Whenever voting is involved, one question is ubiquitous: how can we tell if the results have not been skewed due to bribing the voters?
Democracy is a system where people vote for solving issues or to make governance decisions. In the simplest cases, the majority wins which has its own problems, as we will describe later, if a Sybil attack occurs which is funded by bribery in a system where a small percentage of the total number of eligible voters vote and the cost of bribing is relatively small. But as the system itself is about 2,500 years old, dishonest people have learned how to exploit it.
Money = Power
Let’s examine this scenario: a large group of people vote on whether or not to build a factory near their neighborhood. If the majority of them agrees, the industrial company can buy the land from the city and build the proposed factory. If the majority says no, they have to build elsewhere.
Without going into details, as we’re far from being real estate experts, building a factory near housing isn’t in the best interest of people living nearby. The air might get polluted and the noise might become unbearable. Therefore, no one will want to live there, and, as a consequence, the value of their properties will significantly decrease.
But as Carl Sandburg, an American poet, and writer, three-time Pulitzer Prize winner, wrote:
“Money is power, freedom, a cushion, the root of all evil, the sum of blessings.”
Money IS power in voting schemes. Therefore, what would stop an industrial company that makes millions in profit from bribing the voters to cast a ‘yay’ in favor of their plans? In our case, only a small fraction will be affected by the decision, i.e. those who would live the closest to the planned factory. So, if only 5% of voters really care and the rest are indifferent, the ballot is easily won in favor of building the company’s proposed factory by bribing the indifferent voters.
The problem is that in most cases and with most voting, we can’t be sure that all actors are honest. If bribers target the right people, meaning those who are likely to take the bribe and don’t ask too much for it, the party with more money than the collective is able to influence the results.
The probability of manipulation lowers with the rise of the number of electors but it’s not a guarantee. As 2006 Mexican election showcased, even mass-scale bribery is possible. In this incident, it is believed that 8.8% of the population which wasn’t a beneficiary of mostly two social programmes (“Desarrollo Local Microrregiones” (“Microregion Local Development”), and the “Desarrollo Humano Oportunidades” (“Human Development Opportunities “)) could be influenced. During the time of this election, a Global Exchange observer stated:
“As people came out of the polling station, PRI representatives gave them 100 pesos and a T-shirt.”
A similar scheme appeared in Argentina, Nigeria, Kenya, and even in many developed countries but on a smaller scale — in some states, cities, or towns.
Election Fraud Is a Serious Problem
Regardless of what is being decided by a vote, there should be no room for any type of bribery. On a small scale, the damage can be overlooked — for example, if kids vote on who’s looking in “hide and seek” game. The kid that has the most candy and loves looking, could make others vote for him. How harmful could that be?
But when voting bribery affects governments, this is where we have a problem. It could mean that the well-being of many could be sacrificed for short-term gains of a few. Of course, politics may be run on the edge of what’s legal and what’s not, but at the core, voting should stay free from bribery.
What’s the Solution?
You can’t stop people from taking money for their votes. Although in most jurisdictions, bribery is illegal and consequences are imposed on those caught, but this still requires the ability of catching people in the act and is a problem since most experienced politickers are often aware of intricacies of the system in which they operate to avoid such a scandal. The motivation for bribery might be different but one essential element remains the same: those performing the bribes wish to obtain an outcome in their favor at a relatively low cost by influencing those who lack perspective and to cause them to make a seemingless, harmless “change of mind.”
But we can still mitigate such attacks on democratic systems by utilizing systems and processes that make bribery inefficient. The bribers often consider the economic justification of their actions. If the cost of the bribe is more than what they would be able to gain — they won’t go for it. Obviously, bribery is an issue with nation-states or supernational economic regions with central banks that can inflate the monetary supply at a whim with various policy decisions in order to bribe voters since the cost of the attack does not negatively affect those performing the bribe.
Currently, there’s one main scheme that is considered to ensure the integrity of voting. The commonly used scheme is one that comprises a secret ballot, i.e. every participant is voting privately. Only he can know how he voted and if no one can obtain a proof of the final vote. Why would anyone pay for a vote in a particular direction if it cannot be proven to be cast in favor of the briber? Simply put, a bribe would not work in such a scheme. It is also important to note that properly executed voting within this scheme also ensures that the votes are calculated correctly.
