Legislative research is often criticized for being a “normal science,” engaged in routine work on small puzzles that yield small incremental gains in knowledge within an accepted paradigm. Undoubtedly this is true to some extent, and it is not necessarily to be condemned. But there are also plenty of exceptions, examples of creative new approaches to old puzzles and of discovery of previously unexplored gaps in our understanding. The articles in this issue of the Quarterly provide examples of work that are anything but routine.
Legislators have long complained of the inadequacy of press coverage of their work and scholars have echoed that complaint. These complaints express worry over the effect of reduced press coverage on the relationship of the legislature to the public. But Linda L. Fowler and R. Brian Law take a new tack by identifying an effect of reduced press coverage not on the public but on the internal organization of the legislature. Does reduced press coverage of the Senate committees that deal with national security issues—the Armed Services Committee and the Foreign Relations Committee—reduce their attractiveness to Senators, leading them to redirect their careers? Fowler and Law have undertaken an astonishingly comprehensive compilation of all mentions of Congress in the pages of The New York Times between 1947 and 2006, finding that coverage in 2006 was just a bit better than half of what it had been 60 years earlier. Furthermore, coverage of the national security committees fell even faster. Using two closely related measures of committee attractiveness and comparing the influence of declining media coverage with other influences on committee attractiveness, Fowler and Law conclude that the Senate Foreign Relations Committee has lost visibility and prestige inside the Senate, and that the Armed Services Committee has continuously avoided public attention. They explain that internal rules changes and external events also determine changes in the relative attractiveness of committees, sometimes offsetting the effect of declining media coverage. But the authors note that the decline in press attention to the national security committees has not only diminished the prestige they had within the Senate 60 years ago, but that it has probably hurt the public discourse on national security issues.
The effect of the structure of legislative institutions on their policy outputs is one of the most important and at the same time one of the most difficult research subjects. In the comparative study of U.S. state legislatures, one example is the influence of the professionalization of legislatures on the level of appropriations they authorize. Public efforts to limit spending have driven the movement for term limits, in the belief that a more amateur legislature will be more frugal. The research problem is that professionalization and increasing demands for expenditure may have common causes that cannot be easily disentangled either analytically or practically. Neil Malhotra tries to do that, employing a new method to illuminate a classic problem. He uses propensity score matching to compare spending levels among legislatures existing in contexts that are equally likely to produce professional legislatures although some of the contexts have produced such legislatures and some have not. With control of contexts, he finds no significant relationship between the professionalization of legislatures and the level of expenditures they authorize. He speculates that professionalization of legislatures may be a response to, rather than a cause of, the growth of expenditures. Malhotra’s analysis has practical as well as academic consequences. His method is also applicable to the study of the effects of other institutional characteristics whenever these characteristics are not randomly distributed across cases.
The identification of party influence on voting in legislatures has always been difficult, but that difficulty is greatest in legislatures with young party systems. In such systems various institutional incentives other than party may outweigh party as an influence on voting behavior. Tanya Bagashka uses a novel method to identify the principal determinants of voting coalitions in such settings. She examines individual voting records in the Russian Duma that sat between 1996 and 1999, using discrete latent variable analysis that identifies voting blocs not by formal membership but by similarities in voting behavior among groups of deputies. She finds that although parties were influential in the Duma, there were also important cross-party voting blocs reflecting presidential and electoral incentives. Her general conclusion is that where parties lack stable resources and clear programs, local constituency concerns and powerful presidents can substantially offset party incentives. While this conclusion is not new, Bagashka’s method enables her to estimate the relative strength of these various influences without presupposing the importance of party affiliation.
Analysis of press coverage of Congress usually assumes that the press provides members with a channel for communicating with their constituents and provides constituents with an instrument for monitoring their representatives. Using content analysis of the coverage of 30 members of the U.S. House of Representatives in 2000–01 by local newspapers, Brian J. Fogarty investigates whether and how this happens. He hypothesizes that the press is a strategic actor whose incentive to provide coverage varies with its own interests. He compares coverage of members who were in step with their constituents and coverage of those who were out of step. He finds more thorough coverage of members whose voting records are out of step with their constituents. These members are more newsworthy and they are often found in marginal districts. In providing more coverage of these members, the press may enhance accountability where it most affects election outcomes.
Extensive cross-national research has long demonstrated that differences among electoral systems have profound effects on legislative representation. Philip D. Habel adds an example from within the United States, using a change in the system for electing members of the Illinois House of Representatives to assess the effect on members’ careers. Before 1982, Illinois used a system of cumulative voting that assigned three seats to each district and permitted voters to cast as many votes as there were seats. Voters could cast all of their votes for one candidate, divide them between two, or cast one for each. When Illinois changed that system to the common first-past-the-post system, it provided a natural experiment in which there was a control for all extraneous variables. Cumulative voting makes it more difficult for members to develop a personal vote for themselves and is a source of confusion about responsibility for legislative actions. Hypothetically it should shorten legislative careers while single-member constituencies should favor incumbency. Habel compared legislators’ terms of office for all those Illinois legislators elected between 1966 and 2000, using an event history analysis that compared three competing risks to their tenure: redistricting, reapportionment, and the change in the electoral system. He finds that redistricting had the effect of leading some members to seek a higher office, that all three risks affected the propensity of members to retire, but that the change in the electoral system had the effect of reducing the risk that electoral defeat would end a members’ career. Through this natural experiment, Habel adds evidence that the single-member district system of election protects politicians’ tenure and helps to explain the high incumbency advantage that exists in the United States at both the state and the national level.
Beginning with the November issue, the Quarterly will publish a series of articles on the challenge of identifying legislators’ policy preferences and using these preferences to capture the dimensionality of the policy space. The research in these articles draws on evidence from a variety of legislatures and compares a variety of indicators of members’ ideal points.
Return to August 2008 Titles
Return to LSQ home page