Sunday Night Recap: Don’t Become a Victim of Regime Change

The first step in avoiding a trap is knowing of its existence –
  Thufir Hawat, Dune

While the above quote from Frank Herbert’s “Dune” may be one of the most used by investors and bloggers everywhere, actually following the advice is much more difficult. After all, what can you do to avoid the most common pitfalls the befoul investors when they’re in your own mind. In “Dune” Thufir Hawat is a mentat, a man whose nervous system has been adapted to processing and analyzing vast amounts of data. Relying on statistics and logic, he’s ultimately defeated by an inability to understand human sentiment and the depths an enemy will go to defeat you. Like the good mentat of the novel, you need to study your data carefully but remember that the human mind can struggle with data and often arrive at false conclusions.

One of the most power facets of human behavior in general is the need to “know” what the answer to any or every question is. Lazy journalists and their social media successors have relied on this for sales and ad revenue for years. The S&P 500 was down 3.5% last week and investors big and small are trying to determine why. Some will say a lack of confidence in Europe, others will say spillover from the energy sector is leading to profit taking elsewhere. It could be year-end asset reallocation or political concerns in Russia, or anything you want it to be. But the human mind is programmed for pattern recognitions; it was one of the great survival tools of our species, the ability to look at an event (say a lot of picked-over bones outside a dark tree line), find clues (large animal tracks), and draw a conclusion (something big is in there, stay away!) Today it persists in the need to look at an outcome such as a strong market drop and determine why or how long it can persist for.

As the computing power of our machines has begun to outstrip that we were born with, our ability to process data has increased exponentially, but if your data or process is flawed, so are your conclusions. As an example, I was updating my momentum tables this weekend and trying to come up with what I call “left field” trade ideas when I noticed that Energy Sector Select SPDR (XLE) was trading a level that I never thought possible. Using a short-term range of 5 to 30 days, XLE’s momentum is so low that round it up to the .01% is actually overstating its case. Compared to its prior history, it’s not just cheap it’s CHEAP. On a long term basis (1 to 5 years) I went out ten decimal points and still could find a number above zero. Saying XLE is hated would be understating the case.

Looking at the chart for XLE, the selling pressure hasn’t reached the level of last October but getting very close to it. Having taken out prior support at $75, you can see additional support at $72.50 and then $67.50 that might prove to be the final stopping out point. The question is who is brave enough to try buying there. Given the flood of negative sentiment, most people won’t even consider it. But remember how negative the sentiment was on the real estate sector in 2009? Everyone though they would collapse completely and instead over the next five years ending 12/12/14 they’ve outperformed with the Vanguard REIT ETF up an annualized 17.53% compared to 14.88% for SPY. Or how about Treasuries in late 2013? The iShares Barclays 20+ Year Treasury Bond (TLT) ETF is not up over 27% YTD, nearly 3x the gain of the S&P 500 TR.

What happened?  In a nutshell, there was regime change. At the heart of quantitative finance is a paradox; practitioners rely heavily on statistical theory and the assumption that data is stationary or that the relationship between two variables remains stable and unchanged over time. If variable X changes by so much, it will cause a certain change in variable y and do so consistently over the time frame in question. However, financial data is widely known and understood to be non-stationary; relationships between variables are in a constant state of flux and achieve stability for only limited periods. Despite the tools and techniques to model this instability, at its root, the market is a pool of humans (and increasingly machines) constantly updating their outlook based not only on new data but a change in sentiment which is something machine learning has yet to master. Why do I go into this now? Because I’m curious whether a regime change is about to happen to the S&P 500.

Regime change has already been taking its toll on the performance of active managers in 2014, just look at Good Harbor. The Good Harbor Tactical Core (GHUAX) Fund relies on the asset switching model they developed to rotate between equities and bonds and within equities between large and small cap ETF’s and uses leverage to enhance returns both by leveraging up longs or through inverse positions. During years like 2013 where equities consistently outperformed bonds, firms like Good Harbor can do very well for themselves and pull in a lot of assets, but 2014 has been hard on them. The constant rotation in leadership between equities and bonds in the first half of 2014 along with the consistent underperformance of small caps versus large has left GHUAX down 23% this year and GHUAX, while only a small part of GH’s overall business strategy, has seen assets drop nearly 40%.

