Saturday, April 09, 2016

101ism in action: minimum wage edition


A while ago I went on a rant about the dangers of "101ism", which is a word I made up for when people use an oversimplified or just plain wrong version of Econ 101 in policy discussions. Well, here I have a perfect example for you. And among the culprits was me.

It started when American Enterprise Institute scholar Mark J. Perry tweeted the following graph about minimum wage:


I was annoyed by the word "actually". My current pet peeve is people not paying attention to empirical evidence - I think if you say "actually", there should be more than just a theory backing you up, especially if evidence is actually available. So I started giving Mark a hard time about ignoring the empirical evidence on the minimum wage question. 

That's when Alex Tabarrok jumped in and defended the cartoon, saying that it's just a basic supply-and-demand model:


But that's not right. This cartoon actually doesn't show the basic D-S model at all. Let's look at it again:


The basic, Econ 101 D-S model is a model of a market for a single homogeneous good. In the case of the labor market, that good is labor. There's one kind of labor, and everyone who does it gets paid the same wage. Since the wage in that model is equal to the marginal revenue product of labor, this means everyone's labor generates the same amount of revenue (this is also obvious just from the assumption that labor is homogeneous; if everyone's doing the exact same work, they can't each be generating different amounts of revenue). A wage floor in the basic D-S model will put some people out of work, and will raise the amount of revenue generated by each person who keeps her job, thus raising wages as well.

In the cartoon, however, different jobs are stated to generate different amounts of revenue. Also, the last panel implies that a wage floor leaves the revenue generated by workers unchanged. So while the cartoon and the D-S model both predict that minimum wage causes job loss, it's only a coincidence - they're not the same model at all. 

The cartoon could be trying to portray a sophisticated model of heterogeneous labor in a highly segmented market. Or, far more likely, it could just be some sloppy political crap made by a cartoonist who doesn't remember his intro econ class very well. Either way, Econ 101 it ain't.

When Alex claimed that the cartoon is an "accurate portrayal" of the D-S model, I waved away his protest, basically saying "Who cares, evidence comes first." But (possibly because I had a nasty virus...excuses, excuses), I failed to notice until this morning that the cartoon is not the D-S model at all! I gave it the benefit of the doubt and assumed Alex was right. But Alex must not have been paying close attention - since he teaches the D-S model in online videos, he obviously does know how that model works.

So the cartoonist, and Mark J. Perry as well, are peddling bad economics. But they managed to momentarily convince both me and Alex that they're just peddling good' ol simple Econ 101. How did they do that?

In my case, it was because I committed the fallacy of the converse. I assumed that because the basic Econ 101 model says minimum wages cause job loss, and the cartoon says the same, the latter must be equal to the former. That's like saying "Horses have legs, I have legs, therefore I must be a horse." I suspect that Alex made the same mistake. And so we both gave a stupid cartoon far more credit than it deserved.

This is 101ism at its worst. It got me too, people. It's a plague, I tell you! A plague!


Updates

This post has stimulated a lot of interesting discussion about what the basic Econ 101 supply-and-demand model actually says (see comments, also Twitter).

One point has been that the definition of "the amount of revenue a job generates" - the language in the cartoon - is not clear. I took it to mean "marginal product of labor", but some people take it to mean "average product of labor". Either way, though, the APL generally changes as total labor consumed changes, so the cartoon still doesn't make sense if we define "revenue generated" as APL.

Alex Tabarrok, in the comments, seems to suggest a model in which one "job" is not equal to one differential unit of undifferentiated labor, but actually represents several units. If this is the case, each job will have a different total revenue benefit to employers, and the MPL of a job can't even be well-defined (since it's a discrete unit rather than a differential). So with these definitions, you can definitely say that "each job generates a different amount of revenue". 

But the point is, no matter how you define a job, or the revenue generated by a job, that amount will in general change for each job under a wage floor. The amount of revenue one person's job generates depends on who else is working. That's what Econ 101 teaches - or ought to teach, anyway. And that's what the cartoon gets wrong. It shows a wage floor eliminating every job whose "revenue generated" is lower than the wage floor before the implementation of the wage floor. Actually, basic Econ 101 D-S teaches that a wage floor eliminates every job whose total revenue benefit to employers (the integral of marginal revenue product over some range represented by the "job") is less than the wage floor (representing the cost of hiring the worker) after the introduction of the wage floor. Since the wage floor changes the quantity of labor consumed, and since the marginal revenue product of labor is in general not constant, those things are not the same. 

And that is why the cartoon is a bad representation of Econ 101. Good Econ 101, in my opinion (and probably in most people's opinions), should teach how marginal benefits and costs change according to the quantity consumed. The cartoon shows them not changing. That's not good Econ 101.

84 comments:

  1. The fly in the ointment is the effect of a "mandated" minimum wage (i.e. federal, state and/or local baseline). If the model is true D-S there will not likely be price controls on the upside (Horrors! Stop upward prices in response to heavy demand? Whatcha trying to do, curb profits???) so there is no limit to the price/demand side. Wages march steadily upward as demand for specialized work swells. This, incidentally, accounts for the most extravagant executive compensation packages evah!

