Epistemic status: political navel gazing essay that fails to arrive at a firm conclusion. At the time of publication I’m feeling slightly let down by the result, but I think it’s still worth publishing.
I am in favor of a number of policies (such as drug decriminalization and same sex civil unions) that haven’t been implemented here in Lithuania largely because they are unpopular.
And a small part of me gets angry that the majority even gets a say. That part thinks: “hey most Lithuanians, why do you even hold an opinion on this issue! It doesn’t concern you in the slightest!”1
This seems to have been recognized as a general problem with democracy as a whole. The SEP2 describes it thus:
The idea of “One person, one vote” is supposedly grounded on a commitment to egalitarianism. […] However, it is not clear that giving every citizen an equal right to vote reliably results in decisions that give equal consideration to everyone’s interests. In many decisions, many citizens have little to nothing at stake, while other citizens have a great deal at stake.
Indeed. Is there a solution?
Thus, one alternative proposal is that citizens’ votes should be weighted by how much they have a stake in the decision. This preserves equality not by giving everyone an equal chance of being decisive in every decision, but by giving everyone’s interests equal weight. Otherwise, in a system of one person, one vote, issues that are deeply important to the few might continually lose out to issues of only minor interest to the many (Brighouse and Fleurbaey 2010).
Sounds like a great idea! But:
[…] even if this proposal seems plausible in theory, it is unclear how a democracy might reliably instantiate this in practice. Before allowing a vote, a democratic polity would need to determine to what extent different citizens have a stake in the decision, and then somehow weight their votes accordingly. In real life, special-interests groups and others would likely try to use vote weighting for their own ends. Citizens might regard unequal voting rights as evidence of corruption or electoral manipulation (Christiano 2008: 34–45).
Yep. Even if everyone involved was engaging in good faith, reasonable people would still disagree about what exactly should be recognized as a stake. For instance, while I previously described the Lithuanian majority opposing liberal drug and civil union policies as being low-stake when it comes to those issues, some of those people might respond by claiming that their opposition is based on their strong need to protect their children from being corrupted by drug peddlers and the evil gays, making them high-stake. So this seems like a non-starter.
Likewise, a hypothetical electoral system that estimates stakes based on self-reporting (“please only vote in this election if you believe yourself to have a high stake in the outcome”) would also be useless as everyone would just selfishly declare themselves to be as high-stake as possible.
But what if they didn’t?
Let us imagine a kind of democracy where one’s vote is weighted based on how much one cares about the election. Call it enthusiarchy — rule of those who care.
This is going to be a purely imaginary utopia-question-mark, so just like Plato, I will be handwaving away all implementation problems. Thus:
In our enthusiarchy, the voter’s brain is scanned by an an enthusiasm-measuring device called the enthusiometer.
The enthusiometer is highly precise, undeceivable, and harmless. It cannot be used for nefarious unrelated purposes.
It asks the question “how important is it for you to select this particular candidate on this particular ballot?” and then compels the brain to give an honest, introspective answer.
Perceived electoral impact would probably have to be taken into account, e.g. if you think Candidate X is stellar but has no real chance of winning, while Candidates Y and Z are plausible winners but about equally mediocre, then your enthusiasm for the election as a whole will count as low (despite your high enthusiasm for X)
By “how important”, perhaps it means “how important is it compared to these other important things in your life” or “which percentage of your disposable income would you be hypothetically willing to pay for the ability to cast your vote”.
All voters’ results are then sorted from highest enthusiasm to lowest and divided evenly into ten Enthusiasm Classes. One’s ballot’s value is equal to its class.
So e.g. if you are in the 38th percentile for enthusiasm, that puts you in E-Class 4, which means your vote matters twice as much as an E-Class 2 vote and half as much an E-Class 4 vote.
No drama queens. Your electoral enthusiasm is adjusted to your enthusiasm baseline, so you don’t get rewarded for being a generally neurotic or excitable person.
No utility monsters. Even if the election is the most important event in your life (perhaps you’re one of the candidates!), your E-Class is still capped at 10.
Franchise floor. Conversely, even if you’re the most apathetic person in line to the voting booth, if you manage to get there, you will still be at least E-Class 1.
