Sunday, 3 June 2007

Book Review: Capital Ideas by Peter Bernstein

This is a book nominally for finance aficionados, those interested in the theorists who explained stock and bond markets and laid the basis for index funds, portfolio construction and the explosion in derivatives. Capital Ideas is about the history of the thinkers and some implementers of modern capital markets.

Peter Bernstein, not to be confused with the other popular financial author, William Bernstein (he of Four Pillars fame), writes in a non-technical way about the theories. This is quite an accomplishment, given that modern finance is highly mathematical. In fact, there is only one brief formula in the book, which Bernstein throws in just to show us how complex a formula can be. The Black-Scholes formula for pricing options is the only one in the book (228). There might have been a few more, if only to make a deeper impression on the reader about the difficulty of the intellectual work involved. When doing my MBA in 1982, we went through the mathematical derivation of the Capital Asset Pricing Model, another core financial theory, and I remember feeling quite smart merely at being able to understand and follow the logic. UBC did that because it was a fairly quantitative business school. Most MBA schools at the time did not even go that far, merely presenting the final formula with words to describe its meaning. The Black-Scholes was another order of magnitude more difficult than the CAPM with its reliance on Ito's lemma for the solution and we did not attempt the math. However, to really understand any formula properly in order to apply it intelligently, I maintain that you must be able to do the math, otherwise the intricacies and subtleties are lost. There are likely even today, 30 years later, no more than a few thousand people in the world who truly grasp how options pricing works. Though I consider myself to be of reasonable intelligence, particularly in an academic sense, I feel in awe of the finance theorists that Bernstein pays homage to. It is hard to comprehend the abilities of these individuals. On reading that some of these guys came out of physics, I was reminded of my late wife, who had a doctorate in nuclear physics. She described how she used to pass undergrad physics exams – go to class, listen very carefully to understand what the teacher was saying, remember a few basic principles and formulae, then go to the exam and re-derive whatever additional formula a particular problem might require! It's like any field of endeavour – think what Tiger Woods does with a golf ball - the best are simply amazing to anyone with ordinary abilities. Bernstein might have imparted more of the sense of awe that we should rightly feel.

One of the lasting impressions of the book is amazement at Bernstein himself and his ability to understand and explain the many theories, some of them quite subtle indeed. Not everyone seems to get the message of modern financial theory however, as Bernstein describes in the last chapter. There are many portfolios being managed in the institutional world, which is supposed to be professional, in un-diversified, active trading. Some of the managers probably don't understand the theory properly and so mis-apply it.

There are a number of delightful snippets of information and quotes to give a reflective reader something more to think about:

  • The long run returns of the US market of about 8% annually on a real, after-inflation basis, are unique to the United Sates in the 20th century. No other country has done as consistently well, due to disruptions such as wars, famines, revolutions and the like. (A recent paper by Bernstein suggests that using the number as an assumption about the future rate of return is probably much too high.)

  • Or, for us index investors, the S&P 500, is not a very good index. It doesn't represent the market very well, which is what finance theory says a good index should. Bernstein calls the S&P 500 a “a capriciously structured portfolio” (page 248).

  • Change in financial markets, and in the practices of those in the business, tend to change after major disruptions like the 1973-74 and the October 1987 severe drops. In the interim, people are smug, fat, happy and resistant to change.

  • Financial theorists have been almost exclusively English and American scholars, though that seems to be changing now. And they have been 100% men. Not a woman figures among Bernstein's pantheon. So many theorists were converted engineers, physicists and mathematicians (I think that's where the “improbable origins” subtitle of the book comes from; these guys were just looking for some good problems to work on). It's the ultimate nerd activity.

