Report on Albuquerque Municipal Elections, October 4, 2011

Last Tuesday, October 4, the city of Albuquerque held its non partisan municipal election. The city moved from a traditional precinct based Election Day model with 98 consolidated precinct locations to 49 vote centers allowing voters to vote at any location in the city. The city ballot was pretty thin with only 2 competitive district races, but included some bonds as well as an advisory referendum on whether Albuquerque should keep its red-light camera program. Interestingly, despite the lack of competitive contests, turnout was up on an unusually wet Albuquerque day with approximately 15% of registered voters turning out, which is up about 5% from 2 years ago when the city experienced a competitive three way mayoral race. Thus, the red light camera issue must have mobilized votes (which went down 53% to 47%).

My teams of graduate and undergraduate students in research design courses were out in the field observing voting and the implementation of the new voting system. The city used both ballot on demand printers, electronic poll books, and a network connection to access voter information and upload voting information so that voters could not hop from one vote center to another voting multiple times.

There were quite a number of glitches to the system as they day started. Several presiding judges were unable to login to the system and relatively soon after opening many vote centers’ network went down, preventing them from accessing voter information through the network. In some cases, voters were encouraged to go to other locations or come back later, but in some locations presiding judges just sent voters on their way with no clear advice. To get the system up and running many vote centers had to revert to their local server and not connect to the Internet. This had to be done by technical experts as judges were not trained on this contingency. By about 10:30 AM things smoothed out and systems started functioning at least locally to process voters effectively. Over the course of the day, vote centers appeared to move back and forth between the network and the local server to process voters.

There was a lot of variation in how these problems affected voters. In some places, these problems and other issues, for example Sandia High School also witnessed a fire drill, which led to some incredibly long lines. Some voters waited over an hour to vote. Processing the ballots was the major culprit in these cases as we only saw lines to check-in, but there were never lines for voting or for placing the ballot into the M-100 optical scanner. Although two computers and printers were in every location, some vote centers needed more stations to process voters efficiently. We saw Sandia High School get an upgrade to 3 processing stations around 5:00 PM, a much needed item given the approximately 70 people standing in line waiting to vote.

We also saw quite a bit of variance on voter identification. Albuquerque, unlike the state of New Mexico, has a voter photo id law. While most judges processed voter’s identification without any problems, some presiding judges used strict rules for names. So, if a voter was Bob in the voter registration file, but showed a photo id with Robert, the voter was given a provisional ballot. This also affected a number of senior women, who were registered as Mrs. Joe Smith for example, but of course, have no photo id with that identity. This is a potentially troubling aspect of voter identification laws and the principal agent problems associated with Election Day activities.

It was clear that vote centers are all about location, location, location. Locations on major streets and in major work areas were quite busy. Locations tucked away within neighborhoods and off the beaten path had short lines, if any. We were in one vote center that processed over 1000 people on Election Day and went to another one in the same district that process only a few hundred. Thus, careful planning of locations is essential.