Where is Labor's Share - November 20, 2018
The Good, the Bad, and the Ugly
Last week, Chairman Powell was in Houston meeting with community leaders, touring neighborhoods affected by Hurricane Harvey, and spreading the good word of the Fed. According to a New York Times article covering his visit, much of the dialogue was around or related to inequality, with one goal of his meetings “to demonstrate the Fed’s concern about those who have been left behind by a decade of economic growth”. As we have written about before, this is nothing new to the Fed and something both the Cleveland and Minneapolis have studied. Some new light may have been shed regarding Labor Share by a recent article from the NBER. In “The Micro-Level Anatomy of the Labor Share Decline”, Matthias Kehrig and Nicolas Vincent uncover dynamics that could be considered good, bad, and ugly for those trying to improve labor’s lot.
Focusing on manufacturing, their research finds that although the labor share in manufacturing has declined “by almost 5 percentage points (ppt) per decade between 1967 and 2012”, the median establishment saw an increase in its labor share. Though the increase is a relatively small 0.7% per decade, it is fairly uniform and “this upward trend is present for the vast majority of manufacturing establishments”.
The decline in labor share is “entirely driven by a strong reallocation of value added to establishments with low labor shares”. This dynamic is shown in the left-ward shift in the graph at right. And, despite what standard economic theory might posit, the hyper-productive establishments to the far-left side of the graph (at right) do not pay higher wages.
Perhaps one of the ugliest findings in the paper is that there’s a disconnect between value added and labor input (contrary to models saying that highly productive establishments will expand their workforce and or increase wages). Kehrig and Vincent find asymmetric hiring responses to Total Factor Productivity, where positive shocks leads to weak responses in employment, whereas “negative shocks, in contrast, lead to significant firing throughout the sample.” So, “Why don’t establishments that become highly productive hire more?” Quite simply optimism, “the establishment might be reluctant to act on a positive realization of TFPR that it expects to be temporary”. While this is superficially a sign that employers don’t have much faith in new technology, it could also be an additional nail in the coffin for workers automated out of jobs.
P.S. While we discussed some of the reasons to be optimistic about housing two weeks ago, the data released today, Housing Starts and Housing Permits, now join Existing Home Sales, New Home Sales, and Mortgage Applications in the camp of declining YoY!