GHUAX

The basic logic behind my own momentum model (or anyone’s really) is to identify levels where a security or index can be expected to out or underperform and to develop a trade strategy around it. Basically if short term momentum scores reach a certain low point, a positive return can be expected over a certain time frame. Looking at the S&P 500, momentum scores are close to the lows they set in early October and before St. Louis Fed President Bullard made his comments about changing the timing for ending QE3 that sparked a new rally. My first reaction was to say that equities should be due for a bounce soon that could push us back to the old highs but after creating a few charts my view has begun to evolve.

When I looked at my charts, I noticed that the short term scores haven’t been this low since early October of 2014 and before that February of 2014 but the more interesting discovery was that they really haven’t been this low, this often, since August of 2011 to May of 2012. It’s a document fact that momentum, like volatility, has a tendency to persist for long periods or put more simply, winners will keeping winning for much longer than you can expect. During the period of 8/11 to 5/12, the 30 days return after a hitting a deeply negative score just as often saw negative or at best breakeven performance compared to strongly positive performance in the period both before and after that. Likewise, very strong scores often saw positive performance rather than negative as short rallies cooled off. In other words, there was no consistent relationship between the momentum score and the subsequent return. A similar relationship existed for several months in May to August of 2010 and before that in the 1st quarter of 2009. After May of 2012 the situation changes; deeply negative short term scores saw very strong performance and typically only when scores were at their peak was there a minor pullback of less than 5%. Putting it simply, if you bought when momentum was strongest you might have suffered a small 1% to 3% loss over the next thirty days, but if you simply bought the dip, you made it all back and then some.

What was the likely culprit behind these regime changes? Well consider the only major outside variable, the main driver of equity market returns, over the last five years; the major sentiment boost provided by the Federal Reserve. Take a look at the charts below. QE1 was announced in late 2008 and ran from 12/08 to 03/10 and was followed by a period of weakness when rumors about QE2 began. QE2 ended with a serious market pullback and after a summer of rumors about QE3, the FED unveiled operation Twist II choosing to hold off on unlimited QE until the summer of 2012. After that the rest was history. Why didn’t the market break down further in 2013 when the FED got serious about ending QE3 in 2014? Because it was year-end and no one wanted to sell their positions after such a strong year. Now with heavy volatility and weak performance, investors can’t wait to sell their positions to lock in what gains they may have which could explain the weak performance across nearly all sectors and broad domestic equity markets last week.

QE1QE2

So why is it different now? Because the Fed has been signaling from the summer of 2013 that is very concerned about the effects of low interest rates on risk taking behavior and wants to return to traditional practices as quickly as possible and with the recent performance of the dollar and its impact on commodity price inflation, the Fed now has the cover it needs to wrap up their asset purchases without ever having hit the 2% inflation target. And now, Vice-Chair Stanley Fischer has confirmed that discussions are on-going about dropping the language committing to keep rates low for an extended period as soon as this week’s Fed meeting. Remember, the FOMC may believe raising rates will help increase velocity of money and raise spending because once rates begin to rise, investors will feel more confident that rates won’t rise more. Think about it this way; the pool of investors (individuals or institutions) who sit on cash because they won’t suffer a capital loss from duration if rates rise will feel more confident after the Fed Funds rate goes to .5% or 1% because they think a future rate increase is less likely. The Fed is hoping that the established positive correlation between interest rates and velocity holds true in the future.

Maybe the markets are right to be anxious about the coming end of QE3 after all. In the meantime, maybe it’s time to stop and think about what traps you might be setting for yourself.

One thought on “Sunday Night Recap: Don’t Become a Victim of Regime Change

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s