BEHAVIORAL ECONOMICS: IMPLICATIONS FOR ACTUARIAL SCIENCE ... · BEHAVIORAL ECONOMICS: IMPLICATIONS...

2012 CAS Annual Meeting

November 11-14, 2012

BEHAVIORAL ECONOMICS:

IMPLICATIONS FOR ACTUARIAL

SCIENCE AND ENTERPRISE

RISK MANAGEMENT

Robert F. Wolf, FCAS, CERA, ASA, MAAA

Managing Partner- Casualty Actuarial

Practice, Willis North America

http://www.casact.org/



RECAP OF 5 NECESSARY ELEMENTS OF

PRUDENT ERM CULTURE..AS LEARNED FROM

THE FINANCIAL CRISES

1. Incentive compensation requires appropriate

alignment with desired performance

2. Nobody should have the authority to make decisions without accountability.

3. Do Not Assume we Can Get Rid of the Risk Tomorrow for the same Price as Today

4. Modeling and Management Must consider the Behavioral Decisions of people.

5. Risk Managers Must Question the Answers




MAIN PREMISE

Most risk does not manifest itself

from some exogenous contingent

event but from the behaviors and

decisions of people




QUESTION

• Which of the following

causes more deaths in the

US each year?:

1. Colon Cancer

2. Motor Vehicle Accidents




AVAILABILITY

HEURISTIC

• Vivid Events, Easily Imagined,

Consistent with Memory

Structures are Judged to be

more likely than equally

probably events.




“Each new market crisis demonstrates that correlation in stressed environments is much higher than historical averages would indicate,”

……….

“… the quantification of remote probabilities is more difficult than the quantification of possibilities.”

From the “Credit Crisis Lessons for Modelers” essay by Parr Schoolman, FCAS, Vice president at Aon Benfield Analytics.




QUESTION

• Which Would you Rather

Have?

• $700 for sure

• or 85% chance for $1000?




QUESTION

• Which Would you Rather Do?

• Pay $700 for sure

• Take an 85% chance of

paying $1,000 and 15% of

paying $0




QUESTION

• The U.S. is preparing for the outbreak of a deadly disease that is expected to kill 600 people. Two alternative programs are being considered. Which do you favor?

• Program A: 200 people will be saved

• Program B: 1/3 all be saved, 2/3 all will die




QUESTION

• The U.S. is preparing for the outbreak of a

deadly disease that is expected to kill 600

people. Two alternative programs are

being considered. Which do you favor?

• Program A: 200 people will be saved

• Program B: 1/3 all be saved, 2/3 all will

die




QUESTION

• The U.S. is preparing for the outbreak of a deadly disease that is expected to kill 600 people. Two alternative programs are being considered. Which do you favor?

• Program A: 400 people will die

• Program B: 1/3 probability no one will die, 2/3probability all will die




FRAMING EFFECT

• Framing changes the

“problem”

• Does not change the data

• We are Risk Averse to gains

• We are Risk Seeking to loss




• Based on your audit experience, is the incidence of significant executive level fraud more than 10 in each 1000 firms (1%) audited?

• A. Yes, more than 10

• B. No Less than 10

• What is your estimate of the number of clients per 1,000 that have significant executive –level management fraud?

A Question I asked a Group of

Accountants




A QUESTION I ASKED A GROUP OF

ACCOUNTANTS

Average Answer

16.52





ACCOUNTANTS

• Based on your audit experience, is the incidence of significant executive level fraud more than 10 in each 1000 firms (1%) audited?

• A. Yes, more than 200

• B. No Less than 200

• What is your estimate of the number of clients per 1,000 that have significant executive –level management fraud?




Average Answer

43.11


ACCOUNTANTS




ANCHORING AND INSUFFICIENT

ADJUSTMENT

• We tend to be affected by numbers, even

when numbers aren’t valid.

• They “anchor” our thinking, and we tend

to adjust away from them less than we

should.

• They effect our estimates, even when we

should ignore them


SHOULD OUR INITIAL ESTIMATE=

MEAN?

• Turns into deviation

from anchored/biased

estimate as opposed to

a deviation from mean

My Initial Estimate

Where is the Edge

of the Cliff?

ANCHORING AND INSUFFICIENT

ADJUSTMENT



REGRESSION TO THE MEAN

AND REPRESENTATIVENESS

• If you have a wonderful round

of golf one day, what would

you expect of your next

rounds?




