This paper derives and estimates a unified and tractable model of comparative advantage due to differences in both factor abundance and relative productivity differences across industries. It derives conditions under which ignoring one force for comparative advantage biases empirical tests of the other. I emphasize two empirical results: First, factor abundance- and relative productivity-based models each possess explanatory power when nesting the other as an alternate hypothesis. Second, productivity differences across industries do not bias tests of the HO model in my sample. However, I find weak and mixed evidence that Heckscher-Ohlin forces can potentially bias tests of the Ricardian model.