Minimum Wages: A Puzzle


This, presumably, is a pro-minimum wage cartoon. This, clearly, is an anti-minimum  wage column. But putting our economist’s hat on for a second — aren’t they making the same argument?

This has bugged me about the minimum wage debates for a while. Supporters say, mandating higher wages will lead to more motivated employees, less turnover and training costs, less need for supervision. Opponents say, mandating higher wages will lead to mechanization of jobs currently done by low-wage labor, substitution of capital for labor. But at an abstract level, it seems to me these are just two ways of describing the same thing. Aren’t they both just saying, faster growth of labor productivity? How can the same predicted outcome be an argument both for and against higher wages?

The anti- side seems to be implicitly assuming a fixed quantity of output, so that less labor per unit means less labor overall. The pro- side seems to be implicitly assuming some combination of a flat demand curve and a flattish supply curve, so that a fall in costs leads to a large rise in output. But you’ll still get some fall in employment, in the absence of monopsony or demand effects.

Anyway my interest here is not in what is the right answer. [1] My interest is that supporters of a higher minimum wage invoke efficiency wage arguments without explaining why that should be expected to dampen disemployment effects rather than amplify them. While opponents of higher minimum wages describe faster productivity growth as a cost, without, as far as I can tell, being against it in any other context. This is a case where abstraction — some equations, even one of those awful supply and demand graphs — might actually be helpful.


[1] Just to be clear, I support a higher minimum wage, $15, whatever. A few years ago, I even testified to that effect before the Maryland state legislature.






8 thoughts on “Minimum Wages: A Puzzle”

  1. Perhaps what we’re experiencing in this controversy is the Blind Man being able to simultaneously touch two, rather than just one part of your Abstract Elephant? Perhaps the controversy isn’t really about economics at all.

  2. “How can the same predicted outcome be an argument both for and against higher wages?”

    In my opinion, it is because the argument in itself is nonsensical, so that you can get opposite conclusions from it.

    Suppose that an economy has a yearly income of 100, and a capital allotment of 200. Suppose also that higer minum wages influence the wage share (and not only nominal wages, otherwise it is obviout that the result is just a wash).
    The capital intensity of a business is usually calculated as capital/wages, so that if the wage share in the said economy is 50%, the apparent capital intensity is 200/50 = 4, if the rise in the minimum wage causes an increase in the wage share to 70% the new apparent capital intensity is 200/70 = 3, so the apparent capital intensity actually falls if the wage share increases.
    The true capital intensity, that is capital vs income (and not capital vs wages) stays the same, just with a wage share of 50% the capital value of 200 will represent 100 of cost for the capital producers and 100 of profits for them, with a wage share of 70 it will represent 140 of cost for the capital producers and 60 of profit for them.
    There is no reason to think that income distribution affects the true capital intensity (capital vs. total income), but if anything, since by definition it is capitalists who buy capital goods, and they buy capital goods from the profit share and not from the wage share, even assuming that distribution somehow influences capital intensity a low wage share would cause an higher capital to income ratio, not the opposite (this is what comes from Piketty’s data, although I think that much of Piketty’s capital is just wealth and not real capital, but this is really OT).
    Basically the idea that there is a “supply-side optimal wage share” is based on nothing, so whomever can assume that we are currently above this flogiston wage share or below it or exactly at it and, since we can’t pinpoint it, everything can be true.

    I think that there might be a demand-side argument for a certain wage share, or an import-export argument for the same (and the two might go in opposite directions), but not a supply side argument.

    The anti-rise article makes an argument about different minimum wages for different places, which might make sense if average productivity in different places is so different that different minimum wages cause wildly different wage shares in different places, which would then become an import-export argument against the rise for the poorest places, but it seems to me that this argument is very stretched.

  3. 1) A nitpick, but I would say the Doonesbury cartoon is describing a one-time, permanent step increase in productivity, not faster productivity growth.

    2) There are two ways to look at automation, depending on circumstance. One is devices that provide small efficiency improvements to marginally increase worker productivity and over time reduce the need for companies to hire additional staff (but don’t provoke immediate layoffs). The other is machines that can directly replace workers prompting immediate layoffs. That would be like smart electricity meters replacing meter readers, or Amazon robots replacing warehouse pickers, or Google cars replacing cabbies. In the long term and in the abstract they have pretty much the same effect but in the short term the latter is much worse. The latter provokes a much more negative emotional response.

    3) In some cases increasing human productivity (not through automation but through morale, rest, skills, etc.; or through partial automation instead of near-total automation) will tip the scales toward humans when firms decide whether to use humans or robots, and thus actually preserve/increase the human employment.

    4) My own critique would be that some of the Doonesbury positive effects from the wage increase–in particular the increased retention–seem to mostly be about what happens when one firm increases wages while its competitors do not. With a minimum wage increase the competitors would also increase wages so those effects would not occur.

