Engaging today's political economy
with truth and reason

sponsored by

Does capital capture all the productivity gains of our modern economy, with labor getting nothing? A short case study in how to deal with competing claims

22 Apr 2017

Thanks to our frequent critic Mr. Adams, we have the opportunity to think about how we deal with “facts” that go against everything we know to be true.  When the latest scientific study comes out claiming that global warming is happening (despite no warming in the last 20 years), or that the minimum wage actually doesn’t cause any negative effects, or almost any other competing claim (especially in the social sciences), how do we assess it?  We don’t want to suffer from confirmation bias–which everyone of us has to some degree–where we only look at evidence that supports our prior beliefs.

Mr. Adams’ forte is to ignore the main points of any post, and seek to identify any perceived weakness in the logic, or less than tight argument in any particular sentence, to criticize your Bereans, and especially our employer.  He is nevertheless welcome to post, because despite his animus towards us, his attack is almost always based on positions that many others on the progressive left hold, such as Paul Krugman here.   We can then examine those arguments and understand some of the fallacious thinking that runs deep in our culture.  So this response is not just to more deeply respond to Mr. Adams (I could have continued on that thread), but to illustrate how I think we should approach technical disagreements more generally.  So earlier this week I posted an agreement with Paul Krugman (surprisingly), and in the course of the thread Mr. Adams challenged my views on rewards to labor.  Here is the pertinent part which we’ll delve more deeply into:

“Entry-level positions are therefore paid according to what they produce.” (Haymond)

Nonsense. There used to be something of a connection between worker productivity and worker wages. Like 8-track tapes and maroon bell bottoms, that connection is long gone. Increased productivity flows upward. As labor unions have declined in number and influence, due to right-to-work laws, workers have not been able to stem the tide. (Adams)

In the subsequent back and forth, Mr. Adams triumphantly produced his proof–a government Bureau of Labor Statistics report which does address the issue, while denigrating my source, since it came from YouTube (despite the fact that the narrator is a highly respected economist and the chair of the George Mason University economics department).  Now at the superficial level he seems to have a point–an obviously (?) impartial government video vs.  some YouTube video.  In his words,

I use BLS data. You use Youtube! Case closed.

We shall see.  The arguments can get wonkish, and a comprehensive technical summary can be found from Heritage here, but we’ll review the basic response below.  The BLS report is well done, in the sense that its conclusions logically follow from its presuppositions and initial assumptions.  But its methodology is not appropriate for its conclusion that:

Since the 1970s, growth in inflation-adjusted, or real, hourly compensation has lagged behind labor productivity growth.

Note that even this claim is nothing like the hyperbole (I’m being generous) of our critic Mr. Adams: to say that compensation has lagged productivity (BLS) is quite a bit different than saying there is no longer a connection between the two (Adams).  But we’ll leave that aside. The BLS report methodology is inappropriate for the following main reasons (there are more–see link above to the Heritage summary):

  1. Using conventional but inappropriate inflation index (the Consumer Price Index).   The claim I (and standard economic theory) make is that labor will tend to be paid according to its marginal revenue product (roughly the dollar value of its productivity).  To compare productivity to compensation, you must use the same inflation index, and as Harvard professor Martin Feldstein* has argued, the index should be based on the firm’s output, since the theory’s claim is that workers get a share of the firm’s output.  The theory says nothing about the ability to buy a general basket of goods.  Since general consumer inflation has grown significantly faster over the last several decades than inflation in the firm’s output, this tends to lead to a gap when the CPI is used to deflate output.  As Feldstein concludes, “basic theory reminds that real compensation should be measured using the same price index that is used to calculate productivity.  When this is done, the rise in compensation has been very similar to the rise in productivity.”  Importantly, Mr. Adams own source shows that much of the issue is removed when using comparable price indices; see the 2nd bullet point on page 61 and 62 of the BLS report.  Both Dean Baker and Feldstein dispute the BLS position that labor’s share has declined appreciably.
  2. Comparing compensation to productivity must consider what left-of-center economist Dean Baker calls “usable” productivity, i.e., productivity net of depreciation.  As the economy has undergone structural changes over the last few decades, with more investment going toward short-lived computers and software, additional depreciation is necessary.  As Baker says, neither labor nor capital “can eat depreciation.”  When this is corrected for, Baker says, “the missing wage growth is considerably less of a mystery.”
  3. The BLS does not address the significant structural change in the economy in terms of the average worker.  We have had a significant increase in immigration, women have continued entering the workforce in greater numbers, etc.  This is the key point Prof Boudreaux made–even if the average has stayed stagnant, every one in the actual workforce could have experienced significant gains, if the new entrants to the workforce are starting from a relatively lower position (which is exactly what you’d expect from the demographic changes we’ve seen).

Now I doubt Mr. Adams (or Professor Krugman, or Professor Stiglitz or Senator Warren) will necessarily buy these changes.  But there is a reason why marginal productivity theory of wages is still the basis of what we teach in economics.   Which is why we need to go further whenever we have competing claims to the empirical answer. Notice what I challenged Mr. Adams on previously (which of course he ignored).  Let’s assume he’s right–what would that mean?  Taking the logic of his position–that labor is relatively undervalued compared to capital (and/or management) and has been for decades–this necessarily means that markets for some reason are unable to arbitrage this price differential, and have ignored this money on the table for decades.  Yet arbitrage (buying cheap and selling dear) is the essence of the market process-this is what markets do every moment.  Yet we’re supposed to believe that all these market participants (the people most likely to see arbitrage opportunities) are leaving big money on the table, yet outside observers in academia (with no direct role in markets) see these opportunities clearly.  Since clearly Mr. Adams, Sen Warren and others have seen this profit potential, I cheer them on to commit their own capital to restore equilibrium.  Their failure to do so, on top of the “failure” of market participants to see as clearly as the academics, speaks volumes as the reality of their vision.  This is the same logic as should be applied in minimum wage discussion, comparable worth for women discussion, etc.   If somehow markets are failing, that just means there is a profit opportunity for someone else to exploit.  The fact that neither knowledgable market participants or self-professed experts in academia are willing to commit their own money to profit from this opportunity is highly suggestive that there actually isn’t any price differential to correct.

We shouldn’t be afraid to do the thought experiment of what the world would look like if our intellectual opponents were right.  If the necessary implications of accepting their position are completely implausible (as I argue in this case), while the implications of your own position are highly plausible, then you should feel more confident in believing that your source for data is more reliable.  Its pretty clear to me who is suffering from confirmation bias in this case.

Yes, its great to be a Berean!

* Martin Feldstein is the President of the National Bureau of Economic Research (the organization that calls official recessions). He is a HIGHLY credible source.