The Federal Reserve's impact on stock markets has been a well-documented phenomenon for the last several years. Indeed, strip the performance of the S&P 500 on the days the Federal Reserve makes an announcement and the results look markedly different (For a great site that studies the Fed's impact on equities over various time frames, you absolutely must check out Rob Hanna's Quantifiable Edges site here). It got me to wondering, if there is a seeming edge over one day, does this edge improve with a longer holding period? Or more precisely, how long does the Fed effect last on short-term equity moves? The data seems to suggest that whatever "sugar high" the Fed injects into markets goes away almost immediately, the very next trading day in most cases.
The above chart is piecing together a performance history exclusively for successive holding periods on each Federal Reserve announcement date going back to March of 2004. The idea is to initiate a long position in (SPY) on the close the day before the Federal Reserve makes an announcement and closing the position at the close on the day after the Fed announcement. I then looked at variations of that up to a holding period of 5 days. (I am really a short-term trader, I'm not looking for longer-term implications of Fed policy or how equities respond over longer time-frames…there are too many variables that I can't account for at least). When I did this, I noticed that essentially all of the upside edge in U.S. equities occurred on the very day of the announcement. Now, I wouldn't consider this very groundbreaking stuff, this phenomenon has been well documented as I have previously mentioned.
What I found most interesting is that the edge has consistently given the majority of its gains away the very next trading day in the preponderance of instances. You can see that the line that holds SPY for both trading days significantly under-performs the position that holds just for the day of the announcement and exits on the close of that day. So I thought, why not short SPY on that second day, the day after the announcement? The data seems to suggest this provides an additional edge which appears to be as consistent in the Post-Lehman period as the edge on buying just on Fed days does. (I did test on other holding period combinations up to 5 days and found the aforementioned approach far and away provides the best risk-adjusted outcome).
What I try to do is step back from the data and think about what, if anything, is being implied from the data? There are two implications I find on the very short-term as it relates to Fed announcement days. The first is obviously to not fight the Fed. I think this is something even the layman investor can understand. It seems that, on the date of the announcement at least, Ben Bernanke has successfully conditioned markets to heed his beck and call, much like Pavlov's dog. However, this also relates directly to my second observation, namely, that this Pavlovian response from markets on the day of a Fed announcement is providing opportunity on the short side for those so ambitiously inclined to step out in front of the great Bernanke. Chalk it up to herd mentality I suppose. When enough people start buying into an idea, especially one as obvious as the perception of the Fed driving up equity markets, it's usually a good time to start fading (at least in the short-term).
I make no claims to be an expert of understanding the dynamics of the Federal Reserve and how it can levitate equity markets above levels more reasonably associated with the real economy, but I did find this an interesting study.