I suppose that puts people who didn’t bother to turn out into E-Class 0.
The candidate with the most E-Class-adjusted votes wins.
If there are any obvious superficial design improvements to this that I overlooked (maybe there should be more or fewer E-Classes, or they should be of uneven size, or something) then our imaginary society’s very learnèd mathematicians will implement those very quickly.
If put into practice, would enthusiarchy be a good system?3 I think it would have some upsides:
Adjusting for stakes: enthusiarchy will fix the aforementioned problem of “one person, one vote” not reflecting how much stake one has in any particular election. If you’re highly affected, that will increase your E-Class.
Protection against voter suppression: if you’re inclined to complain, justifiably or not, that the other side is unfairly trying to make your side’s votes less impactful, then you will be high E-Class almost by definition.
Better-informed voters: one’s E-Class will correlate with one’s political knowledge (if you care about something you’re more likely to take the time to read a book about it), making enthusiarchy epistocratic.
More serious voters: people who treat the election as a joke by voting for meme candidates like Vermin Supreme will be low E-Class. At the same time, sincere (as opposed to ironically detached) protest votes will still matter.
Seems nice! But there would surely be downsides as well:
Crying wolf: if a political party wants to increase a certain group’s E-Class, they will try to convince them that the present election is the Most Important Election of Our Lifetimes, even if it isn’t true. If you think political polarization is bad, you should dislike this.
Nihilism: on the other hand, if it benefits them to reduce someone’s E-Class, they might want to amplify messaging about how All Politicians Are The Same, or If Voting Changed Anything, They’d Make It Illegal, leading to undue cynicism and disengagement.
(both of the above already happen all the time, but under enthusiarchy we’d get more of this)
Cults of personality: the way Donald Trump likes to emphasize the size of his supporter rallies (in comparison to his opponents’) makes me think he’d be in favor of enthusiarchy. He probably finds it counterintuitive that both an adoring Trump supporter and a reluctant Biden/Harris supporter get one vote each. So if you dislike Trump-like candidates (as I do), or at least that cultish aspect of theirs, then this should give you pause about the whole idea.4
So do the upsides of enthusiarchy outweigh the downsides?
It’s hard for me to say. The upsides seem good, while the downsides seem bad. Some of them seem in tension (e.g. empowering informed voters vs. empowering cults of personality) and unlikely to be true at the same time.
But I think we should step back and ask: why wouldn’t it be a good idea? After all, democracies are already enthusiarchic to an extent:
Optional voting: since voting in most democracies isn’t mandatory, this means they effectively introduce E-Classes 0 and 1.
Representative (as opposed to direct) democracy: according to polls, a double digit percentage of people in most European countries favor bringing back the death penalty. And yet hardly any politicians are trying to do this. Why not? It’s because of a disconnect between declared and revealed preferences: while many citizens might say they support the death penalty, hardly any of them are willing to bother changing their voting patterns so as to reward candidates who would implement this policy. (Compare this to issues like abortion, where declared popular opinion is much more correlated with actual law.) If this is true, then passing laws via the legislature and not referenda (which is what democracies usually do) is enthusiarchic.
Moving towards an anti-enthusiarchy (a system in which one’s electoral power is inverse to their E-Class) seems obviously bad. But if doing the opposite also isn’t good, then that means that we just happened to have have arrived at a system that is about exactly as enthusiarchic as it should be. What are the odds?
Assuming we do want to be more enthusiarchic, but aren’t about to invent the enthusiometer, how do we go about it?
One can come up with a number of crazy electoral schemes where the voters’ choices might serve as a proxy for their enthusiasm, each more elaborate than the next.5 But I’d like to return to the simplest one.
Earlier I mentioned that an enthusiarchy based on self-reported enthusiasm wouldn’t really work as everyone would just assign themselves to E-Class 10 if they could. But is that really true?
Embarrassing anecdote:6 I’m a university lecturer, and last semester I had lost the sheet where I wrote down three of my students’ grades for a particular public speaking assignment. (I remembered the quality of their work but only vaguely.) There really was no good solution to this, so I decided to just ask them which grade they thought they deserved.7 I explicitly told them that if they believed they deserved a 10 (out of 10), I would enter in that grade no questions asked, and if they were being dishonest then that was between them and their conscience.
