Laws and Roll Calls in the U.S. Congress, 1891–1994

Recent empirical studies of lawmaking activity by legislatures rely heavily on roll call based measures and assume that roll call activity reflects lawmaking activity. We question this assumption for the case of the U.S. Congress. We examine several plausible sources of dissonance between the set of enacted public statutes and the universe of recorded votes in the U.S. Congress, using a comprehensive dataset of public enactments and roll call activity between 1891 and 1994. Because only 11.9% of the bills signed into law receive a recorded vote in the House, only 7.9% receive a recorded vote in the Senate, and only 5.5% receive a recorded vote in both the House and Senate, we provide guidance as to when studying voting behavior is likely a reasonable proxy for lawmaking behavior. There are sometimes important differences between the laws that do and do not receive a roll call that researchers should account for when using roll calls to study lawmaking in the U.S. Congress.