Our democracy is broken. Why can't technology fix it?

If there was any will, there'd be a way.

459 Shares
Share
Tweet
Share
Save
    Image credit:

    Something is rotten in the state of America. For a nation that has for so long promoted itself as a global champion of democratic ideals, we have a rather difficult time practicing what we preach. Outdated election mechanisms like the Electoral College and potential interference from hostile foreign powers aside, Americans have historically proven themselves reticent to participate in choosing their leaders.

    Turnout for presidential elections hasn't topped 65 percent of the eligible population in the past 100 years nor has it even come close to cracking 50 percent for midterms over the same period. During the 2016 election, just 63 percent of the U.S. civilian voting-age population showed up at the polls, according to the US Election Assistance Commission, with just five states -- Colorado, Maine, Minnesota, New Hampshire and Oregon -- managing to break 70 percent participation. Only 42 percent of Hawaiians bothered to vote.

    "If voting changed anything, they'd make it illegal," pioneering feminist Emma Goldman once quipped. But it isn't why we vote that has led to such dismal electoral participation rates but rather how. At the start of the 20th century, America replaced its original oral-ballot method in favor of today's anonymous paper ballots. Aside from allowing early and absentee votes, little of our electoral system has changed in the 118 years since. Granted, 32 states as well as the District of Columbia do allow their residents to vote via more modern methods than punch cards -- email, e-fax or web portal -- but only for local, municipal races.

    The amount of time we spend electing our government representatives is too damn high. Election cycles last for months with candidates and incumbents often announcing their intentions to run a year or more in advance, then furiously fundraising and courting potential voters in the subsequent run-up. Yet even with all that expended effort, only 55.7 percent of eligible voters (that's 86 percent of registered voters) actually went to the polls in 2016, according to the Pew Research Center. But what if Americans had more incentive to vote than simply doing their civic duty? What if, instead, voting were compulsory? It's certainly worked for Australia, where registered voter turnout hasn't dropped below 91 percent since the system was instituted in 1924 and those who do refuse to vote are subject to a minor fine.

    This policy has forced candidates to appeal to the whole of the nation instead of pandering to their political base, leading to more stable, moderate leadership over the past 94 years, Stewart Jackson of the University of Sydney told HuffPost in March.

    "The original intent, or one of the arguments, for compulsory voting was that it would make elections about policy. We'd stop pleading for people to vote and just talk about policy. You wouldn't have to spend all your time with a get-out-the-vote effort," Jackson said. "Now, we do have campaigns about policy. It's not just trying to appeal to a particular sector to vote. You appeal to everybody. The parties have gotten good at appealing to the middle voter."

    What's more, the winning candidates are also afforded a stronger mandate to lead, given that they managed to garner 51 percent of the vote from roughly 90 percent of eligible voters. By comparison, Trump won the support of just 28.7 percent of America's almost 219 million eligible voters in 2016.

    Estonia is different. The EU member country, its western border buffeted by the Baltic Sea, is home to 1.3 million people -- roughly equivalent to the populations of Houston or San Diego. It allows its citizens to vote online in local elections as well as parliamentary and European Parliament races. Its i-voting system was floated in 2001 and first implemented in 2005's local elections, where around two percent of the ballots were cast via the internet. And in 2007, Estonia became the first nation in history to allow for internet voting in a national parliamentary election.

    I-voting leverages the Estonian national ID card, which is also a smart card. It not only serves as proof of identity but also enables residents to remotely authenticate their identities online and generate legally binding digital signatures.

    The i-voting system is a bit different from absentee voting here in the US. Estonians are allowed to vote during the run-up to Election Day, specifically from the sixth day before the election to the fourth. Voters are allowed to change their vote as many times as they like during that period but cannot change or annul their choice via the internet on Election Day itself. That can only be done in person at a physical voting station. Casting your vote there, of course, negates any internet vote you'd cast before, maintaining the "one person, one vote" principle.

