Wednesday, March 20, 2013

Productivity figures have their limits.

We talked before about the fact that output per worker is not the same as the marginal revenue product of labor.

Bob Murphy also points out that composition matters for how much stock we put into productivity figures with the following hypothetical:
In 1968, a nation of identical workers collectively puts in 1 billion labor-hours, to produce 1 billion units of finished consumer goods, at a price of $1 per unit. (There are different types of goods, but they all have the same unit price.) Everyone gets paid $1/hour. This is the only production, so nominal GDP is $1 billion. The government also happens to set a minimum wage law of $1/hour, but this has no effect because that’s what the competitive wage was anyway. (Assume that not only the average product, but also the marginal product, is $1/hour.)  
In 1971, there is a major policy change that unleashes the printing press. The government’s cronies start getting humongous payments of new $100 bills, which they do partially spend on consumer goods, but which mostly go into financial assets, real estate, and precious metals. After decades of this new pattern, we find…  
In 2013, the same workers collectively put in 1 billion labor-hours, to produce 1 billion units of finished consumer goods, at a price of $2 per unit. Everyone gets paid $2/hour. This contributes $2 billion in nominal GDP. However, the fat cats provide “financial services” of $8 billion in nominal GDP, for a total nominal GDP of $10 billion. Since CPI has doubled, in 1968 dollars real GDP is $5 billion, a five-fold increase.  
EPI publishes a paper saying that if the minimum wage kept pace with “inflation,” it would be $2 in 2013. If it kept pace with real productivity growth, it would be $10/hour.
BLS does produce productivity figures by industry and presumably that would stand up to Bob's scrutiny better (I don't personally know how non-financial productivity behaved relative to the economy's productivity).

It's an interesting sort of thought experiment... maybe I'm not understanding it but workers seem to be paying other for consumption goods and fat cats pay other fat cats for financial services? I don't know. Even so, it still shows the limits of these figures especially if you think the financial sector is bigger than it ought to be.

UPDATE:

From the BLS tables -
Nonfinancial productivity grew 125.7% between 1968 and 2011
Total productivity grew 131.5% between 1968 and 2011

So there is some fat-cat finance outpacing but there's still been a lot of productivity growth since 1968.

btw - I know the division chief for major sector productivity at BLS, Susan Fleck. She was my gender micro professor and wrote a recommendation for me for that long term care fellowship I mentioned the other day. She has an essay on the productivity/wage gap here.

13 comments:

  1. Argh! Daniel, just settle this once and for all: Do you think there are at least 1000 workers in the United States who produce more than $20/hour for their employers, yet they are only earning minimum wage?

    (A) If you *do* think that, then I would like to hear your thoughts on how the labor market can be so screwed up. Then, I want you to pitch me a business plan for how we can hire these 1000 (at least) people at $10/hour, and still make $10,000 in pure profit per hour.

    (B) If you *don't* think that, then why aren't you denouncing this EPI study as, at best, terribly misleading and dangerous to low-skilled workers?

    ReplyDelete
  2. I love this "once and for all" language as if I haven't been clear! No, I don't think that (barring weird exploitative modern-slavery type relations which almost certainly exist somewhere but for which the minimum wage is not the solution).

    That leads me to (B).

    Because it's not misleading. The title of the damn paper is advocating a $10 minimum wage. I doubt anyone will mistake that for advocacy of a $20 minimum wage.

    Some people are pretty slow on the uptake, but I think most know the difference between $10 and hour and $20 an hour.

    I didn't read the CEPR thing that Warren was pulling from, but I doubt they said anything like that either.

    ReplyDelete
    Replies
    1. That's to Bob, above.

      I thought he'd like the fact that I agreed with him!

      Delete
    2. The $10 minumum wage proposal gets back close to the highest real minimum wage we have ever had in this country. As noted elsewhere, that was in 1968, which is why both of those studies are framed around 1968. $10 today would match the next highest annual minimum wage in history.

      So basically two left wing advocacy groups would like us to go back to "near max" real minimum wages and argue that its probably okay because productivity has grown a lot since then so its not likely to be too binding. Seems reasonable except this argument misses one important consideration. Was it binding in 1968 when the minimum wage rose 20 cents or 14% from the year before? Was there evidence of a pick-up in unemployment among low skill workers? This is pretty important since the catch-up increase they are advocating this time around is 38%.

