IRV is getting interesting now!

I was interviewed by Martha Brock about the statewide judicial IRV race. Here's a link to the story:

I've been involved in election integrity since 2004, but I actually started getting interested in IRV right after the 2000 election. I first heard about it when I listened to FairVote's Rob Richie talk about it on NPR. Back then I thought it sounded like a method that showed some promise to fix the problems that caused the Gore defeat.

However, I have this terrible problem of not wanting to be a sucker and stake a position on an issue based on something someone else says that might not be true. So I started looking into the Florida debacle and seeing if both reforming the election method (IRV) and the mechanics of the election (the physical casting, counting and auditing of the votes) could work.

The mechanics won out. HAVA was good intentions gone wildly bad with all the money available for electronic touch screen voting machines that had no paper trail. That bit us on the ass in 2004 in our own home state, when nearly 5000 lost votes in Carteret County held up the results of a statewide election for Secretary of Agriculture - and showcased the need for statewide standards for voting machines, counting, and auditing.

Between 2001 and 2004, I sort of lost track of IRV in my desire to get better voting systems, equipment, procedures and auditing requirements. But in the process of working with folks to create what later became one of the country's best election administration systems, I ran into IRV again. But this time it didn't look so good.

The big problem was we looked at the best practices for election integrity and knew that we needed some (if not all) of the following: paper ballots only (DRE sucks even with a paper trail that doesn't print out 100% of the time), good scanners (not ones that don't read some optically visible ink and that read invisible ink - like our old Eagle OpTechs), ballot marking devices like the AutoMARK, counting all votes at the close of election day at the place the votes were cast and not stopping till ALL the votes are counted (prevents election fraud), providing information on under and over-votes for post-election analysis, ballot reconciliation and auditing to make sure there are no fraudulent ballot box stuffing, not using voting systems and software that hasn't been fully tested and certified, escrow software source code, vendor bonding to pay for election mishaps, etc.

The folks working on the Public Confidence In Elections Act knew that they needed tough, high standards for our elections, and they had to resist attempts to weaken the standards from people who wanted to keep their crappy old op-scan and DRE equipment, or from folks who just didn't like paper, etc. One of the first attempts to weaken the Act came from the IRV lobby - they wanted to require that any new equipment either be IRV ready, or be capable of doing IRV sometime in the future. BIG problem was that there was no federally certified election systems, equipment or software capable of doing IRV in 2004. There is none now for that matter. So the folks working on the Act had to make a decision - to weaken the Act to do IRV, or dump IRV. Lots of concessions were made (for instance - DREs with paper trails that don't print out 100% of the votes, pilots for paperless voter verification, etc.) but they stuck fast to dumping IRV. There was just no way to make it work with the Public Confidence in Elections Act then - or now.

After the Act was introduced in the General Assembly in Feb 2005, Paul Leubke introduced his HB1024 entitled "Runoff Changes" presumably to fix the problem folks saw either with the 2004 Supreme Court race for the Orr seat (8-way race with a 23% plurality win) or trying to further eliminate the need for runoff elections (earlier fixes that included lowering the thresholds lessened the number of runoffs and also lowered turnout when runoffs were needed). IRV was supposed to fix that.

The runoff bill dealt with IRV in two ways. The first was a time-limited IRV pilot (two years) in a limited number of communities (up to ten localities one year, and up to ten counties the next), with the SBOE evaluating the elections and reporting the results to the General Assembly. The next was MANDATED judicial IRV under certain circumstances - in elections across the counties (District and Superior Court races) and statewide (Court of Appeals and Supreme Court). The bill really didn't go anywhere in 2005 - mainly because the Public Confidence in Elections Act was needed and passed unanimously and the runoff bill would conflict with major parts of that bill and other election laws. So it was held over to 2006 - the short session. Then many other election measures got added to it, and everyone had to go along with the other parts to pass theirs. So this turkey of a bill got passed.

Although the verified voting community wanted to make some improvements to the requirements for election audits that was part of the 2006 omnibus bill that included IRV, I objected to both of the IRV requirements. Why require a pilot with no built-in way to measure success or failure on one hand, then mandate IRV for judicial elections on the other hand? I remember calling one House member in the summer of 2006 to ask questions about this bill - and this legislator threatened me for daring to suggest that there were bad parts of this bill. Folks only seemed to concentrate on the IRV pilots and forget about the IRV judicial election mandates.

Two IRV elections in 2007 with only one single race using the method (and demonstrating failure) confirmed my worst suspicions about IRV. It caused or showed the potential to cause so many problems with election integrity that we were able to get a pilot extension bill passed with amendments that required voter education programs AND that IRV would have to follow state and federal election laws and guidelines (there were no such requirements in the original pilot). IRV influenced the appointment of County BOE members in 2009 (no IRV supporters made it through). And Cary didn't want to take part in IRV in the 2009 elections. We also got a requirement that any community that wanted to take part in the pilot had to make a public notice in advance of making any decision - so that voters could learn more, ask questions, and make pubic comments.

But nothing was really done nor any attention paid to the mandates for judicial IRV elections. During the pilot, SBOE ED Gary Bartlett admitted that the General Assembly didn't appropriate any extra money to administer or implement IRV elections. That was no big deal for the pilots (only two local pilots in 2007 and one in 2009), but the 2010 election for the Wynn seat on the Court of Appeals led SBOE chair Larry Leake to state in an SBOE meeting that IRV was an unfunded mandate. The SBOE has spent the last couple of months trying to figure out how to do IRV and break the least number of other election laws, regulations and rules.

Of course, over the years there have been many other IRV elections that we could learn from. Mostly we could learn to dump IRV like Aspen is doing, and like Burlington VT and Pierce County WA have done. IRV does not save money, does not increase voter turnout, and is not simple for voters to figure out (in fact the IRV election method is so confusing that the spoiled ballot rate is 300% to 500% higher in some races). But now we have to go through this mess here in NC. Sometimes you have to go through hell to learn something.

None of the candidates really like IRV from what I can tell. They don't know how to deal with it. Two months is not enough time to implement this voting method and educate all the parties. For that matter, trying to use IRV in a May primary when the filing period closes at the end of February. I have gone from supporting IRV to opposing it.

Our Public Confidence in Elections Act is a model election law. It helped us go from a laughing stock in 2004 to being #1 in election audits in 2006, and being one of the top 8 states ready for the 2008 elections. Demos ranked us as one of the 10 best states in protecting voters - but that was not factoring in IRV.

Right now I am not sure how we will end up dumping IRV - will a court case (lawsuit) force its end, or will the General Assembly take it up in January (if not before). But I am sure that this 140 year old voting method that was dumped in the early part of the 20th century in favor of other reforms (like runoff elections) will not be long with us. And our experiences with IRV in NC will help other localities and states keep from making the same mistakes we have.

We can go back to being #1 in election integrity by dumping IRV. We can deal with judicial elections by letting the appointee hold office until the next election in two years - then follow the usual procedure of a November election between the top-two candidates determined by the May primary. We can study and compare the full costs of implementing IRV vs. runoffs and see that IRV will cost more to implement vs. runoffs. Then decide do we want to hold runoffs - and perhaps raise the threshold so that we'll have greater turnout due to more interest in more races. Runoffs are more democratic in some ways than simply letting the plurality winner take office. But those are choices we have to debate and decide on - with all the facts.