    There were already a number of factors in place by 2005 that helped Estonia prepare for national internet-based voting. "Estonia perceives new communication technology very positively; the IT sector is strong there, and Estonians are well-acquainted with computers and the Internet," Daniel Bochsler of the Centre for the Study of Imperfections in Democracies at Central European University notes in a 2010 research study. "Large parts of the population have internet access at home, at work, or in public Internet stations." For example, in 2005, more than three-quarters of Estonians declared their income tax via the internet. The mandatory issuance of national ID cards capable of verifying one's identity online helped as well.

    However, as Bochsler points out, internet usage in its current form is highly socially selective. That is, certain groups -- largely young, affluent, highly educated males who are likely already regular internet users -- are more likely to benefit from the convenience of online voting. "High education and high income are two of the three factors that correlate most strongly with political participation in North America, Western Europe, and in new EU member states," he points out. So while the process of voting has gotten easier in Estonia, this phenomenon might help explain why voter turnout did not significantly increase between 2007 and 2015, only climbing from 62 percent to just over 64 percent even though the percentage of people voting online grew from 3 percent to 30 percent over the same period.

    Of course, the i-voting system has not been without its detractors. In 2014, a team of security researchers from the University of Michigan and the UK-based Open Rights Group published a study stating that i-voting ballots could be changed or corrupted remotely, should one be able to install malware on the system's servers. Both the Estonian National Electoral Committee and Estonian Information System Authority dismissed these claims, arguing that the attack vectors were not feasible in the real world and likely politically motivated given that the study's researchers were connected to the Estonian Centre Party, which has been critical of the i-voting system since its inception.

    In 2016, the University of Oxford released its own independent study of the Estonian voting system. It pointed out that while i-voting "may work well for a close-knit society such as that of Estonia," many of its enforcement policies rely on interpersonal relationships and would need to be further clarified and codified should it be scaled up for larger nations.

    Would such a system work here in the US? In theory, yes. But first a number of aspects would need to be addressed, including electoral infrastructure, ballot security, public and political will for the plan, and access to the system for all Americans.

    In terms of infrastructure, a number of states already allow for some form of online balloting -- albeit only as a means of absentee voting and generally only for military personnel and their dependents under the Uniformed and Overseas Citizens Absentee Voting Act (UOCAVA) of 1986. Arizona, Missouri and North Dakota all allow for absentee ballots to be cast via a web portal. Twenty-one states and DC allow for service members to cast their votes via email or fax, another seven states allow only for fax transmissions and the remaining 19 states demand that physical ballots are sent via the US postal service.

    Online voting can't fix the Electoral College.

    The federal government has toyed with the idea of internet voting. In 2003, the Federal Voting Assistance Program (FVAP) launched the Secure Electronic Registration and Voting Experiment (SERVE). It was designed to enable citizens living abroad, as well as service members and their dependents living either in the US or overseas, to cast their ballots electronically. This experiment would have covered 100,000 Americans living in 51 countries had it ever gotten off the ground.

    Unfortunately, the FVAP quickly ended the experiment after a scathing 2004 exposé by The New York Times. In that article, the NYT cited a panel of four computer-security experts who had reviewed the program and had argued that the $22 million experiment was inherently unreliable and riddled with "fundamental security problems that leave it vulnerable to a variety of well-known cyber attacks, any one of which could be catastrophic." As such, the panel advocated that the experiment be dismantled immediately.

    University of Michigan Professor of Computer Science J. Alex Halderman and students drove this point home in 2010 when, taking up a challenge from DC to see if SERVE could actually be hacked, they managed to not only quickly disable the system but also manipulate votes and even have the system play the University of Michigan fight song whenever a ballot was cast. Halderman and a team of researchers from the University of Melbourne subsequently demolished a similar internet voting system that was to be deployed in the state of New South Wales.

    "Voting over the Internet is a really bad idea," Vanessa Teague, University of Melbourne professor of computer science, told USA Today in 2016. "We haven't yet solved important issues like authentication, dealing with malware, ensuring privacy and allowing voters to verify their votes."

