The Book’s Genesis and Tips on How to Use It

This textbook evolved during the summer of 2019. In February of that year, I accepted a Fulbright Specialist position to teach an intensive short-course in behavioral economics to the faculty at Meiktila University of Economics in the southeast Asian nation of Myanmar. In the project description, Dr. Thida Kyu (PhD economist and Pro-Rector of the university), explained that because it had been cut off from the Western world for so long (roughly 50 years) and had only recently attained (nominal) civilian control of the nation’s government in 2016, Myanmar had a lot of catching up to do, particularly regarding the functioning of its academic institutions. Dr. Kyu was aware of this new field called behavioral economics. She believed its lessons would not only enlighten her faculty and their students, but might also help nudge her country’s fight against poverty onto a more enlightened path policy-wise. I took this to mean, rightly or wrongly, that Dr. Kyu was not looking for another lecture-orientated course, a mere overview of the history, methodologies, and findings of behavioral economics. Rather, her faculty needed a practitioner’s guide, a course that would, as much as possible, engage them with the field’s methodologies and findings through actual practice and firsthand experience—a course that would get them in on the proverbial ground floor of this relatively new field of inquiry.

I began my preparations for the course by doing what I always do when assigned to teach a new course. I sought out existing textbooks. Over the course of my career, I’ve been fortunate to have a wide variety of textbook selections for the fields of environmental and resource economics and microeconomic theory. But not this time. It became apparent almost immediately that if I were to prepare a course geared more toward the practice of behavioral economics, I would need to cobble together material from a host of disparate sources. The book you now hold is the result of this ‘cobbling’ process. It melds Kahneman’s, Tversky’s, and Thaler’s seminal works (along with several other key theoretical and experimental advancements published in a wide variety of journals over the past 50-plus years) with Camerer’s (2003) behavioral game theory text and William Spaniel’s (2011) introductory textbook on analytical game theory. The book also draws from Kahneman’s (2011), Ariely’s (2008), and Thaler and Sunstein’s (2009) New York Times bestsellers Thinking, Fast and Slow, Predictably Irrational, and Nudge, respectively, and to lesser extents, from Levitt and Dubner’s (2005), Gladwell’s (2002), and Harford’s bestsellers Freakonomics, The Tipping Point, and Messy.

In addition to the value-added that comes from having incorporated these works into a single text—in some cases, rendering explicit representations of experiments the authors have merely mentioned and in other cases drawing directly from the original sources cited by the authors—I have included material from works that I consider to be worthy representations of the breadth of behavioral economics as a field of inquiry. In the end, we have before us a book that guides the student through this field no differently than a well-researched guidebook helps the intrepid international traveler navigate a foreign country’s main attractions, and helps the traveler gain knowledge of (and hopefully appreciation for) the country’s history and cultural uniqueness.

As such, this book is not necessarily meant to be read by students from cover-to-cover in chronological order (i.e., first covering the material in Section 1, then the material in Section 2, and so on). Rather, it is possible that what works best for your students is for them to be introduced to the material in a piecemeal fashion. For example, when I taught the course in Myanmar, I included in each three-hour lecture an experiment or two from Section 1 coupled with some of the economic theory presented in Section 2, and either a game from Section 3 or a discussion of empirical research or choice architecture from Section 4. This helped the students engage with each of these facets of behavioral economics for the duration of the course. It also precluded me from front-loading the often fun-filled experiments and games, and leaving Section 2 and 4’s more lecture-orientated (dare I say less-entertaining?) discussions of the theory and empirical research and choice architecture for the last few weeks of the course.

Perhaps most importantly, drawing from more than one section of the book in each lecture facilitates the connecting of an outcome from a Section 1 thought experiment to a laboratory experiment (and an associated, revised economic theory) in Section 2, or connecting an outcome from a Section 3 game to a corresponding result from a Section 4 empirical study. Indeed, the gamut of potential connections that can be made across the topics presented in the different sections of the book is almost limitless. Since economists tend to deal better with finiteness than infiniteness, Appendix E provides what I call a “linkages matrix,” which, provides a structure for identifying connections between the various concepts presented in Chapters 1 – 4 and the experiments, games, and empirical studies discussed in Chapter 6 and Section 4. This matrix is meant to serve as an aid for instructors who adopt this type of piecemeal approach to teaching the course.

