On the 2020 Iowa Democratic Caucuses
February 5, 2020 — Dan Wallach
It’s hard to write a definitive analysis of a developing news story, but here’s what we can say about the Iowa caucuses, and what lessons we can learn to help us in future elections.
First, what exactly happened? As best we can tell, the Iowa Democratic Party (IDP) rolled out some new infrastructure for collecting results from the local caucuses. This app, and its corresponding backend systems, appear to have been developed by a contractor for the IDP with a very short timeline, and a limited budget. According to various news stories, many caucuses either failed to get the app working, or never even tried and instead used a telephone reporting system, which had long wait times. A day later, preliminary results were announced.
What lessons can we take away from this? Here are some questions and answers:
Was the caucus hacked, or was this “just” a failure of some kind?
Unless we learn otherwise, we should assume that Iowa’s issues have nothing to do with malice and everything to do with planning and execution. For example, a New York Times article reports:
“The app wasn’t included in the chair training that everyone was required to take,” said Zach Simonson, the Democratic Party chair in Wapello County.
“When you have an app that you’re sending out to 1,700 people and many of them might be newer to apps and that kind of stuff, it might have been worth doing a couple months’ worth of testing,” said Mr. Bagniewski, the Polk County chairman.
Will this impact who wins Iowa?
Each caucus has a paper tally sheet, and all of those presumably still exist. This means that the IDP will eventually get them tallied, and the correct number of delegates will be allocated to the candidates. Of course, there may be impacts on public perception, but the presence of the paper records at least provides a verifiable backup.
Will this impact other primaries or caucuses?
Primary elections generally use the same vote casting and tabulation systems that are used in the general election in November. Caucuses, on the other hand, are quite different. Some reports are saying that the same app was to be used in Nevada’s caucuses on February 22. It’s safe to assume that they will now be considering alternatives.
What does this mean for online voting or the use of computers in other aspects of elections?
There are several vendors that offer purely online voting systems, whether using web browsers or smartphone apps. When these systems have been subject to independent scrutiny, security flaws are generally discovered. Most concerning of all, these online voting systems have no voter-verified paper ballots, which means that there’s no backup in case the online systems fail. This is why most technology experts are opposed to online voting. A consensus report of the National Academies of Sciences, Engineering, and Medicine recommends that “elections should be conducted with human-readable paper ballots” for exactly this reason.
Of course, computers are essential to rapidly and accurately tabulating ballots. That’s why auditing techniques are being used in many states to ensure that the paper ballots and their electronic scans yield the same winner of the election. These audits ensure that no computer failure or hack could undetectably change who wins the election.
Computers play a role in every other aspect of elections, from getting information to voters about their polling places, tracking whether voters have cast their ballots or not, and of course disseminating the election results. So long as we have those paper ballots, along with procedures to audit them, then errors anywhere else can be detected and corrected.
Perhaps this is the most important lesson from Iowa’s caucuses. Problems may be inevitable, but so long as we have some sort of reliable ground truth, we can always get to the correct answer. And while it may take additional time to get the right answer, that’s preferable to getting no answer at all.