Evaluating Designer Models: Momentum

PDF Version: Choosing Portfolio123 Designer Models Topic B4 – Momentum

Momentum is the factor (style) that in the eyes of many defines Wall Street. It’s where the action is.

• It’s what Hollywood directors have in mind, although they don’t consciously know it, when they stage frantic trading-floor scenes with lots of shouting and gesturing.
• It’s what many internet finance site have in mind, although they don’t know it, when they design web sites that place the hottest most current news stories on top, where its easiest for you to see them (and make it almost impossible for you to find that really crucial story from a month and a half ago that could really help you, if only it hadn’t scrolled its way into oblivion within an hour after it was initially posted).
• It’s why stock charts are so popular.
• It’s what technical analysis is all about.
• It’s why we fuss so much about estimate revision, upgrades, downgrades, guidance and surprises.
• It’s why folks care about “new highs.”
• It’s why so many say you should “sell your losers and ride your winners.”
• It’s an aspect of “noise” (remember early on we covered p=v + n; price=value + noise).

But it’s about more than just action. As noted in a previous document, to say something is a for-real factor or style, I’m also saying there is empirical support for its use in helping investors distinguish among stocks having good or bad future potential. For as many as get this and act on it, there are at least as many who are repulsed by this notion.

• Buy high to sell higher! That’s reckless.
• Past performance doesn’t assure future outcomes (Where have we heard that before?)

So how can we resolve what we see (that it works) with what we think we now (that it shouldn’t work)?

Defining Momentum

Notice how easy it is to say things about momentum before actually defining it. That’s because as soon as the word is uttered, everybody has a gut sense of what it means. But that’s a problem. While there are many ways to define and momentum, many such approaches are flawed, with some of the most intuitive popular definitions being the most financially dangerous.

Momentum starts with the notion that what happened in the past will, in fact, be repeated into the future.

It appears to work like this:

1. Stock A rose very briskly in the past.
2. Because Stock A was strong in the past, we can say it will be strong in the future.

That is dead wrong. Now we remember where we heard about past performance not determine the future. The lawyers make everybody say that. And they do it for good reason; it’s absolutely true.

Momentum actually works like this:

1. Stock A rose very briskly in the past.
2. Stock A rose very briskly in the past because of such and such reason (for example, because earnings rose briskly – but I don’t actually know what the reasons were, I just know they existed)
3. Changes in the business and financial worlds tend to be evolutionary rather than revolutionary
4. Because change is slow, there is a high probability that the reasons why Stock A rose briskly in the past will persist into the future.
5. We can assume those reasons are, in fact, still valid (i.e. that the company will continue to post impressive earnings gains).
6. Because the factors that induced strong performance in the past are still present, Stock A will be strong in the future.

Both logical sequences begin and end in the same place:

• Begin: Stock A was strong in the past
• End: Stock A will be strong in the future.

But the paths that get us from beginning to end are very different and this difference has a dramatic effect on how we apply momentum to our investing portfolios. As noted, the first two-step version is dangerously wrong and can easily blow up your portfolio. You need the full six-step path, and whether you succeed or fail will depend on whether step five is accurate.

Reasons to Believe

Momentum investing will succeed or fail depending on whether the factors that caused the stock to perform as it did continue into the future.

It’s just that simple

Don’t you just love it when a topic can be completely explained in a single sentence!

But it get’s even better: Empirically we can see that there is a long-term upward bias to the economy and to the stock market (based on population growth and technological progress leading to better living standards and better business conditions which leads to better profits which leads to higher stock prices). Contrary examples are aberrations, adjustments needed to correct periodic and sometimes extreme imbalances in supply and demand. Since the future does often turn out to be like the past, momentum often works.

Now, we understand why momentum is so beloved. You can run up a lot of great numbers and make good money, and you don’t even have to bother with Dividend Discount Model and all the other fundamentals that stem from it; momentum investors are, in essence, riding piggyback on the stock market’s mega-trend and on the shoulders of others who are doing the real work of stock analysis. Often momentum investors could care less about what is actually being done in the real-work phase that drives the stocks upward; the free ride they get is more than ample.

