In Chapters 4–6, we argued that pro-Republican bias in nonsouthern congressional elections disappeared in the mid-1960s as a consequence of (1) a change in partisan control of state governments away from unified Republican control, (2) a change in the legal reversion to the redistricting process, and (3) the role of the largely Democratic federal judiciary in supervising the redistricting actions of the states. If our account is correct, redistricting during the 1960s should have affected Democratic and Republican incumbents differently.

Perhaps the simplest way to state our expectations is by contrasting them with Edward Tufte's well-known hypothesis that the wave of 1960s redistricting was mostly an exercise in incumbent protection – a position that fit well with then-emerging evidence that incumbents' margins of victory were increasing. Our thesis is that the wave of 1960s redistricting had a distinct anti-Republican cast to it, making most districting plans less Republican-friendly than their predecessors. In this chapter, we investigate whether the partisan differences we expect appear in the empirical record.

The first purpose of this chapter, then, is to provide further (albeit indirect) evidence for the main thesis of this part of the book. The second purpose is to reconsider some alternative hypotheses about the case of the disappearing bias. The final purpose is to begin to introduce the topic of the next part of the book: how redistricting affected the electoral performance of incumbents and the much-debated rise of the incumbency advantage.

We state our predictions regarding how redistricting should have affected district-level vote shares in the first section. We then test them in the second section, finding that redistricting had a systematically different impact on the two parties' electoral fortunes.