Math Zero vs. Political Zero

It seems I hit a nerve last week when I said:

But if economists say, as they have for centuries, that a minimum wage raises unemployment, reporters treat them skeptically and feel they need to find a contrary quote to "balance" their story.

Turns out the new Democratic Congress is considering raising the minimum wage as its first big action, and political partisans are itching to argue about the subject.   For example, see the commentary at Business Week, Washington Monthly, and Cosmic Variance.

I intended my claim to be interpreted with the usual social science qualifiers.  It was about the sign of an average of economist judgment regarding a point estimate of an all-else-equal, causal long-run effect to be expected on overall employment from raising a minimum wage.   That sign has been consistently negative for over a century, as seen in most basic economics textbooks.   But people instead interpreted me as saying things like:

  1. raising the minimum wage would be bad
  2. every economist dislikes a minimum wage
  3. employment has never risen after a min wage rise
  4. employment in every industry must fall
  5. the effect is always large, no matter how small the rise
  6. the effect is immediate, without delay
  7. no economics model allows the opposite sign
  8. every study’s estimate had a significant coefficient
  9. my favorite model says it, so it must be true

Quite a lesson in the power of misinterpretation (not always accidental).   To set the record straight, most all economists should agree that:

  1. A low enough minimum wage has no employment effect
  2. A high enough minimum wage induces a devastating fall
  3. The effect of a small rise is hard to see amid other effects
  4. Models (e.g., monopsony) sometimes allow positive effects
  5. Far more studies have found negative effects than positive
  6. Due to study errors, we always expect some contrary results

We should also agree that regarding a small (e.g., 10%) rise in the min wage in the U.S. today, plausible models do not allow for a large positive effect, but they do allow for a large negative effect, or for a near zero effect.   

Part of the problem is that ten years ago two respected Princeton economists used impressive statistical techniques to find estimates insignificantly different from zero for the effect of small rises on particular states and industries, contrary to most of dozens of previous studies.   But many others don’t think this changes our total evidence by that much. 

Putting it all together, taking into account both theory and data I say:

Regarding our best estimate of the employment effect of a small min wage rise, while many have recently said this is near zero, more say it is substantially negative, and I have asked around and found *no* economist who says that it is substantially positive. Thus, I conclude, any reasonable average of these estimates must be negative, and has been so for a while. 

It seems to me that I am on safe grounds for any mathematical definition of "zero."   But apparently there is another concept of "zero" out there, because many complain that since some economists say their estimate of the employment effect of a small rise is near zero, I am wrong to say that the economic consensus has long been that a minimum wage lowers employment. 

I’m too close to this to be a good judge of what exactly all this says about bias, but it seems to me that it must say something. 

GD Star Rating
loading...
Tagged as: , ,
Trackback URL:
  • sa

    interesting comment about the effect being -ve no matter how small but certainly not+ve….

  • timely

    When the minimum wage is raised to keep up with inflation, it’s not actually being raised in *real* terms. That’s all the DC wants to do. So all other things being equal, companies should be able to do just as well paying people $7.25/hr today as they did paying people $5.15/hr ten years ago.

    People complain about the price of goods and services going up. However, historically this hasn’t happened, because the price of goods has already gone up — due to inflation. Do you think McDonald’s is charging the same for a Big Mac as it was ten years ago? Of course not, but some of the entry level employees are getting paid the same.

  • http://profile.typekey.com/halfinney/ Hal Finney

    This is a good example of a significant problem which I have run into in getting at the truth on complex issues: it can be very difficult for a layman or outsider to get an accurate view of the scientific or academic consensus on the issue. As Robin’s story illustrates, even insiders can find that they have difficulty accurately summarizing the state of play in a manner that is concise and not misleading.

    Supposing it is true that a “reasonable average” of professional economic opinions is that this particular effect is negative, is that the same as a consensus? How much confidence does this deliver to a layman who only wants to devote ten minutes (or ten seconds!) of thought to the question?

    There are so many complex issues today, and such richness of analysis and expertise available within relatively cloistered communities. All I really want to know is, what is the academic consensus on this issue, and how strong is it? That’s very likely to be a better guide to the truth than something I might come up with on my own. I wish we had better institutions for exposing this information and providing transparency into these professional communities.

  • http://profile.typekey.com/simon112/ simon

    Given that economists’ opinions are not independent, is the average opinion of economists really all that much more accurate than that of a typical econimist? If not, the fact that “some economists say their estimate of the employment effect of a small rise is near zero” should be mentioned by reporters, even if “any reasonable average” is positive.

  • http://profile.typekey.com/robinhanson/ Robin Hanson

    Simon, experts opinions on every topic are correlated, both because they are correlated with truth, and because of other social effects.

  • http://profile.typekey.com/simon112/ simon

    Ok Robin, so do you believe that the truth on economic beliefs tends to be near the average of economists’ beliefs with high probability, or is there a significant chance that the truth might lie closer to some outlying group of economists’ views than the average?

