Do Some Americans Prefer Divided Government and Vote to Create It?

Tests of theories of the electoral origins of divided government hinge on the proper measurement of voter preferences for divided government. Deriving preferences for divided government from voters’ ideological positions or responses to the standard American National Election Studies question inflate the number of people who prefer divided government. We develop two alternative survey measures of preferences for divided government. We evaluate preferences for divided government using these measures across multiple surveys and show, contrary to other measures, that a small but relatively stable percentage of voters prefer divided government and are significantly more likely than other voters to vote for divided government. Relatively rare in earlier periods of US history, divided government has become the new normal. The opposition party has controlled at least one house of Congress in eighteen of the twentyfour congresses between 1969 and 2016. While some scholars claim that divided government has little effect upon legislative productivity (e.g. Mayhew 1991), others contend that more significant legislation is passed during periods of unified government (e.g. Binder 1999, Coleman 1999). Divided government may also influence other areas of governing such as bureaucratic rule-making (e.g. Epstein & O’Halloran 1996, Yackee & Yackee 2009) and the length of time for judicial confirmations (e.g. Shipan & Shannon 2003). The clearest recent example of how divided government can affect government operations is the inability of Merrick Garland to get a hearing for his nomination to the Supreme Court. An unresolved question in American politics is whether divided government occurs because some voters prefer to divide government power between the two political parties. At least since Fiorina (1988) and Alesina & Rosenthal (1989) proposed that voters intentionally divide government, public opinion surveys have sought to determine the proportion of the US public that prefers divided government. The 1992 American National Election Studies found that 32% of Americans thought that “it is better when one party controls both the presidency and Congress,” 40% who believed that it is “better when control is split between the Democrats and Republicans,” with 28% responding that it doesn’t matter. Similar distributions of responses are evident on successive ANES surveys, with 49% of respondents to this question in 2012 expressing support for divided government. These figures raise the question of whether divided government occurs because Americans have preferences for divided government that influence how they vote. Empirical studies of attitudes toward divided government use several different measures of support for divided government and show mixed results. Some studies support the hypothesis that some voters prefer divided government and vote in accordance with that preference (Alesina & Rosenthal 1995, Carsey & Layman 2004, Lacy & Paolino 1998, Lewis-Beck & Nadeau 2004, Mebane 2000, Smith, Brown, Bruce & Overby 1999), but others reject it (Alvarez & Schousen 1993, Beck, Baum, Clusen & Jr. 1992, Born 1994, Burden & Kimball 1998, Burden & Kimball