The Bank of Canada has released a study on the effects of last fall’s short-sale ban, titled Short Changed? The Market’s Reaction to the Short Sale Ban of 2008:
Do short sales restrictions have an impact on security prices? We address this question in the context of a natural experiment surrounding the short sale ban of 2008 using a comprehensive sample of Canadian stocks cross-listed in the U.S. Among financial stocks, which were singled out by the ban in both countries, we observe a significant increase (74 bps) in the difference between the U.S. share price and the Canadian share price. We also observe an impressive and surprising migration of the trading volume from the U.S. to Canada among financial stocks during the ban. Both price and volume effects are reversed after the ban and neither effect manifests itself among the nonfinancial stocks. Our findings support the view that prices reflect a more optimistic valuation when pessimistic investors are kept out of the market by binding short-sales restrictions (Miller (1977)). Our findings also imply that pessimistic investors were more preponderant in the U.S. than in Canada, which is corroborated by the fact that the short interest ratio for our sample stocks was much larger in the U.S. than in Canada prior to the ban.
…
Our findings lend support to an international version of Miller’s (1997) price optimism model in which the degree of pessimism manifested by investors varies across markets. According to Miller’s (1997) model, short sales constraints drive stock prices above their equilibrium value by preventing pessimistic investors from impounding their negative views on the stock price by selling the stock short. In the dual-market setting characterizing the present experiment, an expanded version of this model implies that, under short sales constraints in both venues, a cross-listed stock would trade at a higher price in the market where pessimistic investors are more prevalent and it would trade at a lower price in the market where pessimistic investors are less prevalent.
…
Our findings also lend support to an international version of
the Bai, Chang, and Wang (2006) model in which short sales are either motivated by allocational or by
informational considerations. From this perspective, the price increase that we observe in the U.S. relative to Canada among our treatment group stocks during the ban implies that a greater proportion of short selling activity in U.S. cross listed stocks was driven by allocational, i.e. uninformed investors, than in Canada during our sample period. In summation, our paper contributes to the literature in two important ways. First, by demonstrating, via a natural experiment crafted around cross-listed stocks, that short sales constraints do cause stock prices to trade above their equilibrium value as Miller’s (1977) price optimism theory suggests and, second, by showing how critical the ability to conduct short sales is to arbitrageurs as a mechanism to enforce the law of one price across markets.
This paper joins the collection – I have previously reported the IIROC Report on Short Selling Ban and The Undesirable Effects of Banning Short Sales.