    On the downside, however, economic slowdowns, weather catastrophes, civil unrest, and/or other reasons for unemployment (buggy whips and drive-in movies become obsolete, for instance) a swelling number of job seekers creates a downward-moving feedback loop driving the minimum ever lower. This is why there must be a safety net. Otherwise unemployed people -- very likely overqualified -- will accept ever lower wages for a vanishing pool of available work.

    If the protected minimum rises enough to eliminate the need for social services, those expenses are recovered in lower demand on the tax burden. (Thank you, Mr. Hanauer.) And to the extent those at the bottom have more discretionary income, does that not mean every dollar over the subsistence level will feed the economy?

    I worked my whole life with the working poor and you can be sure that any extra money they received -- gifts, gambling winnings, second or third job, illegal drug transactions, whatever -- every dollar was spent almost immediately. None to my knowledge were feeding a savings account or checking with their stock advisor for advice. If it wasn't used for necessities, any discretionary income went immediately back into the economy.

    ReplyDelete
  2. Well, Noah, you’re sorta right and sorta wrong about this one.

    Sorta right: the workers are homogenous and all receive the same wage.

    Sorta wrong: jobs have varying marginal products in D&S-land. The difference between the wage (set at the MP of the last worker hired) and the products of the inframarginal workers is a sort of rent (not equivalent to profit of course because of fixed costs).

    Thus the problem with the cartoon is that it fails to recognize there is a single wage applying to all jobs in a given labor market, hence rents. Aside from this, I think it does represent the logic it claims to represent.

    But I agree with your original point that “actually” should have something to do with empirical evidence, particularly when the theory in question is dubious. Search models, for instance, are plausible alternatives (better IMO) and are compatible with Card & Krueger-type results. Build in some bargaining and it’s even clearer. I’ve worked with efficiency wage models, which also attenuate the employment loss effect. And there are macro issues with S&D representation of labor markets: the curves are arguably not independent.

    I think the self-certainty exemplified by Mark Perry and Alex Tabarrok in this case borders on the absurd.

    ReplyDelete
    Replies
    1. Dorman, pretty sure you're wrong here. The inframarginal workers have higher *potential* productivity, which is why they keep their jobs and get higher wages when you set a wage floor - the higher potential productivity becomes higher actual productivity. Imagine a McDonald's where you fire half the workers, so the remaining workers now have to staff the cash registers AND the deep fryers. They're each more productive, and their wages go up. But they weren't as productive before the firings, because before the firings you had some people staffing the registers and others staffing the deep fryers. That's what's really happening Econ 101 style in D-S land. The inframarginal workers are working at less than their maximum potential productivity in equilibrium, and the fact that they can do this at a lower effort cost is their producer surplus.

      Delete
    2. Dorman is correct although I am mystified at the unnecessary pot shot. Noah there are lots of models of downward sloping demand not just the one you sketch. See my textbook for one consistent with Dorman.

      Delete
    3. Nah. If you have a D-S model for umbrellas, and the equilibrium price is $10, it doesn't mean that some umbrellas are REALLY AWESOME and some umbrellas are REALLY CRAPPY but both are mysteriously selling for $10. It's an undifferentiated good.

      Labor D-S is no different.

      Delete
    4. And Alex, if you're teaching differentiated goods models as your basic Econ 101 D-S model, you must have a lot of confidence in your students, as well as a highly unorthodox approach... :-)

      Delete
    5. Noah, you are digging a hole. First umbrella you keep in your house. If price is low enough you buy a second for the car. The umbrellas and price are the same but the second umbrella is less useful than the first.

      Delete
    6. Yeah but the umbrellas are interchangeable, you can use either one. Not heterogeneous quality.

      The "digging a hole" troll ain't gonna work on me, sir. ;-)

      Delete
    7. Alex, the minute you buy the second umbrella, the marginal utility of the first umbrella goes down. That's the key here. When you have two identical umbrellas, their marginal utility is the same. It doesn't matter which got bought first!

      Delete
    8. Yes, of course. The umbrellas have the same MU and similarly all the workers are interchangeable and have the same marginal product. But the *tasks* the umbrellas do have different MU. Similarly, as Peter Dorman said, the *jobs* (not the workers) have different marginal products and so when the minimum wage rises the jobs the workers are allocated to changes--just as the cartoon describes.

      Delete
    9. Nope. The jobs are the good in a labor market. And basic D-S is a model of homogeneous goods. Hence, the jobs are homogeneous in that model.

      If the jobs were different, they'd receive different prices, just like umbrellas of different quality receive different prices. There's no reason for firms to pay the same wage for two different jobs with two different marginal revenue products. There's one wage in the basic D-S model of the labor market because the jobs are homogeneous.

