Behavioral finance FAQ / Glossary (Risk)
This is a separate page of the R section of the Glossary
Dates of related message(s) in the
Behavioral-Finance group (*):
Year/month, d: developed / discussed,
00/9i,10d - 01/4i,9i - 02/10i ,11i
- 03/1i,9i - 04/1i,10i - 05/1i,6i,8i
- 07/1i - 08/3i + see uncertainty,
volatility, semi-volatility, stochastic,
random, probability, distribution
Risk as arithmetic
A risk is a random event
That could bring a damage
And which occurrence can be quantified in
* Its frequency
* The size of the loss it could bring.
In other words, a risk can be measured, or at least
estimated,thanks to a probability law or statistical
pattern, at the difference of a pure uncertainty
(see that word).
When money is involved (financial risk), risk is a
quantified loss possibility that
can affect an asset value or a financial operation.
The old china from aunt Martha might break !
The menace addressses usually a downside price possibility but also, in
some operations - short selling - for example, an upside price possibility.
A coffee break before going on:
contemporaneous attempts to tame risk
To buy and to sell risk.
Modern finance has packaged risk into mathematical objects,
based on purportedly measurable and objective probabilities.
One purpose is to change it from a nuisance to a
tradable commodity / asset..
A good intention, don't you think: transferring the risk to
somebody happy to take it... for a price. But are good
intentions always so good (see below)?
Beware, the tradable package can have some leaks,
and get sometimes punctured.
Also, a risk-based financial instrument might not be
transparent about what is stacked inside.
Opaque financial jars! True science or esoteric rites playing
The nuisance might reappear in an unexpected way. The
idea that financial risk can be fully mastered, nearly
eliminated, by sophisticated methods has been proved
That was was seen in the 2007 - .... financial crisis, because of
a misuse, even a delinquent use, of financial techniques such as
debt securitization and credit derivatives, encouraged by a lax
monetary policy (near-free money).
To play on the paradox, those flaws in the system
(and in some theoretical rationalizations) clarify things:
They show that risk can survive in its rawest form:
Yes, social, economic and financial life (and life in
general) will keep being an adventure, as risk (and
uncertainty, its cousin) will survive whatever is
Imprison risk in bottles of highly mathematically
packaged products of financial engineering.
They might fly well, but also explode in flight.
Or avoid it by using an illusory and regressive (but
politically correct) "precautionary principle", or a
range of bureaucratic rules that replace anticipation,
intelligence, independence and courage by those in
charge of monitoring.
Risk and finance
Risk is the game.
Risk, as the possibility of a monetary loss, applies, among other economic
fields,to financial assets and operations.
All assets are risky, in the sense that their price and
return tend to vary more or less dangerously.
Thus the two mainfinancial valuation parameters are
summed up in the risk / return duet.
A financial risk can be understood either as:
The probability (x %) of a negative outcome (see probability,
There can be some illusion here. Contrarily to the platitude,
finance is not a "casino" (a strictly organized place in which
probabilities are known with certainty).
In finance, probabilities, even when presented as scientific, are
just tentatively deduced, usually from past statistics,
whatever those data are worth as regards the future.
Or the "mathematically expected" loss
(the monetary sum of the amounts of possible future losses
each one being multiplied by its probability)
to be compared of course to the expected gains, using the same
Or the maximum possible loss (for ex. a 100% fall in asset prices,
Or the maximum acceptable loss to avoid "total ruin"
Asset markets, and assumptions
about financial risk in those markets
The market risk is supposed to be known
...until you meet it in person.
To buy or sell an asset is to buy or to sell
an expectation of risk and return.
Normally, when you buy capital properties, their present state is only an
What you buy actually are:
The future revenues you expect from those assets, including their future
And the risks of loss they entail.
Assets prices are supposed to mirror the average expectations by all
Also, asset price variations are supposed to measure how those average
That risk / reward ratio is supposed to help in building models of asset
pricing, trading or portfolio diversification.
For this purpose, two assumptions are used
about financial markets:
A supposed correlation between an asset's risk
and its return (higher risk = higher return, see risk premium).
The idea that risk is well-known statistically (using volatility as
a proxy for risk, as shown below).
This is at least partly an illusion that relies too much on standard
distribution laws, while real market data distributions show many
It neglects also the fact that uncertainty (non quantified risk) exists
in all human and social activities as well as in the future value of
Asset risk parameters
Volatility as a practical, but reductive, heuristic
The asset risk parameters differ according to what the financial models is used
But most of them use stochastic (see that word) financial calculations. Their
main (and not always right) assumption is that risk is a purely random
phenomenon, obeying clear random laws.
Here are the most usual financial market risk parameters:
A common (but far from reliable) risk quantification criterion
for asset prices is past volatility.
The use of historical volatility as a proxy for risk is an heuristic
Thus, it might be illusory and verge on a numeracy bias
(see that word).
It is not fully rational to to extrapolate future risk from data
on previous risk, whatever those data are: historical
randomness, past mean-variance, beta...
Also the tenet under which the risk / volatility follows a standard
random law underestimates the risk. It neglects distributions
(fat tails, clusters, asymmetries...) and extreme scenarios, that
only the human mind can imagine as they do not appear in past
Another approach is to take the absolute maximum price variation
seen in the past (extreme risk).
This is closer to the full real risk, but the numeracy bias can
strike here also (see law of "small numbers", "fat tails",
An intermediate solutions is to use a "VaR / Value at risk" model.
Another one is to focus on downside volatility
Last but not least, the CAPM (see that acronym) offered its magical
toolbox with the systematic / specific risk parameters (see
But is risk only a matter of parameters?
Of course, the base rate probabilities (see that pharase) should always
kept in mind when past data exist and are reliable. But be conscious that a
cozy measurable "risk" drawn from past figures might give a false
The danger might strike when the current situation does not fit past
cases, something which happens often in evolutionary systems (see
dynamical systems) where uncertainty dominates.
Also do not exclude extreme risks never seen or rarely seen before
which therefore do not really appear in limited satistics but which
consequences could be dramatic.
What is at play is real uncertainty, given the
whole range of scenarios of good
and bad events that can be built.
Be the scenarist!
Past statistics are crucial, but project you mind into the future.
This is where risk resides, sorry for the platitude!
See rare events
(Specific / systematic) Risk
See risk, beta, CAPM
According to the CAPM (see that acronym),
The systematic risk is the risk, measured by the beta
coefficient (see that word), that is correlated to the whole
market (market index) risk.
It can be reduced by acquiring a portfolio that represents closely
enough the whole market efficient diversification).
The specific risk is the portion of risk of the individual
asset that is not correlated to the market portfolio.
Risk attitude, aversion, preference, profile...
See specific page
Risk perception, perceived risk
See specific page
Risk premium, risk premia puzzle
See specific page
(*) To find those messages: reach that BF group and, once there,
1) click "messages", 2) enter your query in "search archives".
Members of the Behavioral Finance Group,
please vote on the glossary quality at BF polls