Why We Have Never Used the Black–

Scholes–Merton Option Pricing Formula

Espen Gaarder Haug
Nassim Nicholas Taleb

Abstract versions of the formula of Louis Bachelier and Edward O. Thorp (that allow a broad
Options traders use a pricing formula which they adapt by fudging and changing the choice of probability distributions) and removed the risk parameter by using put-call
tails and skewness by varying one parameter, the standard deviation of a Gaussian. parity. The Bachelier-Thorp approach is more robust (among other things) to the high
Such formula is popularly called “Black-Scholes-Merton” owing to an attributed impact rare event. The paper draws on historical trading methods and 19th and early
eponymous discovery (though changing the standard deviation parameter is in con- 20th century references ignored by the finance literature. It is time to stop calling the
tradiction with it). However we have historical evidence that 1) Black, Scholes and formula by the wrong name.
Merton did not invent any formula, just found an argument to make a well known
(and used) formula compatible with the economics establishment, by removing the
“risk” parameter through “dynamic hedging”, 2) Option traders use (and evidently
have used since 1902) heuristics and tricks more compatible with the previous

Breaking the Chain of Transmission adapted in a robust way by a long tradition of researchers and used
heuristically by option book runners. Furthermore, in a case of scientific
For us, practitioners, theories should arise from practice1. This explains puzzle, the exact formula called “Black-Sholes-Merton” was written
our concern with the “scientific” notion that practice should fit theory. down (and used) by Edward Thorp which, paradoxically, while being
Option hedging-pricing, and trading is neither philosophy nor mathe- robust and realistic, has been considered unrigorous. This raises the
matics. It is a rich craft with traders learning from traders (or traders following: 1) The Black Scholes Merton was just a neoclassical finance ar-
copying other traders) and tricks developing under evolution’s pres- gument, no more than a thought experiment2, 2) We are not aware of
sures, in a bottom-up manner. It is technë, not ëpistemë. Had it been a traders using their argument or their version of the formula.
science it would not have survived—for the empirical and scientific fit- It is high time to give credit where it belongs.
ness of the pricing and hedging theories offered are, we will see, at best,
defective and unscientific (and, at the worst, the hedging methods cre-
ate more risks than they reduce). Our approach in this paper is to ferret The Black-Scholes-Merton “Formula”
out historical evidence of technë showing how option traders went
about their business in the past.
was an Argument
Options, we will show, have been extremely active in the pre-modern Option traders call the formula they use the “Black-Scholes-Merton” for-
finance world. Tricks and heuristically derived methodologies in option mula without being aware that by some irony, of all the possible options
trading and risk management of derivatives books have been developed formulas that have been produced in the past century, what is called the
over the past century, and used quite effectively by operators. In parallel, Black-Scholes-Merton “formula” (after Black and Scholes, 1973, and
many derivations were produced by mathematical researchers. The eco- Merton, 1973) is the one the furthest away from what they are using. In
nomics literature, however, did not recognize these contributions, sub- fact of the formulas written down in a long history it is the only formula
stituting the rediscoveries or subsequent reformulations done by (some) that is fragile to jumps and tail events.
economists. There is evidence of an attribution problem with Black- 2
Here we question the notion of confusing thought experiments in a hypothetical
Scholes-Merton option “formula”, which was developed, used, and world, of no predictive power, with either science or practice. The fact that the
Black-Scholes-Merton argument works in a Platonic world and appears to be “
1
For us, in this discussion, a practitioner is deemed to be someone involved in repeated elegant” does not mean anything scientifically since one can always produce a
decisions about option hedging, not a support quant who writes pricing software or Platonic world in which a certain equation works, or in which a “rigorous” proof
an academic who provides “consulting” advice. can be provided, a process called reverse-engineering.

