No Middle Ground: How Informal Party Organizations Control Nominations and Polarize Legislatures

Chapter 3 150 Years of Legislative Party Behavior

chapter 3 150 Years of Legislative Party Behavior The previous chapter largely focused on California’s political history in the 1950s, noting changes in voter and legislative behavior in response to the insertion of party labels on the primary ballot and the banning of cross-‹ling. This analysis painted a picture of politicians as nonideological reelection-seekers. With extralegislative activists excluded from party activity, incumbents moderated and ran the chamber in a way that discouraged partisanship but facilitated their own reelections . The removal of Progressive party-killing reforms cleared the way for local activists to reassert themselves in party nominations, and only then did legislators begin to act like legislative partisans. The availability of 150 years of legislative roll call data makes it possible to look beyond those crucial years of the 1950s and see how assembly incumbents behaved over a larger time frame. We can observe how they reacted to the imposition of Progressive reforms in the ‹rst place, and how they have behaved in recent years. These data also provide a more general opportunity for researchers to understand how incumbents respond to a vast range of institutional conditions . Using long time-series of roll call data, legislative researchers in recent years have drawn all manner of valuable conclusions about legislative behavior, including the in›uence of parties on members’ votes (see, e.g., Krehbiel 2000; Poole and Rosenthal 1997; Snyder and Groseclose 2000). Unfortunately, these studies, enlightening as they are, share a common limitation: a focus on the American Congress. Apart from Jenkins’s 87 (2000, 1999) innovative studies of the short-lived, partyless Confederate Congress, all of these studies examine the national legislature, a body that has enjoyed a functional two-party system essentially since the 1820s. The U.S. Congress has experienced remarkably little structural change in the past two centuries. By contrast, there is a great deal of variance in the legislative structure of statehouses. Several states have highly competitive party systems, while others—notably Rhode Island, Hawaii, and the pre-1970 South— have had essentially one-party rule, and Nebraska has banned parties altogether . Territorial legislatures were often nonpartisan, as well (Espino and Hindman 2007). States offer signi‹cant variations on institutional constraints, including term limits, redistricting practices, salaries, and electoral rules, allowing legislative researchers to make inferences about the in›uence of those constraints on legislative behavior. However, until now, comprehensive roll call data have not existed. One exception is the recent work of Gerald Wright and colleagues (Wright and Osborn 2002; Wright and Schaffner 2002; Wright and Winburn 2002), which utilizes complete roll call datasets from all 99 state legislatures in several recent sessions. But extended time-series for state legislatures have been unavailable , until now. The vast dataset employed here, the ‹rst of its kind for a state legislature , allows us to investigate incumbents’ responses to a variety of institutional conditions. California, after all, has gone from being a virtual one-party state in the early twentieth century, to a two-party state with suppressed partisanship during the cross-‹ling years, and ‹nally to the ‹ercely polarized state that exists today. Using these data, I am able to quantify the levels of partisan polarization in the assembly at various times in history and to measure the impact of various in›uences on legislative behavior. The evidence testi‹es to the strong polarization of the modern assembly, the profound dampening effect that cross-‹ling had on partisanship, and the tendency for incumbents to follow the wishes of whoever can control their fates. the history of the california assembly at a glance Before using the roll call data, however, I attempt to sketch out partisan behavior in the California assembly by examining Speaker elections. The election of a Speaker is a party’s most important vote in any assembly ses88 no middle ground sion—the Speaker exerts a great deal of power over the staf‹ng and leadership of committees, the resources available to individual assembly members, and the fate of legislation (Blair and Flournoy 1967). In a strong party system, the Speaker would be expected to act in a way that bene‹ts his party at the expense of the minority. In the U.S. House of Representatives, for example, Speaker elections have been strictly partyline affairs for over a century, and rare deviations on this key vote are considered punishable offenses. Thus elections for Speaker are a critical indicator...