How to Profit Through Analyzing Investor Positioning in Sector ETFs

|
 |  Includes: SPY, XLB, XLE, XLF, XLI, XLP, XLU, XLY
by: ETFreplay

I have read in some places about how ETF investors, as a whole, tend to be poor timers of the markets. The odd thing is that given that ETF data is readily available and ETFs can easily be shorted – none of the authors of the articles suggested taking advantage of the cited studies. The great investor George Costanza once said, “if everything I do is wrong, then the opposite must be right” -- so let’s look at some data.

I have combined some of the US sector assets of "the big three:" (iShares, SPDR and Vanguard), and will be updating and furthering various ideas on this each month. My goal is to show how ETF investors are positioned versus the index weightings. These sector assets represent $46.2 billion in investor money (the % calculation uses the sum of the big 3 providers sector funds in the numerator -- and the total $46.2 billion asset figure as the denominator). I use the aggregate of the 3 to mitigate the fact that any one ETF provider may simply be losing or gaining market share versus another one in certain funds.

Click to enlarge

(Click to enlarge)

The table below summarizes the largest over and underweights:

Click to enlarge

(Click to enlarge)

Note that Technology is one segment that has a complication, as telecom services is included in S&P Technology Sector but Telecom Services is a separate sector for both Dow Jones indices (iShares) and MSCI (Vanguard). I have included the Telecom Services funds inside the technology sector, as this is how S&P does it and S&P is the standard as far as US sector funds go.

One thing that stands out looking over the data is that ETF investors seem to be pretty consistent on their over and underweights. Energy has been a very popular ETF sector for a long period of time – and it’s also been an underperformer for the past few years.

The consumer segments appear to be consistently unpopular areas for ETF investors (both discretionary and staples).

For one simulation --- I have created a sample look at a strategy involving this data. Simple idea: go equal weight SHORT the most popular sectors and equal-weight LONG the most unpopular sectors. This is not intended to be an actual portfolio – it’s just a simulation of how this strategy is doing this year:
Click to enlarge
(Click to enlarge)

Disclosure: No positions