Over the past quarter century, a significant body of work has emerged examining the relationship between legislators and bureaucrats. Much of this work has been theoretical and primarily focused on relations in the American context. In this issue of the Quarterly, three articles test hypotheses derived from this literature on legislative-bureaucratic relations in arenas beyond those involving the U.S. Congress.
In the first of two articles using data on Brazil, Sylvia Gaylord examines the circumstances under which members of the Brazilian Congress delegate budgetary decisions to the president. The theoretical angle that Gaylord pursues is taken from work centered largely on legislative-bureaucratic relations in the United States. The formal relationship in Brazil, however, differs significantly from that found in the United States in that the president enjoys very strong powers and the Congress very limited powers. The specific question that Gaylord addresses is when do Brazilian legislators opt to pass budgetary legislation that delegates power to the bureaucracy and when do they choose to write legislation that keeps budgetary powers in their own hands. Using data on 53 tax and budget laws passed between 1999 and 2002, Gaylord confirms the larger claim from the literature that legislators delegate when they calculate that it is in their interest to do so. In this case, on direct expenditure measures, over which the president enjoys significant powers, members of Congress elect to delegate considerable spending authority to the bureaucracy. In turn, lawmakers spend much of their time in the capital lobbying bureaucrats about how the appropriated money should be spent. But on legislation involving tax expenditures, over which legislators enjoy much greater control, very little authority is delegated to bureaucrats and lawmakers route the benefits directly to their constituents.
The focus in the second article examining Brazil is on how state lawmakers design bureaucratic institutions. Like the national president, state governors in Brazil enjoy extraordinary powers compared to their counterparts in the United States. In their article, Marcus André Melo, Carlos Pereira, and Heitor Werneck look at the question of institutional design in a political system where the governor, not the state legislature, enjoys most of the power to establish new agencies. Examining 31 independent regulatory agencies created in the Brazilian states between 1997 and 2006, the authors find that heightened electoral uncertainty leads governors to fashion independent regulatory agencies that are insulated from political influence, and thus protecting the new bureaucratic institutions from the governors’ political opponents should they take control of the governorship in the next election. But when a governor’s political future looks more secure, independent regulatory agencies are created with less political insulation, thus allowing the executive to directly influence decision making. By testing theories on the Brazilian system—theories formulated with politics in the United States in mind—Melo, Pereira, and Werneck distinguish between relationships that are system-specific and those that apply generally across these two systems.
The third examination of the relationship between bureaucrats and legislators is drawn from an extensive series of interviews of Michigan state lawmakers conducted in the late 1990s and the early part of the 2000s. These conversations are used by Marjorie Sarbaugh-Thompson, John Strate, Kelly LeRoux, Richard C. Elling, Lyke Thompson, and Charles D. Elder to investigate two separate questions. First, do legislators in one of the more professionalized state legislatures approach oversight of the bureaucracy in the same manner as members of Congress do? Second, has the imposition of term limits changed the importance Michigan legislators attach to oversight activities? This latter question can be addressed because the interviews conducted by Sarbaugh-Thompson and her colleagues bookend the period just before and just after term limits took effect in Michigan. The authors’ analysis of the data coded from these interviews documents that oversight was not of particular interest to Michigan lawmakers prior to term limits, but it became even less important after limits went into effect. Moreover, the variables that are thought to explain the interest of members of Congress in conducting oversight do not appear to have much explanatory power in the case of Michigan lawmakers. Perhaps even more significant, however, is the finding that in the post-term limits legislature, far fewer legislators think oversight is an important activity for the legislature, and, distressingly, a non-trivial proportion do not even recognize it as being within the legislature’s purview. These attitudes are true even for members of the Appropriations Committee—the unit assigned most of the institutional responsibility to oversee the actions of the executive branch. Thus, by changing the career orientation of Michigan legislators through the imposition of term limits, the incentive structure for legislative oversight of the bureaucracy was also altered, undoubtedly in ways that were unintended by the reform’s proponents.
Legislative representation has typically been studied by examining the degree of congruence between a representative’s roll-call voting behavior and the policy preferences of his or her constituents. Using the decennial redistricting of the U.S. House of Representatives as a quasi-experiment, Matthew Hayes, Matthew V. Hibbing, and Tracy Sulkin expand the notion of representation by looking at how lawmakers change their issue agendas as well as their roll-call voting behavior in response to the changing demographic and partisan composition of their districts. Analyzing data on House members who served in both the 107th (2000–01) and 108th (2001–03) Congresses, the authors show that representatives altered the number of bills they sponsored and cosponsored across a range of issue areas in response to demographic changes in the districts they represented, while shifting roll-call voting behavior in reaction to partisan changes in their constituency. Interestingly, the level of responsiveness to the changes in the characteristics of the constituency was greater for a representative’s issue agenda than for his or her voting behavior. It is also notable that the authors find no evidence that members’ electoral concerns influence their representational behavior—those whose new districts made them objectively safer behaved much the same as those who were thrown into greater political peril.
The final article in the issue investigates what motivates members of the U.S. Senate to cosponsor bills. While this question has been addressed many times before, Brian M. Harward and Kenneth W. Moffett offer a novel twist. They examine only the cosponsorship decisions of members of the U.S. Senate who voted in favor of a bill. This approach enables them to distinguish between a senator’s decision to cosponsor a bill from his or her decision on whether to support its passage. Using a data set of 320 major bills introduced in the Senate between 1975 and 2000 that made it to the Senate floor for a vote, the authors report several results that are consistent with other studies of cosponsorship. They find that a senator is more likely to cosponsor a bill introduced by an ideologically similar colleague and that ideologically extreme senators are more likely to cosponsor bills. But the most important finding is that senators increase their cosponsorships in response to increased demands from their constituents. Breaking down constituent demands by policy areas, the authors document such relationships with health care, education, civil rights, and agriculture. Coupled with the previous article in this issue of the Quarterly, this finding emphasizes the importance of decomposing the policy agendas of legislators so we can better understand what drives their behavior in office.
Return to February 2010 Titles
Return to LSQ home page