Hi Ellis [Krauss -- Moderator]! I think the demographic changes are related to
the rule changes in both cases, with some differences. In 19th century Britain,
the expanded suffrage made the old vote-buying practices harder to employ. This,
in turn, set into motion a series of reforms culminating, ultimately, in a
greater willingness to delegate power to the Cabinet and party leaders with
which to set the legislative agenda and formulate party platforms on which to
compete electorally.

In Japan, urbanization, etc. made it more difficult for many LDP members to
divide the vote among themselves using time-tested personalistic campaigning.
(See Cox and Rosenbluth, ELECTORAL STUDIES, forthcoming, on a statistical look
at who broke from the LDP along with Ozawa, Hata, et al; they tended to be
electorally somewhat weaker than those who remained, as you'd expect). In Japan,
even more than in 19th century England, the declining effectiveness of the
personal vote strategy posed a real problem because of Japan's district
magnitude (2-6)--dividing the vote among LDP members was still crucial in most
districts (in England district magnitude was typically 2), even while it was
getting harder to do so. Hence, the electoral rules were an even bigger road
block to party-based campaigning in Japan than in England.

I hope that clarifies what I was trying to say. In Ramseyer and Rosenbluth
[Japan's Political Marketplace: Harvard, 1993], on pp. 20-21, we say "As the
British Parliament enlarged districts and gradually eliminated multi-member
districts over the course of those decades, political parties found they had
greater success appealing to the median voter with policy programs than in
trying to buy off blocs of voters with particularistic favors. Larger districts
made particularism a more costly strategy for individual politicians, because
they had more ground to cover and more individuals to woo for support. At the
same time, the adoption of single member districts made particularism less
necessary..."

Only empirical work, of course, will tell us if events play out in Japan on a
parallel course to 19th century England. A key intervening variable between
electoral rules and candidate strategy is the endorsement process (although that
process is probably somewhat--but not completely!-- endogenous to the rules). I
am planning to spend next year (1996-1997) in Japan investigating endorsement
patterns for both single member districts and list PR districts. That should be
a lot of fun. Your survey project with Steve Reed and the other scholar you
mentioned sounds like its getting at a similar question, and I'll be interested
to see what you find!

Best,
Frances