Somin and Friedman on Voter Ignorance

Ilya Somin and Jeffrey Friedman have been having a dispute about what is the cause of the low level of information that voters possess. Is it rational ignorance, based on the fact that voters know that their votes are unlikely to decide an election? Or is it inadvertent ignorance in the sense that the voters believe they know enough to make a wise decision, even though they are mistaken?

Friedman appears to make a strong criticism of the rational ignorance theory:

Rational ignorance theory is falsified by the fact that 70 percent of voters say that they think their individual votes are “really important,” as I noted in my earlier post. Moreover, as I noted, 89 percent say that influencing government policy is an important reason for their vote. If these findings do not falsify rational ignorance theory, what would?

In other words, since people do not know that their votes are extremely unlikely to influence an election, rational ignorance theory appears to be mistaken.

Somin, however, also makes a strong point:

Friedman’s theory implies that the average voter would not bother to acquire significantly more information about politics if he suddenly learned that he would be part of a small committee tasked with picking the next president of the United States. I think the vast majority of people would take the decision a lot more seriously if that were the case, and would spend a lot more time learning and evaluating political information. Jurors who make decisions in small groups where each vote matters greatly perform better than voters in part for this very reason.

I have a possible resolution of this apparent inconsistency. It draws on the fact that people’s behavior – and their practices – do not always conform to their verbal accounts of what they are doing. So people believe in some significant sense that their votes influence elections and government behavior. But they do not really act as if that was the case. Sure they are willing to incur the minimal effort in order to vote, but they are not willing to incur greater effort to inform themselves in order make a knowledgable decision.

This occurs because people’s practices are informed by the fact that they do not really influence the outcome, even though they think they have such influence. For example, if one was buying a new car or deciding on whether to have an operation, one would make an effort to inform oneself. One would also not intentionally ignore contrary information. Similarly, if one were selecting the next President, one would attempt to become more informed than people do today when they vote.

The bottom line is that people behave differently than their statements indicate. Our behavior is based on habits, which are informed by the results of our actions. When we buy a lemon, we learn to research car purchases in the future. We decide to change our habits. When we vote based on poor knowledge, we have no cause to change our behavior, presumably because we implicitly recognize that our vote did not make a difference.

Mike Rappaport

Professor Rappaport is Darling Foundation Professor of Law at the University of San Diego, where he also serves as the Director of the Center for the Study of Constitutional Originalism. Professor Rappaport is the author of numerous law review articles in journals such as the Yale Law Journal, the Virginia Law Review, the Georgetown Law Review, and the University of Pennsylvania Law Review. His book, Originalism and the Good Constitution, which is co-authored with John McGinnis, was published by the Harvard University Press in 2013.  Professor Rappaport is a graduate of the Yale Law School, where he received a JD and a DCL (Law and Political Theory).

About the Author

Recent Popular Posts

Related Posts


  1. gabe says


    You hit it right on the head!
    What is more, most folks do not think that any policy change will significantly affect them and that they will muddle through. after all, we have always had a sufficiently vibrant economy to allow us to make adjustments. Will this continue, I don’t know.
    It does however, raise an important question – “What is the message that would be most effective in getting voters, low-info and otherwise, to appreciate the values of limited government.”
    I think for the most part, we all miss this. Talk of theories, impersonal statistics mean nothing to most folks. However, if you could show how the vast array of regulations IMPOSE a cost on every citizen and how it affects both their wallets and their choices, perhaps we might make some headway. The opportunities to do so are boundless from those god-awful CFL’s, to ethanol blended gasoline, to licensing fees on building a bloody deck in your backyard, etc etc etc.
    But no! we want to take the high road!!!! Ok, keep losing, conservatives, at least you get to sound intelligent on the talk shows which are ignored by the mass of voters anyway.
    Oh heck, I better get back on my anti-depressants before I really get worked up.
    take care

  2. R Richard Schweitzer says

    You say:

    “This occurs because people’s practices are informed by
    *** the fact that they do not really influence the outcome, ***
    even though they *think* they have such influence.”

    Pray tell what *does* **influence the outcome**?

    Is it your point that “rationality” would call for the seeking of more, or better, or more clearly understood *information* (to be interpreted into “knowledge”) in order to have more or better influence on outcomes?

    Aren’t you really suggesting that voters are only “expressing opinions,” which may or may not be well-grounded on “facts” (information), rather than seeking to exercise any form of “influence?”

    There are other aspects which are posted at Cato Unbound by “Counsellor” to all three points made there.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>