Skip to content

Surprise! Surveys Are “Unreliable Guides” to Actual Voting Patterns.

In 2016 and again in 2018 the state of Washington put carbon tax initiatives on its ballot. Both failed. A paper from 2019 that looks at the results by precinct concludes that opposition was almost entirely explained by ideology: conservatives didn't like a carbon tax while liberals did.

This is boring and unsurprising. But naturally I like it when rigorous research confirms my beliefs, so here's a section from the study on a whole different topic:

Fifth and finally: Why do surveys overstate actual votes for a carbon tax? Using our individual survey data, we estimate that support for I-1631 in Washington is 20 percentage points lower than in other states, controlling for ideology and demographics—but find no such gap for eleven other environmental policies (e.g., the Kyoto treaty).... We interpret the Washington penalty emerging from our difference-in-differences estimator as a campaign effect: respondents exposed to Washington’s two real-world campaigns and actual vote on a carbon tax exhibit lower support than respondents considering a purely hypothetical initiative.

....Thus, while the qualitative patterns in our survey data are consistent with the actual vote in Washington—namely, that a carbon tax is more popular among liberal voters—surveys may be unreliable guides to the absolute number of votes in a referendum.

No shit. This is something that I wish I could pound into the heads of all the liberals who constantly claim that "liberal policies are popular." It's true that they are, but only in polls where people don't really give it much thought.

But the opposition gets a say too, and once something ends up on the ballot (or in Congress) and people start hearing the details of what something will cost and who it will affect and so forth, you're going to lose 10-20 points of support almost instantly. That policy that polls at 70% approval? It might squeak through. But the one that polls at 60% is a deader.

In other words, liberal policies simply aren't as popular as most liberals think. I posted this list of liberal policies a couple of years ago, and I think it deserves reposting. Read it and weep.

37 thoughts on “Surprise! Surveys Are “Unreliable Guides” to Actual Voting Patterns.

  1. Midgard

    The social liberals especially don't get it. Yet, the dccc forced those kind of candidates in during the 2014-20 cycle in many instances. Social Nationalism now.

    1. Crissa

      No, it's not.

      Basically, people support good thing in the abstract, but when doing something might. Have real effects, the falter.

      It's no different than getting everyone to agree upon pizza. In general, pizza will win, but once you start arguing toppings, some will flip against.

  2. jobywalker

    While this may certainly be true, the two Carbon Tax votes in Washington are a very poor example due to the politics around the two different initiatives. The first proposal was a moderate revenue-neutral proposal that was opposed by the left & environmental groups because it did not bring in additional revenue. If the moderate measure had been supported by the environmental groups it would have passed easily.

    The second proposal was revenue positive and the revenue would not flow to the general fund or an independent fund controlled by the Legislature but to an "independent" organization that had little accountability to the voters. This was opposed not just by the right but also by many on the left because of the very obvious governance and accountability deficiencies.

  3. skeptonomist

    Kevin is certainly right that polls are not always reliable, but in this post he doesn't seem to get the dominant principle of American politics over the last 55 years or so, which is that many white people do favor progressive economic politics, but they are also conservative "socially", meaning especially they are racist but also highly partisan on religion. This is especially white wage-earners in the South but also in other areas. These people voted Democratic when there was no question of sharing the goodies with non-whites, but the parties switched on equal rights for non-whites. So of course their votes do not necessarily align with their economic preferences.

    Then there are special interests such as the gun industry which will campaign against candidates with all kinds of negative methods such as character assassination if they are opposed by the candidates.

    There is the question of how a proposition looks in the abstract, such as carbon taxes, and how it will actually affect them personally, which will be in the form of higher gasoline and heating prices (and electricity). Pollsters do not point out the drawbacks of such propositions and then as Kevin says when they actually get on the referendum the negative aspects are fully exposed.

    Aside from the last category, progressive economic policies really do tend to be popular. Democrats are faced with the problem of separating these things from the "social" matters with dominate among many whites. To succeed in the long term they must replace the race war with a class war.

  4. royko

    Your tiered list is actually about what I would expect. I'm disappointed that $15/hr didn't make it to "maybe" (and I doubt the voters would get too upset about it) but otherwise I wasn't surprised and didn't really disagree.

    1. veerkg_23

      The list is from 3 years ago. $15 wage is more popular now, as can be seen from the referendum results, even in places like FL.

  5. bebopman

    And I think it’s not as bad as you or the 2019 survey portrays it. The example I look to, not perfect but as valid as anything, is Florida (yes *that* Florida), where voters approve standalone liberal(ish) proposals and then elect people who do all they can to stop the liberal(ish) proposal the same voters approved. God knows why they vote that way. You would think they would noticed by now that the people they elect oppose the ideas they support.

      1. mudwall jackson

        the dccc has nothing, absolutely nothing, to do with most elections in florida (or any other state for that matter).

  6. jte21

    This is right. I've always said, if we were ruled by polls, we'd be a socialist paradise. Sweden, Canada, and New Zealand would gnash their teeth in envy at us! Unfortunately, we're currently only allowed to have what billionaires, Fortune 500 companies, and whatever embarrassing fraction of the electorate bothers to show up and vote in any given election will let us have, so that's a considerably smaller menu.

    1. Loxley

      'whatever embarrassing fraction of the electorate bothers to show up and vote in any given election'

      Are we forgetting the impact of voter disenfranchisement?

