Chapter One Chapter One Behavioral Foundations

The main objective of this chapter is for students to demonstrate that they can de- scribe 10 specific psychological phenomena that impact the judgments and choices that normal people make about decision tasks that feature risk. These phenomena areBehavioral divided into two groups: heuristics andFoundations biases, and framing effects. After completing this chapter students will be able to: The main objective of this chapter is for students to demonstrate that they can de- scribe1. Identify 10 specific key biases psychological that typically phenomenaarise when people that impact make thejudgments judgments about and risk. choices that2. Explain normal why people reliance make on heuristics,about decision while unavoidable,tasks that feature leaves risk. people These vulnerable phenomena to aremaking divided biased into two judgments. groups: heuristics and biases, and framing effects. After3. Recognize completing that this framing chapter effects students lead willpeople’s be able attitudes to: toward taking risk to be highly dependent on the circumstances in which they find themselves. 1. Identify key biases that typically arise when people make judgments about risk. 2. Explain why reliance on heuristics, while unavoidable, leaves people vulnerable to 1.1 TRADITIONAL makingCORPORATE biased judgments. FINANCIAL DECISIONS AND PSYCHOLOGICAL3. Recognize that TASKS framing effects lead people’s attitudes toward taking risk to be highly dependent on the circumstances in which they find themselves. A wealth of psychological research over the last 50 years provides important in- 1.1 TRADITIONALsights CORPORATE into how normal FINANCIAL people make judgments DECISIONS and choices about risky alterna- tives. Over the same period, research conducted by financial academics provides AND PSYCHOLOGICALimportant insights TASKS into how financial executives should consider making judgments and decisions about corporate risk, and to some extent what executives do in prac- tice.A wealth This of book psychological will help you research to connect over the these last two 50 researchyears provides streams, important and in thein- coursesights intoof doing how sonormal it will people provide make you withjudgments insights and about choices the degree about torisky which alterna- psy- chologicaltives. Over pitfalls the same create period, imperfections research conductedin corporate by decision financial making academics and reductions provides inimportant corporate insights value. into The how book financial will also executives provide guidance should consider about how making to create judgments value byand mitigating decisions vulnerabilityabout corporate to theserisk, andpitfalls. to some extent what executives do in prac- tice.To This help bookfix ideas, will helpconsider you a to small connect example. these Imagine two research a firm streams, is contemplating and in the a newcourse one-year of doing project so it willwhich provide requires you it withto spend insights $3.2 about million the at degree the end to ofwhich this year,psy- inchological the anticipation pitfalls createof receiving imperfections a return inat corporatethe end of decision the following making year. and The reductions project isin cyclical,corporate as value. the financial The book executives will also at provide the firm guidance believe thatabout the how actual to createvalue ofvalue the futureby mitigating cash flow vulnerability from the project to these will pitfalls. strongly depend on the state of the economy duringTo help the nextfix ideas, year. Theconsider project a small also featuresexample. high Imagine risk, asa firmcash isflow contemplating will be espe a- newcially one-year high if the project economy which is requiresstrong but it toespecially spend $3.2 low million if the economy at the end is ofweak. this Typiyear,- incally, the foranticipation projects of of comparable receiving a risk, return the at firm the seeksend of an the expected following return year. of The35 percent. project is cyclical, as the financial executives at the firm believe that the actual value of the1 future cash flow from the project will strongly depend on the state of the economy during the year. The project also features high risk, as cash flow will be espe- cially high if the economy is strong but especially low if the economy is weak. Typi- cally, for projects of comparable risk, the firm seeks an expected return of 35 percent. 1 she77208_ch01_001-031.indd 1 2/7/17 5:25 PM

she77208_ch01_001-031.indd 1 2/7/17 5:25 PM 2 Chapter One Behavioral Foundations 3

