The Myth of the Informed Voter Is Doing Real Work

Decades of survey research say the same thing. Most voters cannot name their representatives, do not know which party controls which chamber, and hold policy views that are unstable across the wording of a question.

The usual response to this finding is despair, followed by a proposal for civic education. Both responses assume that detailed knowledge is what the system requires from a voter, and it is worth asking whether that is true.

What voters reliably do have is an evaluation of whether things are going acceptably and a rough sense of who is responsible. That is a low resolution signal and it is not nothing. It is sufficient to remove a government that has failed visibly, which is the single function that distinguishes a democracy from the alternatives.

The danger is not ignorance, which is permanent and manageable. It is the corruption of the accountability signal, through arrangements that make responsibility impossible to assign. A voter who cannot tell who did it cannot punish anyone for it, and no amount of civic education fixes that.

The despair response assumes the system was designed for a voter who does not exist and has never existed, which makes it a strange thing to be disappointed by.

What a voter is actually asked to do is narrower than the civics account suggests. Not to evaluate a policy, but to decide whether the people currently responsible should continue, which is a judgement about outcomes rather than about mechanisms.

That signal is low resolution and it is not empty. It reliably distinguishes conditions getting worse from conditions getting better, which is most of what a system needs to correct itself.

The failure mode is therefore not ignorance. It is anything that breaks the link between an outcome and the person responsible for it, because the signal depends entirely on that link being traceable.

Federalism is the first complication and the largest. Who does what is genuinely difficult to establish, because a single visible service is often funded federally, administered by a state and delivered by a county.

A voter who is dissatisfied with that service has three levels to blame and no straightforward way to allocate. The blame lands somewhere, and where it lands is closer to random than anyone would like.

How elections are run adds a second layer, because the rules that determine which contests appear on a ballot, and when, are themselves set at the level being judged.

Timing does a great deal of work here. A local election held separately from a national one draws a different and much smaller electorate, and the choice of date is made by the officials it will judge.

The election code governs that choice, which is a good example of the general problem. The rule is public, it is consequential, and it is not the sort of thing a low resolution signal can ever reach.

The remedies that follow are unglamorous and mostly administrative. Consolidate election dates. Make responsibility legible on the service itself. Publish which body funds and which body delivers.

Election administration is where most of that would be done, and it is a field with almost no political constituency, which is why it is chronically underfunded.

None of this makes voters better informed and none of it needs to. It makes the thing they are already good at judging correspond more closely to the thing they are actually judging.

There is a version of this argument used to dismiss the electorate, and it is the opposite of what follows from it. A voter making a coarse judgement on outcomes is behaving rationally given the cost of doing anything else.

The failure belongs to the design rather than the voter. A system that scatters responsibility across three levels of government and then complains that nobody can allocate blame has asked for a signal it made impossible to send.