For example, one of the thought experiments presented in Section 1 exemplifies what Kahneman (2011) and Kahneman and Tversky (1984) originally labeled a “framing effect,” which in turn can lead to a host of biases in choice behavior, such as confirmation bias and representative bias.^[1] Accordingly, in Section 3 we could discuss results from a field experiment that shows how framing the Ultimatum Bargaining game as a “seller-buyer exchange” encourages self-interest (i.e., behavior expected from Homo economicus), while framing the game as a common-pool resource encourages Homo sapiens-like generosity. Empirical research presented in Section 4 demonstrating “loss aversion” on the part of public school teachers in Chicago can also be considered an example of a framing effect as first introduced in Section 1 since the timing (i.e., framing) of bonus payments made to teachers based on improved student performance is the mechanism eliciting the loss-averse behavior. Linkages like these abound across the four sections.

If instead of adopting the piecemeal approach to teaching the course, the instructor prefers a more traditional, chronological approach to presenting the material as laid out in Sections 1 – 4, the annotated course outline provided in Appendix D offer guidance. One outline is designed for a course targeting economics majors, the other for a course targeting non-majors. The main difference between the two outlines is that the former allocates more time to the economic concepts and theories presented in Chapters 3 and 4, while the latter emphasizes the material covering human quirks (e.g., heuristics, biases, and effects presented in Chapters 1 and 2). As the course outlines for both types of courses indicate, the instructor chooses the specific effects, biases, theoretical material, experiments, games, and empirical studies that will be covered in lectures.

In concert with the course outline, figuring out how best to grade students in a course like this can be a challenge, particularly if you decide to administer a demographic survey (Appendix B) on the first day of class. In this case, preserving student anonymity becomes an issue. To deal with this issue, consider creating two separate spreadsheets for the course. One spreadsheet compiles the students’ survey responses and outcomes from the experiments and games. This spreadsheet is linked to the students’ randomly assigned course ID (CID) numbers. The other spreadsheet, which is linked to their university student ID numbers and their names, compiles their performances on quizzes, homework, and exams assigned throughout the semester.

At the risk of sounding draconian, this is a course where it may make sense to base upwards of 50% of a student’s grade upon their in-person attendance, which would entail carefully taking role at the beginning of each class. If the class meets 30 times face-to-face during the semester, for example, their grade attributable to attendance would then drop by 3.33 percentage points for each missed class (excused absences withstanding). Granted, students who foresee having difficulty attending class in-person throughout the semester would likely choose to drop the course immediately. For those students who remain, the remaining 50% of their course grade would then be based upon their quizzes, homework, and exam scores.

The issue of how best to convey written information to the student a priori (i.e., before conducting a given experiment or game) also looms large in a participatory-learning setting such as this, especially if the instructor desires to obtain unbiased responses from the students (or more practically, to control for potential biases). For example, the first set of thought experiments presented in Section 1 is meant to demonstrate firsthand to the students the extent to which automatic, knee-jerk responses from what Kahneman (2011) identifies as the System 1 portion of the brain can result in miscalculations. Students who choose to read ahead (small in number though these types of students may be) potentially skew the distribution of responses away from its otherwise true representation of these miscalculations. Such skewness may be tolerable for strictly educational purposes, where the goal is to demonstrate that at least a certain percentage of students are prone to miscalculation. But if the instructor also hopes to compile student responses into a dataset amenable for statistical analysis, then this type of potential bias draws into question the validity of the data.^[2]

To help control for potential biases associated with students having read ahead about the game or experiment they are now participating in, I recommend including the following question on each Response Card: “Did you read about this topic ahead of time?” (see Appendix A). Answers to this question provide a control for the level of student foreknowledge, which is the potential bias of concern.

I am personally unaware of any studies that have looked at how well students learn the lessons of behavioral economics in a cumulative sense over a span of time (e.g., an entire semester) and across a variety of experiments and games. In other words, I know of no studies that estimate the extent to which individuals who begin a course in behavioral economics as bona fide Homo sapiens evolve toward “Homo economism” in their individual and social choices. The pedagogy promoted in this textbook—in particular, the data it generates—offers instructors the opportunity to empirically test the hypothesis that students make this evolution.