Economics of Culture and Institutions
April 5, 2014
Benjamin Enke and Anke Becker, Bonn Graduate School of Economics; Thomas Dohmen, Maastricht University; and Armin Falk, University of Bonn
Utilizing representative data on 80,000 individuals from 76 countries, Enke, Becker, Dohmen, and Falk study the worldwide distribution of risk preferences. While the age and gender composition of a population predict country-level risk attitudes, a large fraction of the substantial between-country variation remains unexplained by traditional sociodemographic, economic, and geographic variables. In contrast, ancient human migration patterns, as proxied for by aggregate data on human genetic variation, bear a strong association with the between-country heterogeneity in risk preferences. Specifically, the authors show that differences in risk attitudes between populations are strongly related to their Fst genetic distance. This correlation is not only robust across a wide range of regression specifications but also holds when using a measure of predicted genetic distance, which is explicitly constructed to reflect information about the migratory movements of our early ancestors. In addition, the authors provide evidence that the heterogeneity in risk attitudes within a population is related to the diversity of the respective genetic pool. These findings indicate that the specific ways in which mankind migrated out of Africa to populate our planet thousands of years ago have left a footprint in the contemporary cross-country distribution of one of the key economic traits.
Yuriy Gorodnichenko, University of California at Berkeley and NBER, and Gerard Roland, University of California at Berkeley
Gorodnichenko and Roland construct a model of revolution and transition to democracy under an individualist and a collectivist culture. They show that countries with a more individualistic culture, despite being potentially less able to overcome collective action problems, are more likely to adopt democracy faster than countries with a collectivist culture. Empirically, the authors show that there is a strong causal effect from individualistic culture to average polity scores, controlling for other determinants of democracy emphasized in the literature.
Davide Cantoni, University of Munich; Yuyu Chen, Beijing University; David Yufan Yang, Stanford University; Noam Yuchtman, University of California at Berkeley and NBER; and Jane Zhang, Hong Kong University of Science and Technology
Cantoni, Chen, Yang, Yuchtman, and Zhang study the causal effect of school curricula on students' stated beliefs and attitudes. They exploit a major textbook reform in China that was rolled out between 2004 and 2010 with the explicit intention of shaping youths' ideology. To measure its effect, the authors present evidence from a novel survey they conducted among 2,000 students at Peking University. The sharp, staggered introduction of the new curriculum allows the authors to identify the effects of the new educational content in a generalized difference-in-differences framework, taking into account provincial and cohort fixed effects. They examine government documents articulating desired consequences of the reform and identify changes in textbook content and in the college entrance exams that reflect the government's aims. The authors find that these changes were often effective: study under the new curriculum is associated with changed views on political participation and democracy in China, increased trust in government officials, and a more skeptical view of free markets. The results are robust to a wide range of controls and adjustments for multiple hypothesis testing, they are not driven by preexisting trends, and they receive additional support from a variety of placebo exercises.
Ruixue Jia, University of California at San Diego, and Torsten Persson, IIES and NBER
Jia and Persson provide a framework to link the ethnic choice for children with inter-ethnic marriage. Their model is constructed to be consistent with four motivating facts for ethnic choices in China, but it also delivers a rich set of auxiliary predictions. The empirical tests on Chinese microdata generally find support for these predictions. In particular, the authors provide evidence that social norms can crowd in or crowd out material benefits in ethnic choices. They also evaluate how sex ratios affect inter-ethnic marriage patterns and how their effects are strengthened or dampened by ethnic choices for children in mixed marriages.
Xi Chen, Yale University; Ravi Kanbur, Cornell University; and Xiaobo Zhang, International Food Policy Research Institute
It has been widely documented that the poor spend a significant proportion of their income on gifts even at the expense of basic consumption. Chen, Kanbur, and Zhang test three competing explanations of this phenomenonpeer effects, status concern, and risk poolingbased on a census-type primary household survey in three natural villages in rural China and on detailed household records of gifts received on important occasions. The authors show that gift-giving behavior is largely influenced by peers in reference groups. Status concern is another key motive for keeping up with the Joneses in gift giving. In particular, poor families with sons spend more on gift giving in proportion to their income than their rich counterparts, in response to the tightening marriage market. In contrast, risk pooling does not seem to be a key driver of the observed gift-giving patterns. However, the authors show that large windfalls in income trigger the escalation of competitive gift-giving behavior.
Marianna Belloc, Sapienza University of Rome; Francesco Drago, University of Naples Federico II & CSEF; and Roberto Galbiati, CNRS and Sciences Po
For a panel of about 60 Episcopal-see cities (governed by a bishop) over 300 years in medieval north-central Italy, Belloc, Drago, and Galbiati document that the occurrence of an earthquake retarded transition from feudal regime to commune. To provide an interpretation of their findings the authors offer a simple conceptual framework highlighting the basic tradeoffs involved in the process of institutional change when the incumbent political leader is also the religious authority, and derive a number of testable predictions. First, shocks heightening people's religiosity, such as seismic events in the Middle Ages, strengthen the status quo regime and as a consequence retard institutional change. Second, the stronger the shock, the larger is this effect. Third, the effect lasts only in the short run. The authors' interpretation is corroborated by historical evidence and by a number of additional empirical findings. In particular, the negative effect of the earthquake is observed for both destructive earthquakes and for earthquakes that did not result in physical injury or damage to buildings but were still felt by the population, the former producing a larger impact. The effect of an earthquake on the transition probability lasts no more than ten years. Consistent with the idea that an earthquake, in this historical context, represents a shock to people's religiosity, the authors also find that the number of churches built in each city is positively correlated with the seismicity of that city, and the negative impact of earthquakes on institutional transitions is not observed in the group of non-Episcopal cities where the communal movement was also under way in the period considered.
Yann Algan and Elizabeth Beasley, Sciences Po; Frank Vitaro, University of Montreal; and Richard Tremblay, Université de Montréal GRIP
Childhood behavioral skills are strong predictors of adult socioeconomic success, but little is known about how to improve these skills, in particular among children at highest risk of poor adult outcomes. Algan, Beasley, Vitaro, and Tremblay use data from a long-term randomized evaluation of a childhood behavioral skills training program in Montreal to answer this question. They match detailed data on behavior from adolescence with administrative criminal and educational records and self-reported socioeconomic outcomes. As adults, the subjects in the treated group are about 30 percent less likely to have a criminal record, 50 percent more likely to have a secondary diploma, 16 percent more likely to be active full time in either work or school from the ages of 17 to 26, and 68 percent more likely to have ever belonged to a civic or social group. The authors distinguish the different potential channels through which this intervention operates, and present evidence that self-control and social trust are potentially important channels. One possibility is that these behavioral changes in early adolescence (ages 1013) lead to improvements in school outcomes, particularly class assignment, in later adolescence (ages 1417), which in turn lead to improved adult outcomes. Using conservative assumptions in a simple framework, the authors estimate that $1.00 invested in this program yields $4.50 in increased wages.