Alliances and Return Predictability via Alpha Architect
Chinese University of Hong Kong – Department of Finance
Emory University – Department of Finance
University of Hong Kong – Faculty of Business and Economics
May 8, 2014
In a white paper on SSRN, Jie Cao, Tarun Chordia and Chen Lin discuss a trading strategy designed to exploit the information contained in the returns of alliance partners, yields economically and statistically significant returns. A long-short portfolio sorted on lagged returns of strategic alliance partners provides a return of 89 basis points per month that is robust to a number of specifications. Increased correlation in returns after the formation of alliances is driven by increased economic links and the increased probability of mergers amongst alliance partners. Investor inattention and limits to arbitrage may be the source of underreaction of a firm’s returns to that of its partners’.
Fama’s seminal work on market efficiency posits that markets should quickly and correctly incorporate information into prices. More specifically, the weak form of efficiency states that information contained in all past prices should be quickly incorporated into current prices and the semi-strong form of efficiency requires that prices should reflect all publicly available information. A number of papers have documented results that are in conflict with market efficiency, sometimes even the weak form of market efficiency. For instance, the post-earnings-announcement-drift documented by Ball and Brown (1968) finds that earnings surprises impact stock returns for up to nine months; Jegadeesh and Titman (1993) find that returns over the past one year predict future returns for up to a year; Jegadeesh (1990) and Lehmann (1990) document short-term reversals in stock returns; Loughran and Ritter (1995) document lower returns for firms with equity issues; Sloan (1996) finds that stocks with higher non-cash component of earnings earn lower abnormal returns; Ang, Hodrick, Xing, and Zhang (2006) and Cooper, Gulen, and Schill (2008) show that high idiosyncratic volatility and high asset growth, respectively, lead to lower future returns.
The literature has proposed a number of explanations for the violations of market efficiency. Korajczyk and Sadka (2004), Avramov, Chordia, and Goyal (2006), Chordia, Goyal, Sadka, Sadka, and Shivakumar (2009) suggest that trading frictions, including transaction costs and liquidity, impact the incorporation of information into prices.3 The concept of market efficiency with respect to an information set is defined by Jensen (1978) as the inability to profit from that information.4 Daniel, Hirshleifer, and Subrahmanyam (1998) and Barberis, Shleifer, and Vishny (1998) suggest that investors’ psychological biases can impact prices. More recently, Hirshleifer and Teoh (2003), Peng and Xiong (2006), DellaVigna and Pollet (2009), Cohen and Frazzini (2008), Hirshleifer, Lim, and Teoh (2009), and Cohen and Lou (2012) suggest that limited investor attention leads to underreaction of prices to information.
This paper documents the underreaction of a firm’s prices to those of related firms. More specifically, we examine strategic alliances and the impact of a partner’s stock return on that of the firm. Each month, we first sort firms into quintile portfolios based on the last month average returns of their alliance partners. The long-short portfolio is then formed by going long in stocks of firms whose partners had high returns and going short in stocks of firms whose partners had low returns last month. This long-short portfolio provides an average raw return of 89 basis points per month over the sample period from 1991 through 2012. This result is robust to a number of specifications including different adjustments for risk, controlling for different proxies for cross-autocorrelations, excluding partnerships with customer-supplier relationships as well as controls for industry returns. The partner-based trading strategy profits are weaker in recent years and for partners in different industries. The results do not obtain for the largest quintile of stocks. The partner-based underreaction declines monotonically across the size quintiles and exists only for stocks that belong to size quintiles one through four.
The economically large returns to the partner-based trading strategy point to a violation of the semi-strong form of efficiency and raise two questions: (1) why are the returns of alliance partners related, and (2) why is there a lagged response of a firm’s return to that of its partners’? We address both these questions in turn.