72 WILMOTT magazine
 TECHNICAL ARTICLE 3

addition to a collection of assumptions that, we will see, are highly in-

valid mathematically, the main one being the ability to cut the risks
using continuous trading which only works in the very narrowly special
case of thin-tailed distributions. But it is not just these flaws that make it
inapplicable: option traders do not “buy theories”, particularly specula-
tive general equilibrium ones, which they find too risky for them and
extremely lacking in standards of reliability. A normative theory is, simply,
not good for decision-making under uncertainty (particularly if it is in
chronic disagreement with empirical evidence). People may take decisions
based on speculative theories, but avoid the fragility of theories in running
their risks.
Yet professional traders, including the authors (and, alas, the Swedish
Academy of Science) have operated under the illusion that it was the
Black-Scholes-Merton “formula” they actually used—we were told so. This
myth has been progressively reinforced in the literature and in business
Figure 1: The typical "risk reduction" performed by the Black-Scholes-Merton schools, as the original sources have been lost or frowned upon as “anec-
argument. These are the variations of a dynamically hedged portfolio. BSM dotal” (Merton, 1992).
indeed "smoothes" out risks but exposes the operator to massive tail events – This discussion will present our real-world, ecological understanding
reminiscent of such blowups as LTCM. Other option formulas are robust to the
of option pricing and hedging based on what option traders actually do
rare event and make no such claims.
and did for more than a hundred years.
This is a very general problem. As we said, option traders develop a
First, something seems to have been lost in translation: Black and chain of transmission of technë, like many professions. But the problem
Scholes (1973) and Merton (1973) actually never came up with a new is that the “chain” is often broken as universities do not store the acquired
option formula, but only an theoretical economic argument built on a skills by operators. Effectively plenty of robust heuristically derived
new way of “deriving”, rather rederiving, an already existing—and well implementations have been developed over the years, but the economics
known—formula. The argument, we will see, is extremely fragile to as- establishment has refused to quote them or acknowledge them. This
sumptions. The foundations of option hedging and pricing were already makes traders need to relearn matters periodically. Failure of dynamic
far more firmly laid down before them. The Black-Scholes-Merton argu- hedging in 1987, by such firm as Leland O’Brien Rubinstein, for instance,
ment, simply, is that an option can be hedged using a certain methodology does not seem to appear in the academic literature published after the
called “dynamic hedging” and then turned into a risk-free instrument, as event4 (Merton, 1992, Rubinstein, 1998, Ross, 2005); to the contrary dynamic
the portfolio would no longer be stochastic. Indeed what Black, Scholes hedging is held to be a standard operation.
and Merton did was “marketing”, finding a way to make a well-known There are central elements of the real world that can escape them—
formula palatable to the economics establishment of the time, little else, academic research without feedback from practice (in a practical and ap-
and in fact distorting its essence. plied field) can cause the diversions we witness between laboratory and
Such argument requires strange far-fetched assumptions: some liquidity ecological frameworks. This explains why some many finance academics
at the level of transactions, knowledge of the probabilities of future events have had the tendency to make smooth returns, then blow up using their
(in a neoclassical Arrow-Debreu style)3, and, more critically, a certain own theories5. We started the other way around, first by years of option
mathematical structure that requires “thin-tails”, or mild randomness, trading doing million of hedges and thousands of option trades. This in
on which, later. The entire argument is indeed, quite strange and rather combination with investigating the forgotten and ignored ancient
inapplicable for someone clinically and observation-driven standing out- knowledge in option pricing and trading we will explain some common
side conventional neoclassical economics. Simply, the dynamic hedging myths about option pricing and hedging.
argument is dangerous in practice as it subjects you to blowups; it makes There are indeed two myths:
no sense unless you are concerned with neoclassical economic theory. The
Black-Scholes-Merton argument and equation flow a top-down general 4
For instance—how mistakes never resurface into the consciousness, Mark Rubinstein
equilibrium theory, built upon the assumptions of operators working in was awarded in 1995 the Financial Engineer of the Year award by the International
full knowledge of the probability distribution of future outcomes—in Association of Financial Engineers. There was no mention of portfolio insurance and the
failure of dynamic hedging.
3 5
Of all the misplaced assumptions of Black Scholes that cause it to be a mere thought For a standard reaction to a rare event, see the following: “Wednesday is the type of day
experiment, though an extremely elegant one, a flaw shared with modern portfolio people will remember in quant-land for a very long time,” said Mr. Rothman, a University
theory, is the certain knowledge of future delivered variance for the random of Chicago Ph.D. who ran a quantitative fund before joining Lehman Brothers. “Events
variable (or, equivalently, all the future probabilities). This is what makes it clash that models only predicted would happen once in 10,000 years happened every day for
with practice—the rectification by the market fattening the tails is a negation of three days.” One ‘Quant’ Sees Shakeout For the Ages–’10,000 Years’ By Kaja
^
the Black-Scholes thought experiment. Whitehouse, August 11, 2007; Page B3.

