In September, we suggested that the recently imposed bans on short-selling certain stocks would provide academics with a field day as they examined whether such restrictions actually had the intended effects. We compared it to the situation immediately after 9/11 when climate researchers were afforded an opportunity to measure the effect of airplane contrails on ground surface temperatures in the United States.
Well the data is starting to roll in now. And according to a study published by the Cass Business School in London, our contrail analogy may have been a little off the mark. As Ian Marsh and Norman Niemer point out in “The Impact of Short Sales Restrictions“, it is virtually impossible to compare the performance of stocks during the ban with their previous performance histories. Dramatic market volatility that was coincident with the bans meant that the usual caveat “all else being equal” simply didn’t apply. This is like climatologists studying post-9/11 ground surface temperatures in the presence of a coincidental surge in sun-spot activity.
So instead of just comparing the statistical properties of pre-ban returns with those from the ban period, Marsh and Niemer also compare the statistical properties of the “restricted” stocks with those of the “unrestricted” stocks (summarized by us below as “A” and “B” respectively in the diagram below).
In other words, it’s like they’re comparing the ground temperature in the US before and after the absence of airplane contrails and the ground temperature both in the US and Canada.
Everyone seems to have an opinion on the true effects of short-selling on security prices. In part, this results from several different and apparently disparate academic studies over the years. Marsh and Niemer point out that a 1977 study confirmed the prevailing intuition that short-selling pushed prices down. On the other hand, they say, a 2006 study found that short-selling actually put upward pressure on prices since buyers were comfortable that all negative information was already baked into them. Several other studies have been inconclusive with regard to prices but have found that volatility and market efficiency rise in the presence of short-selling.
Unlike ground surface temperatures, the authors find that return behaviour actually changed very little in the absence of otherwise ubiquitous forces (shorting). As they put it:
“We find no strong evidence that the imposition of restrictions on short selling in the UK or elsewhere changed the behaviour of stock returns. Stocks subject to the restrictions behave very similarly both to how they behaved before the imposition of restrictions and to how stocks not subject to the restrictions behave.”
While this seems to put a rest to the assumption that shorting (or the lack of it) has a dramatic effect on price behaviour, there is another important dimension to the debate – market efficiency. Unfortunately for advocates of the shorting-as-market-efficiency-creator argument, Marsh and Niemer actually find that the removal of shorting did not increase serial autocorrelation in returns (a common measure of the randomness of the proverbial “random walk”). In other words, markets were no less efficient without shorting.
In the end, the authors concede that any unique behaviour displayed by short-restricted stocks may have just resulted from sector-specific influences rather than the short-bans themselves. To return to our surface temperature analogy, it’s like a comparison of US and Canadian ground surface temperatures both in the absence of airplane contrails – but with a nasty cold front moving through the US midwest at the same time.
The relatively small size of the short-ban window (31 trading days in this particular study) means that any chill in prices could conceivably be the result of just such a freak mid-western cold snap. To mitigate for this possibility, Marsh and Niemer perform a bunch of other statistical tests. However, none of these change their basic conclusion that the universe seems to want to unfold as it wishes – with or without short-restrictions.