Investor propensity to speculate and price delay in emerging markets

https://doi.org/10.1016/j.irfa.2023.102557Get rights and content

Highlights

  • Evidence of investors' poor portfolio choice affecting price efficiency.

  • Stocks with lottery-type payoffs exhibit greater price delay.

  • Higher market-level propensity to speculate leads to greater price delay for general stocks.

  • Results hold when simultaneously considering MAX and IVOL effects.

  • Results remain robust when controlling for uneven sample size across emerging markets.

Abstract

Stocks with lottery-type payoffs exhibit more pronounced price delay. This finding holds for emerging market stocks even when jointly considering the impact of IVOL. In a cross-market analysis, a stronger market-level propensity to speculate, gauging the strength of investor preference for lottery-type payoffs, is found to delay the price reaction to information for stocks in the market in general. These conclusions remain robust when using a random sample that mitigates the bias from unevenly distributed sample observations across markets. Our findings add to the evidence that investors' asset choices that deviate from ideal portfolio diversification influence the process of stock pricing.

Introduction

The literature has identified a variety of market frictions that impede the process of stock price incorporating information. Those frictions include incomplete information, asymmetric information, short sales constraints, taxes, liquidity and noise trader risk, among others (Brennan (1970), Miller (1977), Merton (1987), Barberis, Shleifer, and Vishny (1998), Barber and Odean (2000), Easley, Hvidkjaer, and O'Hara (2002), Pastor and Stambaugh (2003), Mitton and Vorkink (2007), Goetzmann and Kumar (2008)). In particular, non-diversified portfolio choices, being a common practice among investors, should have profound influence on price efficiency. When investors hold non-diversified portfolios by allocating funds to stocks tilted toward certain characteristics, their portfolio choices may well influence the process of stock pricing when arbitrages are limited or costly (e.g., Barberis and Huang (2008), Kumar, Page, and Spalt (2011)).1

One example that has received great attentions in recent literature is lottery-type stocks. There is extensive evidence that certain investor clienteles have strong preference for extreme payoffs, leading to the over-pricing of lottery-type stocks (e.g., Kumar (2009), Kumar et al. (2011)).2 There are nonetheless few studies investigating whether investor preference for lottery-type stocks is a source of severe frictions affecting price efficiency for stocks. In this study we focus on this issue and further assess whether such stock-level delay in pricing information extends its impact and inflicts on the pricing for the overall market. Note that most current researches focus on one single US market or developed markets on related issues, in this study we exploit the research questions by focusing on emerging markets. Emerging markets are characterized as having institutional factors deterring arbitrages, being dominated by unsophisticated retail investors, being more volatile and less liquid, having weak corporate governance, suffering from greater political risk, being more segmented from world market, and more importantly having socioeconomic characteristics coincide with gambling investors.3 In this study we examine the relation between investor preference and market efficiency by taking advantage of the multi-market platform and characteristics of 32 emerging markets.

The primary objectives of this research are twofold. First, we examine whether stocks characterized by paying extreme returns are slower in their price adjustment to market information. Second, we examine whether the preference for lottery-type payoffs, possibly driven by an investor cohort, extends its influence and affects price efficiency for the overall stock market. That is, we test that the presence of an investor clientele with a strong preference for lottery-type payoffs, which strength will be gauged and termed as propensity to speculate, does not only delay the information pricing for the clientele's preferred stocks but also impede the information pricing process for the overall market after accounting for other relevant market-level institutional factors. This scenario arises when the excessive demand for lottery-type payoffs increases the arbitrage costs and weakens the information environment for the entire market.

This study uses a stock's highest daily return in the month (MAX), as proposed by Bali, Cakici, and Whitelaw (2011), to proxy for the demand for lottery-type payoffs for individual stocks.4 The literature has widely documented the evidence of idiosyncratic volatility (IVOL) puzzle since the study by Ang, Hodrick, Xing, and Zhang (2006), which research finds stocks with higher IVOL tend to have lower next period returns. Hou and Loh (2016) evaluate various explanations for the IVOL puzzle, and find that investor's lottery preference playing a critical role in explaining the puzzle for the US market. Their results suggest the importance of simultaneously considering IVOL when exploiting the impact of lottery-stock preference on price efficiency.

