Home > Growth, Uncategorized > A Piketty Moment

A Piketty Moment

October 25, 2016 Leave a comment Go to comments

Update (10/27): Comments rightly pointed to different deflators for the two series. I added a figure to account for this (thanks!)

Via Mark Thoma I read an interesting Atalanta Fed Comment about their wage tracker, asking whether the recent pickup of wages in the US is robust or not.

The first thing that came to my mind is that we’d need a robust and sustained increase, in order to make up for lost ground, so I looked for longer time series in Fred, and here is what I got


This yet another (and hardly original) proof of the regime change that occurred in the 1970s, well documented by Piketty. Before then, US productivity (output per hour) and compensation per hour  roughly grew together. Since the 1970s, the picture is brutally different, and widely discussed by people who are orders of magnitude more competent than me.

[Part added 10/27: Following comments to the original post, I added real compensation defled with the GDP deflator.  While this does not account for purchasing power changes, it is more directly comparable with real output. Here is the result:


The commentators were right, the divergence starts somewhat later, in the early 1980s. This makes it less of a Piketty moment, while leaving the broad picture unchanged.]

Next, I tried to ask whether it is better for wage earners, in this generally gloomy picture, to be in a recession or in a boom. I computed the difference between productivity (output per hour) and wages (compensation per hour), and averaged it for NBER recession and expansion periods (subperiods are totally arbitrary. i wanted the last boom and bust to be in a single row). Here is the table:

Yearly Average Difference Between Changes in Productivity and in Wages
In Recessions In Expansions Overall % of Quarters in Recession
1947-2016 1.51% 0.40% 0.57% 15%
1947-1970 1.62% -0.14% 0.19% 19%
1970-2016 1.42% 0.67% 0.76% 13%
1980-1992 0.64% 0.84% 0.81% 17%
1993-2000 N/A 0.68% 0.68% 0%
2001-2008Q1 4.07% 1.10% 1.41% 10%
2008Q2-2016Q2 2.17% 0.12% 0.49% 18%
Source: Fred (my calculations)
Compensation: Nonfarm Business Sector, Real Compensation Per Hour
Productivity: Nonfarm Business Sector, Real Output Per Hour

No surprise, once again, and nothing that was not said before. The economy grows, wage earners gain less than others; the economy slumps, wage earners lose more than others.  As I said a while ago, regardless of the weather stones keep raining. And it rained particularly hard in the 2000s. No surprise that inequality became an issue at the outset of the crisis…

There is nevertheless a difference between recessions and expansions, as the spread with productivity growth seems larger in the former. So in some sense, the tide lifts all boats. It is just that some are lifted more than others.

Ah, of course Real Compensation Per Hour embeds all wages, including bonuses and stuff. Here is a comparison between median wage,compensation per hour, and productivity, going as far back as data allow.


I don’t think this needs any comment.

  1. Patrick
    October 26, 2016 at 2:56 pm

    Hello Francesco, Yes, the stylized facts are quite striking. But the next question is, why/how? Is it a decline in unionization?, just due to a rise in income of “the 1%” ?, a compositional effect from a change in output structure at sectoral level with its implied effects on employment…?, a change in ideology and social preferences towards more tolerance for inequality? Any personal views on this? Thanks!

  2. Ilya
    October 26, 2016 at 5:09 pm

    You’re using two different deflators for productivity and wages, the GDP deflator the for former and CPI for latter. We never use different price indexes when comparing to nominal values. Use the same deflator, like the GDP deflator , and the gap goes away.

    • Patrick VB
      October 27, 2016 at 10:36 am

      Hello Ilya, the choice of deflator depends on what phenomenon you want to highlight. Using different deflators, as done here, shows the deviation in the path of the purchasing power of labour income, relative to the path of profits. We see that profits “purchasing power” is outpacing labour “purchasing power”.

      • Ilya
        October 27, 2016 at 6:53 pm

        In principle, there is nothing wrong with what you are doing if you merely wish to show that “the productivity of the economy in general keeps growing, but labor’s purchasing power of consumer goods remains stagnant.”

        But left-wingers almost unanimously use this graph to claim either explicitly or implicitly that the typical worker is not seeing wage gains commensurate with growth in his own productivity, and therefore the marginal theory of productivity is clearly false. But this graph in no way shows this claim. Not even close. Left wingers are being dishonest in this sense (I don’t accuse you), just like right wingers who claim there has been no growth in the huge inequality or that wages keep going up up up for the typical man on the street.

        Even the gap you have left is filled if you use the correct (in my opinion) methodological choices. See…


  3. soleus
    October 26, 2016 at 11:46 pm

    The paper which drew attention to this graph also showed that, up to 2000, most of the divergence could be traced to divergence in the deflators used. The CPI was outpacing the GDP deflator. The illustrative example given was IT equipment which (a) has got relatively cheaper, (b) weighs more heavily in the GDP deflator than in the CPI.

  4. October 27, 2016 at 12:42 pm

    Hi, thanks for your comments. I suspend my “no reply to comments” rule. You are right, I added the Real Compensation line deflated with GDP deflator. Thus, one can go with either Ilya’s or Patrick’s interpretation. Thank you again

  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: