Recent surveys revealed few producers in an export sector participate in trade. Economists explained this result by relaxing their assumption about firms’ operations, to produce a novel observation: Trade liberalization disproportionately benefits the most efficient producers in the sector, while potentially harming the least efficient. Political scientists have begun exploring the consequences of this variation, especially in lobbying. This article explores whether the impact of this finer-grained description of interests can be observed in the later stages of our demand-driven models of the politics of trade. I focus on one case with characteristics favorable to observing intra-industry differences: the American steel industry in Taft's presidency. A trade-based cleavage inside the sector determined firms’ interests. Demands shaped policy, as observed in three pieces of legislation: the Payne-Aldrich Act, reciprocity with Canada, and the 1912 tariff. The first liberalized trade in steel, intensifying competition in the industry. The second promised to do the same, with a similar impact. The third had no effect, however, because Taft vetoed the bill. This case illustrates intra-industry firm heterogeneity can provide additional accuracy, revealing a previously undiscovered cleavage. Nonetheless, preferences alone did not determine policy.