The voting weights implied by the estimated model demonstrate that
early voters have up to 20 times the influence of late voters in the
selection of candidates, demonstrating a significant departure from the
ideal of "one person, one vote."
Here is the paper, I cannot find non-gated versions. If we were aiming for efficiency, and the saving of time, rather than democratic equity, in which state should the first primary or caucus be held?
why not have them in every state on the same day?
If we’re aiming for efficiency and ignoring democratic equity, why are we bothering with an election?
Very interesting paper. Didn’t know that the effects were this large. Will
have to read the paper though.
King rat’s comments are interesting but Tyler’s question still stands.Tim
makes a really interesting suggestion. I think it is quite practical as
well. I will add as a second best solution that states should vote in
order of population size so that these effects are minimized.
The high quality of comments on blogs like these always surprises me.Maybe
the blogoshpere does have a future.
DC?
By the law of large numbers, a random selection of a few thousand voters is nearly as a good as a national primary for determining who is a viable candidate. Efficiency has little to do with it. The nomination process is a much a part of the campaign for the general election as what comes after.
A few things:
Are we assuming all money spent by candidates is an inefficiency? Just the losing ones? just the ones with nothing interesting to say (I’m looking at you, Joe Biden)?
So which do you think is more efficient, the few big candidates spending gobs of cash in CA or Texas, or more candidates spending less cash in a smaller state like NH or SC? I’m inclined to think it’s more efficient to go with a big state and (effectively) have a high monetary barrier to entry.
And awareness and support built up in January can transfer to November, so have the primary where you’ll be campaigning in the general election. Plus, it’s in each party’s best interest to select the candidate who can win the ‘marginal’ states in Novemeber.
So if you’re going for one state, go with Florida. I’m also sure all the news crews and campaign staff will appreciate Miami in December when contrasted with Manchester in January.
(and yes, I’m repeating Mark, but doing it with more words.)
I wish the two major parties would just go about selecting a candidate in a backroom of their national conventions like the old days.
This extended campaigning by way of “primaries” is simply another method of blocking out third party presidential candidates.
Most efficient would be to pick a random set of counties, weighted to match national demographics. The airfare to reach a random sample of individual voters is too high, and, when a candidate addresses an individual instead of a crowd, it is too easy for candidates to make contradictory promises.
What do we mean by “efficiency” in choosing candidates? Do we mean “candidates that would represent what voters generally desire in a set of candidates”? Or “a candidate that represents what voters *of that party* desires? Because if we mean the first, then we pick a battleground state (Ohio, actually. It is the most average state in the union in many respects.), and they’ll pick one set. But if we mean the latter, each party picks a different state to do the vote (in which case we’d need to know more about what each set of party voters in each state wants), based upon the average views of each party.
That said, can I put a word in for the value of inefficiency of choosing candidates? Primaries should take a while; we want candidates to beat each other up, views to crystalize, dirty laundry to be aired. If larger states go sooner, we can have a winner very quickly, and then we have the less-tested candidates in the general.
In terms of representing the “average voter”, Iowa is particularly problematic. Notice how not a single Republican candidate stood by free market principles (which their party is supposed to embrace) when asked about farm subsidies during the CNN/YouTube debate. If there’s a value to having small ‘gatekeeper’ states, they exact a very, very high price for it both from America and from the world (see: global food shortages, agri subsidies as barrier to trade deals).
Going back to the original question, if you are talking about the efficiency of the campaigning process, then the ideal state is one that is geographically compact (so candidates don’t have to spend a lot of time travelling within the state), easy to reach (since most candidates will be from other states), and has a relatively high population density (ability to reach the most voters reached per campaign swing). Alternatively, you could pick a state that is physically larger, but where the population is heavily concentrated in one or two cities (the expectation being that the candidates could ignore the unpopulated areas). On those criteria, I would suggest a first primary in Rhode Island or Connecticut, with the second in Nevada or Arizona.
When there was no incumbent President from their party, 42% of the time the eventual Democrat nominee lost IA or NH. For Republicans it was 50%. What’s a better percentage?
If a state always picked the eventual nominee, people would say it has too much power. If it never picked the nominee, they’d say it is irrelevant. (Candidates would likely agree and avoid the state.) By this measure, I’d say the states are doing their job. They’re testing the candidates, the race is relevant, but not so much that other states are unimportant.
Also, what about experience, culture and institutional knowledge? Are other citizens going to get excited in August (Ames straw poll) about an election in January? What if the state government screws up (Florida)? IA and NH voters get far more 1 on 1 time with candidates. They can judge this year’s candidates against previous years. Changing states would sacrifice this knowledge.
If we’re trying to save time, that mostly means keeping people away from the polls (and choosing a reasonably good candidate that doesn’t make horrible decisions that waste peoples’ time). Since most of the time and money wasted is in the general election, choose either a very conservative or very liberal small state–Idaho, SC, Hawaii, or Vermont. The dems of the conservative state will choose a centrist, while the republicans will choose a lunatic. Similarly, in the liberal state, the dems will choose a lunatic whilst the republicans will choose a centrist. The general election outcome will be certain from the start, so people will not waste their time voting (in either case, the lunatic will lose). And we will have a centrist result.
Comments on this entry are closed.