reflections

01-IntroandPreferences2.pdf

Home >Business & Finance homework help >Operations Management homework help >reflections

BUS143

Introduction and Preferences

BUS143: Judgment and Decision Making

Ye Li

Why you decided to take this class

“Decisions are the essence of management. They’re what managers do—sit around all day making (or avoiding) decisions. Managers are judged on the outcomes, and most of them—most of us—have only the foggiest idea how we do what we do.”

Thomas Stewart Former editor (2002-2008),

Harvard Business Review

BUS143

Decision Making: Two Questions

• Why is decision making difficult?

• What constitutes a good decision?

Decision Making: Good Process

• What is a decision? – A costly commitment to a course of action.

• Outcomes versus Process Outcomes

Good Bad

Process Good

Bad

Bad “luck”

Good “luck”

BUS143

Components of a Good Decision

• I have considered my ABCs – Alternatives – Beliefs – Consequences

• I am devoting an appropriate amount of resources

• I have avoided major decision traps

Decision Making Components: The ABCs

• Alternatives – Identification and articulation – Construction/refinement

• Beliefs – Identification and quantification of uncertainties – Information collection/gathering

• Consequences – Identification of consequences (and objectives

addressed by consequences) – When possible, quantification of tradeoffs among

objectives

BUS143

Decision Making: Good Process

• Putting it all together (for now)…

Good decision making is choosing the alternative that best meets your objectives

in the face of uncertainty about what consequences will ensue.

3 Perspectives on Decision Making

• Normative – How should people make decisions?

 Related concepts: rational; optimizing; forward-looking

• Descriptive – How do people make decisions?

 Related concepts: boundedly rational; limited cognitive capacity; heuristics or rule-based; myopic

• Prescriptive – How can we help people make better decisions? – Prescriptive advice via practical applications, in…

 Management  Marketing  Finance  HR  Life!

BUS143

Example

• Problem – Imagine two 1-mile-long (1.61km) pieces of railroad track, put

end to end, and attached to the ground at the extremes. When it gets hot, each piece of track expands by 1 inch (2.54cm), forcing the pieces to rise above the ground where they meet in the middle. How high will the track be in the middle?

• Normative rule: – Pythagorean Theorem:

• Descriptive reality: – Most people underestimate x. (We anchor on 1 inch.)

• Prescription: – Use normative rule (geometry). Don’t rely on intuition.

More Examples

• Normative rule: – Lighter objects should

be judged as lighter.

• Descriptive reality: – Sometimes our vision

tricks us.

• Prescription: – Use an outside reference

or instrument – Note: Pilots have specific

strategies for counteracting visual illusions

Which box looks lighter?

BUS143

Class Philosophy

• Overarching goal: – Help you to think differently and better about

how you and others arrive at judgments and choices

• Premise 1: Effective management requires a good understanding of how people make decisions (so does effective marketing, investing, etc.)

• Goal 1: Understand what drives decisions. – Traditional view: Informed decision-maker who

decides based on “value” (= preferences × probabilities) – Behavioral view: People evolved to solve problems of

survival and reproduction, not complex business problems

Class Philosophy

• Premise 2: “Flaws” in decision processes are everywhere, and in some cases unavoidable In the land of the blind, the one eyed man is king

– Erasmus

• Goal 2: What are the implications of human decision “anomalies” for… – Managers, investors, consumers, policy makers, etc. – A self-defense guide  Competitive advantage

BUS143

Logistics, Format and Target Audience

• I run this class like an MBA class – Primary source readings (mostly academic papers)  All available on iLearn  no expensive textbooks!

– In-class and online demonstrations – Want to maximize YOUR engagement in the class

• Target market – The intellectually curious – Future management consultants, brand managers,

financial professionals, entrepreneurs, policy makers, lawyers, etc.

My Teaching Philosophy

• I’m here to help you succeed— in class and in life! • I will submit the best grades I can! – BUT, maximum allowable average GPA: 3.25

• The focus is on learning and not testing – I will not give you any busywork. Everything has a

purpose.

BUS143

Administrative Stuff

• Coordination – Class will start on time, will not end late  Content is front-loaded: Basic concepts in first 20 mins

– Lecture slides will be posted 9pm day before class  Please do not upload these (or anything) to CourseHero!

– Please check your email (forward it to your phone!)

