Determinants of Risk Taking: The Important Role of Risk Perception

1
Determinants of Risk TakingThe Important Role
of Risk Perception

• Elke U. Weber
• Columbia Business School
• and Center for the Decision Sciences
• SAMSI Presentation, October 4, 2007

2
Talk Preview

• Economic Theory and Risk Taking
• Risk Taking Attitude and Perception
• Cultural and Personality Differences
• How we Deal with Rare Events
• Conclusions

3
What does economic theory/finance tell us about
the determinants of risk taking?

• Traditional economics/finance models
• Expected Utility (EU) theory
• Risk-return models, like CAPM
• Risk is an (invariant) attribute of the risky
  (investment) option
• Risk Attitude/Risk Tolerance is the only thing
  presumed to vary between individuals

4
Risk Attitude/Risk Tolerance as a Personality
Trait

• While technically only a parameter of a utility
  function, risk attitudes are often given
  interpretations as personality traits
• Psychometric scales measure degree of
  risk-aversion/risk-seeking (Kogan Wallach,
  1962)
• Risk-attitude used as a selection criterion in
  some hiring decisions or client/advisor matchings

5
Problem with Risk Attitude as a Personality Trait

• Risk taking of any given individual varies across
  contexts and situations
• Anecdotal evidence
• manager who is risk neutral with company money,
  but risk averse in her personal financial
  decisions
• rock climber who takes plenty of recreational
  risks but is anxious about missing flights
• Systematic comparison of risk taking financial,
  health, recreational, ethical or social
  decisions (Weber, Blais, Betz, 2002)

6
DoSpeRT Items (Weber, Blais, Betz, 2002)

• Admitting that your tastes are different from
  those of a friend. (S)
• Going camping in the wilderness. (R)
• Betting a days income at the horse races. (I)
• Swimming far out from shore on an unguarded lake
  or ocean. (R)
• Investing 10 of your annual income in a moderate
  growth mutual fund. (I)
• Not wearing a seat belt when being a passenger in
  the front seat. (H)
• Taking some questionable deductions on your
  income tax return. (E)
• Disagreeing with an authority figure on a major
  issue. (S)

7
DoSpeRT Subscale correlations (Weber, Blais,
Betz, 2002)
Study 3
F H E R S Total
F 1.00 .60
H .29 1.00 .60
E .51 .61 1.00 .45
R .36 .34 .34 1.00 .75
S -.07 .06 .04 .13 1.00 .33
8
Recreation targets bungee jumpers hang
gliders scuba divers Gambling targets casino
gamblers Investment targets stock-trading
clubs Health targets smokers gym members
Hanoch, Johnson, Wilke, Psych Science, 2006
9
Problem with single determinant (parameter) for
risk preference

• Need to account for both situational and
  chronic/biological differences in risk taking
• Age and gender differences in sensation seeking
• Hard to do so with a single parameter
• Is risk attitude really the only determinant of
  risk taking that varies between individuals or
  situations?
• Behavioral models of risk taking add determinants
  
• Prospect Theory
• Adds loss aversion and probability weighting
• Generalize risk-return models (Weber Milliman,
  Mgt Sci, 1997)
• Perceived riskiness of risky choice options seen
  as a psychological variable

10
Risk Perception as Mediator of Situational
Differences

• Perception is subjective and varies with
• Expected outcome volatility
• Goals, expectations, and other reference points
• Familiarity with risky option and perceived
  control
• Home bias effects in investing mediated by
  differences in perceived riskiness (Kilka M.
  Weber, 2000 Weber et al., 2005)
• Other affective responses (dread and fear)
• Risk as a feeling (Loewenstein, Weber, Hsee,
  Welch, 2001)
• Personal experience of adverse consequences

11
Risk Perception as Mediator of Situational
Differences

• Perception is relative
• Thurber story
• Two thought experiments
• buckets of water
• two inheritances
• Also true of perceptions of difference/volatility
• Webers law from 19th century psychophysics
• difference in magnitude required to perceive two
  stimuli as different (JND) is proportional to
  absolute stimulus magnitude
• Coefficient of variation (CV) as a measure of
  perceived volatility
• CV standard deviation / expected value
• measure of relative risk risk per unit of return
• Used in many applied areas
• engineering, medicine, agricultural economics

12
Relative-risk coefficient of variation predicts
risky foraging behavior of animals across a
range of situations (S. Shafir, 2000)
13
Risk taking

• Distinguish between
• Perception
• Evaluation of outcomes and probabilities (hot or
  cold)
• Motivation
• Whether risk as perceived is exciting or scary
• Probably related to optimum arousal set points
• Sensation seeking (Zuckerman)
• Perceived-risk attitude (PRA), b
• Willingness to take(X) a(Benefit(X))
  b(Perceived Risk(X)) c

