Technical Trading and Strategic Liquidity Provision
The University of Hong Kong
Shanghai Jiao Tong University (SJTU) - Shanghai Advanced Institute of Finance (SAIF)
March 14, 2012
We develop a dynamic strategic trading model à la Kyle (1985) with risk-averse market makers to study the roles of different investors in liquidity provision over short horizons. Uninformed technical investors infer information from past prices, determine whether prices are overvalued or undervalued, and provide implicit liquidity to the market, whereas informed investors strike a balance between trading on private information and trading on liquidity provision. We show that: First, technical trading enhances market efficiency and market liquidity when market makers faces inventory risk. Second, short-term returns are negatively correlated. Third, informed and technical investors both employ contrarian trading strategies and their trades forecast future returns over short horizons. Fourth, market makers' inventory positions are negatively correlated with past and contemporaneous price changes but are positively correlated with subsequent price changes. Our strategic liquidity provision theory complements and re fines the competitive liquidity provision theory of Grossman and Miller (1988) and the model's predictions are consistent broadly with empirical evidence on short-term return reversals, contrarain trading behavior and return predictability of individual investors and corporate insiders, and empirical findings on specialists' position patterns. Our model also offers one explanation of paradoxical existence of technical analysis in seemingly efficient markets and sheds light on current debate on the Volcker Rule.
Number of Pages in PDF File: 49
Keywords: liquidity provision, technical trading, return predictability, market efficiency
JEL Classification: G11, G12, G14working papers series
Date posted: March 15, 2012 ; Last revised: May 29, 2012
© 2014 Social Science Electronic Publishing, Inc. All Rights Reserved.
This page was processed by apollo5 in 0.375 seconds