    If it’s your own considered opinion as an economist that a rise in the minimum wage will create a significant increase in unemployment, that’s fine. What I object to is you presenting the average of economists’ beliefs as an essentially unassailable authority without justification.

  • http://profile.typekey.com/robinhanson/ Robin Hanson

    Simon, why would you be more interested in my particular opinion than in the distribution of economist opinion? It is a great question to consider: what does the distribution of truth look like, relative to the distribution of opinion. It is a topic I was thinking of posting on someday.

  • http://profile.typekey.com/simon112/ simon

    Robin, the opinion of an economist who thinks independently should be weighted more than one who thinks with the herd (though perhaps less than the herd as a whole)(also, this assumes we don’t have further information on who is likely to be less biased, or smarter, etc.). Judging by some of the stuff on your home page, I think you may be something of an independent thinker. It’s sad to see you overwheighting (in my opinion) the beliefs of experts, because it might lead you to be less creative in the future.

  • Carl Shulman

    I would also be interested in refining the sample of ‘all economists’ to account for a) knowledge of labor economics and b) professional commitments, e.g. published studies supporting one side of the question. The collective opinion of the group of economists who display good factual knowledge in response to a survey about the scientific literature on a question, but have no personal academic stake in the debate (have not authored papers on the subject, have not mentored graduate students involved, etc) would be be a more powerful signal to me than an AEA poll.

    A heuristic for journalists would be to interview three randomly selected economists from a directory of ‘labor economists’ or ‘education economists,’ after excluding the leading contenders in the ‘controversial’ debate in question. This procedure might assuage their concerns of balance without blowing up outlier viewpoints to occupy 50% of airtime.

  • http://profile.typekey.com/simon112/ simon

    Carl, knowledge per se is necessary, but not sufficient to have your own worthwile opinions. Unless the answer to the question is contained within the factual knowledge itself, you have to either think about the implications of the knowledge or get your opinions second-hand from someone else.

  • http://profile.typekey.com/robinhanson/ Robin Hanson

    Simon, it is sad to see you under-weighing the beliefs of experts, because that will surely lead you into belief error.

  • http://profile.typekey.com/robinhanson/ Robin Hanson

    Carl, it would indeed be interesting to compare the quality of different subsets of expert opinion.

  • http://profile.typekey.com/simon112/ simon

    I think you are being evasive here Robin. I actually did have another reason to ask what you personally believed. Assuming that you believe that a higher minimum wage would lead to a significant rise in unemployment, would you change your opinion if the following change were made to the situation?:

    -all the evidence, and theoretical considerations remain unchanged
    -all the other people switched sides (everyone predicting an increase of unemployment now predicts no significant effect, and vice versa)

    I suspect that your opinion would remain more or less the same. I’m not exactly accusing you of bias, since I believe expert opinion is overrated, but I am accusing you of presenting reasons to believe that are different from your own.

    By the way, I myself oppose an increase in the minimum wage, in part because I think that there will probably be enough increase in unemployment to outweigh any benefits.

  • http://profile.typekey.com/halfinney/ Hal Finney

    One problem with your hypothetical, Simon: it’s hard to imagine this alternative world where the evidence is unchanged but all the opinions are reversed. That’s a bit contradictory. It’s easy to play games with counterfactuals but once you start breaking the rules, the worlds become poorly defined.

    But I’m curious, what do you think an unbiased person would do in that situation, someone who saw the truth with unclouded eyes?

  • http://profile.typekey.com/sentience/ Eliezer Yudkowsky

    It appears to me that averaging all expert estimates together (the mean or expectation) implies, under a Bayesian perspective, that you have some kind of Gaussian distribution for your subjective beliefs about their probable errors. In other words, you think that any given economist is about equally likely to be mistaken in one direction or the other, as far as you know; you think that mistakes of larger size have probability proportional to the inverse square of that size; and so far as you know, it doesn’t seem any more plausible that economists would make correlated mistakes than anticorrelated mistakes.

    If you thought that mistakes of larger size had probability inversely proportional to their size, you would take the median of the expert estimates.

    If you didn’t even want to make that much of an assumption, you’d take the modal estimate (a nonparametric approach, and the one that we follow for most scientific disputes because most disputes aren’t about quantitative variables).

    You could make a good case for simplifying the social rule down to “Report the modal opinion only”, because any other method for combining expert opinions has enough flex in it to end up just as controversial as the expert opinions themselves.

  • http://jewishatheist.blogspot.com JewishAtheist

    You’re playing games. Suppose the economists were split 90-10 with 90% saying the minimum wage would not cause unemployment to rise and 10% saying it would cause unemployment to rise 1000%. How does taking the average make more sense than simply presenting the two camps’ opinions?

    Your taking the average is just a game to bolster your belief.