It gets even better if someone is into data mining (building models that successfully use data to pinpoint stocks that did well in the past). Since the past persists more often than not, anything that leads you to past winners can often work in the future.

Now, having provided all that good news, it’s time for the other side . . .

It’s Just That Complicated

In real life, things do change and when they do, momentum investors can and often do experience a lot of pain.

• The best-case momentum scenario when things change is neutral; a model that used to capture stocks benefitting from certain fundamentals is now capturing stocks based on nothing of substance, meaning the list has become a random grab bag.
• The worst-case momentum scenario is when the is negative; a model that used to capture stocks benefitting from certain fundamentals is now capturing stocks getting pummeled by reversals in those same fundamental considerations.

The worst example of the latter scenario may be in fixed income, especially long-term fixed income. That group has benefitted magnificently from a 30-plus year mega-trend of falling interest rates. Momentum there has been a stupendous positive. But if rates start to rise, those same securities have to fall in value, and the longest-term fixed income securities have to get hit harder than anything else.

Making Momentum Work

While nobody can ever assure the outcome of anything, there are things we know we can do to enhance the probability of momentum strategy being used in a way that helps us. Since we know that the success of momentum is tied to the past resembling the future, it stands to reason that the probabilities will move in our favor if we support momentum with factors that shed positive light on the likely sustainability (persistence) of the factors that have thus far been driving share performance.

That brings us full circle right back to basic fundamental analysis, the dividend discount model and all the proxies and approximations used in the real world to identify stocks more likely than not to be priced reasonably relative to their ideal DDM valuations. The difference between Momentum and conventional fundamental-analysis modeling is that we’re not simply specifying fundamental factors but are supplementing them with additional (momentum) factors that provide clues suggesting others in the investment community are, on balance, reacting well to what’s happening.

So obviously, we never actually need to use momentum. It’s just another way of getting at the same considerations we’d otherwise be exploring. Still, momentum can be a very valuable add-on in that it might cue us into investor reactions to unquantifiable analytic factors.

That’s no small thing. Fundamental analysis is necessarily limited to the past. Momentum, capturing the human element, can give us clues to the effect that many others see the trends as sustainable.

So the strategy designers’ tasks are twofold:

1. Define momentum in ways that make it more likely than not that the model is truly capturing investment community assessments of things that have staying power. Hollywood, with its emphasis on cinematography, shows us that being in the moment (reacting quickly to news as it comes out) is the way to do this. Academicians responding to data and study-results as opposed critics’ reviews and box office trends, come up with much different story lines. On the whole, they favor momentum established over intermediate and longer periods as well as not-so-frenetic model refreshing.
2. Accompany momentum items in the model with other items that come directly from fundamentals. Its one thing to infer that strong momentum is associated with favorable quality- and value-factors. It’s another to model it directly.

Measuring Momentum

Our measurement of Momentum for purposes of Designer Models categorization is a two-fold approach.

1. We measure it directly through price-based data. We look at the relationship between the stock price and intermediate-term high, we look at moving averages (again, intermediate term), and we consider pure price momentum.
2. We also consider analyst sentiment, revisions and surprises. This is an indirect measure of momentum; one might even say it’s a catalyst for momentum. It’s useful, though, because it incorporates the impact of judgments regarding things that cannot necessarily be quantified. Many might regard such items as aspects of a different style altogether (e.g., Sentiment). But when viewed this way, Sentiment and Momentum tend to be highly correlated. Seeing, as we do, that Sentiment and pure price momentum point to the same thing, we chose to add the extra measures since they are able to make a unique (qualitative-based) contribution to the picture.

The Other Side of the Momentum Coin

As with the other styles, Momentum is not a singular thing but one end of a continuum. Its opposite is Contrarian, an approach that invokes every bit as much excitement as Momentum.

