Why I oppose IRV for Urbana

Early this week (11/27) I filed a formal objection to including the Instant Runoff Voting for Urbana referendum on the February 2008 primary election ballot. For those who came in late on this issue, I’d like to share my reasons for doing this.

First, a disclaimer. I filed the objection on my own behalf, not as Vice-Chair of the Champaign County Democratic Party, and not at the request of the party or in collaboration with any official affected by the referendum. I object because I believe that IRV is a bad idea for progressive government in my home town. In my opinion, the petition itself and the concept behind it are flawed for technical, mathematical, and political reasons.

Technically, the referendum contained in the IRV for Urbana Petition is not a municipal referendum submitted to the voters for their approval. The News-Gazette reports that the proponents intend it to be a binding referendum, but it has about half the signatures required for such an initiative under 10 ILCS 5/28-7. It may not be an advisory referendum either, since it orders Urbana to enact, repeal, or amend such unspecified ordinances and statutes as required to implement a concept described in the petition. In any case, it is unclear about how it affects primary elections. Does it eliminate them, subject them to IRV, or leave them untouched? These ambiguities prevent voters from making an informed decision on the matter, so the petitions are void and the proposed referendum should not appear on the February ballot. (Since you ask, I think it does not affect primaries.)

Mathematically, the proponents are trying to achieve the impossible. IRV voting is, they claim, eminently more fair and democratic than the traditional “plurality” method. I disagree. Amateurs and experts tried for a hundred years to perfect a satisfactory voting methodology for elections involving more than two candidates. Several ingenious schemes were devised, but all were wanting in some respect. That is, every method proposed could produce results that would violate one of the fundamental principles of “fair elections”. Then, halfway through the 20th century, a young mathematician turned economist effectively ended the hunt by proving (Arrow’s Theorem) that an eminently satisfactory multi-candidate voting methodology does not exist. Since we currently employ a very democratic (understood by all voters) system whose effects and strategies are clear and tolerable, I suggest below that progressives in Urbana continue to make good use of the system we already have. The plurality method can be described as: “Vote for the candidate of your choice. Whoever gets the most votes wins”. Sometimes unsatisfactory? Perhaps, but Arrow showed that all methods are. Simple? Transparent? Easily verified? Universally understood? Absolutely.

Finally, at the practical political level, IRV will accomplish nothing for progressive government in Urbana. Our city has become a model of modern good government, a cosmopolitan oasis in central Illinois. We didn’t need to “fix” the vote-counting process to get this done. We built this city the old-fashioned way, with important issues, excellent candidates, generous donors, scores of organized, hard-working activists, and a progressive local Democratic party. Why should we now dismantle the very system that has worked so well for Urbana?

Change for its own sake is often a waste of resources, and it can sometimes be counter-productive. IRV proponents have the burden of proof to demonstrate that implementing a novel election scheme will produce significantly better results for Urbana voters without exposing us to unacceptable downside risks. I believe they have failed to meet that burden, and I cannot support IRV for Urbana.

8 Comments so far

  1. pattsi, on 11/29/07 at 7:09pm said:

    This proposal just proves one more time that history repeats itself. IRV was the very voting system used during my undergraduate days at the University of Colorado, knowing in those days as the Hare System. Shortly after I graduated, this system was eliminated. Shifting votes in the IRV manner skews the results distictively.
    Pattsi Petrie

  2. Esther Patt, on 12/2/07 at 11:51am said:

    IRV is only a good idea under one scenario: three candidates (no more) and the voter likes 2 of them. In every other scenario it is less democratic and confusing. If the 3 candidates were Barak Obama, Trent Lott and Newt Gingrich, how would you rank your 2nd and 3rd choice? I wouldn’t bother. Many, possibly most voters won’t rank 2nd 3rd or 4th choices.

    What happens if after the “instant” run-off, still no candidate gets 50% plus 1? The ballot question doesn’t address that. What really gets me is that in a 4-way race, we could have a result where the winning candidate was the third choice for many of the people who are recorded as voting for him/her. This would happen if most of the voters who chose the 3rd or 4th place candidate chose the other 3rd or 4th place candidate as their second choice. How can any honest person claim that an elected official has majority support when the candidate was the third choice of perhaps half the voters who “elected” him/her? If these 3rd choice votes put the second runner-up ahead of the one candidate of the four who received the most votes on the first “ballot”, I would call instant run-off “instant rip-off.” It certainly is not closer to real democracy than we have now.

