We reconsider Krugman’s love-of-variety model of trade (Krugman, 1979). We show that initially, as trade costs fall below the prohibitive level, trade leaves everybody worse off. The reason is that, due to the envelope theorem, the gains from trade are second-order, while the utility losses associated with the disappearance of domestic varieties are first-order. This finding is robust. It applies to unilateral and coordinated reductions in trade cost or tariffs, and it extends to heterogeneous firms and arbitrary asymmetries between countries. Even if perfectly free and costless trade dominates autarky, a bit off trade is worse than either of the two.

Please sign up for meetings here: docs.google.com/spreadsheets/d/1uvCvkSKwY8S9Pq7Kk7MT1eG9tIdMA_JMseiQmWjTWqE/edit#gid=0