STAR Voting Cost of Implementation


Clear Ballot's vote tabulators (used in Lane and Multnomah Counties) can be easily adapted to run a STAR election.

 

STAR Voting Cost of Implementation

A number of folks have asked about the cost implications of adopting STAR Voting. This article covers both the long and short-term financial advantages of implementing STAR Voting while taking a look at the components of the cost analysis.

Implementation costs will depend on a number of factors such as location, scale, and election details. Some costs, such as certification, are one time, fixed expenses. Others, such as voter education can very widely and is up to the jurisdiction to choose how much they would like to spend. The analysis below is based on research and estimates for Lane County, Eugene, Multnomah County, and Troutdale, but these citations can be referenced to inform a broader analysis.

In 2017 our team sat down for an hour long, in depth meeting with Tim Scott, Director of Elections for Multnomah County. Scott was reasonably confident that if Multnomah voted to adopt STAR Voting that the implementation costs would be able to be covered by the current election budget without requiring additional funding. He also expressed confidence that it could be implemented in one election cycle and would be ready to go by the next general election.

 

Upfront implementation costs paid for by the county:

Software Update: The election vendor for a given election will need to update their system to run STAR Voting. STAR Voting uses a 5 star ballot which is already commonly used in polling and surveys. Most modern voting machines or tabulators can already read a 5 star ballot, but would require a simple software upgrade to tally those ballots as votes. There are numerous election vendors which can be used and we will cover a couple options below:

ClearBallot:  Lane, Multnomah, and a number of Oregon counties use a system called ClearBallot to design, scan, and tally paper ballots, and while it doesn't support STAR "out of the box," we have checked with the software vendor who has indicated that a STAR voting ballot can be created and scanned using their system. The only missing piece is computing the winner using the rules of STAR Voting.

Using the "Cast Vote Record" output from ClearBallot, Equal Vote has developed an open source proof-of-concept vote counter in the Python programming language. The effort took our intrepid coder a full two hours to write and debug, and the final program was just 40 lines of computer code. This exercise has given us confidence that the software engineering costs will be minimal. When it comes time for the County to work with our software vendor to implement STAR, we have offered this proof-of-concept to help the county to understand the scope of the project and to determine if the eventual quote given is reasonable.

Gravic/Remark OMR: In the interest of keeping costs in check we have also reached out to other vendors so that competing bids can be compared. Gravic has provided a quote for using their Remark OMR software for Eugene City Elections. Lane County Elections (which runs elections for the City of Eugene,) currently uses 4 scanners and this quote is based upon maintaining that capacity. The quote includes four licenses at $4,030, a customer utility which would take the ballot data and calculate both STAR and binary election results for $15,000, as well as an optional $2,000-2,500 if the City would like Remark to draft the ballot design. If the city would like to design the ballots Gravic would consult and then approve the final ballot design at no cost. In total, adopting Remark for Eugene or Lane County elections would represent $19,030 in fixed costs, plus certification. Adding in optional costs such as ballot design and software trainings, Remark software would still come in at under $20k.

Remark was recently used by the Democratic Party of Oregon to tabulate the STAR Voting paper ballots for the Presidential Delegate Elections. Remark was selected as a top option by the Election Integrity Caucus research committee because it met all criteria for best practices, including open source software, independence from the internet, and other election security and user-interface criteria. Remark is compatible with modern scanners and computers and does not require proprietary hardware.

STAR Elections: STAR Elections officiated the 2020 Independent Party of Oregon primary and the star.vote web app behind that election is open sourced, vetted, secure, and reliable. (Star.vote can also be used for informal polling and the public version of the site is the easiest option for those looking to give STAR Voting a quick test drive.)

Anyone wishing to host their own STAR election can currently do so at no cost or STAR Elections can be contracted for more involved or elections. For elections using paper ballots, options include hand counting or scanned ballots, and if desired, results can be tabulated using our spreadsheet tabulator. For those looking to host their own secure elections online, Google Forms can be set up to run STAR Voting.

For vendors looking to add STAR Voting code to their platforms, a number of open source templates and scripts exists in a number of languages which can be used for reference.

Certification: The most significant cost of adding STAR voting to an election vendor's system is the cost of certifying the software. ClearBallot’s Oregon representative has estimated the cost of re-certification to be approximately $50,000. Certification for a STAR Voting specific upgrade is a one time cost -at most- which would come from the county elections budget. Once a vendor is certified to run STAR Voting in one locality others which adopt STAR will not need to pay this fee again. Note that election vendors update software and re-certify on a regular basis anyways, so including STAR Voting code in an upcoming re-certification may or may not represent an additional cost beyond what was already covered in the election budget. This may be a point which could be negotiated.

