Additional Considerations Maximizing Competitiveness: Too Much of a Good Thing?

Since the Graduated Random System delays the need for candidates to raise and spend money, possibly many more candidates would toss their hats in the ring than under the current system. This would hopefully include candidates having talent but only modest means who would otherwise be discouraged by the high entry costs of the current system. Conceivably, this might mean that voters would be confused by a huge flood of candidates, a cacophony of disparate political messages.

There were twelve Republicans who started out to seek the 2000 nomination. Half of them dropped out before the Iowa caucuses because they did not raise enough money in the pre-season (also referred to as the Money Primary). Without big bucks, they knew they were going to get blown out by the second week of March, so there was no point in going on. We have about the same number of major Democratic candidates for 2004. Historically, this is not an unusually large number to start out with. While it is likely that a system less driven by money would encourage a somewhat larger initial field of candidates, perhaps as many as 18, only a half-dozen or so will win, place, or show in a sufficient number of the first month’s contests. The rest will have demonstrated their non-viability and will drop out of the race. Over the next couple of months of primaries, this half-dozen should slowly dwindle, until perhaps two to four candidates remain viable going into the ninth round.

Is it more or less likely under Graduated Random System versus the current system that a candidate who might be highly competitive in a later large-state primary and the national general election would be prematurely discouraged and eliminated by the outcomes of early small-state primaries? At first blush, one might regard it as somewhat more likely due to the “smaller-earlier” structure of the Graduated Random System; however, how does one define “small” versus “large?” Virginia is the 12th most populous state, and under the Delaware Plan it would vote in the last of four rounds. On the other hand, in the Graduated Random System, Virginia would be eligible to vote in the second of ten rounds. If a candidate expected to win Virginia in the second round, he or she should not be so discouraged by outcomes in the first round as to drop out of the race. So faint a heart ought not to become president, or as Mark Antony said in Shakespeare’s Julius Caesar, “Ambition should be made of sterner stuff.”

The current system demands that a candidate prove his or her viability in Iowa or New Hampshire, because shortly after these come Mega Tuesday and Super Tuesday. However, because of the gradually increasing size of primary contests under the Graduated Random System, the first round of primaries would not be do or die states.

Competitiveness in the later rounds would be a function of several factors. First of all is the quality of the campaigning. If a candidate is successful in early rounds, this is likely due in part to his or her message resonating with the voter, together with a well-organized campaign that is effective in getting that message out. A successful campaigner won’t change message; he or she will keep going with what works, expecting that it will keep working. Similarly, a competent campaign organization is unlikely to “unlearn” from its early successes and be less competent in later stages of the campaign. Rather, a successful organization will learn lessons from wins and losses alike and transmit the benefit of those experiences to the later contests.

Independent of this is the “bandwagon effect,” in which the herd mentality uncritically votes for the front-runner.

Last, but not least, of course, is what Jesse Unruh, former Speaker of the California Assembly, called “the mother’s milk of politics.” The time-structure of the Graduated Random System gives candidates time to collect contributions and arrange loans for the big, final push at the end of the primary season. The intrinsic qualities of the successful campaign organization, together with the “bandwagon effect,” increase the likelihood that early successes at the polls will translate into money to fight the final, large battles. Thus, candidates are less dependent on raising huge sums of money the year before the primary season. Currently it is this so-called “money primary” that determines who will actually face the voters. Half of the twelve candidates who started out to seek the 2000 Republican nomination dropped out of the race before the Iowa caucus and the New Hampshire primary. The Graduated Random System greatly reduces the power of this money-driven (plutocratic rather than democratic) down-selection process, and allows candidates to “earn as they go.”

The continuous, elegant Graduated Random curve is based on a straightforward mathematical hypothesis, that such a curve should provide the healthiest growth medium for a broad spectrum of candidates, the perfect political Petri dish. However, simple mathematical models rarely reflect the complexity and inelegance of the real world. Thus one might be led to ask, would some lumpy, discontinuous, step-function alternative, although mathematically ugly, be more optimally adapted to the objective of producing a party candidate most fit to compete in the general election? Why not require at least one large-population state early in the process to achieve a shake-out of the David Dukes and Pat Buchanans before resuming the smooth, monotonic sweep up the Graduated Random System’s electoral curve; in essence, why not put at least one speed bump in the system early on?

The counterargument is that, is an early speed bump necessary? The David Dukes and Pat Buchanans are like hardy weeds that occasionally grow in the cracks of the pavement, but never spread out to cover the landscape of presidential politics, for the politics of fear and anger usually stunt themselves. But a big speed bump in 1968 or 1972 might have prevented a Gene McCarthy or a George McGovern from flowering. The current system does not seem to produce the McCarthy’s or McGoverns anymore, but the Dukes and Buchanans are still with us. This seems like the worst of both worlds. We do not need to erect new barriers to candidacies, we need to tear down existing ones. In any case, while many find the Dukes and Buchanans objectionable, some do not, and they have as much right to access the democratic process as anyone else.

