Reform the caucus system

David Yepsen has a good column in Sunday's Des Moines Register urging Iowa's political parties to improve the caucus system. He reasons that Iowa is less likely to retain its first-in-the-nation status if our state parties do not correct some of the flaws in the caucus process.

I would go further and state that Iowa does not deserve to remain first unless the parties make some changes in the caucus system. Actually, if I were in charge of reforming the nominating process, I would ban caucuses for the purposes of presidential selection. The parties in Iowa will never adopt primaries, though, because of New Hampshire's law stating that it must hold the first primary.

After the jump I'll go over the reforms Yepsen proposes, which would go a long way toward addressing the flaws in the Iowa caucus system. I will then add a few ideas of my own.

For background, here are links to the diaries I wrote last year on the Iowa caucus system:

How the Iowa caucuses work, part 1 (basic elements of the caucus system)

How the Iowa caucuses work, part 2 (corrects an error in part 1 and discusses who is over-represented and who is under-represented when delegates are counted)

How the Iowa caucuses work, part 3 (why it's hard to turn out caucus-goers)

How the Iowa caucuses work, part 4 (more about why caucus turnout is low)

How the Iowa caucuses work, part 5 (on second choices and caucus math)

How the Iowa caucuses work, part 6 (on how precinct captains help their candidates before caucus night)

How the Iowa caucuses work, part 7 (why it's hard to figure out how well the candidates are doing in Iowa)

How the Iowa caucuses work, part 8 (on the many ways to win your precinct)

How the Iowa caucuses work, part 9 (analyzes common arguments made in favor of the caucus system, along with my response to those arguments)

Yepsen offers seven reforms to the Iowa caucuses:

Involve county auditors

I'm with Yepsen here:

In Iowa, county election officials should run caucuses just as they run primaries and elections. They should report results at each county, just as they do other elections.

The poll workers who staff the polling places on Election Day could be recruited to staff each caucus. This injects an element of fairness, objectivity and expertise into the process.

Conduct secret-ballot voting

Iowa Republicans already do this, but Democrats have to stand in a group with other supporters of their candidate. Again, I'm with Yepsen--let people make their choices without having to tell all the neighbors.

Report the number of caucus-goers who supported each candidate

Again, the Republicans already do this, but the Iowa Democratic Party refuses to release those raw numbers. Instead, we hear about "delegate equivalents," which obscures the number of people who preferred each candidate. Statewide, Bill Richardson only received about 2 percent of the delegates, and Joe Biden only 1 percent, but it's clear they had a lot more support than that.

In my precinct, about 3 percent of the caucus-goers stood up for Dodd, 8 percent stood up for Biden, and 10 percent stood up for Bill Richardson. However, since none of those three were viable, their supporters did not "count" toward the candidates they liked best.

The Iowa Democratic caucuses could be a glorified straw poll, like the Iowa Republicans hold, and that would be fine with me.

Allow absentee ballots

Yepsen notes that Maine allows absentee ballots for the caucuses. This would enfranchise people who are housebound or cannot get off work on caucuse night.

Yepsen says absentee ballot voters would not be able to realign if their first choice were not viable, but I don't see why we couldn't have absentee ballots that allow voters to mark a first, second and third choice. These voters could be realigned in absentia.

Pick a new day or time

Again, I agree with Yepsen: it would be preferable to hold the Iowa caucuses on a Saturday night after sunset, which would reduce the number of people who can't come because of work obligations.

Ban same-day registration

Yepsen thinks that "Allowing someone to just show up, claim residence in the precinct and participate is an invitation to fraud."

He would like to see new registrations cut off a week or two before the caucuses, so that the precinct organizers would have an up-to-date list of who was eligible to participate.

I am not sure what I think about this idea. In principle, I support same-day registration, and while I think some people did fraudulently caucus in January, I am not convinced that their numbers were high enough to justify doing away with same-day registration.

Set national standards

Yepsen would like to see consistent rules about whether independents and crossover voters are eligible to caucus, or whether caucus-goers have to be members of the party in question.

There are good arguments on both sides of this question. In theory, it would be nice to have one set of rules across the country, but I don't think courts would uphold that reform. There are longstanding precedents in election law allowing states to set their own rules for primaries. If state parties sued to retain their right to set rules for caucuses, I suspect they would be successful.

I do agree with Yepsen that "the parties need to consider uniform caucus tabulation systems, appeals processes and criminal penalties for falsifying caucus results, just as it is a crime to falsify primary-election results."

