Abortion Rights in the US: Congress' Policy Responsiveness to Public Opinion

Abortion is a controversial and frequently debated topic in US Congress; however, literature on predictors of voting behavior by Members of Congress on the issue is scant beyond the late 1990s. In the aftermath of the Dobbs v. Jackson Women's Health Organization decision, Congress has contemplated the introduction of national policies on abortion to reduce the major power that the Supreme Court has placed on states. As a consequence, understanding Congress' responsiveness to public opinion on abortion-related decisions is relevant to potentially forecast the next evolution in the landscape of abortion rights in the US. The aim of this study was to investigate the responsiveness of Congress' decisions to public opinion on abortion. Four hypotheses were tested: H1) constituencies' opinions on abortion were expected to be correlated to Members of Congress' voting behavior; H2) partisanship was expected to be significantly correlated to Members of Congress' voting behavior, potentially offsetting the influence of constituencies' opinions; H3) the degree of belief in biblical literalism, used as a proxy for the strength of religious conservativeness in state/district, was expected to be correlated to Members of Congress' voting behavior; H4) electoral competitiveness was expected to enhance Members of Congress' responsiveness to constituencies' opinions. The analysis was based on abortion voting records between the 111th (2009-2010) and 117th (2021-2022) Congresses for both the House of Representatives and the Senate collected by the National Right to Life Committee (NRLC) and data on public opinion on abortion, voting behavior, and belief in biblical literalism collected in the relevant pre-electoral time periods by the American National Election Studies (ANES) survey. The empirical strategy consisted in estimating pooled ordinary least squares regression models using Members of Congress’ NRLC pro-life score as the dependent variable, with state/district and Congressional year fixed effects. Robustness checks were implemented using generalized linear models with logit link. As for the House of Representatives, results showed that public opinion on abortion lost statistical significance in explaining NRLC score once average NRLC score by party members was controlled for; moreover, partisanship was the single consistently significant predictor while religious conservativeness and electoral competitiveness did not appear to be relevant in explaining NRLC score. Looking at the Senate, however, public opinion on abortion by party voters remained statistically significant in explaining NRLC score also once average NRLC score by party members was controlled for; also in this case, religious conservativeness and electoral competition did not exhibit consistent explanatory power. In conclusion, partisanship appeared to be the key predictor of voting behavior by Members of Congress on abortion-related decisions, although differences in the role of public opinion emerged between the House of Representatives and the Senate.