Extras: David Henderson on Robert Solow

David Henderson shares his insights about Nobel Laureate Robert Solow (August 23, 1924 – December 21, 2023) the man, the MIT professor, and author of the model of economic growth that bears his name. In his conversation with Juliette Selgrin Henderson critiques Solow the man and explores some limitations of the Solow Growth model and how the study of long-run economic growth continues to evolve. 

You can listen to the podcast here: David Henderson on Robert Solow

I read about Robert Solow and grew to admire this dedicated academic who was born and raised in a middle income family in Brooklyn, NY. He came of age during the Great Depression and was guided by a devoted English teacher to apply for scholarships to Harvard University. As the scholarship ran out he worked two jobs and kept terrific grades as he studied economics. The route to MIT was direct and Henderson points out that his model and claim to fame happened mid-career followed by 40 remaining uneventful academic years.  

Henderson views Robert Solow as a great thinker and beautiful economic writer, though not a great economist. He also shares stories of Solow engaging in unflattering behavior through some interactions with his peers such as Milton Friedman and Robert Lucas. Who would say, “Milton Friedmans are bad for the economy and for society?”

As a teacher of economics I was especially interested to learn more about Robert Solow and his model. I have taught the Solow Growth Model for years as a straightforward way to understand what drives long run economic growth. Labor, technology, and capital (which is subject to depreciation and diminishing returns) are the factors involved with growth. The Cowen/Tabarrok Principles of Economics textbook and affiliated digital MRUniversity explains the model that I will summarize briefly.  

Y represents total GDP, or the output of a country, and F as a function of the three inputs. A is for the technical knowledge that increases productivity, K is for physical capital and eL represents education times labor.  If it can be assumed that A and E are constant, then the focus can be on K, and the formula can be simplified to F (K). This explains the marginal product of capital, or the additional output that results from each new tractor added to a new farming enterprise. The first tractor is enormously productive in breaking the land, the second tractor less so, the fourth, and fifth, even less. More capital increases output, but at a diminishing rate. Increasing investment is needed to grow the capital stock, while also accounting for the wearing out of these tractors over time. New capital is more valuable than old capital.

Substantial investment and the marginal product of capital explain the tremendously high growth rates of post war Germany and Japan as new capital created significant “catch-up” growth. Least developed countries can also have high rates of growth as compared to the more stable low single digit growth of wealthier, capital-intensive economies. 

In a marvelous  2016 interview with the MIT InfiniteHistoryProject, Solow describes how the Harrod-Domar growth model helped him formulate his idea. “Economic history was indeed a record of fluctuations as well as of growth, but most business cycles seemed to be self-limiting. Sustained, though disturbed, growth was not a rarity.” Solow’s model suggested the tendency for growth, yet the slowing of growth over time.

On occasion, I have seen the name Swan added to the model. Henderson explains that the Solow Swan Model is a more accurate name, as Australian economist Trevor Swan came up with it at the same time. However, a great fault in the model was discovered. Henderson reads from Jeff Hummel, another economist’s paper, that a really poor assumption on Robert Solow, and apparently also on Trevor Swan’s part, was about the homogeneous nature of capital and its tendency to wear out in an identical way. Henderson points out that capital is “heterogeneous as hell” and that labor also has variances in productivity and that the model is misleading and still misleads people to this day. Henderson would like us to forget about it! If not this model, then which model should we embrace and teach?

Feeling guilty about my role in teaching the faulty model, I listened eagerly for an update. Here is where Juliette Sellgren and David Henderson discuss the potential strength of no model at all.  After all, Adam Smith’s long-lived theory of division of labor is described through the elaborate pin factory example through his keen observation, without any numbers. 

I’m left wondering: 

Want to learn more?
Robert Solow on Growth and the State of Economics, an EconTalk podcast.
Robert Solow, Growth Theory and After, Solow's Nobel prize address.
David Henderson on Economists' Nobels, Obituaries, and More, a Great Antidote podcast.
David Henderson on Disagreeable Economists, an EconTalk podcast.
Browse Henderson's entire collection of EconLog posts.