Investor Fear Illustrated: Changes in Long/Short Allocation
We’ve talked about how investors react to bear markets before: volatility increases, overnight gaps become larger, and the market falls much faster than it rises.
Here’s another observation to add to the mix: in bear markets active traders swing much more quickly between long and short. You might be tempted to say that’s savvy speculators capitalizing on higher volatility, but based on the fact that traders are overwhelmingly bad at what they do I would call it what it is: fear and panic.
The graph above shows the S&P 500 index (grey, 1-month moving average) versus the 1-month average absolute daily change in investors’ long/short allocation (red) based on Rydex’s S&P 500 2X mutual funds (our trading vehicle of choice).
When the red line is low, it means investors are making very small day-to-day changes to their long/short allocation, and when it’s high, they’re swinging quickly from long to short and back again. See geek note at end of post re: Rydex MFs and how this data is calculated.
During the 2003-07 bull (middle of chart), investors traded like molasses, changing their daily view infrequently. That doesn’t imply that they were on the right side of the market – in fact Rydex traders spent a good part of the bull market net short. But it does imply that whatever position they took, they were slow to change the outlook.
Compare that to the 2000-02 and 2008-09 bear. Investors became hyperactive, making as much as 40x larger day-to-day swings in their long/short allocation. They didn’t do a good job at being on the right side of the market, but they were very quick to change their outlook (usually, to their detriment, chasing recent returns).
Last note: it’s interesting that the red line has stayed so high through the 2009 bull. I think that has to do with how quickly it began (leaving a lot of folks late to the party and scrambling to go long) and how strongly it persisted (leading folks to frequently try betting against it). Both were unprecedented since the 1930’s.
Happy Trading,
ms
Geek note: the data in this post is calculated using Rydex’s long/short S&P 500 2X mutual funds (not to be confused with ETFs). A unique feature of these funds is that they provide a once-per-day snapshot of assets invested in both the long and short flavors of the fund (because they only trade once-per-day at the close and cannot be traded short). We can then use those asset totals to determine the net direction this group of active traders is expecting the market to take the following day. Click to read more about why we use leveraged mutual funds in our own trading.
. . . . .
To stay up to date with what’s happening at the MarketSci Blog, we recommend subscribing to our RSS Feed or Email Feed.
Filed under: Stock Market Mechanics, VIX & Volatility | 4 Comments
Isn’t it somewhat misleading to use asset totals? Because of the market going up/down the asset totals will change even if there was no change in allocation. Shouldn’t the metric to consider be changes in total assets/NAV?
RE to JB: very good thought, but the difference isn’t significant enough to much impact the results here. Put another way, if total assets in each fund were frozen at inception, there would be movement in the red line as funds gained and lost, but the red line would almost always be below 1% and would have only momentarily spiked up to about 3.5% during the depths of the most recent crises.
I showed it this simpler way b/c it would be confusing enough already to folks.
Thanks for the comment,
michael
Great visual…like the distinction that traders weren’t necessarily right or wrong, but quick to change their mind due to the emotion/volatility involved