A very simplistic incentive model of student behavior would here have predicted that all of them would have requested a 10. But what actually happened is that the three students said they believed they deserved the grades 8, 8 and 7. Later I ended up finding that grade sheet and discovered that their real grades were 8, 8 and 6, respectively. Pretty close!
If a secret ballot asked me the question Which E-Class would you like to be counted as for this election?, I think I would at least think about it and not just mindlessly select the highest number. Wouldn’t you?8
So all of that was just a bunch of speculation. Could we test any of this?
A pretty ambitious way to do so would be to attach an opinion survey to ballots.9 On a scale of 1 to 10, how important do you consider this particular election? Then that would give us enthusiasm data correlated with voting preferences.
(of course, this would never be implemented; so an alternative would be to attach the survey to exit polls instead)
The data would let us know which voters have a perceived greater stake, which would let us model the results of a hypothetical enthusiarchic election. Then we can decide if we prefer those results, and possibly start designing a system of self-reporting, elaborate proxies, or get to work on that brain scanner.
I guess I’m out here proposing what they call a “research agenda”. Get on it, political scientists!10
Compare “nothing about us without us”.
I’ll admit that I myself am not very directly impacted by these issues, so it would be hard for me to beat luxury belief accusations here. Still, I have reasons to believe that those that are more impacted are more likely than not to agree with me on these issues.
This isn’t the first time I’m relying on the SEP in one of my essays. Honestly, I really like this encyclopedia; I’m surprised I don’t see it referenced more, even by bloggers who aren’t above citing Wikipedia.
Needless to say, everything in this section is super speculative. I deleted several instances of “presumably” here, but I did that solely for legibility.
On second thought, if a particular voter is reluctantly pro-Harris, but enthusiastically anti-Trump, I suppose they’d still be high E-Class.
Here are some I came up with. I doubt any of them would be a good idea to implement as written, but maybe someone reading this footnote could bounce off them and come up with something better, in a brainstorming session kind of way:
Pay to play: essentially a tax where people pay for extra ballots. To counter the objection that this only will empower rich people, we could make the tax progressive, though then again poor people (who are most often claimed to be politically underrepresented) might be unwilling to spend even a little money.
Nonfungible luxuries: kind of the inverse of the above. If we paid money to low-enthusiasm voters not to vote, that would just make the poor forego political representation to get their basic needs met. But what if instead we paid them in attractive but “useless” goods that cannot be sold, like nontransferable movie tickets?
Tax designation: Lithuania is a country with a percentage tax designation system, i.e. one can choose to allocate a percentage of one’s taxes to an NGO (among other things). (You pay the same amount of money in taxes either way, but you can choose who gets a small part of it.) So maybe in an enthusiarchy you could forfeit this choice in exchange for extra ballots?
Obstacles: maybe you could get extra ballots by doing some useless task, like waiting in a room for an hour? This one seems like it would just reward people who have nothing better to do; also I hate it on a gut level as I have a visceral revulsion to unforced inconvenience.
Accumulated E-Class: every election you show up for, you get a number of ballots that you can choose to spend on the present election or save them for a future occasion. If you really believe the current election is the most important one ever, you’ll be willing to spend all of your ballot “savings” on it, otherwise, you might want to hold onto them. [EDIT:
in the comments points out that this concept already exists and is called quadratic voting.]
Real facts slightly modified for simplicity.
As seen in this video.
I don’t like how this empowers a minority of totally shameless liars who wouldn’t even hesitate to make the greediest choice. But how large is that minority, really?
I kind of think we should do these ballot surveys more often for all kinds of different questions, actually.
Or point me to some research that’s already been done. I spent 5 minutes entering various keywords into Google Scholar and didn’t find anything, so I assume it doesn’t exist, but even I could be wrong!
I searched for references to quadratic voting, and was surprised to find you didn't discuss it (or any discussion about it):
https://en.wikipedia.org/wiki/Quadratic_voting