Concluding Thoughts

When I started this blog (way back on that other page), I finished the intro with questions on why and when bubbles start and burst and whether or not they will continue to occur, before setting off on a meandering journey through the social sciences.  To summarize my findings I would say that:


Given the power of social proofs , it seems highly likely to me that herding  plays a major role in how a bull run becomes a bubble.  As speculators attempt to suss out the Keynesian beauty contest , members of the general public begin to enter the market.  Their lack of knowledge makes them vulnerable to miscalibrated  experts (mavens ), messages in the media (connectors ) and talented salesmen  - collectively, the 20% of the population who do 80% of the work , according to Gladwell.


After the tipping point , the bubble, fuelled by our built-in inability to adequately frame context , has the big mo  on its side.


But eventually, the biggest investors start to get anxious, remembering the backwards induction argument (from the Brunnermeier video earlier on) and realising that no-one knows when this bubble is going to burst.  They decide to liquidate their position.  The connectors  start to spread the word and the salesmen  suddenly stop offering 125% mortgages and the whole process reverses.

POP!

Why did this whole thing started in the first place?  It could well have been a big act of risk homeostasis  - maybe insurance has suddenly taken on a major role (like before The South Sea Bubble in 1720) or Black and Scholes have developed this great new way of pricing options contracts that mean institutions have more cash available or maybe JP Morgan have developed a brand new derivative instrument that enables securitization (just like prior to the housing bubble in the early 21st century).


This, of course, is re-enforced by centuries of human intuition that leads us all to think that because we can't remember an example  of when this kind of disaster happened before, there's probably not much chance of it happening again!


This all suggests that once enough time has passed after the 2007 housing market crash, investors will convince themselves that the next major bubble is in fact not a bubble at all, screening all available information with another process that has served mankind well for thousands of years - confirmation bias .


Forever blowing bubbles?  I think that this is somewhat inevitable, unless of course the financial world stops ignoring an entire branch of science that contains countless examples supporting the idea that we are only rational within limits; contradicting the very foundation of economic theory - efficient markets hypothesis .


To wrap up, I'll leave you with another couple of quotes from Dan Gardner's book:

"We are only 'boundedly rational' - rational within limits.  The fact that we are often rational is the reason why conventional economics often works; the fact that there are limits to our rationality is the reason why conventional economics occasionally falls on its face"

"It's easy to fall into dispair.  If the economic turmoil we are experiencing is rooted in psychology, what can we do about it?  Human nature can't be changed.  Is there nothing we can do to lessen our vulnerability to economic mania?

In fact, there is.  Human nature may be unchangeable but much of the environment in which people live is of our own making.  It can be changed.  'If we want to prevent things like this from happening again, we have to create tools that are designed for humans and not Homo economicus,' says Cass Sunstein, whose book Nudge, co-written with Richard Thaler, is about how to do just that"

A Word from Daniel Kahneman...

Miscalibration

Miscalibration is the technical term used to desribe overconfidence. One of the most notable studies showing miscalibration was coordinated by Stuart Oskamp in 1965. The purpose was to show the possible consequences of perfect information. Oskamp's study examined the responses of a panel of expert psychologists when presented with a sample case study.

Dr. Stuart Oskamp

In the initial stages, the psychologists were provided with a short summary detailing the condition of one of Oskamp's patients, an army veteran named Joseph Kidd. After reading this summary paragraph, the experts were asked to diagnose Kidd's condition and answer a short questionnaire. In addition, they were also asked to predict their own performance on the questionnaire.



After this stage was complete, Oskamp provided the members of the panel with two pages of additional information about Kidd and asked them to repeat the process.  After this stage, another two pages of information were provided and then finally the whole patient file.



Logically, you would expect that the more information an expert is provided with, the more accurate their results. However, Oskamp found that, over all stages of the process, the performance of the experts on the diagnosis questionnaire improved by a mere 1% from 25% to 26%. This suggests that we are mislead in the value of additional information when attempting to solve a difficult problem.



Another interesting discovery was the change in the experts' prediction of their own scores - 25% initally, then 40%, then 50% and finally, closer to 100% when the whole patient file was available. The moral of the story is that additional information does not improve the results of experts when confronted with a difficult problem but does improve experts' confidence in their results.