Suppose that the economy was stagnant last year, and that economists’ current $3.4 million. Notably, $3.4 million exceeds the $3.2 million required capital investment, outlook for next year is that there is a very high chance that growth will continue which according to traditional textbook analysis suggests that the financial executives 2 Chapter One to be stagnant. Suppose too that most economists agree that there is a low chance should adopt the project. of moderate growth and a very low chance of a boom. The consensus view is that There are two important psychological points to note in the example. The first a Supposerecession that is thenot economylikely, but was if stagnantone materializes last year, andit is that half economists’ as likely to current be severe point pertains to the fact that the financial executives were not given quantitative prob- outlookas mild. for next year is that there is a very high chance that growth will continue abilities, but instead had to make their own subjective judgments about what numbers to beThe stagnant. financial Suppose executives too that typically most economists rely on these agree forecasts that there to form is a lowtheir chance own judg- to use, based on a mix of qualitative and quantitative information available to them. ofments, moderate which growth they andcombine a very with low otherchance information of a boom. such The asconsensus recent stock view market is that per- Notice from Exhibit 1-1 that executives assign a probability of 6 percent to a boom, a formance.recession isIn notthis likely, regard, but the if stockone materializes market increased it is half by as25 likelypercent to duringbe severe the last which is much higher than the 1 percent they assign to low-to-moderate growth. Given as mild.twelve months, the stock of the average firm in their industry increased by 18 per- that economists characterized low-to-moderate growth as being more likely than a cent,The andfinancial the stock executives of their typically own firm rely increased on these byforecasts 12 percent. to form their own judg- boom, financial executives’ probability assignment might reflect judgmental biases ments,In assessingwhich they their combine proposed with other project, information the executives such as estimaterecent stock that market its cash per- flows or a psychological tendency to overweight probabilities of extreme events. A similar formance.will be $46.9 In this million regard, in thethe stockevent market of a boom, increased $31.9 by million 25 percent if the during economy the exhibitslast remark applies to the probabilities that executives assign to a recession. twelvemoderate months, growth, the stock$1.9 ofmillion the average if the economyfirm in their is stagnant,industry increased−$8 million by 18in per-the case The second point concerns the gains and losses associated with the project, where cent,of a andmild the recession, stock of theirand −$13own firm in the increased case of bya severe 12 percent. recession. the gain is expressed as “payback” defined as cash flow minus initial investment. InPut assessing yourself their in the proposed position project, of a financial the executives executive estimate who has that the its responsibility cash flows of See the rightmost column in Exhibit 1-1. Notice that the probability of the project willdeciding be $46.9 whether million to in adopt the event this ofproject. a boom, Suppose $31.9 millionthat you if placethe economy a lot of exhibits trust in the incurring a loss is 93 percent, with gains being generated if the economy grows. moderatecash flow growth, estimates, $1.9 asmillion well ifas the in theeconomy economists’ is stagnant, general −$8 assessment. million in the Would case you Psychologically, being super sensitive to losses might lead the executives to reject offeel a mild that recession,you have enoughand −$13 information in the case toof makea severe an recession.informed decision about whether the project, despite its positive net present value. toPut adopt yourself the project? in the position of a financial executive who has the responsibility of The two points just mentioned relate to the primary focus of this chapter, whose decidingThe standardwhether totextbook adopt this approach project. to Suppose project selectionthat you place revolves a lot around of trust discounted in the purpose is to introduce ten specific psychological phenomena that underlie the be- cashexpected flow estimates,cash flows. as Inwell theory, as in the economists’computation general of expected assessment. cash flows Would would you re- havioral approach to corporate finance. To be sure, there are other psychological feelquire that quantitative you have enough probabilities information for the to makealternative an informed states ofdecision the economy, about whether not simply phenomena discussed in the book, besides the ones introduced in this chapter. How- tothe adopt qualitative the project? descriptors mentioned above. Therefore, financial executives who ever, the ten phenomena that form the heart of this chapter provide the foundational soughtThe standard to base theirtextbook decision approach on the to kindproject of selectionreasoning revolves found in around traditional discounted corporate material for everything else in the book. expectedfinance cashtextbooks flows. would In theory, need the to computation convert the ofeconomists’ expected cash qualitative flows would descriptors, re- Notably, these ten phenomena impact investors just as they impact the managers quirealong quantitative with other probabilities information for such the alternativeas stock market states of performance, the economy, intonot simplynumerical of a firm. Investors also form judgments about firms’ future cash flows, especially theprobabilities. qualitative descriptorsImagine that mentioned they do above.so, and Therefore, assign a probabilityfinancial executives of 6 percent who to a when they face choices about trading the securities of those firms. An important soughtboom, to 1 base percent their to decision moderate on thegrowth, kind of90 reasoning percent to found a stagnant in traditional economy, corporate 1 percent issue that arises in the behavioral approach is that both managers and investors often financeto a mild textbooks recession, would and need2 percent to convert to a severe the economists’ recession. Exhibit qualitative 1-1 summarizesdescriptors, the need to consider the psychological phenomena impacting other people, as well as alonginformation with other in this information example. such as stock market performance, into numerical themselves, when making their own judgments and decisions. probabilities.Based on Imagine the executives’ that they probabilitydo so, and judgments,assign a probability the expected of 6 percent cash flow to a from boom,the project 1 percent would to moderatebe $4.5 million, growth, whose 90 percent present to avalue stagnant at 35 economy, percent turns1 percent out to be Dual Systems to a mild recession, and 2 percent to a severe recession. Exhibit 1-1 summarizes the heuristic In this chapter you will learn about biases associated with heuristics, or rules of thumb, information in this example. A rule of thumb used to upon which people rely in order to arrive at their judgments and make decisions, as EXHIBIT 1-1 Forecasted CashBased Flows on for the a Hypothetical executives’ Project, probability with judgments, Associated theGains expected and Losses, cash Conditional flow from on arrive at a judgment or well as how people’s choices are influenced by the manner in which they frame their make a decision. the Future theState project of the wouldEconomy be Along$4.5 million, with Economists’ whose present Qualitative value atProbabilistic 35 percent Assessments turns out to andbe decision tasks with respect to gains and losses. Accordingly, the remainder of this Managers’ Quantitative Probabilistic Assessments frame chapter is organized into four main sections, the first three of which are respectively Synonymous for EXHIBIT 1-1 Economists’ Probability Managers’ Probability Cash Flow devoted to biases, heuristics, and framing effects; these sections introduce and de- Future State ofForecasted Economy Cash FlowsAssessments for a Hypothetical Project, withAssessments Associated Gains andCash Losses, Flow Conditional(Gain/Loss) on description. the Future State of the Economy Along with Economists’ Qualitative Probabilistic Assessments and scribe 10 specific psychological phenomena. The fourth section discusses psychologi- Severe recessionManagers’ QuantitativeHighly unlikelyProbabilistic Assessments 2% –$13.1 –$16.3 cal and neurological frameworks to explain the factors underlying people’s judgments Mild recession Not likely 1% –$8.1 –$11.3 System 1 and decisions about risk, as well as ways to improve both of these tasks. Stagnant economy Economists’Very high chance Probability Managers’ Probability90% $1.9 Cash Flow–$1.3 Many aspects of the 10 phenomena take place below the level of consciousness. FutureLow-to-moderate State of Economy growth AssessmentsLow chance Assessments1% Cash Flow$31.9 (Gain/Loss)$28.7 The fast, automatic, and in- Boom Very low chance 6% $46.9 $43.7 tuitive processes associated In this regard, psychologists suggest that we can think of our brains as having two Severe recession Highly unlikely 2% –$13.1 –$16.3 with thinking. systems, an intuitive system they call System 1, which is fast, and a more deliberative Mild recession Not likely 1% –$8.1 –$11.3 Stagnant economy Very high chance 90% $1.9 –$1.3 Low-to-moderate growth Low chance 1% $31.9 $28.7 Boom Very low chance 6% $46.9 $43.7

she77208_ch01_001-031.indd 2 she77208_ch01_001-031.indd2/7/17 5:25 PM 3 2/7/17 5:25 PM

she77208_ch01_001-031.indd 2 2/7/17 5:25 PM 2 Chapter One Behavioral Foundations 3