Of course, the test of any hypothesis is data. It turns out that the graduated curve corresponds to a surprising high degree to historical data from an era when competitive campaigns were much more common than they are today, as shown in the following figures:

Could marginal candidates with narrow appeal, who might otherwise be weeded out in an early large-state primary, instead win blocks of votes and gain momentum in small states unrepresentative of the national electorate, making him or her hard to stop at a later stage? If one looks at the Graduated Random System’s process as a type of stratified sequential sampling, and if the first stage small-state samples are grossly unrepresentative of the U.S. electorate that will decide the final outcome in the general election, is it possible that the Graduated Random System would select out more wheat and select in more chaff in the early primaries, resulting in a higher likelihood of survival of the unfit for combat in the general election?

How credible is this scenario? Patrick Buchanan is sometimes cited as an example of a marginal candidate with narrow appeal who won New Hampshire--a small, unrepresentative state--in 1996. Yet Robert Dole had little problem stopping him in later primaries. If Buchanan had turned out to be hard to stop, would this not mean that he was in fact not a marginal candidate with narrow appeal? In the final analysis, it is (or ought to be) the electorate that determines who is marginal and who is not, rather than the pundits.

In 2000, the Democratic Party took the position that too much competition for the presidential nomination would put it at a disadvantage. Having won the two most recent presidential elections under an increasingly front-loaded system, and confidently looking forward to a third victory, this time behind Al Gore, the DNC Rules and Bylaws Committee attitude was “don’t fix it if it ain’t broken,” in utter denial of the reality that the system is indeed broken. Its report to the national chair in April 2000 stated that “criticism about a front-loaded schedule has to be weighed against a process that seems to be working well. In the last few cycles, the current system has allowed the Democratic Party to identify its presumptive nominee early. As a result, the process has helped the Party unify behind its nominee and focus its resources on the general election.... Therefore, at this point, there is little incentive to change the system.” (Democratic National Committee 2000, 11, 16).

Al Gore started out as the heir-apparent and easily captured the nomination. However, he went on to run an abysmal campaign against George W. Bush, squandering the advantage of being the sitting vice-president during the longest economic boom in memory. In retrospect, Democrats have to wonder whether Bill Bradley, whose challenge to Gore was blown out early due to front-loading, might have been a better choice for the party. A more deliberative process would better ensure that the party is represented by the best candidate, rather than the best fund-raiser or the heir-apparent or the pundit-anointed front-runner.

This disappointing experience in 2000 with the front-loaded system did not change the Democratic Party’s mind, however. Looking forward to 2004, their overriding political concern was the sky-high popular approval that President George W. Bush has enjoyed since the Al Qaeda attacks of September 11, 2001, and throughout the wars in Afghanistan and Iraq. “[Democratic] Party leaders anticipated an uncontested nomination on the Republican side and did not want to grant the opposition an additional competitive advantage by prolonging the search of their candidate (Parshall and Mattei 2002, 9).” However, the economy had been sluggish, there had been corporate scandals in the energy industry to which both the president and vice-president have business and campaign contribution ties, and there was also a controversy over whether intelligence on Iraqi weapons of mass destruction was politicized and distorted to justify the war. Despite these political liabilities, Bush had no opposition for the 2004 Republican presidential nomination. The front-loaded schedule is a major reason why. It squeezes what otherwise could be a legitimate and well-reasoned dissent within the Republican Party into an untenable blip in time. It gives an insurmountable advantage to the front-runner, and a sitting president is the ultimate front-runner. Were both parties to adopt a schedule more permissive of competition, challenges to an incumbent would be more likely.

The anticipation of “no contest” in one party incentivizes the other party to consciously craft (or leave be) a system designed to end the process as early as possible. But what evidence is there for the idea that parties are best served by narrowing the field of candidates and restricting political debate? After all, it took Franklin Roosevelt four ballots to win the Democratic nomination in 1932, yet he went on to defeat Herbert Hoover in one of the biggest landslides in American history. In the eight most recent presidential elections, shown in Table 13, it can be seen that in all three cases in which a party nomination was uncontested, the general election was won by that party. However, of the five elections in which the nomination was contested on both sides, in all four cases in which the nomination in one party was clearly the more contested in terms of the number of candidates, the general election was won by that party (the 1988 nominations were nearly evenly contested in both parties). Thus the historical data suggests that the next best thing to an uncontested nomination, which can only occur in the case of a popular incumbent, is a highly contested nomination. Such a contest generates more media attention, more interest in the party’s candidates, more diversity of issues, and hopefully in the end, a more fully vetted political platform. Since a contested nomination is statistically more likely than an uncontested one, it behooves the parties to have a more competitive system to better ensure the nomination of the most electable presidential candidate.

One can draw an analogy between the current strategy in presidential politics and nuclear strategy during the Cold War. Essentially, front-loading has collapsed the strategic space in political planning into a first-strike, launch-on-warning scenario. It is the American electorate that gets nuked every four years.