WILMOTT magazine 73
 • That we had to wait for the Black-Scholes-Merton options formula to In the late 1800 and the early 1900 there were active option markets
trade the product, price options, and manage option books. In fact the in London and New York as well as in Paris and several other European
introduction of the Black, Scholes and Merton argument increased exchanges. Option markets it seems, were active and extremely sophis-
our risks and set us back in risk management. More generally, it is a ticated in 1870. Kairys and Valerio (1997) discuss the market for equity
myth that traders rely on theories, even less a general equilibrium options in USA in the 1870s, indirectly showing that traders were sophis-
theory, to price options. ticated enough to price for tail events.7
• That we “use” the Black-Scholes-Merton options “pricing formula”. There was even active option arbitrage trading taking place between
We, simply don’t. some of these markets. There is a long list of missing treatises on option
trading: we traced at least seven German treatises on options written
In our discussion of these myth we will focus on the practice driven
between the late 1800s and the hyperinflation episode8.
bottom-up literature on option theory that has been hidden in the dark
One informative extant source, Nelson (1904), speaks volumes: An op-
recesses of libraries. And that addresses only recorded matters—not the
tion trader and arbitrageur, S.A. Nelson published a book “The A B C of
actual practice of option trading that has been lost.
Options and Arbitrage” based on his observations around the turn of the
twentieth century. According to Nelson (1904) up to 500 messages per
hour and typically 2000 to 3000 messages per day where sent between
Myth 1: People Did not Properly “Price” the London and the New York market through the cable companies. Each
Options Before the Black-Scholes-Merton message was transmitted over the wire system in less than a minute. In a
heuristic method that was repeated in Dynamic Hedging by one of the
Theory authors, (Taleb,1997), Nelson, describe in a theory-free way many rigor-
It is assumed that the Black-Scholes-Merton theory is what made it possible ously clinical aspects of his arbitrage business: the cost of shipping
for option traders to calculate their delta hedge (against the underlying) shares, the cost of insuring shares, interest expenses, the possibilities to
and to price options. This argument is highly debatable, both historically switch shares directly between someone being long securities in New
and analytically. York and short in London and in this way saving shipping and insurance
Options were actively trading at least already in the 1600 as described costs, as well as many more tricks etc.
by Joseph De La Vega—implying some form of technë, a heuristic method
to price them and deal with their exposure. De La Vega describes option 7
The historical description of the market is informative until Kairys and Valerio try
trading in the Netherlands, indicating that operators had some expertise to gauge whether options in the 1870s were underpriced or overpriced (using
in option pricing and hedging. He diffusely points to the put-call Black-Scholes-Merton style methods). There was one tail-event in this period, the
parity, and his book was not even meant to teach people about the great panic of September 1873. Kairys and Valerio find that holding puts was
technicalities in option trading. Our insistence on the use of Put-Call profitable, but deem that the market panic was just a one-time event:
parity is critical for the following reason: The Black-Scholes-Merton’s “However, the put contracts benefit from the “financial panic” that hit the market in
claim to fame is removing the necessity of a risk-based drift from the September, 1873. Viewing this as a “one-time” event, we repeat the analysis for puts
excluding any unexpired contracts written before the stock market panic.”
underlying security—to make the trade “risk-neutral”. But one does
Using references to the economic literature that also conclude that options in
not need dynamic hedging for that: simple put call parity can suffice
general were overpriced in the 1950s 1960s and 1970s they conclude: “Our analysis
(Derman and Taleb, 2005), as we will discuss later. And it is this central shows that option contracts were generally overpriced and were unattractive for retail
removal of the “risk-premium” that apparently was behind the decision investors to purchase”. They add: “Empirically we find that both put and call options
by the Nobel committee to grant Merton and Scholes the (then called) were regularly overpriced relative to a theoretical valuation model.”
Bank of Sweden Prize in Honor of Alfred Nobel: “Black, Merton and These results are contradicted by the practitioner Nelson (1904): “. . . the majority of
Scholes made a vital contribution by showing that it is in fact not neces- the great option dealers who have found by experience that it is the givers, and not the
sary to use any risk premium when valuing an option. This does not takers, of option money who have gained the advantage in the long run”.
8
mean that the risk premium disappears; instead it is already included in Here is a partial list:
the stock price.”6 It is for having removed the effect of the drift on the Bielschowsky, R (1892): Ueber die rechtliche Natur der Prämiengeschäfte, Bresl.
value of the option, using a thought experiment, that their work was Genoss.-Buchdr Granichstaedten-Czerva, R (1917): Die Prämiengeschäfte an der Wiener
originally cited, something that was mechanically present by any form of Börse, Frankfurt am Main.
trading and converting using far simpler techniques. Holz, L. (1905): Die Prämiengeschäfte, Thesis (doctoral)–Universität Rostock
Kitzing, C. (1925): Prämiengeschäfte: Vorprämien-, Rückprämien-, Stellagen- u.
Options have a much richer history than shown in the conventional lit-
Nochgeschäfte; Die solidesten Spekulationsgeschäfte mit Versicherg auf Kursverlust,
erature. Forward contracts seems to date all the way back to Mesopotamian
Berlin.
clay tablets dating all the way back to 1750 B.C. Gelderblom and Jonker Leser, E, (1875): Zur Geschichte der Prämiengeschäfte
(2003) show that Amsterdam grain dealers had used options and forwards Szkolny, I. (1883): Theorie und praxis der prämiengeschäfte nach einer originalen
already in 1550. methode dargestellt., Frankfurt am Main.
Author Unknown (1925): Das Wesen der Prämiengeschäfte, Berlin: Eugen Bab & Co.,
6
see www.Nobel.se Bankgeschäft.

