Auctions of divisible goods: on the rationale for the treasury experiment

We compare a sealed-bid uniform-price auction (the Treasury'sexperimental format) with a sealed bid discriminatory auction(the Treasury's format heretofore), assuming the good is perfectlydivisible. We show that the auction theory that prompted theexperiment, which assumes single-unit demands, does not adequatelydescribe the bidding game for Treasury securities. Collusivestrategies are self-enforcing in uniform-price divisible-goodauctions. In these equilibria, the seller's expected revenueis lower than in equilibria of discriminatory auctions.