      Delete
    10. marcel proust9:13 AM

      Isn't is simpler to think think about this in terms of neo-classical production function, with homogeneous labor and homogenous capital? Holding K fixed, then the MP of less L is greater than the MP of more L? No need to worry about different tasks or differentiated labor. It would fit the McDonald's example well: the number of fryers and cash registers is unchanged, the number of undifferentiated works is less, and their MP is now higher than before.

      Delete
    11. Paul Claire3:31 PM

      Ok, why the hell should I believe ANYTHING economists say when you can't even get something THIS INSANELY SIMPLE agreed on?

      Delete
    12. Anonymous8:31 PM

      Because this isn't an "INSANELY SIMPLE" topic? (And, because Noah and Alex are more or less bloggers with PhDs, and about fifty steps removed from the economists doing the research on this frontier?)

      Delete
    13. I have a sense this depends on mg productivity vs average productivity. Think about firms and mgc costs as analogy. If number of firms fixed, and decreasing returns to scale, mgc costs higher than average costs and firms make profit. Free entry sets mgc costs equal to average costs and so no profit. I think inframarginal workers must be producing more if average productivity higher than mgc productivity... competition between firms shoul bring average productivity in line with mgc productivity...

      Delete
    14. Ok, why the hell should I believe ANYTHING economists say when you can't even get something THIS INSANELY SIMPLE agreed on?

      Well, I remember a physics prof of mine used to go around asking the other profs simple-sounding questions about intro topics, and getting all kinds of conflicting answers...

      Delete
    15. And, because Noah and Alex are more or less bloggers with PhDs, and about fifty steps removed from the economists doing the research on this frontier?

      OK, but I strongly suspect that people on the frontiers of micro research don't spend much time thinking about the meaning of the simple models taught in 101 classes... :-)

      Delete
    16. Anonymous5:30 AM

      Not at all. The Chicago tradition (in the vein of Becker, and later, Murphy) is all about thinking hard about the most basic of models you can work with, for better or for worse.

      You're a perfectly fine blogger. But the econ blogosphere gives readers the deeply false impression that wars in comments sections are representative of how the academic world behaves. It makes people believe 'Austrian' bloggers and think they represent some kind of real faction in our discipline!

      Delete
    17. Inframarginal differences were explained in my Ph.D. program using a river. Labor is homogeneous, but it is a lot harder to farm the land that is further from the river. The MPL will be much higher for farms on the river bank, but it is the last viable farm furthest from the banks that will set the wage = MPL for the inframarginal firm. The best farms will be highly profitable with MPL > wage.

      Delete
    18. the econ blogosphere gives readers the deeply false impression that wars in comments sections are representative of how the academic world behaves. It makes people believe 'Austrian' bloggers and think they represent some kind of real faction in our discipline!

      Unfortunately this is true. I keep saying that "Austrian econ" isn't econ, but it's hard to get that message through...

      I haven't read Murphy, I will check him out.

      Delete
    19. I'm with Dorman and (gah) Tabarrok. The demand for labor is analogous to consumer benefit. The first worker or unit purchased is worth more than the second, etc. Otherwise why would willingness to pay be higher on the margin the fewer the quantity purchased. If you buy a dozen eggs, they are identical but you'd pay more for the first than the twelfth, if you were buying them singly. In equlibrium the employer gets rents for labor other than the n'th one hired. Similarly, workers other than the n'th are paid more than their supply price.

      Full disclosure, I believe none of this. But that's what the textbooks say.

      Delete
    20. So the cartoon is faithful to the cockamamie theory.

      Delete
    21. At least there is one thing you economists can agree on---Austrians are not real. Now if you could finally get to the bottom of a single good D-S model!

      Delete
    22. Anonymous6:52 PM

      Dorman, Tabarrok, Ben Wheeler, Anonymous @ 12:15 am, and MaxSpeak have all explained the Econ101. Noah erred. End of story.

      Delete
    23. Guys, see the update at the bottom of the post.

      Delete
    24. I must side with Noah in this - Econ 101-world's D-S model assumes homogeneous product (labor), and it assumes that under perfect competition each unit of labor receives it's MPL. In 101-world, any unit of labor whose product exceeds the wage would automatically know to go to another firm (no search costs) that would be equal time/distance from his home (homogeneous employer), and the new firm would automatically pay his/her MPL consistent wage (because firms would be competing for his/her higher productivity). No employer rents here in 101-world!

      Therefore, the initial equilibrium would represent only all labor whose MPL is equal to the prevailing wage given the particular intersection point of the D-S curves.

      With the added above equilibrium wage floor, labor would still be demanded of this homogeneous labor force because of the inverse of the principal of diminishing marginal utility (the 9th unit of labor would produce more MPL in a given employment market than the 10th or 11th).

      In the cartoon at issue, each house would actually occupy a separate D-S curve equilibrium. Many/most of the occupants of those houses depicted as winding up underwater would, in fact, be able to establish beachheads on dry land because of the increased marginal utility resulting from the smaller labor force demanded leaving very few who remain underwater.