To properly incorporate both MAX and IVOL in the regression analysis, this study considers the following issues. First, as indicated by Hou and Loh (2016), the MAX variable is a range-based measure of volatility and highly correlated with IVOL. Second, the negative impact of MAX on price efficiency may well be non-linear and concentrated on stocks with particularly high MAX. We revise the measures as follows in testing their impact on price delay accordingly. First, we simultaneously include in the regression an IVOL measure that is orthogonalized against MAX. The ensuing regression coefficient of MAX then measures its impact net-of-IVOL effect. Next, instead of using the continuous MAX variable, we apply an indicator variable, which is assigned a value of one when the stock has a MAX in the top quintile. Note that such treatment better accommodates some cross-market institutional differences.

This study proposes a market-level measure for investors' propensity to speculate (PTS), which is estimated by the overpricing of lottery-type stocks that is predicted by the culture factors whilst controlling for the institutional factors related to limits to arbitrage for the market. The measure is designed to assess the extent to which investors in a market pursuing such lottery-type stocks while isolating it from the cross-market institutional differences in limits to arbitrage.

To gauge the speed of adjustment to market information, this study applies the price delay measures revised from those by Hou and Moskowitz (2005). Our delay measures further consider stock price reactions to both the local and the global market information, as well as possible asymmetric speed of adjustment to good news versus bad news. Both an R2-based delay measure and a regression-coefficient-based delay measure are employed in our testing (Saffi and Sigurdsson (2011), Bae, Ozoguz, Tan, and Wirjanto (2012), Boehmer and Wu (2013)).

Using the emerging market data from 1980 to 2016, we have several important findings. First, the R2-based delay measure (Delay1) finds that emerging markets exhibit a weighted-average of 22% to 40% of stock returns attributable to lagged global and local market information. The regression-coefficient-based delay measure (Delay2) finds that the average price delay varies across emerging markets, with 54% to 70% of stock return responsiveness being made toward lagged market information. Our study also shows that the coefficient-based Delay2 measure captures the phenomenon that emerging market stocks tend to respond slower toward global information than to local information, and slower toward bad news than toward good news. Our analysis also shows that stocks of smaller size, lower turnover, less analyst coverage, lower beta risk, lower liquidity, higher MB, lower earnings, and higher leverage tend to show slower reactions to market information, a finding similar to those for the US market.

Second, our results find that lottery type stocks, defined as those having higher MAX values, tend to suffer from more pronounced price delay in emerging markets. This suggests that investors' lottery-stock preference induces a barrier for stocks to incorporate market information in the pricing process. A similar association with price delay is found for IVOL, after controlling for the MAX effect. Overall, our findings suggest that lottery type stocks or stocks with a high idiosyncratic volatility, even when being jointly considered, introduce significant limits to arbitrage in emerging markets.

This study further examines partitioned samples. We re-group the sample according to FTSE country classifications, which provide more refined and time-varying country classifications for our sample emerging markets. The positive relation between MAX/IVOL and price delay remain for ‘Advanced Emerging’ as well as for ‘Secondary Emerging’ countries. In addition, we also group the sample markets based on their geographical regions, which naturally breed similar institutional and investor behavioral characteristics. This regional analysis finds that the relation between lottery type stocks and price delay is weak for Latin American markets, suggesting a link between speculative trading behavior and geographic attributes.

Finally, we provide empirical evidence corroborating the argument that the presence of an investor clientele with a strong preference for lottery-type stocks aggravates the price delay for the overall stock market. That is, investors' preference for speculative trades, as measured by the market's PTS, affects price discovery for stocks in general in the market. Specifically, even for those non-lottery-type stocks traded in a market having a PTS of one standard deviation higher than its median, they also suffer an average of 16.01% more information delay relative to similar non-lottery-type stocks traded in a median-PTS market, when controlling for country factors including short sales restrictions and price limit implementation. These results are related to theories of incomplete information and limited stock market participation (see Merton (1987), Shapiro (2002), Hou and Moskowitz (2005)). The above conclusions remain robust when using a random sample that mitigates the bias from unevenly distributed sample observations across markets. To the best of our knowledge, this is a new finding to the literature regarding the impact of investor preference on the price informativeness of general stocks in a market.

