I often read advocacy to restrict money in politics. Emphasizing attempts to restrict it may amount to a misdirection of effort. What may be more effective is to arrange matters in such a way that the money and advertising won’t have so much effect on the elections.
There are two points to argue: (1) trying to restrict money has little effect, and (2) another approach may have great effect if we can put it through.
As for (1), I will just quote Jim Mueller (of Wisconsin) [link]:
Political money flows like water around, over, under obstacles put in its way [my emphasis]. One can see this in “amateur” sports. On page C5 of 9/10/13 Wisconsin State Journal reported allegations about the Oklahoma State Football Program. The allegations included pay to “amateur” players for performance, special personnel to tutor and complete work for the players and professors who gave passing grades for little or no work, selective enforcement of positive drug tests, hostesses having sex with players being recruited.
The amounts involved in College Football are tiny in comparison to what the government controls. Even if the Move to Amend effort is successful in amending the constitution, legislation has to be passed and regulators have to enforce the laws. If that happens, there will be more creative ways to circumvent the regulations and make the money flow to people willing to be influenced [again, my emphasis].
Jim goes on to offer a solution in how legislatures would be chosen and how they would operate, and I am in agreement with his proposal; however, I want to talk about single-winner elections, for in case such continue to exist in the US. In doing so, I enter into branch (2) of my argument, i. e. the subargument to convince you that a method is available to decouple money from having any effect on elections.
All single-winner elections at all political levels in the US are currently conducted by the choose-one plurality voting system, also called “First Past the Post” (FPTP) [note 1]. As pointed out by Warren D. Smith, Ph.D. (math) and his co-authors, FPTP is particularly vulnerable to money influence. Please see the writing by Smith that I cite first, and another writing in which Shentrup and Jennings join with Smith to make a similar argument:
- Why score voting should decrease the importance of money in elections
- Duopoly Must Go: An Appeal for Score Voting
“[It may be] more effective to try to reduce the inherent importance of cash, than to wage a potentially futile battle to [restrict money].” — Jennings, Shentrup, and Smith
These authors present reasoning to convince us that FPTP is vulnerable to money influence and that Score Voting, also called Range Voting, is a solution, will resist money influence and put the general public in power rather than the oligarchs.
I want to call to your attention some specific characteristics of the two voting systems that they compare.
FPTP does not give the voters equal power to one another. To explain why I think this, I will repeat an argument from Mark Frohnmayer (of Oregon) and from his father.
Consider an example where three candidates, call them A, B, and C, are running for a single-winner office (e. g. governor of a State). Let’s say that a million voters would be happy with A and equally unhappy with either B or C. So they vote for A; the system (FPTP) allows them to express their exact position, and it gives their expression full weight. But suppose some voters have the opposite opinion. They would be roughly equally happy with B or C, and for them, the election of A would be the end of the world. Call them the “anti-A faction”. I use the term ‘faction’ to indicate any group of voters who have the same values, regardless of whether they are organized or even know anything about one another. How many voters must the anti-A faction contain in order to balance the million pro-A voters? With FPTP, the answer is about two million, because the anti-A faction’s vote will be split between B and C. And so I’m arguing here that since it takes two million in one faction to balance one million in the faction that has the exact opposite valuation on the candidacies, each individual voter in the anti-A faction has only half the political power of each individual voter in the pro-A faction. Therefore, FPTP does not attach equal weight to each vote, QED.
I will now try to demonstrate to you that there is a formal constraint that a voting system has to pass if it gives the votes equal weight. This constraint is that for every possible vote in the system, there is another possible vote that would exactly cancel the political effect of the first vote, its antivote if you will. This is the Frohnmayer balance constraint and it was described by Mark Frohnmayer’s father. When I say “exactly cancel” (or “exactly balance”), here’s what I mean in detail. Suppose some number of people have already voted in the election, and you and I (two individuals) are the last voters, and we are on our way to the polls. Let us say that if neither of us votes, the outcome will be such-and-such. But let us say that if you make it to the poll and I don’t, your vote will sway the election, change the outcome. Someone who won now ties, or someone who ties now loses, or someone who would have lost now ties, or someone who would have tied now wins. Now I arrive and cast my vote. Can my vote cancel yours? If it cancels, that means that the former outcome is restored, the outcome that would have occurred had neither of us voted. Otherwise, what power relations have played themselves out? You were able to move the needle, and I was not able to move it back. That implies that you had more power than I had. Therefore, a voting system that does not meet the Frohnmayer balance constraint does not weigh the votes equally, does not accord equal political power to the voters, one voter to another [note 2].
