Party Control and Partisan Bias in 1980s Congressional Redistricting

We estimate bias and responsiveness in 1980 and 1982 congressional elections for the nation as a whole and for states grouped according to control of the governorship and state legislature and according to legislative versus court control of the redistricting process. We find that in states where one party controlled the redistricting process, partisan bias moved in a direction favoring the party controlling the redistricting, but the effect was large only for states under unified party control. However, at the national level, redistricting was almost a wash, with a slight pro-Democratic bias in 1982 which was lower than that in 1980. These findings contradict views widely expressed in the national press about a strong pro-Democratic national gerry-mander, and also contradict the findings of Abramowitz (1983) that partisan bias in 1980s congressional redistricting was important only in Democratic-controlled states. We show that Abramowitz's conclusions rest on an improper method of estimating bias, one that confounds bias and responsiveness (swing).