In a world were we have access to more information than ever before, overconfidence is rife among experts in the financial sector. And when these experts are then given a podium and are allowed to broadcast to the nation, they are given the opportunity to sell their idea, which could spread through the population until it reaches a tipping point and becomes a commonly held belief. An example of this came in 1997 when Chancellor of the Exchequer, Gordon Brown stated in a budget statement that "Britain will not return to the boom and bust of the past".   This statement would later enforce the belief that the UK property market was not experiencing a bubble and would continue to rise in the future. This was overconfidence at the time and in retrospect, it just seems outright ridiculous but it's entirely plausible that it had a part to play in further inflating the UK property bubble. I think it's also worth noting the importance of confirmation bias in re-enforcing a belief like that.

The Example Rule

The example rule has profound implications for the way we think about risk. Simply put, it suggests that if we can easily recall an example of something, then the perceived risk is greater.



It is easy to see how this rule survived the process of natural selection. Imagine a caveman walking in a forest.   He sees another member of his tribe mauled by a bear. It serves that caveman well to remember the incident, as he stands a better chance of survival if he avoids that part of the forest. Conversely, if he knows of another part of the same forest where he has no example of anyone being attacked, the odds are that this part of the jungle is safer.

Levels of Earthquake Risk - California

Modern day examples include studies of insurance rates in earthquake prone areas such as California. Research shows that demand for (and price of) insurance is greatest immediately after an earthquake. This is because people can easily recall an example and so judge risk to be higher.  Similarly, the price of insurance decreases over time, as the last earthquake becomes a distant memory.




But this is inherently irrational. These people are thinking with their gut. If they thought with their heads, they would realise that statistically speaking, the chance of back-to-back earthquakes occurring is low and it makes sense to wait until the price comes down.  If this market behaved rationally, we would see insurance rates rising when there has been a long interval between earthquakes; surely another one is due soon.




Applying this rationale to my theme, should enough time pass between bubbles such that investors cannot quickly recall an example of the last one and the devastation it caused, their assessment of the risk involved may be greatly impaired.   Combine this effect with that of confirmation bias, traders could conceivably convince themselves that there isn't a bubble in the first place and the result is highly risky behaviour, contributing to the expansion of a financial bubble.

Two Heads are Better than One



René Descartes

The workings of the human mind have posed puzzling questions for many great thinkers, from the great philosophers of ancient Greece, to Descartes who hypothesised that the body and mind are separate.  In more recent times, philosopher Gilbert Ryle mocked Descartes' idea, calling it "the ghost in the machine".  In the six decades since, science has made great progress in understanding how humans think and this progress supports Ryle.  Current research suggests that the brain is an entirely physical entity, subject to the process of natural selection.  This discovery has lead to a new field within social science - evolutionary psychology.







In his book "Risk: The Science of Politics and Fear" , Dan Gardner provides an in-depth commentary on the latest theories on the human mind.  He explains how the brain is essentially divided into two systems.  The first system is subconscious, what we commonly call "gut instinct".  Gardner describes this system as "the more ancient.  It is intuitive, quick, and emotional".  The second  system is the one referred to when someone tells you to "use your head".  This is conscious thought.  Gardner calls it "calculating, slow and rational".








The rest of the book is then focused on the problems that arise as this ancient, evolved way of thinking struggles in modern society.  I think, of the features associated with this system, two in particular are worth mentioning, so I'll introduce one now and the other in subsequent posts.




Confirmation Bias



Once a belief is in place, we screen everything we see and hear in a biased way, paying attention only to the information that serves to strengthen our beliefs.




The evolutionary rationale for confirmation bias is that the collection of examples not only strengthens our beliefs but makes us better at arguing the point and convincing others.  But this trait leaves us somewhat blind to the logic of counterexamples.  Try it yourself:



Is the following statement true?

"No C are B; all B are A; therefore some A are not C."

Apparently, fewer than 10% of people figure out that it is.  Interestingly enough, the problem seems much easier when A, B and C are swapped for something more real:

"No flying things are penguins; all penguins are birds; so some birds cannot fly."

If everyone is prone to confirmation bias  then it follows that financial traders may hold beliefs that are irrational (or were once rational but aren't any longer) and convince themselves that they are perfectly rational.  Such beliefs could then influence behaviour and this kind of behaviour could most certainly contribute to a bubble.