74 WILMOTT magazine
 TECHNICAL ARTICLE 3

The formal financial economics canon does not include historical the most sophisticated, according to Nelson. It could well be that World
sources from outside economics, a mechanism discussed in Taleb War II and the subsequent shutdown of option trading for many years
(2007a). The put-call parity was according to the formal option literature was the reason known robust arbitrage principles about options were for-
first fully described by Stoll (1969), but neither he not others in the field gotten and almost lost, to be partly re-discovered by finance professors
even mention Nelson. Not only was the put-call parity argument fully such as Stoll (1969).
understood and described in detail by Nelson (1904), but he, in turn, Earlier, in 1908, Vinzenz Bronzin published a book deriving several op-
makes frequent reference to a 1902 book by Higgins. Just as an example tion pricing formulas, and a formula very similar to what today is known
Nelson (1904) referring to the earlier works of Higgins writes: as the Black-Scholes-Merton formula. Bronzin based his risk-neutral
It may be worthy of remark that ‘calls’ are more often dealt than ‘puts’ the option approach on robust arbitrage principles such as the put-call pari-
reason probably being that the majority of ‘punters’ in stocks and shares ty and the link between the forward price and call and put options—in a
are more inclined to look at the bright side of things, and therefore more way that was rediscovered by Derman and Taleb (2005)9. Indeed, the put-
often ‘see’ a rise than a fall in prices. call parity restriction is sufficient to remove the need to incorporate a fu-
This special inclination to buy `‘calls’ and to leave the ‘puts’ severely alone ture return in the underlying security—it forces the lining up of options
does not, however, tend to make ‘calls’ dear and ‘puts’ cheap, for it can be to the forward price10.
shown that the adroit dealer in options can convert a ‘put’ into a ‘call,’ a Again, in 1910 Henry Deutsch discussed put-call parity but in less de-
‘call’ into a ‘put’, a ‘call o’ more’ into a ‘put- and-call,’ in fact any option into tail than Higgins and Nelson. In 1961 Reinach again described the put-call
another, by dealing against it in the stock. We may therefore assume, with
tolerable accuracy, that the ‘call’ of a stock at any moment costs the same as
parity in quite some detail (another text typically ignored by academics).
the ‘put’ of that stock, and half as much as the Put-and-Call. Traders at New York stock exchange specializing in using the put-call par-
ity to convert puts into calls or calls into puts was at that time known as
The Put-and-Call was simply a put plus a call with the same strike and Converters. Reinach (1961):
maturity, what we today would call a straddle. Nelson describes the put-
Although I have no figures to substantiate my claim, I estimate that over 60
call parity over many pages in full detail. Static market neutral delta per cent of all Calls are made possible by the existence of Converters.
hedging was also known at that time, in his book Nelson for example
writes: In other words the converters (dealers) who basically operated as mar-
ket makers were able to operate and hedge most of their risk by “statically”
Sellers of options in London as a result of long experience, if they sell a Call,
straightway buy half the stock against which the Call is sold; or if a Put is hedging options with options. Reinach wrote that he was an option trad-
sold; they sell half the stock immediately. er (Converter) and gave examples on how he and his colleagues tended to
hedge and arbitrage options against options by taking advantage of op-
We must interpret the value of this statement in the light that stan- tions embedded in convertible bonds:
dard options in London at that time were issued at-the-money (as explic-
Writers and traders have figured out other procedures for making profits
itly pointed out by Nelson); furthermore, all standard options in London writing Puts & Calls. Most are too specialized for all but the seasoned pro-
were European style. In London in- or out-of-the-money options where fessional. One such procedure is the ownership of a convertible bonds and
only traded occasionally and where known as “fancy options”. It is quite then writing of Calls against the stock into which the bonds are convertible.
clear from this and the rest of Nelson’s book that the option dealers If the stock is called converted and the stock is delivered.
where well aware of the delta for at-the-money options was approximately Higgins, Nelson and Reinach all describe the great importance of the
50%. As a matter of fact at-the-money options trading in London at that put-call parity and the need to hedge options with options. Option traders
time were adjusted to be struck to be at-the-money forward, in order to where in no way helpless in hedging or pricing before the Black-Scholes-
make puts and calls of the same price. We know today know that options Merton formula. Based on simple arbitrage principles they where able to
that are at-the-money forward and not have very long time to maturity
have a delta very close to 50% (naturally minus 50% for puts). The options 9
The argument of Derman and Taleb (2005) was present in Taleb (1997) and in O'Connell
in London at that time typically had one month to maturity when issued. (2001) but remained unnoticed.
10
Nelson also diffusely points to dynamic delta hedging, and that it Ruffino and Treussard (2006) accept that one could have solved the risk-premium by
worked better in theory than practice (see Haug, 2007). It is clearly from happenstance, not realizing that put-call parity was so extensively used in history. But
all the details described by Nelson that options in the early 1900 traded they find it insufficient. Indeed the argument may not be sufficient for someone who
actively and that option traders at that time in no way felt helpless in ei- subsequently complicated the representation of the world with some implements of
ther pricing or in hedging them. modern finance such as “stochastic discount rates”—while simplifying it at the same
time to make it limited to the Gaussian and allowing dynamic hedging. They write that
Herbert Filer was another option trader that was involved in option
“the use of a non-stochastic discount rate common to both the call and the put options
trading from 1919 to the 1960s. Filler (1959) describes what must be con-
is inconsistent with modern equilibrium capital asset pricing theory.” Given that we have
sider a reasonable active option market in New York and Europe in the never seen a practitioner use “stochastic discount rate”, we, like our option trading pred-
early 1920s and 1930s. Filer mention however that due to World War II ecessors, feel that put-call parity is sufficient & does the job.
there was no trading on the European Exchanges, for they were closed. The situation is akin to that of scientists lecturing birds on how to fly, and taking
Further, he mentions that London option trading did not resume before credit for their subsequent performance—except that here it would be lecturing them
1958. In the early 1900, option traders in London were considered to be the wrong way.
^