But in a world where everyone carries a camera and video recording device in his pocket — a smartphone — the briber could obtain proof of the vote for which he paid. As such, the risk of bribery still exists.
Thankfully, this scheme proves to work most of the time as the problem can be barely observed in most countries. The risk may be higher in developing countries but solving that is some more complex problem, probably individual to every case.
As another solution to the problem of bribery, many point to Distributed Ledger Technologies. But can they really be used to solve this problem?
Voting in DLT
We like to see the technology as a solution to many problems. It surely is, but first, it needs to overcome its own problems. Let’s be honest — voting held on committee-led protocols could be even more dangerous than it is now.
In these protocols, choosing the committee involves voting. Then, the committee processes transactions and decides which to accept, based — roughly speaking — on a majority vote. And so we have, theoretically, two points where the actions could be dishonest. An adversary could influence voting for the committee and then issue double-spending transactions, knowing they will be validated by the committee of his choice. And if the committee is small enough, he could just focus on bribing its members anyways.
As already mentioned, the main motivation for bribery is economic gains. Hence, if bribing is not profitable, there is no point in doing it. When designing consensus protocols, one has to keep in mind that part of the users might be dishonest. The network has to tolerate these bad actors and be able to operate correctly, despite their actions.
But it still might not be enough.
The Concept of DAO
A Decentralized Autonomous Organization is a new type of business model that originated with Bitshares and has grown increasingly popular with dApps building on Ethereum. Just as for driverless cars and IoT factories that require minimum staff, a DAO is a commercial, or non-profit, organization put on autopilot.
In this model, a series of Smart Contracts perform almost all of the tasks that normally would require a human workforce. The shareholders can vote with their tokens on the main issues but most of the time, a DAO makes decisions, executes payments, and shares information with contractorsautonomously and transparently, according to the shareholders’ will.
After ensuring that the DAO works exactly as desired, it’s a dream for the entrepreneurs. They create the company’s logic and, barely doing anything, watch as their numbers grow while enjoying life on their personal island.
Unfortunately, it’s just as with the invention of dynamite — it could be created with the best intentions in mind, but there will always be people who would like to use technology for their own, selfish interests, no matter how much harm they can deal to the world.
A Dark DAO is using the same scheme as DAO but for malicious actions. And here we go back to the voting. With the on-chain voting, even with the secret ballot, a Dark DAO could be a set of Smart Contracts that pay voters automatically after discovering they voted under their influence. In a nutshell, this is just a mechanism that allows automated bribing. Even worse — there is no way to prevent it or stop it, as it operates as a smart contract. No one has to know anyone — it’s just the technology helping bribers to achieve their goals.
Thus, elections in DLTs might be even more susceptible to bribery than in the real ones. It could make bribing more convenient than it has ever been!
What Is Deniable Re-Voting?
In the search for accessibility and convenience, many countries enable the option of e-voting which basically means voting through the Internet. In such a situation, the risk of bribery is even higher as the secret ballot is much harder to achieve in an online scenario.
That’s why another bribery prevention scheme has emerged, although it’s still a theoretical one. It’s called “Deniable Re-voting.” Here, the user can vote multiple times and only his last vote counts. In particular, under a briber’s custody, a voter could make one vote, collect the reward, and then, just before the election is over, secretly vote again, invalidating the previous vote! Getting proof of the voter’s choice is extremely hard in such a situation. It’s quite a complex solution, so if you want to know more about that, here’s a very interesting paper on that concept.
It seems deniable re-voting could become a solution to digital voting, although there is a lot of research to be made before it solves bribery.
How to Avoid Corruption?
Voting is seen as one of the use cases of blockchain and other DLTs. It could become a major part of future democracy as it would provide a transparent ecosystem in which any malicious action would not have an effect on the voting results.
However, the DLTs have to sort out the problem of bribery in voting internally, first. This technology is both the blessing and the curse, considering the topic we’re describing here. But if these problems can be solved for DLTs, they can work anywhere.
And this is what we are working on as well in Aleph Zero. How are we solving that? Disclosure won’t happen overnight, so follow us on the platform of your choice as we disseminate the evidence of our research.