Biases in Turnouts – philosophical discussion by Greg Wolfe

This essay was inspired by the following conversation:

A: It looks like the fact the USA does not have compulsory voting, universal suffrage, and automatic universal registration, causes large biasing of the vote.
B: But that bias could be beneficial, from a utilitarian standpoint.

I agree with B's statement, but I think that it is important to note that we are at the edge of a slippery slope. Going down that slope for a moment...:

  1. There is an argument that choosing to participate indicates that a voter cares and is knowledgeable (⇒ disfavor mandatory voting)
  2. There is an argument that being able to get over minor voting barriers indicates that a voter cares and is knowledgeable (⇒ favor keeping election day as a single inconvenient day [Tuesdays] as opposed to having a longer period or weekend voting, favor voter ID requirements)
  3. There is an argument that having access to excess resources indicates that a voter is more vested in the system and has more socially useful views (⇒ favor poll tax, voter tests, property requirements)
(I) is a fine step for me. (II) is distasteful but I'm willing to listen and might be convinced. I find (III) completely unacceptable.

In my view the utilitarian worldview of what is best for society is not enough. There also needs to be a principle-based "fairness doctrine" that encourages participation and doesn't raise extraneous barriers to voting. Some reforms might be worthwhile even if they have little or negative impact from a utilitarian viewpoint. For example, I would support women's suffrage even if it turned out to be harmful from a utilitarian viewpoint.

To attend more to our current political situation: I support measures to reduce election fraud even when the fraud levels are low enough or unbiased enough not to have a major clear impact on actual results.

So what is the "take-away lesson" here?

When comparing the impact of competing factors [like women's suffrage vs election fraud] there are at least three measures:

  1. How influential (on altering election winners) is the factor?
  2. How beneficially influential is an factor? (utilitarian view)
  3. How strong is the principle of the factor?

All are interesting. For example, it is an interesting type-1 statement that black votes in the USA currently have more winner-altering effect than women's votes, even though from a pure type-3 point of view women's votes are more important since there are more women.

(1) is the easiest to measure. (2) can be hard to measure (or even to define) in the real world, but in some cases can be easy to measure in computer simulations. (3) is subject to the criticism that your moral principles may differ from mine... the basic principle I have in mind is that it is immoral to deprive people, against their will, of equal political rights.

Return to main page