WILMOTT magazine 75
 hedge options more robustly than with Black- Scholes-Merton. As already Referring to Thorp and Kassouf (1967), Black, Scholes and Merton
mentioned static market-neutral delta hedging was described by Higgins took the idea of delta hedging one step further; Black and Scholes (1973):
and Nelson in 1902 and 1904. Also, W. D. Gann (1937) discusses market If the hedge is maintained continuously, then the approximations
neutral delta hedging for at-the-money options, but in much less details mentioned above become exact, and the return on the hedged position is
than Nelson (1904). Gann also indicates some forms of auxiliary dynamic completely independent of the change in the value of the stock. In fact,
hedging. the return on the hedged position becomes certain. This was pointed out
Mills (1927) illustrates how jumps and fat tails were present in the lit- to us by Robert Merton.
erature in the pre-Modern Portfolio Theory days. He writes: “A distribu- This may be a brilliant mathematical idea, but option trading is not
tion may depart widely from the Gaussian type because the influence of mathematical theory. It is not enough to have a theoretical idea so far re-
one or two extreme price change.” moved from reality that is far from robust in practice. What is surprising
is that the only principle option traders do not use and cannot use is the
approach named after the formula, which is a point we discuss next.
Option Formulas and Delta Hedging
Which brings us to option pricing formulas. The first identifiable one
was Bachelier (1900). Sprenkle (1962) extended Bacheliers work to assume Myth 2: Option Traders Today “Use” the
lognormal rather than normal distributed asset price. It also avoids dis-
counting (to no significant effect since many markets, particularly the
Black-Scholes-Merton Formula
U.S., option premia were paid at expiration).
James Boness (1964) also assumed a lognormal asset price. He derives
Traders don’t do “Valuation”
a formula for the price of a call option that is actually identical to the First, operationally, a price is not quite “valuation”. Valuation requires
Black-Scholes-Merton 1973 formula, but the way Black, Scholes and a strong theoretical framework with its corresponding fragility to both
Merton derived their formula based on continuous dynamic delta hedg- assumptions and the structure of a model. For traders, a “price” pro-
ing or alternatively based on CAPM they were able to get independent of duced to buy an option when one has no knowledge of the probability
the expected rate of return. It is in other words not the formula itself that distribution of the future is not “valuation”, but an expedient. Such
is considered the great discovery done by Black, Scholes and Merton, but price could change. Their beliefs do not enter such price. It can also be
how they derived it. This is among several others also pointed out by determined by his inventory.
Rubinstein (2006): This distinction is critical: traders are engineers, whether boundedly
The real significance of the formula to the financial theory of investment rational (or even non interested in any form of probabilistic rationality),
lies not in itself, but rather in how it was derived. Ten years earlier the same they are not privy to informational transparency about the future states
formula had been derived by Case M. Sprenkle (1962) and A. James Boness of the world and their probabilities. So they do not need a general theory
(1964).
to produce a price—merely the avoidance of Dutch-book style arbitrages
Samuelson (1969) and Thorp (1969) published somewhat similar op- against them, and the compatibility with some standard restriction: In
tion pricing formulas to Boness and Sprenkle. Thorp (2007) claims that addition to put-call parity, a call of a certain strike K cannot trade at a
he actually had an identical formula to the Black-Scholes-Merton formula lower price than a call K + K (avoidance of negative call and put
programmed into his computer years before Black, Scholes and Merton spreads), a call struck at K and a call struck at K + 2 K cannot be less ex-
published their theory. pensive that twice the price of a call struck at K + K (negative butter-
Now, delta hedging. As already mentioned static market-neutral delta flies), horizontal calendar spreads cannot be negative (when interest
hedging was clearly described by Higgins and Nelson 1902 and 1904. rates are low), and so forth. The degrees of freedom for traders are thus
Thorp and Kassouf (1967) presented market neutral static delta hedging reduced: they need to abide by put-call parity and compatibility with
in more details, not only for at-the-money options, but for options with other options in the market.
any delta. In his 1969 paper Thorp is shortly describing market neutral In that sense, traders do not perform “valuation” with some “pricing
static delta hedging, also briefly pointed in the direction of some dynamic kernel” until the expiration of the security, but, rather, produce a price
delta hedging, not as a central pricing device, but a risk-management of an option compatible with other instruments in the markets, with a
tool. Filer also points to dynamic hedging of options, but without show- holding time that is stochastic. They do not need top-down “science”.
ing much knowledge about how to calculate the delta. Another “ignored”
and “forgotten” text is a book/booklet published in 1970 by Arnold
Bernhard & Co. The authors were clearly aware of market neutral static When do we value?
delta hedging or what they call “balanced hedge” for any level in the If you find traders operating solo, in a desert island, having for some to
strike or asset price. This book has multiple examples of how to buy war- produce an option price and hold it to expiration, in a market in which
rants or convertible bonds and construct a market neutral delta hedge by the forward is absent, then some valuation would be necessary—but then
shorting the right amount of common shares. Arnold Bernhard & Co also their book would be minuscule. And this thought experiment is a distor-
published deltas for a large number of warrants and convertible bonds tion: people would not trade options unless they are in the business of
that they distributed to investors on Wall Street. trading options, in which case they would need to have a book with

76 WILMOTT magazine
 TECHNICAL ARTICLE 3

fractal distributions are such that the effect of the Central Limit Theorem
are exceedingly slow in the tails—in fact irrelevant. Furthermore, there is
sampling error as we have less data for longer periods, hence fewer tail
episodes, which give an in-sample illusion of thinner tails. In addition, the
point that aggregation thins out the tails does not hold for dynamic hedg-
ing—in which the operator depends necessarily on high frequency data
and their statistical properties. So long as it is scale-free at the time period
of dynamic hedge, higher moments become explosive, “infinite” to disal-
low the formation of a dynamically hedge portfolio. Simply a Taylor ex-
pansion is impossible as moments of higher order that 2 matter
critically—one of the moments is going to be infinite.
Figure 2: A 25% Gap in Ericsson, one of the Most Liquid Stocks in the World. Such The mechanics of dynamic hedging are as follows. Assume the risk-
move can dominate hundreds of weeks of dynamic hedging. Courtesy Yahoo! free interest rate of 0 with no loss of generality. The canonical Black-
Scholes-Merton package consists in selling a call and purchasing shares
of stock that provide a hedge against instantaneous moves in the security.
offsetting trades. For without offsetting trades, we doubt traders would
Thus the portfolio π locally “hedged” against exposure to the first mo-
be able to produce a position beyond a minimum (and negligible) size as
ment of the distribution is the following:
dynamic hedging not possible. (Again we are not aware of many non-
blownup option traders and institutions who have managed to operate in ∂C
the vacuum of the Black Scholes-Merton argument). It is to the impossi- π = −C + S
∂S
bility of such hedging that we turn next.
where C is the call price, and S the underlying security.
Take the discrete time change in the values of the portfolio
On the Mathematical Impossibility of
∂C
Dynamic Hedging π = −C +
∂S
S

