- After a 103% stock market advance over a mere 18 months, many investors are looking for signs of a significant correction.
- That is particularly true now in this time of year when many corrections have historically occurred, combined with the Fed talking about tapering and the COVID Delta variant still wreaking havoc.
- Corrections can come in 2 ways: a time-based correction, a price-based correction, or both. Are we in a mid-cycle time-based correction? Only time will tell (no pun intended).
- Remember that corrections can and do happen frequently in the context of a long-running upward trend.
Technicians often define market regimes as either trending or mean-reverting, and they have specific tools for each approach. I used to be a mean-reverter, always wanting to be that contrarian who would show the herd just how smart I was.
About the expert
Jurrien Timmer is the director of global macro in Fidelity's Global Asset Allocation Division, specializing in global macro strategy and active asset allocation. He joined Fidelity in 1995 as a technical research analyst.
These days, I am more chill and have learned to trust the market gods and follow the math. To be sure, there are times when there are screaming excesses evident, and those are worth playing for, but those signals are fairly rare.
Most of the time the signals are much smaller, which means that the risk of being out of the market (and losing out on the compounding magic) is greater than getting caught in a squall. Remember, the stock market's value proposition of a 10% return with an annualized volatility of 15 includes the myriad corrections that happen all the time. The annualized volatility refers to the Volatility Index, a measure of expected volatility. It reflects the range of market volatility investors expect in the near future. It tends to go up when the S&P 500 goes down. March 2020 saw a spike of 82.69. The lowest reading was 9.14 in 2017.
For the most part, my investment process these days is to try to stay on the right side of the secular trend, and to use that trend as context to help navigate the market's many inflection points. Without context, most indicators are reduced to a coin toss.
Anyway, the temptation to mean-revert is alive and well right now, especially given 1) we have entered the seasonally vulnerable September-October window, 2) the Fed is about to taper, and 3) the Delta variant is wreaking havoc.
And all this following a 103% advance in less than 18 months. This has been the strongest and fastest and most stimulus-fueled recovery in history, and it's certainly understandable to want to take some chips off the table, or at least to rebalance. Rebalancing is always a good idea.
With the market having fewer and fewer drawdowns since last fall, it's understandable to think back to similar low-vol periods (1993, 2013, 2017) and wonder what kind of volatility event is lurking around the corner. Interestingly, all 3 of these periods were followed by Fed tightening.
For sure, the seasonal window has investors worrying that the time for a correction is now. But while it's true that many of the worst drawdowns in history have happened during the September-October time frame, on average, this window is more often just a bump in the road in the absence of a larger bearish narrative.
Also, while the September-October time frame often produces the worst declines, they are often followed by the best recoveries. Plus, the actual batting average of weekly returns during this seasonal window is barely below 50% (the worst week is 48%). This shows that the seasonal window is skewed by outliers, which suggests that it is less seasonal than it appears.
But let's assume for a moment that the stock market has reached another inflection point and that we are at the start of a correction (mid-cycle or worse). What would happen next is far from uniform.
Corrections can take many forms: price or time, or both.
Examples of time-based corrections are 1994 (a year of sideways while the Greenspan Fed doubled short rates), and 2015 (when the Yellen Fed tried to normalize policy while China was going through a growth slowdown).
A good example of a price correction is late-2018, when the Fed was deemed to have overshot and the market suffered a swift 20% drawdown.
Time corrections can be barely visible at the surface. Case in point, the S&P 500 has obliterated the 2010 mid-cycle correction analog. Back in 2010 the S&P 500 went through a 16% correction from April to July on liquidity fears as the Fed ended the first round of QE.
But beneath the surface we can see how poor the market's internals have been this summer. At one point in early August, only 30% of the S&P 500's constituents were above their 50-day moving average. The index climbed to ever higher levels through the month with at most 63% of constituents trading above their 50-day moving average—compared to 84% earlier this year.
We can see below that large-cap value stocks have been going nowhere for about 6 months now.
The reason we can't see this correction is that the large growers remain on their relentless advance.
In fact, the year-over-year change in the value/growth ratio is playing out exactly as it did in 2009/2010.
Are we in a mid-cycle time-based correction? The headline indexes say no, but the internals say "maybe." We will only know in hindsight.
Either way, I continue to have high conviction that we are in a secular bull market, and that it's far from over.