  • http://www.hedweb.com/bgcharlton Bruce G Charlton

    It has been fascinating to se this minimum wage issue shape up in the USA. I think it is, on the whole, a healthy sign that the economic perspective is given a public airing because minimum wage has been fairly frequently raised here in the UK without any public discussion.

    On the other hand, I think progress in modernizing societies depends on separating social functions such as economics (and science, and education, and the military) from direct political control. The current movement for ‘raising the minimum wage’ is at root an imposition of politics upon economics, and using a quite different system of evaluations than economics; rather analogous to the government deciding on what type of antibiotics should be prescribed for pneumonia, or the method by which school children should be taught to read (both of which pretty much happen in the UK!)

    It does not really make much sense for politicians (or journalists) to delve into economic discussions and controversies about minimum wages, when it is obvious enough that raising the minimum wage is a _political_ rallying cry, designed to symbolize government _concern_ with poverty plus or minus inequality. RtMW is designed as a consciousness-raising measure, an attempt to re-shape the political agenda and priorities – it is certainly not designed to have a specifically economic effect.

    RtMW will not have a beneficial effect on raising the living standards of the poor (because it will probably do nothing or more likely make things worse in this respect). Nobody who understood these matters and actually wanted to help the poor/ reduce inequality would actually choose RtMW as a method of doing it – it is a strategy, a means to an end. In a sense it is an anti-economic policy, because it implicitly suggests that political goals should trump economic evaluations.

    At root this is a moral debate. Politicans who favour RtMW feel that it is morally correct to enforce a higher minimum wage because we/they should be morally concerned about the poor/ inequality, and that economic reasoning should _not_ be applied to these discussions.

    This is why poverty campaigners focus on Africa (which has a relatively small population – albeit many countries), rather than Asia (which has most of the world’s population in a few countries). In Asia poverty is being rolled back with amazing rapidity – perhaps one of the greatest achievements in world history? But in Asia poverty is conceptualized and addressed economically – which is of no interest to moral campaigners. Meanwhile in Africa, everything is moralized using non-economic evaluations, and the poverty just keeps getting worse (Greg Clark’s forthcoming book of genius (IMO) – Farewell to Alms – suggests that the current poverty in countries like Malawi is worse than has ever previously existed on this planet; thanks to well-meaning interventions such as modern medicine and Western ‘aid’ – non-economically applied).

    SO: I think poverty should be regarded primarily economically. The cure for poverty is wealth, and that is the domain of economics.

  • Marc_Geddes

    The impression created by Robin’s previous statements were at best very misleading. The *implied* impression was that raising minimum wage causes unemployment. What was not mentioned is the existence of a threshold and it’s only after the threshold is breached that the negative effects are seen. Raising minimum wage below the threshold causes *little or no* effect on employment. And by all estimates the US minimum wage is far too low, given that mimimum wage increases have failed to keep pace to inflation (i.e it’s far below the threshold at which further raises would cause unemployment). In Robin’s own words:

    “To set the record straight, most all economists should agree that:
    A low enough minimum wage has no employment effect”

    Note that this fact alone completely contradicts purist Libertarian ideology.

  • http://profile.typekey.com/simon112/ simon

    Hal said: But I’m curious, what do you think an unbiased person would do in that situation, someone who saw the truth with unclouded eyes?

    It depends on why they believe what they believe: whether they believe what the experts say or their own judgement. Which strategy is best depends on how reliable the person’s own judgement is compared with the average of “experts”. For most people most of the time believing the experts is almost certainly the best bet. What I claim but Robin seems to deny is that there are people for whom relying on there own judgement is the better strategy for some questions. Of course, if he believes that the average of experts always trumps one’s own personal judgement, this is inconsistent with him ever taking a controvertial stance on any question in economics.

  • Marc_Geddes

    Yes simon, taking an ‘average’ of experts views is misleading, since this average would be skewed by the views of a few dissenters. I would expect that the ‘majority view’ of the experts would give me a more accurate view. The ‘majority view’ of US economists is that you could substantially increase US minimum wage and produce zero addtional effect on unemployment. The reasoning for this (which looks very solid) is, as I and someone else pointed out, US mimimum wage has not kept pace with inflation is already far below (for instance) current European norms.

    From the wiki:

    “The current (July 2006) minimum wage in France is set at 8.27€ (EUR) per hour (~10.46 US dollars)”

    “The current minimum wage in the UK for adults aged 22 or older is £5.35 ($10.01US), compared with $5.15 in the US.”

    “(In the US) The minimum wage fell about 29% in real terms between 1979 and 2003. For the median worker, real hourly earnings have increased since 1979, however for the lowest deciles, there have been significant falls in the real wage without much fall in the rate of unemployment.”

