Biases Flashcards
Dan Ariely - illusions and bias
its easy to use hindsight to see the mistakes we make with visual illusions its not that easy to see the cognitive mistakes we make as they are much less obvious
“even Einstein would probably be fooled by those tables” - Thaler and Sunstein
“our understanding of human behaviour can be improved by appreciating how people systematically go wrong”
Why do we have these biases?
the problem with rationality is choice:
- if we’re totally rational we would weigh up all the available information, decline probabilities and utility and then make rational decisions
the reality is that we are faced with 1000’s of decisions and choices and we simply can’t compute every single equation
Bounded rationality - Simon Nobel price winner 1955
refers to cognitive limitations facing decision makers in terms of acquiring and processing information
as a result we tend to use heuristics in decision making
Heuristics are methods for arriving at satisfactory solutions with modest amount of computation
ie instead of making a comprehensive search for the options that maximise benefits that people satisfice
we satisfice because we have limited cognitive abilities and are making decisions in complex scenarios - bored too easily
Simon recognised that we don’t necessarily make decisions in the way in which traditional economics had described
Heuristics are necessary and pragmatic and can often result in quicker and effective decisions being made.
The problem with heuristics
they can lead to biases meaning systematic errors in how we Make decisions
- we are in possession of all the necessary information to make rational decisions with consistent preferences
- we don’t have the time to take into account all the information to make rational decisions so we have to rely on short-cuts/heuristics
- we rely on heuristics and our emotions which affect our decisions
Difference between bias and random error?
biases are predictable, there is a pattern to them and they’re consistent. A pattern of human errors that affects us all
Availability bias
relates to the process of judging frequency by the ease with which instances come to mind (the availability of information)
people believe that events are more frequency or more probably if they’re easier to remember
Kahneman on availability
“when faced with a question people relating to the frequency go a category we can make judgements based on the ease of which something comes to mind”
we substitute an easier question from a difficult question
examples of availability bias
after 9/11 in the US peoples estimates for the danger of flying increased massively
as a result, more people decided to drive instead leading to an increase in road deaths
road deaths greater than casualties of 9/11 itself - ITNU
using availability bias
understanding can help save marriages or prevent fights during group work
when working on a group presentation you’ll retrieve and be aware of your contribution more easily than that of others
“you are likely to have this feeling even when each member of the team feels the same way” - Kahneman
earthquake insurance will be higher just after an earthquake relative to 7 years after
Overconfidence/optimistic bias
“we tend to exaggerator our ability to forecast the future, which fosters optimistic overconfidence” - Kahneman
positives of overconfidence
Martin Seligman speaks of positives for optimism in terms of success and mental wellbeing
the market values overconfidence: if CEO predicts share price of company highly to investors
“optimism is highly values, socially and in the marketplace; people and firms reward the providers of dangerously misleading information more than they reward truth tellers” - TF&S
4/10 new businesses fail in their first year and 9/10 after several - its optimistic bias that leads entrepreneurs to undertake risk so is good for capitalism
“economic dynamism of a capitalist society” - TF&S
their confidence in future success sustains a positive mood that helps them obtain resources raise morale and enhance prospects of prevailing
“when action is needed, optimism, even of a mildly delusional variety, may be a good thing” -TF&S
AIG overconfidence
gov bailed the company out for the $180bn - highest gov bailout
needed as it invested highly in a one way bet
CEO said that he couldn’t see them losing a single dollar in the investment
overconfidence can lead to individuals and organisations believing that they understand and are aware if all the risks that they are facing and can lead people to underestimate risk with dangerous consequences
overconfidence and economics
confidence is implying behaviour that goes beyond the rationale approach to decision making
when people are confident, they go out and buy, when they aren’t they withdraw and sell
Barber and Oden - confidence
studied gender differences in the area of finance
analysed data from a brokerage company
study found that men traded 45% more than women
didn’t actually lead to higher results, but actually a lower expected utility
Confirmation bias
we have a natural tendency to search for confirming rather than disconfirming evidence
we also have a tendency to look for information which confirms a belief that we already hold - self-serving bias
leads to people to overweight information which confirms their prior views, and to underweight information which disconfirms these views
“people seek data that are likely to be compatible with the beliefs they currently hold” - TF&S
confirmation bias example
betters asked how confident they were in their first horse winning before and after they had placed their bet
- more confident afterwards as they’re looking to rationalise their decision
Ellsberg problem
urn contains 30 red balls and 60 balls that are either white or yellow - in any ratio
which gamble do you prefer?