Finally, we discuss the severe flaw in the dynamic hedging concept. It as-
sumes, nay, requires all moments of the probability distribution to exist11. By expanding around the initial values of S, we have the changes in
Assume that the distribution of returns has a scale-free or fractal the portfolio in discrete time. Conventional option theory applies to the
property that we can simplify as follows: for x large enough, (i.e. “in the Gaussian in which all orders higher than S2 and disappears rapidly.
tails”), P[X>n x]/P[X>x] depends on n, not on x. In financial securities, say,
where X is a daily return, there is no reason for P[X>20%]/P[X>10%] to be ∂C 1 ∂2C 2
π = − t − S + O(S3 )
different from P[X>15%]/P[X>7.5%]. This self-similarity at all scales gener- ∂t 2 ∂S2
ates power-law, or Paretian, tails, i.e., above a crossover point, P[X>x]=K
x−α . It happens, looking at millions of pieces of data, that such property Taking expectations on both sides, we can see here very strict re-
holds in markets—all markets, baring sample error. For overwhelming quirements on moment finiteness: all moments need to converge. If we
empirical evidence, see Mandelbrot (1963), which predates Black-Scholes- include another term, of order S3 , such term may be of significance in
Merton (1973) and the jump-diffusion of Merton (1976); see also Stanley a probability distribution with significant cubic or quartic terms.
et al. (2000), and Gabaix et al. (2003). The argument to assume the scale- Indeed, although the nth is is smaller font derivative with respect to S
free is as follows: the distribution might have thin tails at some point (say can decline very sharply, for options that have a strike K away from the
above some value of X). But we do not know where such point is—we are center of the distribution, it remains that the delivered higher orders of
epistemologically in the dark as to where to put the boundary, which S are rising disproportionately fast for that to carry a mitigating effect
forces us to use infinity. on the hedges.
Some criticism of these “true fat-tails” accept that such property So here we mean all moments—no approximation. The logic of the
might apply for daily returns, but, owing to the Central Limit Theorem, Black-Scholes-Merton so-called solution thanks to Ito’s lemma was that
the distribution is held to become Gaussian under aggregation for cases the portfolio collapses into a deterministic payoff. But let us see how
in which α is deemed higher than 2. Such argument does not hold owing quickly or effectively this works in practice.
to the preasymptotics of scalable distributions: Bouchaud and Potters The Actual Replication process is as follows: The payoff of a call
(2003) and Mandelbrot and Taleb (2007) argue that the presasymptotics of should be replicated with the following stream of n dynamic hedges, the
limit of which can be seen here, between t and T
11
Merton (1992) seemed to accept the inapplicability of dynamic hedging but he perhaps ⎛

n= T t 
thought that these ills would be cured thanks to his prediction of the financial world  ∂C 
“spiraling towards dynamic completeness”. Fifteen years later, we have, if anything, Lim ⎝  (S − S )
∂S S=St+(i− 1) t ,t=t+(i−1)t
^
t+it t+(i−1)t
t→0
spiraled away from it. i=1

WILMOTT magazine 77
 Such policy does not match the call value: the difference remains
stochastic (while according to Black Scholes it should shrink to 0). Unless
Order Flow and Options
one lives in a fantasy world in which such risk reduction is possible12. It is clear that option traders are not necessarily interested in probability distri-
Further, there is an inconsistency in the works of Merton making us bution at expiration time—given that this is abstract, even metaphysical for
confused as to what theory finds acceptable: in Merton (1976) he agrees them. In addition to the put-call parity constrains that according to evidence
that we can use Bachelier-style option derivation in the presence of was fully developed already in 1904, we can hedge away inventory risk in
jumps and discontinuities—no dynamic hedging— but only when the options with other options. One very important implication of this method is
underlying stock price is uncorrelated to the market. This seems to be an that if you hedge options with options then option pricing will be largely
admission that dynamic hedging argument applies only to some securi- demand and supply based14 This in strong contrast to the Black-Scholes-Merton
ties: those that do not jump and are correlated to the market. (1973) theory that based on the idealized world of geometric Brownian motion
with continuous-time delta hedging then demand and supply for options
simply should not affect the price of options. If someone wants to buy more
The Robustness of the Gaussian options the market makers can simply manufacture them by dynamic delta
The success of the “formula” last developed by Thorp, and called hedging that will be a perfect substitute for the option itself.
“Black-Scholes-Merton” was due to a simple attribute of the Gaussian: This raises a critical point: option traders do not “estimate” the odds
