Empirical evidence for the United States shows that many antidumping petitions are withdrawn before the International Trade Commission and the International Trade Administration complete their investigations. Prusa (1992) argues that petitions are used by domestic industries to threaten and induce foreign industries into a collusive agreement exonerated from antitrust concerns because of US trade laws. In his model, all antidumping petitions should be withdrawn, which is not the case. This paper provides a model in which only some petitions are withdrawn and Prusa's result is just a special case. The decision to withdraw a petition depends on two key parameters: the coordination cost and the bargaining power of domestic and foreign industries. A new dataset is constructed to test the theoretical model on the US experience for the period 1980-1992. The econometric analysis supports the theoretical conclusions of the model. This finding is consistent with the hypothesis that the antidumping law is used as a collusive device.