The Economist as Bankrupt

Philip J. Davis

SIAM News, Volume 27, Number 9, November 1994


Irving Fisher: A Biography By Robert Loring Allen
Blackwell, Cambridge, Massachusetts, 1994, 324+xv pages, $39.95

In December 1929, in Des Moines, Iowa, two months after the big crash, Irving Fisher, professor of economics at Yale, mathematician, economist, writer, lecturer, columnist, inventor, prohibitionist, health enthusiast, crusader and public figure extraordinaire, gave the prestigious Gibbs Lecture of the American Mathematical Society. Fisher began the lecture, entitled "The Applications of Mathematics to the Social Sciences," with an apotheosis of Josiah Willard Gibbs, who had been one of his teachers. He then cited a few of the early names and concepts of econometrics and wound up with the prediction that the future would see many applications of mathematics to the social sciences.

Fisher had just lost or was about to lose a fortune in the crash (an estimated $10 million, in 1930 dollars!), and he never, thereafter, got out of debt. It is clear that his oracular power as regards the future role of mathematics in the social sciences was stronger than his ability, with or without mathematics, to read the market.

I have a soft spot for Irving Fisher. Although I never knew him, he played a considerable role in my becoming a mathematician. It came about in this way. When I was in high school, my geometry teacher gave me a copy of Phillips and Fisher's Elements of Geometry (1897), a rather more difficult text than the one we were using in class. This book had an appendix with a collection of problems. I went through these problems systematically.

Eventually, I came to a problem I couldn't solve. As a high school student, I worked on that problem for months without being able to solve it. Later, in college, I still couldn't solve it. Many years later still, I perceived the relation between that problem and the discrete Fourier transform, and I wrote a little paper on the topic. I may yet write more on it. Fisher's problem was both a spur and a permanent bequest to me. Perhaps if I go to the Fisher Archives and look in Box 24, File 358, where his "Senior Math Prize Problems" are stored, I shall find out how Fisher himself solved the problem.

But back to the man. The late Robert L. Allen, who was a professor of international economics and economic history at the University of Missouri, has written a very readable and carefully researched book, one that will be enjoyed by a wide spectrum of readers. There is some discussion of theoretical economics in it, so that we learn about both the man and his work, but I, who consider myself a babe-in-the woods as regards economics, had no difficulty in following these portions.

Irving Fisher (1867-1947), born in the Catskill Mountains, was the son of a Congregational minister. He should not be confused with Sir Ronald A. Fisher (1890-1962), the biostatistician whose name is perhaps a bit more prominent in applied mathematics circles.

Irving Fisher's genealogical line can be followed back on the male side to 16th-century Palatinate Germany, where Fisher was Fischer. He grew up in Peace Dale, Rhode Island, a small, paternalistic mill village, founded and run by the Hazard family, where his father had a congregation. Fisher was a pukka Yale Man, both undergraduate and graduate--Skull & Bones, crew, and all the rest. He married Margaret Hazard, daughter of the wealthy mill owner Rowland Hazard, and he always thought of Peace Dale as his home town.

Fisher was an assistant professor of mathematics at Yale in 1892 when he co-authored the elementary geometry book that played a role in my life. He spent 1894 in Europe, studying and poring over notes from the lectures of Helmholtz, Schwarz, Frobenius, and Poincaré,

By 1895 he had switched out of mathematics and into economics. By 1898 he had become a full professor, and then he fell ill with tuberculosis. In a remarkable recovery based on good advice, luck, and extraordinary will power, he worked his way successfully out of a three-year bout with this terrible illness. The experience turned him into a lifelong campaigner for good health habits. He returned to the Yale faculty in 1903.

Fisher didn't smoke, and he drank no alcoholic beverages, coffee, or tea. He didn't eat chocolate or pepper. He ate salad for breakfast. (So do I.) He was almost a vegetarian. For a while, he thought that a diet of bananas and peanuts contained all the ingredients necessary for life, and he estimated that a person could live on this diet for $35 a year. Whether he considered this a practical suggestion is doubtful; in any case, it was surely an anticipation of the kind of problem that would become standard in courses in linear programming: Find the minimum-cost diet that includes all the necessary dietary elements and excludes those that are counterindicated. In many different ways, Fisher's ideas were in advance of their time.

Fisher was not a modest man. His ego was, in fact, enormous. He had plenty going for him and he knew it, and his desire to convince others and convert them was tremendous. He thought of himself as a knight in shining armor, and contemporaries agreed: "a modern Parsifal," was Harvard economist Joseph Schumpeter's description; "pure in heart and noble in purpose," is Robert Allen's addition.

Fisher had little sense of humor; everything was serious to him. No existing photo shows him smiling. He was an optimist. (Who says that optimists must smile?) He believed in the powers of auto-suggestion as expounded by Emile Coué, a French psychotherapist whose writings were all the rage in the 1920s.

