Sunday, May 9, 2021

The role of wages in the 1955-57 inflation

The first result returned by my search for the creeping inflation of 1955-57 was Employment, Growth, and Price Levels: The effects of monopolistic and quasi-monopolistic practices, dated September 1959, from the Joint Economic Committee of Congress.

The text returned by the search:

But the major thesis of this paper is that the creeping inflation of 1955 – 57 is different in kind from such classical inflations , and that mild inflation may be expected in a dynamic economy whenever there occur rapid shifts in the mix of final ...

which turns out to be part of a statement by Charles L. Schultze, who we saw this past January in another publication of the Joint Economic Committee.

The inflation was "different in kind" from the classic demand-pull inflation, Schultze says. He gives me something to live for.

Schultze and Google Search open a door and leave it open for me and my thinking on the cost-push problem. And then Schultze says

Similarly there is no attempt here to prove that autonomous upward pressures of wage rates have had no impact on the price structure. Such pressures may have played a role in recent inflation.

And I suddenly wanted to show that wage rates played no role in that inflation. I don't know where this comes from, but my mind went instantly to Components of Corporate Cost, from 2010, where I show corporate compensation of employees falling as a share of corporate costs (as measured by corporate deductions) for the 1948-2007 period.

And then instantly to my list of FRED data that I usually use for labor productivity...


 ... and yes, the list has business sector compensation and business sector current dollar output. And if I look at the ratio of those two I can see the nominal cost of labor relative to the nominal price of output. And yes,

Graph #1

labor cost goes down from start to finish, so: No, wages have not been gaining on prices. And hey, that graph looks an awful lot like labor share.

Graph #2

Yes it does. Exactly like Labor Share.

The other components that make up the price of output are nonlabor cost, and profit. I found that data not long ago. I'll have to find it again.

But oh, there is a sharp down-and-up after the 1954 recession. That'll be involved in the 1955-57 inflation. Here, look at 1950-1962:

Graph #3

The plotted line drops down to a low point after 1954. That low point is First Quarter 1955, early in the year-long rise of the labor force. The line reaches its next high in second quarter 1956. Most of that increase occurs in 1956, the year after the year-long rise of labor force participation.

Checking compensation per hour:

Graph #4

The low point in the middle of the 1954 recession is 1954 Q1. The line drifts down to a low in 1955 Q4, then rises to a peak of more than 8% in 1956 Q4. So wages did go up, but not until 1956. This confirms what we saw on Graph #3.

Wages didn't go up until 1956. But prices were already going up early in 1955:

Graph #5: Three Measures of Inflation, 1950-1962


It wasn't wages that got the 1955-57 inflation going. It was the year-long increase in labor force participation that occurred in 1955. And the unusually large increase in employment in 1955 and '56 and into '57:

Graph #6

The large increase in new, unskilled workers. They came at a bargain price, but hiring them led to a fall in productivity that increased business costs and started the 1955-57 inflation.

2 comments:

The Arthurian said...

For US manufacturing, "percent change from year ago" of Total Labor Cost Per Unit of Output was still negative as of December 1955, after a whole year of employment expansion and labor force growth.

If it had gone positive, one could try to say that the hiring led to wages pushing costs up, and that it was still rising wages that caused the inflation. But the number was still negative, so that argument does not stand up.

The Arthurian said...

I have recently been looking at the lag created by the "percent change from year ago" calculation. The lag appears when the month-on-month inflation rate is coming down but the "from year ago" calculation still has the high inflation months in the year it considers.

I will have to look, but that the "percent change from year ago" data for US manufacturing (noted in my comment above) will show a similar lag.

I will have to look, and reconsider the 1955-57 inflation.