      Delete
  3. I don't know Noah, when I googled your blog on minimum wage, this post appeared:

    http://noahpinionblog.blogspot.com.au/2013/12/wage-subsidies.html

    Which seems to counteract all that you say here. Is the empirical evidence that minimum wages don't destroy jobs so strong? It seems not. If you advocated wage subsidies instead of minimum wages, why now go back on it?

    Or perhaps is to avoid being in accord to right wingers? I do not understand why it's cruel to advocate the common sense view that minimum wages don't necessarily benefit the people they inted to help (unemploymentrates are after all much higher for young people with low qualifications, the target population for this policy) and look for alternatives.

    ReplyDelete
    Replies
    1. Michael Yost11:30 PM

      "young people ... the target population for this policy"

      That's not really correct though. Most people affected by minimum wage increases are at least in their 30s. (http://www.epi.org/publication/wage-workers-older-88-percent-workers-benefit/)

      Delete
    2. It's true in the sense that young, low-skill workers are directly inhibited from using low wages as an edge.

      Delete
  4. How about 101ism on Rent controls? I stumbled across this article today, when reading about rent controls (and the basic 101 logic behind it), and now I'm not sure what I should think.

    https://psmag.com/in-defense-of-rent-control-3cb453119116#.nc6zcdcei

    ReplyDelete
    Replies
    1. I've been thinking the exact same thought amidst the news that previous econ model conclusions about the effects of minimum wage laws don't turn out to be empirically true

      Delete
  5. Anonymous11:28 PM

    Discussions like this make me a bit nuts, for several,reasons but one of which is the reification of a complex interaction we choose to bundle and call a "job."

    Sometimes reification ("thing-making") works fine. You can talk about how many beds a hospital has, shorthand for the beds, staff, physicians and so on needed to handle a particular number of a usual mix of patients. You can talk about this many "head" of cattle because everyone knows the heads come complete with bodies.

    The simplification in this case leaves out crucial elements (like the bodies of those cattle) and then behaves as though only the heads are being moved through the ups and downs of the economic model. But the unseen bodies are still there, hungering and mooing, while economists and policy wonks wonder where all the dung is coming from.

    Employers get far more than consistent hamburgers for the wages they pay. One thing they get is a society largely free of roving bands of raiders, and largely free of huge destitute populations cultivating new and interesting plagues.

    Paid or not, people will always exert energy towards goals. If not employed, they don't power down like a misplaced iPhone.

    Noni Mausa

    ReplyDelete
  6. The cartoon should read "people who can produce more than $15/hr of value" and "people who can't produce more than $15/hr of value" and it would be exactly right.

    ReplyDelete
  7. Anonymous12:15 AM

    Your understanding of the 101 model strikes me as odd. If everything was paid to MP, then there would be no profit. Thus the MP of inframarginal hours must not just be hypothetical. The reason they are paid below their MP is just the law of one price which apply to homogeneous and interchangeable goods(producers, in that case, workers are price-takers and cannot affect the price of their the goods they sell). If they could all get paid to MP, this would indeed be a differenciated good model, or monopoly model(or whatever). As Atabarok is saying the same good is used for different things. The first hour is the most useful(in a mcdonald, for instance, serving clients), the last one is the least useful(e.g. cleaning) and must be equated to wage( because of declining MP, and allowing for fractionnal hours of work). All this works even if workers have the same skills and are interchangeable. And is consistent with the cartoon(hours below the MP of minimum wage are cut back).

    ReplyDelete
    Replies
    1. Anonymous12:30 AM

      I think you're going to have to admit that you screwed up on this one. You should have kept questionning the factual basis of the 101ist notion that the right model that applies to wages is the econ 101 D-S one instead of trying to go for such a (weak) gotcha.(because it's just not reflected in the evidence and we have better models that are more consistent with them)

      Delete
    2. "If everything was paid to MP, then there would be no profit."

      Well, exactly! In equilibrium, there *are* no pure profits: capital gets paid its MP, labor gets paid its MP, and there is nothing left over to call "profit."

      Delete
    3. The employer incorporates the cost/return of capital in their "cost" calculations, so that they will employ labor in equilibrium whose MP covers this as well. This return to capital is the firm's profit in accounting terms, it's just that the accounting profit includes no rents.

      Delete
    4. The employer incorporates the cost/return of capital in their "cost" calculations, so that they will employ labor in equilibrium whose MP covers this as well. This return to capital is the firm's profit in accounting terms, it's just that the accounting profit includes no rents.

      Delete
    5. Anonymous3:34 PM

      Gene Callahan: i may be mistaken, but i think this is confusing ''marginal productivity'' with ''average productivity''. If everything is paid to AP, then you're right, no profit left over in equilibrium. Thinking in terms of declining MP, there's no reason to conclude that there is no profit in the short term(market) equilibrium. As i said, different hours have different MP(with marginal product being higher for the first hours) and the reason why everybody is paid to the MP of the last hour is that we could employ say ''the 10th hour'' to do the job which is done by the 1st hour(which leads to price-taking behavior). The marshalian notion that there is no profit left in the long run is another kind of equilibrium(because firms keep entering the market as long as there is economic profit to be made). I think all this is consistent with the model being taught at the elementary level(and the cartoon). Of course some of what is normaly called ''profit'' gets called ''return to capital'' in the model, but the firm itself is trying to maximize profit(by assumption).