Speaking of reforming the Iowa caucus system, here are two changes I would like to see that Yepsen did not mention:

Abolish the 15 percent threshold for viability at the precinct level

As I mentioned earlier, the Iowa Republican caucuses are basically a straw poll conducted by secret ballot. The Iowa Democratic Party has caucus-goers divide into preference groups and count. Only candidates supported by at least 15 percent of the people in the room are considered "viable." There is a realignment period, during which people can switch to a different candidate. The second and final count at a precinct caucus determines the number of county delegates awarded to each candidate. Then, the county delegate totals for each candidate are translated into the number of state delegates we would expect each candidate to win.

The viability threshold is fundamentally unfair. Why should my vote count toward the candidate of my choice only if at least 15 percent of my politically-active neighbors agree with me?

Disparities in turnout at the precinct level and the number of county delegates awarded to each precinct introduce other types of distortion.

My precinct awarded six Polk County delegates and had 293 caucus-goers. Another suburban precinct had about 425 caucus-goers and awarded seven county delegates. It took 44 supporters in my precinct to get one Polk County delegate, but in my friend's precinct, a candidate needed 64 supporters to be viable.

A different suburban precinct had almost as many caucus-goers as mine (about 275), but awarded only four county delegates instead of the six delegates assigned to my precinct.

One Des Moines precinct had about 210 caucus-goers and awarded three delegates. More than half the attendees were in the Obama group. However, Clinton and Edwards were both viable (with the support of at least 15 percent of the people in the room). So, Obama, Clinton and Edwards each got one county delegate from that precinct.

I would rather see everyone's vote count the same at a caucus. Precincts would still elect delegates to county conventions, as the Iowa Republicans do, but the reported results would be the raw numbers of supporters for each presidential candidate. In the previous example I described, Obama would still get credit for having more supporters in the room, even if he didn't win more county delegates than Clinton or Edwards.

If the Iowa Democratic Party is unwilling to do away with the 15 percent threshold, and unwilling to stop reporting the results in terms of state delegate equivalents, I propose a different reform:

Prohibit supporters of viable candidates from realigning

The Nevada Democratic Party adopted a caucus rule stating that only supporters of non-viable candidates, or those who caucused as "uncommitted" during the first division into preference groups, are allowed to realign.

That would greatly reduce the kind of gamesmanship I described in this diary on second choices and caucus math. I'll give you two examples that occurred this past January. In a Clive precinct that awarded five delegates, it appeared at the first count that Obama would receive 2, Edwards 2, and Clinton 1. The Obama and Clinton precinct captains "donated" enough people to make Bill Richardson viable, so that the delegates went 2 for Obama and 1 each for Edwards, Clinton and Richardson.

In a Waukee precinct that awarded four delegates, more than half of the people stood in the Obama corner. Clinton and Edwards were also viable, and those groups sent enough supporters to the Richardson group to make him viable. That meant that Obama, Clinton, Edwards and Richardson each got 1 delegate from the precinct. A friend and first-time caucus-goer who was in the Obama group at that precinct was still visibly angry when she told me the story a week later.

If the Iowa Democratic Party adopted this reform, there would still some room for gamesmanship. A precinct captain could ask some supporters to caucus as "uncommitted" at first, creating the opportunity to move them to another group. However, I believe the Nevada rules would make it more difficult to pull this off. It shouldn't be easy to deprive a rival of a delegate by moving some of your supporters to a third candidate.

Reforming our nominating process will require much more than tinkering with the caucus system. Perhaps other states should get to go first, although I would hope that the early nominating contests would always be in small states.

I would also like to see other changes, so that a candidate could not lose the popular vote in a state but win more delegates there.

In addition, I believe it was wrong for Obama to emerge from some low-turnout caucuses with as large a net delegate advantage as Hillary won in the Ohio primary blowout.

Something needs to be done so that a few thousand caucus-goers in a small state don't wield as much influence over the presidential selection as hundreds of thousands of primary voters in a large state.

How would you like to change the caucus system or the nominating process?

  • entice participation...

    My wife came up with a simple and elegant solution, but one that would involve changing the rules for everyone, not just Iowa.

    I'd proposed the idea of a lottery for first-in-the-nation status, but she proposed that first-in-the-nation status be granted to the state the turned out the highest percentage of voters in the general election.

    This rewards political participation with media attention (and dollars) in the next election.

    I'm absolutely in love with this idea. Iowa could still maintain it's first-in-the-nation status, but we'd have to earn it.

Login or Join to comment and post.