A: 1/3 chance if red - £100
B: £100 if white drawn - between 0 and 2/3 prob
C: £100 if red or yellow drawn - between 1/3 to 1 prob
D: £100 if white or yellow - prob 2/3
most people prefer A over B and D over C
but this is inconsistent
if you think a red ball is more likely than a white ball and you chose gamble A then you should also choose gamble C not D
Ambiguity aversion
in decision theory and economics, ambiguity aversion is a preference for known risks over unknown risks
and ambiguity-averse individual would rather choose an alternative where the probability distribution of the outcomes is known over one where they’re unknown
“risk greases the wheels of a free-market economy; uncertainty grinds them to a halt” - Nate Silver
Representativeness Bias
refers to when we make decisions and focus on the similarity of the description to the stereotype ignoring the base rate
if we are faced with a difficult question we substitute it with an easier question and answer that instead
e.g if looking at probability of a situation occurring, we look for what the description of the event most similar to
“base rates should dominate you estimates” - TF&S
we should anchor our judgements on the probability of an outcome on a plausible base rate
“be aware of the ease with which we can try and look for answers which sounds more likely but are not backed up by fact” - Kahneman
Conjunctive Fallacy
when people judge the conjunction of two events (use bank teller and feminist example) to be more probable than one of the events in a direct comparison
another example of representativeness
the most representative outcomes combine with the personality description to produce the most coherent story, which is most plausible but not necessarily the most probable
- use computer science example too
The Gambler’s Fallacy
urn contains 10 balls, half red, half blue
once selected, balls placed back
I make 3 draws, all are red
for the 4th draw: blue is more representative of the population than drawing red but not more probable
another example of the representative heuristic
relates to evolutionary psychology as in the past, there was a good reason to think that a series of events would be broken - e.g for meteorological events
Present bias
relates to time inconsistency
if you prefer 3/4 chocolate now to a full bar in 3 weeks you should prefer 3/4 of a bar in 3 weeks over a full one in 6 weeks but you don’t
we have a very high discount factor for now compared to later
- preference for immediacy of rewards
- we are disproportionately impatient in the short term but more patient when preparing for future
- our emotions get in the way as are real and tangible, in the future things are less tangible
- this makes our decisions for the future emotion-free and we can act more rationally (marshmallow experiment)
present bias in many disguises
e.g the environment and chopping down trees and in over-spending, overfishing and predicted empty oceans by 2048
financial innovations to fuel our hyperbolic discounting desire
- Apple Pay, Amazon swipe to pay etc
US citizens have more than a billion credit cards equating to 4 credit cards per person
research shows that people spend more when they have access to credit cards
Prelec and Simester experiment
held an auction for Harvard students for tickets to Red Sox and a Celtic game
randomly asked if they’d pay by cash or credit card - with a cash machine available
- tickets for the Celtic game went for more than double with credit cards whilst for the baseball game went for 75% more
this is because credit cards allow us to consume now but delay the pain
we get all the benefits of the consumption now but don’t get the associated loss
“with a credit card that loss is not as vivid or visceral” - Ariely
contrasts with the downside of losing cash in payment
Astraction
- paying for things is easy, frictionless and also thoughtless
- when we pay by cash, we have to take out and feel the cash, a pain which we don’t feel with CC’s
“our lack of awareness about spending may be the scariest thing” - Ariely
Ariely believes what we end up doing is increasing the time between consumption and payment and we decrease the attention needed to make a payment
Status Quo bias
an implication of loss aversion is our strong tendency to remain at the status quo
- disadvantages of leaving loom larger than the advantages