This research explores the role of investors' preferences for speculative assets in pricing stocks. Taking advantage of the international market platform, this study evaluates the impact of lottery-stock preference on price delay both at the individual stock level and at the market level. The evidence conforms to our main proposition that an investor clientele with a lottery-payoff preference in their asset choices reduce the informational efficiency of prices for individual lottery-type stocks and for the overall market.

Preference for lottery-type stocks has long been found correlated with investors' social, religious and cultural background (Kumar (2009), Kumar et al. (2011)). The demographics of investors in emerging markets tend to correspond to those characteristics delineated by those studies. The ensemble of emerging markets across continents offers a multinational sample with rich diversity in social and cultural backgrounds, in economic and financial development, and in industrial structures. These markets breed distinct habitats with varying preferences for speculative payoffs and levels of financial market efficiency, and naturally serve as our research subject to examine whether investors' speculative choices harms market information efficiency. Our study contributes to the literature in providing evidence on the impact of investor preference in asset choices on market efficiency with a general sample while without suffering from the sample dilutions by large-size developed stock markets.

Finally, our evidence also adds to the literature on the IVOL puzzle. Recent studies find that the IVOL puzzle can be largely explained by MAX (Hou and Loh (2016)). Our evidence shows that IVOL, even after being purged of the MAX effect, still acts as a limit to arbitrage in emerging stock markets, when price delay is used as an evaluation media.

The rest of the paper is organized as follows. Section 2 develops our hypotheses. Section 3 describes the data and our empirical methods. Section 4 presents results on the effect of lottery-type preference on price delay. Section 5 examines the impact of propensity to speculate on the price delay and re-examines our results with a random sample. Section 6 concludes.

Section snippets

Development of research hypotheses on price delay and lottery-type stocks

There is a large literature which identifies a variety of market frictions that impede the process of stock price incorporating information, including asymmetric information, taxes, liquidity and noise trader risk, the number of analysts following, institutional ownership, trading volume, and short sales restrictions/costs (e.g., Brennan (1970), Miller (1977), Brennan, Jegadeesh, and Swaminathan (1993), Barberis et al. (1998), Chordia and Swaminathan (2000), Easley et al. (2002), Pastor and

Data

This study assembles a sample of emerging markets based on those defined by Morgan Stanley Capital International and those considered in the literature (Morck, Yeung, and Yu (2000), Bekaert and Harvey (2000)). Those markets are selected as the sample primarily based on data availability, including the sufficiency of firm-level data provided by Datastream and Worldscope. Our sample thus covers 32 emerging markets from Europe, America, Africa, and Asia. Hong Kong and Singapore are included as

Proxy measures for lottery-type stocks (MAX) and idiosyncratic volatilities (IVOL)

Kumar, 2009 and Kumar et al. (2011) define lottery-type stocks as stocks that pay low while having very low probability of very high payoff. Bali et al. (2011) propose a direct measure, MAX, which assesses the magnitude of prior highest return of a stock arising from investors' demand for lottery-type payoffs. Stocks with high MAX, i.e., extreme positive payoffs, are characterized as lottery-type stocks. We follow Bali et al. (2011) and define MAXi,m as the maximum daily return for stock i

Price delay and investor propensity to speculate

The regional analysis suggests that the speed of adjustment toward local and global market information depends on both firm-specific factors and country-specific factors. One of the primary interests of this study is to examine whether an emerging market breeds investor clienteles exhibiting gambling attitudes, which is revealed by the lottery-stock premium, and their speculative trading behavior posits a further barrier to timely reflection of common information in stock prices. We explore the

Conclusion

Investors may allocate funds to stocks with speculative characteristics, such as lottery-type payoffs or high idiosyncratic volatility, making their portfolios deviate from being well-diversified. A strong presence of such portfolio choice will delay the information efficiency of stock prices. Emerging markets, with a history of high volatility and high skewness in stock returns and investor clienteles with socioeconomic characteristics inclined to show a strong preference for lottery payoffs,

Declaration of Competing Interest

Chin-Wen Hsin: I have nothing to disclose.

Shu-Cing Peng: I have nothing to disclose.