  • http://profile.typekey.com/simon112/ simon

    Marc: I don’t know why you are addressing me, but anyway: pointing out that the US has a lower minimum wage than Europe is not a good argument for the minimum wage not having an effect since Europe does indeed have much higher levels of unemployment. Also, the failure of unemployment to drop in the US could be due to increases in other costs of employment besides wages, or due to a drop in the productivity of the least skilled workers. The drop in wages for the lowest deciles is evidence that at least one of these is in fact the case.

  • http://kybernetikos.com kyb

    It seems to me that the arguments that sprung up around this are a good example of your original point. Do you think there would have been as much argument if a physicist had said what you said? As far as I understand it your claim was very narrow and specific. There is a consensus that raising the minimum wage has a negative effect on employment rates. People automatically interpret this as you saying that therefore one should not raise the minimum wage, and that is a whole different question, involving many variables beside the employment rate that economists don’t agree on. It reminds me a lot of the controversy around the claim that abortion has decreased the crime rate. You can be aware of this phenomenon and still not be in favour of abortion.

    I don’t really think this is a difference between economists and physicists though. I think that no scientist is given much credence when their theories disagree with peoples opinions. Theoretical physicsists have the advantage of rarely coming up with theories that challenge normal peoples opinions, so they just say “that’s cool”. People argue and look for balancing opinions when they perceive a scientific theory as challenging their beliefs (whether in fact it actually does or not). Look at evolution.

    The difference is between normal speech and scientific speech. In normal speech, mentioning a benefit of a strategy is usually done by an adherent to the strategy and with the sole purpose of promoting it. Policy decisions are separate from scientific discourse, and although they should be influenced by them, often a group of people will decide that despite all the tangible benefits demonstrated by science, there are intangible costs that outweigh them. They don’t have to be logical in a strict sense of the word.

    Perhaps the best way is to put a disclaimer on every example of scientific speech. Something that shows that while what you are saying is considered by you to have been demonstrated empirically, positive findings do not necessarily indicate a personal endorsment of any policy nor do negative results necessarily indicate that a policy should not be persued.

    So anyway, back to your original point, I don’t think anyone respects physicists more than economists, it’s just they disagree with them less often.

  • http://profile.typekey.com/robinhanson/ Robin Hanson

    It seems there are are lot of Californians or people who stay up late here. :)

    Simon, I agree with Hal that your counterfactual is not obviously coherent. But on topics where I know some evidence but experts know a lot more evidence, I would certainly go with expert opinion. It is an interesting question to ask whether one should weigh independent-minded experts more than others; it is not obvious to me. Also, it can be socially valuable to me to publicly explore an unusual opinion, even if in the end I think it more likely false than true.

    Eliezer, the modal opinion is not robust to redefining the parameter in question. Of course neither is the mean. A simple median here is for a substantial negative effect, by the way.

    JewishAthiest, if 90% of experts thought a substance was harmless, and 10% thought it would kill me, I’d want to know the mean more than the median.

    Marc, it is just not true that the US min wage is clearly so low as to have no substantial effect, and this is not the majority view of economists.

    Kyb, yes, at root the reason physics and economics are treated differently is that people care more about economics topics.

  • Physicist

    Oh for heaven’s sake. Can we please stop comparing economists to physicists? But why, you might ask, should we not compare the two?

    Nuclear Weapons, Lasers, Computers etc.
    Show me an economic theory that produces equally impressive results, then we’ll talk. Until then let go of the analogy. It makes you look ridiculous.

  • http://profile.typekey.com/robinhanson/ Robin Hanson

    Physicist, our world is rich, and some of that wealth is embodied in technology, and physicists were part of the process that produced that technology. But academic physicists can claim only a tiny part of the credit for computers. Our social institutions, which allow us to coordinate together to specialize, to invent, to build, to distribute, etc. deserve a lot of the credit. And economics deserve a tiny part of the credit for those institutions. Which tiny part is larger overall, is hard to say.

  • physicist

    That’s a rather amusing rationalization.
    Look I will be the first to admit that physicists benefit a great deal from the institutions in modern America. Were I living in Mali instead, I’d probably be a peasant farmer. That having been said, to suggest that the contributions to modern life from physicists and economists is even remotely comparable strikes me as laughable.
    A little thought experiment:
    Go back in time to 1900 and kill every physicist on earth and burn all their records. How does the world turn out?
    Now go back in time again to 1900 and instead kill every economist on earth and burn all their records. Which alternate world would you rather live in?

    Now consider what the response of the majority of Americans would be to that counterfactual and you’ll gain some insight into why they treat physicists and economists differently.

  • http://kybernetikos.com kyb

    Physicist: It’d be interesting to trace those alternate histories. I’m not involved enough in economics to know what net positive effect economists have had, but I don’t think the physicists are responsbile for as much as you seem to think. Many of the important inventions of the 20th century were made by engineers, chemists and most recently biologists. Physicists are often involved in inventions by accident, or due to other hobbies. Inventions spring from necessity (and so, most naturally to engineers rather than physicists), not theory.

