Gerrymandering, incumbents and their vote share in the United States of America

This paper investigated the effect of gerrymandering on the vote share of incumbents in the United States of America. Gerrymandering is the practice of redistricting to have an advantage during the next elections. This research is done because gerrymandering still is an issue today and the incumbent re-election rate in the last elections was 95%. This raises the question where this high re-election rate comes from. It is interesting to see whether gerrymandering helped incumbents to win in the past. The data used consist all the winner of district elections between 1974 and 2008. For every district we looked up if redistricting took place in redistricting years, which are always the third year of a decade. To see whether cases of redistricting were actually gerrymandering, the Reock Ratio is used. The first step to look into this issue was to check when gerrymandering is done. Redistricting processes only take place every ten years and are typically done by incumbents who think they might lose the next elections. This fact is checked upon and correct, because by a probit regression we found that incumbents who face a larger vote share loss, have a higher probability to gerrymander. Next, the electoral effect of gerrymandering were investigated. To do this, a sensitivity analysis is made to check whether gerrymandering has an influence for a longer period than only the elections after redistricting took place. Since the results did not give enough reasons to believe it did, we concentrated on the years around redistricting. For these years the effect of gerrymandering on vote share were either not significant, very small or even negative. This result is consistent with earlier research on this topic