you can express any probability distribution in terms of Gaussian, even if it of rare events by pricing out-of-the-money options. They just respond to
has fat tails, by varying the standard deviation σ at the level of the density supply and demand. The notion of “implied probability distribution” is
of the random variable. It does not mean that you are using a Gaussian, nor merely a Dutch-book compatibility type of proposition.
does it mean that the Gaussian is particularly parsimonious (since you have
to attach a σ for every level of the price). It simply mean that the Gaussian
can express anything you want if you add a function for the parameter σ ,
Bachelier-Thorp
making it function of strike price and time to expiration. The argument often casually propounded attributing the success of option
This “volatility smile”, i.e., varying one parameter to produce σ (K), or volume to the quality of the Black Scholes formula is rather weak. It is partic-
“volatility surface”, varying two parameter, σ (S,t) is effectively what was done ularly weakened by the fact that options had been so successful at different
in different ways by Dupire (1994, 2005) and Derman (1994, 1998), see time periods and places. Furthermore, there is evidence that while both the
Gatheral (2006). They assume a volatility process not because there is nec- Chicago Board Options Exchange and the Black-Scholes-Merton formula
essarily such a thing—only as a method of fitting option prices to a Gaussian. came about in 1973, the model was "rarely used by traders" before the 1980s
Furthermore, although the Gaussian has finite second moment (and finite all (O'Connell, 2001). When one of the authors (Taleb) became a pit trader in
higher moments as well), you can express a scalable with infinite variance 1992, almost two decades after Black-Scholes-Merton, he was surprised to find
using Gaussian “volatility surface”. One strong constrain on the σ parameter is that many traders still priced options “sheets free”, “pricing off the butterfly”,
that it must be the same for a put and call with same strike (if both are and “off the conversion”, without recourse to any formula.
European-style), and the drift should be that of the forward13. Even a book written in 1975 by a finance academic appears to credit
Indeed, ironically, the volatility smile is inconsistent with the Black- Thorp and Kassouf (1967) -- rather than Black-Scholes (1973), although
Scholes-Merton theory. This has lead to hundreds if not thousands of papers the latter was present in its bibliography. Auster (1975):
trying extend (what was perceived to be) the Black-Scholes-Merton model to Sidney Fried wrote on warrant hedges before 1950, but it was not until 1967
incorporate stochastic volatility and jump-diffusion. Several of these re- that the book Beat the Market by Edward O. Thorp and Sheen T. Kassouf rig-
searchers have been surprised that so few traders actually use stochastic orously, but simply, explained the “short warrant/long common” hedge to a
volatility models. It is not a model that says how the volatility smile should wide audience.
look like, or evolves over time; it is a hedging method that is robust and con- One could easily attribute the explosion in option volume to the com-
sistent with an arbitrage free volatility surface that evolves over time. puter age and the ease of processing transactions, added to the long
In other words, you can use a volatility surface as a map, not a stretch of peaceful economic growth and absence of hyperinflation. From
territory. However it is foolish to justify Black-Scholes-Merton on grounds the evidence (once one removes the propaganda), the development of
of its use: we repeat that the Gaussian bans the use of probability distri- scholastic finance appears to be an epiphenomenon rather than a cause of
butions that are not Gaussian—whereas non-dynamic hedging deriva- option trading. Once again, lecturing birds how to fly does not allow one
tions (Bachelier, Thorp) are not grounded in the Gaussian. to take subsequent credit.
We conclude with the following remark. Sadly, all the equations,
12
We often hear the misplaced comparison to Newtonian mechanics. It supposedly pro- from the first (Bachelier), to the last pre-Black-Scholes-Merton (Thorp) ac-
vided a good approximation until we had relativity. The problem with the comparison is commodate a scale-free distribution. The notion of explicitly removing
that the thin-tailed distributions are not approximations for fat-tailed ones: there is a
the expectation from the forward was present in Keynes (1924) and later
deep qualitative difference. Another problem is that finance, unlike Newtonian mechan-
by Blau (1944)—and long a deep call, or long a call short a put of the same
ics, has never been reliable.
13
See Breeden and Litzenberberger (1978), Gatheral (2006). See also Bouchaud and
14
Potters (2001) for hedging errors in the real world—a predecessor paper to this one. See Gârleanu, Pedersen, and Poteshman (2006).