Acknowledgments

Chin-Wen Hsin is from the College of Management, Yuan Ze University, 135 Yuan Tung Road, Taoyuan, Taiwan (e-mail: [email protected]) and Shu-Cing Peng is from the Department of Finance, National Central University, Taiwan (e-mail: [email protected]). We are grateful for helpful comments and suggestions from K.C. John Wei and seminar participants at National Chengchi University and the 2016 Financial Management Association Annual Meeting (FMA). Hsin thanks the Ministry of Science

References (78)

  • S.S. Deb et al.

    Are price limits really bad for equity markets?

    Journal of Banking & Finance

    (2010)
  • D.W. Diamond et al.

    Constraints on short-selling and asset price adjustment to private information

    Journal of Financial Economics

    (1987)
  • J. Egginton et al.

    The robust “maximum daily return effect as demand for lottery” and “idiosyncratic volatility puzzle”

    Journal of Empirical Finance

    (2018)
  • W.M. Fong et al.

    Investor sentiment and the MAX effect

    Journal of Banking & Finance

    (2014)
  • M. Gu et al.

    The role of analysts: An examination of the idiosyncratic volatility anomaly in the Chinese stock market

    Journal of Empirical Finance

    (2019)
  • K. Hou et al.

    Have we solved the idiosyncratic volatility puzzle?

    Journal of Financial Economics

    (2016)
  • W. Hung et al.

    The MAX effect: Lottery stocks with price limits and limits to arbitrage

    Journal of Financial Markets

    (2018)
  • A. Kumar

    Do the diversification choices of individual investors influence stock returns?

    Journal of Financial Markets

    (2007)
  • A. Kumar et al.

    Religious beliefs, gambling attitudes, and financial market outcomes

    Journal of Financial Economics

    (2011)
  • R. Morck et al.

    The information content of stock markets: Why do emerging markets have synchronous stock price movements?

    Journal of Financial Economics

    (2000)
  • G.V. Nartea et al.

    Do extreme returns matter in emerging markets? Evidence from the Chinese stock market

    Journal of Banking and Finance

    (2017)
  • M. Scholes et al.

    Estimating betas from nonsynchronous data

    Journal of Financial Economics

    (1977)
  • M. Seasholes et al.

    Predictable behavior, profits, and attention

    Journal of Empirical Finance

    (2007)
  • M. Seif et al.

    Limits to arbitrage and the MAX anomaly in advanced emerging markets

    Emerging Markets Review

    (2018)
  • C. Walkshäusl

    The MAX effect: European evidence

    Journal of Banking & Finance

    (2014)
  • H. Wang et al.

    Reference-dependent preferences and the risk–return trade-off

    Journal of Financial Economics

    (2017)
  • A. Watanabe et al.

    The asset growth effect: Insights from international equity markets

    Journal of Financial Economics

    (2013)
  • S. Yao et al.

    Idiosyncratic skewness, gambling preference, and cross-section of stock returns: Evidence from China

    Pacific-Basin Finance Journal

    (2019)
  • A. Zhong et al.

    The MAX effect: An exploration of risk and mispricing explanations

    Journal of Banking & Finance

    (2016)
  • A. Ang et al.

    The cross-section of volatility and expected returns

    Journal of Finance

    (2006)
  • M. Baker et al.

    Benchmarks as limits to arbitrage: Understanding the low-volatility anomaly

    Financial Analysts Journal

    (2011)
  • T. Bali et al.

    A lottery-demand-based explanation of the beta anomaly

    Journal of Financial and Quantitative Analysis

    (2017)
  • B. Barber et al.

    Trading is hazardous to your wealth: The common stock investment performance of individual investors

    Journal of Finance

    (2000)
  • N. Barberis et al.

    Stocks as lotteries: The implications of probability weighting for security prices

    American Economic Review

    (2008)
  • S. Bartram et al.

    Why are US stocks more volatile?

    Journal of Finance

    (2012)
  • G. Bekaert et al.

    Foreign speculators and emerging equity markets

    Journal of Finance

    (2000)
  • K. Bergsma et al.

    Short interest and lottery stocks

    Financial Management

    (2019)
  • M. Blume et al.

    The asset structure of individual portfolios and some implications for utility functions

    Journal of Finance

    (1975)
  • E. Boehmer et al.

    Short selling and the price discovery process

    Review of Financial Studies

    (2013)
  • Cited by (0)

    View full text