Ageing and productivity: Economists and others

Daniel S. Hamermesh, 20 February 2013

a

A

Sixty years ago, Harvey Lehman published a path-breaking book examining the lifecycle of productivity in various fields, scientific, humanistic and artistic (Lehman 1953). He demonstrated the now widely accepted conclusion that the contributions of mathematicians and people in mathematics-related disciplines peak very early in their careers. Lehman also showed that artists and humanists in many cases achieved their greatest successes much later in life. How do economists stack up along the age-productivity dimension, and how has that been changing? What does this tell us about the reasons why older people’s productivity declines?

The pattern changes

Table 1 Age distributions of authors in top US economics journals, 1963-2011

Year of
publication

Age (percentage distributions)
<=35 36-50 51-60 61+
1963 50.5 45.3 2.4 1.8
1973 61.5 32.6 5.9 0
1983 48.5 47.2 3.5 0.8
1993 49.8 43.1 5.6 1.5
2003 36.8 50.4 10.7 2.1
2011 33.0 48.1 13.0 5.9

Source: From Hamermesh (2013). The results are based on the American Economic Review, Journal of Political Economy and Quarterly Journal of Economics. The population excludes all Presidential and Nobel addresses, comments, replies and notes. The age of one author in 1963 was unavailable.

Table 1 shows the distribution by age of authors of articles published in the top three general US-based economics journals, recognised fairly widely as three of the top five journals in the field (Hamermesh 2013). Studies published in these journals are the most prestigious and have the demonstrably largest impact on scholarship throughout the economics profession. Two things are noticeable from the table:

  • First, it is clear that through the 1990s top-level economic research was very much a young person’s game – almost nobody over age 50 published a paper in these premier outlets;
  • Second, this fact has changed in the last 15 years;

Although 80% of authors are below age 50, well above the fraction of PhDs teaching at North American and European universities, the percentage of ‘old’ (ages 51 years old and above) authors has increased from 5% to nearly 20%. Why?

Explaining the change

Simple economic theory and a lot of evidence for the entire work force (Mincer 1974) make it clear that there should be, and is, an inverse-J shaped relationship between age and productivity. When you’re young, it pays to invest to improve your skills so that you can earn more over a long remaining work-life. When you’re older, why bother? Improving skills is hard work, forces you to forego current earnings, and you don’t have much time to benefit from the improvements in your skills. This holds for economists as well as anybody else – internal economic incentives to produce are important.

Outside opportunities matter too – as an economist ages and has an established reputation, he or she has more outside opportunities such as consulting or academic administration. Technology also matters – younger scholars are tooled up in work at the frontiers of knowledge, tools that are hard for older scholars to acquire. The general observation about youth being more important in more mathematical endeavours is also demonstrated in economics – Nobel laureates whose prize-winning work was more mathematical tend to have been younger when they did that work (Galenson and Weinberg 2005). Sadly, one’s physical energy, one’s “stick-to-it-ness” and one’s ability to concentrate also diminish with age and reduce one’s ability to do the hard work required to innovate.

I have seen no research that distinguishes the relative importance of these four explanations. But why has the pattern changed recently? The profession is not becoming any older, nor are economists in their 50s and early 60s now any healthier than their predecessors a generation ago. Nor are the incentives to produce changing; even though a few economists are staying on the job longer, real earnings still stop rising for most of us by our late 50s.

Technology slowdown

What has changed? I believe the answer lies in the changing nature of technology in the profession, a slowdown in the expansion of the technological frontier. The training in a top-notch PhD programme today is not that qualitatively different from what I received in the late 1960s (I was born in 1943); my training then would in every respect have been unintelligible to an economist trained in the early 1920s. In some fields the slowdown in technology is obvious. In my own field of labour economics, for example, there has, if anything, been technological regress.

What does this discussion tell us about the economics profession? There’s hope for us older economists – the slowdown in technological advances has made the profession less like pure mathematics, more like a humanistic field. Old folks are demonstrably more able to compete at the frontiers of research than before. For younger economists, it might be a bit depressing. They no longer have the same advantage of the novelty of their skills as my generation did – the earlier unlikelihood that an ‘old guy’ would be intellectually readily substitutable for them.

Conclusions

In no way should the implied slowdown in methodological advance be viewed as negative for the profession as a whole. For the role of economics in society, the question is whether the profession is keeping up with the problems of an evolving complex society, not how it solves them. While one might despair of our progress in understanding issues and offering solutions for macroeconomic difficulties, the remarkable advances in the application of microeconomic ideas to real-world problems should be reassuring.

The general question of what causes old folks to ‘poop out’ has not been solved, any more than the specific question of why older economists slow down. As an economist, I like to think that the role of incentives, both the economic returns to working and the increasing marginal value of household time (because fewer years of life remain and assets are higher), matter a lot. But physical ability matters, and so does the rate of technical progress (as Jones 2009 shows). It would be great to find examples that allow us to attribute the depressing decline in productivity with age to its varied causes.

Author’s note: This discussion is based in part on some of the evidence provided in my article, “Six Decades of Top Economics Publishing: Who and How?” , forthcoming in the Journal of Economic Literature, March 2013.

References

Hamermesh, Daniel S (2013), “Six Decades of Top Economics Publishing: Who and How?” Journal of Economic Literature 51.

Jones, Benjamin (2009), “The Burden of Knowledge and the ‘Death of the Renaissance Man’, Is Innovation Getting Harder?”, Review of Economic Studies 76(1), 283-317.

Krapf, Matthias (2012), “Age and Complementarity in Scientific Collaboration.” Unpublished paper, University of Konstanz.

Lehman, Harvey C (1953), Age and achievement, Princeton, NJ, US, Princeton University Press.

Mincer, Jacob (1974), Schooling, Experience and Earnings, New York: Columbia University Press.

Weinberg, Bruce A and David W Galenson (2005), “Creative Careers: The Life Cycles of Nobel Laureates in Economics”, NBER Working Paper No. 11799, November.

Topics: Frontiers of economic research, Productivity and Innovation
Tags: academia, age, economists, Nobel, research, technology

Sue Killam Professor in the Foundation of Economics, University of Texas at Austin; Professor of Economics, Royal Holloway University of London

Vox Talks