If you have account already, Log in here first

Log In
  • en
    • cs

Alternative approaches to short inefficient ETF – pair trading (1/3)

15 Mar 2018,

Dear readers, in today’s article I want to introduce you to a slightly different approach of trading ETF inefficiencies. So, let’s take a short break from the option constructions. First, I would like to follow p with the article in which I have described a very basic approach – direct ETF short. However, as I have mentioned in that article, it is (without strict money management) a relatively aggressive trading strategy, where the rapid movement of the underlying asset can be very unpleasant.

Today, I want to describe more conservative trading approach, and this is pairing of inefficient ETFs. What is it about? Its logic is basically very simple. As you already know from previous articles, we focus on markets – ETFs, which are characterized by inefficiencies – either those that have been declining for a long time or their mechanism of operation is inclinable to a “black swan” events and whose value can from day to day be reduced by higher tens of percent. Of course – in the case of leveraged ETF inefficiencies are increased by several times, so you will probably not be too surprised that we will meet them in our selection again.

Unlike the classic pair trading, when we buy one title and sell the other with the same value, we are going to sell both titles of the inefficient ETFs.


Trading itself is quite simple.

  • First of all, we have to choose a suitable pair – the ideal one is from the same “stable”, ie from the same ETF manager. One ETF should follow the underlying market as LONG, while the second ETF should follow the same underlying market as SHORT (the description of the ETF is typically Bull x Bear or Long x Short)
  • We will open the same – meaning nominal value – short position on both ETFs. The same nominal value means that if one ETF costs for example 10 USD and the other 20 USD, then to have the same nominal value, let’s say $ 10,000, I must short 1000 pieces of the first one and 500 pieces of the second ETF.
  • Surely you will wonder what to do when the price changes. With the changing price, of course, the value of the nominal is changing, but we need to have it more less the same, so that our position is balanced. Consider, for example, a situation where after a significant movement – from the initial $10,000 short of both ETFs, the nominal of the first one will be 14,000 USD, while the other $ 6,000. So, it is clear that we have a very unbalanced position. It is therefore necessary to rebalance the trade – ie to equal the nominal on both sides. Typically, this is done by another shortening of an adequate number of the second ETF or by closing part of the first ETF trade, or by the combination of both. The goal is clear – equal the nominal value to the same level.
  • Regarding the closing of the trade – here I use 2 exit methods:-Exit at predefined profit against the nominal of the trade-Time stop loss (typically up to 1 year)

Selection of suitable ETFs for the pair shorting

Basically, I use the following criteria for the selection:

  • borrowing fee – the cost for borrowed shares, simply the cost of holding a short position. I have already described that it is very important to know the exact value when it comes to a short of ETF or shares, because this can very much influence the performance of the strategy, especially in the longer term. Ideally, this cost should not exceed 5% per annum in the case of ETF.
  • Volatility – the volatility of underlying instruments – the higher is the volatility, the better for us. We simply do not want the underlying market that “goes in one direction”, we need one that is volatile, and ideally goes for a longer time to the side.
  • Liquidity – respectively the number of available shares for short – for some instruments this may be a problem, the daily minimum volume should be at least 1 million.


As you surely think, ETFs linked to volatility (VXX, UVXY, SVXY, …) that I have described in the previous chapters are out of this candidate selection. The borrowing fee is very high here. That is why we are trading these volatile ETFs not directly through the short ETFs, but by opting strategies speculating on the decline in the market.

On the contrary, as a potentially interesting candidate, the leveraged ETF (3x) appears to fill the volatility criteria. But there is a need to pay attention to the cost and short-term stock availability (which is, in essence, are very closely related).

You can find a list of interesting leveraged ETFs (3x) at ETFdb.com.

As examples of potential candidates who meet the liquidity criteria, I state the following:

  • FAS and FAZ
  • TNA and TZA
  • NUGT and DUST
  • ERX and ERY
  • UGAZ and DGAZ
  • TQQQ and SQQQ
  • SPXL and SPXS
  • JNUG and JDST
  • GUSH and DRIP
  • UWT and DWT

and a few more examples.


In any case – as you can see in the next part of this series, it is still necessary to carry out a thorough analysis of the shorting cost and volatility of the underlying instrument for the final selection.

At the end – as a small example of an interesting pair, the chart NUGT (marked by candle chart) and DUST (marked by blue line). Note that there has been some discrepancy within a relatively short time (22.75% vs. 30.30%). But more about it in the next part.


Obr.1: Graf vývoje NUGT a DUST od 1.1.2018

Figure 1: Development graph NUGT and DUST since 1.1.2018 (source: finance.yahoo.com)





Option report

Get your report from Jan Kaska's trading every week!


Leave a Reply

Your email address will not be published. Required fields are marked *