    Of course, any positives you give to the physicists should be weighed against the negatives – nuclear weapons for example.

  • passing by

    “That having been said, to suggest that the contributions to modern life from physicists and economists is even remotely comparable strikes me as laughable.”

    Indeed, physicist, I agree. But I’m chuckling at the pretensions of the physicists.

    I count economists like Adam Smith through the recently departed St. Milton as having a much more profound an impact on many more lives than physicists over that same time. When you look around in wonder at modern life, thank economists and philosophers who thought hard and long how to advance the concept of individual liberty, rule of law, representative democracy. Without those advancements, none of this wealth would be here. And without the wealth, physicists would still be at the level of opining Greek philosophers: interesting and wild imaginations but no way to test any of them.

    To the point of Kyb, though, I think physicists do serve an essential role in uncovering basic laws that allow other displines to apply knowledge. For example, I guess, but cannot say for certain, that Newton’s ideas permit engineering as we know it today.

    Still, if we had to lose one or another, I would give up physicists first, engineers second and economists last. Without economists to remind people about counterintuitive effects of our collective decisions, our whole wealth-generating system will erode pretty quickly, as people make apparently “just” social policies that utterly destroy incentives to strive. All the engineers and physicists in the world won’t save you from the dark ages imposed by the resulting collapsing economies.

    You cannot operate a modern mixed economy on the basis of E=MC(2) or F=MA, not matter how sublime those ideas.

  • passing by

    Oh, and re this: “Nuclear Weapons, Lasers, Computers etc.
    Show me an economic theory that produces equally impressive results, then we’ll talk. Until then let go of the analogy. It makes you look ridiculous.”

    Echoing Samuelson’s response to Stanislav Uliman, I’d put some of Ricardo’s ideas up against those any day. You want impressive results arising from theory? There they are.

    In fact, at risk of engaging in a “post hoc, ergo prompter hoc” fallacy, I genuinely wonder whether you would have those material items without the insights of Ricardo and his ilk. After all, humans were around a long time before we very recently became wealthy enough to allocate resources to basic research. And the wealth came only after we translated certain key insights about human economic interactions into consistent social policy.

  • http://profile.typekey.com/simon112/ simon

    “But on topics where I know some evidence but experts know a lot more evidence, I would certainly go with expert opinion.”

    I agree, most of the time. That’s why I said some people _for some questions_ – you get to pick which questions you will defer to the experts on and which questions you think you can do better on.

  • http://robertvienneau.blogspot.com/ Robert

    The following has nothing to do with any imperfections in competition, principle agent problems, transaction costs, etc.:

    “But, as economic theory has learned since the 1930s, the pattern of activities adopted in the face of long-run factor-price changes can be complicated and counterintuitive. Consequently, the long-run demand for factors can be badly behaved functions of factor prices.” — Michael Mandler, _Dilemmas In Economic Theory: Persisting Foundational Problems Of Microeconomics_. Oxford, 1999. p. 34.

    I could also cite Garegnani (1970), which argues that the classical economists did not explain prices by supply and demand functions, and that they were correct.

    In short, Robin Hanson’s claims about price theory are badly out-of-date.

  • http://profile.typekey.com/robinhanson/ Robin Hanson

    Robert, I don’t think I made any claims here about price theory per se.

  • Douglas Knight

    Robin Hanson, author of “He who pays the piper must know the tune,” strikes me dumb:
    “Simon, it is sad to see you under-weighing the beliefs of experts, because that will surely lead you into belief error.”

    On Card & Krueger:
    I don’t give much weight to an individual study and I think fancy statistics are ways of hiding data-mining (more from the investigator than from the reader), but they also did a meta-analysis which is much more relevant to this blog: they made accusations of publication bias. Those statistics certainly made an impression on me.

  • Marc_Geddes

    Note that the UK, with a minimum wage nearly *double* (yes double, 100% greater) than the Federal minimum in the US had a *lower* unemployment rate than the US last year. Funny that.

    ‘passing by’, I really can’t let a couple of your remarks pass ;) This one in particular had me in fits of laughter:

    “Without economists to remind people about counterintuitive effects of our collective decisions, our whole wealth-generating system will erode pretty quickly, as people make apparently “just” social policies that utterly destroy incentives to strive. All the engineers and physicists in the world won’t save you from the dark ages imposed by the resulting collapsing economies.”

    There’s an interesting graph that our esteemed Robin Hanson showed us earlier this year where he charts the estimated amount of wealth in existence over time (human history). Through-out most of human history in many places the ‘just social policies’ you so seem to dislike were non-existent. Fans of the free market had everything they could want… no minimum wages, no welfare, no big government etc etc. But all through that time the graph of human wealth is almost completely flat. Then shortly after the rise of big government and the implementation of just social policies (i.e the modern nation state) the graph of wealth in existence suddenly explodes. Funny that.

