Why Local Government is Unrepresentative and Uncreative

Many thanks to Harvard's David Schleicher for letting me know about his wonderful new paper, "Why Is There No Partisan Competition in City Council Elections?". This isn't (only) dry political science, it can also be read as a call for a radical shake-up of the way that democracy works in cities. Let me skip straight to the most exciting idea:

A state could also take a more dramatic step: it could bar any party that chooses to register for national and state elections from getting a place on the ballot for local elections... Such movement would require parties seeking to be competitive in local elections to express policy preferences on local issues, which would permit all voters to use the party identification of candidates as a good heuristic for their policy views on local issues. Because of the use of single member districts and first-past-the-post voting, local elections would likely result in a two party system at the local level, too, but with parties that provide relevant heuristics for the policy beliefs of the candidates.

I love this idea. Everybody complains about how politics is broken in Albany, but as Schleicher says, New York state elections are positively dynamic compared to elections at the city level. The party affiliation in any given seat never changes, which means that the only election which matters is the primary, in which only the most hardcore local party activists ever vote. And more to the point, the differences between Republicans and Democrats nationally simply aren't the kind of differences which make much sense at a local level:

Beliefs about local politics do not strongly track beliefs about national politics. There is substantial evidence that Democratic or Republican voters (and politicians) in any locality, who form relatively coherent ideological blocs on national issues, do not form coherent ideological blocs about local politics. Put another way, the information that a candidate is a Democrat or a Republican in a local election does not reveal much information about her beliefs about local issues.

The results are far-reaching, and very important:

If the model is correct, there are dramatic implications. First, and most importantly, it shows local elections are very inefficient means of translating voter preferences into government policy. That is, local government does not meet the most basic definitions of democracy--it does not provide voters with the ability to replace incumbents with opponents with different views and to have their views represented in local policies. Further, a system that retards party competition also removes from local politics the forces that create new political coalitions, investments in the development of political ideas, and new leaders. The lack of parties makes city government both unrepresentative and uncreative.

What's interesting is that at the mayoral level, where individuals really do run on their own ideas rather than on their party affiliation, people with new ideas - Ken Livingstone, Mike Bloomberg, even Rudy Giuliani, for that matter - can and do get elected. But the only time people can vote for those ideas is during mayoral elections. When it comes to electing a city council, the paradigm of voting for national parties effectively quashes any hope of originality or democracy.