Suppose that the economy was stagnant last year, and that economists’ current $3.4 million. Notably, $3.4 million exceeds the $3.2 million required capital investment, outlook for next year is that there is a very high chance that growth will continue which according to traditional textbook analysis suggests that the financial executives to be stagnant. Suppose too that most economists agree that there is a low chance should adopt the project. Behavioral Foundations 3 of moderate growth and a very low chance of a boom. The consensus view is that There are two important psychological points to note in the example. The first a recession is not likely, but if one materializes it is half as likely to be severe point$3.4 pertains million. to Notably,the fact that$3.4 the million financial exceeds executives the $3.2 weremillion not required given quantitativecapital investment, prob- as mild. abilities,which but according instead tohad traditional to make textbooktheir own analysis subjective suggests judgments that the about financial what executives numbers The financial executives typically rely on these forecasts to form their own judg- to use,should based adopt on the a mix project. of qualitative and quantitative information available to them. ments, which they combine with other information such as recent stock market per- NoticeThere from Exhibit are two 1-1 important that executives psychological assign points a probability to note inof the6 percent example. to Thea boom, first formance. In this regard, the stock market increased by 25 percent during the last whichpoint is muchpertains higher to the than fact thethat 1 the percent financial they executives assign to low-to-moderatewere not given quantitative growth. Given prob- twelve months, the stock of the average firm in their industry increased by 18 per- thatabilities, economists but insteadcharacterized had to make low-to-moderate their own subjective growth judgments as being aboutmore whatlikely numbers than a cent, and the stock of their own firm increased by 12 percent. boom,to use, financial based executives’on a mix of probabilityqualitative andassignment quantitative might information reflect judgmental available to biases them. In assessing their proposed project, the executives estimate that its cash flows or aNotice psychological from Exhibit tendency 1-1 that to overweightexecutives assignprobabilities a probability of extreme of 6 percent events. to A asimilar boom, will be $46.9 million in the event of a boom, $31.9 million if the economy exhibits remarkwhich applies is much to higherthe probabilities than the 1 percentthat executives they assign assign to low-to-moderate to a recession. growth. Given moderate growth, $1.9 million if the economy is stagnant, −$8 million in the case Thethat secondeconomists point characterized concerns the low-to-moderate gains and losses growthassociated as being with themore project, likely wherethan a of a mild recession, and −$13 in the case of a severe recession. the boom,gain is financial expressed executives’ as “payback” probability defined assignment as cash flowmight minus reflect initial judgmental investment. biases Put yourself in the position of a financial executive who has the responsibility of Seeor the a psychologicalrightmost column tendency in Exhibit to overweight 1-1. Notice probabilities that the of probability extreme events. of the A project similar deciding whether to adopt this project. Suppose that you place a lot of trust in the incurringremark a applies loss is to 93 the percent, probabilities with thatgains executives being generated assign to aif recession. the economy grows. cash flow estimates, as well as in the economists’ general assessment. Would you Psychologically,The second being point concernssuper sensitive the gains to andlosses losses might associated lead the with executives the project, to wherereject feel that you have enough information to make an informed decision about whether the theproject, gain despiteis expressed its positive as “payback” net present defined value. as cash flow minus initial investment. to adopt the project? TheSee twothe rightmostpoints just column mentioned in Exhibit relate 1-1. to the Notice primary that focusthe probability of this chapter, of the projectwhose The standard textbook approach to project selection revolves around discounted purposeincurring is to aintroduce loss is 93 ten percent, specific with psychological gains being phenomenagenerated if thatthe underlieeconomy thegrows. be- expected cash flows. In theory, the computation of expected cash flows would re- havioralPsychologically, approach to being corporate super sensitivefinance. toTo losses be sure, might there lead are the other executives psychological to reject quire quantitative probabilities for the alternative states of the economy, not simply phenomenathe project, discussed despite inits thepositive book, net besides present the value. ones introduced in this chapter. How- the qualitative descriptors mentioned above. Therefore, financial executives who ever, theThe ten two phenomena points just thatmentioned form the relate heart to of the this primary chapter focus provide of this the chapter, foundational whose sought to base their decision on the kind of reasoning found in traditional corporate materialpurpose for iseverything to introduce else ten in specific the book. psychological phenomena that underlie the be- finance textbooks would need to convert the economists’ qualitative descriptors, Notably,havioral theseapproach ten phenomenato corporate impactfinance. investors To be sure, just asthere they are impact other thepsychological managers along with other information such as stock market performance, into numerical of aphenomena firm. Investors discussed also formin the judgmentsbook, besides about the firms’ones introduced future cash in thisflows, chapter. especially How- probabilities. Imagine that they do so, and assign a probability of 6 percent to a whenever, they the face ten phenomenachoices about that tradingform the the heart securities of this chapter of those provide firms. the An foundational important boom, 1 percent to moderate growth, 90 percent to a stagnant economy, 1 percent issuematerial that arises for everything in the behavioral else in theapproach book. is that both managers and investors often to a mild recession, and 2 percent to a severe recession. Exhibit 1-1 summarizes the need toNotably, consider these the tenpsychological phenomena impactphenomena investors impacting just as theyother impact people, the as managers well as information in this example. themselves,of a firm. when Investors making also their form own judgments judgments about and firms’ decisions. future cash flows, especially Based on the executives’ probability judgments, the expected cash flow from when they face choices about trading the securities of those firms. An important the project would be $4.5 million, whose present value at 35 percent turns out to be Dualissue Systems that arises in the behavioral approach is that both managers and investors often heuristic In thisneed chapter to consider you will the learn psychological about biases phenomena associated impacting with heuristics, other people,or rules asof thumb,well as A rule of thumb used to uponthemselves, which people when rely making in order their to own arrive judgments at their andjudgments decisions. and make decisions, as EXHIBIT 1-1 Forecasted Cash Flows for a Hypothetical Project, with Associated Gains and Losses, Conditional on arrive at a judgment or well as how people’s choices are influenced by the manner in which they frame their make a decision. Dual Systems the Future State of the Economy Along with Economists’ Qualitative Probabilistic Assessments and decision tasks with respect to gains and losses. Accordingly, the remainder of this Managers’ Quantitative Probabilistic Assessments frameheuristic chapterIn this is chapterorganized you into will fourlearn main about sections, biases associated the first withthree heuristics, of which or are rules respectively of thumb, Economists’ Probability Managers’ Probability Cash Flow SynonymousA rule of for thumb used to upon which people rely in order to arrive at their judgments and make decisions, as arrive at a judgment or devoted to biases, heuristics, and framing effects; these sections introduce and de- Future State of Economy Assessments Assessments Cash Flow (Gain/Loss) description. well as how people’s choices are influenced by the manner in which they frame their make a decision. scribe 10 specific psychological phenomena. The fourth section discusses psychologi- Severe recession Highly unlikely 2% –$13.1 –$16.3 cal anddecision neurological tasks with frameworks respect to togains explain and thelosses. factors Accordingly, underlying the people’s remainder judgments of this frame Mild recession Not likely 1% –$8.1 –$11.3 and chapterdecisions is organizedabout risk, into as wellfour mainas ways sections, to improve the first both three of theseof which tasks. are respectively SystemSynonymous 1 for Stagnant economy Very high chance 90% $1.9 –$1.3 Manydevoted aspects to biases, of the heuristics, 10 phenomena and framing take placeeffects; below these the sections level ofintroduce consciousness. and de- Low-to-moderate growth Low chance 1% $31.9 $28.7 The fast,description. automatic, and in- scribe 10 specific psychological phenomena. The fourth section discusses psychologi- Boom Very low chance 6% $46.9 $43.7 tuitive processes associated In this regard, psychologists suggest that we can think of our brains as having two with thinking. systems,cal and an neurological intuitive system frameworks they call to Systemexplain the 1, which factors is underlying fast, and a people’s more deliberative judgments System 1 and decisions about risk, as well as ways to improve both of these tasks. The fast, automatic, and in- Many aspects of the 10 phenomena take place below the level of consciousness. tuitive processes associated In this regard, psychologists suggest that we can think of our brains as having two with thinking. systems, an intuitive system they call System 1, which is fast, and a more deliberative