Risk Homeostasis

Risk homeostasis  is a theory on the evaluation of risk, developed by Queen's University (Ontario, not Belfast!) professor Gerald J.S. Wilde .  The basic idea is that when a way to reduce risk is introduced into a system, something inherent in human nature leads us to take greater risks in another area of that system.  


To illustrate the point, let's take a look at Wilde's now famous 1990s study of taxi-drivers in Munich, where  ABS (anti-lock braking system) was fitted to half of the cars within a fleet of taxis and at the end of the study, the crash rates of the two sets of taxis were compared. 

To briefly outline the impact of ABS, in June 1999, the National Highway Traffic Safety Adminstration (NHTSA) found that, on average, ABS increased stopping distances by 22% (on a gravel surface).  That 22% could be the difference between having an accident and escaping with a near miss, so you would think that the taxi drivers with ABS in their cars would have a lower crash rate but this wasn't to be the case. 



The crash rate remained the same for both sets of drivers, suggesting that because drivers knew they had better brakes, they could drive faster, tailgate other cars and rely on their brakes to get them out of trouble.  This is risk homeostasis - increasing the riskiness of your driving because risk is reduced in another area; in this case, by the braking system.



And if you think about it, we all do it!  Look out for it next time you're crossing the street.  If you're at a busy pedestrian crossing, odds are someone standing beside you will be texting a friend whilst listening to their i-pod.  They rely on the "green man" to get them across the road safely, paying no attention to the flow of traffic.  This is another example of risk homeostasis.  How often do you see someone cross the road at a place where there is no crossing, behaving in the same way?  If the same care and attention was displayed crossing the road in both instances, accidents at pedestrian crossings would be virtually non-existant but unfortunately, it isn't and they aren't.  In 2005 for example, the AA found that fatalities on pedestrian crossings accounted for 2.2% of total road fatalities in the UK.
http://www.theaa.com/public_affairs/reports/aa-pedestrian-crossings-survey-in-europe.pdf



I think that this result is an important one for the field of finance, particularly in wake of the 2008 financial crisis.  The development of the mortgage backed security (MBS) and the credit default swap (CDS) meant that financial institutions could hedge their risk through securitization.  But when the risk in this area was reduced, what happened?  The development of sub-prime mortgages; institutions lending money to people who were previously deemed too high-risk to be considered for mortgages, and the fuelling the property bubble in the US.

For a more detailed read on the Munich taxi-driver study and risk homeostasis  in general, I'll refer you to Malcolm Gladwell  once again!

Shopping for Jackets and Calculators

In 2002, Daniel Kahneman (a psychology professor at Princeton University) was awarded the Nobel prize for his work (with Amos Tversky) in behavioural economics.  Their work sparked the new discipline of prospect theory  - a theory that suggests that people do not always make the optimal decision when it comes to risk.




The following example is quoted in their journal article "The Framing of Decisions and the Psychology of Choice"  published in Science  in 1981:


Imagine that you are about to purchase a calculator for $15.  The salesman informs you that the calculator you wish to buy is on sale for $10 at another branch, located 20 minutes drive away.  Would you make the trip to the other store?


When Tversky and Kahneman posed this question, 68% of respondants said they would make the trip. 


The pair then rephrased the question:



Imagine that you are about to purchase a Jacket for $125. The salesman informs you that the jacket you wish to buy is on sale for $120 at another branch, located 20 minutes drive away. Would you make the trip to the other store?

Logically, this is the same problem as the calculator - a saving of $5 is there to be had.  But the results in this case showed that only 29% were willing to make the trip.





In the same paper it is noted that it is the percentage saving that customers tend to take into consideration:



"...the variability of the prices at which a given product is sold by different stores is roughly proportional to the mean price of that product."



and


"Individuals who face a decsion problem and have a definite preference (i) might have a different preference in a different framing of the same problem, (ii) are normally unaware of alternative frames and of their potential effects on the relative attractiveness of options, (iii) would wish their preferences to be independent of frame, but (iv) are often uncertain how to resolve detected inconsistencies."



