Stats › Options, shorts and positioning › Short interest
Short interest predicts returns equal weighted and predicts nothing value weighted
| Short interest cutoff | Equal weighted | t | Value weighted | t |
|---|---|---|---|---|
| At least 2.5% of shares | -0.48% | -2.79 | +0.16% | 1.02 |
| At least 5.0% | -0.69% | -3.53 | -0.14% | -0.76 |
| At least 10.0% | -0.78% | -2.87 | -0.27% | -0.86 |
| At least the 95th percentile | -0.54% | -2.67 | +0.24% | 1.24 |
| At least the 99th percentile | -1.25% | -4.42 | -0.38% | -1.09 |
Four factor intercepts in per cent a month. Exchange listed shares from 1976 to 2002 and the electronic market from July 1988, with combined results over 174 months from July 1988. Two of the five value weighted portfolios have a positive sign. The paper states plainly that value weighted portfolios show no reliable underperformance and that no listed exchange value weighted intercept has a t statistic of minus two or worse.
Heavily shorted stocks underperform by 0.48 to 1.25 per cent a month on an equal weighted basis with t statistics between 2.7 and 4.4. On a value weighted basis not one of the same portfolios has a t statistic worse than minus two.
Members only
The rest of this page is for members
Below this point there are 5 sections, 1 table and 3 named sources, roughly 950 words of it. Every figure carries the source it came from and the date the data is from.
You can keep browsing every statistic in the library for free. The intro and the summary are always open.