    So ‘passing by’, it appears that your own ideas on economics went out of date sometime prior to the 19th century…

  • http://kybernetikos.com kyb

    Marc: I hope you factored population growth into that analysis.

  • http://robertvienneau.blogspot.com/ Robert

    “…economists say, as they have for centuries, that a minimum wage raises unemployment…” According to Robin Hanson, that is not a statement about price theory. It seems to me to be a false statement about how economists have understood wages, at least, under, say, some interpretations of classical economics. See, for example, Garegnani (1970). Frank Hahn would also disagree with Robin Hanson’s statement, albeit on different grounds.

  • Michael Sullivan

    I posted a comment over at Drum’s journal that you may not have read, as it looks like you gave up on that comment stream when you said something about posting this piece just before my comment showed up.

    I’ll post some of it at the end here because I still see those sections as relevant, but I’m going to change my tack on the main thrust a bit based on your post here.

    I agree that your assertions about the state of economic discipline are reasonable characterizations. I can even agree that you can get from there to an assertion that “economists agree that a minimum wage raises unemployment” given certain contextual caveats. The problem is that in this argument you are suggesting that if you say that in a lay context (where we are almost certainly talking about a particular proposal which is likely to be a small increase and have a small, zero or uncertain effect), it is unreasonable for someone to counter with an economist who says that such an increase will probably have no significant effect on unemployment. Because that’s the sort of “balance” that one would usually see in that spot. If you commonly see the media quoting cranks who claim that the minimum wage is likely to *increase* employment, then your demand for a cite of an economist who believes this in response to Drum’s argument that it’s an unsettled question is more plausible.

    Averaging the respected practicioners claims is one way to get an estimate of “reality”, but unless you get to that average by combining samples from essentially equivalent experiments, your average doesn’t have any more claim to Truth[tm] than any particular estimate on the spectrum (which includes zero when talking about typical min-wage proposals).

    Drum’s assessment of the economics is reasonable. His critique breaks down because the physics you cited is every bit as unsettled as the economic question on minimum wages (really far more unsettled), and yet, as you say, whatever J.Random Physics PhD. has to say about some new theory, even when it is fairly off-the-wall speculation, will almost always be taken as TheTruth[tm] by a science reporter.

    Here’s what I said over at Drum’s journal about that:

    ——-
    To me, the scandal here is as much the completely uncritical and uneducated treatment of physics and medicine results as gospel, as a lack of respect for economic opinions. That’s not to say that there isn’t a great deal of ignorance, even willful ignorance, about economics, including a lot of people who think they know much about the subject that ain’t so. But the minimum wage looks like an iffy example of that, given the C&K study. Free trade would probably be a better example. In general though, economics suffers from modelitis. Most of the very clear results come from very clear models rather than mounds of empirical trials, and there is much more question about models’ predictive value than there is about the models used in standard physics.

    Admittedly much of what gets treated as “fact” in science journalism is not standard at all, but new, cutting edge results which are as likely to get overturned or radically re-interpreted as stand for a long time. But I’d argue that the too skeptical approach to economics is, if anything, a bit closer to optimal truth seeking behavior than is the reverence accorded physics and other disciplines.
    —-

    I stand by that. I too have issues with the idea that you have to “balance” every claim. Some claims are nearly universally supported in the discipline and offering the same outlier to everybody as if the issue is truly unsettled is foolish. Failure to differentiate between difference of degree and differences of kind is another example. But the treatment I’ve seen of the minimum wage issue tends not to be of this sort, because nobody argues that there is an employment increase, they merely argue that the employment costs are near zero or otherwise small and outweighed by benefits. And they can find plenty of real economists to support those positions. They don’t have to dig up the same three or four everytime who make a career out of being dissenters like you see in the ID debates.

  • passing by

    Marc Geddes:

    “Fans of the free market had everything they could want… no minimum wages, no welfare, no big government etc etc.”

    *Everything?* Um, how about freedom to contract, or any freedom at all, for that matter?

    I am not sure those countless generations of slaves, serfs, conquered subjects, and just plain old peasants would agree that they lived in eras where free market transactions were the norm. Actually, I’m guessing they would tell you they lived in a very regulated society…on pain of death.

    Accordingly, I’m not sure all those nasty, brutish and short years of our ancestors are exactly a good model of what benefits free markets offer us.

    True, if I had to choose between Mongols pillaging my town and and over-regulated economy , I would likely choose the latter. But that is a false choice, isn’t it?

  • http://profile.typekey.com/robinhanson/ Robin Hanson

    OK, I’ve been too lenient with all this generic political sniping. Keep your comments related to the post or they’ll be deleted.

