Replication data for: Procedural Cartels in Texas: A Note

In this work, we test the Procedural Cartel Thesis of Cox and McCubbins (2002, 2005). In this model, one may view agenda control as a natural monopoly or naturally subject to cartelization. While voting power in the world’s democratic legislatures is always distributed equally (one legislator, one vote), the power to determine the order of business and other matters of procedure tends to be concentrated in the hands of a relatively small group of “senior partners” in the governing coalition. As originally stated, this thesis presumes (1) an assembly that has full control over its internal organization and does not face constitutional or other entrenched provisions that seriously hamper its ability to delegate agenda powers; and (2) a majority coalition or party that uses the rule-making power of the legislature to cartelize the agenda. The Texas state legislature is interesting in that it violates both assumptions. First, Texas’ constitution endows the Lt. Governor with extensive agenda-setting powers. Thus, as in Brazil (cf. Amorim Neto, Cox and McCubbins 2003), agenda cartels cannot be easily formed without the active leadership of an external actor (the Lt. Governor in Texas, the president in Brazil). Second, because the Lt. Governor has exogenously entrenched power, and because the Lt. Governor is separately elected, the majority in the Texas Senate may or may not be of the same party as the Lt. Governor. Thus, majority status and agenda power can be separated in Texas (and other similar cases, such as Brazil), in a wa y that they cannot in the U.S. House of Representatives (and other similar cases, such as the British Parliament). To test for the existence of agenda cartels, we usually examine the majority (and minority) party’s roll rate: the percentage of bills that pass contrary to the expressed preferences of a majority of the majority party. In Texas, we have collected legislative roll call votes from both chambers of the legislature for the years 1981 through 1999. In particular, we sampled every roll call vote on the third reading of every bill during the 67th through the 76th regular sessions in each chamber.1 In addition to reporting roll rates by majority status, we also consider how “divided government” (e.g., a Lt. Governor of a party different than that of the majority in the Senate) affects party roll rates