she77208_ch01_001-031.indd 2 she77208_ch01_001-031.indd2/7/17 5:25 PM 3 2/7/17 5:25 PM

she77208_ch01_001-031.indd 3 2/7/17 5:25 PM 4 Chapter One Behavioral Foundations 5

System 2 system they call System 2, which is slow. Although both systems engage in calcu- Diagnostic Question The slow, deliberate, con- lation, we are consciously more aware of the calculations when they are done by Consider the following four specific events that might happen to you during your 4scious Chapter processes One associated with thinking. System 2, such as computing the present value of a stream of cash flows. The cal- lifetime and answer the question that appears thereafter. culations performed by System 1 are more automatic, such as when making a turn System 2 systemwhile drivingthey call a System car or running 2, which to is catch slow. a Althoughball. both systems engage in calcu- 1. Being fired from a job The slow, deliberate, con- lation,The we operations are consciously that System more aware1 makes of theare calculationsquite impressive, when giventhey are the done absence by of 2. Your work recognized with award scious processes associated conscious calculation. At the same time, human intuition is imperfect, which is why 3. Living past 80 with thinking. System 2, such as computing the present value of a stream of cash flows. The cal- culationswe are prone performed to make by Systemerrors in 1 judgmentare more automatic, and choices such that as arewhen imprudent. making a It turn is easy 4. Having your car stolen whileto suggest driving that a car the or remedy running is to to catch rely ona ball. System 2 for judgment and decision making. However,The operations System that 2 requires System much1 makes more are effort quite thanimpressive, System given 1, and the we absence typically of lack Compared to other people of the same gender as you in this class, what do you consciousthe mental calculation. resources At and the capabilities same time, necessaryhuman intuition for System is imperfect, 2 perfection. which is why think are the chances that each of these events will happen to you? The choices we areThe prone term to “dual make systems”errors in judgment is shorthand and choices for the that System are imprudent. 1/System It 2 is conceptual easy range from much less than average, through average, to much more than average. toframework. suggest that An the important remedy is toreason rely onto Systemstudy behavioral 2 for judgment corporate and decision finance making. is to inves- Enter a column of numbers from 1 to 4 on a blank page, and record your answers However,tigate how System dual systems2 requires influence much more judgments effort than and System decisions 1, and that we are typically made lackin a cor- next to each event number. For example, consider event 1, being fired from a job. If theporate mental financial resources environment. and capabilities In this necessary regard, formuch System of what 2 perfection. is taught in traditional you think that being fired from a job is as likely to happen to you as to anyone else, corporateThe term finance “dual systems”textbooks is pertains shorthand to processes for the System that System 1/System 2 carries 2 conceptual out. Behav- record a 7 beside event 1. framework.ioral corporate An important finance addsreason explicit to study consideration behavioral corporate of System finance 1 issues is to to inves- the mix, 1. 100% less (no chance) 9. 20% more tigatewith howthe intent dual systems to help influencemanagers judgmentsmake better and decisions. decisions that are made in a cor- 2. 80% less 10. 40% more porateThe financial next three environment. sections discuss In this the regard, results much of experimentsof what is taught that psychologists in traditional have 3. 60% less 11. 60% more corporateused to uncover finance textbooksthe 10 psychological pertains to processesphenomena that that System form 2 the carries heart out. of thisBehav- chapter. ioralIn order corporate to appreciate finance addsthe discussion explicit consideration that follows, of it Systemis vitally 1 issuesimportant to the to mix,complete 4. 40% less 12. 80% more withthe behavioralthe intent to questionnaire, help managers which make is better available decisions. in Additional Resources to Chapter 1. 5. 20% less 13. 100% more ThisThe questionnaire next three sections can be discuss downloaded the results from of experimentsthe book web that site psychologists at www.mhhe.com have 6. 10% less 14. 3 times average used/shefrin2e. to uncover After the completing 10 psychological the behavioral phenomena questionnaire, that form the please heart continueof this chapter. your read - 7. Average 15. 5 times average Ining order of this to appreciatechapter, which the discussion provides athat detailed follows, discussion it is vitally of eachimportant of the to 10 complete phenomena. 8. 10% more the behavioralEach of the questionnaire, next three sections which is isavailable divided in into Additional subsections, Resources with toeach Chapter subsection 1. Thisintroducing questionnaire one ofcan the be ten downloaded foundational from psychological the book web phenomena. site at www.mhhe.com The subsections Discussion /shefrin2e.feature a three-pronged After completing 3-D the structure, behavioral namely: questionnaire, description, please diagnostic continue your question, read- and In the preceding diagnostic question, people can rate how likely they are to experi- ingdiscussion. of this chapter, A subsection which provides begins awith detailed a brief discussion description of each of aof psychological the 10 phenomena. phenom- ence particular events relative to other people who are similar to them. A rating of enon.Each This of the is next followed three sectionsby an example is divided of intoa diagnostic subsections, question with each that subsection psychologists 7 means that a person feels that an event is as likely to happen to them as to anyone introducinghave used toone identify of the tenthe foundationalphenomenon psychological in question. Thephenomena. ensuing Thediscussion subsections indicates else in similar circumstances.1 Some of the events are favorable, and some are unfa- featurehow people a three-pronged generally 3-Drespond structure, to the namely: question, description, and what diagnosticthe response question, pattern and means vorable. The unfavorable events are being fired from a job and having your car stolen. discussion.for the phenomenon A subsection being begins discussed. with a brief description of a psychological phenom- The favorable events are your work recognized with an award, and living past 80. enon. This is followed by an example of a diagnostic question that psychologists The four events under discussion are part of a general study involving 42 life 1.2 BIASES have used to identify the phenomenon in question. The ensuing discussion indicates events, of which 18 are favorable and 24 are unfavorable. The 42 events are de- how people generally respond to the question, and what the response pattern means scribed in the Additional Resources to Chapter 1, which is available online at the for the phenomenon being discussed. book web site. The questionnaire you answered as part of this chapter comprised Bias A bias is a predisposition towards making a specific type of error. A predisposition toward an 18-question subset of the set of 42. If everyone held objectively correct beliefs, 1.2making a BIASES specific type Excessive Optimism then the average response across the class for all events should be 7. Typically the of error. average rating for the unfavorable events is below 7, while the average rating for Description 2 Bias A bias is a predisposition towards making a specific type of error. the favorable events is above 7. This means that people believe that unfavorable A predisposition toward Psychologists have concluded that people are excessively optimistic, by which they events are less likely to happen to them than to other people, but favorable events making a specific type Excessivemean unrealistically Optimism optimistic. More precisely, people tend to overestimate how fre- are more likely to happen to them than to other people. While this may be true for of error. Descriptionquently they will experience favorable outcomes and underestimate how frequently some people, it cannot be true for everyone. The general conclusion is that people they will experience unfavorable outcomes. tend to be excessively optimistic.3 Psychologists have concluded that people are excessively optimistic, by which they mean unrealistically optimistic. More precisely, people tend to overestimate how fre- quently they will experience favorable outcomes and underestimate how frequently they will experience unfavorable outcomes.

