A History of the Theory of Investments: My Annotated Bibliography (a review)

In this annotated bibliography, Mark Rubinstein selects and explores what he considers to be the path-breaking books and research papers in the historical development of investment theory. The book is divided into three periods—“ancient” (pre-1950), “classical” (1950–1980), and “modern” (post-1980)— which contain, respectively, 37, 116, and 27 entries. Rubinstein discusses each, either individually or in small groupings, in entries of 1–10 pages.

The main value of A History of the Theory of Investments: My Annotated Bibliography is not as a didactic device but as a reference tool. For the most part, Rubinstein presents the publications in chronological rather than topical order. He makes a few key points about each topic but does not go into enough depth for the reader to come away with comprehensive knowledge. Those who are already familiar with the material in a particular book or paper will find Rubinstein’s discussions useful refreshers. For others, however, most of the treatments will be too brief for benefit. This outcome is to some extent unavoidable because the book’s scope is extremely wide and only so much commentary can fit into a single volume.

Nevertheless, some of the material could easily have been left out. For example, the book appropriately discusses the influence of the 15th century’s Luca Pacioli on the development of accounting but then devotes 10 pages to a treatment of accounting issues, from which Pacioli is entirely absent. Another example is the Coase theorem, which, although certainly seminal in nonfinancial economics, is arguably out of place in the present context.¹ The justification for its inclusion—to be compared with the Franco Modigliani and Merton Miller theorems on the basis that both Coase and M&M are about irrelevancy—is too much of a stretch for this reviewer.

Similarly superfluous are mathematical proofs that Rubinstein provides for some propositions in the papers he discusses. Examples include a proof of the additivity property of duration and a “proof that quadratic utility or multivariate normality implies mean–variance preferences.” Many of these are Rubinstein’s own proofs rather than those of the original authors, and often they are too opaque to provide insight. A reader desiring a proof can simply refer to the original paper. The space allotted to proofs could have been used, instead, to include additional papers.

The book leaves out too many papers to be considered definitive. Rubinstein acknowledges as much by saying he simply “hasn’t gotten to” some papers yet, even though they warrant inclusion. He asks the reader to suggest papers to be added to future editions of the book.

The selections that compose the book lean strongly toward financial economics. As Rubinstein readily admits, the book includes no papers on behavioral finance (the term is even absent from the book’s “Index of Ideas”). Another topic ignored entirely is fundamental indexation, the debate over which may have begun too recently for inclusion in the book. This reviewer also would have liked to see more publications related to the debate over the M&M irrelevancy propositions, which Rubinstein correctly lists under the rubric of corporate finance but, curiously, not of investments. Also, almost no articles are to be found from the entire 60-year and more history of the Financial Analysts Journal (the only one among the 180 main entries is by Rubinstein himself). This reviewer would humbly recommend some of the articles in the FAJ’s January/February 1995 50th anniversary reprint issue or some of the articles that, like Rubinstein’s, have won Graham and Dodd Awards.²

Nevertheless, A History of the Theory of Investments includes most of the books and papers one would expect to find in such a volume, and Rubinstein’s discussions are, for the most part, enlightening. Along the way, we learn a number of interesting tidbits. Financial practice led to the development of probability theory, rather than vice versa. The first formal analysis of option-style derivatives dates back to Dutchman Johan de Witt in 1671, who also discovered what today we call “adverse selection.” Risk aversion dates back to Bernoulli in 1738. Diversification as a risk management tool dates back to the Talmud. Although it is perhaps well known that Louis Bachelier in 1900 (in “undoubtedly the finest thesis ever written in financial economics”) anticipated the mathematical basis for the Black–Scholes option-pricing formula, less well known is that Bachelier was, in turn, anticipated in 1863 by another Frenchman, Jules Regnault. Rubinstein even explains the origin of the word “bankruptcy.”

Rubinstein makes the important point that the origin of scientific ideas is often misattributed. Indeed, a useful consequence of reading the book from cover to cover is the realization that great ideas often build on one another. For example, Rubinstein cites a wonderfully articulate paragraph by John Burr Williams in 1938 that contains all the essential truths of M&M’s subsequent proposition about the irrelevancy of the capital structure. In turn, both the Williams and the M&M formulations were arguably founded on Irving Fisher’s (1930) theorem of separation between financing and production. Another example is the puzzle of the equity risk premium. Although usually thought to have been posed first by Rajnish Mehra and Edward C. Prescott in 1985, it had, in fact, been observed, albeit not labeled as such, more than 20 years earlier following the publication in 1964 of equity rates of return by Lawrence Fisher and James H. Lorie.

Helpfully, each main entry in A History of the Theory of Investments provides complete bibliographical details of the original publication and those, if applicable, of modern translations and reprints. Locating the actual works is thereby greatly facilitated. In conclusion, even though it is not comprehensive, the book is a valuable reference tool that belongs in the library of any serious student or practitioner of finance.

–J.H.T.