• What you will be graded on 20% - Participation 40% - Four 600-word write-ups 10% - Weekly web assignments 30% - Take-home final exam

20%: Participation • Purpose: 1) Engage students and get different perspectives to improve

learning (and to know when students are confused). 2) Get practice for jobs and MBA classes (participation % is sometimes as high as 50%!). 3) 80min of lecture is boring! – Attendance is required for learning but not enough! – Participation means raising your hand and contributing to discussion.  Goal is to

hear from at least one-third of students every class! – For some types of questions, it’s possible to be wrong This is useful for learning

and should not make you afraid! – You should also ask ME questions (there are no bad questions) – Doing readings before class makes it easier to contribute

 I expected readings to take between 2-5 hours a week. – Last quarter, average participation score was 17/20! (so do not stress)

• I will use a seating chart to learn names but you pick your own seat – Your seat in class will be permanent for the rest of the quarter

• Electronics policy: Absolute no phones! (I can see you using them) – Leave them at home if you don’t have the self-control to not use them. – Computers and tablets for note-taking You must sit in laptop section

BUS143

40%: Four Write-Ups • Purpose: Encourage deeper thought on the readings to stimulate

better class discussion.

• 600-word limit (penalties for going over) each write-up – Business memos must be concise and to the point. People are busy and

won’t read anything beyond first page! – Shorter write-ups (<450 words) tend to lack detail. Aim for 500-600. – Do NOT include an introduction, background, or conclusion!

 These are not standard English essays  No need for definitions (we know the material) but show your understanding by

correctly applying the concepts and provide concrete examples

• NO need for outside research – Should be based on class concepts and readings!

• Clarifying a possible misconception: This is not a writing class – Yes there is a lot of writing, but the purpose is to communicate your

understanding and ability to apply class concepts. – Style, grammar, etc., only important in terms of me being able to

understand you

Write-Up Logistics • Due on Tuesday at 9pm each week on iLearn

– 1 point bonus if turned in by 9pm Sunday, and feedback guaranteed by the weekend

– 2 point penalty after 9pm Tuesday • First write-up is to be done in randomly-assigned groups • Pick any 3 of the remaining 6 weekly topics (no topics weeks 9 or 10)

– Can only submit current week’s topic – Weighted by how well you do (WARK: Weighting after results known)

 Best = 40%, 2nd = 30%, 3rd = 20%, worst = 10%

– Optional: Submit a 5th write-up and the lowest score will be dropped

• Scores out of 25 (but they are not percentages!) – Most scores will be in the 19-22 range. 20 and 21 are most common – 17 or less  =( , 18-19  major errors, 20-21  good with some errors,

22-23  great, minor errors, 24-25  near perfect, great examples – Individualized feedback to improve for next time (via iLearn) – Please read the syllabus for more details on what we’re looking for!

BUS143

10%: Web assignments

• Purpose: Demonstrate ideas we will cover in class. Seeing is believing. – Short web assignments due each Friday at 3pm

 Available at end of prior week of class  I have to analyze the data for class, so be on time!  Submissions after deadline not be included in the class data (and no

chance for extra credit)  After first class of the week  ½ credit  After second class of the week  ¼ credit.

– Complete BEFORE doing readings. – No formal grades (no right or wrong answers)

 No credit if you clearly just rushed through the survey without any thought.

– At least one chance for extra credit each week, sometimes more!

30%: Take-Home Final Exam

• Purpose: Test your understanding of ALL class materials. No midterm!

• Take-home case final – Available last class and due June 10th – 2000-word write-up analyzing a real business case using

concepts from this course – May complete the final in pairs or alone (90% in pairs) – More details in last few weeks of class

• Note: If you don’t know anyone in this class, get to know your groupmates! – I also want to meet each group! Groups 1 and 2: please see

me after class today!

BUS143

Up to 5%: Extra credit (see syllabus) • Purpose: Encourage you to see the world in terms of class concepts.

• Find a recent (within 1 month) example from current news, ads, etc., that illustrates an idea or ideas from the course – One paragraph explanation of why this is a good example of something we

discussed in class. – Submit example by email (see syllabus for format)

• Great examples earn up to 1% extra credit (~50% past success rate) – Maximum of 5 submissions. Maximum of 5% total extra credit in class

• HINTS 1. Articles about research or that use class concepts directly are not good

examples. The idea is that I might use the example in my slides! 2. The example should very clearly illustrate the class concept even before I

read your explanation. 3. If you specifically go looking for examples/articles, you probably won’t find

good ones. The best examples are just things you happen upon in everyday life.

