Skip to content

A fill-in-the-blanks contest: Attributing the persistence of the $7.25 minimum wage to “the median voter theorem” is as silly as _______________________

My best shots are “attributing Napoleon’s loss at Waterloo to the second law of thermodynamics” or “attributing Michael Jordan’s 6 rings to the infield fly rule.” But these aren’t right at all. I know youall can do better.

Background here.

For some relevant data, see here, here, here, and here.

P.S. I get it that Cowen doesn’t like the minimum wage. I have no informed opinion on the topic myself. But public opinion on the topic is clear enough. Also, I understand that I might be falling for a parody. Poe’s law and all that.

P.P.S. Yes, this is all silly stuff. The serious part is that Cowen and his correspondent are basically saying (or joking) that what happens, should happen. I agree that there’s a lot of rationality in politics, but you have to watch out for circular reasoning.


  1. Tiago says:

    I don’t believe in the general explanatory power of the median voter theorem for policy choice, but am surprised that you don’t, given your paper on the rationality of voting. Do you think MVT fails in this specific instance or in general? And if so, how does this square with the rationality of voting?

    • Andrew says:


      I like the median voter theorem. It gives us a lot of intuition. And when the median voter theorem fails, that’s interesting too. The median voter theorem serves as a baseline, so that deviations from the theoretical result need explaining. See this article, for example.

      The persistence of the $7.25 wage cannot be attributed to the median voter theorem. The median voter would want a minimum wage of something like $12 or $15.

      Regarding the relevance of rationality theories to the understanding of voting, see section 5.2 of this article.

    • M says:

      Cowen has chosen this hill to die on. His blatant disregard for the MASSIVE body of research on the subject is becoming increasingly pathetic. He’s way, way out of line with the mainstream on this issue. His position as the Internet’s Loudest Economist is becoming increasing damaging for the field.

      Most of the remaining debate around minimum wage in the literature revolves around arcane methodological issues — the kind of issues that you (Andrew) are probably interested in! I’d recommend you look into this a bit more, I bet you’d enjoy it.

  2. Dave says:

    It’s only been three months. Would we expect all of those things listed to have changed by now? Maybe the administration is trying to be deliberate and take its time. It has two years to get its preferred policies passed.

  3. Joshua says:

    The funny part is that this announcement was botched.

    They may well walk it back, look like they’re already starting to do so, and they’ll likely increase the limit to 125k in October.

    This relates to your earlier post about how people update their theories when subsequent evidence makes their mistaken notion of a casual mechanism (or direction of cauality) evident.

    I think there should be a law where people have to specify where their theories fit with a model of cauality.

    For example, would reverse cauality apply, and if so why not?

    IOW, if the Biden administration increases the refugee limit to 125k, will Cowen reverse the cauality to say that the Biden administration is much more committed to ideals than he previously thought? And if so, why not? Why wouldn’t the causal mechanism work in reverse?

    Just funny that Cowen’a theory could just be blown up by a statement being poorly worded. Now he can just say “Nevermind.” if his theory is incorrect.

    Freedom from the constraints of a full model of cauality is a big problem, imo. People can just assert it exists merely by an observation of correlation. Often they don’t even feel constrained by an explanation of a plausible mechanism of cauality.

    In this case, what is the causal mechanism by which the Biden administration wouldn’t live up to their pledge of increasing the limit to 60k? Cowen asserted one (cowardice in the face of political pressure) w/o any explanation for what might back up that assertion of cauality. It’s basically unfalsifiable.

    As soon as I heard of the announcemet, I said that makes no sense. There’s no political advantage, and what would the logic possibly be along with a pledge to increase the limit to 125k in October (unless that’s just a lie). But obviously this would be a bad political move – the exact opposite of Cowen’s conjecture of cauality.

    But he will see this as proof he’s right if it goes on one direction and just irrelevant if it goes in the other direction. He’s got no skin in the game, plus it’s just weak logic.

