Skip to content

30/30/40 Nation

Barack Obama’s win has a potentially huge effect on policy. The current budget negotiations will affect the level and direction of government spending and on the mix of taxes paid by different groups of Americans. We can guess that a President Romney would have fought hard against upper-income tax increases. Other areas of long-term impact include the government’s stance on global warming, foreign policy, and the likelihood that Obama will nominate new Supreme Court justices who will uphold the right to abortion announced in Roe v. Wade.

When it comes to public opinion, the story is different. The Democrats may well benefit in 2014 and 2016 from the anticipated slow but steady recovery of the economy over the next few years—but, as of November 6, 2012, the parties are essentially tied, with Barack Obama receiving 51% of the two-party vote, compared to Mitt Romney’s 49%, a split comparable to Al Gore’s narrow victory in 2000, Richard Nixon’s in 1968, and John Kennedy’s in 1960. Over the next few months, you will be hearing a lot about Obama’s non-mandate, and rightly so.

But here I want to talk about a slightly different split of the voting-eligible population: the approximately 30% who voted for Obama, the nearly identical number who chose Romney, and the 40% who did not vote at all or who voted for minor-party candidates. . . .

See my Daily News article for the full story.

For more background, see this post from a few years ago on the differences between voters and nonvoters. I really do think it would be a good idea for us to talk less about 50/50 America and more about the 30/30/40 split.


  1. How do you feel about making voting *mandatory*? That’s the state of affairs in some countries. Not really an option here in the US, but interesting. On the substantive point: I suspect that the Leavitt-style economic argument weighs heavily, in the sense that many voters just don’t think it could possibly be worth TWO HOURS of their time. In many cases they are right. I really don’t know how people with real jobs (ie, not me) vote.

  2. David says:

    I would like to see the US implement:
    (a) mandatory voting + a voting holiday
    (b) a “None of the Above” option on every ballot question

    None of the Above would not mean anything… it could be the exact same as not-voting but it would be tallied. So the final count might be 40% None of the Above 30% Obama, 29% Romney, 1% 3rd Party, in which case Obama wins, but the public has overwhelmingly expressed displeasure with the two parties, and that might enable third parties to get more donations/support as well as whip the major parties into shape.

    On the other hand if it is 50% Obama, 49% Romney, 1% 3rd Party… then we at least know that the real political issues in this country is between Republicans and Democrats and that the parties are doing what the public wants.

    Its also infuriating to wait in line for an hour and see a single name for half the races because the local races are uncontested. None of the Above would be a good way for people to exprss their displeasure with not having any choice, without negatively affecting the process in those cases.

    • Anonymous says:

      Denmark has a “blank” option. Basically what you are proposing. Sure other countries have it as well.

  3. Fernando says:

    Why not take a random sample of 1000 Americans and let them decide the election. At least the vote would be representative. It would also be a whole lot cheaper.

    Getting everyone to vote is grossly inefficient.

  4. Lex says:

    George Friedman made a good point on this very topic a few weeks ago:

    From the end of the article:
    The United States has elected presidents with the narrowest of margins and presidents who had far less than a majority. In many countries, this might reveal deep divisions leading to social unrest. It doesn’t mean this in the United States because while the division can be measured, it isn’t very deep and by most, it will hardly be remembered.

    The polls say the election will be very close. If that is true, someone will be selected late at night after Ohio makes up its mind. The passionate on the losing side will charge fraud and election stealing. The rest of the country will get up the next day and go back to work just as they did four years ago, and the republic will go on.

  5. And there is a time when a pie chart makes good sense!

  6. There is a downward trend in participation: “As of today, Mitt Romney won 2.1 million fewer votes than McCain in 2008, 1.2 million less than Kerry in 2004” via

  7. Chris says:

    I found the article confusing.

    If it’s reasonable to guess non-voters reject the red-blue political system (paragraph 6), why would you then assign them to the red-blue political system (paragraph 8)?

    • Andrew says:


      The bit about how these people would vote, if they were to vote, is based on survey response. But I think that if they did all vote, the parties themselves would change (on economic issues, both parties would move to the left), so in that sense perhaps my counterfactual is misleading.

      • Christian Hennig says:

        I’d guess that people who don’t vote are also less likely to respond to surveys, and those who in fact do respond to surveys are a rather biased selection of those who don’t vote. Of course I haven’t surveyed the others so I don’t know exactly but intuitively “not voting” and “not responding” seem related enough to suspect such a thing, or not?

        • Andrew says:


          Yes, I am pretty sure that people who are less interested in politics are less likely to respond to opinion polls. But this doesn’t make these polls useless; survey organizations try their best to correct for nonresponse bias.

          • Christian Hennig says:

            I’m not saying they are useless but I have my doubts about how reliable this is, particularly in this special case. Because I suspect that for a considerable portion of people belonging to the non-voting group their reason to “not vote and not respond to anything else either” dominates everything that can be observed and used for bias correction.

