Signs and Portents

Category: Note, Narrative, Game Theory

epsilon-theory-signs-and-portents-december-1-2014-the-omen

Young nanny: Look at me, Damien! It’s all for you.  

[she jumps off a roof, hanging herself] 
– “The Omen” (1976)  

epsilon-theory-signs-and-portents-december-1-2014-bag-of-bones

When one has little faith, one must survive from day to day signs.

– Stephen King, “Bag of Bones” (1998)

epsilon-theory-signs-and-portents-december-1-2014-batman

Criminals are a superstitious cowardly lot, so my disguise must be able to strike terror into their hearts. I must be a creature of the night, black,  terrible … a … a …  


– Bob Kane and Bill Finger, “Batman” (1939)

When clouds appear, wise men put on their cloaks;

When great leaves fall, the winter is at hand;
When the sun sets, who doth not look for night?

― William Shakespeare, “Richard II” (1595)

Alas, why gnaw you so your nether lip?

Some bloody passion shakes your very frame:

These are portents; but yet I hope, I hope,

They do not point at me.

― William Shakespeare, “Othello” (1603)

Destiny does not send us heralds. She is too wise or too cruel for that.

― Oscar Wilde (1854 – 1900)

Like the criminals that Bruce Wayne fought as Batman, we investors are a superstitious, cowardly lot. We are constantly ascribing way too much import to this sign or that sign, constantly freaking out over the meaning and significance of this market event or that market event. It doesn’t help that the financial media world has devolved into fiefdoms of rah-rah soothsayers on the one hand and doom-seeing end-timers on the other, so that whatever our predispositions might be we can easily find Voices of Authority to read the entrails to our liking. And it really doesn’t help that we are in the midst of the greatest crisis of faith in the markets since the 1930’s, so that – as Stephen King wrote – we survive by looking for day-to-day signs to show us what to do.

And yet sometimes a little freaking out over the signs and portents is clearly the right thing to do. Sure, if your nanny declares her loyalty to your adopted-under-mysterious-circumstances devil-child as she hangs herself outside the nursery window it’s probably a case of mental illness, but I’d also listen a little more closely to what that pesky priest says. If you’re Pierce Brosnan in the “Bag of Bones” mini-series and you think that your dead wife is sending you cryptic messages via a handful of refrigerator magnets … well, maybe you should drive into town and buy more refrigerator magnets, see if she’s got anything interesting to say. If you’re Desdemona and you’re worried that Othello’s lip-biting is a sign that he’s about to fall into a jealous, murderous rage … well, maybe you should run out of the room instead of hanging around to see if you’re right.

It’s a tough call, evaluating what’s a “true” sign and what’s a “false” sign. Are we being foolish to sell our energy stocks after oil prices took another big hit, or are we reading the market’s tea leaves correctly and saving ourselves a lot of future pain? Are we acting as Shakespeare says any wise person would in a knowable and deterministic world, by putting on our cloaks as clouds appear and looking for the night as the sun sets? Or are we mistaking our play-acting market world for the real world, putting on our cloaks as the projectionist shows us a picture of clouds and looking for the night as the stage lights dim?

Here’s the Epsilon Theory answer: the latter mistake is 1,000 times more common than the former wisdom, and the vast majority of investors would be better off if they never read the newspaper and never turned on the TV. Why? Because what they think is a “sign” is actually a signal, neither true nor false in and of itself but only more or less influential in changing their mind and other investors’ minds about the world. (for more on signals and Information Theory, see “Through the Looking Glass” and “The Music of the Spheres”) Signals are constructed. Signals are malleable. And unless you are focused on how and why signals are constructed and shaped, you will be whipsawed. You will be shaken out. You will be roped in. You will catch a falling knife. Pick your own analogy or metaphor … there are a million to choose from and anyone who has spent any time at all in the market has experienced most of them. We’ve all been there.

Case in point: why are many investors puking energy sector stocks today? It’s not because they have a detailed cash flow model of the specific companies they’re selling and have calculated the incremental earnings impact of oil prices moving from a $70 handle to a $60 handle. It’s also not because there’s some credit freeze roiling financial markets and a careful balance sheet analysis shows imminent dividend cuts or debt stress throughout the sector. Will lower oil prices over a long period of time hurt earnings and crimp growth for the entire sector? Well, sure. That’s kinda what it means to invest in a cyclical stock, and if this comes as a surprise to you then I really don’t know what to say. Will lower oil prices over a long period of time create balance sheet distress in the energy sector’s more levered, go-go stocks? Absolutely. If you’re not stress testing the balance sheet, capital allocation, and distribution coverage models of the energy stocks you own, then you’re not doing your job as a risk manager. But neither earnings risk nor balance sheet risk explains why you see a spasm of energy sector selling today or back in October.