Progression for each week

Wednesday

Web assignment link posted

Additional videos posted (if any)

Can start readings after completing web assignment

Friday 3pm

Web assignment due (before doing

reading)

Plan 2‐5 hours per week for readings

Plan 3‐6 hours per 600‐word write‐up

Sunday 9pm

Write‐up early deadline

(Tuesday 9pm is normal deadline)

Lecture slides posted (for help in taking

notes)

Monday & Wednesday

Class 9:30‐10:50; 11:00‐12:20

Office Hours MW 12:30‐1:30, Th 9:30‐10:30

BUS143

My blunt view on you liking this class

• (I think) MOST of you will enjoy this class, and (I hope) many will love it

• However, some of you may not, especially if… – You do not like doing potentially difficult (non-textbook)

readings before learning the topic in class – You do not think you can write concise write-ups based on

those readings and open-ended prompts – You do not enjoy class discussions – You are not here to actively learn (or are just here because

you heard the class is easy)

• This class is designed to challenge you but it will be very rewarding to the intellectually curious!

Outline of Topics Domain-General,

Foundational Stuff • Two Brains • Preferences (Loss Aversion,

Prospect Theory) • Uncertainty, Heuristics, Biases • Overconfidence • Mental Accounting • Choice Context • Choice over Time • Memory, Emotions, Social

heuristics

Bringing It All Together and Reinforcement • Prediction: Models vs.

Experts • Organizational Decision

Making – “Into Thin Air” Case

• Happiness • Nudging (AKA choice

architecture) • Behavioral Finance • Deciding Better

BUS143

The Balance of Good Decision Making

“Decision-making is a struggle between emotion and intellect. You must use intuition and analysis, not intuition or analysis. When you use both together, you achieve the optimal result.”

-Andy Grove

Address to Harvard Business School Class of 2000

Introduction: Two types of decision makers

• Homo sapiens (humans) – Limited processing

capacity: memory, attention, will

– Satisfices: Aims for “acceptable” level of performance, defined over limited problems

– Has social concerns, positive and negative

• Homo economicus (econs) – Infinitely sensitive,

incredibly smart • Know precisely what they

want • Know exact tradeoffs

– Maximizes across all decisions

– Purely selfish and greedy, pursues wealth

BUS143

Introduction: Your two brains

• System 1: Intuitive – Fast

– Automatic

– Effortless

– Associative

– Difficult to modify

Introduction: Your two brains

• System 2: Deliberative – Slow – Needs to be learned – Effortful – Non-emotional – Deliberately controlled – Easy to modify – Rule-governed

BUS143

The Modular Brain

• System 1 evolved over millions of years to solve simple decisions quickly. – e.g., Food, Fight, Flight, Mate?

• The more “effortful” System 2 is a much more recent evolution – Prefrontal cortex deals with planning, personality, social

“control”, etc. – What happens with more complex decisions: choosing a

mortgage, stock, or employee?

System 2 as an integrator and editor of System 1 output

• Keep this in mind whenever you have questions: – About rationality – “Why are people so stupid?”

• The big question: Do people always figure out when System 1 makes mistakes? Does System 2 always dominate?

BUS143

Agenda: Preferences

• Review: normative theory vs. behavioral theory • Applications of loss aversion

– Endowment effect – Pricing – Sunk costs – Defaults – Persuasion – Multi-attribute reference levels – Targets

Normative Theory: Subjective Expected Utility

• People assess options relative to their net worth

• People pick the best option that maximizes utility across all possible actions, times.

