Does teaching enhance research in economics? A few more thoughts.

Siegfried, John J.

William Becker and Peter Kennedy have assembled a series of entertaining anecdotes about ways teaching has complemented the research of successful scholars. The examples are numerous, varied, and interesting. There is no doubt that the research productivity of many serious scholars has been stimulated by their teaching responsibilities.

There is a danger, however, of drawing unwarranted conclusions from these interesting vignettes. The successful scholars surveyed by Becker and Kennedy may be individuals of such innate curiosity and creativity that they are stimulated by virtually anything and everything. These may be people, to steal unabashedly from the title of Dan Hamermesh's (2004) wonderful little book, who see economics everywhere. If that is the case, the role of teaching in their research cannot be distinguished from the roles of reading, listening, watching, pondering, eating, drinking, smelling, singing, running, golfing, driving, and perhaps even dreaming.

The individuals surveyed by Becker and Kennedy may have been stimulated at least as much, or perhaps even more, by spending time in informal discussions with colleagues, scanning the New York Times and Wall Street Journal, viewing CNN and C-SPAN, or gazing out the window of their office, but Becker and Kennedy did not ask them how they perceived the research productivity of discussing, reading, watching, and looking. These may be people, like Hamermesh, who, if he spent a week at the zoo, probably soon would be writing about the economics of butterfly beauty, the labor supply of elephants, and the economics of bear hibernation.

A control group with a level of curiosity and imagination similar to those surveyed would be needed to tease out the distinctive effect of teaching on research. Matching curiosity and imagination across individuals would be challenging because there is no accepted measure of curiosity or imagination. It might be reasonable to assume that attributes like curiosity and imagination remain constant in individuals across time and then to compare the stimulation successful scholars received from different working environments--some teaching and some not--they have experienced throughout their careers. The experiences of Herb Grubel and Peter Lloyd reported by Becker and Kennedy suggest the potential value of asking economists who have spent a considerable amount of time in environments both with and without teaching responsibilities, say at a university and at a research think tank, to reflect on which portions of their careers were more intellectually productive, and whether they believe the productivity differences were caused by their environment.

One cynical interpretation of the survey results is that they show that there are smart, exciting, and intellectually curious students at Harvard, Yale, MIT, and Stanford, where many of those surveyed work. Of course, Becker and Kennedy understand this as thoroughly as the rest of us. The interesting question, however, is whether the same scholars, were they not to have enjoyed the privilege of teaching the highest quality students in the world, would have had as productive research careers as they did while rubbing shoulders with the best students in the world. If not, one wonders if teaching is sufficiently valuable that the scholars would have been willing to pay for the privilege of associating with their students?

Becker and Kennedy find that successful scholars believe teaching has enhanced their research. This evidence, however, is insufficient to answer some relatively important questions about the relationship between teaching and research productivity. Becker and Kennedy's evidence shows that teaching is an input into some research success. But, does teaching always lead to successful research? I doubt it. If we find, as I suspect would be true, that sometimes teaching enhances research, whereas at other times it does not, the interesting question then devolves to the frequency with which teaching enhances research. On that question we remain in the dark, because the survey (intentionally) was not random. It did not include teachers who have not been successful scholars.

Another important question is whether successful scholarship can be produced without any exposure to teaching, that is, is teaching an essential input for producing scholarship? If teaching is not an essential input, what are substitutes for it and how much do they cost relative to teaching? In other words, what is the relative scholarly marginal productivity of an hour's worth of teaching vis-a-vis an hour spent pursuing other activities that might also complement research?

A bit unexpected is the way in which the survey respondents seemed to uniformly interpret teaching to mean undergraduate teaching. Most respondents work at universities with large economics Ph.D. programs. It is surprising, therefore, that there is so little mention from the respondents of cross-fertilization between Ph.D. thesis supervision and faculty research, especially so since the knowledge and level of analysis conducted by Ph.D. students is closer to that undertaken in faculty research than is the content of typical undergraduate instruction.

Finally, I offer a category of how teaching enhances scholarship that is not mentioned in the paper, namely, the future research collaboration between scholars and their students. Several examples illustrate the productivity of such connections: Robert Shiller and his former student, John Campbell; Edward Prescott and his former student, Finn Kydland; James Tobin and his former student, William Brainard. Scholarly progress resulting from a collegial relationship initiated through student-teacher contact deserves no less acclaim than the numerous happy and productive marriages between former students and their teachers.

Reference

Hamermesh, Daniel S. 2004. Economics is everywhere. New York: McGraw-Hill/Irwin.

John J. Siegfried, Department of Economics, Box 1819-B, Vanderbilt University, Nashville, TN 37235, USA. Received March 2005; accepted May 2005.