Stansbury and Summers on the declining bargaining power of labor
In one of the best papers of the year, Anna Stansbury and Larry Summers present what is to me the best non-“Great Stagnation” story of what has gone wrong, and I have read many such accounts. Here is their abstract:
Rising profitability and market valuations of US businesses, sluggish wage growth and a declining labor share of income, and reduced unemployment and inflation, have defined the macroeconomic environment of the last generation. This paper offers a unified explanation for these phenomena based on reduced worker power. Using individual, industry, and state-level data, we demonstrate that measures of reduced worker power are associated with lower wage levels, higher profit shares, and reductions in measures of the NAIRU. We argue that the declining worker power hypothesis is more compelling as an explanation for observed changes than increases in firms’ market power, both because it can simultaneously explain a falling labor share and a reduced NAIRU, and because it is more directly supported by the data.
There is a good deal of critical thinking about how different macroeconomic trends fit together, and a willingness to consider disconfirming evidence, so I do recommend you read through this one.
I have five main worries about the argument:
1. Rather than labor losing bargaining power, I think of the key development as “management measuring the marginal product of labor more precisely.” Admittedly that does lower the bargaining power of the majority of workers, given the 20/80 rule, or whatever you think the proper proportions are (Stansbury and Summers themselves presumably are underpaid, but in general wage dispersion has been going up in high-skilled sectors).
A minority of highly productive workers have much more bargaining power than they did before, which doesn’t quite fit the “lower bargaining power per se” hypothesis. And under my interpretation, easier unionization may not be much of a solution, since the problem here is the actual reality of who produces what. Consistent with my view, labor’s share is not really down if you consider the super-talented labor/owners/capitalists who start their own companies. That is a return to labor as well.
2. It is a noted advantage of the Stansbury and Summers approach that is explains the now-lower natural rate of unemployment. The puzzle, I think, is to explain both lower NAIRU and the slower labor market matching observed over the post-2009 labor market recovery. Their hypothesis seems to predict a higher degree of worker desperation, and thus quicker matches, than what we actually observed.
If you think, as I do, that employers are now better aware of the diversity of worker quality, and that only ex post do they learn that quality, employers will be more careful upfront, which probably does slow down matching speeds, thus fitting the data better.
3. If you play down market power, and postulate a fall in the share of labor, you might expect investment to be robust, but measured investment clocks in as mediocre. The authors discuss this point at length on pp.45-46 and offer multiple rebuttals, but I suppose I still think the first-order effect here ought to be stronger than what we (seem to) observe.
4. If corporate profits are so high, how is this consistent with the persistently low demand postulated by Summers’s “secular stagnation” hypothesis? The paper does consider this question very directly on p.56, but I genuinely (just as a matter of grammar) do not understand the answer the authors are suggesting. Here goes:
A fair question about the labor rents hypothesis regards what it says about the secular stagnation hypothesis that one of us has put forward (Summers 2013). We believe that the shift towards more corporate income,that occurs as labor rents decline,operates to raise saving and reduce demand. The impact on investment of reduced labor power seems to us ambiguous, with lower labor costs on the one hand encouraging expanded output and on the other encouraging more labor-intensive production, as discussed in Section V.So,decreases in labor power may operate to promote the reductions in demand and rising gap between private saving and investment that are defining features of secular stagnation.
I suppose I had thought of low rates of profit as a (though not the?) defining feature of secular stagnation, but again I may not have understood this passage correctly.
5. Matt Rognlie found that the decline in labor’s share went to housing and land ownership, not capital.
In any case, here is a whole paper full of economics, go and enjoy it.