Congressmen, Perquisites, and Elections.
This paper examines the so-called "perquisites/constituency service" hypothesis that purports to explain incumbents' electoral margins in congressional elections. First, it reviews the individual-level and aggregate-level studies, pointing out serious problems in measurements and methodology and concluding that existing literature does not and cannot support the thesis. Second, using 1984 CPS/NES data, the paper models the voting decisions in 1984, finding that respondents' decisions were affected by shared partisanship with the incumbent, ideological compatibility, challenger spending, assessments of President Reagan, and (a dubiously measured) recollection of the incumbent's having done something for the district. Individual casework experience did not affect the voting decision. Third, with data gathered from over 200 House offices, the percentage of the vote won by incumbents in the 1982 election is modeled as a function of 1982 casework loads, staff allocation, mailings, and several baseline variables. The findings are that casework loads, trips home, and mailings did not affect the vote, but that district partisanship and ideology, redistricting, national electoral tides, and occasionally district education levels did. We conclude that, based on existing evidence, the perquisites and casework hypotheses have not been demonstrated and that to explore the theses further, panel data are needed.
|Main Author:||McAdams, John.|
|Other Authors:||Johannes, John.|