ETF Country Plus Strategy Insights: Crossing the Zero Divide (Part 1)

James Kimball | July 12, 2015

Both ETF Country models lagged their benchmarks by a significant margin this week. The major freefall in Asian equities did us no favors, which was the main contributor to the underperformance this week. However, with Monday, we look to have a couple of fresh trades on.

The U.S. markets had tumultuous week, with large negative and positive gaps both being attributed to all sorts of news out of Greece and China. Chinese markets appeared to have brought a halt to the recent selloff, while in Europe, the Greek drama continues on, though it appears likely we will see some form of a resolution this weekend. Markets appear to be holding support near the lower end of their recent trading ranges.

We had a number of position changes this week (the article below attempts to address why we are seeing this). And we have a couple more to start next week. The model will be selling EFZ and buying EWQ (France) and EWL (Switzerland).  We recommend that you log into the website to see all the current holdings and pending changes for Monday.


This Week’s Strategy Lesson: Crossing the Zero Divide (Part 1)


We have seen a recent increase in the number of trades and rotations in our ETF models. This is being driven by a couple factors. The broader markets have been in a wide-sideways range for months now. They have tried to breakout up and repeatedly failed as they have repeatedly come down to support and held. When trends fail to breakout it can lead to some higher than average turnover, but there is one other thing coming into play: the average TSI score in our model universes has gone negative.

Rules and Effect

One of the rules of the ETF models is that we will not hold an ETF with a negative TSI score. Most of the time, particularly in bull markets, this little rule will have no effect, as there will be plenty of ETFs with positive scores and the “leaders” should be fairly obvious. But during market corrections and bear markets, this rule becomes much more significant – particularly during the transitional period where our “leaders” and holdings might cross this threshold.

We don’t spend much time or focus on the average TSI score of the model universes. In many respects, this number should be reasonably correlated with the TSI on the S&P 500 index (Index performance is roughly the average of all the sectors and industries represented). However, it may be useful in helping us gauge the overall state of the markets and when we might expect to start seeing a higher level of turnover.


The chart above tracks the average TSI score of the investible universe (18 ETFs) of the Sector Conservative/Moderate model. You have to go back to 2011 to see a period where we spent significant amount below the zero line. (Not entirely coincidentally, the 2010 and 2011 corrections were also related to Greek and European debt issues, among other things).

The point when the models change their long-short bias will not always coincide with the average going negative, as the actual alternative and inverse ETFs we use to provide us this exposure are not all equally correlated to the overall market.

Alternative and Short ETF Options

Each model has access to a variety of alternative and inverse ETFs. The Sector models have a short S&P500 ETF, the unlevered “SH” or levered “SDS” depending on which model, and likewise, the unlevered “TLT” and levered “TMF” for treasuries, an instrument that often tends to provide timely protection against market corrections. In addition, energy and commodity related ETFs can often have counter cyclical movements.

In the Country model, in addition to the same exposure to U.S. treasuries, they also have access to EFZ, an ETF that tracks the inverse of an index that tracks the developed markets of the world excluding the U.S. and Canada.

The Global Macro model has several additional inverse and alternative ETFs including precious metals (SLV and GLD), the volatility index (VXX), and one that tracks the inverse of the emerging markets (EUM).

So depending on the model, our current holdings, and the relative TSI of the alternative and short-biased ETFs, transitional periods could be rather smooth or it could cause a lot of friction in our holdings.

Best case scenario for the models would be a clean bounce off these market lows and onto new highs. However, if the markets continue to chop around at these exact levels, we could see a number of our holdings continually skirt the zero TSI line.

Over the next couple of weeks, we are going to dig a little deeper into the historical archives of the models to see how they behaved during some of these transitional periods in the past and see what we can learn that might be useful going forward.

The Current Condition of the Model

Our three positions are EWJ, EFZ, and CASH.  On Monday, both models will move to EWJ, EWQ, and EWL.

Stay tuned to daily updates for any position changes.

Here is a summary of the weekly performance of all the ETFs that the strategy monitors:


Best wishes for your trading,

James Kimball
Trader & Analyst