ABSTRACT
The classical gains-from-trade conjecture, first formulated in the eighteenth century, was finally given a thorough Arrow-Debreu proof in 1972. It was then and remains virtually the only easy-to-interpret comparative-static proposition valid for any Arrow-Debreu economy.2 Since 1972 the proposition has been repeatedly challenged, sometimes on grounds that in no respect violate Arrow-Debreu assumptions; and therein lies a puzzle on which we focus in the first part of our paper. The remaining and most recent challenges rest on the assertion that some of the Arrow-Debreu assumptions are unnatural and unnecessarily restrictive or on the imposition of artificial restrictions on the set of permissible compensatory transfers; these challenges will be examined in the remainder of the paper. It is our general conclusion that none of the challenges can be sustained.
