Can Technology Save Democracy?

Representative democracy may have run its course. It’s time for liquid democracy.

Can Technology Save Democracy?
Illustration: Oliver Munday

On March 11, in a state parliament election in West Australia, 24 candidates made only one campaign promise: If they won, they promised to vote on every bill according to the wishes of their constituents, as determined via an app called Flux. (While the votes are still being tallied, it looks unlikely that any will win.)

Flux’s app is one of a handful of new platforms that aim to use technology to let people participate directly in politics, at scale. All are premised on the fact that–around the world–representative democracy isn’t working well. But technology could potentially help end corruption and lobbying, allow people to delegate votes to trusted friends rather than politicians, and empower experts in a field to meaningfully impact policy.

Does Democracy Work?

In 2015, shortly after Donald Trump announced that he was running for president, polls found that only 19% of Americans trusted the government “always” or “most of the time.” (The survey has not been repeated, but presumably, the numbers have not improved.) Only 11% approved of Congress.

Those numbers are historic lows; in 1958, when a poll first asked the question, 73% of Americans said that they could trust the government most of the time. The results can be partisan–people are less likely to trust the government when the opposing party is in power, and Republicans are less likely to trust government, in general, than Democrats. But the overall message is clear. Most people don’t think democracy is working in its current form.

A Princeton study that looked at 1,779 different policy issues found that economic elites and business groups had substantial impacts on U.S. policy, while citizens and mass-based interest groups had little or none. Lobbyists often write bills.

Congress may spend more time fundraising than actually writing laws. Gerrymandering–redrawing congressional districts to favor one party–makes it possible for representatives to essentially choose the people they’re representing, rather than voters choosing representatives. Active voter suppression, often of minorities, means, unsurprisingly, that fewer people vote.

The problems may stem from our form of government. “The problem, fundamentally, is representative democracy,” says Nathan Spataro, cofounder of both the Flux political party in Australia and XO.1, the startup making the software that powers the Flux app. “It is not that your politicians are corrupt, it’s that the politicians are corrupt because of the system. You don’t have to look far to watch how politicians start their career, and how then the system fundamentally changes them by the time they get to the end of it.”

Liquid Democracy

True direct democracy, in which every member of a society votes on everything, could eliminate the problem of lobbying, but has rarely existed. In ancient Athens, assemblies made up of all the citizens gathered to make decisions (women and slaves were not allowed to be citizens). In some Swiss cantons, citizens can participate directly in local government. In both cases, though, issues facing voters were relatively simple and limited in scope. While direct democracy might be the ideal–a government that’s literally by the people and for the people–it’s hard to scale up. In a large society with complex issues, it isn’t possible for even the most dedicated individual to keep up with every possible item that requires a vote–or have an informed opinion about them.

Representative democracy, which ideally solves that problem, also struggles with size. “One of the key problems of the U.S. political system is that it runs into scaling limits,” says Bryan Ford, a computer scientist who leads the Decentralized/Distributed Systems lab at the Swiss Federal Institute of Technology.

Sixteen years ago, Ford began thinking about what he calls delegative democracy, now also known as liquid democracy. “The whole idea of delegative democracy is to try to create a representative system that responds to the needs of individuals but also scales,” he says. “In some sense, delegative or liquid democracy is an approximation to the completely impractical idea of fully participatory, direct democracy.”

It works like this: Rather than asking citizens to vote on every issue, it gives each person the power to vote or to appoint a delegate to vote for them. Unlike a typical representative, that delegate could be changed at any time depending on the issue.

If you feel like you understand a particular issue, you can choose to vote on related proposals. For subjects you don’t know much about, you can appoint a friend or a known expert in the area. It solves the problem of education; no single person–including representatives in the current government–can be well-versed on everything.

“It’s just not possible to educate millions of people in these kinds of issues,” says Spataro. “So what we do is leverage the skills and talents that people in society have already got to make policy decisions. If a piece of legislation came up about aeronautical engineering regulation, would you want pilots and engineers voting on that, or your local butcher or lawyer?”

Voting Tech In Action

One early experiment began in Sweden in 2000, where educators and students created a platform called Demoex to let citizens in a Stockholm suburb comment and vote on local issues. In 2002, a 19-year-old student from the Demoex party won a seat on the city council, promising to vote based on decisions from the online platform. She was reelected in 2006, and another representative from the party won in 2010. (However, when the party failed to win a national election, it folded.)

Others have used similar tools, though not fully in the way that some in the liquid democracy movement now envision. In Germany, the Pirate Party, which advocates for online freedom, has used digital platforms to choose candidates and issues. In Italy, the populist Five-Star Movement used online tools in similar ways.

Now the technology is much more sophisticated. XO.1 uses bitcoin’s blockchain technology as a secure foundation for its platform; the blockchain records data across a decentralized network of computers, so voting results can’t be changed. Bitcoin can only record a small number of transactions at a time–usually about three or four votes a second–so XO.1 also developed a second layer of its technology that can work at scale. In a stress test on March 8, they proved that the system can handle more than 1 million votes a minute, or 1.5 billion votes in 24 hours.

The company plans to use what’s called a private audit log, so an independent third party can verify links between someone’s real identity and their blockchain identity. No one else will be able to link the two; XO.1 patented a two-step process that uses something called an “oblivious shuffle” to ensure that each vote comes from a registered voter, while protecting the anonymity of those votes.

“If you imagine you get 10 people in a circle, and each person writes a note–maybe their favorite color–and shuffles these notes between each other. At the end of the process . . . each person knows that their note was included and there were 10 notes,” says Max Kaye, CTO of XO.1. “The end result is a collection of votes that’s associated with a collection of identities, but not individually associated.”

