Chocolate Consumption and Noble Laureates

Abstract

At first glance, the positive correlation of chocolate consumption per capita and the accumulated number of Nobel laureates per capita, as reported by Messerli (2012), seems spurious. The intention of this paper is to check whether this correlation vanishes if relevant covariates and a sophisticated estimation method are employed. A two-stage Heckman selection model is estimated where the stock of Nobel laureates per capita is regressed on chocolate consumption per capita, as well as the number of published scientific papers, R & D expenditures per capita and GDP per capita. In addition, coffee and tobacco consumption per capita are also included. In contrast to the expectation, it is found that a positive correlation between chocolate consumption per capita and the stock of Nobel laureates per capita persists although more recent data and relevant covariates, as well as a sophisticated estimation method are used. In addition, a further negative correlation between coffee consumption per capita and the number of Nobel laureates per capita is detected. Albeit no clear causal relationship between chocolate and coffee consumption seems to exists, it remains unclear whether the effects are caused by hidden variables or by chance. However, the results are limited due to the lack of individual consumption data.

Keywords: chocolate consumption, coffee consumption, scientific papers, R & D expenditures, Nobel laureates, spurious correlation