  • http://profile.typekey.com/robinhanson/ Robin Hanson

    Michael, yes, my primary complaint was about the *relative* treatment of speculative physics and economics. So regarding that issue I don’t mind if the media finds “balance” with opposing views in economics, if they would also do so in speculative physics. I’m not convinced of your summary of the media on the min wage, that it all acknowledges the sign and just argues about the size of the effect. Otherwise people wouldn’t jump on me so vigorously for making my claim only about the sign.

  • Marc_Geddes

    passing_by, as to the historical graph of human wealth versus time again, the fact the big explosions in wealth all occurred after technological revolutions spread to large numbers of people (i.e. the agricultural, industrial revolutions etc) suggests that the effects of economic policy on wealth creation are completely swamped by factors like technology and education any way.

    Robin, Getting back to something on topic, possibly the reason why economists don’t get as much respect from the general public is due to three factors (It seems to me): (a) Economics is a lot more complicated than physics, (b) It’s a lot more domain specific and (c) There is much more motivation for hidden agendas and special interest groups to distort or bias results.

    Economics ain’t physics. For physics it turned out that the basic abstract principles behind the universe were very simple and has wide applicability. Economic models however are more complex and have narrower applicability. There are all sorts of confounding variables and the complexity of human nature to deal with. Results obtained in one narrow social context don’t necessarily apply to another. Change the background social conditions even slightly and some particular model might fail. Given all this some might say we’re better off with plain old common sense. Finally because economics is strongly associated with politics there are reasons for greater distrust due to distortion from special interest groups.

  • http://robertvienneau.blogspot.com/ Robert

    At http://robertvienneau.blogspot.com/2006/12/wages-and-employment-not-determined-by.html, I provide some references arguing that the intro textbook story about minimum wages is invalid and incoherent on its own terms.

  • Barry

    “But if economists say, as they have for centuries, that a minimum wage raises unemployment, reporters treat them skeptically and feel they need to find a contrary quote to “balance” their story. ”

    A standard liberal critique of the mainstream media (AKA ‘liberal media’) is precisely that. Given an overwhelming consensus of experts, reporters will strenuously seek opposing quotes for ‘balance’, even if those come from outright intellectual prostitution brothels and astroturf organizations (e.g., CEI, AEI, Heritage, CATO, TCS).

    Prominent recent examples would be evolution and global warming; a bit back would be supply-side economics. Of course, the difference would be that very, very few prominent biolgists or climatologists would support fraud and junk science, while ‘expert’ support for Reagan/Bushnomics was plentiful.

  • Richard Dobson

    As others have pointed out, economics is argued over heavily because it is nearly impossible for most people to look at economic theory without considering the way in which it is likely to drive policy. Theory that supports policies which are (for possibly unrelated reasons) considered harmfull is likely to be opposed, this is a flaw in human nature, but it is pervasive and certainly not limited to economics.

    Secondly, people are often skeptical about economics because, it does often seem that economic decision making, even by supposed experts, is driven by ideological bias. For recent examples of this see supply side economics, or Stiglitz detailed account of the failure of the washinton consensus policies in Asia and Russia. Or another example, look at how liberalising the Californian energy market, led to massive fraud, failures in supply and increased cost. All of these are examples of people having “faith in free markets” or in a particular set of polcies (washington consensus) rather than sitting down and really thinking about how the particular markets involved would react to the proposed policies.

    Thinking based on ideology or personal or class interests occurs at both ends of the economic spectrum, both economic liberalizers and socialists are guilty of it.

  • http://www.intheagora.com/archives/2007/01/the_minimum_wag.html In the Agora

    The Minimum Wage

    The Democrats were carried into power this past election largely upon an anti-incumbent tide, but one should also acknowledge that House Speaker Nancy Pelosi put together a modest package of legislative priorities in order to give the Dems some substan…

  • http://t-a-w.blogspot.com/ Tomasz Wegrzanowski

    I’m just going to boldly defy the textbook economics and say that all economists for centuries have been wrong, and here’s why I think so.

    Everything I can think of indicates that demand for unskilled labour (let’s use this label for all kind of cheap hard-to-outsource labour – the kind that typically gets near minimum wage) is highly inelastic. So if legislation forces costs of unskilled labour higher by X%, demand will drop by Y%, for Y being much lower than X. For the sake of example let’s make X=20%, Y=2%, but the argument follows the same way.

    If minimum wage is below market rate for a particular kind of unskilled labour, raising it a bit won’t have any effects. If it’s above market rate it will have an effect of increasing costs of unskilled labour. Worker gets some part Z% of this cost, the rest going for taxes, and all kind of expenses related to employing someone. Let’s assume the Z% doesn’t significantly change before and after introduction – share of admin costs will most likely go down, but taxes can go either way, so let’s just hold Z% constant.

    So far we have two big effects: Y% decrease in employment in unskilled labour sector, and X%-Y%-Y*X% increase in income for people in this sector, even including those that just lost their jobs. With numbers from the example it’s 2% drop in employment and 17.6% increase in income for the working poor. This money isn’t taken from thin air of course, it’s taken from everyone, and demand reduction effect is already included here. As a side note nobody would need to lose a job because of it, the working poor as a group would just work 2% fewer hours, what can mean some people losing jobs, or some people going part-time, or any combinations of two.