she77208_ch01_001-031.indd 4 she77208_ch01_001-031.indd2/7/17 5:25 PM 5 2/7/17 5:25 PM

she77208_ch01_001-031.indd 4 2/7/17 5:25 PM 4 Chapter One Behavioral Foundations 5

System 2 system they call System 2, which is slow. Although both systems engage in calcu- Diagnostic Question The slow, deliberate, con- lation, we are consciously more aware of the calculations when they are done by Consider the following four specific events that might happen to you during your scious processes associated Behavioral Foundations 5 with thinking. System 2, such as computing the present value of a stream of cash flows. The cal- lifetime and answer the question that appears thereafter. culations performed by System 1 are more automatic, such as when making a turn while driving a car or running to catch a ball. 1. BeingDiagnostic fired fromQuestion a job The operations that System 1 makes are quite impressive, given the absence of 2. YourConsider work the recognized following withfour awardspecific events that might happen to you during your conscious calculation. At the same time, human intuition is imperfect, which is why 3. Livinglifetime past and 80 answer the question that appears thereafter. we are prone to make errors in judgment and choices that are imprudent. It is easy 4. Having your car stolen to suggest that the remedy is to rely on System 2 for judgment and decision making. 1. Being fired from a job However, System 2 requires much more effort than System 1, and we typically lack Compared2. Your towork other recognized people of with the award same gender as you in this class, what do you the mental resources and capabilities necessary for System 2 perfection. think3. are Living the pastchances 80 that each of these events will happen to you? The choices range from much less than average, through average, to much more than average. The term “dual systems” is shorthand for the System 1/System 2 conceptual 4. Having your car stolen framework. An important reason to study behavioral corporate finance is to inves- Enter a column of numbers from 1 to 4 on a blank page, and record your answers tigate how dual systems influence judgments and decisions that are made in a cor- nextCompared to each event to other number. people For of example, the same consider gender eventas you 1, in being this class,fired fromwhat doa job. you If porate financial environment. In this regard, much of what is taught in traditional youthink think are that the being chances fired that from each a job of isthese as likely events to willhappen happen to youto you?as to Theanyone choices else, corporate finance textbooks pertains to processes that System 2 carries out. Behav- recordrange a 7 from beside much event less 1. than average, through average, to much more than average. ioral corporate finance adds explicit consideration of System 1 issues to the mix, 1. 100%Enter lessa column (no chance) of numbers from 1 to 9. 4 20% on a moreblank page, and record your answers next to each event number. For example, consider event 1, being fired from a job. If with the intent to help managers make better decisions. 2. 80% less 10. 40% more The next three sections discuss the results of experiments that psychologists have you think that being fired from a job is as likely to happen to you as to anyone else, 3. 60% less 11. 60% more used to uncover the 10 psychological phenomena that form the heart of this chapter. record a 7 beside event 1. 4. 40% less 12. 80% more In order to appreciate the discussion that follows, it is vitally important to complete 1. 100% less (no chance) 9. 20% more 13. 100% more the behavioral questionnaire, which is available in Additional Resources to Chapter 1. 5. 20%2. 80% less less 10. 40% more This questionnaire can be downloaded from the book web site at www.mhhe.com 6. 10%3. 60% less less 14.11. 3 60%times more average /shefrin2e. After completing the behavioral questionnaire, please continue your read- 7. Average4. 40% less 15.12. 5 80%times more average ing of this chapter, which provides a detailed discussion of each of the 10 phenomena. 8. 10% more Each of the next three sections is divided into subsections, with each subsection 5. 20% less 13. 100% more introducing one of the ten foundational psychological phenomena. The subsections Discussion6. 10% less 14. 3 times average feature a three-pronged 3-D structure, namely: description, diagnostic question, and In the7. Averagepreceding diagnostic question, people15. 5 cantimes rate average how likely they are to experi- discussion. A subsection begins with a brief description of a psychological phenom- ence8. particular 10% more events relative to other people who are similar to them. A rating of enon. This is followed by an example of a diagnostic question that psychologists 7 means that a person feels that an event is as likely to happen to them as to anyone have used to identify the phenomenon in question. The ensuing discussion indicates elseDiscussion in similar circumstances.1 Some of the events are favorable, and some are unfa- how people generally respond to the question, and what the response pattern means vorable.In the The preceding unfavorable diagnostic events question, are being people fired fromcan rate a job how and likely having they your are carto experistolen.- for the phenomenon being discussed. Theence favorable particular events events are yourrelative work to otherrecognized people with who an are award, similar and to livingthem. Apast rating 80. of The7 means four thatevents a person under feels discussion that an eventare part is as of likely a general to happen study to involvingthem as to 42anyone life 1 1.2 BIASES events,else ofin similarwhich circumstances.18 are favorable Some and of 24 the are events unfavorable. are favorable, The and42 eventssome are are unfa de-- scribedvorable. in the The Additional unfavorable Resources events are tobeing Chapter fired from1, which a job isand available having your online car atstolen. the bookThe web favorable site. The events questionnaire are your work you recognized answered withas part an award,of this and chapter living comprised past 80. Bias A bias is a predisposition towards making a specific type of error. an 18-questionThe four subsetevents ofunder the discussionset of 42. Ifare everyone part of aheld general objectively study involving correct beliefs, 42 life A predisposition toward events, of which 18 are favorable and 24 are unfavorable. The 42 events are de- making a specific type Excessive Optimism then the average response across the class for all events should be 7. Typically the of error. averagescribed rating in the for Additional the unfavorable Resources events to Chapteris below 1, 7, which while is theavailable average online rating at forthe Description the bookfavorable web site.events The is questionnaireabove 7.2 This you means answered that aspeople part ofbelieve this chapter that unfavorable comprised Psychologists have concluded that people are excessively optimistic, by which they eventsan 18-questionare less likely subset to happenof the set to ofthem 42. thanIf everyone to other held people, objectively but favorable correct beliefs,events mean unrealistically optimistic. More precisely, people tend to overestimate how fre- are thenmore the likely average to happen response to acrossthem than the classto other for allpeople. events While should this be may7. Typically be true forthe average rating for the unfavorable events is below 7, while the average rating for quently they will experience favorable outcomes and underestimate how frequently some people, it cannot be true for 2everyone. The general conclusion is that people they will experience unfavorable outcomes. tendthe to favorablebe excessively events optimistic. is above 7.3 This means that people believe that unfavorable events are less likely to happen to them than to other people, but favorable events are more likely to happen to them than to other people. While this may be true for some people, it cannot be true for everyone. The general conclusion is that people tend to be excessively optimistic.3