      Delete
  8. I recently read your post about incorporating simple empirics into 101 (yeah, I only stop by every-once-in-awhile). Today I was thinking that the minimum wage section of 101, where most text books drop the classic supply and demand analysis and just move with nary a hint of discussion, would be a really great place to throw in some simple and famous empirics (i.e. Card and Krueger). So I came here to be late to the party and post a comment about something that has probably already been covered.But then I am greeted by a post about people abusing their 101 education in a debate about the minimum wage. Perfect.

    Thinking about 101 and the minimum wage also made me wonder why the monopsony case is rarely if never taught in 101. I mean, it has been floated as one possible reason for C&K's results (probably not), and there is some reason to believe that certain labour markets could become captured by large employers such that something resembling monopsony could exist.

    Imagine a section in 101 where the case of a price floor is shown in both competitive supply and demand and monopsony augmented with a discussion of Card and Krueger (method and interpretation) for a little food for thought. The basic, baseline DD they do in that paper is certainly simple enough for people in 101 to understand. Plus, you could show a simple graph that explains the common trends assumption (and possibly the problem with violations of it) which I am sure at least some students could grasp.

    This might be of great service to the world in general. Those who enter debates about the minimum wage with their 101 war chest will hopefully have a more well rounded view of the topic and we will be less subjected to the posts of the form "C'mon guys, It's simple supply and demand...".

    ReplyDelete
    Replies
    1. Undergrad econ profs seem to be panty-piddling scared of teaching more than a couple classes on monopsony and market power. I guess it's too hard a topic.

      Delete
    2. Nathanael1:39 AM

      Employers notoriously compare notes on wage scales. When there's spare labor around, they often tacitly agree to keep wages low, and the same as each other. It's the cartel variant of monopsony.

      Delete
  9. Great post Noah, as your favorite neighborhood econo-wannabe, I must say I love it when people actually give a fair and accurate analysis of minwage, even just a simple partial analysis as you have done here.

    Minwage is a complex issue that involves stuff like philosophical dialectic, as presented masterfully by Steven Keene in his youtube lecture "Dialectics as a foundation for dynamic nonequilibrium monetary economics"

    Is Minwage a wage standard or a productivity standard? It must necessarily be both. But the connection between wages and productivity is mediated by business and capital, while the standard is created through public political channels. Complex checks and balances happen to allow labor and capital to work things out.

    Some people say that workers should naturally be able to negotiate a fair market wage, but fail to recognize that politics is an important channel whereby people participate in market outcomes.

    Most of the time workers are willing to meet the productivity standard or are already there, and simply need to negotiate a matching wage. However, not everyone(including yours truly) likes this productivity standard.(I'd rather have smaller rents, and smaller costs than higher wages)

    I've written a lot about minwage, but perhaps the most lucid thing I wrote was this letter my local paper published in print and online: http://www.deseretnews.com/article/865603601/Why-a-local-minimum-wage-makes-more-sense.html

    Keep up the work talking about minwage. It's a topic that needs more coherent analysis, people who think the Card-Krueger study gives the full picture, are usually unduly politically motivated.

    As I understand it, the 101 Supply/Demand model could apply to labor in a situation I call "commodity labor". Both workers and jobs are basically interchangeable. This happens in "assembly-line" like operations whether it's making burgers or cars. There's no real connection between the worker and the job.

    These type of operations can usually scale production to match demanded output, which tends to make the demand for labor inelastic, which actually matches some empirical observations like card and kruger did looking at fast food chains.

    So inelastic labor demand would match empirical observations of relatively little impact on employment levels of the minimum wage.

    Bear in mind, the model would predict no impact if the minwage is below the current market equilibrium. But as you point out, most of the time we have different workers and different jobs.


    ReplyDelete
  10. This comment has been removed by the author.

    ReplyDelete
  11. Is there actually any employer who manages all the "shadow" assumptions behind D-S model?

    ReplyDelete
  12. "American Enterprise Institute scholar"

    One of those words does not belong with the others.

    ReplyDelete
  13. The ultimate measure of the minimum wage is net social benefits. So the correct cartoon would put net social benefits along the land instead of houses. The key is to set the minimum wage where net social benefits are the greatest.

    ReplyDelete
  14. Here is how the minimum really-really works (warning: advanced taxi driver wonkish level):

    If a minimum wage hike is priced right it will bring in more dollars for fewer hours of work -- but the employment picture doesn't end there. The extra dollars will be diverted from purchases higher wage folks would have made to buying decisions lower wage people make.

    If lower wage people make exactly the same buying decisions as higher, then, the effect on employment will be null (it's a little trickier, but that's the idea). The difference will be in consumption: the lower wage will consume more and the upper less (that's the idea).

