We find no support for the thesis that social cleavages as a whole have declined in magnitude in U.S. presidential elections since 1960. There have in fact been increases in two of the four major social cleavages we examine (race and gender), small decreases in one (religion), and no net change in the fourth (class). In terms of the rank order and relative magnitudes of each of the cleavages, we find, not surprisingly, that race has by far represented the largest cleavage during the period since 1964. African-American voters have overwhelmingly embraced the Democratic Party (in both relative and absolute terms), while white voters have moved away from that party. The next largest cleavage has been the religious cleavage, with the class and gender cleavages following (in order). Most interestingly, the ordinal ranking of the race, religion, class and gender cleavages has been stable since 1964 (although we will see in chapter 8 that this did not hold in 1996).
Because our analyses investigate the interrelationship between social cleavages, they help us to better understand some of the claims made in the literature about the impact of the rise of the race cleavages on other social cleavages. With respect to the race/class thesis, we find no support for the strong version of this thesis which asserts that the emergence of the race cleavage leads to a lasting dealignment of the class cleavage. However, we do find evidence supporting the weak version of the thesis, namely, that when the race cleavage grows during a particular election, the class cleavage experiences a corresponding shrinkage during that election. While some presidential elections have clearly been marked by the prominent use of campaign messages that activate the racial resentments of some whites, our results also imply that the zero-sum relationship between the race and class cleavages can be produced even in the absence of such explicit cues. 35 Nevertheless, despite the powerful impact of the race cleavage, our analyses also provide evidence that even large increases in this cleavage in future elections will be insufficient by themselves to displace the other three cleavages.