she77208_ch01_001-031.indd 4 she77208_ch01_001-031.indd2/7/17 5:25 PM 5 2/7/17 5:25 PM

she77208_ch01_001-031.indd 5 2/7/17 5:25 PM 6 Chapter One Behavioral Foundations 7

excessive optimism The original study on excessive optimism was conducted in the 1970s and used Diagnostic Question People overestimate undergraduate students at an American university as subjects. The general results Based on your own knowledge, and without using any additional sources, what 6how Chapter frequently One they will experience favorable have been replicated many times and are robust, applying both to undergraduate is your best guess about the length, in miles, of the Amazon River? This is a outcomes and underestimate students and to working professionals of all ages, both in the United States and in- difficult question for most people to answer, and it is unrealistic to assume that excessivehow frequently optimism they will ternationally.The original Instudy respect on excessive to the four optimism life events was discussed conducted here, in the typical 1970s averageand used class they will provide an accurate response. For this reason, suppose that we also ask Peopleexperience overestimate unfavorable undergraduateresponses for studentsthe question at an about American being fired university lie in asthe subjects. range 3.8 The to 5.2. general This results range cor- for a confidence interval, meaning a low guess and a high guess, with the width of how frequently they will outcomes. haveresponds been to replicated judgments many that times are between and are 44 robust, percent applying and 18 bothpercent to undergraduateless than the aver - the interval reflecting a person’s degree of confidence in his or her response. For experience favorable 4 outcomes and underestimate studentsage, respectively. and to working Typical professionals ranges for of the all other ages, three both questionsin the United are 7.5States to 9.1and for in- your example, we might ask people to structure their confidence intervals so that they how frequently they will ternationally.work being recognizedIn respect to with the anfour award, life events 7.1 to discussed 7.9 for living here, past typical 80, averageand 5.1 classto 6.3 for feel 90 percent confident that their low and high guesses would bracket the answer experience unfavorable responseshaving your for thecar questionstolen. For about these being ranges, fired everylie in theunit range deviation 3.8 to from 5.2. This 7.0 correspondsrange cor- to they would find by conducting an Internet search about the length of the Amazon. outcomes. responds10 percent to judgments in probability. that are between 44 percent and 18 percent less than the aver- If you have not yet done so for the Amazon River question, try it now by providing 4 age, Psychologistsrespectively. Typicalhave identified ranges for a separate the other concept three questions of optimism, are 7.5 called to 9.1 “dispositional for your your best guess, a low guess, and a high guess, structured to define a 90 percent workoptimism,” being recognized which is about with anhaving award, a positive 7.1 to 7.9 general for living outlook past on80, life. and Dispositional5.1 to 6.3 for op- confidence interval. havingtimism your can car be stolen.measured For theseon a scaleranges, of every 0 to 100,unit deviationwith scores from above 7.0 corresponds 50 corresponding to A confidence interval for a single question is somewhat informative. More infor- 10to percent optimism in probability. and scores below 50 corresponding to pessimism. mative is to ask people to provide best guesses and confidence intervals for a series PsychologistsDispositional have optimism identified and a separate excessive concept optimism of optimism, are called positively “dispositional related, but of difficult questions, say 10, so that we can point out that a person who is well cali- optimism,”weakly so. which For is undergraduate about having a positive finance general majors, outlook a 10-point on life. increaseDispositional in disposi-op- brated would expect that for nine of the ten questions, the “true” numbers provided timismtional canoptimism be measured on average on a scale increases of 0 to the 100, probability with scores attached above 50 to corresponding favorable events by an Internet search would lie within the confidence intervals. toby optimism approximately and scores 3 percent, below 50 and corresponding decreases the to pessimism.probability attached to unfavorable eventsDispositional by approximately optimism 2 andpercent. excessive However, optimism the correlation are positively between related, dispositional but Discussion weaklyoptimism so. and For excessive undergraduate optimism finance is low. majors, Moreover, a 10-point for an increase international in disposi- group of An Internet search for the length of the Amazon River yields an answer of tionalinvestment optimism professionals, on average theincreases correlation the probability is effectively attached zero, suggestingto favorable the events need for 4,000 miles. If 4,000 lies between your low guess and your high guess, give bycaution approximately when basing 3 percent, statements and decreases about excessive the probability optimism attached on measures to unfavorable of disposi- yourself a hit. Otherwise, give yourself a miss. The answers to the 10 difficult eventstional by optimism. approximately5 2 percent. However, the correlation between dispositional questions in the behavioral questionnaire can be found in endnote 7 at the end of optimismThe diagnostic and excessive question optimism for optimism is low. Moreover, that involves for anliving international past the age group of 80 of is of this material. People who are well-calibrated should expect to score nine hits for investmentspecial interest. professionals, In practice, the correlation assessments is effectively of length zero,of life suggesting can be used the needto proxy for for the 10 questions.8 cautionoptimism. when The basing use ofstatements this proxy about has excessiveled to the optimismconclusion on that measures excessively of disposi- optimis- When people set their confidence intervals too narrowly, their hit rates can 5 tionaltic people optimism. are more likely to believe that future economic conditions will improve. be expected to be less than nine. Such people are overconfident about their TheyThe arediagnostic also more question entrepreneurial for optimism and that work involves more living hours past than the people age of who 80 is are of less knowledge. When they set their confidence intervals too widely, their hit rates specialoptimistic. interest.6 In practice, assessments of length of life can be used to proxy for can be expected to be 10. Such people are typically underconfident about their optimism. The use of this proxy has led to the conclusion that excessively optimis- knowledge. ticOverconfidence people are more likely to believe that future economic conditions will improve. How well do people do on these questions? Typically, the most frequent number TheyPsychologists are also more have entrepreneurial found that people and work are generallymore hours overconfident than people who when are it less comes of hits, and the average number of hits, is about 4, with the mean being approxi- 6 overconfidence optimistic.to their knowledge and ability to complete difficult tasks. Overconfidence is a mately 4.5. For a typical class, the standard deviation is approximately 2.5: the re- People know less than they Overconfidencebias that pertains to how well people understand the limits of their knowledge, sponse associated with being well calibrated lies two standard deviations above the think they know and view their own abilities, or both. most frequent number of hits. That is, when it comes to difficult questions, people themselves as better than Psychologists have found that people are generally overconfident when it comes they actually are. are typically very overconfident about their knowledge. They do not realize how overconfidence toDescription their knowledge and ability to complete difficult tasks. Overconfidence is a little they know. People know less than they biasPeople that whopertains are overconfidentto how well people about theirunderstand level of the knowledge limits of theirthink knowledge,they know more In respect to the overconfidence questions, what is the percentage of people think they know and view their own abilities, or both. themselves as better than than they actually know. People who are overconfident about their abilities think responding who are well-calibrated? The answer turns out to be about 3.5 percent. they actually are. Descriptionthey are better than they actually are. This overconfidence does not necessarily Virtually everyone else is overconfident. Nevertheless, occasionally someone will mean that these people are ignorant or incompetent. It just means that in their own achieve a hit rate of 100 percent and appear to be underconfident. But this is a rela- overconfident Peopleeyes they who are are smarter and better about than their is level actually of knowledge the case. think they know more tively rare occurrence.9 thanPsychologists they actually know. test for People overconfidence who are overconfident about knowledge about their by asking abilities knowledge- think In order to test for overconfidence in respect to ability, psychologists often pose theybased are questions better than such they as the actually following. are. This7 overconfidence does not necessarily the following question to people. mean that these people are ignorant or incompetent. It just means that in their own eyes they are smarter and better than is actually the case. Psychologists test for overconfidence about knowledge by asking knowledge- based questions such as the following.7