  • The value and significance of research is not immediately apparent, recognized or adopted: Markowitz' paper on portfolio selection languished nearly 10 years after its publication before people took much notice. Or, Black and Scholes submitted their paper on options pricing first to a Chicago University journal and then to one at Harvard, who both promptly rejected it. (220)

  • Things can change permanently: in the years after the 1929 crash stocks were considered so risky that the dividend yield on stocks was nearly three times the interest on savings accounts. In the the 1950s stock prices overtook bond prices for the first time in the history of over 100 years of the US market. That relationship has never returned. “Growth had replaced solvency in market valuation.” (160)

There are amusing and useful sayings for the investor:

  • “Diversification depends more on the way individual securities perform relative to one another than it does on how many assets the investor owns. In Markowitz's terminology, “It is necessary to avoid investing in securities with high covariances among themselves.”” (50)

  • “Diversification takes all the fun out of investing.” (53)

  • “The riskiness of a portfolio depends on the covariance of its holdings, not on the average riskiness of the separate investments. A combination of very risky holdings may still comprise a low-risk portfolio so long as they do not move in lockstep with one another – that is so long as they have low covariance. (54)

  • The market is so hard to outguess because so many people are out there doing the guessing. (134)

  • “The only thing the investor should worry about is how much any asset contributes to the risk of the portfolio as a whole.” (188)

The book has an extensive bibliography with details of all the major papers and books, providing a handy reading list for the essential sources of modern finance up to the date of the book, which was published in 1992.

Buy the book at


Luc said...

Thank you for the review.
I suggest that you provide a link to the book. As an Amazon affiliate you could make some coffee change.

Anonymous said...

Great review.

Sorry to hear that your wife is not around.


CanadianInvestor said...

Thanks Mike for the sentiment. She was a wonderful woman whom I was privileged to know and share many happy years with. She didn't deserve to leave us so soon but life isn't always fair. Her legacy is, along with scientific papers, our children.

Luc, I keep waiting for the Amazon book selector to adapt itself and automatically list a link to the books I review but I guess it isn't that sophisticated and has not done that so I may yet do the manual link.

Anonymous said...

I share Mike's sentiments.

I hope that more than a few thousand people know how to derive the B-S model - otherwise I'm in pretty exclusive company and I only have limited post-grad math knowledge.

The problem with the Black-Scholes formula is in its assumptions. It assumes constant volatility as well as log-normal distributed returns, both of which have been proven false for most of the securities whose derivatives are priced using B-S. By the way, most modern financial theory is based on similar assumptions, which is part of the reason such theories have not mitigated risk as much as one would have thought.

While researchers haven't conclusively locked down an accurate model depicting the randomness of stock returns, there has been some advanced on many fronts, including fractals, chaos theory and nonlinear filtering. This is part of the reason why quant shops are gobbling up Math PhD's - so they can take advantage of mispricings due to the use of an inappropriate model.

Nonetheless, this book sounds like it was a good read.

CanadianInvestor said...

Investoid, you may be too modest about your level of knowledge.

Bernstein mentions in one sentence the Cox-Ross-Rubinstein option pricing model published in 1979 but doesn't discuss it in comparison to Black-Scholes. He also devotes a paragraph to Mandelbrot and the chaos theory. I can imagine a writer so well informed as Bernstein (that's the milieu he came out of and still publishes research about) would have had a tremendous problem deciding what to put in and what to leave out. There could be so much more. One must also keep in mind that this is a history book not a finance book. It won't make one a better investor per se.

Funny how all those math PhDs did not keep BMO from losing $680 million in commodity trading. Few people, except perhaps other math PhDs at other banks who took advantage to eat their lunch, can tell when they got it wrong ... well, everyone can tell afterwards when the spectacular mess happens. Guess that's why the market is quite efficient, the smart are squashing the ignorant constantly.

Anonymous said...

I learned the CRR model too - it's a simplified version of the B-S model (it involves pricing binary options that are valued 1 or 0 at expiry).

Funny, I just made that point on a finance forum - quants may be in high demand, but they're only working with models at the end of the day, and those models are only so close to reality, and reality can change on them at any time. It's why companies like Barclay's has their quants completely revise their models all the time.

Wikinvest Wire

Economic Calendar

 Powered by Forex Pros - The Forex Trading Portal.