How much is enough? An investigation of nonprofessional investors information search and stopping rule use

Highlights

• We investigate the stopping rules used by nonprofessional investors.

• Stopping rules determine when the information gathered is sufficient to terminate search.

• Stopping rule use affects the amount and type of information gathered.

• Investors include minimal financial information in their search.

Abstract

Regulators are concerned that the information overload in the current Internet-based disclosure environment may cause investors to overlook important information. To gain a better understanding of the information set gathered by investors, this study incorporates theories from information systems research to examine the cognitive stopping rules used by investors to terminate information search. We survey nonprofessional investors to gain insight into what information they gather and when they determine they have enough information to stop searching and make an investment decision. Demographic analysis shows that investor characteristics are associated with the particular stopping rule used. In addition, results show that the stopping rule used affects the amount and type of information gathered. We find that, in general, investors include very little financial information in their search, and the amount gathered depends on the stopping rule employed. Our results call into question the decision usefulness of accounting information for nonprofessional investors and should be of interest to accounting information systems researchers, regulators, and accounting practitioners.

Introduction

The Securities and Exchange Commission (SEC) (2012) considers information search a critical part of the investor's due diligence process. Investor demand for easier and timelier access to information has led to a significant increase in the quantity of Internet-based information (e.g., investor news, company websites, and social media platforms) available for investors to search. An unintended consequence, however, is that nonprofessional investors¹ are overloaded with information. The SEC is concerned with information overload and warns of the potential negative capital market effects if investors “overlook or do not take the time to study valuable information because there is simply too much information to try to engage it constructively” (Paredes, 2013). SEC Chairman Mary Jo White called for disclosure reform to address the information overload problem and investigation into what information investors find necessary and what information investors do not want (White, 2013). In other words, the actual information that investors find useful to the investment decision making process is an unanswered empirical question. To shed light on the issue, this study investigates investors' information search processes.

The search process is key to understanding what information investors consider decision useful.² Any decision making task is preceded by a search for information that the individual considers useful. In situations of information overload, such as an investment task, a plethora of information is available which must be reduced, or filtered, in order to be used effectively. The process used to filter the information will likely have a significant impact on the information set gathered, the information set ignored, and, ultimately, investment-related judgments and decisions. We rely on findings from information systems research and examine the filtering mechanism used by investors that leads to information search termination (i.e., the stopping rule) and the end result—the information set gathered.

Academic research regarding investors' information search strategies is scant and findings are limited by the methodologies employed. For example, archival studies offer evidence of aggregate investor information search volume via EDGAR and Google around corporate event dates (e.g., Drake et al., 2012, Drake et al., 2014), but archival methods cannot identify individual decision processes or investor characteristics that influence search. The most common survey method is to provide investors lists of information types and sources to evaluate (e.g., Elliott et al., 2008), and experimental studies (e.g., Cianci and Falsetta, 2008, Rose et al., 2010, Arnold et al., 2011) vary specific information cues presented to participants in order to isolate variables of interest that impact search.³ However, the information sets used in these studies are determined by the researchers a priori and it is not clear if investors would include the same information in a more natural (i.e., not primed) setting.

Information processing theory suggests that investors may not be able to consume the amount of information provided in prior studies. Specifically, individuals' cognitive capacities are limited and they often use heuristics, or decision rules, to lessen the cognitive burden of decision making tasks (Miller, 1956, Chewning and Harrell, 1990, Stock and Harrell, 1995, Tuttle and Burton, 1999, Pennington and Tuttle, 2007). However, prior survey research often uses long lists of items which far exceed an individual's processing capacity. Thus, it is unlikely that investors have the capacity to use all the information provided in these studies. We posit that investors' information search will reveal a more limited set of information that the investors find useful to the investment decision.

To manage information search, eventually the investor must stop the search. Information systems research defines stopping rules as heuristics individuals use to determine when the information gathered is sufficient to terminate information search and move on to the decision making stage of the task (Nickles et al., 1995, Pitts and Browne, 2004, Browne and Pitts, 2004). The use of a stopping rule involves determining sufficiency through a comparison of the information gathered to some pre-defined criteria. When the criteria of the stopping rules are met, the search for information stops. Prior research does not consider what stopping rule investors are likely to use. We suggest investors will determine sufficiency of information by using the criteria of one of the following stopping rules: absolute standard, differences, or single criterion. For an absolute standard stopping rule, the criterion is a certain amount of information and/or a pre-determined list of information items. If investors use an absolute standard rule, they would search until the pre-determined amount of information or list of information items is attained. The criterion for a differences stopping rule is the incremental value of an additional information cue to the overall view of the information set. If investors use a differences rule, they would search until the overall representation (or mental model) of the company stops shifting and stabilizes. The single criterion stopping rule considers a single information item (e.g., stock price history) and/or related facts about that item as sufficient information. If investors use a single criterion rule, they would search until the information on that criterion or characteristic is attained (see Table 1).