 

Upfront implementation costs paid for by the jurisdiction adopting STAR:

Voter Education: Adopting a new voting method requires an educational campaign to ensure that voters are comfortable with the new system and will trust the election results. This is an expense which historically has varied widely in terms of how much money is spent and who funds the education campaign. Fargo, North Dakota (which is slightly smaller that Eugene) was the first to use Approval Voting for city elections just this year. According to campaign manager Jed Limke, roughly $35k was spent on the educational campaign between the time when the initiative passed and the first election. Almost all of that was paid for by the campaign and their non-profit, not by the city.

STAR Voting is a type of preference voting, like Ranked Choice Voting, and we recommend basing budgets for education campaigns on Ranked Choice efforts in areas with a similar population, then subtracting the value of resources outlined below which will be provided by Equal Vote.

The Equal Vote Coalition is committed to working to ensure that STAR Voting is implemented properly and voters understand how the system works and how to properly fill out their STAR ballots. The coalition’s multi-pronged plan will significantly mitigate the cost burden on the taxpayers. Here's our plan:

  1. Provide multi-lingual educational materials to the county for how STAR Voting works and how voters should be instructed to fill out the ballot. This will save the county the significant cost of starting from scratch when drafting their own explanatory statements for the ballot and voters' guide.
  2. Print and online materials and media. We plan to make video and print materials for voter education available through the STAR Voting and Equal Vote Coalition web sites and we will offer those same resources to the City or County directly as well at no charge.
  3. Provide clinics for political parties and other interest groups. We expect that political interest groups will be on the "front lines," explaining how STAR works to their constituent groups. We plan to provide educational materials and templates, free of charge, to all local political organizations.
  4. Provide candidate clinics for those running for office under the new system. Candidates running in the first round of STAR elections will be the true educators about how STAR works, and we expect that candidates will be the most effective advocates for thoughtful voting using the new system. The Equal Vote Coalition will provide these clinics, free of charge, for political candidates.
  5. Press outreach. The Equal Vote Coalition is committed to advocating that our local media organizations - newspapers, TV stations, podcasts, radio shows, etc. are well-versed in the STAR system and we'll push them to report heavily on STAR in the form of public service announcements leading up to the first 2022 STAR vote.

These upfront costs are more than mitigated by the tangible and intangible long term savings outlined below.

 

Long Term Savings:

Election cost savings over time

According to the Secretary of State, the cost for recent elections (not counting money spent by candidates) is between $1.71 and $1.91 per voter. This includes printing, mailing and counting costs for our vote by mail system. Local non-partisan elections in most of Oregon, (including Eugene, Lane County, Portland, Multnomah County, and Troutdale,) currently use the Jungle Primary and Top 2 general election process and City Council, County Commissioner, and Mayor races take place on the primary ballot, as well as a few others.

Current STAR Voting ballot initiatives would move all races to the general election which require a charter amendment to change. The few remaining races, such as Judge and School Board elections, could be moved to the general election by ordinance, and doing so would improve the cost savings of STAR dramatically.

If all local nonpartisan races are run just a single time in November, this means that all nonpartisan and minor party voters need not receive ballots nor voter guides for what would, in Lane County, be a primary only for major party affiliated voters. As of our latest voter registration statistics, 83,119 of Lane County's 254,589 voters are not affiliated with a major party. Using an average cost per voter of $1.81, this represents a potential savings of $150,445.39 for Lane County taxpayers, every two years.

These cost savings will be particularly significant when all races in the county use STAR Voting, but even during the transition phase, when some races are using it and others are still considering switching over there will be notable cost savings from reduced printing costs for both the ballot and the voter guide. For the ballot itself, cost savings will occur any time the ballot itself is reduced from three pages to two, or from two pages to one. For the voter guide, less candidates and races represent a savings, as does eliminating redundant costs from having some candidates listed in both the primary and general election voter guides.

Campaign finance and the influence of money in the political system

If no candidate wins outright in May in our current system, the top two candidates have to run another election in November, meaning they campaign for another six months and raise a ton more money, giving special interests even more of a hold on political outcomes. This fact was made clearly apparent by the races in Lane County in the 2018 election cycle. Joe Berney and Sid Leiken, who competed for the Springfield seat on the County Commission raised a combined total of $235,727.84 to compete in a single May election, while Heather Buch and Gary Williams, who had to compete twice in East Lane County, raised approximately twice as much money as their one-election counterparts.

When using STAR Voting for local nonpartisan offices, there is always just a single election in November. The benefit to city and county residents on the whole is enormous when hundreds of thousands of extra special interest influence dollars are not a part of the process.

 

STAR Voting: better outcomes, less expensive

When weighing new proposals, cost is a critical component. This analysis shows that, on the balance, STAR Voting is a sound investment which carries low cost risk upfront and may provide significant savings over time, both to taxpayers and by reducing the influence of special interest money in the political process.