    An online voting system "relies on the user's computer being trustworthy. If a virus can intercept a vote at keyboard or screen, there is basically no defense," David Dill, a Stanford computer scientist, told MIT Technology Review in 2012. "There are really fundamental problems. Perhaps a system could be tightened so some particular hack won't work. But overall, systems tend to be vulnerable."

    Dan Wallach, a professor in the systems group at Rice University's Department of Computer Science, echoed those sentiments. He hopes that an American version is still "a long, long way, because the Estonian system's not particularly secure."

    "Voters' computers are a cesspool of malware and bad configuration and lack of security patches."

    He pointed out that since the system relies on physical ID cards with smart chips, "the computer knows which votes are yours. That's bad from a security perspective, never mind that the Russians also just tried to do a denial of service attack on the entire Estonian internet." This system also is susceptible to bribery and coercion attempts. "I can take your national ID card away from you," Wallach stated. "Without it, you can't vote."

    Americans' computers would be equally vulnerable to online electoral interference, Wallach argued. "Voters' computers are a cesspool of malware and bad configuration and lack of security patches," he said. "That's not acceptable when you're planning a national election that foreign nation-state adversaries might have an incentive to tamper with."

    "We want to make sure that the system is secure, accurate, recountable, accessible and transparent," Jeanette Senecal, the senior director of elections for the League of Women Voters, told Engadget. "So that you can really see from beginning to end and ensure the security of the data. Especially right now, given what happened in the 2016 election and the investigations that are going on there, and the fact that the intelligence community did say that state [voter registration] systems were hacked.

    "I think that online voting is further away now than ever, for us as a country," she continued. "And there's a lot of issues we would have to work through if that was ever something that was going to move forward."

    Then again, it's not like the e-voting machines we use today are any better. Ensuring the accuracy of online banking and shopping transactions is a completely different breed of problem than online voting. In the case of banks and vendors, losses from fraud or the actions of a malicious third party can be written off. But not so with voting. You can't just go losing votes. The ballot being cast must be verifiable, and there must also be a physical record of the vote in case the losing candidate demands a recount. Without these, there would be electoral chaos, throwing the results of the race into question and undermining the public's trust in our democratic process.

    Our current crop of voting machines simply cannot ensure that level of trust. According to a 2015 report from the Brennan Center for Justice at New York University School of Law, an estimated 42 states are using ballot machines that are more than a decade old. What's more, 13 states employ electronic voting machines that do not produce paper ballots, making post-election audits impossible. And given that 20 percent of voters in 2016 cast their ballots on such machines, according to The Washington Post, we have no way to guarantee that their votes were not tampered with by the Russians or even simply "flipped" due to the decrepit programming of the voting machines themselves.

    The good news is that there are ways to protect against such shenanigans. States can, and generally do, perform security tests on voting machines and systems in the run-up to elections; however, the scope and depth of those tests differ depending on which state is doing the testing. Ohio, for example, employs the National Guard to test its public voting systems while others, well, do not. Post-election audits are another key to guaranteeing the accuracy of the machines' results, but unfortunately, only New Mexico and Colorado currently employ post-election audits that are capable of detecting cyberattacks.

    Some US counties are considering integrating auditing capabilities directly into the voting machines. Wallach has spent the better part of a decade working with election officials in Travis County, Texas, where Austin is located, to create the STAR-Vote system. "That stands for secure, transparent, auditable and reliable," Wallach quipped. "You know how computer scientists love acronyms."

    After spending a long weekend with cryptographers, statisticians, systems engineers and people versed in human-computer interfaces, as well as the election officials who run Travis County's office, the team came up with a preliminary design that is extremely easy for voters to use. "From a voter's perspective, I'm interacting with a touchscreen computer that prints a ballot," Wallach explained. "Then I put the printed ballot in a ballot box. That's the user experience."

    However, the backend is far more complex. "The security perspective is really interesting because we have two separate ways of adding up the votes. We have a box full of paper. Adding up boxes full of paper is annoying, right? It's slow. You have to handle the paper. But it's very secure. It's hard for Vladimir Putin to tamper with a box full of paper," Wallach said. "On the flip side, we have electronic records. Those are super easy to add up. On the other hand, it's easy for Vladimir Putin to tamper with electronic records."