78 WILMOTT magazine
 TECHNICAL ARTICLE 3

strike equals a forward. These arbitrage relationships appeared to be well

[22] Gârleanu, N., L. H. Pedersen, and A. M. Poteshman (2006): “Demand-Based Option
known to practitioners in 1904. Pricing”. Working Paper, New York University - Leonard N. Stern School of Business.
This is why we call the equation Bachelier-Thorp. We were using it all along [23] Gatheral, J. (2006), The volatility Surface, Wiley.
and gave it the wrong name, after the wrong method and with attribution to [24] Gelderblom, O. and Jonker, J. (2003) “ Amsterdam as the cradle of modern futures
the wrong persons. It does not mean that dynamic hedging is out of the ques- and options trading, 1550–1650” , Working Paper.
tion; it is just not a central part of the pricing paradigm. It led to the writing [25] Haug E. G. (1996): The Complete Guide to Option Pricing Formulas, New York:
down of a certain stochastic process that may have its uses, some day, should McGraw-Hill.
markets “spiral towards dynamic completeness”. But not in the present. [26] Haug, E. G. (2007): Derivatives Models on Models, New York, John Wiley & Sons.
The authors would like to thank Ed Thorp and Marty O'Connell for [27] Higgins, L. R. (1902): The Put-and-Call. London: E. Wilson.
technical and historical discussions. [28] Kairys, J. P. and N. Valerio (1997) “The Market for Equity Options in the 1870s”
Journal of Finance, Vol LII, NO. 4., Pp 1707–1723.
[29] Keynes, J. M. (1924): A Tract on Monetary Reform. Re-printed 2000, Amherst New
York: Prometheus Books.
[30] Mandelbrot, B. (1997): Fractals and Scaling in Finance, Springer-Verlag.
REFERENCES [31] Mandelbrot, B. (2001a): Quantitative Finance, 1, 113–123.
[1] Auster,R. (1975): Option Writing and Hedging Strategies, New York: Exposition Press. [32] Mandelbrot, B. (2001b): Quantitative Finance, 1, 124–130.
[2] Bachelier, L. (1900): Theory of Speculation in: P. Cootner, ed., 1964, The Random [33] Mandelbrot, B. (1963): “The Variation of Certain Speculative Prices”. The Journal of
Character of Stock Market Prices, MIT Press, Cambridge, Mass. Business, 36(4):394–419.
[3] Bernhard, A. (1970): More Profit and Less Risk: Convertible Securities and Warrants. [34] Mandelbrot, B. and N. N. Taleb (2007): “Mild vs. Wild Randomness: Focusing on Risks
Written and Edited by the Publisher and Editors of The Value Line Convertible Survey, that Matter.” Forthcoming in Frank Diebold, Neil Doherty, and Richard Herring, eds.,
Arnold Bernhard & Co., Inc. The Known, the Unknown and the Unknowable in Financial Institutions. Princeton,
[4] Black, F., and M. Scholes (1973): “The Pricing of Options and Corporate Liabilities,” N.J.: Princeton University Press.
Journal of Political Economy, 81, 637–654. [35] Merton R. C. (1973): “Theory of Rational Option Pricing,” Bell Journal of Economics
[5] Blau, G. (1944–1945): “Some Aspects of The Theory of Futures Trading,” The Review and Management Science, 4, 141–183.
of Economic Studies, 12(1). [36] Merton R. C. (1976): “Option Pricing When Underlying Stock Returns are
[6] Boness, A. (1964): “Elements of a Theory of Stock-Option Value,” Journal of Political Discontinuous,” Journal of Financial Economics, 3, 125–144.
Economy, 72, 163–175. [37] Merton, R. C. (1992): Continuous-Time Finance, revised edition, Blackwell.
[7] Bouchaud J.-P. and M. Potters (2003): Theory of Financial Risks and Derivatives Pricing, [38] Mills, F. (1927) The Behaviour of Prices, National Bureau of Economic Research.
From Statistical Physics to Risk Management, 2nd Ed., Cambridge University Press. [39] Nelson, S. A. (1904): The A B C of Options and Arbitrage. New York: The Wall Street
[8] Bouchaud J.-P. and M. Potters (2001): “Welcome to a non-Black-Scholes world”, Library.
Quantitative Finance, Volume 1, Number 5, May 01, 2001 , pp. 482–483(2). [40] O'Connell, Martin, P. (2001): The Business of Options , New York: John Wiley & Sons.
[9] Breeden, D., and R. Litzenberger (1978): “Prices of State-Contingent Dlaims Implicit in [41] Reinach, A. M. (1961): The Nature of Puts & Calls. New York: The Book-mailer.
Option Prices”, Journal of Business 51, 621–651. [42] Rubinstein M. (1998): Derivatives. www.in-the-money.com.
[10] Bronzin, V. (1908): Theorie der Prämiengeschäfte. Leipzig und Wien: Verlag Franz [43] Rubinstein M. (2006): A History of The Theory of Investments. New York: John Wiley & Sons.
Deticke. [44] Ruffino, D. and J. Treussard (2006) “Derman and Taleb’s ‘The Llusion of Dynamic
[11] Cootner, P. H. (1964): The Random Character of Stock Market Prices. Cambridge, Replication’,: a Comment”, Quantitative Finance, 6, 5, pp 365–367.
Massachusetts, The M.I.T. Press. [45] Samuelson, P. (1965): “Rational Theory of Warrant Pricing,” Industrial Management
[12] De La Vega, J. (1688): Confusión de Confusiones. Re-printed in the book: Review, 6, 13–31.
Extraordinary Popular Delusions and the Madness of Crowds & Confusión de [46] Sprenkle, C. (1961): “Warrant Prices as Indicators of Expectations and Preferences,”
Confusiones edited by Martin S. Fridson, 1996, New York: Wiley Publishing. Yale Economics Essays, 1(2), 178–231.
[13] Derman, E. and I. Kani (1994) “Riding on a Smile”, Risk 7, 32–39. [47] Stanley, H.E. , L.A.N. Amaral, P. Gopikrishnan, and V. Plerou, 2000, “Scale Invariance
[14] Derman, E. (1998), “Stochastic Implied Trees: Arbitrage Pricing with Stochastic Term and Universality of Economic Fluctuations”, Physica A, 283,31–41.
and Strike Structure of Volatility”, International Journal of Theoretical and Applied [48] Stoll, H. (1969): “The Relationship Between Put And Call Prices,” Journal of Finance,
Finance 1, 61–110. 24(5), 801–824.
[15] Derman, E., and N. N. Taleb (2005): “The Illusion of Dynamic Delta Replication,” [49] Taleb, N. N. (1997): Dynamic Hedging. New York: John Wiley& Sons.
Quantitative Finance, 5(4), 323–326. [50] Taleb, N. N. (2007a): The Black Swan. New York: Random House.
[16] Dupire, B. (1994): “Pricing with a Smile”, Risk 7, 18–20. [51] Taleb. N. N. (2007b): “Scale-Invariance in Practice: Some Questions and Workable
[17] Dupire, B. (1998): “A New Approach for Understanding the. Impact of Volatility on Patches”, working paper.
Option Prices”, Discussion paper Nikko Financial Products.
[52] Thorp, E. O. (1969): “Optimal Gambling Systems for Favorable Games”, Review of the
[18] Dupire, B. (2005): “Volatility derivatives modeling”, www.math.nyu.edu/ carrp/
International Statistics Institute, 37(3).
mfseminar/bruno.ppt.
[53] Thorp, E. O., and S. T. Kassouf (1967): Beat the Market. New York: Random House.
[19] Filer, H. 1959: Understanding Put and Call Options, New York: Popular Library.
[20] Gabaix, X. , P. Gopikrishnan, V.Plerou & H.E. Stanley,2003, “A Theory of Power-Law [54] Thorp, E. O. (2002): “What I Knew and When I Knew It—Part 1, Part 2, Part 3,”
Distributions in Financial Market Fluctuations”, Nature, 423, 267–270. Wilmott Magazine, Sep-02, Dec-02, Jan-03.
[21] Gann, W. D. (1937) How to Make Profits in Puts and Calls, Pomeroy, WA: Lambert [55] Thorp, E. O. (2007): “Edward Thorp on Gambling and Trading,” in Haug (2007).
Gann Publishing Co. [56] Thorp, E. O., and S. T. Kassouf (1967): Beat the Market. New York: Random House.
W

WILMOTT magazine 79