A closer look at return predictability of the US stock market: evidence from new panel variance ratio tests JAE H. KIM*† and ABUL SHAMSUDDIN‡ †Department of Economics and Finance, La Trobe School of Business, La Trobe University, Bundoora, VIC 3086, Australia ‡Newcastle Business School, University of Newcastle, Callaghan, NSW 2308, Australia (Received 5 February 2014; accepted 19 December 2014) This paper examines the return predictability of the US stock market using portfolios sorted by size, book-to-market ratio and industry. We use novel panel variance ratio tests, based on the wild bootstrap proposed in this paper, which exhibit desirable size and power properties in small samples. We have found evidence that stock returns have been highly predictable from 1964 to 1996, except for a period leading to the 1987 crash and its aftermath. After 1997, stock returns have been unpredictable overall. At a disaggregated level, we find evidence that large-cap portfo- lios have been priced more efficiently than small- or medium-cap portfolios; and that the stock returns from high-tech industries are far less predictable than those from non-high-tech industries. Keywords: Fama–French portfolios; Market efficiency; Monte Carlo experiment; Panel data; Wild bootstrap JEL Classifications: C12, G14 1. Introduction Under the efficient market hypothesis, stock returns are purely unpredictable since stock prices fully and instanta- neously reflect all available and relevant information (Fama 1970). When the market is weak-form efficient, stock returns show no autocorrelation and cannot be predicted by exploiting past price information.§ This hypothesis has been tested extensively for decades, but the empirical results have been mixed (see Park and Irwin 2007, Yen and Lee 2008, Lim and Brooks 2011). While early studies have found little evidence of autocorrelation in stock returns (Fama 1970), a large body of research that emerged in the mid-1980s has provided evidence of return predictability such as the momentum effect (Jegadeesh and Titman 1993) and return reversal effect (Debondt and Thaler 1985). Recently, there have been claims that return predictability fluctuates over time. Malkiel (2003) argues that return pre- dictability may arise over time since the collective judge- ment of market participants cannot be always correct, which is complemented by Timmermann’ s(2008) finding that forecasting models reveal stock return predictability only during certain pockets of time. These claims are in line with Lo’ s(2004) adaptive markets hypothesis, which asserts that market efficiency is highly context-dependent and dynamic. Thus, return predictability can arise from time to time as boundedly rational market participants adapt to changing market conditions.¶ Kim et al. (2011) provide empirical evidence that supports the adaptive markets hypothesis for returns on the Dow Jones Industrial Average (DJIA) index, where return predictability changes over time depending on the prevailing market conditions. They also report a considerable decline in the return predictability from 1980, possibly due to a range of technical innovations *Corresponding author. Email: J.Kim@latrobe.edu.au §Portfolio return autocorrelation arising from market microstruc- ture biases such as non-synchronous trading and bid-ask bounce is viewed as spurious autocorrelation. On the other hand, return autocorrelation arising from time-varying expected returns and partial price adjustment is termed as genuine autocorrelation (see Campbell et al. 1997). In this paper, we use the term weak-form efficiency and return predictability interchangeably, as autocorrelations in US stock returns are primarily attributed to partial price adjustment rather than market microstructure biases or time-varying expected returns (see Mech 1993, Anderson et al. 2013). ¶Hommes (2001) provides a survey of papers which model finan- cial markets as complex adaptive systems. © 2015 Taylor & Francis Quantitative Finance, 2015 Vol. 15, No. 9, 1501–1514, http://dx.doi.org/10.1080/14697688.2014.1002419