      Delete
  3. I think Bob's point is that it is misleading to use the percentage by which the economy-wide average productivity went up to justify saying that the (inflation-adjusted) minimum wage should have gone up by the same percentage (disregarding is-ought issues of course), because the latter then for a crucial part is dependent on the assumption that the average productivity of workers at the minimum wage level (disregarding for the sake of the argument whether this is about one industry or for minimum-wage level jobs in all industries) went up by the same percentage as the economy-wide average and *that* assumption is nowhere argued for (and would also be implausible given basic economic principles)

    But Warren does make this inference, and then goes on to use its "conclusion" to challenge that employer. Moreover, from the conclusion of that inference the idea of a minimum wage adjusted for that economy-wide average productivity gain is a natural (though not necessary) one (though she does not literally state that idea (in this case that the minimum wage should be $22).


    I haven't read the EPI study itself but in any case I think the part of Bob's question that you left out, namely the one that pertains to Warren's statements may be more difficult to answer for you. It's basically: If you don't agree with (a) then why aren't you denouncing Warren's statements as, at best, terribly misleading and dangerous to low-skilled workers?

    ReplyDelete
    Replies
    1. 1. Right. Read my post from this morning where I argue this before Bob even wrote this post. See this post where I'm agreeing with Bob on another reason why the average productivity figure poses complications. I am not missing this point - I have been making this point several times now.

      2. I did answer that part of Bob's question. See my response following "That leads me to (B.)".

      Delete
    2. 1. I read that post. I read this post. I agree with nearly all of the points you make in them. I also agree with you that whales are mammals and not fish (though we may differ about dolphins). But these points are not what the discussions are about.

      In particular, both Bob and I claim that Warren's statements are at the very least ambiguous and likely intentionally misleading. Moreover, I claim that if you listen carefully to what Warren says she's not just being misleading but she actually liyterally makes a false inference when she says:

      "As productivity goes up, that is, as workers are producing more, then the minimum wage is gonna go up *the same*. And if that were the case, the minimum wage today would be about $22 an hour. So my question mr. something, with a minimum wage of $7,25 an hour, what happened to the other $14,75?”
      So let's analyze two stages of her statements here.
      1. First, the false inference: she herself ties the average productivity growth to the productivity growth at the minimum wage level by saying the minimum wage would go up the same, which assumes that the average productivity in minimim-wage jobs went about by the same rate as the economy-wide average productivity. But the latter assumption is wrong or at the very very very least unargued for. (and that is what Bob wanted to bring out with his scenario you discuss in this post).
      2. Then, on the basis of this false inference, it is nautral for the listener to think that the 2013 minimumwage equivalent of the 1968 minimum wage is $22 or at least substantially higher than the actual current one, especially when she aks the emplyer dude what happened to the $14,75 difference.
      Now she doesn't say something like 'you damn employers stole the difference and raising the minimum wage to $22 would merely be reredistributing that money back to the employers without any nasty consequences", and by not explicitly saying stuff like that she leaves open the possibility that there are other kinds of explanations for why $14,75 are missing or why the $14,75 number may not mean what people may think it means. And there being a difference between economy-wide average productivity increases and average productivity increases in minimum wage jobs - something you discuss in your posts and something Bob also discusses - is one possible explanation for the difference between the $22 and the actual current minimum wage. We agree with you on that! But it would be hella difficult for Warren to agree with that because she herself made the inference that allowed her to arrive at $22 by explicitly (well, as an unstated assumption required for her conclusion) assuming that the average productivity increases were the same economy-wide and in the minimum wage industries.

      You dig?

      Delete
    3. I understand what you've been claiming. I'm disagreeing with your claims about her inferences here. I don't think you have given any convincing reason to think that she is doing the sort of accounting exercise you're proposing she is doing - instead I think she is making a point about real wage stagnation.

      If you were right and she's making some kind of precise accounting statement, then what would be so absurd about proposing a $22 minimum wage? Why is nobody doing that? It doesn't really hold up.