    What textbooks of economy get wrong is focusing solely on the smaller demand drop effect, and completely ignoring a lot more significant transfer of wealth effect.

    If this transfer of wealth has even modest positive impact on economic growth, or it changes structure of demand to increase demand for unskilled labour, it can be easily a lot stronger than naively calculated reduction in demand.

    There are two more effect here – increasing value that unemployed people can get from getting a job, and decreasing the value unskilled people get from increasing their marketable skills. The first is mostly a good thing, the second is mostly a bad thing.

    I don’t know if transfer of wealth increases economic growth, or changes demand structure, or how it quantitatively make people go up the employment scale. I know that any of these effects can easily overwhelm small drop in demand for unskilled labour, and that’s why we need hard empirical data instead of this naive theory – the thing is we don’t have much macroeconomic hard data on anything, there are just too many things happening simultaneously, and we’re limited to observing a tiny not-even-close-to-independent sample.

    Even if this caused tiny increase in unemployment, I would be surprised if this transfer of wealth didn’t have a lot stronger beneficial social effects (reducing all kinds of poverty-related bad thing like crime, children malnutrition, and so on) than negative effects from tiny unemployment increase.

  • http://occludedsun.wordpress.com Caledonian

    Simon, it is sad to see you under-weighing the beliefs of experts, because that will surely lead you into belief error.

    Robin, it is sad to see you giving so much weight to qualifications without first determining whether they represent real expertise.

    The opinions of experts may be correlated because they converge on truth. There may be other reasons. A look across human history indicates that people socially recognized as experts were often not only mistaken, but did not know what they were talking about.

  • Mark Buchanan (another physicist)

    I’m surprised I feel moved to comment, but the discussion of physics vs economics is just too interesting to leave alone.

    My opinion as a physicist is that much of physics, especially the physics that gets lots of popular press, is highly speculative and should be treated as such. We don’t know how many dimensions the Universe has, we only have some interesting and extremely speculative theories that suggest certain answers. Does anyone really believe otherwise? We have some idea of the age of the Universe, though the accepted estimate has changed significantly in the past decade. Might it change again by a factor of 2 or 10 or 1000 over the next ten years? Probably not, but this is speculative territory and I wouldn’t be shocked (just mildly, and pleasantly, surprised).

    But then we have other areas of physics, atomic physics, condensed matter physics, fluid dynamics and statistical mechanics, etc, where we have very accurate theories that make wonderfully accurate predictions. All this has been built up over a long period by repeated testing of ideas against experiment. This success (which has made a great deal of modern technology possible), isn’t traceable to the brilliance of physicists as people, but to the effectiveness of the scientific algorithm for developing knowledge, in which both speculation and empirical test play equally important roles.

    I think economics doesn’t get quite as much respect because it hasn’t been as strongly committed to testing its theories empirically, and rejecting those ideas that do not work. Serious people are still quibbling over the Efficient Markets Hypothesis, despite absolutely overwhelming evidence against it. It’s taken 30-40 years for it to become acceptable for economists even to begin studying systematically the way people really make decisions, in contrast to what perfect rationality would imply.

    I also find it more than a little strange that economics papers so often adopt an extremely formal mathematical tone, with theorems and lemmas, etc, much more so that do papers in theoretical physics. My suspicion is that this is in part to give the illusion of scientific certainty, and therefore to confer authority on the theory and its developers. Much more impressive would be theories for basic economic phenomena that met very stringent empirical tests, theories to which every economist could point and say “this clearly works”. As far as I understand, there is just no such thing.

    One other comment regarding what journalists should believe when asking the experts about how things work. A good journalist doesn’t just ask the experts for their opinions, but for the reasoning behind those opinions. If you find that lots of experts have different opinions, and different reasons behind them, you begin to suspect you’re dealing with a field in which there isn’t a great deal of sound knowledge (in my opinion). And if an expert gives bad reasons, or incoherent and inconsistent ones, you’ve got reason to suspect their motives.

    So, short answer, forget about string theory and dimensions of the universe. On far more basic issues, physics has very powerful and well tested theories and economics has nothing remotely comparable. This might be because the social world is much more complicated. It is definitely NOT, in my opinion, because physicists are smarter than economists, but may also in large part be due to some particularly damaging theoretical paradigms in which much of economics has been stuck for a long time (rational expectations, fixation on equilibria, etc.)

  • nolrai

    wait there are people out there that belive that raising the minimum wage, will decrese unemployment?
    I have been for raising the wage, but always despite the fact that it will destroy some peoples jobs. (I just think keeping wages artificially high is good for society).

  • Pingback: Overcoming Bias : The Felt & The Unfelt