, and then you sell it, and that just makes you look like an idiot, right? It's a very uncomfortable thing to do, and frankly, for most professional investors, it's a very unintuitive thing to do, because most of us are kind of naturally born and bred contrarians. We at GMO believe very much in contrarian value-based investing.

So, the pain point for momentum is so high that when you see these drawdowns as you saw, for example, coming out the market bottom in 2009, when momentum underperformed, the wipeouts are just too painful. So, there's a limit to the arbitrage there and people can't exploit it, at least can't exploit it completely so it disappears.

Zimmerman: What are some of the factors that actually culminate in the momentum effect--and we've talked a little bit about some behavioral finance inputs--but ... to me one of the most provocative parts of your paper is the chart that shows the way in which price momentum seems to anticipate earning success.

Hancock: Right, there is sort of a fundamental argument ... and a behavioral argument, and they both work. The fundamental side of it is that you if you, for example, line up stocks based on trailing 12 month return and look at the stocks that are in the best 10%, on average, they've delivered 20% earnings growth, well above the market. The flipside of that, the stocks that had the worst momentum, have earnings decline, and the relationship is pretty smooth between them.

So, what that means is, markets are actually reasonably efficient at a first order. The stocks that people bid up in price and go to slightly higher valuation multiple as a result, are actually going to deliver higher earnings growth. So, the sort of first fear that if you buy momentum you are buying expensive stocks and will therefore underperform, hasn't been true. In fact, there has been enough extra earnings growth leftover for you to make money.

Zimmerman: Right. Two other aspects I want to ask you about. So, one of the weaknesses of momentum is its susceptibility to inflection points. When there's a dramatic turn in the market, as there was in 2008, when momentum performed quite poorly, and then in 2009, when the turn was in the positive direction for the market overall, but not for momentum as a tactic, which underperformed relative to the broader market. Are there ways that investors who might want to use a momentum strategy in at least a part of their portfolio, can they mitigate that inflection point vulnerability?

Hancock: Momentum is a strategy which in isolation is very volatile and vulnerable, but can be paired with contrarian strategies, value-based strategies that tend to work at exactly the other periods. Turning points tend to be great for value. Long-running trends are what run away from a value manager, and that's when momentum works.

So, the strategies that GMO manages, we'll typically have 70% or so of the assets in a more contrarian mean-reversion-oriented value strategy, 30% in momentum, and when one's zigs, the other zags. So, yes, they're kind of working at cross purposes. If you had perfect foresight about which way the market would go, you'd use just one or the other, but we don't, most people don't. I think a blend between the two gives you a much smoother ride.

Zimmerman: Yes. Diversification not just in terms of style or market-cap orientation, but the strategy. It's a different kind of strategy that you can complement a more fundamentally focused portfolio with, which is exactly what you do in the ones that you run for GMO, right? There's a sleeve that's dedicated to a price momentum strategy and the bulk of the assets are in this contrarian value style.

It's the case--and I think that this is also documented in your paper--that momentum is negatively correlated with value. Investors, you look at the time series, and it seems to be the case that value outperforms growth over the long haul. Should investors consider momentum as a proxy or perhaps even a substitute for the growth sleeve of their portfolio?

Hancock: We think so. I mean, the problem with traditional growth investing is it's very hard to forecast growth, and so people who try tend to overpay, even if they're somewhat right about growth, they pay such a high valuation multiple that they end up underperforming as a result.

Momentum, as we discussed, does forecast growth to an extent and seemingly doesn't overpay for it. So, it gives you that exposure to growth-like characteristics that--most importantly, the negative correlation to value--but has done it without underperforming the market. So, it's been very powerful.

Zimmerman: Very interesting, and I think a lot of folks who may be fundamental investors, but are aware somewhat of the momentum effect, might want to consider that as a strategy for at least a part of their portfolio.

One last question, there is the academic research, and now there's quite a big library of academic research around the momentum effect, very well documented, under laboratory conditions. But if you were to implement some of the strategies that are documented in your actual portfolio, the transaction costs would be huge. How do you manage that and the sleeves that are price-momentum-based in your GMO portfolios?

Hancock: I think huge is a bit of an overstatement. Momentum is a higher turnover strategy, but the momentum flavor we use at GMO and is typical for managers like us is more of a 12-month return. So, that will have a turnover of 100%-150%, a year, which is high but it's not awful. The other downside is higher turnover level you also, since you're chasing stocks that have gone up, you tend to pay higher commissions.

For a quantitative manager, one of the virtues is that we tend to hold lots of positions. Or maybe to make a virtue out of a vice, we tend to hold a lot of positions in our portfolio, which means any one name, our pressure on that stock will be relatively less. So, we tend to have lower trading costs, because we have lots of low-confidence ideas that we trade a little bit into, rather than a few favorite stocks that we're really chasing. It's definitely something you have to monitor. It's one of the reasons why at the margin you use 12-month price momentum rather than six-month price momentum, for example, is to try to find the sweet spot, but you do have to be willing to pay some of that turnover because the one way you can really get in trouble with momentum investing is holding on to stocks too long; a the momentum stock that went up in price has become expensive and doesn't have momentum anymore is not what you really want a lot of exposure to in your portfolio.

Zimmerman: Right. I think you made the point in the panel discussion, too, that it's a strategy that needs to be ruthlessly applied and very consistently applied.

Hancock: That's right. You have to be willing to do the uncomfortable thing of, essentially, in some cases, buying high and selling low. If you buy a momentum stock that went up and then sell it when you are down. And momentum investing got a bad name in the tech bubble, I think, not because it was wrong to buy tech stocks in 1999, but the people who are momentum investors but didn’t admit to it, didn’t sell them after the market turned. They rode it down not just in 2000, but in 2001 and 2002. You really have to be strict about your sell discipline, which is why I think a quantitative application of investing is a good, effective technique.

Zimmerman: I think John Montgomery would agree with you and might say that it's takes emotion out of the process.

Hancock: Yes, emotion can be the enemy.

Zimmerman: Absolutely, Tom, thank you very much for being on the panel and for being with us today.

Hancock: It's been my pleasure. Thank you, Shannon.