    To ensure that hostile nation-states keep their fingers off the scales of American democracy, Wallach's team employed a bunch of modern cryptographic tools, specifically homomorphic cryptography. "The homomorphism is that you can add the cyber text together without decrypting [it] first, which means anybody can look at the encrypted ballots and add them up and get an encrypted total without being able to decrypt, which is super cool," Wallach beamed. "The election officials have the key material to do the decryption. It's like a public-key crypto scheme. Election officials can decrypt a total and provide a proof to you that they decrypted it correctly."

    Election officials can then draw a sampling of paper ballots, typically as few as 30 of them, and compare it against the encrypted electronic records to ensure that they sync. "That sampling process is something that was designed in California and is now used in Colorado and is spreading to other states slowly but surely," Wallach said. "We can use a statistical sample to make sure that the paper artifact for the electronic artifacts matches up.

    "We were doing blockchain back before blockchain was hip," he continued. "Each of these voting machines has an internal hash chain of all of the votes that it's cast." The receipts that voters take home with them are therefore basically a hash point, allowing them to prove that they did in fact vote without revealing who they voted for -- mitigating threats of bribery and coercion.

    "We were doing blockchain back before blockchain was hip."

    What's more, the system also defends against electronic interference. Say, for example, a bad actor was able to corrupt the programming of a STAR-Vote machine so that when a voter selected candidate Alice, it would print Alice on the paper ballot but secretly record a vote for candidate Bob on the digital ballot. Wallach points out that standard sampling audits should be able to catch such shenanigans, but "that's not very satisfying. It would work, but we want more."

    With STAR-Vote, election officials would be able to scan the barcode on the voter's paper ballot, which would send a signal to the voting machine itself instructing it to decrypt the corresponding digital ballot. "If you voted for Alice, the screen said Alice, the paper said Alice, but the cyber text said Bob," Wallach explained, "then this process would allow you to catch a machine with what amounts to a signed confession of its cheating."

    Unfortunately, even the most clever of ideas can be felled by a swift kneecapping by bureaucracy. Specifically, Travis County couldn't find a vendor willing to actually produce the machines. "Let's just call it unfortunate," Wallach said. "But there's nothing that says that a state or federal organization couldn't pick up the baton and complete this thing and make it an open-source solution. One of the pragmatic aspects of [STAR-Vote] is that it'll run on completely off-the-shelf hardware."

    Another issue is that even if states were motivated to update their voting machines and harden them against outside interference, where are they going to get the money? A 2017 study from the Brennan Center estimates that updating the balloting system nationwide would cost us $1 billion. Just updating the paperless machines would run between $150 million and $400 million.

    Doing so would require an act of Congress on par with the Help America Vote Act of 2002, which ponied up $3.6 billion to update the previous generation of outdated voting machines. Even that act was not without its shortfalls. While Congress did provide the funds, it did not set updated standards by which the new voting systems needed to conform. "[States] bought new equipment before [Congress] worked out good standards," Wallach said. "The equipment that they bought wasn't very good to begin with -- that applies to security analysis as well as just engineering. The quality of the components and how they were put together was poor. What we're seeing is these machines that were purchased in the early 2000s are all physically failing now."

    And in the current political climate, with Congress incapable of passing even mundane legislation, the prospect of procuring those sorts of funds is wildly unrealistic. Even if it did suddenly start throwing money at the problem, it wouldn't make a difference in the short term.

    "Realistically, [voting reform] is the sort of issue that flies beneath the radar of most voters."

    "Between now and November, even if Congress magically coughed up a couple billion dollars, we simply couldn't spend it fast enough," Wallach said. "The expertise isn't there; the equipment that's for sale is not good enough. Even if a county or state did buy a ton of gear, even if a vendor could somehow deliver it on time, the planning and the training and the testing wouldn't be there." To be fair, Congress did authorize $380 million in the latest omnibus spending bill for states to improve their election security. However, most of those funds are going toward things like "better network security, intrusion detection, fire walls, training, some degree of software and hardware updates," Wallach explained. "If we're lucky, a little bit of threat analysis and planning [as well].