  4. John Hall says:

    The way Cowen uses median voter theory is a short hand for a more complicated argument. The simplest median voter case is a direct democracy. Then, you can consider a unicameral Congress with representatives elected by population. Then, you get to the more complicated bicameral system we have. So what is the equivalent of a median voter in our actual system? The limiting factor is what gets passed through the Senate, so really a raw measure of public opinion does not measure the median voter. Rather, you need to aggregate by state. In how many states do they favor raising the minimum wage to 15.

    • Andrew says:


      I’d guess that raising the minimum wage has majority support in at least 45 states, probably 50. If the minimum wage were raised to $12 or $15, I could accept the argument that the median voter is being served. But not at $7.25.

      Also, regarding your general argument, you have to watch out that it’s not a tautology: the government is democratically elected so whatever the government does reflects the will of the majority. But, in that case, a statement that “X demonstrates the median voter theorem” is empty, as it could be applied to any X.

      • Dzhaughn says:

        What, in your view, is stopping those 45 states from adopting a $15 minimum wage?

        Even in solid blue Washington state the minimum wage is $13.50. We have direct referenda all the time, so it’s not the legislators being bought. And why isn’t it $20? I’m sure you’d get 55% or better majority polling for $20 in this state.

        I would turn to the MVT to explaint that. Apparently your answer is different. What is it roughly?

        • Andrew says:


          There are a zillion reasons why a policy supported by a majority of the voters doesn’t get enacted. Here are a few:
          – legal restrictions (not an issue with the minimum wage, but this can arise in other policy settings)
          – stickiness (passing a law takes effort, and there can be many veto points)
          – expert opinion (legislators can be convinced based on policy arguments that an idea that’s popular now could have negative consequences)
          – political calculation (maybe it’s more popular to raise the wage a dollar at a time rather than all at once)
          – asymmetrical support (a policy can have broad lukewarm support and narrow intense opposition)
          – party politics (a policy can have majority support but not within the party in power)
          – influence by campaign contributors
          – political or economic ideology
          – bundling with other issues.
          I’m sure there are some other items I’ve forgotten. The point is, there are lots and lots of reasons why policies don’t get enacted. The median voter is just one part of the picture.

      • John Hall says:

        The tautology point is a good one, as are the ones you made below.

        I had a hard time finding state-by-state polling on the minimum wage. However, I think a useful proxy [1] is how many states have enacted minimum wages higher than the current minimum. I count twenty-nine, based on wikipedia [2]. This would suggest that the median state supports a higher minimum wage than the Federal level. However, that does not imply support for a $12 one or a $15 one. In fact, only ten have higher minimum wages than $12 and none currently have a higher minimum wage than $15. Granted, there are counties and municipalities within some states that have set higher minimum wages.

        [1] I think it’s a useful proxy because it accounts for the complexity of state-wide politics better than a poll does. Polls can be unreliable. If you poll people to ask them how much they want a $10,000 check from the government, you might get more than 50% support from the population, but that doesn’t mean they will be happy with their representatives when the bill comes due.

    • Fred says:

      The median states according to 2020 election would be Florida and North Carolina.
      Conveniently for us, Florida had Amendment 2, $15 Minimum Wage Initiative in the same ballot box.
      Floridan voters chose Trump over Biden by 51%-48% split while passing the $15 minimum wage initiative by 61%-39%.

      The funny thing is that you could make an argument about how the problem of minimum wage supports the median voter theorem,
      not because it didn’t increase in the first three months of Biden presidency,
      but rather because it is extremely likely to increase before 2022.

  5. Jonathan (another one) says:

    Surely you don’t th0ink that the position of the NATIONAL Median Voter affects anything. There are 50 median voters, right? And it’s median VOTERS, not Median People Being Polled. It’s also complicated by the fact that voters don’t vote item-by-item, but for parties. So it’s complicated. So I don’t think what Tyler and Angus is refuted by “popular wisdom.” It remains to be seen, of course, but Biden’s actual policies are going to have to change a lot more before I’ll believe that that it has to be something other than the will of the Median Voter. (Note that is not support for the Median Voter Theorem as anything other than a null hypothesis.)

    That said, I’ll give it a try:
    How about: “…is as silly as believing that the guy falling off a twenty story building is still in pretty good shape as he passes the fifth floor.”

