Article Preview
Top1. Introduction
Traditional models in the finance literature are based on a premise of economic rationality, which argues that each individual processes all available information, filter the information according to the set of rational preferences and maximize utility. However, there is a growing literature showing that individuals may exhibit less than fully rational behavior. Individuals estimate probabilities poorly, have narrow confidence intervals, overstate their abilities, are reluctant to realize their losses and trade too much.1
The results seem to be no different for professionals. Professional managers perceive their judgment to be too exact, sell winners quicker than losers, invest in familiar stocks, and have narrow confidence intevals.2 However, studies mostly focus on top level managers such as CEOs or on finance professionals. Yet, we do not know much about whether lower level managers or professionals in other industries like the pharmaceutical industry are affected by the irrational decision making process. It may be likely that because of the nature of their responsibilities and actions, finance professionals and top level managers are more subject to irrational behavior due to, for example, the self-attribution bias which is mainly attributing successful outcomes to own skill and blaming unsuccessful outcomes on bad luck.3 Therefore, it is worth examining the possible biases, if any, of managers outside of the finance industry and also of lower levels as these lower level managers are important components of the daily running of the firm. Moreover, it would be interesting to look at the responses of pharma professionals who are functional in the health sector to specific questions such as the ones on diseases and options for treatment for such diseases.
In this study, we use the a survey methodology that follows Tversky and Kahneman (1979) Prospect Theory and Tversky and Kahneman (1992) Cumulative Prospect Theory which may help us explain the decision patterns of the respondents. However, our survey differs from these studies in the sense that we ask the survey questions to the same group of respondents rather than different groups in order to allow for the learning effect. Learning effect may result in the respondent realizing that the sets of questions that are framed differently are essentially the same and enable them to change their answers accordingly which may end up in less biases observed.
There are a total of 80 responses to our survey who are the employees of a multi-national pharmaceutical company and our findings show that prospect theory is able to predict the actions of the professionals in general. However, our findings do not provide evidence for the framing of outcomes which may be due to the learning effect. Tversky and Kahneman (1981) ask the question set to two different groups of respondents in their original study, whereas we ask the questions to the same group who can see both versions of the question and may change their answers retrospectively. This learning effect may be a result of the structure of the questions asked. The questions for framing of outcomes are shorter and easier to spot compared to the other question pairs or trios such as preference reversals or framing of contingencies. Therefore, the respondents can realize that the questions are essentially the same and change answers. Finally, our findings show that demographics, experience and field of work influence respondents' answers in general.
This study complements the behavioral literature by examining the decision making patterns professionals outside of the finance industry, namely the pharmaceutical industry, and of lower levels in order to shed light on to what extent and how these professionals are affected by behavioral biases. Moreover, it tests for the learning effect by asking the questions that are framed differently to the same groups of respondents.
The paper is organized as follows. Section 2 presents the related literature and hypotheses. Section 3 describes the data. Section 4 presents the empirical results. We present our conclusions in Section 5.