Recent Financial Books
By Nassim Taleb
Derivatives Handbook: Risk Management and Control
Robert J. Schwartz and Clifford Smith Jr., eds. New York:
John Wiley and Sons.
This is an informative and broad compilation of articles on the various facets of market risk management. It is not just a good book—it is the only one of such breadth. It ambitiously covers areas from control and accounting to regulation and hedging. I disagree with a small minority of the chapters (mostly articles related to value-at-risk), but the rest range from the mildly factual to the luminous (the latter applies to the papers on regulation by Merton Miller and Christopher Culp and to the chapters of the Metalgesselshaft debate).
On the surface, the book mixes articles written by practitioners and academics on the overly babbled topic of market risk management. We have been flooded with conferences on risk management with few concrete things to say—merely platitudinous sermons proffering the financial markets equivalent of "brushing your teeth greatly reduces the incidence of tooth decay.” The pleasant surprise in this volume lies in the (relative) dearth of such confident platitudes.
Risk management is at its best when we look at the structure of the market system and, using common sense, study ways to protect it from what we do not know, while keeping in mind the limitations of our knowledge and the weaknesses of the instruments used in its acquisition. It is at its worst when it focuses on what I call casual and arrogant measurement, when we become presumptuous over the extent of our knowledge—with people uncritically offering scientific measurement without having given much thought to the inductive problems this entails. Bad risk management reeks of the bad side of predictive economics, so aptly dubbed "scientism.” This book largely consists of good risk management; most of the references to value-at-risk are tolerable and honest, written by authors who do not seem to take the concept very seriously.
The book includes seven sections of unequal length. The first discusses derivatives risk and control. The second discusses legal risks, and it is informative, at least to a reader without legal training. The third discusses risk measurement, and is the only section that contains misleading articles (a topic to which I will return). The fourth discusses areas of risk oversight. The best sections are the fifth, sixth and seventh, which discuss regulation, transparency and disclosure rules, and the Metalgesselshaft debate, respectively.
The most valuable part of the book, and one sufficient reason to read it, is its reprinting of the discomforting Metalgesselshaft debate. The points of contention that involved the pairs Culp and Miller, Mello and Parsons, and Canter and Edwards are presented almost in full. It is indeed a strange and depressing debate as the protagonists all write with the unmistakable stamp of scholarship. Its depressing aspect lies in the possible interpretation of the disagreements between financial economists. Such disagreements make the profession sound like a mushy soft science fraught with ideological divisions, something it has been striving to eliminate. The impression it conveys, beyond the elements of the debate, is the towering image of the Chicago economists, particularly when one includes the reflective and provocative articles on the economics of regulation by Miller and Culp. While the two authors, in my opinion (and probably in the opinions of those who have lost money hedging a crude oil forward contract), are totally wrong in their Metalgesselshaft conclusion, they do not fail to impress the reader with their intellectual depth and economic erudition. We do not identify the mark of a great thinker when he or she is right—any glib, shallow thinker who has a good argument can defend it persuasively. The mark of true thinkers, as Aristotle intimated, is when they are plain wrong but inspire in one so much respect that they almost make one revise his or her own logic.
I wish such debate between financial economists did not take place so publicly. I also wish that my hero, Merton Miller, one of the representatives of a Chicago insight that harks back to Knight, Stigler, Coase and Friedman, did not get involved in the defense of the activity of a hedger with whose definition of hedging the current market structure so heavily disagreed. The only methodological protective belt traders have is the dogma that he who loses money is unconditionally wrong and the market unconditionally right, rather than the opposite. It is something we cannot question without threatening the foundations of our profession. The implication is something called a stop loss, which is executed as a way to ensure survival and force such revision of beliefs. Even Merton Miller can be wrong: A stop loss is there to protect us from the hypotheses that we accept as gospel. Once such a protective belt is lifted, Miller's argument can become potent. Likewise, Miller has argued elsewhere that Orange County should have refrained from closing down its positions, which, in retrospect, would have paid off, except that such statement violates the protective belt.
|We have been flooded with platitudinous sermons proffering the financial markets' equivalent of "brushing your teeth greatly reduces the incidence of
I have little disagreement outside of two articles. The first is an article entitled "Report Card on Value-at-Risk,” by Tanya Beder. It is sufficiently thoughtful and informative to include the fact that every year at least one market experiences a 10-standard-deviations move (she does not discuss the more deleterious cases of joint moves of a vector of securities). This information should be obvious to anyone who devotes part of his or her time to the observation of markets. The argument that follows, however, does not bring the observation to it logical conclusion. A 10-standard-deviation move, according to the models, should only be expected to occur every few billion-billion years (the model has one chance in several billion-billion of being correct). I do not understand why she writes so confidently in her concluding statement that, supplemented with other methods, value-at-risk gets high marks. The author shows that the measurement of the parameters will exhibit extreme variability according to the length of the time window used (the properties of a 100-day series can be markedly different from, say, those of a 101-day series). This is clearly to mean that either the data being used are not stationary (and the value-at-risk number is meaningless) or that the process used to determine value-at-risk is grossly mis-specified.
