By Phil Mackintosh, Chief Economist, Nasdaq
With hearings in Washington DC this week into the GME short squeeze, we decided to take a look at two related data points:
- What happens to stocks that rally significantly in a short period of time? The answer is: More of them stay up than you would expect.
- What has happened to short interest since the squeeze? The answer is: Lots of the most shorted stocks saw short interest fall, according to the latest data.
What happens to stocks that double in a week?
We looked at all the times in 2020 where stocks doubled in the space of a rolling five-day window.
After limiting our sample to common stocks that had at least a full 60 days of trading after the initial 100% increase, there were 371 unique observations. However, 40% of all instances occurred during the initial market-wide v-shaped rebound in markets following the Covid-19 selloff, as stimulus was added to the system.
In this analysis, we took those “COVID recovery” stocks out. What was left were 225 different NMS stocks that doubled (or more) in price over a rolling five-day window in 2020.
We all recently watched the Reddit-fueled rally in GME, which pushed the stock price from around $40 per share to over $400 per share, only to fall quickly back to almost $40 per share again. That has left some asking: Why wasn’t trading halted to protect investors?
The answer, it turns out, is more complicated. It also poses a warning to those who want to implement volatility halts for single stocks without disrupting markets unnecessarily.
When we looked at the subsequent performance of all the stocks with rapid rallies, we actually see that the majority held most of the gains, even months later.
Chart 1: Stocks with 100% price gains in one week often held most of those gains
We group the stocks into deciles based on their returns roughly two months later (Chart 1). This shows that:
- In almost all decile groups, the following week sees average prices pullback.
- But over the next two months, many deciles hold (most of) the remaining gains.
- While the stocks with the strongest rallies tend to resume their outperformance (decile nine and 10), this vertical axis is on a log scale, so the best (tenth) decile of stocks triple in a week and then roughly triple again.
- Only the worst (first) decile portfolio fell back below their prices at the starting of their five-day doubling window.
One thing that is reasonably consistent when you look at these fast-rising tickers is that there is usually a corresponding material news event. A handful of these cases were biotech companies that had positive news on drug tests; however, that wasn’t always the main catalyst. For example, three stocks in the top decile (10) were:
- Seres Therapeutics (Nasdaq: MCRB), which announced positive trial data on a new drug.
- Workhorse Group (Nasdaq: WKHS), whose initial rally was further boosted by its inclusion to the Russell 3000 index and an announcement of new company financing.
- SPI Energy Co. (Nasdaq: SPI), which rallied on the announcement of plans to build electric vehicles (EVs).
There are plenty of valid reasons for a stock to double in short time. Knowing that, ensuring that volatility halts designed to protect investors don’t incorrectly (and unfairly) slow trading in other companies whose valuations have also increased becomes much more complicated.
The squeeze is off
Part of the buying driving GME higher was a classic “short squeeze.” That’s when the stock price goes up rather than down like the shorts expect, causing margin and collateral calls (and loses) on their short positions. Eventually, losses can lead to shorts exiting their position by buying it back, adding to the short-term demand for the ticker, and in turn, potentially causing more shorts to cover.
We talked recently about why being 100% short actually means other investors are 200% long, which explains why there are usually still long investors willing to lend. But we didn’t talk about how short interest data is disclosed differently from long positions.
- Long holdings are disclosed by underlying institutional investors each quarter, with a multi-week lag mostly in form 13Fs. So, companies know details about who owns how much of their stock, but it is updated infrequently.
- In contrast, short interest data is disclosed in aggregate to Finra every two weeks and published with a one-week lag. So, we know total short interest data more frequently, but nothing about who holds the largest short positions.
We have updated short interest data showing what happened to the shorts in late January. It shows that many of the most shorted stocks saw a significant reduction in short interest during the prior two weeks. GME, in particular, fell from around 90% to 30% of market cap shorted.
Chart 2: The most shorted stocks mostly saw significant declines
It turns out the stock going against the trend, Atossa Therapeutics (ATOS), is also a stock with a 100% return in five days (which brings us back to the data in Chart 1). Consistent with our findings for many of the stocks in Chart 1, ATOS rallied on news. Specifically, Atossa announced “substantially positive results” from its Australian Phase 2 clinical trial of Endoxifen, with early participants seeing a reduction in tumor cell activity.
However, there are a number of hurdles remaining before the drug receives FDA approval. This is an example of buyers and sellers both trying to place a value on the future of the company. That’s an important way that markets allocate capital.
Why is this important?
There are already reports of investors losing money on their GME trading in January. Some would say that was inevitable, and the market should have worked better to protect those investors. Ironically, others have said the market should have allowed all investors to trade as they wanted.
They can’t all be right. Data shows that deciding which stock needs halting in the middle of a rapid rally is even more complicated than a single GME case study might suggest.
