Even Money: Tied Chambers, Power Sharing, and Party Eects in State Legislatures

In American legislatures, the party leadership has a wide variety of powers, including agenda-setting, selecting committee chairs and doling out committee assignments to members. Whether the leadership is able to use institutional rules to produce outlying committees, and as a result, outlying legislation is the subject of much debate among scholars of Congress. Because the majority party and the party leadership are constant within a Congress, it is difficult to discern independent “party effects” at the federal level. States, with significant variation in majority party size, institutional rules granted to the leadership, and ideological cohesion, offer a better venue for studying party effects. We leverage a unique institutional situation, tied chambers, to exploit the variation available at the statelevel while accounting for cross-sectional differences. Many state chambers have an even number of legislators, and partisan ties occasionally occur. In many states, power sharing agreements require leadership rotation, and give each party control over different committees, or switch leadership at time intervals. We connect these agreements to committee composition and legislative outcomes. Specifically, we expect that committee composition will reflect the outlying ideological preferences of the party with control over them, rather than the chamber median. We additionally expect bills reported from these committees to reflect different party agendas, leading to intraand inter-state variation in the size of winning coalitions. Oregon found itself in an unprecedented situation on the night of November 3, 2010. Election returns indicated that Republicans had picked up six House seats to tie the Democrats 30-30 in the sixty seat chamber, and amazingly, a victory by Republican challenger Alan Olsen in the 20th district also seemed to tie the Senate 15-15. After all the votes had been counted a few days later, Democrats ended up pulling out another close Senate race, putting the partisan balance in the Senate at 16-14. Still, the Oregon House chamber remained tied until the 2012 election, forcing the parties into a power sharing agreement. As unlikely as it to have two tied chambers, Oregon is not unique in ultimately having one of its two legislative houses in partisan deadlock. Unlike the United States Congress, many state chambers have an even number of seats, and as a result, there have been 13 instances of tied chambers since 2000, according to the National Conference of State Legislatures (NCSL) (See Table 1 below). Further, the NCSL reports that ties are occurring more commonly, with at least one every year since 1984 (in even-numbered election years). The Alaska Senate for example, was tied for four years across two elections, and even states with large populations and relatively professionalized legislatures like New Jersey, Michigan, Washington, and Pennsylvania have had tied chambers in recent years. A tied chamber forces the parties to create power sharing agreements, rotating turns dictating the legislative agenda, splitting committee assignments more evenly, alternating times during which each party presides, and forcing legislative compromise. While the parties and their leadership may find tied chambers a frustrating inconvenience, they offer a unique opportunity for researchers to test theories of party influence in legislatures. A tied chamber is a case where there is effectively no majority party, allowing us to compare legislative organization and outcomes to other state legislatures which operate under traditional majority party rule. Legislative scholars have long sought evidence that parties are able to use their majority power to produce non-median outcomes in majoritarian legislatures. One of the fundamental debates in the study of legislatures is the extent to which the median controls policy outcomes. In majoritarian legislatures, the median must be included in any winThe Oregonian, “Oregon House tied 30-30, waiting on the Senate.” Janie Har, November 3, 2010, accessed at http://tinyurl.com/msz35rr on March 28, 2014, and Crosscut.com, “Oregon will likely have a legislative tie, in both Senate and House.” Floyd McKay, November 3, 2010, accessed at http: //tinyurl.com/jwscgvo on March 28, 2014. See the National Conference of State Legislatures information page, “In Case of a Tie,” at http: //tinyurl.com/m9uzebq.