    In practice, consumers tend to aim their purchases somewhat more towards vendors who hire in the same pay spectrum. Poor people buy other poor people's second-hand cars -- used car lot too pricey. Middle wage people buy off the used car lot. Higher wages put you in the showroom.

    1/11/14, NYT article "The Vicious Circle of Income Inequality" by Professor Robert H. Frank of Cornell: " .. higher incomes of top earners have been shifting consumer demand in favor of goods whose value stems from the talents of other top earners. ... as the rich get richer, the talented people they patronize get richer, too. Their spending, in turn, increases the incomes of other elite practitioners, and so on."
    http://www.nytimes.com/2014/01/12/business/the-vicious-circle-of-income-inequality.html?src=me&f_r=0.

    Upshot, a higher minimum wage – if priced right – should predict a few more customers at Mickey D’s and a few less at Olive Garden. Likely that’s what we saw in Card and Krueger (1992) -- studying firms with “giant” 33% labor costs no less.

    Further clarification: if orange sellers and apple sellers started out charging the same prices but orange sellers toyed with their price until they could get more money for fewer oranges from apple sellers, then, the businesses that apple sellers prefer to buy from, cabbage sellers, would suffer – but the firms that orange sellers buy from, lettuce sellers, would prosper.

    ReplyDelete
  15. There are two problems:

    1. Econ 101ism: Ramming a complex problem into a far-too-simple straightjacket.
    2. Blind empiricism: Putting excessive weight on the arguments of an economist with some data and a regression equation.

    ReplyDelete
    Replies
    1. Well, given the wide variety of theories we have on hand to predict the effects of minimum wage, don't you think we should choose to use one that matches the evidence?

      Delete
    2. Stephen, are you really so obtuse as to imply that minimum wage unemployment skeptics are relying on one guy with "some data and a regression"? This has been a persistently studied topic, over a variety of contexts and specifications, with a lot of null results.

      Delete
    3. Or, you know, you could talk about the large number of empirical studies that came before C&K that zeroed in on a concensus around a 10% increase in the min wage leading to a 1 - 3 percent decline in youth employment AND the large body of empirical evidence that came after C&K that has mostly found estimates in the same range as before or estimates that showed no employment effects.

      Then you could talk about all the empirical work that has looked at other possible effects of a min wage increase. Think job tenure, total wage bill, wage compression, wage increases away from the lowest end of the wage distribution, etc.

      And then maybe we could talk about how this all pertains to the models.

      Delete
    4. Of course C&K was neither the first nor last word. It certainly was one of the first credible words--the empirical literature leading up to it was poor quality for the most part. There's a reason it's highly cited, and it's not just because it found no effect. In fact, if you read the first paragraph of the paper, they point this out.

      The very mixed later literature, which includes both precise effects both of disemployment and null effects across a variety of measures and contexts, should point to a nuance that "101-ism" is insufficient for. Framing the literature, as Stephen does above, and as do empiricism-only-when-it-suits-me pundits like Tabarrok, as "mountains of theory vs. one guy with a regression" is either completely ignorant or deliberately misleading.

      Delete
    5. My bad, Dave, I assumed you were responding to me, but my guess is you were responding to Stephen. (Either way, I agree with your post!)

      Delete
    6. Was definitely responding to Stephen. I agree with your characterization of his comment being obtuse. There is plenty of credible evidence out there that suggest minimum wage increases have only small (null?) to moderate employment effects and don't cause the unmitigated disaster that the 101 theory warriors suggest.

      I also agree that the research after C&K is far more credible. I think when considering the post C&K literature people should consider that the real minimum wage had been falling for the better part of a decade before C&K. This would probably mean that the minimum wage wasn't super binding in the eras when the studies were performed and this might have something to do with the null/slightly positive estimates in that research.

      And of course, none of that literature looks at minimum wage increases as large as the current call for a movement to $15/hr would be. But still I think the empirics thus far help ease fears of huge employment effects.

      Delete
    7. Sorry, you guys didn't get it. What I mean is that quantifying the effects of an increase in the minimum wage is a big job. This is a general equilibrium problem, with many possible effects, and some of the most important may be ones that you will only see over a long period of time. You need a complicated dynamic structural model, and you need good data. I think it's a hard case to make that we are so confident about the effects that we can make solid policy recommendations. A theorist friend of mine - albeit a very cynical character - likes to say that there is no empirical work in economics that has convinced anyone of anything. Maybe he's right.

      Delete
    8. Anonymous11:38 PM

      I find S.Williamson the most convincing here and declare him the winner over Smith, Perry, Tabarrok. Walras >> Marshall everywhere and always.

      Delete
    9. Re: Steve W., my micro prof Mancur Olson thought that about econometric work too -- nobody is ever convinced who doesn't want to be.

      Delete
  16. Noah, I posted a better cartoon honoring the ecosystem of the coastline. Mark Perry's cartoon is childish.

    http://angrybearblog.com/2016/04/better-cartoon-min-wage.html

    My message to Mark Perry... It's the ecosystem stupid.

