In searching for a frequently updated source of information on the ever-changing primary calendar for 2008, I stumbled across the website for the National Association of Secretaries of State (NASS). As with a handful of other organizations (ably covered here by Subversive), the NASS has a panacea for the calendar mess - a rotating regional primary/caucus calendar.
While one could reasonably assumes that an organization of secretaries of state could come up with a tenable plan for an election calendar, this idea fails on at least three fronts.
First, the NASS plan would continue to give the first voice to Iowa and New Hampshire, "based upon their tradition of encouraging retail politics". This position assumes two things: first, that these states actually encourage retail politicking, and second, that retail politics is a good thing.
The American Prospect's Paul Waldman takes down the first assumption in the case of Iowa(covered by Markos here). Iowans are not smarter than the rest of the country, they do not have a greater knowledge of politics, and they turn out in embarrassingly low numbers. An estimated 100,000 people turned out for the 2004 Iowa Democratic caucus, which would represent somewhere in the 5% range of the voting age population. Yet the next morning, the traditional media will report the anointing of the winner, lending dramatic publicity and fund-raising advantages to that candidate while simultaneously dismissing the also-rans.
Worse, this pittance of the population cannot even come close to representing the demographics of the United States. An appropriately balanced poll commissioned from Gallup that had a sample of 100,000 people would be remarkably accurate (an estimated +/- 0.3%). But in a self-selected survey, which is essentially what a caucus is, there is very little hope of achieving a representative result. This is especially true in a state like Iowa, which is so unlike the rest of America:
Percentage white (2000 Census): Iowa 94.9%, USA 80.2%
Percentage Hispanic (2000 Census): Iowa 3.7%, USA 14.4%
High school graduates (2000 Census): Iowa 86.4%, USA 80.4%
Unemployment (August 2007): Iowa 3.9%, USA 4.7%
And don't forget, Iowa gave its eight electoral votes to Gerald Ford and Bob Dole in 1976. What a bellweather.
Further, changes in American culture and advances in technology have radically altered the way people communicate with each other and with their elected representatives. Instantaneous information (macaca, anyone?) makes it possible for voters to be affected by events hundreds of miles away. It also gives them the chance to make decisions based on raw feeds, and not on information or events filtered through traditional media. Retail, door-to-door, hand-to-hand polticking is antiquated; like Iowa, it is no longer representative of where we are as a country.
Second, creating a rotating system but then limiting it to geographic regions is self-defeating. The four regions depicted in this plan are similar enough, and different enough from each other, that even a regional focus for the first primary of any given year gives a disproportionate amount of influence to one area, one culture, one way of life, one set of demographics. In the election where the northeast region chooses first, the Democratic Party could get saddled with an exceptionally liberal candidate who cannot sufficiently compete nationally. In certain electoral circumstances, the same could happen to Republicans if the southern states chose first. In addition, the number of delegates to be earned in these regions is not balanced, from a low of 964 in the Midwest to a high of 1207 in the South. Combined with the maintained primacy of New Hampshire and Iowa, this can perpetuate the problems of undue regional or state influence.
Third, and perhaps most fatally for the NASS plan, the regional rotating primary concept relies on the cooperation of legislatures, executives, and other decision-makers in 56 different jurisdictions. (The NASS even offers Model Primary Legislation (pdf) for this purpose.) Any one of these jurisdictions could start the tumbling dominoes again by making a unilateral power play to advance their own selection date. There is nothing in the model legislation, nor could there be anything created, that binds the states and other delegate selectors to this one method. Any region, blessed by lottery to be the first selectors in 2012, may enjoy their status so much that they abandon the rotating system and move their own primaries earlier. I would not expect that such a system could survive four consecutive elections, enabling each region a chance to go first one time, before some state jumped out of line again.
Whether it be a rotating, rolling, regional, or national primary system, or some other method, there are really only two options for implementation. One possibility is discipline from the two national party committees. Using their own criteria, each committee could set their national primary calendar, refuse to recognize any delegates that are chosen outside accordance with that calendar, and show the fortitude to enforce and protect the system if challenged. After all, the parties are private organizations, and have the right to decide for themselves how to choose the candidates that bear their standards.
(Worth noting: [1] Both parties have already shown inconsistencies and weaknesses in the enforcement front. [2] The holier-than-thou looks from the Republicans over Democratic threats to unseat Florida delegates are spurious. Democrats are threatening entire delegations that are not selected according to their criteria. Republicans are only planning to withhold half of all "improper" delegates. However, given the fact that there are approximately 4,323 Democratic delegates, and 2,302 Republican delegates, the effect on the process is mathematically similar.)
The second possibility is action from the federal government. While this option is fraught with dicey political challenges, it is the more likely of the two choices to have strong and lasting effects. While elections, like education, are considered by most to be local fixtures, the federal government has ways to influence the process. Federal legislation is needed to correct a number of ways in which we conduct national elections - standardizing identification, ballot types, recount methods. Election calendars are one more thing that can be added to this list. Standardization of all of these problems can reduce the costs of elections, decrease voter confusion or voter fatigue, and increase interest and participation in the process.
Unfortunately, I have expended so much mental energy dabbling in the land of the theoretical that I have few resources left to tackle the questions of practicality and action. Whatever changes come, they should have some minimum characteristics. The entire nation should have an identical, or at least equally random, opportunity at influencing the process. The changes must come from an entity large enough and powerful enough to bind all the states and other delegate selectors to its process. The process must reflect and employ the campaign and communication methods that are currently common to most of the country. And the end result should encourage confidence, clarity, and participation in the process.