The 2008 Democratic presidential primary in California will be conducted on a two-tier semi-proportional basis. As described in the California Progress Report, 241 delegates will be allocated in California’s US House districts and 81 statewide. A threshold of 15% applies in each tier. The district magnitude at the district level will range from 3 to 6, “based on a formula of total population and the average for Democratic candidates in the last two presidential elections.”

With a high threshold in the statewide tier and low magnitude in the district tier,¹ this system is not meaningfully “proportional,” but it has the potential to allow a candidate with well under 15% statewide but with local concentration of support to win some delegates. Of course, it is hard to say whether any such candidate exists (Richardson in some Latino-majority areas?, Kucinich in San Francisco or the North Coast?), because when was the last time you saw a presidential pre-candidate preference poll conducted at the level of congressional districts?

The CPR comments:

A candidate can very well win the state as a whole and not win a majority of delegates.

Well, duh. It is not a [plurality] winner-take-all system. However, if a candidate had a majority of the statewide vote, she (or, less likely in 2008, he) would surely win a majority of the delegates. And the candidate with the plurality is sure to be significantly over-represented–very likely with a majority–unless the result in statewide votes is very close, or there is a significant regional variation.² Neither of which is likely.

For all its flaws–and the Democrats’ system has many–this is vastly better than what the Republicans will use in this state for their 2008 presidential primary: District-level block plurality, with no variation in magnitude. The absence of magnitude variation means that House districts with few Republican voters are vastly over-represented. The winner-take-all rule means that voters have strong incentive to vote strategically, yet they will have little information on which to base such strategy. Unlike in the Democratic primary, your candidate, if not the plurality winner, will win no delegates from your vote. So, California GOP voters, do you know who of Romney, Giuliani, Huckabee (etc.) might be in serious contention to win the plurality in your House district, or to be most likely to surpass the potential leading candidate you least like? I did not think so.

A while ago the LA Times had an article about the Ron Paul campaign attempting to take advantage of these rules’ overrepresentation of San Francisco Republicans, and the city’s libertarian bent, to win some delegates for Paul there.

The CPR conveniently put up a link to the Democratic candidates’ campaign offices. Many of these candidates could get some district-level delegates, even if (as I suspect is likely) only two (or at most three) win any of the 81 statewide delegates. (California field offices are listed, however, only for Clinton and Obama.)
_________

1. With such low magnitudes, the 15% district threshold is not likely to be operative, though if a 6-seat district were quite fragmented in its candidate preference, it could come into play. [↩]
2. For example, if the statewide leader racks up huge majorities in big urban areas with 6-seat districts (some of which should have more than 6 seats if the magnitude allocation strictly followed Democratic voter populations), but falls below 15% in many other districts, while the second candidate statewide wins one or two delegates even in the leader’s strongholds, while beating the leading candidate in most of the three-seaters. It is hard to imagine such a scenario materializing, interesting though it would be. [↩]