• Diminishing marginal utility of wealth explains why people are risk averse

Total Wealth

U ti li ty

Utility of gamble: 50% of winning $140 50% of losing $100 (assume you have $1,000 saved) u($1,000) .5*u($1,140)+.5*u($900)

$900 $1000 $1140

BUS143

Are you an Econ? Rabin’s calibration theorem

• A person who turns down this gamble – 50% chance of +$10.10

– 50% chance of -$10

• Must turn down – 50% chance of -$800

– 50% chance of +$3,494

• And also must refuse – 50% chance of -$1,000

– 50% chance of +$10,000,000,000

(or more) (Rabin 2000, Econometrica)

Behavioral Theory: Prospect Theory (Kahneman & Tversky 1979)

1. Relative Evaluation: Value is judged relative to a reference point (xr), NOT total wealth.

2. Loss Aversion: Losses loom larger than gains.

– V(x) = xi–xr if x >r

– V(x) = λ(xi-xr) if x < r

3. Diminishing sensitivity.

– V(x) = xα where α < 1

λ = how much steeper slope is for losses

GainsLosses

reference point

Value

Change in wealth

BUS143

Loss aversion: Defined

• Reactions to losses are more intense than reactions to gains of the same magnitude

• Potential costs, efforts, and sacrifices are weighted more heavily than potential benefits, rewards, and opportunities

Measuring Loss Aversion

You indicated whether you would play a series of gambles involving potential losses

• If the coin turns up heads, then you lose $15; if the coin turns up tails, you win $90

• If the coin turns up heads, then you lose $30; if the coin turns up tails, you win $90

… • If the coin turns up heads, then you

lose $90; if the coin turns up tails, you win $90

• If the coin turns up heads, then you lose $105; if the coin turns up tails, you win $90

BUS143

Framing Matters: Mortality Rate McNeil, Pauker, Sox, & Tversky (1982)

• Surgery: Of 100 people having surgery for this condition, 10 die in surgery or in the postoperative period, 32 have died within a year, and 66 have died by the end of 5 years.

• Radiation: Of 100 people having radiation therapy for this condition, 0 die in treatment, 23 have died within a year, and 78 have died by the end of 5 years.

Framing Matters: Survival Rate McNeil, Pauker, Sox, & Tversky (1982)

• Surgery: Of 100 people having surgery for this condition, 90 live through the postoperative period, 68 are alive at the end of the first year, and 34 are alive at the end of 5 years.

• Radiation: Of 100 people having radiation therapy for this condition, 100 live through treatment, 77 are alive at the end of the first year, and 22 are alive at the end of 5 years.

BUS143

Loss Aversion in Marketing: Endowment Effect (Thaler)

• Why do businesses offer free trials?

• Loss aversion when trial ends! – Also, laziness/forgetting to cancel (inertia) – One reason stores offer free returns

• Endowment effect: People overvalue what they own – People won’t trade random lottery tickets with each other – Your examples?

• Rocksbox $21 membership credit or StitchFix $20 styling fee credit (credits are use it or lose it)

– This is also why Marie Kondo is a thing (hard to give up stuff)

What are some other examples of setting reference points strategically?

• HINT: This is what part of Write-up Topic 1 is about

• Some examples: – MSRP’s – Company earnings expectations (profits $/share) – Negotiating

BUS143

General Strategies for Reference-Setting

• Provide people with frames that look good – But be careful of high expectations

• Discover customers’ reference points, which may differ by segments – Your current customers – Competition’s current customers (switchers) – New customers (to the product category)

• Be sensitive to when reference points are updated

Multi-dimension Reference Dependence

• Each dimension has its own λ and reference point • Example:

– How much would you pay for one more year of life expectancy?

– How much would I have to pay you to have one less year of life expectancy?

• More loss aversion for attributes that are: – Important – Hedonic (related to happiness) – Difficult to tradeoff for money

BUS143

Loss aversion across attributes: Car buying

1. You are about to buy a new car and are considering a model that does not have/has side-impact airbags. These are bags which will deploy in case of a collision, lowering the chance of injury. Another model of the same car, identical in every other respect, has/does not have side-impact airbags.

2. You are about to buy a new car and are considering a model that gets 25/33 miles per gallon. Another model of the same car, identical in every other respect, gets 33/25 miles per gallon.

• How much more would you be willing to pay for that car? • How much cheaper would that car have to be, for you to

switch?

