The 1990s redistricting aimed to increase the number of minorities in the United States Congress. The group most affected by this objective was African Americans. Between 1991 and 1993 the number of African Americans elected to the House in the South increased by more than ten. This increase in the number of African American House members coincided with the "Republican revolution" in Congress. Many scholars and political observers have suggested a relationship between the creation of majority-black districts and the increased conservatism in Congress.

I would argue that there is in fact a direct relationship and an indirect relationship between the creation of majority-black districts and increased conservatism in the House. The direct relationship focuses on the voting behavior of the House member. When majority-black districts are formed, the surrounding districts are "bleached", because they have lost a substantial portion of their black population. Many scholars contend that House members in these "bleached" districts will change their voting behavior to reflect better the interests of their new constituency, which has a higher white population proportion than previously. Therefore, as a result of the creation of one majority-black district, there may be several House members who become less liberal in their voting behavior due to the loss of a large liberal voting bloc. 

There is also an indirect relationship, and this focuses on the partisan makeup of the House. The 1994 election marked the first time in over 40 years that the Republicans dominated the House of Representatives. Some political scholars have suggested that the creation of the majority-black district influenced that shift in dominance. The "residual" districts, those subsequent districts that have lost a large portion of black population to the majority-black district, are now more attractive to the conservative Republican candidate. 

Recognizing that voters are rational in their selection of elected officials, the individual running for office must also act rationally. In both the roll-call voting behavior component as well as the candidate emergence component, the desired end is electoral success in the form of reelection or election. As rational actors the incumbent House members, candidates, and prospective candidates are expected to behave in a way that maximizes their chance of election or reelection. In both instances the individual running must choose what means will lead to the desired end. 

When focusing on the voting behavior component, the desired end is reelection. Accepting the notion that the primary goal of incumbent House members is reelection, it follows that their actions while in office will be geared toward maximizing their chances of reelection. House members maximize their reelection chances through constituency service, bringing federal money and projects to the district, and by representing their constituents' interests in Washington. The representation of constituents' interests is reflected in the congruence between the constituency interests and the House members' policy decisions as manifested in their roll-call voting behavior. 

Due to the 1990s redistricting, several districts underwent substantial changes in boundaries and composition. What happens to voting stability after the changes occur? Do incumbent House members change their behavior, and is this change a "stable" 

one? As rational actors working toward the goal of reelection, most incumbent House members would be expected to alter their voting behavior to reflect the changes in their constituency (Glazer and Robbins 1985; Whitby and Gilliam 1991; Fleisher 1993). Most previous works on race-based redistricting find that House members were responsive to changes in their district composition that occurred after the 1990s redistricting (Overby and Cosgrove 1996; Lublin 1997). 

In the candidate emergence component, the desired end is election. Behaving rationally, prospective candidates will search for an electoral context in which the possibility of electoral victory is high. High quality challengers will not expend time and energy competing in an election that they have no real chance of winning. The strategic challenger will wait until the "time" is right to forge a serious campaign. 

After redistricting, prospective challengers examine the new political environment in order to see if the time is right for them to run, and if electoral success is possible. Redistricting gives rise to several concerns that a prospective candidate must weigh in order to determine his probability of electoral success. Potential challengers ask questions such as how the districts have changed in terms of partisanship and race. They may also be concerned with what the final map will look like, and when it will be completed. As Representative Vin Weber (R-Minn.) explained in 1991, several potential candidates told him, "If I don't know what the district is going to look like, I'm not going to run" (Congressional Quarterly 1991:3325). In the 1990s, the potential candidates also had to be concerned with whether a court challenge might change the existing district. 

The potential challenger would likely pay close attention to the voting behavior of the incumbent to see whether his/her behavior changes as a result of constituency changes. If the incumbent's behavior does not appear to change, the prospective challenger may see an opportunity to defeat the incumbent. In addition, the redrawing of congressional lines may lead to increased electoral vulnerability for incumbents in general, due to the fact that their geographic and, more importantly, reelection constituencies may have changed significantly.

The idea that all of these factors would affect if and when a potential candidate ran for congressional office is commensurate with Jacobson and Kernell's (1983) strategic politician theory. Their theory asserts that "[p]oliticians do act strategically. Their career decisions are influenced by their assessment of a variable political environment." Thus, prospective candidates weigh the costs and benefits of running for office at a particular time. I assert that quality challengers should have emerged at an increased rate in 1994 and 1996. In several of the southern states the district plans were not finalized until shortly before the 1992 elections, making it impossible or unattractive for a candidate to run in the election. Instead, the number of quality challengers in 1994 and 1996 was high because candidates waited until the time was right to run for election. If this assertion were accurate, the following factors would influence quality challenger emergence. 

First of all, the more quality challengers would have emerged in states where race-based redistricting occurred. With the creation of majority-black districts, several congressional districts in a state were restructured. These restructured districts attracted more quality challengers who hoped that shifts in district boundaries would lead to a decline in incumbency advantage. Incumbent House members who previously relied on their name recognition and their record of service and voting behavior were now thrust into a "new" district. In this district the constituents may not have been aware of the incumbent's voting record or name. Because of this, the incumbent became electorally more vulnerable.

Furthermore, within these states, more quality challengers should have emerged in districts where the incumbent did not adequately change his voting behavior to reflect the interests of his new constituency. Particularly in districts with large decreases in their black constituency composition, there may have been too large a gap between the incumbent House members' ideology and the ideology and policy preferences of the new constituency. Anticipating the incumbent's inability to accurately represent the interests of his/her district, a more conservative quality challenger may strategically choose to run against the incumbent. By the same token, if an incumbent House member actually does not change his voting behavior to reflect the interests of the constituency, he too will be challenged by an individual who recognizes this discrepancy. 

Incumbent House members as well as prospective House candidates are rational actors whose goal is electoral success. Both incumbent members and prospective candidates select the method or means that will yield them the largest probability of success. The most efficient method for incumbent House members is to vote the preferences of their constituency. The most efficient method for prospective House candidates is to choose elections where the conditions are favorable for electoral victory. 

These profit maximizing decisions were made in the context of the 1990s redistricting. Incumbent House members who lost substantial portions of their black constituency due to redistricting became more conservative in their voting behavior. It is likely that this increased conservatism in roll-call voting behavior in the House was then compounded by the recognition on the part of strategic Republican politicians that the new districts were fertile for quality conservative candidacy. 

Congressional Quarterly, Inc. 1991. Congressional Quarterly Weekly Report 49(45): 3325.

Fleisher, Richard. 1993. "Explaining the Change in Roll-Call Voting Behavior of Southern Democrats." Journal of Politics 55:327-41.

Glazer, Amihai and Marc Robbins. 1985. "Congressional Responsiveness to Constituency Change." Journal of Politics 47:258-72.

Jacobson, Gary C. and Samuel Kernell. 1983. Strategy and Choice in Congressional Elections. New Haven: Yale University Press.

Lublin, David Ian. 1997. Gerrymander for Justice? Racial Redistricting and Black and Latino Representation. Princeton: Princeton University Press.

Overby, Marvin and Kenneth Cosgrove. 1996. "Unintended Consequences? Racial Redistricting and the Representation of Minority Interests." Journal of Politics 58:540-50.

Whitby, Kenny and Franklin Gilliam. 1991. "A Longitudinal Analysis of Competing Explanations for the Transformation of Southern Congressional Politics." Journal of Politics 53:504-520.