No, the selling is because the dominant Common Knowledge regarding energy sector stocks is that they move up and down with the price of oil. Common Knowledge is not what everyone knows; that’s the consensus. Common Knowledge is what everyone knows that everyone knows, and it’s the driving force behind the Game of Sentiment. Everyone knows that everyone knows energy stocks are tied to oil prices, we just took another sharp leg down in oil prices, and so energy stocks must be sold. The fact that energy stocks are down “proves” the relationship (a wonderful example of Soros’s concept of reflexivity), which adds to the selling. And “Even After Selloff, Energy Stocks Attract Few Buyers” because, as the WSJ breathlessly announces, “prices could soon plumb new depths.” Or not, but … hey, all the better to set-up that “rebound that no one was expecting” story. Until that story is written, any oil price increases are merely because “traders who had bet on lower prices locked in gains.” I find it awfully telling that this WSJ article now titled “U.S. Oil Prices Trade Higher After Selloff” was originally titled and archived as “Oil Slides as Market Struggles To Get Grip” (you can track URL’s to identify this stuff), but then the market failed to cooperate and they had to change the title!

A couple of Epsilon Theory points on all this.

The reality (not that it matters) is that energy stocks are barely correlated with the price of oil, and their correlation with each other is barely driven by oil prices. We’ve run some basic regression analyses on MLP portfolios, and since 2012 only about 7% of the total return profile of MLP’s can be “explained” (statistically speaking) by change in oil prices. The largest explanatory factor is just the S&P 500, with about 4 times the “power” of oil prices to predict MLP prices. My interpretation is not that a rising overall market is “causing” MLP stocks to work, but that the same non-fundamental monetary policy-driven forces that are driving up the overall market are also at work in the MLP space. MLP’s have both growth and yield – the two rarest things in a Fed-dominated world – so whatever market dynamics work for stocks overall have really worked for MLP’s.

For another perspective, take a look at this recent piece by Ed Tom and the Credit Suisse Equity Trading Strategy team, titled “What’s Driving Energy Sector Correlation? (Hint: It’s NOT Oil)”. Ed and his team do stellar econometric analysis of equity market derivative contracts, which means that their papers typically need some translation into plain English. Here’s the skinny: most investors think that energy stocks traded off in unison in October because they’re highly correlated to the decline in oil prices. Not true. Yes, there’s some correlation to oil, but what’s really driving this across-the-board decline is the fact that “long energy” has become a very crowded trade. So if you get a signal that spooks the long energy crowd, you’re going to get a mad rush of investors heading for the exit even if the signal isn’t truly that relevant for the fundamentals or the historical beta of energy stocks. When a trade is crowded on the long side, everyone has an itchy trigger finger to sell.

So what does matter? How can we improve our investing around energy sector stocks by thinking about oil prices as a malleable signal that drives sentiment dynamics (at least in the short and medium term) rather than as a deterministic and inexorable sign of things to come? I think what happens from here depends on the strategic interaction of four factors:

  1. How crowded is the trade (still)? The good news here is that the Credit Suisse team believes a lot of the air was let out of the long-energy crowded trade balloon in October. I think that’s probably true, although I certainly wouldn’t call it un-crowded. I also think there’s a tremendous amount of air in the more general “the Fed has got your back” crowded trade balloon, which is worrisome for all equity market sectors, including energy.
  2. What’s the investing DNA – value or growth – of the majority of energy sector holders? The notion of population dynamics and evolutionary theory is something I explored earlier this year in Epsilon Theory (here and here), and it’s a topic that I’m going to refocus on in 2015. The basic idea is that different investors have different linguistic grammars (value investors possess a mean-reversion grammar, while growth investors possess a momentum grammar), and that for a stock or sector to “work” you need the dominant Narrative grammar to fit the dominant investor type.
  3. How is the oil price Narrative framed – supply/demand fundamentals or monetary policy? I wrote about this at length in last week’s Epsilon Theory note, so won’t repeat all that here. Everything I wrote then remains true: the supply/demand fundamentals Narrative is now ascendant and I suspect will remain so for at least a couple of weeks, maybe longer. That’s important because at current supply/demand projections it’s hard (not impossible, but hard) for oil prices to get much below $70 and stay there for a long time if you believe in this Narrative. A fundamentals-driven “explanation” places a martingale on oil prices that does not exist with monetary policy-driven “explanations”.
  4. What will China and the US do with their monetary policy, and what will Saudi Arabia and Russia do with their foreign policy? Hey, your guess is as good as mine. I don’t have a crystal ball on outcomes or timing, but this is where my risk antennae are focused 99% of the time.

I don’t have settled answers to any of these questions. And I don’t have a predictive model (a risk-based econometric analysis), because not only don’t I have settled answers, but I don’t even have a sense of potential outcomes or rough probability distributions for #4. I know that’s unsatisfying to many readers (certainly it’s unsatisfying to me!), but you have to take what the market gives you, not what you wish were there. My goal is not to be a hero and make bold predictions in the Golden Age of the Central Banker. My goal is to be a survivor. My goal is to play the game a bit better than the crowd by paying attention to the construction and shaping of market signals, all the while keeping my attention focused on how politicians and bankers wrestle with a global debt crisis. Call it being reactive if you like. I prefer to call it Adaptive Investing, and that’s what I want to communicate with Epsilon Theory.

epsilon-theory-signs-and-portents-december-1-2014.pdf (161KB)