If there are only two factors, they have to be complements. If there’s more capital, the wage has to rise. Now imagine that…you can take some of the stock of machines and, by designing them appropriately, you can have them do exactly what labor did before…When capital is reallocated to substituting for [replacing] labor, the stock of effective labor rises and the stock of conventional capital falls, and so wage rates fall. Third, the capital share, understood to include the total return to capital of both varieties, rises. That’s just a corollary of output rising and wages falling. This pattern is similar to what we have seen take place. I suspect that this reflects the nature of the technical changes that we have seen: increasingly they take the form of capital that effectively substitutes for [replaces] labor.
Pointer from Tyler Cowen, who I’m surprised did not make a bigger deal about it.
My one quibble/criticism is that this describes a closed economy. In the real world, with China and India developing, factor-price equalization is at least as important as factor substitution. To put this another way, include those countries when you calculate trends in labor’s share of income.
Also, Summers writes,
Where production has taken place in the classic way we teach, productivity growth has continued. There has been progress. Real wages measured in those terms have increased substantially. It’s just that a larger and larger share of our economy is in sectors that are not well thought of as widgets produced by competitive firms. They are sectors where property rights, scarcities, intellectual property, and the like are of fundamental importance.
My take on this is to be wary of talking about “the” real wage. Your real wage is much higher if you abstain from making extravagant use of modern medicine and private colleges. See The Reality of the ‘real wage’.
He concludes by raising the issue that Nick Schulz and I called the New Commanding Heights. Summers writes,
Whether the expansion of those sectors as a share of the economy necessitates a growing share of the public sector in the economy, or whether the share of healthcare and education that takes place in the public sector should decline will be a matter of great public debate. As a country, and not without controversy, we do not seem to be moving toward a smaller public role in healthcare. Nor do other countries in the world. But that will, perhaps, change over time.
Summers: “When capital is reallocated to substituting for [replacing] labor, the stock of effective labor rises and the stock of conventional capital falls, and so wage rates fall.”
Is this supposed to be obvious? Not to me.
Almost all capital replaces labor. Sometimes, the only practical way to do something is to use a machine, say by using hammers (simple machines) to drive nails, to build houses. Workmen could use rocks instead of hammers, but no quantity of workmen could do the job needed. In this sense, hammers substitute for a large amount of ineffective labor.
Nail guns then substitute for much of the prior labor needed to use manual hammers. Fewer workmen are needed to build the house. Is this substitution of capital for labor a bad development? Not for house buyers.
Is it bad for the workmen? That is not predictable from first principles. Maybe the remaining workmen are paid more, and maybe they are paid less. Maybe some are paid more, and some less.
Does the number of such workmen decrease? Again, not predictable. Maybe sales of the now cheaper houses go up enough to employ even more workmen than before. So, the use of labor saving nail guns reduces the need for labor, but the expanding sales of houses may increase the need for labor.
How can a supposedly smart person make sweeping assumptions about capital and labor, then reach a conclusion that is opposite to the actual experience of our society. From the start of the industrial revolution, machines have substituted for labor and made everyone much wealthier, even those people who cannot manage to swing a hammer.
EasyOpinions blogspot com
This is a good point. I think you in fact have to make it a bit stronger. ALL capital replaces labor when that capital is employed. Isn’t this definitional to productivity gains by labor through the use of capital?
Summers’ is indicative of a fallacious line of economic reasoning that seems to be on the rise. Specifically, failure to think past the immediate, temporary, but technically true response to see further to the dynamic, long-term, “unseen” response. A person creates a tool, which *saves* labor, as the fruits of production increase for the ultimate consumers. If that sentence reads “replaces” instead of “saves”, we lose a crucial part of that story. A laborer lost some or part of his job is not the economically interesting part of the story. This is true whether or not the saved labor is in the traditional production of widgets by competitive firms. At what point will most people realize that manufacturing is this century’s version of the last two centuries’ agriculture?
One of the absolutely key yields from saved labor is using that freed resource to come up with ideas of new and better ways to employ SAVED labor.
The ‘modified production function’ was the key part. ‘Pure Capital’ can be imagined as being able to escape diminishing returns, create a kind of ‘specialized division of labor’ within itself, and so reallocate itself into ‘labor-like’ and ‘productivity-enhancing-type’ forms of capital.
The question is, “What Would The CEO Do?”
In the traditional production function, he faces the familiar productive-factor-capital vs. labor trade-off, compares capital costs with wages, and finds the optimal point on the curve. The trade-off changes over time with technological innovation, but not *too* quickly.
But now the CEO expects a dramatic acceleration in Summers’ lambda factor. Or at least opportunities to develop such high-lambda capabilities.
So he is tempted to divert current capital flows to invest in these future high-lambda capabilities and find a new optimal production-factor allocation. To do this, he has to try and estimate expected demand, expected wages, the expected regulatory climate (self-driving trucks, trains, and planes?) and *current* interest rates.
Current corporate debt interest rates (and spreads) are extremely low, which makes capital and investment seem cheap. Technology is getting cheaper in real terms, whereas total-compensation for employees is stagnating or getting more expensive. There seem to be some high-risk, high-reward automation opportunities.
If I were a CEO facing this kind of future, where the pace of lambda-acceleration is accelerating but a highly uncertain pace, I would very, very reluctant to enter into any ‘sticky’ long-term relationships with new employees. I would want maximum flexibility in the event of high levels of disruption in my industry.
My prediction is that companies that fail to go ‘all in’ on automation as quickly as possible and hang on to their labor force will be quickly overtaken by entrepreneurs building new companies based on an all-automation vision from the start, so that they never have to deal with major labor frictions.
The future is exhilarating and frightening!
The old joke applies: There is a factory that employs one worker and a dog. The worker’s job is to feed the dog. The dog’s job is to keep the worker from messing with the robots. Ha. Ha.
What should the labor share of the factory’s output be?
On the global view point, the cost of living is much lower in developing countries. Thus workers in two countries whose productivity is the same may be in vastly different situations. One can survive earning her marginal product, while the other is effectively unemployable. The day when there are no more people to convert from subsistence agriculture to modern services or manufacturing will level the global playing field, but there is an effectively unlimited supply of robots.
We should improve her skills and remove every block to creating more jobs, but this problem is bigger and longer term than that. Ask yourself, can we continually improve her or her children’s productivity faster than that of her robotic competitors?