          • Andrew says:


            Maybe you’re right, but there is a literature on this (follow the links above). The political scientists who study nonvoters might be completely full of crap, but I’d be inclined to trust their consensus until convinced otherwise.

      • Chris says:

        I guess I still haven’t quite grown out of my idealistic youth. When I read the part about rejecting the red-blue system, my mind raced ahead to a story about disrupting the 2 party system & bringing other parties into play. When you didn’t go there, it threw me for a loop.

        The more I think about this, it seems you are operating under the assumption that the US system is unstable with more than 2 major parties. If this is true, then it does make sense to think about what the non-voters would choose if forced to. However, if the US system is only stable with 2 options, and the two parties move to attract non-voters, then wouldn’t these platform changes just create a new group of non-voters? In your example, if both parties moved further left on economic issues to attract non-voters, do they create a new group of non-voters who are way right on economic issues?

        Wow, I’m really in over my head now…

        • Christian Hennig says:

          Andrew: I apparently can’t respond directly under your last one, so here:
          1) Thanks, I actually wanted to express (and somehow forgot it when I typed my last response) that I’m really interested in how they do this.
          2) It wasn’t really necessary to make “completely full of crap” out of “I have my doubts”, or was it?

          • Andrew says:


            I don’t think these researchers are using any special techniques; they are just assuming that the standard survey adjustment techniques will give reasonable answers. I don’t think the adjustments will be perfect but I have no reason to believe they will be far off.

  8. Toni Verbeiren says:

    Just a thought experiment:

    Perhaps we are looking at a 2-‘particle’ system in thermal equilibrium. Or expressed in terms of game theory, both parties have optimal information about each other and have thus reached equilibrium. Winning the elections, then, is a matter of fluctuations.

    From this perspective, the non-voters do not matter very much…

    • Andrew says:


      1. It can take awhile to reach equilibrium.

      2. If everyone voted, the equilibrium would be in a different place, hence different policies.

  9. Wayne says:

    Something in your Daily News article reminds me of the Ben Franklin quote: “When the people find that they can vote themselves money, that will herald the end of the republic.”

    If poll wait times could be minimized, and the day declared a mandatory holiday, I agree that a “None of the Above” would be a great idea. Though mostly something for pundits or perhaps third parties to think about. It would be amazing if teeth could be added to the election such that if the winning party’s votes don’t exceed “None of the Above” votes, there would be some kind of restriction applied to their governance, so they’d actually care about dissatisfied voters.

  10. Jeff says:

    Maybe instituting mandatory voting and providing a “none-of-the-above” option would actually drive a lot of people into choosing red of blue. Who wants to be known as a “none-of-the-above” person? You would have the same effect you have from people who answer poll questions: they might not have thought about what is being asked, but they feel compelled to provide an answer. After all, an election ballot is nothing more than a multiple choice question.

  11. jrkrideau says:

    A rather off-topic question though perhaps wait times are relevant: Why are there such long line-ups for voting? It seems to happen every election in the USA.

    I vote in Canada and while we seem to have a very simple ballet compared to the US ones, See the sample ballot here
    I have never had to spend more than 3-5 minutes in line to vote at a regular poll. I think it took almost 10 minutes, once, at an advanced poll

    Given past experience, why would not the election officials simply add more resources to each polling station to reduce wait times?

    • Andrew says:

      You ask, “why would not the election officials simply add more resources to each polling station to reduce wait times?” I suspect that politicians have no great desire to change the system under which they were elected.

  12. jrkrideau says:

    Oh right. Again, a difference in our systems. At the Federal level Elections Canada is an independent body and tends to pride itself on being totally neutral though we always hear grumbles. I have never checked but assume it is the same for the provinces.

    Generally the idea seems to be to get the maximum number of voters out. The previous Chief Electoral Officer was big on getting the homeless and prison inmates to vote IIRC.

    Therefore I suspect the tendency here would be to expand facilities as much as possible to reduce wait and travel times.


    The position of Chief Electoral Officer (CEO) was created in 1920 by the Dominion Elections Act. The Chief Electoral Officer is appointed by a resolution of the House of Commons. He or she reports directly to Parliament and is thus completely independent of the government of the day and all political parties. The CEO serves until retirement at age 65 or resignation. He or she can be removed from office only for cause, by the Governor General after a joint request following a majority vote by the House of Commons and Senate.

  13. Dr. Boli covered this with the headline: “UNDECIDED WINS IN LANDSLIDE.”

  14. […] out this graph from Andrew Gelman¬†(whose book Red State Blue State is another must-read on the statistics of […]

  15. […] 5. ¬†40% of Americans did not vote. […]

Where can you find the best CBD products? CBD gummies made with vegan ingredients and CBD oils that are lab tested and 100% organic? Click here.