Violations of sound empiricism are not the exclusive province of practitioners. An article by the academics Jordan and Mackay on implementing value-at-risk for equity portfolios presents a technique for the calculation of the magic number, after establishing (using their admittedly casual test) that equity portfolios tend to be normally distributed, an assumption without which their paper would be worthless. Theirs is, ironically, the only mathematical paper in the book and the only paper that used the formal tone of science. They use a Kolmogorov-Smirnoff type test of normality for a time series, a test so conditional as not to be acceptable in the social sciences. A Kolmogorov-Smirnoff-type test can only show normality conditional on the time series having reached its equilibrium frequencies. In other words, it can tell you that the data would be normal if they included all possible regime switches and if the generating process was pure of what economists call "peso problems.” Take a market that is well-behaved but has rare and violent jumps. The Kolmogorov-Smirnoff test will pronounce the series normal if the jump is not included. The difference between conditional normality and normality is not trivial: It makes the test a useless tautology for time series and is, like most tautologies, dangerous. In addition, the authors hint that, even in cases where individual stocks present non-normality, a portfolio combining them tends to be normally distributed. They obviously have not studied the harrying properties of a portfolio with negative and positive weights—the attributes of a dealer book.
Among the articles worth reading is a characteristically crisp paper by Alan Greenspan that, not unexpectedly, is written in a way that makes one read it at least twice looking for hints. Also there is a practical paper on operations risk by Wendy Brewer that provides down-to-earth measures to control the significant exposure to the back office.
Black-Scholes and Beyond
Neil Chriss. New York: McGraw Hill.
This book, which proffers a pleasing intellectual style, is the rehabilitation of binomial-tree. It is an introduction to option pricing that is extremely modest in its aims and does exactly the job it intends, perhaps better than any other book on the subject.
An impressive aspect of the author is that, although he is a mathematician with first-rate credentials (University of Chicago doctorate, Harvard and Institute of Advanced Studies postdoctoral studies), the text does not include one misplaced buzzword or one display of showoff-iness. Not having much to prove to the reader, he embarked on a pedagogical venture that aims more at teaching than impressing—like great pedagogues. It is no accident that he had recourse to the trees. Lattice trees, which I will discuss next, have three attributes: simplicity, honesty and computational superiority.
First, binomial trees open up option theory to just about anybody with intellectual curiosity. Why? Because they draw not on advanced mathematics but on simple arithmetic, and are therefore accessible to any hard-working person equipped with probabilistic intuition. They allow readers to go far in their study of options without the formalism or the language of mathematics. Some recent authors confuse readers with a great deal of complicated mathematical properties using a language borrowed from topology and measure theory, apparently unaware that all that is significant in these properties had been illuminated in the 1970s by John Cox, Steve Ross and Marc Rubinstein, among others. It is a sad fact that sophistication has been put to the service of tenure-seeking swagger rather than the advancement and the dissemination of knowledge.
Second, trees accordingly prevent someone from concealing nonsense. This is a refreshing method, considering the current excesses in our environment. As Marc Kritzman writes on the cover of the book, readers do not have to "decipher the mathematical symbolism that usually permeates writings on this topic.” Literary scholar George Steiner holds that a language that needs to be deciphered can be a language used for obfuscation rather than communication (language as a weapon to protect a territory, here an intellectual franchise). Furthermore, James Buchanan (Nobel Memorial Prize in Economics, 1986) believes that much of economics has been invaded by clueless savants who are unaware of elementary economic propositions, and satisfy their lack of confidence by using an overly complicated language. Take the following simple test, recommended by the great J.M. Keynes: find an option paper written in a complex symbolic language—that is, with a high mathematics-to-English ratio (lattice trees and arithmetic equations count as English; topology counts double). Translate it into plain English, and see if you express a "so what ?” Often the ideas appear to be a combination of the tautological and the circular, and the authors seem to be trying to hide from themselves the fact that they have not given much thought to what an option means.
Third, trees provide us with solutions that continuous time (and more complex) methods cannot provide, and allow us to fit the volatility surface. We can use time-dependent parameters in them. Trees have an excellent track record, particularly in numerical solutions to American options.
This remains a technical, not an economics, book—and a good one at that.