    • Andrew says:


      The median voter theorem can be valuable as a null hypothesis or, as I’d prefer to call it, a reference model. Then again, there are some cases where policy is far from the median voter, or far from the median voter in most states, etc., and this is one such case.

  6. John says:

    It seems to me that Cowen’s post was directly regarding immigration policy, not the minimum wage. He posted a tweet that included a connection between MVT and immigration policy. That tweet also included the minimum wage (which Cowen did not directly comment on).

    It seems disingenuous / misleading for Andrew’s post to read as if Cowen directly connected MVT and minimum wage (even if that is his position, it’s not clear from the MR post).

    • Andrew says:


      I have to admit I never know what to make of it when someone describes me as “disingenuous,” which according to the dictionary, means “not candid or sincere, typically by pretending that one knows less about something than one really does.” I feel like responding, truly, that I was being candid and sincere! But of course once someone accuses you of being insincere, it won’t work to respond in that way. So I can’t really do anything with that one.

      I agree that Cowen’s post focused on immigration policy. Public opinion on immigration policy is much more fluid than on the minimum wage, to the extent that it’s hard for me to place immigration policy in a median-voter framework. The minimum wage example is much cleaner which is why I used it in my post.

      • John says:

        Fair enough! Thanks for the reply, I would be interested in hearing more about public opinion re: immigration policy in the future here. And yes it does make sense why you used minimum wage in the post, and I in fact agree with your assessment.

        And to be clear, I did not intend to describe you personally as disingenuous. Just the construction / tone of the post (specifically in the P.S. and P.P.S) and their relation to the original MR post. I believe that you are candid and sincere! But you do seem to enjoy taking shots at Cowen now and again ;)

    • Joshua says:

      John –

      > It seems to me that Cowen’s post was directly regarding immigration policy, not the minimum wage. He posted a tweet that included a connection between MVT and immigration policy.

      I agree. I was confused by that aspect of Andrew’s post also.

      That said, Cowen was wrong (or at least likely wrong) on multiple levels in his post even when you focus on the immigration policy aspect.

  7. zbicyclist says:

    Shouldn’t we be weighting these voters by their expected campaign contributions? That would move the median quite a bit.

    • jim says:

      Up here in the great Rainy Woods, there are a lot of wealthy people thumping for higher min wage, and in fact that’s one reason the WA state has among the highest min wages in the country (CA is apparently a few dimes ahead of us now). So I think there’s plenty of money to be had for raising the min wage.

      I don’t think donations are the problem. The problem is that in the face of a near economic collapse, when many small businesses have already gone under and many many more are in deep shit – small businesses being the biggest employers of people on minimum wage – and the government is already hot on new taxes for business, doubling the minimum wage is a risky bet, so risky even Democrats can smell trouble.

    • David Chorlian says:

      Would this be the median donor theorem?

  8. Peter Dorman says:

    In terms of the current battle lines, I see two broad camps. One says we’re in a democracy more or less, and public opinion is well reflected in public policy. The MVT is a particular way to pin this down, but not the only one. The other camp is that we are in an oligarchy in which the opinions of anyone not in the upper bracket (however delineated) are largely immaterial, at least in “political economy” issues. The poster boy for view #2 is Gilens and Page, which has been the subject of a lot of research back and forth. I rather like the recent work of Jarron Bowman, “Do the Affluent Override Average Americans? Measuring Policy Disagreement and Unequal Influence”, which appeared last year in Social Science Quarterly. This would be a great topic for this blog to dig into; I’d love to know if my take is shared by others.

    Anyway, wouldn’t the minimum wage, both nationally and in most states, be evidence for the oligarchy view?

  9. jim says:

    I guess this puts us square back in the sites of the big question:

    Do opinion surveys mean anything?

    Even though 2/3 of Americans purportedly support raising the min wage to $15, only very slightly more than half of them voted for Biden. So at the least 15% of Americans don’t support the concept enough to suffer the consequences.

    With regard to the median voter theorem, who is actually in that “median voter” group with respect to the political decision making? Many states already have a min wage over $12, so how much does their opinion really count for a national wage increase? According to my quick assessment, 38% of voters already have min wage above $11.75, and CA, WA and a few other states are over $13.

