Home » Business

Risky Risk Management: Exploiting The Predictable Risk Management Behavior Of Short-Sellers

Updated on

Risky Risk Management: Exploiting The Predictable Risk Management Behavior Of Short-Sellers

xinyao huang
Western Asset Management

James Clunie
Jupiter Asset Management Ltd.

Lana Yan Jun Liu
Northumbria University – Newcastle Business School

May 27, 2016


The literature suggests that short-sellers mitigate the risk of large losses through the use of stop losses. This predictable behaviour exposes short-sellers to predatory trading risk. This paper examines a strategy of predatory trading against short-sellers and finds that positive abnormal returns can be earned by anticipating such short-covering.

Risky Risk Management: Exploiting The Predictable Risk Management Behavior Of Short-Sellers – Introduction

Short-selling is a tool for expressing negative investment opinions, as well as an essential part of arbitrage trading strategies, which exploit price anomalies and help to set market prices (Miller, 1977; Diamond and Verrechia, 1987; Angel et al, 2003; Engelberg, et al 2012, Boehmer & Wu 2013). Short sellers undertake two types of trade: short selling and short covering1. Both types of trades potentially have price impact. However, the literature tends to focus on the first part of the short-sellers’ trading activity, i.e. the opening of the position. More recent studies have investigated the covering of the short position. For example, Boehmer (2015) stipulates the existence of these two types of short-trading activities and highlights a lack of empirical studies on the covering of short position. This study helps to fill this gap in the literature, by providing new insights on short covering trades, the risks involved, and the corresponding market impact on stock prices. Specifically, we link short covering trades to short-seller’s responses to losses and examine stock returns around short covering.

Empirical studies show that short-sellers are not prone to the behavioral bias known as loss aversion. Instead, they crystallize small loss and cover losing position systematically. For example Gamboa-Cavazos and Savor (2007) investigate the factors influencing changes in short interest by using monthly US data on short interest. They find that short-sellers cover their positions after stock price increases and conclude that short-covering is not motivated by expected stock returns. They argue that capital constraints or myopic loss aversion could account for this behavior. Clunie et al. (2009) study short-sellers behavior and their responses to losses by using daily UK data over a three-year period. They find that short-sellers react to their own ‘mark-to-market’ losses, rather than simply to the change of stock price. They identify that greater losses trigger stronger reactions among short-sellers. More recently, Boehmer et al (2015) examine data on large short positions in the Japanese stock market. They provide detailed evidence of short-covering trades and find significant positive returns to be associated with such trading activities. Their finding indicates that short sellers elevate stock prices by covering their large short positions.

In practice, risk control mechanisms, such as a stop loss rule which forces investors to close out their positions when they suffer losses up to a certain loss level, are used by short-sellers to contain their risks. This is consistent with the empirical evidence. Unlike long-only investors, whose losses are limited to the invested capital, short-sellers theoretically face the risk of unlimited losses since there is no upper limit to the price at which stock can be traded. The use of stop losses leads short-sellers to seek to cover their short positions once a pre-defined loss has been observed. This systematic short-covering behavior mitigates the risk of large losses, but also unintentionally presents a form of predictable behavior, which can be exploited by other market participants.

The objectives of this research are to examine the risk of predictable short covering behavior and to inform a practical trading strategy that exploits this predictability. Specifically, we ask the following questions: 1, does the use of stop losses present a form of predictable behavior? 2, can traders earn significant abnormal return by predating the expected short covering? 3, how to construct a profitable predatory trading strategy? With the use of a stock lending dataset from Data Explorers2 and stock returns via Datastream, we estimate the cost basis of short positions and study the reaction of short-sellers in response to losses over the period from September 1, 2003 to May 3, 2007. Our finding is consistent with the literature, in that short-sellers are found to cover their positions routinely once they fall to a mark-to-market loss. This systematic short covering (forced buying) is associated with market impact on stock prices. We observe significant abnormal stock returns around these events, similar to the pattern observed by Chen et al (2006) where predators profit from the predictable tracking error constrains faced by index fund managers. Based on these observations, we develop a simple trading strategy that can be employed to take advantage of short-seller’s predictable short-covering behavior.



See full PDF below.

Leave a Comment