PHYLOGENETIC COMPARATIVE METHODS IN CROSS-CULTURAL ECONOMIC RESEARCH

This dissertation explores the origins and evolution of culture across societies, introducing phylogenetic comparative methods to cross-cultural economic research using ethnic-group-level data. The first chapter examines headhunting in Austronesian societies as a cultural adaptation to frequent inter-tribal warfare, functioning both as a mechanism for warrior training and a system of rewards for skill development. Using phylogenetic comparative methods and ethnographic data, we find that headhunting was significantly more prevalent in societies exposed to frequent warfare. Bayesian correlated evolution models suggest that its adoption followed increases in conflict, while its decline preceded reduced intergroup violence.

The second chapter addresses a key methodological challenge in cross-cultural economic research—cultural non-independence due to common ancestry. We show that phylogenetic regression, combined with global language trees, effectively accounts for this issue. Using Murdock’s Ethnographic Atlas, we demonstrate that economic, institutional, and cultural characteristics tend to be more similar among societies with closer ancestral ties, violating standard independence assumptions. Simulations reveal that phylogenetic correlation leads to severe inefficiency of the standard OLS estimator compared to phylogenetic generalized least squares (PGLS). Reassessing recent published results, we find that PGLS estimates are consistently smaller and less statistically significant than their OLS counterparts.