James R. Rogers

Bicameral Sequence: Theory and State Legislative Evidence

Theory: Voting sequence is a strategic variable between bicameral chambers. Decisions by chambers to acquire costly information affect the payoffs chambers receive from sequencing choices. Thus, it is also a strategic variable between bicameral chambers. Hypotheses: (1) Because of incentives to take advantage of one chambers informational expertise when chambers share relatively homogeneous policy preferences, house chambers in legislatures dominated by a single party will initiate adopted legislation in greater proportion relative to legislatures that are not dominated by a single party. (2) A cost advantage in acquiring information grows with an increasing disparity in the number of legislators between bicameral chambers. Lower information acquisition costs in the larger bicameral chamber account for a persistent first-mover advantage for that chamber. Methods: Hypotheses are derived from a game theoretic model in which informational endowments are endogenously chosen by bicameral chambers and in which voting sequence is endogenously determined. Implications of the model are tested with an original data set developed from the session laws of thirty-three state legislatures. Results: Under specified conditions, bicameral chambers sequence themselves to take advantage of one chambers informational expertise. Under other conditions, bicameral chambers will not take advantage of a chambers informational expertise. The main theoretical results conclude: (1) that one-party legislatures take greater advantage of informational expertise relative to divided-party legislatures; (2) that senators have an incentive to specialize in more risky policy environments relative to house members; and, (3) that there is an informational basis for house chambers to initiate more adopted legislation than senate chambers. The empirical results are consistent with hypothesis 1. The results are consistent with hypothesis 2, although an alternative (noncontradictory) theory might also account for the result.

Free Riding in State Legislatures

This paper applies the theory of team production to thelegislative process and tests the empiricalimplications of that theory using a uniquestate-level data set. Empirically testablehypotheses are derived from the theory,including predictions that free riding willincrease as chamber (or majority) sizeincreases, and that legislativeinstitutions (such as committee structureand staffing levels) can mitigate freeriding. The hypotheses are tested using apooled state-level data set developed fromthe session laws of twenty-three statelegislatures. The empirical results areconsistent with the major hypotheses.

The Impact of Bicameralism on Legislative Production

It is generally accepted by scholars, as well as by cameral partisans, that adding a second chamber to an otherwise unicameral legislative process will decrease the volume of laws that a legislature enacts. This study challenges the conventional wisdom. First, I offer a simple theoretical argument that shows that when second chambers can originate as well as reject legislation, bicameralism will have an indeterminate impact on legislative production. Second, I provide historical data gathered from the four U.S. states that have experienced cameral transitions. Although very rudimentary, the historical evidence, when coupled with the theoretical argument, raises serious doubt regarding the traditional claim that bicameralism reduces the production of legislation.

Empirical Determinants of Bicameral Sequence in State Legislatures

In a previous article (Rogers 1998), I showed that the bicameral chamber that originates enacted legislation tends to realize policy outcomes closer to the preference of its median legislator than does the chamber that votes second on legislation. All things being equal, this �first-mover advantage� implies that each chamber could be expected to originate roughly half of all enacted legislation. But all other things are not equal in U.S. state bicameral legislatures. Drawing on an expanded dataset, I innovate and test a number of additional hypotheses related to bicameral voting sequence. My results account for the effects of constitutional, institutional, and electoral variables on bicameral sequence.