      I think you are taking a rhetorical point - what happened to the difference - and imbuing it with a lot more technical significance than it ever came with in the testimony.

      I think she posed the "what happened to the other 14" to the economist, not the restaurant owner. That was a softball sort of point about wage stagnation. The real beauty of the testimony was where she took Dube's research and really took apart the restaurant owner's claims with it.

      I think you are investing more significance in this 14 line than it really played in the exposition of the argument in the testimony.

      Delete
  4. DK wrote: "I understand what you've been claiming. I'm disagreeing with your claims about her inferences here. I don't think you have given any convincing reason to think that she is doing the sort of accounting exercise you're proposing she is doing"

    Like, like... I mean, but... I friggin' spelled it out for you... I quoted her extensively. I showed what she said. And how her argument went. It was right there. On the screen. And I showed it. I showed it, Daniel.... so showed it...

    what else can I do. what.

    Dk wrote: "If you were right and she's making some kind of precise accounting statement, then what would be so absurd about proposing a $22 minimum wage? Why is nobody doing that? It doesn't really hold up."

    yes, oh Lord yes, that's exactly my point, don't you see. You're confirming it. Yes, if I am right in what she said then yes there would not be anything absurd about proposing a $22 minimum wage. it would be a natural thing for the listener to think after what she said. That's what I wrote.


    DK wrote: "I think you are taking a rhetorical point - what happened to the difference - and imbuing it with a lot more technical significance than it ever came with in the testimony."

    No, my crucial point, the thing I have tried to make clear so many times comes before the $14 point. It was that she makes an invalid inference by assuming that economy-wide average productivity increases have been "the same" in minimum wage jobs. Only on the basis of that can she even come up wioth the $22 number (that in turn, as mentioned above, creates a misleading impression in the ears of the unsuspecting listener) and after that the $14 difference.

    DK wrote: "I think she posed the "what happened to the other 14" to the economist, not the restaurant owner."

    Yes, you're correct about this. And I was wrong and careless in saying that.


    See Daniel, that's how somebody can admit that they were wrong about something! That's how it's done.

    She did use the existence of a difference between the current minimum wage and a hypothetical higher minimum wage though when she challenged the employer, and her case for that higher hypothetical minimum wage was at least in the argument as she presented it here based on the $22 figure.

    DK wrote: "That was a softball sort of point about wage stagnation. The real beauty of the testimony was where she took Dube's research and really took apart the restaurant owner's claims with it."

    You and I have different ideas about beauty.
    Also in this case because there seems to be a considerable power differential between the senator and the restaurant owner in their respective roles in this setting.


    "I think you are investing more significance in this 14 line than it really played in the exposition of the argument in the testimony."

    Well, I hope I clarified this above by making clear that the crucial point is that she relied on an unwarranted assumption (about average economy-wiode and minimum wage jobs' respective productivity increases) in making her inference and so in arriving at the $22 (that causes the misleading impression in the ears of the listener about what minimum wage would be just and/or unproblematic) to begin with.

    ReplyDelete
    Replies
    1. Don't get angry about it. You and I listened to the exact same testimony. I'm not sure what quoting it does if we disagree in our read of the same quote.

      Look, I've noticed a tendency for our conversations to go in circles. Not sure why. It's frustrating to me too.

      re: "Yes, if I am right in what she said then yes there would not be anything absurd about proposing a $22 minimum wage."

      And since nobody seems to think it's not-absurd shouldn't it follow that you probably aren't right? Have you considered that you might not be?

      re: "It was that she makes an invalid inference by assuming that economy-wide average productivity increases have been "the same" in minimum wage jobs."

      But she never assumed this. You read this into what she said. She never ever said this. If she said this I'd agree it'd be wrong.

      re: "See Daniel, that's how somebody can admit that they were wrong about something! That's how it's done."

      Oh quit being a jackass.

      re: "She did use the existence of a difference between the current minimum wage and a hypothetical higher minimum wage though when she challenged the employer, and her case for that higher hypothetical minimum wage was at least in the argument as she presented it here based on the $22 figure."

      Right, that's the point of citing productivity in the first place! To do this exercise! It's not exactly the marginal productivity figure we'd want, but you use it to illustrate that an (admittedly crude) measure of productivity is outpacing wage growth.

