The City of Minneapolis will be conducting a hand count of its first ranked choice voting elections in 2009. While voting equipment is used to tally RCV elections in other US jurisdictions, none was certified for use in this year’s election in Minneapolis. The city’s election officials have stated that they look forward to the possibility that certified equipment to be in place for the next scheduled election in 2013.

It’s important that every precaution be taken with something as fundamental as the right to vote and we believe the City has designed a manual count procedure for this year’s election that is as transparent, verifiable, complete and thorough as possible. There are checks and balances built in throughout the process and two-person teams will perform each step of the process, which means that there will be oversight at every step along the way.

While the procedures the City has adopted go beyond what is necessary to assure a fair and complete count and we have recommended and advocated for more time-efficient procedures, we appreciate the reasons why the city is taking this approach for the 2009 elections when Minneapolis voters will be using RCV for the first time.

First, the procedures provide a guide for a recount should one be needed in this or a future election. Second, the count ensures consistent information recorded for all candidates and races. Third, the consistency and thoroughness of the count will provide the information necessary to fully evaluate voters’ first-time experience using ranked choice voting in all races and their effective use of rankings in casting their ballot. Fourth, the information gathered from the process, including the hands-on experience of both voters and election officials, should provide a basis for improving and streamlining the process even further in future elections.

Current equipment will be used at the polls to provide voters with important error notification and to tally first choices. As the City has always done, it will aggregate first choices for all races and submit these data to the Secretary of State’s office for publication to the public on Election Night. This means that outcomes for races in which a candidate reaches the threshold will be known on a “preliminary and unverified” basis on Election Night.

Note that the election night results will provide totals of first, second and third choice rankings in each race. But only the first choice totals are usable totals. The candidates, media and the public should be advised that they cannot infer or predict outcomes through raw totals of rankings.

The City will begin the hand count the day following the election and report results by race as each race is completed, with the initial ward results reported by November 13th and final city-wide races reported by December 21st. The long timeline is based on an estimated 90,000 ballots, which is high for a municipal election year. Lower turnout will reduce the timeline of the count.

The order in which the wards and city-wide races will be counted has been determined by random draw. We understand that the City made this decision to ensure fairness, but we believe that the ward races that did not result in a candidate who exceeded the winning threshold according to the preliminary machine totals on Election Day should have been first in line for the count. The order in which to have counted this group of wards could have been determined randomly, but doing this group first as a set would have produced results more quickly for races that did not have “preliminary” results on the Election Night and eased concerns about the hand count timeline.

The way the count is designed, the Park District races will be aggregated and reported when the precincts comprising these districts are counted. Similarly, the city-wide mayoral, Park Board at-large and Board of Estimate and Taxation races will be reported last, after all of the precincts are counted.

It’s important to understand that the time-intensive part of the hand count is upfront. All of the ballots will be counted at the precinct level and triple sorted to create unique groups of ballots with like rankings for each race. Once this sorting process is completed, ballot summary sheets will be entered into a spreadsheet, which will be used to calculate the results for each race. This final step will go quickly.

The order in which the wards will be counted and reported and other detail will be available on the city’s website at http://www.ci.minneapolis.mn.us/elections/.