The "Baqaee-Farhi approach" and a Russian gas embargo

In a controversial policy paper, Bachmann et al. (2022) argued back in March 2022 that the economic effects for Germany of a complete immediate stop of energy imports from Russia would be small, between 0.5% and 3% of GDP. Baqaee et al. (2022) even presented 0.3% GDP loss in the case of an embargo as the headline number in a follow-up report for the French Council of Economic Analysis (CAE). This note argues that these estimates are both problematic from a scientific point of view, and also strongly biased towards finding small effects of a gas embargo: this is true of the (so-called) "Baqaee-Farhi approach" arriving at 0.2-0.3% of GDP, the "production function approach" arriving at 1.5% to 2.3% of GDP, as well as the "sufficient statistics approach" (also based on Baqaee-Farhi) arriving at 1% of GDP. This note argues that Olaf Scholz was correct in saying that the mathematical models which were used "don't really work" here, and tries to explain why. In any case, these models do not permit such categorical statements.