Turnover effect: risk or mispricing?The role of retail investors

  • Despoina Kentrou

Student thesis: Unknown


The empirical relation of high turnover - low future returns, i.e. the turnover effect, became known after Datar et al. (1998) proposed turnover as an alternative to the bid-ask spread measure for liquidity. Turnover is the number of shares traded divided by the shares outstanding. The negative relation of returns and turnover expresses the illiquidity risk- returns relation. However, apart from the illiquidity risk premium advocates, other studies approach turnover and returns as being jointly determined by the same market dynamics, such as investors' sentiment and demand (e.g. Lee and Swaminathan, 2000) which can move prices away from fundamentals, causing mispricing effects. While the debate of illiquidity risk or mispricing remains unresolved, additional evidence about turnover's impact on the effect of other known market regularities, such as momentum (Hvidkjaer, 2008), urges a deeper understanding of how and why turnover associates with returns.This thesis examines the nature of the turnover effect and contributes to the debate of risk or mispricing which is vivid in all areas of asset pricing. To my knowledge, it is the first study which empirically distinguishes between individual and institutional investors (clientele approach) in order to break down the turnover effect and explore its nature and mechanism. The clientele approach is flourishing in the literature on market anomalies and behavioural finance but remains less popular in non-market anomalies relations. This thesis shows the existence of a clientele effect disguised in the turnover effect and it thus highlights the usefulness of a clientele approach in asset pricing research in general. While the focus of the thesis is not to investigate the motivations for retail trading or the interactions with institutions, it contributes to research about retail investors, bringing evidence that the presence or absence of them plays an important role in the predictability of stock market returns. Finally, it points to the need for further research in behavioral finance and microstructure towards reaching a better understanding of clienteles and return generating mechanisms, to help build more efficient asset pricing theories and models. The analysis uses all common/ordinary NYSE and AMEX stocks from 1993 to 2011, and employs intraday data from Trades and Quotes. The results do not coincide with a risk compensation explanation and support a mispricing effect. Findings show that the turnover effect, in the period before the economic crisis in 2007 (within the crisis there is no evidence of a turnover effect) is very sensitive to the presence or absence of retail investors trading the stock. In the cross-section the turnover effect is driven by stocks traded heavily by retail investors, while it vanishes for stocks where institutional investors prevail.
Date of Award1 Aug 2013
Original languageEnglish
Awarding Institution
  • The University of Manchester
SupervisorStuart Hyde (Supervisor)


  • turnover effect, liquidity, retail investors, clientele effect, individual investors, mispricing effects

Cite this