Loss aversion across attributes: Your data (medians)

• Price to buy side-impact airbags: $700 • Price to give them up: $2500 • λ(air bags) = 2500/700 = 3.6

• Price to buy better gas mileage (2533): $1200 • Price to give it up (3325): $2000 • λ(gas mileage) = 1.66 • Assuming $3/gallon, 10k miles/year, $291

difference a year

BUS143

Applying Prospect Theory: Approaches to Risk Attitudes

• In expected utility theory, it is all about the shape of the utility function – Convex: risk-seeking – Concave: risk-aversion

• In Prospect Theory, it depends on the reference point – For gains, risk-aversion – For losses, risk-seeking

Total Wealth

U ti li ty

GainsLosses

People hate losses and try to avoid them

• Risk attitude depends on reference point – For gains, risk-averse – For losses, risk-seeking

GainsLosses

BUS143

Risk Attitudes for Gains and Losses Example 2: Risky gambles

A: A sure gain of $240 B: A 25% chance to gain $1000, and a 75% chance to gain

nothing

C: A sure loss of $750 D: A 75% chance to lose $1000, and a 25% chance to lose

nothing

E: A 75% chance of losing $760 and a 25% chance of gaining $240

F: A 75% chance of losing $750 and a 25% chance of gaining $250

$240 $240

‐$1000 $0

‐$760 $240

$0 $1000

‐$750 ‐$750

‐$750 $250

75% chance 25% chance

E: A & D

F: B & C

75% chance 25% chance

Why this is inconsistent

BUS143

Sunk Cost Fallacy

• Normative: Econs only includes future costs and benefits in computing net present value (NPV)

• Descriptive: Humans pay attention to sunk costs – Related to loss aversion: we hate to “close” a

mental account at a loss – Leads to throwing good money after bad

Sunk Cost Fallacy: Your answers

As the president of an airline company, you have invested $100 million of the company’s money into a research project. The purpose was to build a plane that would not be detected by conventional radar. When the project is 95% completed, another firm begins marketing a plane that cannot be detected by radar. Also, it is apparent that their plane is much faster and more economical than the plane your company is building—pretty much better in every important way.

The question is: Should you invest the last $5 million of research funds to finish your plane?

BUS143

Sunk Costs in Professional Sports

• Do sunk costs affect play time? • Case study: Kwame Brown (2001-2013), 6’11 C

– Drafted by Washington Wizards 1st pick in 2001 (under team president Michael Jordan)

– Average 4.5ppg, 3.5rpg as a rookie – Played for Lakers from 2005-2008, averaging 7.4ppg – SIX other teams gave him a shot! – Career averages: 6.6ppg, 5.5rpg, 0.6bpg

• Camerer & Weber (1999) found that highly drafted players (who are paid a lot of money) get more playing time, controlling for talent

• Any of your own examples of sunk cost fallacy? (sports or otherwise)

Why do these countries differ so much?

Proportion of people classified as organ donors

4.25

27.5

17.17 12

99.98 98 99.91 99.997 99.5 99.64

85.9

100

BUS143

Manage defaults wisely…

• iPhone and AT&T’s billing – Many people were surprised

to see bills dozens of pages long.

– One user received a large box with over 300 pages.

– Most entries: “1 kb download (time) $0.00.”

• Why? The default at signup was for itemized bills!

Loss Aversion, Compensation and Targets

• Many people work to a target. • Sometimes these are explicit:

– Sales targets – Productivity quotas

• Sometimes these are self imposed: – Income or savings goals – Effort goals

BUS143

How will targets affect behavior?

• Hint #1: The target is a reference point • Hint #2: Not meeting the target is a loss

• So, first, people will work much harder until the target is met – But then they’ll slack off

• Question: Why is so hard to get a cab when it rains in New York?

A general point about goal-setting and goal-striving

– Targets above current performance…

– Motivate people (more) to reach them

– Motivate them less when they are reached

– Should be aware of both effects on motivation

– Note: Extremely high targets, not met, can be demotivating

BUS143

Summary: Loss Aversion Matters

• The power of the status quo as a reference point – Endowment effect – Defaults – Sunk costs

• Framing effects (changing reference points) – Targets/aspirations – Risk preferences

• Marketing and persuasion – Market segmentation – Product positioning

Ye’s Keys: Topic 1

1. Fast, automatic System 1 is audited by slower, analytic System 2, but not always that successfully.

2. Loss aversion makes people reluctant to switch from the status quo (e.g., endowment effect, trial pricing, defaults) and makes them do things they otherwise wouldn’t (e.g., sunk cost effect, increased risk taking)

3. Losses are in the eye of the beholder, and (largely) in the hands of the reference- setter. You can use framing to change reference points and therefore behavior.