What they are saying is that context  influences consumer behaviour.  If the same theory is applied to financial markets, we would surely see traders paying over the odds when the orders they place are of considerable size.  Or indeed, house buyers wrapping up negotiations early when they could have squeezed a saving of a couple of thousand dollars out of the deal.  Such behaviour would exacerbate the gap between asset price and intrinsic value, expanding the bubble.

The Power of Context

John Darley and Daniel Batson were two Princeton University psychologists who devised one of the most well-known social experiments of modern times, using the parable of The Good Samaritan to illustrate their point.

The experiment involved a group of theology students who were asked to prepare a seminar and then walk over to a nearby building to present.  Along the way, each student would pass an actor, slumped in an alley, pretending to be in distress.  The purpose of the experiment (unknown to the participants) was to see whether or not the students would stop.

Darley and Batson also split the group of students in two.  The first group were informed “it will be a few minutes before they’re ready for you, but you might as well head over now.” 

The second group however, were given a starkly different send off - “Oh, you’re late. They were expecting you a few minutes ago. We’d better get moving.”

The results of this study showed that of the first group, 63% of students stopped to help the man.  Of the second group, only 10% stopped to help.  In fact, in one instance, a student was actually on his way to give a talk on the parable of the Good Samaritan and actively stepped over the man in distress to get there on time!

This experiment shows the effect that context  has on human behaviour.  In the context of being in a rush, the behaviour of theology students differed greatly to those in the alternative context - having plenty of time to spare.  The intuition here is that as humans, we don’t behave in the same way (perhaps even rationally...) all of the time. Our behaviour is very much influenced by our environment.

The following is a clip from a meeting of the Institute of New Economic Thinking.  The speaker is Markus Brunnermeier, a leading academic in the field of financial crises, who gives some insight into the thinking of traders when it comes to financial bubbles.  He mentions a few of the concepts I've blogged about so far such as greater fool theory and indirectly, the Keynesian beauty contest.  I think it also shows how trader behaviour is altered when there is a change in context.  Rather than refusing to purchase assets that appear over-valued, traders consider purchasing in the context of a financial bubble.

The Tipping Point

Epidemiology is the study of how something spreads throughout a population.  Traditionally, the subject of such analysis is related to healthcare e.g. disease outbreak or surveillance but in 2000, Canadian journalist Malcolm Gladwell released his international best-selling book "The Tipping Point"  which examines how epidemiology can be used to understand how social phenomena can permeate through a population in the same way that viruses do.

Gladwell defines the tipping point as "the moment of critical mass, the threshold, the boiling point." That is to say, the moment when the idea really explodes.

He goes on to describe three rules of epidemics that lead to an idea reaching this tipping point and going viral:



1) The Law of Few



This rule states that there are three kinds of people needed for an idea to spread; connectors (who have wide social networks), mavens (information specialists) and salesmen (who have the power to persuade others).

Relating this to financial bubbles and specifically the recent US housing bubble, who could have taken on each of these roles?



Connectors - in my mind, this has to be the media.  Throughout the housing boom, news reports were positive about the housing market and television ads etc made the public very much aware of mortgage deals on offer.



Mavens - the technical experts in this case must be investment banks and other financial intermediaries who developed mortgage backed securities (MBS) and pricing models.



Salesmen - in this case, they could be those who sold mortgages to the public in the first place.




2) The Stickiness Factor



This rule addresses how sticky  the idea is.  I think earning great returns on an investment is probably about as sticky  as it gets - the chance to live the American dream!




3) The Power of Context

(I'm going to put this in a separate post)


Whilst Gladwell has come under some criticism for this theory, it does seem to add up in this case: banks develop a method of securitization involving property and increase activity in the market.  House prices begin to rise and the news starts to make headlines.  The public become aware of the news and confidence in house prices rises.  Then the institutions that offer mortgages find new products and attempt to outdo each other by offering 125% mortgages etc.  House prices rise and rise and the idea that this is a safe investment and a great way to make money and improve lifestyles spreads.  And this idea travels further and further through society to a point where everyone has a big mortgage and anyone with large savings is interested in property development.  And so the bubble expands as the idea works its way through the population.  From the graph below, there looks to be a tipping point at around the same time Gladwell's book was published in 2000.

A similar tipping point  then occured in 2007, when the market went the other way, as beliefs were reversed.