    ReplyDelete
  17. Kudos for recognizing a self-mistake and setting the record straight! Kudos also for exposing the equivocation fallacy of some who should know better (you know who you are) trying to change a basic model while calling it the same name. Tsk.

    ReplyDelete
  18. "everyone's labor generates the same amount of revenue (this is also obvious just from the assumption that labor is homogeneous; if everyone's doing the exact same work, they can't each be generating different amounts of revenue)."

    ^ This "obvious" reasoning makes no sense. If two factories have machines of different qualities, the same labor inputs to each will generate different revenue.

    ReplyDelete
    Replies
    1. What revenue does the HR department generate? Or the CEO? Part of the front-line worker's wage is confiscated by other actors within the firm.

      Delete
  19. This comment has been removed by the author.

    ReplyDelete
  20. Empirical evidence tells you how minimum wage set at 2/3 of median wage works? Do tell.

    ReplyDelete
  21. Anonymous3:38 PM

    Noah, Congratulations for making it past the generalization about how "progressives" all think. Hearty. I noticed a couple proper uses of the it's contraction. Was that needling?

    ReplyDelete
  22. A worker's wage isn't determined by marginal product of labour, workers are price-takers in a monopolistic market, and labour demand is inelastic.

    Also, why do economists assume that the labour demand curve doesn't move down RIGHTWARD below the minimum wage, considering all workers need to meet their autonomous consumption and you can't do that at 40 hours a week at $7/hr.

    What I want to know is why economics professors insist on teaching this patently false nonsense to undergrads. First year physics profs don't teach their students that the world is flat and the sun is a golden chariot, even though that would make their subject matter easier.

    Are economics professors teaching patent nonsense because it's what's demanded by all the little neo-Nazi Murray Rothbards in charge of state education policy? Because sorry dude, you econ profs have created all those Ayn Rand wannabes by teaching this crap to every polisci, engineering, commerce and finance student who wants to graduate with an econ minor.

    ReplyDelete
    Replies
    1. Labour SUPPLY curve trends down rightward below the minimum wage, not demand curve.

      Delete
  23. Sorry Noah, I'm with Alex on this one. Equal marginal products and wages are not incompatible with jobs having different average productvities in Econ 101. You should stick to "Evidence first", that is the right intuition.

    ReplyDelete
    Replies
    1. Manuel, whether we use average productivity as the definition of "how much revenue a job generates", the cartoon is silly because this changes when a wage floor is imposed... ;-)

      Delete
    2. Anonymous3:57 PM

      Exactly, Noah! Thanks for being the only economist I've seen admit that productivity measurements change when wage changes. And while productivity is reduced as wages increase, it also increases as wages are reduced. So if a large number of intermediate-level products suddenly are created with much, much lower wages, productivity increases substantially.

      Delete
    3. @Amateur economist.....Huh?

      Delete
    4. Hmm, I believe now I see your point (sorry, should've read your entry more carefully). But as a humble Econ 101 lecturer I believe your point is too subtle for Econ 101. You seem to object to the use of "jobs" in the cartoon meaning something like different qualities of workers, but I still believe Alex is right that what the cartoon tries to represent is jobs performed by workers ("tasks") that have different values in the market. In Econ 101 you usually put emphasis that D for labour is derived from D for product.

      Delete
  24. Noah: Last I checked, the principal costs of food at a restaurant that pays minimum wage is labor, cost of food and land rent.

    If employers need to raise minimum wage, why isn't the primary effect, over time, to decrease rent? After all, if the McDs closes down, who's going to take its place -- a restaurant that caters to a population that can pay more? Seems unlikely that you can do this everywhere. It seems to me, instead, that the new tenant will point to his P&L statement and tell the landowner that he needs to reduce his rent if he wants to lease the land.

    ReplyDelete
    Replies
    1. Nathanael1:34 AM

      I think you're actually right. But rent is a whole 'nother issue, because building supply is restricted by very strict zoning.

      Delete
  25. Nathanael1:32 AM

    Forget Econ 101. Here in the real world there are only four ways in which raising the minimum wage can *ever* cause job loss; three of them *practically never happen*.

    #1. At $10/hour it's cheaper to have a human do the job; at $15/hour it's cheaper to buy a machine to do the job. This does not happen. If a machine can do the job, it usually costs way less than any employee and is being used already.
    #2. The business has appallingly low profit margins. The difference between $10/hr and $15/hr is the difference between profit and loss. This also doesn't happen very often, unless the business is *extremely* labor-intensive. Nobody goes into busineses because they expect such tight profit margins; they go into businesses with fat profit margins.
    #3. The business owner is an idiot. The business owner was not using the optimal number of employees before, and reduces the number of employees because they read about the rise in the minimum wage, though they could have done so earlier. Or the business owner is even more of an idiot; the business owner was using the optimal number of employees before, and reduces the number of employees becuase they read about the rise in the minimum wage.
    #4. At $15/hour it's cheaper to relocate the factory to Mexico (or China, or Idaho) where there's a lower minimum wage. This actually happens, but not in Seattle. Seattle is a high-cost-of-living, high-cost-of-property area, so any business which would have relocated for this reason has already relocated.