    "It's a Band-Aid on top of what we've got now," he concluded. "Realistically, [voting reform] is the sort of issue that flies beneath the radar of most voters," Wallach argued. "Once you get past who's running for president and who's running for Congress and Senate, once you get below that, maybe governor, voters are not particularly well informed in most cases and will vote by party affiliation."

    A paralyzed Congress and potentially compromised president aside, American voters themselves have become ambivalent toward even registering to vote in the first place. This not only lowers the population of prospective voters but also skews election results, which, in turn, impacts future policy decisions. Specifically, requiring people to manually register themselves causes the voting population to overrepresent the views of older, whiter and richer Americans while underrepresenting younger, less-affluent people of color.

    To combat this trend, as of 2014, some 21 states have implemented online-registration systems -- that covers roughly 47 percent of the American population. "Online voter registration saves taxpayer dollars, increases the accuracy of voter rolls, and provides a convenient option for Americans who wish to register or update their information," a Pew Charitable Trusts study from that same year states. However, there are still a number of issues with these systems that can be addressed to further improve their effectiveness.

    Most of these systems force citizens to have either a state-issued ID or driver's license in order to use them, a requirement that disproportionately impacts low-income, minority, elderly and disabled individuals. The problem is further exacerbated when politicians put their thumbs on the scales of democracy in order to gain an advantage, as Alabama Republicans did in 2015 when they closed 31 DMVs in majority poor and black counties.

    This is exactly the sort of shenanigans that the League of Women Voters is resisting. "We've really been pushing to have online registration programs be more inclusive than what they currently are," Senecal said, "making sure all NVRA -- National Voter Registration Act -- agencies are tied into the online system.

    "In reality, if you're a first-time voter, under the Help America Vote Act, you're required to show identification the first time you go to vote, so why can't you be on the rolls?" she continued. Senecal pointed out that the entire reason for allowing people to register to vote at the same time that they apply for a driver's license is so that there will be a signature on file when the person goes to vote. "But if you're already showing ID," Senecal said, "just collect the signature in person on Election Day."

    Beyond that, select states are beginning to utilize novel methods to increase voter-registration levels, including advocating for same-day registration, which a 2015 University of Florida study suggests boosts participation. "Election Day registration is the reform that has repeatedly demonstrated through research and actual election implementation to increase turnout," Senecal argued. "Especially in communities that are underrepresented."

    However, actually motivating voters to get out to the polls offers unique challenges. The League of Women Voters argues in favor of early-voting practices such as polling places "that include evening and weekend hours, especially the weekend before Election Day," Senecal said. "As an example, in New Hampshire there's no early voting; they just had this northeaster [on March 13th] and there was a little town election yesterday, and you couldn't go vote. Many people couldn't go vote."

    Voting advocates in California, for example, are in the midst of a registration drive through the state's county jails as part of the American Civil Liberties Union's Unlock the Vote campaign.

    "No one should be denied their constitutional rights," Los Angeles Supervisor Mark Ridley-Thomas, who co-authored the plan, told the Los Angeles Times in February. "I think voter registration efforts in the jails ought to be viewed as a significant piece to anti-recidivism and reentry." So far, more than 600 inmates have registered to vote, out of an county jail population of 17,500. Only those serving time for low-level offenses like parole violations are eligible, mind you. Those serving time in state or federal prisons (as well as inmates found to be mentally incompetent) are ineligible.

    "I think voter registration efforts in the jails ought to be viewed as a significant piece to anti-recidivism and reentry."

    Even without considering voting rights, incarceration causes other electoral issues -- specifically, where these people are incarcerated. Since the decennial census counts prisoners as residing in the counties where they're incarcerated rather than where they lived before, when those inmates are released back into society, they are moving back into areas where their votes aren't represented.

