The predictive nature of short interest on market returns and the effect of short selling on volatility, liquidity and price discovery with some international evidence
Bibliographic record
Abstract
I look to explore the findings of Boehmer et al. (2010) and in particular test across model specification investment horizon and across countries. This in turn means I look at whether short sellers are informed traders and if there is evidence that short sellers engage in market manipulation. I also look at whether and to what extent short sales affect liquidity, price discovery, volatility and cross-section of stock returns. The main novelties of research of this PhD are that I employ a new more efficient asset pricing model to the findings of Boehmer et al. (2010), I explore also a time period after publication to observe market efficiency and that I explore an international market. I am also one of few studies to study volatility, liquidity and price discovery in regards to a short sale ban in the UK. \n \nFirstly, I look at whether the strategy of Boehmer et al. (2010) is valid when another most recent and efficient model is used to adjust for risk premium. Boehmer et al. (2010) used the Fama and French Three Factor Model with Momentum to adjust for risk premium, in my case I use the Fama and French Five Factor model to adjust for risk premium. The strategy of Boehmer et al. (2010) involves going long the top percentile of stocks ranked by short interest and going short the bottom percentile of stocks ranked by short interest and rebalancing each month based on new short interest data. I find that heavily shorted stocks underperform lightly shorted stocks. The Fama and French Five Factor model holds a high positive alpha for lightly shorted portfolios on top of both higher excess and raw returns for lightly shorted portfolios compared to heavily shorted portfolios. However, the short component of the Boehmer et al. (2010) strategy yields a positive return, therefore I advise to go long the top five percentile of stocks. The Fama and French Five Factor model does indicate there is good news in short interest albeit without a short component. \n \nSecondly, I look at whether the strategy of Boehmer et al. (2010) is still valid after publication in the US stock market or whether an efficient market has caused arbitrage to take place and make this strategy redundant. I again find that heavily shorted stocks underperform lightly shorted stocks, however the strategy of Boehmer et al. (2010) is not completely valid as the short component of the strategy yields a positive monthly return. It is still however valid to go long the top five percentile of stocks and I believe that this underperformance of heavily shorted stocks means arbitrage has not taken place. \n \nThirdly, I also look at whether the strategy of Boehmer et al. (2010) is valid in another OECD country such as Canada. This is used as an indicator of the international OECD market based on short interest data availability. I find again that heavily shorted stocks underperform lightly shorted stocks, in line with findings consistent with previous literature. However, the strategy of Boehmer et al. (2010) is again completely not valid as the bottom five percentile of stocks have a positive monthly return. I advise going long the top five percentile portfolio as the best strategy, in line with my findings regarding arbitrage in the US stock market. There is no valid reason in holding a long/short portfolio if the short side is yielding a positive monthly return. I find the US stock market outperforms the Canadian stock market on average over the February 2010 to July 2017 period. \n \nFourthly I look at the relationship between liquidity, volatility and price discovery with short selling. I use the UK financial stocks short sale ban of the 2007-2009 financial crisis in order to explore the effects of short selling on liquidity, volatility and price discovery. I employ a control portfolio as well to see the effects of the short sale ban. I employ a GARCH model to explore the effects of short selling on volatility. I find that volatility is not affected during the short sale ban, this in turn questions the significance of short sale bans like many other studies before mine have done. I employ a Bid-Ask Spread Model to explore the effects of short selling on liquidity. I find that the Bid-Ask Spread Model shows good fit regression wise and that liquidity deteriorates during the short sale ban period. Lastly, I employ a Wald-Wolfowitz Runs Test to see fat tails in its distribution, this shows the effects of price discovery on a short sale ban. I find that price discovery deteriorates during the short sale ban period.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.004 | 0.024 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.003 | 0.003 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.003 |
| Insufficient payload (model declined to judge) | 0.009 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".