Social Proofs

Social proofs form the basis for the term herding, which has become popular when describing financial bubbles. 


In his book Influence: The Psychology of Persuasion, Dr Robert Cialdini devotes an entire chapter to the power of social proofs.  The concept comes from a human awareness of the views of society as a whole and a reliance on the wisdom of crowds : the idea that if an opinion is held by a majority, it is likely to be the right one.  This feature is amplified in times of uncertainty; if a person has difficultly coming up with their own decision or view, they will comply with others more easily.






Cialdini recalls both incidents from the public sphere as well as experiments carried out by psychologists,who collect empirical data to show the biases asserted by individuals when placed in a group.  The strongest examples are particularly dark, describing the mass suicide of The People's Temple at Jonestown, Guyana ; where in 1977, 910 people drank from a vat of strawberry-flavoured poison and the murder of New York citizen Catherine Genovese in 1964; where 38 witnesses failed to call the emergency services.

Dr Robert Shiller

Extending this line of thought to financial markets has been the work of Robert Shiller.  In 2005, Shiller teamed up with Karl Case and surveyed new home buyers in San Francisco and found that the average person expected house prices to continue rising at a rate of 14% per year for the next decade.  Indeed, about a third of respondents expected to see a 50% rise in house value per year.  As these people were not market experts, this example shows how housing market optimism had become the socially accepted position and contributed to a housing bubble.

The Big Mo!

In the 1960s, a new term crept into sports reporting in the United States.  This term came to describe the extra energy a team would show when they were 'in the zone'.  They were said to have the driving force of momentum on their side - "The Big Mo"



It didn't take long for this term to catch on and it has since come to be used in all sorts of different contexts including political campaigns, social upheavals and economic cycles as well as (you guessed it!) financial bubbles.

Perhaps most notably, the term was used by Mark Roeder (a former UBS Bank exec.) in 2010 when he stated that "The Big Mo" played a pivotal role in the 2008 global financial crisis:

"...recent technological advances, such as computer-driven trading programs, together with the increasingly interconnected nature of markets, has magnified the momentum effect."



Similarly, The Economist published an article entitled "The Big Mo" (available here) which discusses how momentum in markets might explain financial bubbles and contravene efficient markets hypothesis.

Whilst contributions to the concept of momentum are numerous, the best known interpretation is Sir Isaac Newton's Second Law of Motion, which takes the form:

                                                                              F = ma

where F is the force, m is mass and a is acceleration.

The intuition here is that as mass or acceleration (or both) increases, the greater the force.  That is to say, if an object is bigger, or moving with a greater acceleration, stopping it will require greater force.



In 1982, psychologists John Nevin, Charlotte Mandel and Jean Atak, wrote a paper called "The Analysis of Behavioural Momentum", in which they explored why certain behaviours can become persistent over time.  They developed a method of applying Newton's equation to human behaviour and the way we resist change.



Could this be applied to financial markets?  What if the object (F) is the force with which traders hold a certain belief e.g. the housing market is the most profitable and the least risky market.  The mass (m) could be the number of traders and the acceleration (a) the rate of trades taking place.  As more and more traders come around to thinking this way, the asset price soars but so does F.  That is to say, if an idea already has momentum amongst traders, it will take a lot of convincing to change their minds depending on how many of them there are and how much trading they're doing.



If this is in fact the case, the next logical question is what causes the idea to spread in the first place?

Greater Fools and Beauty Contests

"Who's the more foolish, the fool, or the fool who follows him"

Star Wars fans may recognise these words as the ones uttered by Sir Alec Guinness in his role as Obi-Wan (Ben) Kenobi in Episode IV: A New Hope.  This thought-provoking concept forms the basis of one of the earliest attempts to explain financial bubbles in terms of social psychology - greater fool theory.

The idea here is that an investor (the fool) will pay for an asset, knowing full-well that the price is too high.  However, he has confidence in being able to sell it on at a later date, at an even higher price, to another investor (the greater fool).  Although popular among laymen, the theory lacks any empirical clout.




A similar concept is that of the Keynesian beauty contest. 

The idea is to imagine a competition where you are asked to pick the six most beautiful women from a larger pool.  If you pick the 6 most popular women, as decided by every entrant in the competition, you win a prize.