Prior information systems research finds that characteristics of the decision maker (e.g., experience) and of the task influence stopping rule use and that the specific stopping rule used affects the amount and type of information gathered (Pitts and Browne, 2004, Browne and Pitts, 2004, Browne et al., 2007). However, it is unclear whether results from prior research will generalize to an investment decision making task as the investment task is inherently different than the tasks used in prior research. For example, investors often conduct information search online and have immediate access to a vast amount of information from multiple sources (Kelton and Yang, 2008). This unlimited amount of information makes it possible for investors to gather similar quantities of information regardless of the stopping rule used. For example, a single criterion investor may gather a similar quantity of information cues about a specific investment characteristic (e.g., revenues) as an absolute standard investor with a pre-determined list of information requirements. In addition, the stock investment task has a relatively higher degree of risk, as compared to tasks used in prior research, which may result in different stopping rules being used and differing effects on information search outcomes (Browne et al., 2007). Investors are typically loss averse (Tversky and Kahneman, 1991) and respond to increased risk by collecting more information during search (e.g., Blay et al., 2012). It is thus unclear whether and how stopping rule use will impact investor information search.

We draw from research in information systems and psychology to investigate the association between stopping rule use and the quantity and type of information investors gather. The theory of representational congruence (Chandra and Krovi, 1999) predicts that in order to reduce cognitive effort and improve decision making, individuals will seek out external information that “fits” their decision strategy or internal problem representation (i.e., representational congruence). In the context of investor information search, we suggest that investors will gather information (quantity and type) that best fits the criteria required by the stopping rule. An absolute standard rule uses a reference point of a threshold of necessary information or particular information items and thus focuses search more on the quantity of information than a differences rule or single criterion rule. Thus, we predict that absolute standard investors will gather more information during search than those who use a differences rule or a single criterion rule. In addition, a differences rule uses a reference point of the incremental value of an additional information cue to the overall representation (or mental model) of the company. Thus, we expect investors using a differences rule to gather less detailed information that has already been analyzed as it better fits their development of an overall view of the company. Accordingly, we predict that absolute standard investors will gather more detailed information (i.e., financial) than differences investors. The amount of detailed information gathered by a single criterion investor will depend on the criterion. For example, an investor searching on a criterion of earnings per share will likely gather more detailed information than an investor searching on a criterion of corporate social responsibility. Thus, it is less clear whether single criterion investors will gather more or less detailed information than absolute standard and differences investors.

Our survey methodology differs from prior accounting research in that we gather open-ended survey responses about nonprofessional investors' information search and stopping behaviors. We used Qualtrics Panel, a third-party online survey administrator, to recruit nonprofessional investors with significant investing experience to participate in the study. Participants were instructed to assume they have available funds on hand to invest and are in the stage of gathering information about companies in order to ultimately make a stock purchase decision. Participants responded to four open-ended questions designed to elicit their information search processes. Independent research assistants analyzed participants' responses to identify the cognitive stopping rules used and the quantity and types of information participants reported gathering. Unique key words regarding information gathered were identified and counted in total for each participant. Key words pertaining to information derived directly from financial statements and/or ratios derived from financial statement items were counted as detailed financial information. By using open-ended survey questions, we overcome limitations in previous research by allowing the participants to respond without any prompting or priming to better determine the stopping rules used and the amount and type of information gathered as described by the participant.

Results show an association between the choice of stopping rule used and the total amount and type of information investors report gathering during information search. Specifically, investors using an absolute standard rule indicate gathering more information items and more detailed (i.e., financial) information than those using a differences rule. On average, investors describe gathering five pieces of information, which is consistent with findings from human information processing research that individuals are unable to process more than seven information cues at one time (Miller, 1956, Chewning and Harrell, 1990, Stock and Harrell, 1995, Tuttle and Burton, 1999, Pennington and Tuttle, 2007). Additionally, even though investors using an absolute standard rule report using the most financial information, the overall average is only one piece of financial information. Results also show that investor characteristics are important determinants of the choice of stopping rule. Specifically, older investors with greater risk tolerance are more likely to use absolute standard versus differences stopping rules, while investors with greater risk tolerance are more likely to use single criterion than differences stopping rules.

Our results have important implications for regulators and accounting practitioners as they seek ways to improve the usefulness of accounting information. The SEC provides tools for investor education including information about the importance of a 10K and how to use EDGAR to research investments (Securities and Exchange Commission, SEC, 2015). However, our results suggest that most of the investors in this study do not include any financial information in their information search, which is consistent with recent research questioning the decision usefulness of financial information (Young, 2006, Williams and Ravenscroft, 2015). We provide evidence that the amount of financial information gathered is dependent on the stopping rule used. Our results confirm regulators' concerns that the current information overload environment leads some investors to ignore certain information (Paredes, 2013) and should inform regulators as they consider ways to address and improve the disclosure environment.

The results of this study also have important implications for accounting information systems (AIS) research. Browne et al. (2007) call for investigation of stopping rule use in other tasks and for further understanding of the implications of such use. Our study and results respond to this call by extending information systems theory to the AIS literature and the investment decision making context. Our task is unique, as compared to tasks used in prior research, thus our findings contribute to the information systems research. In addition, our study adds to the growing body of accounting research on nonprofessional investor decision making. Archival research shows that investor information search, on an aggregate level (i.e., volume of search), is associated with specific characteristics of the firm, including corporate events and the strength of the information environment (Drake et al., 2014). Prior experimental and survey research identifies individual characteristics influencing investors' information search, such as investor affect (Blay et al., 2012) and experience (Hodge and Pronk, 2006, Elliott et al., 2008). We extend this literature with evidence of the stopping rules used by investors and how these rules influence investors' information sets.

The remainder of the paper is organized as follows. Section 2 discusses the prior research and presents the hypotheses development. We describe the research methods in Section 3 and our results in Section 4. Section 5 provides a discussion of the results and Section 6 discusses the implications and limitations of the study.