Market Cap Arbitrage: SPY vs IWM

The spread between domestic large cap stocks ($SPY) and small cap stocks ($IWM) has reached a statistically significant level.  These ETFs have an economic and statistical correlation which bounds their relative performance and there is good evidence to suggest they mean revert from extreme circumstances. Over the few months the large cap $SPY has greatly outperformed the small cap $IWM. It may be time to take a contrarian view and anticipate this spread to revert to the mean.

There are a couple ways to play this relationship. Most simply, by initiating a short position in $SPY and a long in $IWM, your trade will be anticipating of the spread reverting to the mean and generating a profit. For those with a positive outlook on the market as a whole, this setup might indicate that you may want to take some profits in large cap positions and cycle some of the capital into smaller cap companies looking for the IWM to play catch-up.

Here’s the daily chart with green shading showing when the $SPY and $IWM traded 2-standard deviations from the 1-year mean.


This spread can mean revert back to the mean in several ways: large caps ($SPY) can correct while small caps ($IWM) just moderately underperform; small caps ($IWM) can outperform, while large caps ($SPY) consolidate — or some combination of the two. Keep in mind that regardless of how statistically unlikely, macro trades like this can continue to push further away from their mean territory for many reasons, so keep that in mind when planning your timing and position sizing.

Below, we’ve included the statistics for typical 20-day trades. As you can see if you had initiated a short in the $SPY and long in the $IWM each time this relationship stretched to this degree you have a winning trade 8 of 10 times.

Below is a chart of the relative spread $SPY/$IWM with respective Alpha Curves and statistical table via



Disclaimer: Nothing on this site should ever be considered to be advice, research or an invitation to buy or sell any securities, please click here for a full disclaimer.

blog comments powered by Disqus
Dynamichedge Blog