Keynes noted that there are different orders of strategy to selection:

Order 1) 
Pick the women that you find to be most beautiful

Order 2) 
Pick the women that you think other people will find to be most beautiful

Order 3) 
Pick the women that you think other people will think other people will find to be most beautiful

(eh???)

In otherwords, in the third order, you assume that everyone else is chosing based on what they think the average will be i.e. they will selecting using the second order rather than the first.  In the same way, the fourth order would assume that everyone else is thinking in the third order and so on. 

Keynes believed that the same principal applied to financial markets, whereby traders valued prices at the level they preceived to be the average opinion, rather than what they thought was the actual, intrinsic value.

Both of these theories seem to imply that dealer behaviour is not always consistent, leaving room for speculation and the inflation of a bubble.

Money Doesn't Grow on Trees

A simple statement that everyone in Northern Ireland will be aware of, having most likely received it in the form of a mildly agitated response from a parent after a request for money, usually as a child (or perhaps not!).  As a consequence, I think that it would be fair to say that the vast majority of those in today's society would accept the notion that money doesn't grow on trees, you can't get something for nothing and if it sounds too good to be true, it normally is.



It seems that all of these notions were set to one side for a short while in late 2001 as a 'money tree' fad swept across Counties Antrim and Down.  Being from that part of the world myself, I remember this episode quite well and recall wondering how it worked as it seemed to fly in the face of all those clichés I mentioned in the previous paragraph.  For anyone not familiar with the story, I'll break it down using the diagram below:

Here we have JB.  JB has worked his way up to the top level of the money tree since his friends MD and AP, who form the next level down,  have managed to invite two friends each, who form the next level; in this case, JC, TS, GS and RP.  Once everyone in this tier has found two friends to join in, the system begins to pay out. 

 

In the case above we have TH and KP already in this tier.  Once six other investors pay the £3,000 entry fee, JB receives all eight entry fees, walking away with £24,000 - an incredible 700% in profits!  Each member in the tree then moves up a level; MD and AP move to the top level and continue the recruitment process until each person on the bottom level brings in two friends - that's sixteen more people, therefore £48,000 split between MD and AP, so they also leave with their 700% profit and so the process continues.  So long as people keep joining the tree, investors keep progressing up the levels and eventually reach payout - SIMPLES! 

Except it's not that simple is it? As with pricing bubbles, as soon as supply outweighs demand, problems occur. And this is what happened in Northern Ireland. Eventually, people ran out of friends and family to approach and the process of recruiting for the next level stalled and the system failed.  (Either that or people lost the belief that they would make it to the top tier.)  Anyone sitting on the tree lost their £3,000 and as the whole operation was conducted in pubs and restaurants, with each investment classed as a 'gift', there was nothing they could do about it.  Ouch!

More details of this story can be found in the BBC archives: http://news.bbc.co.uk/1/hi/northern_ireland/1647715.stm



Relating back to the title of this post, if people are generally smart and know that money doesn't grow on trees, they must know that a scheme like this is destined for failure, just as traders must know that the price of houses or the value of dot-com businesses can't keep rising forever!  Yet these things keep happening and people, companies and even countries still end up on the verge of financial ruin as a result of the bursting bubble.  I believe that the social sciences can offer (at the very least a partial) explanation for why this is.

An Introductory Overview

As bubbles don't have a unanimously accepted explanation within the world of finance, economist opinion varies quite a bit. Schools of thought range from denial of their very existence to debates about whether or not bubbles are rational. Furthermore, others look to the social sciences to explain pricing bubbles in terms of human psychology and behaviour; and it is this particular area that I find most intriguing.

As a person with something of a casual interest in areas such as epidemiology and evolutionary psychology, I'm interested to see how some of the ideas discussed in books on my bookshelf at home might apply to this particular area of finance. Over the next few weeks I'll take a look at well discussed aspects like greater fool theory and herding, as well as ideas that maybe aren’t quite so well renowned within economics; risk perception, social proofs and the role of context. 



Each of these will shed a little light on issues like why bubbles start, when/why they burst, when they are likely to occur in the first place and why we are destined to see the same newspaper headlines reporting the occurrence of this destructive phenomenon in the future.