![Sample our Behavioral Sciences journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/110801/TOC-Subject-Sample-2021-Behavioral-Sciences.jpg"})
Abstract
The literature has established that retail investors are “net buyers” of attention-grabbing stocks. In this study, the authors utilize a unique dataset of actual information about 290,000 household investment accounts and track their “net buying” decisions with a focus on their economic and demographic characteristics. Unlike previous research, the authors focus not only on net buyers of attention-grabbing stocks, but also on net sellers of such stocks. They find that factors such as financial experience, wealth, consulting with advisors, and other individual characteristics, indicative of investors’ sophistication, account for the differences in the net buying decision. Specifically, the authors find that more trading experience and a lower tendency for home bias are associated with net selling during months when stocks attract a great deal of attention, and with net buying during months when they are paid less attention. The authors document that investors who trade in months of less attention are more experienced, engage more in complex trading, have less of a home bias tendency, are wealthier, and have a higher income than those who trade during the highest attention-grabbing months. Finally, the use of financial advice varies not only between households, but also between months in which stocks receive a great deal or little attention.
Notes
1 Retail investors behave differently from institutional investors. Specifically, in contrast to individuals who exhibit traits such as attention-based buying, institutional investors trade less aggressively during attention-grabbing months, and are net buyers on days of abnormally low trading volume (e.g., Barber and Odean 2007; Ruan and Zhang Citation2016).
2 As of January 2018, the share of Google’s search engine in the search market is about 87% (e.g., https://www.statista.com/statistics/216573/worldwide-market-share-of-search-engines/).
3 Our sample includes details about employment status and the professions of individual investors. Individuals with professional occupations that imply higher levels of education are classified as sophisticated investors and those with nonprofessional occupations are assumed to be unsophisticated (Dhar and Zhu Citation2006; Haziza and Kalay Citation2014).
4 The occupations considered professional in this study include the following: engineer, architect, business administrator, microbiologist, doctor, economist, accountant, agronomist, banker, lawyer, reporter, organizational consultant, pilot, tax advisor, judge, psychologist, architect, biologist, academic staff member, lecturer, computer programmer, social worker pharmacist, advertiser, land assessor, portfolio manager, investment advisor, educational counselor, director, and credit counselor.
5 We use the Google SVI to capture attention. More details appear in the Data section.
6 We include the VIX in the model because previous research regards the VIX as an indicator of sentiment (e.g., Baker and Wurgler Citation2007), and documents the negative relationship between investor sentiment and future equity returns (e.g., Fisher and Statman Citation2000; Fernandes et al. Citation2014). In Bekaert et al. (Citation2011), as well as in Silvennoinen and Thorp (Citation2013), the VIX is regarded either as a proxy for risk aversion or worldwide investor sentiment. In addition, researchers have found that an increase in the VIX increases co-movements in the OECD markets (e.g., Cai et al. Citation2009; Min and Hwang Citation2012). Therefore, VIX may be indicative about the effect of investor sentiment on investors’ buy and sell decisions.
7 The Tel Aviv Stock Exchange does not use ticker symbol for stocks. Therefore, Israeli investors are likely to search for a firm’s name to find stock-specific information on Google.
10 Our sample also includes an additional 16,630 accounts that are marked as managed by professional advisors separately. We duplicated the estimation presented in Table 1 with respect to these accounts, and estimated the correlation among Google SVI, trading volume, and buy-sell imbalances in these managed accounts. The earlier literature has established that compared with the buying behavior of retail investors, the buying behavior of institutional investors' is less influenced by attention (Barber and Odean 2007). Our results are in line with these findings. We find that for accounts managed by portfolio managers the correlation between attention and net buying holds in the weak sense only. To wit, while for retail investors the correlation among Google SVI, trading volume, and buy-sell imbalances is positive and significant for almost all of our sample stocks; in contrast, for managed investors' accounts, we do not find such a significant correlation.
11 We exclude accounts worth more than 100 million NIS (worth about 25 American cents each at the time of writing), because portfolios of wealthy investors are usually professionally managed. Therefore, we would expect them to be traded more like institutional portfolios than like the portfolios of retail individual investors.
12 The average exchange rate for the sample period is $1 = 3.96 NIS.
13 The notation “M” denotes millions. Since in this study millions is associated only with the monetary sum, for simplicity we omitted the notation of the domestic currency (NIS).
14 Jianakoplos and Bernasek (Citation1998), Sunden and Surette (Citation1998), and Barber and Odean (Citation2001) reported different investment behaviors among men and women depending on marital status.
15 Men and male as well as women and female are used interchangeably in the article.
16 We label only those investors who engaged in one of these actions at least 10 times during the sample period as sophisticated because we want to include only those investors who are willing to participate in such complex trading on a regular basis.