Genes, History and Economics

Abstract: 1. Introduction This dissertation consists of six chapters that span a very diverse set of topics. Yet, it has two unifying themes, economics and biology, that tie it together. The first four chapters present the principal findings from a project that was initiated jointly with David Cesarini and Magnus Johannesson, and that applies the twin method from behavioral genetics to economics. The last two chapters instead use a simple regression framework and evidence from biological anthropology to investigate recent claims regarding the effects of child bearing and past slave trades.2. Genes and economics There is a small, but rapidly growing, literature studying the genetic and environmental origins of economic behavior and outcomes (Bowles et al., 2005; Beauchamp et al., 2011). Until recently, this literature focused exclusively on outcomes, and in particular income. In chapters 1-4 we instead focus on economic behavior and decision-making.Previous behavioral genetic work outside the domains of economics has changed the way that we think about a number of behavioral traits. In this literature it is typically found that i) variation is heritable ii) genetic factors are more important than family environment iii) a large fraction of variation cannot be explained by neither genes nor family environment (Turkheimer, 2000; Plomin et al., 2009). However, compared to many other disciplines, and psychology in particular, economics is lagging behind. In fact, as recently as 2009 the leading text book in behavioral genetics described economics as "still essentially untouched by genetic research" (Plomin et al., 2009, p. 353). Hopefully, the chapters in this dissertation can help to improve on this somewhat unsatisfactory state of the art.Chapters 1 and 2 study economic decision-making in the laboratory using the twin method. More specifically, we study the ultimatum and dictator games alongside risky gambles, using same-sex twin pairs as our subject pool. Given a few additional assumptions, the fact that identical twins have, in expectation, a twice as high coefficient of genetic relatedness as fraternal twins implies that we can study the genetic and environmental contributions to variation in behavior by studying twin correlations in observed choices. Chapters 3 and 4 apply the same method to actual portfolio choices associated with a far-reaching pension reform, as well as to a set of standard behavioral anomalies. Taken together, these four chapters provide strong evidence in favor of the hypothesis that genes influence economic decision-making. Thus, economic behavior does not appear to be much different from other types of behavior.3. Economics and history The last two chapters of the dissertation turn to the past, rather than genes, in an effort to evaluate recent findings regarding two important welfare outcomes. In chapter 5 we investigate Nunn’s (2008) claim that past slave trades had a negative impact on current economic performance in Africa. By extending the sample period back in time we demonstrate that this relationship was not significant in 1960. In addition, by applying Nunn’s method to an episode of large scale slave raiding in Italy, we demonstrate that there exists a similar negative relationship across Italian regions, although it becomes insignificant when geographical controls are included. Intriguingly, going back to 1960, the coefficient on slave raids for Italy also has a similar time trend to that for Africa. Taking these facts, and our reading of the historical and anthropological literature, which is much different from that of Nunn, into account we do not find much support for the hypothesis that the African slave trades had a negative impact on current economic performance.Finally, chapter 6 investigates the large and negative relationship between giving birth to a son, rather than a daughter, and maternal longevity that was documented in a Sami hunter-gatherer population from Finland (Helle et al., 2002). Using a substantially larger sample of pre-industrial Swedish Sami we find no evidence in favor of such a relationship.4. Brasklapp Five of the chapters in this dissertation (Ch. 1-4 & 6) are slightly altered versions of previously published papers (Wallace et al., 2007; Cesarini et al., 2009 a, b; 2010; 2011). Unfortunately, the fact that earlier versions of the chapters were prepared as separate articles for five different journals means that they can at times appear both repetitive, and in terms of notation and formatting, somewhat inconsistent. I apologize to the reader for these inconveniences.