Nassim Taleb, senior adviser at Paribas, is the author of Dynamic Hedging: Managing Vanilla and Exotic Options. The opinions expressed in this article are the author's and do not necessarily represent those of Paribas.
Oh No! The Sky Is Falling Again!
By Stephen Rhodes
Apocalypse Roulette: The Lethal World of Derivatives
Richard Thomson. London: Macmillan U.K. (Not yet published in the United States.)
Consumer Alert: Don't be put off by the Chicken Little title of Richard Thomson's enjoyable new book on the business of derivatives. True, any book named Apocalypse Roulette: The Lethal World of Derivatives would seem to be intent on pushing the buttons of the mass-market audience who only know derivatives from the 64-point screaming tabloid headlines decrying the demise of Barings, Orange County and Sumitomo. Nevertheless, Roulette is ultimately not so much an end-of-the-world-is-near manifesto as it is a fine reporter's survey of the unique personalities that drive the derivatives markets.
Last year's big-buzz book on derivatives, Frank Partnoy's F.I.A.-S.C.O.: Blood in the Water on Wall Street, was a compulsively readable mole-on-the-inside account of a year on Morgan Stanley's fixed-income derivatives desk, a micro-view of the market. By contrast, Roulette is the macro-view of this world, studded with some juicy tidbits about the legendary players drawn to this game.
First, of course, the author trots out the obligatory gloom-and-doom. The first two chapters lay the foundation for the potential destruction of the global economy—first with fact, then with fiction. After an account of recent derivatives debacles, Thomson launches into a fictional minute-by-minute account of a Nick Leeson-like character who single-handedly loses $6 billion in 48 hours by trading equity futures and options on the Sao Paulo and Hong Kong Futures Exchanges. Naturally, cheap margins and lax trading control are blamed. The black hole of losses at this single institution results in a domino effect that brings down several investment banks, the Sao Paulo and Hong Kong exchanges and the Bank of England. A global Depression follows.
Fair enough—derivatives are potent. Thomson draws sober analogies between financial instruments and weap-ons of mass destruction. "Think of derivatives as a loaded gun,” he writes. "You can use it in self-defense or for murder.” Oy. The author also likens them to "high explosives,” "nuclear power” and a "glorified casino.”
At the same time, however, Thomson undercuts his book's central thesis with the balanced reporting that comes naturally to him as the former deputy city editor of The Independent on Sunday. The number $64 trillion is repeatedly invoked as the total notional amount in outstanding derivatives contracts as of 1996—60 times the size of Britain's economy, Thomson notes. Yet later in the book, he concedes that notional is not truly "value-at-risk” and, accordingly, the number is misleading when represented as a meaningful indicator of the perils derivatives pose to the world economy.
Also, while Thomson decries the greed and stupidity that brought down the venerable House of Barings in his home country, he neglects to acknowledge the resilience of a global market that so efficiently absorbed the shock waves of such an implosion. When the author holds up the infamous 1636 Tulip Bulb Mania in Holland as an example of how an out-of-control derivatives market can destroy the lives of ordinary citizens once the bottom drops out of the market, the reader thinks more of the super-heated U.S. stock market than, say, the global interest rate swaps market.
|Apocalypse Roulette: An Excerpt
|In 1986, [Bankers Trust] hired a young currency options trader from Salomon Brothers called Andy Krieger who had graduated from the Wharton School after studying Sanskrit and philosophy at the University of Pennsylvania. He quickly became one of the most aggressive dealers in the world, with the full sanction of Bankers. While most of the bank's currency traders had an upper dealing limit of $50 million, Krieger's was in the region of $700 million—around a quarter of the bank's capital at the time. By using options, Krieger could leverage this exposure to many times that size ($100,000 of currency options would buy control of $30 million to $40 million in actual currency). In 1987 he did this to launch a speculative attack on the New Zealand dollar. If his own claims can be believed, he sold short the entire money supply of the country. In a matter of hours, the NZ dollar plunged 5 percent against the U.S. dollar. It was enough, at any rate, to draw an angry complaint from the New Zealand central bank. But with typical arrogance, Sanford later turned this on its head. "We didn't take too big a position for Bankers Trust,” he grumbled, "but we may have taken too big a position for that market.” It was New Zealand's fault, in other words, for being too small to cope with Bankers.
