Saturday, January 03, 2009

The Right Agenda: Fix the Primary System

It should be clear by now that the primary system for the 2008 election was a disaster.

States competing for influence pushed their primaries earlier and earlier. Some states awarded delegates proportionally, while others gave all delegates to the winner. In many cases, state parties wrote the rules to favor whichever particular candidate their leadership supported. The media devoted most coverage to the candidates that it declared "frontrunners", and ignored or scorned the rest. Exit polls revealed that a significant minority of voters in party primaries were democrats or independents. A significant number of voters also appeared to have no clue what the candidates actually stood for. For example, a fair number of McCain voters listed immigration restriction as their most important issue.

What can be done to fix these problems?

The primary schedule. The scheduling problem is admittedly difficult. States have pushed their primaries earlier and earlier because states that have early primaries have disproportionate influence on the selection of the nominee. This is because the media has created the notion of "momentum", in which a candidate who does well early supposedly is much more likely to win, even if only a tiny fraction of delegates have actually been selected. This becomes a self-fulfilling prophecy, with candidates who don't win the first couple primaries being hounded about when they will drop out and dismissed as unviable.

The difficulty is compounded by the fact that it is state parties and/or governments that actually set the primary dates. They have no incentive to do what is in the "national interest". National parties can allocate delegates, but they cannot stop states from having early primaries. Only the federal government could do that, but that would be an unacceptable and unconstitutional inference in parties' internal business.

One proposal to "fix" the primary system is to create a single national primary. While this might be "fairer", it would be even less likely to choose good nominees than the current system. This is because in a national primary, the media would have more influence over public perceptions of the candidates, as they could not do as much retail campaigning per person. Such a system would also give establishment candidates with large early fundraising an advantage, since advertising would be relatively more important.

Another idea would be to leave the primaries spread out, but rotate the order in which the states go. This is probably the best schedule in theory, but it is unlikely to happen because Iowa, New Hampshire and other states wouldn't go along with it for the reasons mentioned above. A variation on this system that preserved the privileged status of Iowa and New Hampshire would be tolerable.

But as we shall see, the best solution to this problem involves not merely changing the schedule of the primaries, but changing their composition.

Party rules. Some states award delegates proportional the votes received by the candidates, others award all the delegates to the popular winner, and others are somewhere in between these two. Moreover, state party establishments in some states changed the rules at the last minute to benefit their favored candidates. In 2008, several states whose establishments favored Rudy Giuliani made their states winner-take-all. After Giuliani bombed out early, these rules ended up helping John McCain, who won most of the winner-take-all states while Mitt Romney won most of the proportional states.

It is probably unnecessary to attempt to impose uniformity in delegate allocation from the top down. It would be far better for the national party to simply require states to put their primary rules in place early, and not let them change them later on. The presidential primary rules should be finalized at least two years before the first primary is held. For the 2012 election cycle, that would mean by the end of 2009. This is before candidates have declared for office and it would the sort of last-minute shenanigans that happened in 2008. This would also benefit conservatives since party establishments tend to support more moderate establishment candidates than the grassroots does.

Establishment control. While the primary system may appear democratic on the surface, it is actually controlled by a fairly small elite. To see this, consider how the race for the nomination plays out. No candidate can hope to personally reach enough voters to win the nomination. They are largely dependent on media coverage and advertising. The media typically anoints a few candidates as 'frontrunners' who will get most of their attention. The others struggle to get noticed and when they are, it is usually with a dismissive comment that they have no chance of winning.

Polls are often used to justify this treatment. But early polls, months before most voters begin paying attention, simply reflect name recognition. This in turn is a reflection of previous media coverage. For example, in 2008 Rudy Giuliani led polls for months, but he only cracked double digits in one state in the actual primaries.

The other major factor in becoming a frontrunner is fundraising, particularly early fundraising. While candidates may eventually raise a fair amount from small-dollar contributors, their early fundraising usually comes from a relatively small number of large-dollar donors. These donors are particularly concentrated on Wall Street. The relative support for the bailouts of the banks and automakers amongst Republicans, despite the fact that the automakers employ a lot more people, reflects this influence.

Only twice since World War II has the establishment lock on the Republican presidential nomination been broken, in 1964 and 1980 (not counting 1984). Two other times conservatives came close, in 1976 and 1952, when the nomination was actually stolen.

Weakening the establishment hold on the Republican presidential nomination is inherently related to the issue of the composition of the primary electorate.

The primary electorate. Exit polls have revealed several facts about who voted in Republican primaries that ought to disturb conservatives. One is that a substantial minority of the voters in Republican primaries were self-identified democrats or independents. While we don't know the motives of all these voters, we do know that some portion of them intend to sabotage the process. In 2000 in Michigan, there was an organized effort to sabotage the Republican primary by voting for John McCain over George Bush.

The influence of these voters tends to select more moderate and less conservative nominees. This might be tolerable if such nominees had better chances of winning general elections. But the opposite is true. Nominees who run as conservatives consistently do better in November.

The other problem is that many voters seem quite uninformed about the candidates that they are voting on. A fair number of McCain voters cited immigration restriction as their top issue.

Both these problems can best be addressed by narrowing the pool of people who select the Republican presidential nominee. What is needed is a way to screen out most democrat infiltrators and also discourage the entirely uninformed voters. Actually, such mechanisms already exist in several states. There are caucuses, which require voters to actually show up to a meeting. There are conventions, which require being proactive enough to get appointed as a delegate and go to a meeting, possibly in another city. There are also closed primaries, which at least require that you take the trouble to identify with a party before the day of the election.

All these mechanisms, particularly the first two, would increase the influence of the grassroots of the Republican party, who are more likely to be active and informed about the candidates. This would help to nominate more conservative candidates, who are also more electable.

Adopting a caucus or convention would also do a lot to mitigate the primary scheduling problem. More informed voters would be less likely to swing en masse to whichever candidate had 'momentum', according to the media, after winning or doing 'better than expected' in one primary. This in turn would decrease the relative influence of early states over later states and so mitigate the rush to move primaries earlier and earlier.

Adopting a caucus or convention system is particularly necessary for the 2012 election cycle. In 2012, the democratic nomination is all but certain to be uncontested. If Republicans don't fix their nomination system, democrats will be free to sabotage it without missing any elections on their side.

The 2012 election may seem a long way off, but the best time to address this issue is now, when the new leadership of the Republican party is being elected. There are contested races for Michigan Republican Chairman and Republican National Chairman. Has anyone asked the candidates where they stand on this issue?

1 comment:

Unknown said...

Thank you for bringing to light the primary problem. I believe the national party should have more influence in how the state parties. It is a national election. I also think we should look at a three tier primary process. An initial national primary, to eliminate state influence and to narrow the field to 3 or 4 top canidates. Allow a sufficient time for campaigning and then have a final national primary to select the final canidate. This may not be the answer and there maybe many negatives. But to your point something must be done.

On another note, I have taken a stand to turn our congress around. We now have only 38 solid conservative senators and 144 conservative representatives, based on past voting records. Visit therightagenda.com, subscribe and let's build a coalition to make a conservative difference.

Thank you for your article.
Tony Corsaut