    "That means that the people who have voting rights in those areas have a greater percentage of power than other people in their state," Senecal explained. For example, "if there's 200,000 people per district and 50,000 of them are in various prisons, that means 150,000 people are represented by a congressperson versus 200,000 in another district."

    California has also recently launched a new program that pre-registers teens to vote, building off the National Voter Registration Act of 1993. On Jan. 1st, 2019, it will begin automatically registering 16- and 17-year-olds when they get their driver's licenses, enabling them to vote as soon as they turn 18. The program is expected to enfranchise as many as 200,000 people annually. Governor Jerry Brown signed a bill in February codifying the pre-registration program into law as an expansion of the state's new Automatic Voter Registration system, making California the second state to do so.

    Oregon was the first. Its Automatic Voter Registration system went live in Jan. 2016. According to The Nation, "Eligible voters who have a qualifying interaction with the DMV are notified by mail that they will be added to the voter rolls, unless they decline registration or opt out within 21 days by returning a postcard to the state's election authorities." Nearly every other state in the nation (save for California and North Dakota, which do not require registration in order to vote) conversely demands that people take affirmative action to register themselves.

    Of the 288,516 Oregonians who registered to vote for the first time in 2016, 66 percent of them were signed up through the AVR. What's more, 36 percent of those signed up through the AVR -- some 67,902 people -- went on to vote that year. The result: Between 2012 and 2016, voter turnout in Oregon increased by 4 percent, compared with just 1.6 percent nationally over the same period.

    The push for automatic registration appears to be gaining momentum. As of 2018, nine states and DC have implemented AVRs with 15 more having submitted legislation to install or expand these systems this year. Another 10 states are already debating similar legislation introduced in previous congressional sessions, according to the Brennan Center.

    Researchers are turning to algorithms to prevent gerrymandering.

    In the end, our electoral system remains in the same state of disrepair as the rest of our aging infrastructure. Just like Mississippi's bridges, our voting operations are out of date, grossly underfunded, running on equipment decades past the end of its service life and now pose a hazard to the citizens they were designed to serve. This has resulted in anemic turnout for any race other than marquee presidential contests, and voters' lack of interest is compounded by both bureaucratic and systemic roadblocks that prevent sizable percentages of the US population from casting their votes. These low turnouts have only served to further entrench partisan politics, shifting the focus of elections from substantively debating policy to simply riling up one's political base hard enough to get them to the polls. In turn, this entrenchment has led to even sleazier partisan tricks such as the gerrymandering of congressional districts, splitting and concentrating political opposition to more easily remain in power.

    But for all the electoral challenges the republic currently faces, numerous nonpartisan organizations and individuals are working tirelessly to improve our voting system. The League of Women Voters has long lobbied for automatic voter registration, a valuable enfranchisement tool that drastically increases the political voice of America's most-often overlooked populations -- a mechanism that states across the nation are increasingly adopting on their own -- while the American Civil Liberties Union's Unlock the Vote campaign has helped register almost 1,000 more. Rice University Professor Dan Wallach and his team have spent the better part of a decade building the next big thing in secure voting machines. The STAR-Vote system offers the possibility of elections free from foreign interference -- if only they could find a vendor to produce it. And Wendy K. Tam Cho, a professor at University of Illinois at Urbana-Champaign, has brought congressional redistricting into the Supercomputing Age with the development of PEAR. But despite its ambitious goals, that project must reckon with technical issues of its own before being brought online for the public's benefit.

    These are all examples of the reforms that can be accomplished by putting country over party. But while they're certainly not alone, these efforts aren't nearly enough. American democracy cannot flourish with only a fraction of its citizens remaining politically engaged. In order to scale these solutions to the point where they can positively impact both political institutions and the people whom those institutions are supposed to protect, all Americans must make their voices heard to their elected officials, whether through voting or by protest. Because if we don't fix our democracy, nobody else will.

    Comments to this article were available for the first 24 hours after publication, and have since been closed. We appreciate all of your participation in the discussion.

    From around the web

    ear iconeye icontext filevr