The two writings I cited above in which Warren D. Smith is either the author or a co-author, they compare Score Voting to FPTP, and they argue that FPTP is very money-vulnerable and that Score is not. I note that FPTP does not give equal weight to the votes and Score does. I want to generalize and propose that this is the key difference between those systems that causes one to be money-vulnerable and the other to be money-resistant, that one does not accord equality and the other does. I want to suggest that other systems that, like FPTP, do not accord the votes equal weight, also are money-vulnerable, and I want to suggest that other systems, like Score, that do accord the votes equal weight, are money-resistant. I feel that when a system permits inequality, that inequality functions as a crack into which the oligarchy can insert its wedge in order to burst the system apart as a tool of democracy and create oligarchy instead of democracy. I don’t have a proof of this contention, but I’m suggesting to you that it is likely true based on the examples.
The simplest voting system that provides equality is Approval. Simplicity may be necessary to sell changes to Americans. However I think another system, a balanced multiround system, may provide slightly better power relations. Arguably with Approval or more generally Score, in order to give any of your support to a compromise candidate over the worst candidates, you have to dilute some of your support for your favorite candidates. I think a system I describe solves that.
By the way, I oppose the most popular alternative voting system, IRV, on the grounds that it does not provide equality; it does not meet even the formal balance constraint. IRV is a multiround system, which may be a win (see previous paragraph), but the problem with it is that each round is not fair, gives more power to those who support a candidate (and oppose the others) than to those who oppose a candidate (and support the others), as FPTP does.
[note 1] The election for President of the United States (POTUS) is said to require a majority. Indeed it requires a majority in the electoral college. However, the elections of the electors goes basically by FPTP in every State. So, the election of POTUS inherits all the problems that a choose-one system brings, including extreme vulnerability to money influence.
Even though I argue that for a voting system to provide equality, it must meet the balance constraint, even among systems that do meet that formal constraint, I contend that there can be a difference in degree to which they give everyone the same power. Systems meeting the balance constraint give two voters the same power if those voters are in full disagreement with one another. But it seems to me that such systems may fail to give two voters who are only in partial disagreement with one another equal power.
To explain this, I start by referring to a voting system that I think gives everyone equal power: Score Voting also known as Range Voting. Not only does Score give equal power, but it is also interesting to use as a prototype in which it is possible to cast descriptions of a number of other voting systems, as we can view them as restricted versions of Score. These systems have restrictions on the forms of their ballots, but the ballots that they do allow map to Score ballots in such a way that if the election is tallied as in Score, it produces the same outcome as the voting system being described would produce.
A vote in Score is a point in a vector space whose number of dimensions equals the number of candidates. If the middle of this space is the origin, and we picture a vector as an arrow, a rational voter will make as long an arrow as possible, and it will have some angle from the origin. Score allows all angles (to some granularity). All rational votes are simply angles.
Now consider the plurality/antiplurality voting system. The way this works is the ballots are restricted to either support one candidate and oppose all the others, or oppose one candidate and support all the others. This meets Frohnmayer balance. Manifestly, every vote in this system has its antivote. However, consider a four-candidate election in which I favor candidates A and B and I oppose candidates C and D. Score would let me vote that way, but plurality/antiplurality prevents me from doing so. On the other hand if you, say for example, like A, B, and C but oppose D, plurality/antiplurality accepts your exact expression of your sentiment. So, I conclude that even though plurality/antiplurality meets formal balance, it doesn’t gives our positions equal power, because it allows one position to be expressed on the ballot and the other not. Generalizing, we can see that some systems are better at providing equality than others even if both systems being compared meet the formal balance constraint.
The space of permitted vectors under Score is the whole surface of the hypercube (some people want it to be a sphere instead), but that of plurality/antiplurality is a spiky thing: you can only vote in the center of each face, for or against one candidate; you can’t go into the corners and support a different count of candidates than one or all-but-one.
[end of note 2]