    However there are some very plausible positive effects from higher low-end wages that could increase productivity at all firms: with higher wages workers would have a much easier time handling their personal expenses and thus be much less stressed out and mentally fatigued (poverty is very stressful); with higher wages–and, critically, less of a gap between them and wealthier people–workers would be happier and more satisfied and have higher morale (as Doonesbury says); with higher wages they might be able to stop working a second job, or be able to hire a babysitter, which would drastically decrease their fatigue and stress.

    5) Speaking of which, if some workers quit their second jobs after a wage increase, then that could offset a hiring decrease resulting from productivity increases.

    6) You have set demand effects aside, but Trudeau says, and I think, and I expect you also think, that a higher minimum wage would increase consumer demand.

    1. 1. Sure. Sloppy language on my part.

      2. I’m not sure. The question in my mind is whether it’s reasonable to lump improved worker motivation/retention/etc., on the one hand, and mechanization, on the other, as just two forms of higher labor productivity. I’m still struggling with this, but I don’t think the factors you suggest are what make the distinction, if there is one.

      3. Well, I don’t know. Someone is still making the robots. Presumably robot-making is a higher productivity activity than burger flipping, but in the aggregate that just brings us back to the original question — under what circumstances will an increase in labor productivity amplify the disemployment effects of a higher wage, and under what circumstances will it dampen them? Only in this case the rise in labor productivity is about shifts between sectors rather than within an individual firm.

      6. You have to be talking about a pretty broad wage increase before demand matters. For a minimum wage, let alone the wage choices of an individual firm, it seems pretty unlikely that it could be important.

  4. I am sorry – I meant may need to right click and save as ‘file.pdf” with the proper PDF extension before reading for the last linked file.

  5. I wonder, though, “what if” by mid century virtually nobody will need to work?

    I am just saying that this is happening, you don’t have to believe me. As most Americans likely have never worked in fields where the state of the art changes radically from one year to the next, they don’t have the same context.

    Historically, among my circle of friends and colleagues its not been something we debate much if ever. Its a fact. People are solving problems that stand in the way and more and more work is being automated. We need to shift away from a society that values people solely by their paid work, because within our lifetimes, most people won’t work. The longer we put off having a realistic viewpoint of a future without work, the more unprepared we will become for it. My suggestion is that we try to transition away from a ‘money is everything’ point of view, in some formal manner. That requires rejecting the pending trade deals which basically represent a global coup with the objective of declaring that money is everything. The antidote to that is its opposite.

    The EU TiSA “mandate document” I posted in another thread contains a sort of coded outline of TiSA’s goals and it is in some ways an outline of the pact which attempts to force a radical globalization of service jobs, with one goal clearly being lowering wages, not something I think will be positive. They are attempting to preserve what I consider to be bad, predatory business models in the face of their increasing irrrelevance, and make impossible the best business models, such as single payer healthcare (which is forbidden by multiple means in both GATS and TiSA), and worker owned businesses, that work. The deals attempt to proscribe all adjustments which from the people’s viewpoint could improve stability and prosperity and allow us to adjust to the changes – such as rapid improvements in productivity, because their benefits would be shared and the risks made less destructive. Instead it seems very clear to me that they are trying to lock in bad policy – carve it in stone with the secret agreements, and force the world into a race to the bottom. That is especially what ISDS, giving corporations property rights in bad policy, does. ISDS, indirect expropriation and so on try to do that, taking away the right to regulate.

    1. Oak-

      I’m glad to have you reading an commenting here. But I had to delete another of your comments that was just a very long cut-and-paste from a document posted elsewhere.

      1. I’m very sorry about that I wont do that again.

        Could I ask your readers to read an East African document which is kind of relevant to this issue – minimum wages in the US and it kind of speaks for itself- however its another slightly problematic PDF on my older machine.. the document is entitled:

        Trade in Services Agreement (TiSA)

        An Information Note on Current Scenario and Scope for Future

        Prepared by
        Tanvi Sinha, on behalf of
        CUTS International, Geneva
        July, 2014

        There are also some relevant bits of information here:

        Basically I am afraid that since tiSA is going to be merged with GATS, the WTO will gain jurisdiction and so something like what they did to El Salvador in the 1990s, I know that many politicians privately, even Dems, want wages to fall not rise.

        Basically, TiSA is an continuation and hoped for legitimization of GATS which has been 20 years in the makings, its an attack on skilled workers wages.. Like GATS it attempts to completely block, forever, single payer health care and attempts to do the same to public education and it effectuates a global trading of jobs for markets. It changes procurement to make it so domestic spending will be unlikely to lead to more jobs for American firms than staffing firms from other countries, if we want those jobs wages will have to fall a lot- Jobs currently occupied by people who have no idea this is coming. The media has been virtually completely silent for 20 years, on this.

Comments are closed.