The S&P 500 (SPY) had a sharp pullback in the middle of October mostly on market fears that the end of the Federal Reserve’s quantitative easing program would dry up liquidity and cause stocks to fall. There was good reason to fear this; SPY has fallen at the end of every previous QE program, and few arguments could be made for why this wouldn’t happen again. As a result, the markets turned ugly, bottoming 7.5% down from their peak at 2011:
Click on image to zoom
The sharp reversal was thanks to one event only: the Bank of Japan surprised markets by announcing it would raise its target for monetary expansion to 80 trillion yen ($720 billion), up about 23% from previous expectations.
Shares have steadily risen since then, even as headlines about Japan’s monetary policy have faded. Earnings season has so far been strong. A total of 461 SPY companies have reported, with 6.9% year-over-year earnings growth and 4.1% year-over-year revenue growth. Over half have beaten revenue estimates, and over 70% have beaten earnings. The TTM P/E ratio for SPY remains dangerously close to 20, a psychologically significant point and a ceiling that has not been broken since January 2010, which was a significant factor in stocks’ mid-October decline. The P/E ratio now is between 19 and 19.8, depending on your data source, up from below 19 that we saw last month.
However, the forward P/E ratio is substantially lower. According to one estimate (Birinyi Associates), the forward SPY P/E ratio is 16.65, and that estimate has fallen substantially from estimates before the earnings season, if adjusted to current price levels. If that forward estimate is accurate, SPY is still far from the psychologically significant 20 barrier, and will probably stay below that barrier if there are no significant price spikes before the next earnings season.
If this steady bull trend is priced into the price of SPY earnings, how can we account for the dip in October? Fears of a fall in liquidity account for most of it, but fears of a slowing U.S. economy aggravated the issue. At the time, many market participants pointed to oil, which was on a steady slide downwards. WTI futures were falling to the $80 point, where many analysts suggested oil production would become unprofitable.
Click on image to zoom
However, the slide in equities and oil prices diverged in the middle of October, and they are now reversed, suggesting that the correlation between the two was spurious. Oil has fallen 6.2% since stocks were at their lowest, while SPY has risen 9%.
The narrative about oil has changed; now cheap oil is seen as a tailwind to aggregate consumer demand in the U.S., meaning more people will spend more with the money they saved from cheaper energy costs.
The transition from one viewpoint on oil’s slide to another can be quantified; it’s in the VIX. This volatility index is a good indicator of how the market is feeling—whether a new trend is perceived as a headwind or tailwind to SPY. The fears of valuations, the worries of cheap oil producing deflation, and concerns about the end of QE converged to cause the VIX to spike over 26 in the middle of October:
Click on image to zoom
The VIX reverted to the mean (it’s below 14 now). It always does. The problem is that sometimes the VIX takes longer to revert to its mean than other times. In late 2008, the VIX didn’t revert to the mean until over a year later. This time, it took a couple weeks.
Another problem with the VIX is that its mean changes. Its 30-day SMA has been falling since that spike in 2008, and is now in the mid 13 range; last year it was on a steady slide down from 15.4, where it started in 2013. And that was considered historically low—which it was:
Click on image to zoom
This is a headache for anyone who wants to sell volatility. On the one hand, the trade is enticing because of the downward slope; on the other hand, spikes can make a short on volatility extremely dangerous to a portfolio. Here statistical tools to manage risk, including (but definitely not limited to!) VAR can help. But the good news is that the headache is getting smaller; spikes are getting smaller and smaller, and shorter lived. Selling volatility became incredibly profitable in 2012, and has been even more lucrative since.
And that brings us back to QE. Each spike in 2010, 2011, 2012, and 2013 were related to changes or perceived changes in QE. Even the recent fall in the VIX spike was a result of Japan’s QE boost, not fundamentals. If the Fed removes QE entirely and keeps it off in 2014, selling volatility will become even riskier even as the chart looks more and more enticing. Yet mean reversion is one of the surer bets out there, so understanding how the VIX operates during pullbacks and panics remains a significant lesson for anyone actively investing in stocks, either on the long or short side.