    The key point here is that the number of employees required to do the work is *inelastic* -- it's determined by the production process, it doesn't change when the wages change.

    ReplyDelete
  26. Subscribe with Epic Research Private Limited and get essential market news with lots of trading knowledge, which is helpful in the investment.

    ReplyDelete
  27. Anonymous1:54 PM

    "There's one kind of labor, and everyone who does it gets paid the same wage. Since the wage in that model is equal to the marginal revenue product of labor, this means everyone's labor generates the same amount of revenue "

    This is nonsense. If true, it means the demand curve for labor would be a horizontal line. H

    In the simple perfectly competitive market model the demand curve is downward sloping. Why? Because some jobs (perhaps those in firms employing larger amounts of capital) are more productive than the market wage. That firm, like every other firm, will continue to hire workers until their marginal revenue product equals the market price for labor. But it isn't as if each worker produces an identical amount even though they are paid that way. If a worker producing more balks they can be replaced with another identical worker. All workers get the market-clearing wage; all workers do NOT produce the marginal revenue product.

    ReplyDelete
  28. I don't understand general equilibrium, but as a curious scientist currently working through some actual ECON 101 lectures online, I'd like to add that lecture 2 on supply & demand seems to undermine the arguments critical of the minimum wage. For something that educated economists seem to be debating, the logic seems stunningly simple. I'm new to this, and very happy to have logic corrected if I'm wrong.

    Take the idea that the minimum wage hurts those its intended to help.

    The revenue of low wage workers = price*quantity=(hourly wage)*(hours worked).
    Therefore, for a small change in wages,
    (change in revenue)/(change in wages)=(hours worked) + e*(hourly wage)
    where e=(change in hours)/(change in wages) is the elasticity of demand for labor (defined as a negative number in the lectures I was taking).
    If e>-1 (i.e. if e is not very negative), then increasing the minimum wage increases the total revenue, when summed over all low wage workers.

    So, when people suggest that, according to ECON 101, raising minimum wage hurts people in the low-wage labor market, are they saying that they know elasticity <-1? Contradicting this doesn't require evidence that elasticity is particularly small (aren't elasticities >-1 pretty routine?). As I understand it, a market with e<-1 is one that's so elastic that there's no benefit to producers who form a cartel and jack-up prices. This seems odd: if it's at all difficult to get significant p values for empirical estimates of e then the confidence limits on e must get near zero. Are the data so useless that confidence limits on e extend from near-zero all the way past -1?

    Perhaps critics are saying that we should put aside the total monetary value of low-wage work - perhaps they instead object to the minimum wage based on distributional concerns. Consideration of non-monetary issues seems reasonable to me, e.g. perhaps the lower quality of life effects for those made unemployed outweighs the extra dollars to those who keep their jobs, even if the total revenue of low-wage workers has increased. But I think this depends on (i) the magnitude of the job loss vs the magnitude of the wage increase, and (ii) some value judgements. If I see these questions definitively answered by the remaining ECON 101 lectures, I'll be extremely impressed. More importantly, I read once that...

    the average burger-flipping job turns over about 3 times/year.

    (my personal experience with folks I know working minimum wage jobs is consistent with this - they switch jobs a lot).
    In this case, raising the minimum wage will mean spending more time between jobs, and getting a higher wage when working. If e>-1, then low wage workers get more money for slightly less hours worked/year. This doesn't seem like a policy hurting the people its trying to help.

    I've focused on the interest of low-wage workers, because I see critics claiming that minimum wages hurt low-wage workers. Perhaps if you add in the interests of folks with higher incomes, then minimum wages do hurt overall GDP. If concerns about the interests of higher-wage workers are a necessary part of the criticism of minimum wages, it'd be nice to see the critics acknowledge this.

    Steve H.

    ReplyDelete
  29. Oops. Forgot to normalize in that formula, but doing so doesn't change logic.

    ReplyDelete
  30. I'm confused about the notion of the "amount of revenue that a job generates," and Noah's comment above about how this will change as the wage floor changes.

    Presumably this is a case of English words not meaning the same thing as technical terms from a model (which IMO is a significant source of confusion) but isn't the issue with raising the minimum wage precisely that revenue has NOT changed, but expenses have increased?

    If a McDonalds can sell 1,000 hamburgers per day at $5 each, then its maximum revenue is $5,000 per day. Let's say it needs a staff of 10 people to produce those 1,000 hamburgers. Now if it's the case that if I only had 9 people then I could only produce 900 hamburgers per day then I guess you could say that the marginal revenue generated by that 10th job is $500, but I don't think that's how it actually works in the real world. For one thing I'm probably more interested in the marginal profit than the marginal revenue. Is that adding unnecessary complexity to the model, or is it essential to understanding what's really going on?

    ReplyDelete