Krieger resigned the following year in disgust at the ingratitude of his employers who had paid him a mere $3 million for his efforts which had netted the bank a profit of more than $300 million. But after his departure an odd thing happened. Regulators discovered discrepancies in the way Bankers valued its currency options portfolio. The bank was forced to admit that $80 million of foreign exchange trading income had disappeared and that it had deliberately overstated its 1986 earnings. It seemed, on the face of it, that the bank had been simply unable to understand Krieger's complex options positions. Dealers in other banks, however, wondered how it was that Krieger's options portfolio only maintained its value as long as Krieger himself was in charge of it. Whatever the reason for the readjusted profit, the whole episode was a serious embarrassment. An even bigger embarrassment, though, should have been that Bankers had been willing deliberately to publish figures that were inaccurate by $80 million as if it didn't matter. But the bank showed no sign of blushing.
This was yet another warning of its attitude: anything goes as long as the suckers don't find out. Bankers was lean and very, very mean and there were no bigger suckers in sight than in the derivatives markets.
Excerpted from Apocalypse Roulette: The Lethal World of Derivatives, by Richard Thomson. Reprinted with permission by the publisher, MacMillan U.K.
Even Thomson himself concedes, "Derivatives are not bad in themselves…used properly and wisely, they are an undoubted benefit.” How much of the $64 billion on the books, then, represents proper and wise use?
The book soars when Thomson turns his formidable narrative powers on the people drawn to the derivatives game. Invoking the names of present and future derivatives hall-of-famers, Thomson provides a cascade of snapshots that define the people who created this business from ground zero in just two decades.
His "Short History of Derivatives and Speculation” details the genius of the philosopher Thales, who made a fast fortune in 330 B.C. by soliciting and purchasing cheap options on the use of olive-oil presses that permitted him to corner the market on the use of the presses during a bumper olive crop. The anecdote unintentionally positions Thales as the original thinking-man-propeller-head-quant-jock who applied his analytical powers to create wealth where there was none.
The studs and duds of derivatives invariably come off as complicated, fascinating individuals. CSFP's CEO Allen Wheat is chilly, but visionary—two-thirds of the way through a year of record revenues, Wheat threatened to fire the entire derivatives unit if they failed to double revenues by year's end. (They succeeded.) Ex-Kidder CMO trader Mike "The Arm” Vranos is brilliant but vicious—the former weight-lifting champion brought five deals to market in a single week worth about $9.2 billion, but was known to smash chairs and phones into pieces during tense moments. Former Bankers Trust currency trader Andrew Krieger was religious but hyperaggressive: soon after he quit in disgust after BT paid him a mere $3 million bonus against the $300 million profit he derived for the firm, regulators uncovered an $80 million "overstatement” of his portfolio's value.
Also entertaining is Thomson's account of Howard Rubin, who lost $250 million in a single day trading mortgage-backed derivatives for Merrill in 1987 and hid the evidence in a desk drawer. (He was instantly hired by Bear Sterns, where today he is considered a big hitter.) Others in cameo roles include the mysterious "Hot-Tub Sheik,” who moves the currency futures markets with a single phone call to the trading floor, and, of course, George Soros, Joe Jett, Leo Malamed, Robert Citron and David Askin. Inexplicably, the book includes no mention of Howard Sosin, formerly of AIG Financial Products, a Columbia b-school professor who is widely credited with transforming mere theoretical models into lucrative high-margin business lines for Wall Street firms.
For the most part, Richard Thomson's skill in presenting the human element of the derivatives business compels the reader to keep flipping pages until late into the night. Apocalypse Roulette is a must-have for the derivatives practitioner's personal library, although the inflammatory title may tempt you to display it on your bookshelf with the spine facing in.
Apocalypse Roulette is not yet published in the United States. Readers can order a copy by calling the publisher's U.K. office at 011-44-1256-329-242 during business hours.
Stephen Rhodes is a lawyer specializing in derivatives at a global financial firm. His Wall Street novel The Velocity of Money was published by William Morrow last winter and is available at amazon.com.
CIBC's 1998 Yearbook
Managing Financial Risk: 1998 Yearbook
Charles Smithson. Toronto: Canadian Imperial Bank of Commerce.
For several years running, Charles Smithson, managing director at CIBC's School of Financial Products, has been producing one of the most interesting ongoing projects in financial publishing. As a supplement to his textbook, Managing Financial Risk, he releases a yearbook each spring that outlines some of the major derivatives events of the previous year.
It's a good way to get a summary of the most important developments in the field and catch up on the news you may have missed. A chapter reviewing financial markets, for example, summarizes the chronology of last year's Asian crises, and touches on financial prices in a dozen or so markets around the world.
Elsewhere, the book reports on developments in the credit derivatives and electricity markets, the results of several surveys, and provides updates on major regulatory and accounting events. A chapter on dealers was particularly interesting, as it summarized all the year's bad news about dealer losses, as well as notes on mergers and dealer rankings.
CIBC promises to post an electronic copy of the book on its web site (www.schoolfp.cibcwm.com).