14
Perceived Risk Attitude (Weber, Blais, Betz,
2002)
Domain averse neutral seeking
F 44 72 0
H 45 70 1
R 43 71 0
E 49 65 2
S 50 66 0
15

• Some effects attributed to risk attitude due to
  non-attitudinal factors
• Perceptions of risk (and of return)
• Probability weighting
• Loss aversion

16
Cultural Differences in Risk Taking Mediated by
Differences in Perceived Riskiness

• Study of investment decisions made by Chicago and
  Shanghai business students, where Chinese
  appeared less risk averse than Americans
• Chinese had greater WTP for same investment
  options
• Cushion hypothesis
• Cultural collectivism provides implicit insurance
  against catastrophic losses, so risks are
  objectively reduced
• Investors with larger collectivist networks
  (mostly Chinese) perceived risks of investment
  options to be lower
• Risk perceptions and risk taking only different
  between cultures for outcome dimensions that are
  transferable
• For money, but not for health or grades
• No cultural differences in attitude towards
  perceived risk, b
• WTP(X) EV(X) bR(X), where R(X) is stated
  subjective perceived risk of investment option X,
  not Var(X)

17
Not all differences in perceived riskiness are
rational/justifiable

• Gender differences in risk taking
• Womens pension investments are more conservative
  than those of men
• Women take fewer risks than men in all domains,
  except for social risks
• But also perceive the risks to be greater in
  those domains than men do (Weber, Blais, Betz,
  2002)
• No gender differences in perceived risk attitude

18
Gender differences in perceived risk

• Probably not biological, since restricted to
  cultural majorities
• White males perceive risks to be lower than any
  other group (Flynn et al., 1994)
• Sociological/psychological explanation for group
  differences in perceived risk
• Majority members do not take risks, they have
  the power to manage risks
• Perceived controllability one of the
  psychological risk dimensions identified by
  Slovic et al. in 1960s and 70s
• Driving seen as safer than flying
• Emotional mediation of effect
• Responsible for false generalization to domains
  where no gender differences in controllability
  exist (investment decisions)

19
Risk taking in decisions from description vs.
decisions from experience

• Decisions from Description
• outcome distribution fully described
• numerically or graphically
• almost exclusively studied in human choice
  
• Weber, Shafir, Blais (2004)
• Decisions from Experience
• outcome distribution initially unknown
• knowledge of outcome distribution acquired by
  personal exposure in repeated choices
• strong recency in effect of prior observations
• exclusively studied in animal choice
• recent resurgence of interest in reinforcement
  learning models
• Camerer Ho (1999), March (1996)

20
Effect of rare events

• Rare events are
• Overweighted in decisions from description
• Captured by decision weight function of Prospect
  Theory
• Underweighted in decisions from experience
• Unless they recently occurred, in which case they
  are strongly overweighted (Hertwig et al., 2004)
• Risk taking under the two information conditions
  can be very different as a result, when outcome
  distributions are skewed
• Most investment decisions based on a combination
  of outcome distribution description and personal
  experience
• Personal experience typically the stronger
  determinant of choice
• More emotionally engaging, vivid

21
Decomposing risk taking into risk perception and
risk attitude What does it buy us?

• Improved understanding of individual or group
  differences in risk taking
• Not all differences in risk taking are due to
  differences in attitude towards risk (liking it
  or disliking it), though some are
• Probably related to biological differences in
  optimal arousal set points
• Differences in perceived risk can be due to
• Stable differences in sociocultural environment
• Transient situational characteristics
• Differences in perceived risk can be
• Justifiable and realistic responses to
  differences in opportunities or constraints
• Less justifiable myopic emotional reactions to
  events or emotional overgeneralizations

22
Summary

• Talk tried to introduce you to a different
  perspective on risk taking (in financial and
  other contexts)
• Psychology component of risk taking is not
  restricted to individual or group differences in
  risk attitude
• Richer framework that unconfounds differences in
  risk perception from differences in true attitude
  towards risk resolves many apparent paradoxes
• Inconsistent risk attitudes across domains
• Rabins (2000) calibration theorem

23
Possible Take Away

• Better appreciation of the influence of
  non-analytic processes on risky decisions
• Better appreciation of the malleability of
  perceptions of risk
• Risk is not an immutable attribute of a risky
  choice option
• Mode by which information about outcome
  variability is acquired, familiarity and
  emotional comfort, and expectations and other
  relative comparisons can play important roles