(Curiosity) The Rationality of Disenfranchised Non-Voting

In college, I took a poli-sci class on the American Presidency. One of the things we learned was that the president’s party tends to lose seats at midterm elections and that this effect is often more pronounced with a second-term president. (Note: this is a tendency rather than a rule.)

That was TWENTY YEARS AGO I learned this. I don’t understand why everyone is confused by the 2014 midterms. This is not a new thing.

You all can believe whatever you want, but let me give you something to think about.

In statistics, there is a measure called degrees of freedom. It is the number of ways some observation is free to vary. American voters have very few degrees of freedom. They can vote for the Democrat (which she will interpret as a validation of her policies), they can vote for the Republican (which he will interpret as a validation of his policies), or they can not vote.

Technically you could also write in a name, but in a strong two-party system, the effect on the composition of the next Congress is the same as not voting.

There are no other options.

Imagine going to the grocery store and finding only two choices. Of anything. It’s an effective way to insulate voters from altering or even influencing the function of the black box, which is the system to which your vote is an input.

In general dynamics (and operations research), you model a function as:

input(s) –> box –> output(s)


where each box could be “opened” to reveal another dynamical system of various interconnected inputs, outputs, and boxes.

Your vote is an input, but what happens to it is a result of the electoral system. That system has certain characteristics: low degrees of freedom, for example (only two candidates).

People think a vote is a vote is a vote, but it’s not necessarily true. In the first place, congressional districts are gerrymandered all to hell. A great many are carved such that your vote means little. Not nothing, but very little.


Those races that are actually contestable are won by money more often than not, and this effect has quickened since the 2010 Supreme Court decision in Citizens United v. Federal Election Commission. Whenever I make that claim, people will often bring to mind one or two cases where the candidate who won a particular office did not necessarily spend the most, but such examples are misleading.

If you regress total expenditure across all elections in a particular cycle, you see a strong positive correlation between total money spent and electoral success. Politicians and the professional marketers who get them elected know that money is not a guarantor of success, but it’s a HUGE boost, and certainly some minimum is necessary (where that minimum is still a boat load of cash).

So practically speaking, any candidate is dependent on money, which means any candidate is dependent on people who have money. This need not be rich people. If LOTS of middle-income folks give middle-income amounts, the effect is the same. But that requires a degree of widespread political motivation that, while certainly possible, is rare.

The best way to think of it is NOT whether your vote, or any vote, counts. The 2000 presidential election notwithstanding, most votes are counted correctly. That is, over the course of your life something like 95% of your votes will be tallied. (I forget the exact number.)

What disenfranchised voters understand, even if they can’t put it in these words, is that their tallied vote actually has very little to no impact on the composition of the Congress — again, because of how it is processed by the box, by the system: the choice of two candidates, both products of a political machine, both largely dependent on the same class of interested wealthy benefactors, chosen by an electoral system that doesn’t count votes directly but instead aggregates them into gerrymandered sums.

If that’s not bad enough, a recent Princeton study found that when the interests of the people differed from those of the ruling class, actual implemented policies overwhelmingly favored the elite, and that this was true regardless of which political party was in power. So not only is your vote unlikely to impact the output of the election — the composition of the Congress — it has near-zero impact on the output of the whole shebang: laws and policies and, ultimately, people’s lives.

In a world where not everyone cares to be politically active — which is the world we will always inhabit — such a system is stable because it gives the illusion of participation while concentrating power.

In psychology, there is a bias called the fundamental attribution error, which is one of my favorites. It’s a big topic, but basically it says that people (read: you and me) tend to attribute positive outcomes to our own worthiness and negative outcomes to the nastiness of the world.

An ideologue — that is, someone who is strongly partisan — interprets electoral success as a validation of their cause, as the Republicans are doing now and as the Democrats did in 2008. Whenever a party is ousted — as they surely must be at some point — the fundamental attribution error says it’s because the bad guys cheated, or people suddenly turned stupid (where they had been so brilliant before), etc.

But looking at the “meta level” — at the election process itself, at the low degrees of freedom, at gerrymandering, at the influence of money, at the composition and therefore the actions of Congress — you see that changes in party often have very little to do with changes in anything else.

I’m not saying you shouldn’t vote. It’s still a good thing to do, and the more people who do it, the better for all. But if voting is rational, that’s because it’s most akin to zero-cost betting, where a vote is like a free lottery ticket. Under the present oligarchic system, you almost certainly won’t win, but it can’t hurt.