she77208_ch01_001-031.indd 6 she77208_ch01_001-031.indd2/7/17 5:25 PM 7 2/7/17 5:25 PM

she77208_ch01_001-031.indd 6 2/7/17 5:25 PM 6 Chapter One Behavioral Foundations 7 excessive optimism The original study on excessive optimism was conducted in the 1970s and used Diagnostic Question People overestimate undergraduate students at an American university as subjects. The general results Based on your own knowledge, and without using any additional sources, what how frequently they will Behavioral Foundations 7 experience favorable have been replicated many times and are robust, applying both to undergraduate is your best guess about the length, in miles, of the Amazon River? This is a outcomes and underestimate students and to working professionals of all ages, both in the United States and in- difficult question for most people to answer, and it is unrealistic to assume that how frequently they will ternationally. In respect to the four life events discussed here, typical average class theyDiagnostic will provide Question an accurate response. For this reason, suppose that we also ask experience unfavorable responses for the question about being fired lie in the range 3.8 to 5.2. This range cor- for aBased confidence on your interval, own knowledge, meaning aand low without guess and using a high any guess,additional with sources, the width what of outcomes. responds to judgments that are between 44 percent and 18 percent less than the aver- the interval reflecting a person’s degree of confidence in his or her response. For 4 is your best guess about the length, in miles, of the Amazon River? This is a age, respectively. Typical ranges for the other three questions are 7.5 to 9.1 for your example,difficult we question might ask for peoplemost people to structure to answer, their and confidence it is unrealistic intervals to soassume that they that work being recognized with an award, 7.1 to 7.9 for living past 80, and 5.1 to 6.3 for feelthey 90 percent will provide confident an accurate that their response. low and For high this guesses reason, would suppose bracket that wethe alsoanswer ask having your car stolen. For these ranges, every unit deviation from 7.0 corresponds to theyfor would a confidence find by conductinginterval, meaning an Internet a low searchguess and about a high the guess,length with of the the Amazon. width of 10 percent in probability. If youthe haveinterval not reflectingyet done soa person’sfor the Amazon degree of River confidence question, in tryhis itor now her response.by providing For Psychologists have identified a separate concept of optimism, called “dispositional yourexample, best guess, we mighta low askguess, people and to a structurehigh guess, their structured confidence to intervalsdefine a so90 that percent they optimism,” which is about having a positive general outlook on life. Dispositional op- confidencefeel 90 percent interval. confident that their low and high guesses would bracket the answer timism can be measured on a scale of 0 to 100, with scores above 50 corresponding Athey confidence would find interval by conducting for a single an questionInternet search is somewhat about the informative. length of the More Amazon. infor- to optimism and scores below 50 corresponding to pessimism. mativeIf you is tohave ask not people yet done to provide so for the best Amazon guesses River and confidencequestion, try intervals it now by for providing a series Dispositional optimism and excessive optimism are positively related, but of difficultyour best questions, guess, a saylow 10,guess, so that and we a high can pointguess, out structured that a person to define who a is 90 well percent cali- weakly so. For undergraduate finance majors, a 10-point increase in disposi- bratedconfidence would expect interval. that for nine of the ten questions, the “true” numbers provided tional optimism on average increases the probability attached to favorable events by an InternetA confidence search interval would for lie awithin single thequestion confidence is somewhat intervals. informative. More infor- by approximately 3 percent, and decreases the probability attached to unfavorable mative is to ask people to provide best guesses and confidence intervals for a series events by approximately 2 percent. However, the correlation between dispositional Discussionof difficult questions, say 10, so that we can point out that a person who is well cali- optimism and excessive optimism is low. Moreover, for an international group of An brated Internet would search expect for that the for length nine of of the the ten Amazon questions, River the “true” yields numbers an answer provided of investment professionals, the correlation is effectively zero, suggesting the need for 4,000by anmiles. Internet If 4,000 search lies would between lie within your the low confidence guess andintervals. your high guess, give caution when basing statements about excessive optimism on measures of disposi- yourself a hit. Otherwise, give yourself a miss. The answers to the 10 difficult tional optimism.5 questionsDiscussion in the behavioral questionnaire can be found in endnote 7 at the end of The diagnostic question for optimism that involves living past the age of 80 is of thisAn material. Internet People search who for are the well-calibrated length of the Amazonshould expect River to yields score an nine answer hits for of special interest. In practice, assessments of length of life can be used to proxy for the 4,00010 questions. miles. If8 4,000 lies between your low guess and your high guess, give optimism. The use of this proxy has led to the conclusion that excessively optimis- Whenyourself people a hit. set Otherwise, their confidence give yourself intervals a miss. too The narrowly, answers their to the hit 10 rates difficult can tic people are more likely to believe that future economic conditions will improve. be expectedquestions in to the be behavioral less than questionnaire nine. Such people can be arefound overconfident in endnote 7 at about the end their of They are also more entrepreneurial and work more hours than people who are less knowledge.this material. When People they whoset their are well-calibrated confidence intervals should expecttoo widely, to score