It’s a way around the problem of coercion: If a particular vote was associated with a particular person, someone else could force them to vote in a certain way, because they’d have proof after the fact.

The Flux app will also have a “duress” pin code, so if someone’s standing over your shoulder telling you to vote for a candidate or policy, the app will show a random result instead of confirming your own vote. “If you can’t prove how someone voted, you can’t coerce them,” says Kaye.

A True Global Democracy

In Silicon Valley, the Y Combinator-backed nonprofit Democracy Earth is also building a liquid democracy platform. On November 8, 2016, as Americans were voting, the organization released Sovereign, their open-source platform that they envision being used first by organizations to make governance transparent and auditable and accountable.

The startup tested its platform in a pilot in October 2016, when Colombia held a referendum to accept or reject a peace agreement between the government and guerrillas in the country’s long-running civil war. Many Colombians live outside the country and some couldn’t participate when the government decided not to reopen voter registration in time for the referendum; the pilot let them symbolically vote.

As a test of liquid democracy, it also let people delegate votes and vote separately on different parts of the referendum, rather than a simple yes or no. The majority of people voted “yes” on the referendum, but “no” on a key point that would have allowed the guerrillas to participate politically–a nuance that the actual vote (which rejected the peace deal) lacked. As the founders wrote in a Medium post after the referendum, this type of less polarized decision-making could help make divisions between voters less bitter.

The angry and perplexed reactions of both post-Brexit and post-Colombia show how clueless people were about what others were thinking and feeling. It will be hard to overcome these limits if we insist on an infrastructure and a culture of participation in politics that permits only a binary vote every four years. Polarization, apathy, and powerlessness are outcomes of a democracy in which sovereignty is reduced to a paper in a ballot box.

Liquid or delegative democracy isn’t guaranteed to work, of course. In California, where citizens can vote directly on initiatives and referendums, special interests spend hundreds of millions to sway voters. In 2016, for example, the drug industry spent $109 million to campaign against Proposition 61 (it lost). In total, campaigns for and against statewide measures spent nearly $500 million on the election. If big business stopped using lobbyists to chase politicians, they could still pour money into influencing voters or their delegates.

Another issue: If people can choose delegates to vote on their behalf, those delegates may not be diverse. In Iceland, after the banking collapse, Parliament began crowdsourcing a new constitution and randomly selected 950 citizens to work on it. But when the public voted on a group of citizens to actually draft the new constitution, most were from the capital and none were working class.

In Germany, the Pirate Party struggled with some people getting disproportionate power. “If an individual can only delegate to one person . . . people kind of pick the most popular superstar on that topic, and then the decision-making power can become super-centralized,” says Ford. “I think there was one professor in Germany that become the liquid democracy superstar, and he got a completely outrageous disproportion of votes.” (This is a problem, Ford says, that can be addressed in how software is designed, potentially by asking voters to delegate to multiple people rather than a single person).

Participation is another challenge. In Iceland, the general public was invited to comment on the draft constitution, but only a relatively small percentage of people did. Those with the most time and interest had a greater say. If even presidential elections suffer from poor turnout, getting citizens involved in smaller policy decisions would obviously be much harder, even with the convenience of an app.

The biggest challenge, though, is fundamentally shifting how government works and how it relates to the citizens it ostensibly represents. “I think we need to build a new model that makes the existing model obsolete,” says Democracy Earth cofounder Santiago Siri. “That’s how technology and innovation operates, and we have the means to do it with blockchain technology right now.”

Siri and his cofounders initially tried to win a seat on the city council in Buenos Aires, but after witnessing corruption–a federal judge, for example, asked them for a bribe when they wanted to run in the 2015 election–and after losing in previous elections, he now believes that the system needs to fully change rather than trying to change it from within. His ultimate vision: a global democracy powered by blockchain, rather than governance by individual countries. Technology would securely manage identity, voting, and representation, replacing passports and traditional elections, and allowing people to make decisions globally. That change would obviously take time.

“We are aware that political time moves at glacial speeds,” says Siri. “Sometimes what is needed for a big change is generational change . . . The thesis that we have for our notion of post-nation-state governance comes from the fact that we are already seeing ways of transacting online that are completely disruptive to the central control of banks and nation states.”

For those, like Flux, that are trying to insert liquid democracy into the current system, the challenge is winning elections, though it seems unlikely they’ll prevail in this most recent one. Some of the changes that would make it easier for candidates to win, such as limits on campaign spending, are difficult to enact without already having a seat in government.

But if it works–even, initially, at a small scale–it has the potential to make progress happen more quickly. “Right now, we have a society that is every day coming up with great ideas to a lot of the problems that we’re facing,” says Spataro. “Government is actually one of the biggest barriers to implementing some of these solutions . . . for us, Flux is trying to empower people to take personal responsibility for making a difference in their societies.”

He also believes that the current system of representative democracy has to be fully replaced, but thinks that the best way to do that is by getting people elected to parliament who can then lead a full shift to a new system. He’s optimistic that it can happen relatively soon.

“I think technology and society are advancing far too fast for this change to not happen sometime in the next century,” says Spataro. “We certainly hope much sooner than that . . . we hope to start converting at least smaller, more forward-looking countries to this model within two decades.”

Read all the 2017 World Changing Ideas

Correction: This article has been amended to reflect Max Kaye’s correct title as is the CTO of XO.1.

About the author

Adele Peters is a staff writer at Fast Company who focuses on solutions to some of the world's largest problems, from climate change to homelessness. Previously, she worked with GOOD, BioLite, and the Sustainable Products and Solutions program at UC Berkeley.