Gustav Alexandrie on Model Thinking

Thanks for writing this! Here are some further reflections: As the paper points out, there is a lot of uncertainty associated with the Barro-Becker model and its parametrization, so one should probably not put too much weight on it. Moreover, the model was designed for the very specific purpose of explaining fertility choices/population growth; I would not necessarily expect it to be a very good model for estimating parent-to-child altruism. For estimates of the intergenerational altruism parameter, it might be worth looking at is Schwarze and Winkelmann’s 2001-paper "Happiness and altruism within the extended family". They estimate that the intergenerational altruism parameter is between 0.04 and 0.25, using survey data from the German Socio-Economic Panel. Another consideration that is worth mentioning is that the intergenerational altruism parameter, as used in the Barro-Becker model, also reflects discounting. The Barro-Becker model is not an overlapping-generations model: the parents live for 25 years and are then replaced by their children, who live to 25 years, etc. Therefore, in the model, parents' only opportunity for being altruistic is to give up some of their own current consumption and save it for the sake of their children's consumption in 25 years. A standard 2% rate of pure time preference on its own implies that utility in 25 years from now is valued at only 60% ≈ 0.98^25 of current utility.