臺大管理論叢第31卷第3期

140 The Study of Gambling Preference, Trading Pattern and Excess Comovement in Lottery-Like Stock Returns to its 52-week high). What we reveal can help determine the common factors that exhibit systemic effects on the return comovement of lottery-like stocks and identify situations in which the value of lottery-like stocks is more likely to be overestimated. 5. Originality/Contribution The contributions of this study are as follows. First, Kumar, Page, and Spalt (2016) employ the ratio of Catholics to Protestants in the local population (CPRATIO) around a company’s headquarters to determine the overall gambling propensity of investors. However, this inference is made under the assumption of local bias. By contrast, our study explores the relative importance of investor types in the excess return comovement of lottery-like stocks based on order submission behavior. Therefore, this study provides more direct evidence on the relationship between gambling preference and the excess return comovement of lottery-like stocks. In addition, this study uses retail investors’ order flow of lottery-like stocks to measure the stock-level gambling trading, but the factor does not reveal any relationship between return comovement and stock issuers’ expenditure on R&D or the quality of financial statements. This finding is inconsistent with previous studies, indicating that CPRATIO represents only the gambling culture of certain areas although it systemically affects residents and corporate culture, making CPRATIO an indirect indicator of the overall gambling propensity of investors and local companies’ willingness to tolerate risk and make gamble-like decisions. Conversely, the variables used in our study may not have a direct connection to companies’ gambling culture and thus do not affect their gamble-like decisions as seen in previous studies. Second, studies have identified factors that contribute to institutional investors’ preference for lottery-like stocks (e.g., Agarwal, Jiang, and Wen, 2020). This study goes further and is the first to reveal institutional investors’ correlated order flow of lotterylike stocks. These institutional investors may split orders for stealth trading, withholding information, or reducing price impact. Their behaviors explain the lack of a direct effect on the excess return comovement of lottery-like stocks, despite the correlation among foreign institutional investors’ trading decisions regarding different stocks. These findings contribute to the literature on the role of institutional investors in the trading of lottery-like stocks. Finally, this study is the first to investigate the effect of stock-level gambling sentiment on the excess return comovement of lottery-like stocks. Past studies have

RkJQdWJsaXNoZXIy ODg3MDU=