Running ahead (or more realistically, alongside or even a bit behind) the crowd feels exhilarating. Standing firm against the crowd also gives investors a pretty good adrenalin rush. Few things in life evoke more excitement than the notion of the romantic rebel.

Those who choose this approach have plenty of cheerleaders egging them on. The Wall Street herd is ignorant. Analysts are corrupt. Mr. Market (the mythical being created by Ben Graham and popularized by Warren Buffett) is manic-depressive. You’ve heard it all. You know the drill. It’s an aspect of fundamental analysis. It’s also an aspect of technical analysis (oscillating indicators, buying oversold stocks, etc.).

As with Momentum, though, the reality can be much more an exercise in mundane grunt work. Momentum works when you support it with indications that favorable fundamental trends are likely to persist. Contrarian strategies work when supported by indications that unfavorable trends are likely to reverse.

In both cases, it’s the fundamentals that matter. And in both cases, practitioners need to put emotions and stereotypes aside and work with day-to-day reality.

The reality is that being a contrarian today is much, much harder than ever before. That’s because it’s next to impossible today to argue that the market is wrong because “they” don’t know. Essential facts are embarrassingly easy to accumulate nowadays and can be done the low, low price of zero.  So if you see a stock that’s depressed and you think the market has it wrong, your first reaction should be, not to buy, but to refresh your own analysis. When money is on the line, there’s no role for ego (“I’m right, the world is wrong”).

This is not to say the world is always right and that you’re always wrong to disagree. Anything is possible. And at times, you and the world may agree on facts but differ in interpretation or in terms of investment goals. That’s fine. But that’s much different from a knee-jerk rejection of consensus views. It’s OK to disagree with the market – but only after you roll up your sleeves, do the grunt work, and understand how and why the disagreement exists and assess it in a thoughtful way. There was once a day when information was hard to come by and one could readily assume that you have simply outworked the ignorant rabble. Those days are gone. Nowadays, you can’t make a nickel by outworking Mr. Market. You have to work smarter and that brings us back to the core of contrarian investing; the willingness and ability to articulate criteria that give you objective reason to believe trends will reverse.

The Strategy Designer’s Role and the Subscriber’s Role

A lot of what I said addresses what strategy designers should be doing. Obviously, one can’t expect a Designer Models model subscriber to replicate that. Instead, Subscribers have their own tasks.

1. Since you aren’t going to/can’t analyze the stocks in the portfolio, you can and should at least do that for the broader investment environment. Step one is to understand the market and develop an assumption regarding whether trends in general are more likely to persist or reverse. Having done that, your next order of business is to identify models with compatible Momentum style scores (high if you expect economic and market trends to persist, less so or the opposite if you expect them to reverse).
2. If you want or are open to high Momentum scores, your next task is to pick and choose among models with appropriate (as defined by you) Momentum ranks. If you favor high momentum, it’s reasonable to favor models with strong live-money performance during good times; these are likely the ones by designers who have tapped correctly into what Momentum is about. (NOTE: It is absolutely critical; that backtest or simulation results play no role whatsoever in this decision; it’s too easy for those with strong empirical skills to define models that identified known winners from the past. The question is whether their algorithms can identify unknown-ahead-of-time future winners.)
3. If your inclinations are more neutral or contrarian, you may face a tougher challenge. There may not be sufficient opportunity to observe these strategies through live-money reversal periods. This is where you may have to downplay what you see in model performance (even with respect to live-money performance, which can be expected to lag during times that favor momentum). You may, instead, wish to pay more attention to verbal discussions by designers, either in posted model descriptions, in the community, or in response to inquiries from you. Look for indication that the designer is in tune with the needs of success in the contrarian world, a willingness of the designer to discuss how they are going about seeking clues that the stock are better then the market assumes. High rankings in terms of Quality scores and Value scores that are no worse than neutral will likely help since many things they can do in this regard will be show up that way.

Moving On

We’ve now covered the three Designer Models styles. Next, we’ll move on to Size, which may or may not be a style/factor based solely on rhetorical preferences.