CONFIRMATION BIAS

• We tend to seek out

confirming evidence for our

hypotheses

• We tend to ignore/discount

disconfirming information




HINDSIGHT BIAS

• Once we know something, we

cannot remember the situation

when we did not know it




DOLLAR AUCTION

• Bidding starts at $1. Proceeds

with increments of $1

• Highest bidder pays what they

bid and receives $20

• 2nd highest bidder pays what they

bid




ESCALATION OF

COMMITMENT

• Commitment of Resources to a course of action

• Action does not produce a return, but may

produce a loss

• Decision is made to commit further resources in

order to turn the situation around

• Process may repeat and “escalate” several

times as additional resources are invested




WHO MADE THE QUOTE

• “ A severe depression like that

of 1920-21 is outside the

range of probability”




WHO MADE THE QUOTE

• “ A severe depression like that

of 1920-21 is outside the

range of probability” • Harvard Economic Society- Weekly Letter

9/16/29




WHO MADE THE QUOTE

• “There is not the slightest

indication that nuclear energy

will ever be attainable. That

would mean that the atom

would have to be shattered at

will”




WHO MADE THE QUOTE

• “There is not the slightest indication that nuclear energy will ever be attainable. That would mean that the atom would have to be shattered at will” • Albert Einstein, 1932




WHO MADE THE QUOTE

• “If anything remains more or

less unchanged, it will be the

role of women”




WHO MADE THE QUOTE

• “If anything remains more or

less unchanged, it will be the

role of women” • David Riesman, Harvard Sociologist, 1967




WHO MADE THE QUOTE

• “640K is enough for everyone”




WHO MADE THE QUOTE

• “640K is enough for everyone” • Bill Gates, 1981




OVERCONFIDENCE

• We are paradoxically more confident in

estimates of events we are less familiar

with than with events we are more

familiar with

Expertise

Confidence




…But So What ?????????

YES VERY INTERESTING…..




TO MAKE CONVINCING

RATIONAL ARGUMENT…

• You Need…

• Sharp Analysis and

• Clear Presentation




1. Search Search Search for Disconfirming Information • Private Collection of Information

• Structure Meeting – Make Key Decisions Early

• Ask Outsiders; Maximize Diversity

• This helps overcome representativeness, hindsight, confirmation and overconfidence bias traps

TOOLS




“CHIEF SKEPTICAL

OFFICER?” “When a business line brings a new idea to

the CEO, he should be able to ask, ‘Have you run this past the Chief Skeptical Officer and does she concur with this proposal?’ The CSO (could also be referred to as the Common Sense Officer) might not always be popular, but the improved decisions made will allow the CEO to more confidently execute the company’s strategic plans.”

…Max Rudolph’ s Essay on Financial Crises..Lessons Learned




2. Refine/Reframe/Flip the Problem • Redefine the old problem. It is truly a new one.

• Re-assign Accountability

• Set Limit of Commitment

• Avoid Public Comment Too Early

• Flip the Problem and See if You are Having the Same Discussion and/or reaching the same decision.

• This helps overcome framing, escalation of commitment biases

TOOLS




ERM STRATEGY- DECISION-

MAKING

• Understand Decision Biases and Our Predilection to be Influenced by them

• As decision makers, we all deviate from rationality

• We are influenced by the format of information

• Heuristics are useful but open us up for biased decisions




ERM STRATEGY- TOOLS TO

OPTIMIZE INFLUENCE

• Harness Framing

• Highlight Sunk Costs

• Prompt Vivid Stories

• Harness Availability




….BUT WHAT DO YOU GET WHEN

YOU CROSS AND ACTUARY WITH AN

ANTHROPOLOGIST?

..




….HOW WE BEHAVE IN GROUPS AND

HOW WE FORM GROUPS BASED ON

OUR VIEWS OF RISK?

……WE HAVE AN INCOMPLETE STORY

SEE INGRAM’S COLLABORATIVE

COLLECTION OF WORK

..




THIS IS THE SLIDE TITLE

Thank You!!!!


BEHAVIORAL ECONOMICS: IMPLICATIONS FOR ACTUARIAL SCIENCE ... · BEHAVIORAL ECONOMICS: IMPLICATIONS...

Documents

Transcript of BEHAVIORAL ECONOMICS: IMPLICATIONS FOR ACTUARIAL SCIENCE ... · BEHAVIORAL ECONOMICS: IMPLICATIONS...