Dear Yasunobu,

Thank you for your thoughtful reply. I feel as though I am at a conference or workshop and benefitting from an excellent dialog!

As to the size of the floating voter portion of the electorate, you are doubtless right. I also agree that voter mobility, level of social organization and integration, and (perhaps--I'm less sure here) media influence are important influences in determining the number of floaters. On the other hand, I think the electoral system also exerts an independent influence. Here is how it might work: Every tie to an organized group comes at some cost to the policy flexibility of the candidate/party relying on those hard votes. In putting together a large enough pastiche to get past the post, the candidate relying on this hard vote strategy therefore begins to lose a coherent policy platform. An entrepreneur who is able to provide a more compelling, perhaps ideologically consistent, appeal should have the advantage. That, in any case, is at least a logical possibility.

You are surely right that campaign finance laws also matter. But it seems to me that, since desparate-enough politicians will always find a way to cheat, it is even more important to change their incentives. A system that advantages politicians with a broad appeal over those with a cobbled-together coalition of hard votes should have some effect independently of the penalties for cheating.

As to political culture, I'm sure there is also something there. Interestingly, however, it is said that the political culture in Whig England was extremely personalistic. For an interesting account, see John Ferejohn, "Rationality and Interpretation: Parliamentary Elections in Early Stuart England," in Kristen Monroe, ed., THE ECONOMIC APPROACH TO POLITICS, Harper Collins, 1991. Ferejohn doesn't mention electoral rules, there, but it seems plausible that electoral rules and political culture are interactive. For an argument that stresses electoral rules, see Thomas Rochon's article in John Campbell's 1981 edited volume in the University of Michigan Japan Studies series (sorry I don't have the full citation).

Split-ticket voting is certainly possible, but one effect of the new rules should be to strengthen the party's platforms. Cox points out that cross-plumping was much more common in pre-reform England before the parties were able to present coherent appeals. Another instructive case would be Germany; Kathleen Bawn has a piece on that in her dissertation (Stanford, 1992) but I don't know if it has been published. Perhaps in AJPS in the last year or so.

Best regards,
Frances