I calculate a series of capital stock per employee -- you can only do it with annual data. But the change in this series lagged one year is a fantastic determinate of productivity growth in regression models.
this series expanded steadily and rapidly from 1950 to 1980. But in the 1980s it stagnated. It rebounded in the 1990s capital spending boom but since 2000 it has stagnated again.
But this strongly supports the argument that productivity growth will continue to slow.
Technical detail. Over the long run the quarterly growth in hours worked reported in the employment data correlated very closely with the hours worked data in the productivity report. But this cycle the employment data has shown much stronger growth then the productivity data in hours worked. this quarter the employment data showed hours worked up at a 1.25% rate while the productivity data has nonfarm business hours falling 0.2%.
that is why the report was such a surprise to the market.
the employment data is for production employees and the productivity data is for all employees. What the data implies is that we are now seeing much slower growth in nonproduction or salaried workers. I see two causes for this. One for large corporations we are seeing a flattening of the management structure. How does this relate to outsourcing-- maybe it is because firms now outsource many professional services, like accounting, etc., that they use to do inhouse. Second, in recent years we are seeing small firms growing faster then larger forms -- but small firms also have a smaller management structure and overall pay lower salaries.
But both os these developments may show how the middle class squeeze is impacting the productivity data.
Thoughts?
Unit labor costs in manufacturing rose 2.7% in the first quarter compared to the headline number of 0.6%.
the more I look at this data the stranger it becomes.
I have been talking to staffing companies who are always optimistic, but are currently swearing that the job market is very hot. I also talk to people who are looking for work and they tell me there are jobs, but that they don't even think about suggesting a pay hike.
Unemployment rate low, job growth meager, GDP slowing, yet jobless claims the last couple weeks are down..
Spencer, I'm with you....
Spencer: good points. I wonder as well to what degree the discrepancy between the employment data and the productivity data may be a function of the organizations that the non-farm business sector excludes, like non-profits and government. If much of the growth in hours is due to those latter sectors, they wouldn't show up in today's productivity figures.
Financial enterprises are explicitly excluded, but does this mean that any and all financial consideration that a company undertakes is excluded from its calculation of its product/services?
With its growing portion of our economy, doesn't this exclusion, no matter how it is accomplished, distort and underestimate the actual products and services of thie economy?
Last bothersome thing: labor is not management and yet with the increasing displacement of labor from capital/machinery in the product (maybe not so much in services) and increasing amount of untallied hours from management assembling this capital, don't we have a built-in positive productivity bias? Would this bias show up more in times of crummy business investment when management is doing nothing other than buying its stock back?
Commenting by HaloScan
