Short interest is the number of shares outstanding on the short side. There is a claim that this may be foretelling the end of our long bull market. Does this even make sense?
Short interest is an indicator that is sometimes used by traders, and traders look at a lot of indicators, even ones that we could call fundamental ones like this. Some trading books mention it, and stock markets do publish data on short interest, which may have a use in limited situations, such as with a stock under a lot of pressure or with an IPO that may have a disproportionately large amount of it, such as Beyond Meat had at one time.
You never want to trade this indicator alone though no matter how extreme the readings may be, although readings do have to be pretty extreme for this to be relevant. This information can be used at times in combination with the other things that you are looking at to take into account what is called a short squeeze.
In all cases, we need enough short interest to make a material difference in price should it rise enough. This means a big percentage of shares traded need to be short, which is indeed pretty unusual.
With a short squeeze, if there is a lot of short interest, like there was with Beyond Meat, and the price keeps going up, this short interest will need to be covered, which places more upward pressure on the stock’s price. Shorted shares require a share purchase to cover them, where the short trader repays the stock borrowed by buying it, and this increases the demand for it, which is bullish.
Now that we have an idea what short interest means, how do we get from a trader’s tool that is very occasionally used to one that can predict long-term market direction, such as the end of a 10-year bull market?
The first thing to realize with short interest is that this is not something that is the least bit meaningful when applied to the market as a whole. Stock exchanges do release this data, and the Nasdaq releases it twice a month instead of one, and also is comprised of stocks that are more volatile on average and therefore more prone to shorting, so we’ll look at their data to get a better idea of how useless this is for investing or predicting long-range trends.
The average number of shares shorted per day on the Nasdaq is about half a million. The market itself has an average of 1.5 billon shares change hands each day. The short shares therefore only represent about 1/3000 of the overall market, an utterly insignificant amount.
This ratio does go up and down all the time, but whether it is 1/3000 of the overall market, 1/2000, or 1/4000 cannot possibly matter because all of these numbers are far too small to have any effect on anything.
The long interest, on the other hand, the 2999/3000 fraction of this, does matter of course and this is what drives stock prices. Given that the long interest is on average 3000 times more influential than the short interest, a tail this small could not possibly wag the dog.
When we look to interpret short interest, it’s preferable that we look at not the number of short shares out there but at what is called days to cover, which essentially is the average length of time that they are held. You take the total number divided by the average daily short volume to get this, and this average ranges from 3 to 6 days typically.
So now you have 1/3000 of the market which only holds these shares for a few days being predictive of major stock market trends somehow. It is simply not possible that this could matter at all to investors because the scope of influence is a completely meaningless one and one as far away from the scope of investors as you could get.
Short Sellers Do Not Possess Magical Powers
Could it be that these traders somehow have special insights that the market does not have? There is a view among some that there are crystal balls out there and we just need to look hard enough to find them, the holy grail of trading and investing. Although this is a myth, people who trade short tend to be at least more knowledgeable than investors are.
Traders don’t look beyond the very short term though, so at best, we’re measuring very short-term trends here, if people are out of these trades so quickly. At the moment, the average holding time for these shares is just 3 days, but somehow, they are predicting the future way beyond this.
To be fair, we still should look at the arguments in favor of this, thinking that this tail does indeed wag the dog, as absurd as this may be. Professor Matthew Ringgenberg of the University of Utah is the champion of this idea. He has stepped up once again to speak on the topic, and in this case to issue a warning to us, as the numbers aren’t looking good to him at all lately.
In his paper on this that he published in 2014, he pitted short interest against popular investing ratios such as price to earnings and price to book value, in the period between 1974-2014, and found that short interest came out on top, and in his mind, this makes this indicator the best one out there.
Ratios like this run counter to trends though so that says nothing. During bull markets, they go down, and during bull markets, they go up, so if you’re looking to use lower ratios to predict bullishness and higher ones to predict bearishness, you bump your head. A lot of people bump their head on these though because they don’t understand this.
Since short interest at least moves in the same direction as trends, where we expect this to increase in bearish times and decrease in bullish times, that would be more than enough to handily beat these ratios, which place us on the wrong side of trades overall.
If we are going to use these things, we want to align our interpretations of these ratios with reality, where we’re buying as they go up and selling as they go down, and this would have presented a much bigger challenge to the professor’s view.
The error in reasoning here is to assume that this small sample of indicators that he looks at are all there is, and therefore when his beats the others it deserves to be declared the best out there. Long interest, the behavior of the market itself, didn’t make the list though, and this indicator, price that is, isn’t just predictive of what we want to know, it is what we want to know itself.
Where is this magic supposed to come from? He claims that short sellers are more in the know because it costs more to short and shorting has “infinite risk.” Short sellers on average, being traders, do tend to be a lot more focused on markets than investors are, but at best this means that they are better at shorting than investors are, and this has nothing to do with long-term trends at all, which they do not partake in or anything close.
This also tells us that the professor misunderstands the comparative risks of shorting as it is practiced, and the only approach that would have infinite risk is one where one had unlimited funds and took on unlimited downside risk and stayed in positions to infinity if necessary.
These are traders though who are only in positions for a few days on average, and will head for the hills when things even start to go wrong, unlike investors on the long side, who are actually the ones that take on more risk, and far more risk in fact. Shorting is less risky overall than going long because the risk of markets crashing is considerably higher than markets exploding upward in a manner that could not be managed, which never happens.
Normal Deviations in Meaningless Indicators Are Not So Scary
The concern that Ringgenberg has right now is that the short interest at the end of August was 20% higher than it was to start the year. 20% is not a big number at all for a variation in short interest though, and we often see moves month to month larger than this.
We started the year at 3.72 days to cover, and at the end of August, this went up to 4.17. In case you are too worried about this, the latest number is 3.72, right where we started the year. Perhaps this was only a false alarm. We had numbers a lot higher than 4.17 earlier in the year so if 4.17 alarms us, imagine how we felt when this was in the 6’s.
These numbers bounce around this much all the time and really don’t tell us anything useful due to their lack of correlation with price. Between January and mid- April, short interest went from 3.72 to 6.50 and the Nasdaq went up by 20%. However, along the way you would have gotten a lot of mixed signals as this number jumped up and down quite a bit.
During the fourth quarter selloff last year, short interest went from 4.72 in mid-October to 3.72 at the end of December. We might think that, given these two instances, this might be a good contrarian indicator, but the results aren’t really correlated very well with anything overall, not that we should really expect them to be.
Ringgenberg tells us that this 20% dip and short interest allegedly spiking lately indicates the risk of a significant market downturn, although with this number moving back down to exactly where it was at the start of the year, we can no longer claim that short interest is spiking lately. This is not just things calming down in September, as if there was a spike here, it was between mid-April and mid-June, not that these spikes mean anything really.
For people who plan on getting out if a real bear market comes, one beyond the ups and downs that we’ve seen lately, it makes more sense to go with real indicators, ones that actually speak to the risks of this, like price for instance.
If we really do want to be guessing at where we are going, we’re going to at least need to see it happening in real life or at least starting to happen. It’s hard to imagine something less relevant than short interest as far as predicting long-term stock market trends or any trend for that matter, so we can confidently just look away here.