Behavioral finance FAQ / Glossary (Rare)
This is a separate page of the R section of the Glossary
Related message(s) in the
Behavioral-Finance group (*)
Year/month, d: developed / discussed,
(risk of) Rare events
See fat tails, numeracy bias,
memory, small numbers,
scenario, extreme, peso
Damocles sword? Or exclusive bonanza?
About rare events, do not dream too much about the nice ones,
And do not exclude the dangerous ones from your nightmares!
They happen only when they occur
(Pierre Dac, a French humorist)
A popular allegory for extremely rare "singularities" is "the 100 year storm".
In statistics and probabilities , rare events (aka black swans):
Are usually not visible in short historical series, which
can make most people believe ...that they will never happen.
Apart some anecdotal minor ones, are usually
excessive in their scope and effects (for example Himalayan or
abyssal market prices),
Average-sized events or beings are rarely ...rare, except in
populations made of only dwarfs or giants.
But already occurred once or a few times
in the past,
And/or are practically certain to occur one day, and maybe more
often than the "normal law" of distribution (see distribution)
Is it important?
Rare disasters are disasters anyway. Big ones sometimes!
Nobody will consider rare / singular events a problem...
when they are nice events.
But some of them might be real disasters.
Rare events, and among them the most exceptional ones, are an important
In "dynamical systems" (see that phrase), they are common, notably in
Mother Nature and in human societies.Here, rare events tend to strike
more often than common distribution laws predict.
Those strange occurence:
Either cannot be predicted at all, except as fantasies (might white
elephants be also black swans?).
=> Some creative thinking is needed to make "improbable but
possible" future scenarios.
Or can be predicted via long time statistics, but some cognitive factors
make them usually overlooked and underestimated (see "extreme"
and "fat tails") or (less often) exaggerated.
If most people think that according to the "normal law",
a specific disaster can take place once in a billion years,
but if actually it has a good chance to occur
in the next five years, we have a problem!
Not to include them in prediction models and in strategic planning, make
that worst case scenarios, plan B and other precautions against contingencies
are swept under the carpet.
Economics and finance are among the human activities that can be
affected by such episodic phenomena.
Some crucial cases
Better be ready for worst case scenarios!
It does not mean you cannot settle on a volcano to grow beans on
its fertile lava,
But be aware of the danger, decide if it is worth the risk, and
prepare for contingencies!
To anticipate rare events (including some that never happened but seem
possible in extreme scenarios) and simulate them to be ready, is crucial in
various fields of activities.
This is the case for example for airplane crashes, oil tanker catastrophic spills,
Emergency plans must be ready in advance whatever the rare
The economy, as well as finance, had their share of rare events all
along their history, and they can be disastrous:
Financial markets offer a well-known example:
stock crashes .
They are far from being unpredictable, even if you cannot know the
However rare they are, nobody can say that the last one was, well,
...the last one.
But when an uptrend persists for a while a common belief that a crash
cannothappen grows and spreads.
This is because
* Many people get accustomed to (or envious of) its benefits
* Market guru get accustomed to the efficient market dogma that
considers that market prices are based on fair values.
The economy also is prone to some unanticipated crises.
The possibility of "systemic" (= generalized to all institutions
and players) economic or financial catastrophes should be kept
One cause could be a sudden liquidity squeeze
(see that phrase).
The rare events trap #1: too short statistics
Rare events can be more frequent than recent measures show, and than
Less rare than rare?
In various human and social areas, rare events are not that "rare".
In statistical distributions (if the sample is big enough and covers enough
time), they can be seen :
* under the shape of "fat tails"
* as well as extremely "long tails" (see those phrases).
Some economist consider that present social evolutions favor some "law
of extremes" (see extreme).
In other words there is a tendency to extreme economic discrepancies,
although they might be only transient phenomena in the general evolution.
Observers might not see those transient / rare (or more decisive) phenomena
on short time data series on stock price volatility for example.
Therefore to rely only on recent data can entail:
False conclusions about the real risks (for investors in the
An omission to integrate rare events into mathematical economic /
financial models (see "model"), making them imperfect and dangerous.
The rare events trap #2: just neglect
Ignoring rarity ...at your own peril
A more mundane hidden trap is that most people do not imagine those rare
events could happen.
They do not anticipate them as they never experienced them,
or, if they have seen them, because their memory (see that word) faded.
Thus they are overconfident.
They neglect to take precautions , to get
prepared (protection, escape road), even when signals
of possible trouble appear.
=> They get trapped when the situation turns bad.
=> Or they are not ready to seize the opportunities
they might bring.
Rationality commands to have a safety margin and contingency plans
in prevision of totally destructive dangers, however infrequent they
The opposite trap: pervasive small risk fright
Focusing on the bad trees in a rich forest.
On the other hand, individuals and societies can suffer from (or be manipulated
into) paranoid frights against risks that have very small negative
effects, or even against invented dangers. Booh!
Also, the prospect theory (see that phrase) shows that the loss aversion curve
is very steep near the "reference point", therefore for small losses.
Any loss, however small, has usually an emotional impact
This dramatization of losses explain also
- millenary panics based on flimsy analyses,
- or to a smaller degree, abusive uses of the "precautionary
principle" that paralyzes innovation, initiative and scientific quests.
In contrast, as said above, there is some rationale, while accepting to
take risks, in not neglecting to make some preparation against
identifiable identified infrequent albeit extreme dangers.
(*) 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
This page last update: 09/08/15