    Arguably, states with low min wages have low min wages for a reason: they need a low min wage to attract employers. Is it possible that the states with the lowest wages are the states where people least want an increase? Sure, it would be great if you were a federal min wage worker to get an increase, but if you’re garment maker in a state with a $7.25 min wage, you can kiss your job goodbye if wages double.

    There is no such thing as a “median voter”, unless the issues are specifically defined for each voter.

    • Andrew says:


      I agree with your last paragraph. Issues are bundled. This also answers the implicit questions in your earlier paragraphs. Voters don’t get to construct their ideal candidates; instead, they have to choose between two options. Biden offered some very popular policies such as the minimum wage and also some less popular policies. Also, Biden wants to be reelected. If he thinks a high minimum wage will be bad for the economy, or if if he doesn’t want to burn all his political capital on this particular issue, that’s his call. There are reasons why, even though a higher minimum wage is popular, it hasn’t happened yet. My point is that these reasons have little if anything to do with the median voter and a lot to do with all sorts of other concerns, as I noted in one of my comments above. That’s what I thought was silly about Cowen’s post: the assumption that just because a certain policy was not enacted, that this could be attributed to the median voter, of all things.

  10. Dzhaughn says:

    I would just add your PPS is way off, in my opinion.

    Cowen would not believe half those things “should happen” in the sense of being good ideas. Whether they reflect “the will of the demos” is not a question that interests him. He would say not enough people think about public choice theory, however.

    True, he would point out that $15 is more than the median wage in Mississippi, and would not believe that setting the mimimum wage to that level will mainly result in a raise for half the work force. That’s about the only one; he finds both stimuli wasteful, is very immigration positive, anti gitmo, pro trade.

  11. Russell says:

    For better or worse, policy rarely tracks public opinion. In situations like minimum wage, it is probably best it does not. I wouldn’t want to enact a policy based on public opinion of the downsides are not evaluated. In my experience, most voters do not have a good understanding of the potential costs.

    I’m agnostic on the minimum wage (which I believe is a better description of Tyler Cowen’s position rather than he not liking it) because the empirical evidence is mixed, our prior should be it will have at least some negative impacts if its sufficiently meaningful (demand curves slope downward), and in general even the researchers of the issue are all highly partisan. So we have a pretty good prior in terms of prices and then all the empirical evidence seems like it is gathered with people proving their priors, as opposed to understanding the true mechanics.

    The debate on minimum wage often seems more like debates on theology combined with some fancy Greek symbols.

  12. Anonymous says:

    Minimum wage is a restriction on those with few skills or little reputation, making it harder for them to improve their social status.

  13. Andrew says:

    Russell, Anon:

    I think we’re all in agreement here that the persistence of the $7.25 minimum wage is not an instance of the median voter theorem. It’s the opposite: a policy that persists despite there being a nearby alternative that is much closer to the position of the median voter. As noted above, there are many reasons that this can happen, including the idea that some legislators and policymakers would prefer to keep the wage at $7.25 because of economic or political arguments. That’s fine: the position of the median voter is only one of the influences on policy. My problem was not with Cowen and his correspondent opposing or being agnostic on the minimum wage; my problem was their assertion that the persistence of the $7.25 minimum wage demonstrated the median voter theorem. That’s just silly. It should be fine for people to support a policy without having to impute the median voter to their position. Lots of unpopular positions are perfectly reasonable!

    • Jonathan (another one) says:

      I don’t think they’re exactly attributing the failure to the mean voter theorem. I take what they’re saying as a criticism of the notion that the two parties are that different or that changes in which party gets in makes much of a difference in actual policy enacted. The minimum wage is actually not the best example here… immigration is a much better one. But the argument (vis-a-vis the minimum wage) is that sharply different differences between the party platforms lead to little change in actual policy, even when the party in power changes. The argument is that parties are so close (because of the MVT) that who wins an election is random. But the actual policies that can be enacted are still determined by the MVT. The “strength” of the MVT in that case is partly ironic… you can win power by ignoring it, but you can’t do the things you’d like.

Leave a Reply

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.