Financial markets are often compared to casinos. This parallel
was first made by John Maynard Keynes and is still repeated in
the media today. It is, in our view, from a historical point of
view that the comparison is with financial risk management – at
least until 1952. A shared history so far links financial risk
management and gambling.

Probabilities, a very recent invention »

The notion of quantifiable risk, now at the heart of finance,
dates back to the mid – 20th century-long after financial
markets began to flourish. It is based on the concepts of
probability, which curiously did not emerge until the 16th
century, thanks to the study of games of chance.

However, gambling existed long before our era. The ossicles,
precursors of the dice, are painted on Greek vases and the walls
of Egyptian tombs dating from 3,500 BC. Since the ossicles do
not have the same size faces, it would have been difficult to
calculate probabilities from this instrument.

The Greeks, however, would have used dice, both for divination
and for playful purposes. Despite their advanced mathematical
knowledge, they have never been interested in probabilities,
which today form the basis of risk management. Bernstein
attributes this to the fact that the future was at the time the
reserved domain of the gods. To know, we went to see an oracle
and not a philosopher or a mathematician (often the same
person).

Similarly, the Greeks maintained a love for purity and
mathematical demonstration. The probabilities hardly correspond
to this characterization. Today’s mathematicians have
always tended to view chance as an unenforceable field of study.

It was not until the 16th century that this branch of
mathematics appeared, and the research of the Milanese physician
Girolamo Cardano. It is in the context of his addiction to
gambling that he develops an analysis based on probabilities. It
is funny to see that he specifies in his book, ” if the
dice are not stacked.”

Later, probabilistic research is often carried out as a more or
less playful “sidebar” by scientists of the time in
response to requests from hard-core players. Galileo thus worked
on the subject of probabilities at the request of his employer
The Grand Duke of Tuscany, Cosimo II-according to some
historians without much pleasure. In France, the Chevalier de
Méré will get in touch with the famous Blaise Pascal and Pierre
De Fermat to find out how to divide the winnings when a game of
chance is stopped during the game.

It is from Daniel Bernoulli that the notion of risk appears.
Again, it is due to a hypothetical Game of Chance, called the
St. Petersburg Paradox. Let’s imagine the next coin toss.
You bet an initial bet, which we cash in. We throw a coin. If
the face appears, we’ll pay you $ 1, and we’ll stop
playing. If not, we’ll raise the currency. If the face
looks, we pay€ 2, and we end the game. If not, we’ll build
the coin. If the face appears, we pay you 4€, and so on.

So your expectation of gain is infinite. How much would you be
willing to pay to play this game?

Daniel Bernoulli postulated that a reasonable man would not be
willing to put more than 20€ in this game (in the first paradox,
the amounts were in ducats). His idea is: between receiving 1€
or 2€, there is a big difference – much more significant than
between 1 000 and one 001€. Following the same reasoning, most
people will, therefore, prefer to keep their 20€ than to flip a
coin for a potentially infinite win.

Sort of ” better one yours than two you’ll
get.” The idea of quantifiable risk was born – humans do
not like variability.

Funnily enough, this age-old probabilistic enlightenment
research using gambling has left its mark today. Probabilities
are always taught in the form of games of chance.

Even more impressive, “games of Chance” are still
used today in research, to understand the attitude of
individuals towards risks. They are always a preferred method of
studying risk.

A lottery to test “temperance,” a feature of risk
aversion (Nousair et al., 2014). The left option is less risky.
Noussair et al. Two thousand fourteen

Risk management in finance today

The history of financial risk management and gambling is very
clearly separated in practice in the second half of the 20th
century – after the death of Keynes in 1946. One may, therefore,
wonder what Keynes would think today of the profound
mathematization that revolutionized finance after his death.

Before the second half of the 20th century, decisions in market
finance were made without the use of generally accepted
mathematical models. It is, therefore, generally intuition,
“the gut,” or heuristics of choice, more or less
arbitrary, that prevails.

The French Louis Bachelier made the first attempt at
mathematization in 1900. His thesis received a very mixed
reception by the famous Poincaré. It is the fact that this young
mathematician attacks a field far removed from traditional
mathematics which condemns him to a weak recognition. Paul
Samuelson returns to this line of research in the 1960s and
shows that stock prices follow a random pattern, implying that
the distribution of returns is (almost) normal. The recognition
of Louis Bachelier– who also died in 1946-was thus posthumous.

But it was in 1952 that a real revolution in financial risk
management took place. Based on the idea that variability is a
thing to be avoided and the expectation of gain a something to
be maximized, Harry Markowitz defines the problem of Management
in his thesis as a simple problem of optimization. Modern
finance was born. Funnily, Milton Friedman, who was on Harry
Markowitz’s Jury and who would later win the Nobel Prize
in Economics, told him in his defense :

“Harry, this is not an economics thesis, and we
can’t give you a Ph.D. in economics for something that
isn’t economics. This is not math, and this is not the
economy; it is not even the management “.

“Harry” did get his Ph. D. and the Nobel Prize in
economics for his work. These are now part of the economic and
financial fundamentals.

The following decades saw the rise of mathematical models in
risk management, together with the development of computer
science. Risk perception, measurement, and control are
incredibly dynamic areas of research today. Increasingly
multidisciplinary research-sometimes involving mathematicians,
managers, psychologists, and neuroscientists – could lead to new
leaps forward in understanding the risk for both what it is and
what it means to investors.

Perhaps a final word is needed on the difference between market
finance and casino. First, there is an economic interest in
investing – this money is useful to the economy, which is not
the case with a casino bet. Indeed, it should not be forgotten
that the issue of securities by companies, financial statements,
and other economic agents allows them to finance themselves. The
gain on this investment is linked to the first prediction of the
state of the economy in general and of one or more companies in
particular in the future, and thus rewards a generation of
information useful to society. Finally, statistically, a
diversified portfolio of several securities will have positive
long-term earnings expectation. A casino offering the same
services would go bankrupt.