By now, it is quite clear that the rules of the democratic primary process should be reformed (note: they should also be followed, pending a reform). The task of reforming the primary process will be, to paraphrase Hobbes, nasty, brutish and long, and I can predict with confidence that whatever compromise emerges in the future will be far from perfect although, hopefully, an improvement upon the current system. Whatever happens in the future, I think it is important to learn from this years' primary campaign. Below the fold are some of the lessons to be learned, as well as some suggestions for the future
From the 2008 primary campaign, we learn that:
- The more complicated the rules of a process, the more likely it is that it will be perceived as unfair by some people (this is true, I think, even if one were to trust campaigns not to muddy the waters). Multiple metrics, multiple voting systems, hybrid systems, superdelegates, add-on delegates, all contribute to the possibility that supporters of one candidate will feel their candidate has been treated unfairly. Of course, the implied requirement of simplicity needs to be weighed against other considerations; eg some states will want to have primaries and others, maybe for good reasons, will want caucuses.
- Fights over superdelegate support are ugly. Even taking away that guy in California who wanted money for his support, Virgin Island's Kevin Rodriguez's erratic behaviour and the Bill Richardson "Judas" mess, the whole process of candidates cajoling superdelagetes into supporting them is unseemly and undemocratic. It smacks of insiderism and favor trading. In the best of all worlds, superdelegates would be eliminated; maybe we can at least hope that only elected federal officials are allowed to be superdelegates.
- It is possible to have a primary where all states count, without precluding an insurgent campaign from winning, and without having retail politics in some states. In fact, this is what happened this year! The argument against having a few small states go first is that they decide, unfairly and at the expense of other states, who the nominee is. The argument in favo of small states first is that 1) they allow for retail politics and 2) they make it possible for a candidacy that is not well funded to succeed, on the basis of early wins in small states. This year, almost all states had an important say (ironically, the states with the least weight in the process were Florida and Michigan). I think this has to do with proportional delegate apportionment and also, this year, with the fact that there were a lot of very good candidates and also a stupendously stubborn/resilient one. In any case, I think that the 2008 primaries prove that you can have the best of both worlds, that is early primaries in small states and the benefits they bring AND every state being important in the primary process.
- The knives come out when the primary is down to 2 people. The campaign only became really negative when Edwards dropped out. The larger the amount of candidates running, the less worthwhile it is for one of them to go negative. The lesson from this is that, in as much as possible, the primary calendar should encourage a three-or-more way race for as long as possible. Long primaries have mostly positive effects, the negative effect of long primaries is the negative campaigning. In as much as possible, the rules of the primary process should discourage negative campaigning.
- Republicans will ratfuck open primaries. On the other hand, requiring registration as a democrat to participate in a primary increases democratic registration. I think closed primaries are a no-brainer (on the other hand, I think it should be easy, not hard, to register as a democrat).
Based on these lessons, I would like to put forth the following suggestions:
I. The DNC should mandate closed primaries.
II. The winner of the popular vote should be the nominee. This would be good for keeping the process simple (lesson 1), would eliminate superdelegates (lesson 2)and would discourage negative campaigning in a 3 or more person race, because negative campaigning depresses turnout for the offending candidate as well as the offended candidate. Of course, winning states would have an impact on the media narrative.
III. I think the role of small states should be to vet a wide field of candidates and reduce a field of ten or so candidates to a field of three or four. The role of big states should be to choose among the remaining candidates. This seems consistent with the "best of both worlds" idea of lesson 5. Small states would have the opportunity, through retail politics and cheaper media markets, to prop up candidates with low name recognition and low campaign funds- small states should go first. Big states would have an equally important role, that of deciding among the candidates who have passed the small state sieve.
I would like to suggest two primary calendars that would achieve this idea:
The small to big calendar
This calendar would only have three dates, each three weeks apart.
Date 1 (say, beginning of february): all the offshore territories, AK, HI, OR, ID, NV, UT, MT, WY, NM, ND, SD, NE, KS, OK, IA, AR, MS, KY, WV, SC, District of Columbia, DE, CT, RI, VT, NH, ME
Big establishment candidates would likely try to win across the board, small insurgent candidates would concentrate on one or two states in order to establish their viability. Perhaps a third category would emerge, that of regional candidates who can claim, say, "I can win in the west" or "I can win in the south". From a field of ten or so candidates, maybe five or four candidates would still be on their feet
Date 2 (end of february, beginning of march): WA, AZ, CO, MN, WI, MO, LA, IN, TN, AL, VA, MD, MA
I think this "medium states" primary would reduce the field to two or, hopefully, three candidates
Date 3 (end of march): CA, TX, IL, MI, OH, PA, NY, NJ, NC, GA, FL.
Big states decide the nominee in the end.
The regional calendar
This calendar would retain a feature of the small to big calendar, the big states go at the end. Small states would be grouped into regions, and each primary date would feature two regions (so that, each primary date, candidates would have a stimulus to focus on a single region, leaving the other to other candidates, thus, in theory, keeping a field of three or more candidates for a longer time) . The nice thing about this idea is that it would lead to debates about regional isssues in the primary process. This calendar would have X dates, two weeks apart (regions that have smaller states go first, they are all roughly equal in size):
Date 1: beginning of january
the Midwest: ND, SD, NE, KS, MO, IA
the Northeast: CT, MA, VT, NH, ME
Date 2: Middle of january
the northwest: WA, OR, NV, ID, MT, WY
the mid atlantic: DE, MD, VA, the District of Columbia
Date 3: beggining of february
the southwest: AZ, UT, CO, NM
appalachia: WV, KT, TN, SC
Date 4: middle of february
the gulf coast: OK, AR, LA, MS, AL
the great lakes: MN, WI, IN
Date 5: beggining of March
The big states: CA, TX, IL, MI, OH, PA, NY, NJ, NC, GA, FL.
...
So those are my thoughts, for what they're worth