      1. jte21

        Some people certainly are thwarted by some of the roadblocks thrown up to registering and getting to the polls on election day. In tight races, like we saw in GA and AZ last year, it can sway an election, which is why Republicans are scrambling like crazy to put up more. That's dwarfed, however, by the larger issue of voter apathy, particularly in state, local, and off-year elections. Politicians can safely disregard polls and do the bidding of their corporate and PAC paymasters when they know barely half of the electorate is going to be assed to respond either way.

  7. Midgard

    To my point above, a large reason why Democrats don't run "FDR" New Dealers, whether more moderate or populist, is because of northeastern and West Coast elites power in the party. So running a anti-ltbg pro 15 dollar min Democrat in Tennessee doesn't happen. The party is shut out. Republicans had the same problem with neoconservatives. It wasn't building up a large coalition so Putin/Trump came up with Kosher Nationalism to try and racially motivate voters.......but it only works so far, especially considering Trumpism is basically a reworked neoconservatives. Once white workers struggle, it would struggle as debt markets collapse. A temporary solution in maintaining some power aka you get Greeneberg, Gosar and Gaetz among others, while the national party makes gains/pushes for colored voters..

    1. Crissa

      Citation never included, of course. In one post you're calling Democrats nazis and in another you're saying we refuse to run them.

      Pick a lane, but why do we have your racist tripe?

  8. D_Ohrk_E1

    I think you missed the message. It's not about the popularity of liberal policies; it's about risk-aversion triggering a default to existing policy, which by extension, is about the need to shift the Overton Window to drive policy changes.

    1. golack

      Excellent point.
      There's a lot of stuff crammed into the "only in polls where people don't give it much thought". I think the problem is more due to "But the opposition gets a say too..." If good faith arguments are being made--great. If the opposition is mainly fear-mongering, then that's not good.

      1. Atticus

        Correct. Fear mongering is bad. It's also bad when one side accuses the other of being racist, a xenophobe, a misogynist, etc. Liberals are often too quick to apply these labels when they are often not warranted. One hypothetical case in point...would it be possible for a republican to want to enforce our immigration laws and beef up security at the border without being labeled a racist or xenophobe by some liberals?

        1. azumbrunn

          The problem with this example is that "beefing up border security" is popular with the GOP precisely BECAUSE it is ineffective at enforcing immigration laws. The key to enforcement is going after employers; everybody knows that. And the GOP never even considers doing that: It would hit their donor base plus it would solve the immigration issue thereby taking a powerful campaign issue away from them.

          1. Atticus

            I'm all for going after employers. I agree that would be effective. But till think we need to do a better job securing the borders. We can have reasonable discussions as the best methods of enforcing immigration laws. It's just that some on the left seem to instantly call anyone racist who want to do anything.

  9. Loxley

    'In other words, liberal policies simply aren't as popular as most liberals think.'

    It's disappointing, Kevin, that you leave out the obvious and reliable mass-propaganda campaign that occurs once something actually gets on a ballot. MILLIONS were spent to kill the carbon tax initiative in washington state. And, that you imply that the reality (facts) about an initiative in an of themselves make it less popular. Also not true.

    Therefore, it's more accurate to say, that liberal policies are popular and good policy in the real world most of the time- but conservatives are gullible.

    1. azumbrunn

      It is possible to win legislative fights for the left if two conditions are met:
      1. The proposals need to be carefully designed and must avoid any obvious flaws that can be exploited by opponents (this is necessary anyway if the legislation is to actually work properly).
      2. The doctrinaire left needs to be on board with an incremental proposal. As long as we have people who vote against a 12 dollar minimum wage because they prefer 15 we will never even get to 12.
      Even under these conditions it is hard, but maybe not quite as hard as Kevin suggests. This topic is of course one of his hobby horses that he takes out for a ride once in a while.

      1. azumbrunn

        To add to this: One of the Washington carbon tax proposals violates condition number 1 in the above reply, the other violates condition number 2.

  10. Toby Joyce

    Kevin needs to factor in that policies that are over 50% are popular until Fox and right wing media get loose on them.

  11. JonF311

    "Repealing Obamacare" is a conservative initiative, not a liberal one. It's the odd one out on the list (and it's in the No section)

  12. JonF311

    "Repealing Obamacare" is the odd one out on the list-- as phrased, it's a conservative policy initiative, not a liberal one.

  13. James Wimberley

    If rejoining the Paris Agreement were a non-starter at 53% support, it's a puzzle why the policy did not hurt Biden in the election, and why pushback to his carrying it out on Day 1 has been negligible. Is there even snark from Trump?

  14. Vog46

    Funny thing is is that people, for the most part are friends with people who may have differing views on things. Politicians on the other hand are fueling a "how can you be friends with them" attitude.
    I am not embarrassed to say I have both republican and democrat friends. They tend to leave me (The independent voter) alone. I have told people from both parties no signs on my lawn - no stickers on my car etc.
    I have voted split ticket for decades.
    And let's be honest here. I have over my long life voted for people who turned out to be idiots. I have told the Mrs I have to learn from those mistakes. Which she just laughs at. She says "Hon, if you had learned from every mistake you've made in your life so far you'd be considered a genius by now - and you're not even close......" She then giggles and walks off.........

    But I have heard of people from BOTH parties lying to phone pollsters. They admit they do it and for some reason enjoy doing it.
    Pollsters are viewed just a half rung above robo-callers.
    Nobody has any use for them anymore

Comments are closed.