Fisher, like Harvard mathematician William Fogg Osgood (author of Lehrbuch der Funktionentheorie), his exact contemporary, and like the younger Norbert Wiener, wore a beard, feeling that it would enhance his professional image. In an age when most men were clean-shaven, a beard in the academic U.S. was a status symbol of the scholar steeped in European Wissenschaft.

Fisher wrote more than 30 books. If you flip through The Making of Index Numbers (1923), for example, you will find that he was looking for the ideal way to compute index numbers. The book was a major contribution to the field. (Now that indices of this, that, and the other are almost as commonplace as the air we breathe, it came as no surprise to me, on a recent trip to California, to have stumbled across an Index of Laidback-ness.) Fisher also worked on the theory of interest rates. He built a hydrodynamic model of equilibrium and explained it in his mathematical investigations of the theory of value and price.

James Tobin, Yale Nobelist in economics, has opined: "In his writings on capital and interest, Fisher had laid the basis for the investment and savings equation central to modern macroeconomic analysis. Had Fisher pulled these strands together into a coherent theory, he could have been an American Keynes. . . . Fisher would have done it all himself."

It is clear from what has just been said that, willy-nilly, we are living in a Fisherized world.

During the years from 1925 to 1929, Fisher earned a fortune from his invention of a rotary index card file and from his investments and stock market manipulations. From 1929 to 1932, he lost a fortune in the stock market crash. Although the loss affected him psychologically, Irving Fisher--always the optimist--was never one to jump out the window.

His precrash financial dealings and interpretations got him in trouble with the IRS. (Irving Fisher, Al Capone; does it make any difference to the IRS? Or to a computer?) He went to tax court and lost. By January 1931, the IRS told him to pay up or else. He was lucky that his wealthy sister-in-law (Caroline Hazard, president of Wellesley) was willing to bail him out a bit, although she lost patience with him, and her sisterly-in-law feelings diminished considerably when he kept coming back for more.

In 1934, after a year of haggling, Yale bought his expensive home and rented it back to him on a life-tenancy basis. Even this arrangement proved too expensive, and Fisher paid Yale in IOUs instead of cash. Naturally, the Yale administration was irritated and began to think of him as a crank and a nuisance instead of the genius that he undoubtedly was. (Yale University is now reprinting his major works.) As Allen says, "His efforts made him into the country's most well-known and unsuccessful monetary reformer whose poor judgement cost him his fortune, his businesses and his home."

Never a mere theoretician, Fisher always wanted to transform his theories into policy. Strapped though he was financially, he knocked on the door of the White House and was admitted. He got on fairly well with President Franklin D. Roosevelt on a personal level; he thought of himself as a presidential adviser--after all, he was "the greatest American economist that America has produced" (Fisher himself would have agreed with these words of Joseph Schumpeter). As it turned out, his batting average with respect to recommendations that were actually accepted by Roosevelt was only so-so.

Let us now perform a Gedanken experiment: If Fisher had had available all the tools--hardware, software, models, theories of chaos, Lyapunov exponents--that have been developed since his death, would he have been able to ward off his staggering losses? Since, as I suggested earlier, I am not out of kindergarten when it comes to The Market, I put this question to someone who in his various books has written much about the efficacy of econometric models.

John Casti responded as follows: "I'm fairly certain that if Fisher had been the only one in possession of modern computer methods and techniques of chaos, fractals, et al., in 1929, he certainly would have been able to employ them to advantage in reversing his losses. The tools would almost certainly help someone who has a differential access to them.

"But how helpful would these methods be in predicting price movements if 'everyone' has access to them? Here the verdict is pretty unclear. What happens depends upon the decisions made by everyone in the system. If, for example, everyone has a system that says the market is going up and acts on that prediction, then the market goes up and they gain confidence in their predictors. So, are these predictors any good or not?"

What is the moral to the story of the bankrupt econometrician? Fisher was not the first genius to die in debt. Mozart wound up in a pauper's grave. Thomas Jefferson's personal library was auctioned in order to pay his creditors. While he was alive, Fisher endured plenty of ridicule from laughing skeptics: "Physician, cure thyself," they said.

Although it was not the case until 250 years ago, mathematicians generally believe in the separation of mathematics and morals. They turn experience into definitions. Responding to the question just posed, they might answer, following Richard von Mises, that "random" means that any expert who claims to have a system might very well lose his shirt in betting against the Universe. Or, following Martin-Lof, Kolmogoroff, and Chaitin, the actual play of the market is the least complex description of itself that is possible.

And with these high-flown, existential, and inconclusive observations, I leave to SIAM News subscribers the pleasure of reading the full and fascinating story of Irving Fisher in Robert Allen's wonderful book.


Philip J. Davis, professor emeritus of applied mathematics at Brown University, is an independent writer, scholar, and lecturer. He lives in Providence, Rhode Island, and can be reached at AM188000@brownnvm.brown.edu.


Reprinted from SIAM News
Volume 27-9, November 1994
(C) 1994 by the Society for Industrial and Applied Mathematics
All rights reserved.