(Bloomberg Opinion) -- The crazy quilt Americans call their election process has produced another nail-biter, with delays caused by everything from postal bottlenecks and voting machine malfunctions to an unexpected shortage of printer ink. The way votes are collected and counted in the U.S. makes recounts, lawsuits and sometimes lingering doubts over an election’s true outcome all but inevitable: You don’t have to be Donald Trump to worry that errors could have tilted a tight count toward the opposing side.
There’s a 21st-century cure for these ills. It’s pandemic-proof, and it solves a couple of other important problems, rendering vote-buying pointless and allowing voters to change their mind if they’ve voted early and some late-October surprise makes them regret their choice. It’s called internet voting, and although it’s widely considered infeasible because a lot of people don’t trust the government to organize it and the technology to perform as expected, this distrust isn’t an insurmountable obstacle. Working around it is an engineering challenge.
E-government pioneer Estonia has been using electronic voting since 2005, when it became the first government to use internet voting nationally. In its 2019 parliamentary election, 44% of voters cast their ballots electronically. Estonia is small, which makes all kinds of e-government easier to scale. It has another advantage, in that 46% of its residents trust the national government. And there’s another reason Estonia has forged ahead with voting online, notwithstanding the kinds of vulnerabilities that have prompted high-trust countries like Norway (68%) and Switzerland (85%) to abandon their e-voting efforts: A clandestine subversion of an Estonian election is too much trouble for too little reward for any major power, even its neighbor Russia, which has perpetrated blunt-force hacking attacks in the past.
In these respects, the U.S. is about as far from Estonia as you can get. It is big, public trust in its government is near record lows, and its elections present a high-stakes, high-reward target for foreign adversaries. Yet notwithstanding the huge challenges of moving its voting system to the internet, the U.S. has good reasons to make the switch. Not only would it address many of the problems laid bare by the debacle of this year’s contest, but it would also bolster the U.S.’s increasingly shaky standing as a technology leader and the standard-bearer for democracy worldwide.
A good starting point for a spec comes from Melanie Volkamer of the Technical University in Darmstadt, Germany, and Oliver Spycher and Eric Dubuis of the Bern University of Applied Sciences in Switzerland, who proposed some technical features of a hypothetical perfect voting system.
One is splitting computational tasks among a number of organizations to make collusion difficult — a feature that mimics the U.S. election system’s decentralization, which most Americans rightly recognize as a strength. As an extra safety measure, the organizations could be nominated by the political parties.
Other recommendations include enabling every voter to verify that her vote has been counted as intended and to change the vote any number of times before a deadline. Estonia has tried its hand at these tasks and posted the code on GitHub, but U.S. engineers, of course, could — and probably should — try alternative approaches as well. The challenge is to make the information accessible to the voter herself but not to others; technology used in cryptocurrencies could come in handy here. Some U.S. states already have already experimented with blockchain-based voting for overseas-based military personnel: It has reportedly worked for West Virginia and for the city of Denver.
Another recommendation from the Volkamer paper is tailor-made for divided societies with mutual distrust among political “tribes.” The German-Swiss team proposed allowing “independent implementations of voting client software” — that is, different voting apps developed and distributed by political parties or groups and thus more trusted by their supporters than any government-issue app. All of them would feed the data into a single back-end for tabulation.
Sound daunting? Let me tempt you: Would you like to vote from your favorite device using an official application provided by your party, change your mind as many times as you like before Election Day, and check that your most recent vote is entered correctly into a system that would count the votes in a distributed fashion, without allowing any of the processors control over the final results? Or would you like to keep the current choice between dragging yourself to a polling station in the middle of a pandemic or mailing off a ballot — perhaps only to have it ignored by court order?
The standard response from e-voting opponents to any plans to build such a system is that end-user devices — our phones and laptops — are the weakest link. They could be infected with malware to attack the entire system, and they could be hacked to de-anonymize votes. Besides, no one has ever built a completely secure e-banking system, for example, and hundreds of thousands of identity theft complaints are filed with the U.S. Federal Trade Commission every year. But the benefits of hacking a bank are obvious even if the data of only a few clients are compromised. With an election, scale is more important. A well-designed, highly distributed system could make localized hacks about as useful as attempts to subvert the current voting machines in one county or another.
Designing and testing out a trustworthy e-voting system would require a big investment of money, time and engineering resources, not to mention a huge public information campaign. But a process that produces the dispiriting tableau of workers trudging through absentee ballots as partisan poll watchers scream from the sidelines, days after an election, is a poor advertisement for the American way.
U.S. leaders and engineers shouldn’t be deterred by other countries’ difficulties in modernizing the voting process. In fact, any responsible U.S. government should see it as a project akin to NASA’s Apollo space program. I know: Just because you can send a man to the moon using a computer with one-millionth the memory of a smartphone doesn’t mean you can put a secure, anonymous ballot box in the hands of every American voter. But if Americans want a chance to refurbish their image and remind the world that they are indeed special, here’s a good place to start.
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
Leonid Bershidsky is a member of the Bloomberg News Automation team based in Berlin. He was previously Bloomberg Opinion's Europe columnist. His Russian translation of George Orwell's "1984" is due out in early 2021.
©2020 Bloomberg L.P.