  3. Al Klein, on 12/3/07 at 5:37am said:


    Yes, Thomas Hare proposed a “Single Transferable Vote” (STV) system in the 19th century to elect members of parliament. I assume it was used that way at Colorado, to elect members of student government? I’m not surprised that college students would quickly learn how to game STV and skew the results, as you observed.

    Instant Runoff Voting” (IRV) today is essentially STV employed to elect a single winner. It might better be called “Instant Sequential Runoff Voting” to distinguish it from the traditional (plurality) Runoff Voting used to pare down a long list of candidates before asking the voters to make their final selection. That method trims several candidates at once from the bottom of the list; “IRV for Urbana” trims them sequentially from the bottom of the top. Another difference: Runoff Voting gives voters time to re-evaluate the reduced list before the final vote is taken. To save money, the IRV computer will be programmed to transfer your vote “instantly” to as many less desired candidates as required to both trim the list and produce the winner in one fell swoop.

    PS: An “Instant Runoff” may not produce timely results, but that’s another story.

  4. Al Klein, on 12/5/07 at 7:12pm said:

    An Electoral Board hearing was held Tuesday (12/4) about my objection. Since only one of the legal arguments was discussed at the hearing, I’ll confine myself to that.

    The argument is: The IRV petition appears to call for a “binding referendum” for a municipality. State law requires that a referendum binding the City of Urbana be supported by a petition from 10% of its registered voters (about 2,000). The petition filed by “IRV for Urbana” purports to have only 1,062 signatures. Therefore, the petition is void, and the referendum cannot be held.

    Durl Kruse, leader of the IRV group, confirmed that the petition did call for a binding referendum. Mr. Kruse stated that Mark Shelden had assisted him in the preparation of the petition and had suggested the lower signature requirement. Mr. Shelden acknowledged making that suggestion, but testified that he had also advised Mr. Kruse to check everything with the State Board of Elections. Mr. Kruse said he had sought other counsel instead before filing the petition.

    The hearing was recessed until next Tuesday 10 a.m. Thursday 12/13.

  5. curious, on 12/6/07 at 1:12am said:

    Thanks for the update. I don’t see how the referendum can go forward without the collection of new signatures for another election given the facts. It’s interesting that Kruse blames Shelden, yet Shelden gave Kruse the best advice he could–to talk to the State Board of Elections. Kruse admits not following this advice. Ignorance of the law is never a defense in the courtroom.

  6. curious, on 12/6/07 at 1:20am said:

    After further thought, the right thing to do would be for Kruse to withdraw the referendum. If he doesn’t, then he’s saying that he doesn’t care whether the statutorily designated percentage of voters are consulted before a binding referendum goes forward. It’s also not like he’s off by a couple signatures…he’s off by 50% (1000 people). To do anything less than withdraw is insulting to the citizens of Urbana.

    In my opinion this is just further evidence that this referendum was put forward with very little forethought or advance research.

  7. pattsi, on 12/7/07 at 5:01pm said:

    Even though all of this conversation may end up being moot, I want to point out another aspect of this system. When it was used at my undergraduate college, many times the candidate’s campaign materials would state “vote for me for #2.” This is an excellent strategy to win under the Hare system. So think about this related to the democracy of voting no matter how many candidates for a position.

  8. Al Klein, on 12/13/07 at 2:35pm said:

    The Urbana Election Board ruled 3-0 today (12/13) that the IRV referendum cannot be held on 2/5/08 because it was not requested by the minimum number of voters required by law for a binding referendum.

    All credit for this victory goes to my able counsel, Ruth Wyman of the Kirchner Law Office.

    At a noon meeting of the LWV, Durl Kruse and I presented our views on the IRV concept (under severe time constraints). The meeting was recorded, and I presume it may appear on UPTV Channel 6.

    Critiques invited.

    Update 12/15/2007:

    The official findings and decision of the Board were delivered to me this morning. The gist is:

    The Board FOUND:

    “That it is uncontroverted that Respondents’ petition contains a little more than half of the signatures (1,062) required by statute.”

    “That Respondents’ testimony and argument are not persuasive, in that, they apparently seek to have this Board ignore the relevant and controlling state statute without providing any case law or contravening statute(s) that allows or mandates it to do so.”

    The Board DECIDED:

    “That because the numbers of signatures for placing a binding public question on the ballot to change the way a home-rule municipality elects its officers is set by statute and therefore, not able to be amended by this Board, it is despositive of this matter. Consequently, this Board need not decide on the Petitioner-Objector’s other objections.”

    ” … Therefore, the referendum question put forth in Respondents’ petition may not be placed on the February 5, 2008 ballot.”

Leave a comment

You must be logged in to post a comment. Don't have an account yet?
Register here.