Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
In a recent article, Ananyev and Guriev (2019) used Russian gross regional product (GRP) data to argue that income changes strongly affected generalized trust during 2008–9 crisis. I show that GRP changes were a poor proxy for regional household incomes during the crisis. Substituting household incomes for GRP eliminates all statistical significance of the relevant coefficients and in some specifications even changes the coefficient signs.