Section 4 – Empirical Studies and Field Experiments

In Sections 1–3, we learned something about how the field of behavioral economics has evolved: from laboratory experiments undertaken with individuals and games played by groups of individuals to revised theories that canonize the specific ways in which Homo sapiens’ choice behavior deviates from the rational-choice model of Homo economicus. In Section 4 we consider novel empirical studies and field experiments that have been conducted in a myriad of settings, some of which test (with larger samples of Homo sapiens) the main tenets of behavioral economic theory. In particular, they test our penchants for loss aversion, reference dependence, time inconsistency, and generally our reactions to how choice situations are framed (e.g., narrowly versus broadly). These studies seek answers to the same types of questions the laboratory experiments and games of the previous two sections posed to economists and psychologists; how exactly do Homo sapiens’ choice behaviors deviate from those of Homo economicus in a variety of real-life situations that often have pronounced social consequences? What might explain these deviations?

Other empirical studies conducted by behavioral economists test for evidence of sub-optimal behavior among Homo sapiens that expresses itself as discrimination, racial bias, corruption, crime, waste, homelessness, drunk driving…I think you get the point. These behaviors manifest themselves as social ills. Less severe forms of sub-optimal behavior (e.g., underperforming students and teachers, insufficient savings and investment by households and businesses, tax evasion, and low voter turnout) have also received attention from behavioral economists. These types of behaviors lead scholars and policymakers to ask what social constructs (or, what Thaler and Sunstein (2009) call choice architecture) might be developed to mitigate the negative social consequences associated with these types of sub-optimal behavior.

Referring to the diagram presented in the This Book’s Approach section, Section 4 pertains to the diagram’s middle portion, where we investigate empirical studies that have been undertaken in the real world and the corresponding choice architecture that has been developed—studies and architecture spawned by the now famous experiments, games, and concomitant theory of behavioral economics.

For the most part, these empirical studies are based not on experiments run in laboratories but rather on experiments run in the “fields” of our daily lives. We call them “natural experiments” or “field experiments.” The findings we obtain from these studies can inform public policy and point us (or, as Thaler and Sunstein would say, nudge us) toward crafting more enlightened policies that reflect and channel our idiosyncrasies, biases, foibles, and social norms (i.e., our humanity) for improved social outcomes. As you proceed through what is essentially a menagerie of different behavioral economics case studies, bear in mind that one of the goals of this section is to introduce you to research methods as well as attendant empirical results. Along the way, you will be introduced to the myriad approaches researchers use to present their statistical results.

Note that, unlike a normal textbook chapter, this section is not necessarily meant to be read from beginning to end. There are too many disparate studies included herein, with no particular theme to tie them all together. Rather, only those studies deemed relevant for your course—whether this relevance is predetermined by the course’s instructor or the student’s personal judgment—are meant to be chosen from the wide universe of studies compiled below, and conceivably linked with the concepts and topics from Sections 1 – 3 that were previously covered in the course. The Linkages Matrix contained in Appendix E can help you with making these linkages.