      That's my impression. You seem convinced she is using it far more literally. But just quoting her doesn't demonstrate that because my impression was formed from the exact same transcript!

      re: "Also in this case because there seems to be a considerable power differential between the senator and the restaurant owner in their respective roles in this setting."

      I agree, and the way politicians can bully witnesses makes me cringe. I look at it this way, the restaurant owner was trying to play economist, which was why it was nice that she brought the economist in to provide a counter-argument. It's one thing to browbeat a witness when they're providing good testimony. It's another thing to counter them when speaking outside of what they can really speak to.

      re: "Well, I hope I clarified this above by making clear that the crucial point is that she relied on an unwarranted assumption"

      Well it was never a mystery that you thought she was relying on that assumption, Narrator. You've said it many times now. I wasn't and still am not convinced that she really was - that's the whole problem!

      Delete
  5. "Don't get angry about it. You and I listened to the exact same testimony. I'm not sure what quoting it does if we disagree in our read of the same quote."

    I know. It's like, this is where explanation comes to an end. This is where we hit rock bottom, where things should be immediately evident and they are, but not for you, somehow.

    "Look, I've noticed a tendency for our conversations to go in circles. Not sure why. It's frustrating to me too."

    I know why.

    "And since nobody seems to think it's not-absurd shouldn't it follow that you probably aren't right?"
    No, what follows is that Warren is trying to have it both ways: she is creating an impression in the progressive listener's mind (the people she wants to appeal to) that the comparison with 1968 and the productivity increase since then show that the current minimum wage is way too low and may need to be closer to $22, while at the same time technically leaving herself room - if ever confronted about it - to deny that she said this or implied this or that this was what she had in mind.

    "Have you considered that you might not be?"
    At first this was hard to do, since it seemed so very obvious that this is what Warren was doing and I had to really strain myself to consider ways in which she could be interpreted so that she's not being manipulative etc. So I carefully listened to what she said, re-constructed the argument, read your posts, re-listened and re-read and re-analyzed, but I simply cannot understand how one can reasonably interpret what she is doing in the way you suggest.

    "I look at it this way, the restaurant owner was trying to play economist, which was why it was nice that she brought the economist in to provide a counter-argument."
    1. But the restaurant owner didnt have the luxury that she had, of bringing an economist to back him up
    2. I think it's relatively easy for somebody who is experienced at this sort of thing to make their inexperienced opponent come off badly.
    3. Even if I grant you these points it's still hardly a thing of *beauty* what Warren did with the restaurant owner.

    "Oh quit being a jackass. "
    Easy, Narrator, easy...

    "Right, that's the point of citing productivity in the first place! To do this exercise! It's not exactly the marginal productivity figure we'd want, but you use it to illustrate that an (admittedly crude) measure of productivity is outpacing wage growth.

    That's my impression. You seem convinced she is using it far more literally. But just quoting her doesn't demonstrate that because my impression was formed from the exact same transcript!"
    I know, but how exactly that formation process takes place in you is another matter.


    "But she never assumed this. You read this into what she said. She never ever said this. If she said this I'd agree it'd be wrong."
    Just listen to what she says. It cannot be reasonably interprwted or even heard in a different way.

    "Well it was never a mystery that you thought she was relying on that assumption, Narrator. You've said it many times now. I wasn't and still am not convinced that she really was - that's the whole problem!"
    I know. And I for one give up.

    ReplyDelete
  6. Tuгn on a floureԁ ѕurface and
    knead for 10 mіnutеs, adding more flour as needed.
    Layeг bottom ωith as much or as little pizza sauсe aѕ you liκe.

    Foг ѕevеral weеks I had bеen seeing this commeгcial οn teleνіsion telling about a new pizza and since I
    love ρizza I ԁecided to ρuгchase one on mу next shopping tгip.


    Loоk at my blog post ... friendfeed.com

    ReplyDelete
  7. Question: Does Murphy's fable work if the money is created, not by the gov't, but by private banks? After all, most of our money is bank money, anyway. Thanks.

    ReplyDelete

All anonymous comments will be deleted. Consistent pseudonyms are fine.