their nine hit hits rates for 8 optimistic.6 canthe be 10expected questions. to be 10. Such people are typically underconfident about their knowledge.When people set their confidence intervals too narrowly, their hit rates can Overconfidence Howbe expected well do people to be lessdo on than these nine. questions? Such people Typically, are overconfidentthe most frequent about number their Psychologists have found that people are generally overconfident when it comes of hits,knowledge. and the Whenaverage they number set their of hits,confidence is about intervals 4, with toothe widely,mean being their hitapproxi- rates overconfidence to their knowledge and ability to complete difficult tasks. Overconfidence is a matelycan 4.5.be expected For a typical to be class,10. Such the peoplestandard are deviation typically is underconfident approximately about2.5: the their re- People know less than they bias that pertains to how well people understand the limits of their knowledge, sponseknowledge. associated with being well calibrated lies two standard deviations above the think they know and view their own abilities, or both. most frequentHow well number do people of hits.do on That these is, questions? when it comes Typically, to difficult the most questions, frequent numberpeople themselves as better than of hits, and the average number of hits, is about 4, with the mean being approxi- they actually are. are typically very overconfident about their knowledge. They do not realize how Description littlemately they know.4.5. For a typical class, the standard deviation is approximately 2.5: the re- People who are overconfident about their level of knowledge think they know more Insponse respect associated to the with overconfidence being well calibrated questions, lies what two isstandard the percentage deviations ofabove people the than they actually know. People who are overconfident about their abilities think respondingmost frequent who are number well-calibrated? of hits. That Theis, when answer it comes turns outto difficult to be about questions, 3.5 percent. people they are better than they actually are. This overconfidence does not necessarily Virtuallyare typically everyone very else overconfident is overconfident. about theirNevertheless, knowledge. occasionally They do not someone realize howwill mean that these people are ignorant or incompetent. It just means that in their own achievelittle athey hit rateknow. of 100 percent and appear to be underconfident. But this is a rela- eyes they are smarter and better than is actually the case. tively rareIn respect occurrence. to the9 overconfidence questions, what is the percentage of people Psychologists test for overconfidence about knowledge by asking knowledge- In responding order to test who for are overconfidence well-calibrated? in The respect answer to ability,turns out psychologists to be about 3.5 often percent. pose based questions such as the following.7 the Virtuallyfollowing everyone question elseto people. is overconfident. Nevertheless, occasionally someone will achieve a hit rate of 100 percent and appear to be underconfident. But this is a rela- tively rare occurrence.9 In order to test for overconfidence in respect to ability, psychologists often pose the following question to people.

she77208_ch01_001-031.indd 6 she77208_ch01_001-031.indd2/7/17 5:25 PM 7 2/7/17 5:25 PM

she77208_ch01_001-031.indd 7 2/7/17 5:25 PM 8 Chapter One Behavioral Foundations 9

Diagnostic Question that you are asked to select those cards, and only those cards, that will determine Relative to all the people in the class, how would you rate yourself as a driver? (1) Above whether the hypothesis is true. That is, please select the minimum number of cards 8 Chapter One average? (2) Average? or (3) Below average? Here average is defined as the median. that will enable you to determine whether or not the hypothesis is true. Of the four cards, which would you turn over to verify the hypothesis? DiagnosticDiscussion Question Discussion RelativeBy definition, to all the the people median in the lies class, exactly how would in the you middle, rate yourself with the as population a driver? (1) Above equally di- a average?vided on (2) either Average? side. orThe (3) point Below of average?the last question Here average is that is verydefined few as people the median. rate them- In this card task, most people turn over the card with the , and some turn over the card selves below average. Instead, almost everyone rates themselves as either above with the 2 as well. The proportion turning over just the a and the 2 tends to vary between better than average Discussionaverage or average. This notion of overconfidence is sometimes called the better 30 and 70 percent. Typically less than a third choose to turn over just thea and the 3. effect Bythan definition, average the effect. median lies exactly in the middle, with the population equally di- Yet, turning over only the a and 3 turns out to be the correct answer. This is because People tend to rate them- videdIn on a typicaleither side. class, The 45 point to 55 of percentthe last questionmight rate is that themselves very few aspeople above rate average them- and the efficient way of testing the validity of the hypothesis is to follow the “innocent until selves above average rather proven guilty” maxim and turn over only the cards that might falsify the hypothesis. than below average. selves30 to below 45 percent average. might Instead, rate themselvesalmost everyone as average. rates themselves The proportion as either of above those rat- better than average averageing themselves or average. as belowThis notion average of overconfidenceare fewer in number, is sometimes typically called 10 percent the better or less. Consider in turn the falsification potential of each card. Suppose we turn over the effect thanNotably, average for undergraduates,effect. the responses for above average tend to be at the high card featuring the a. We will find either an even number or an odd number. If we find People tend to rate them- endIn ofa typical the range. class, Some 45 to people 55 percent fault might the wording rate themselves in the question as above for average not specifying and an even number, we have evidence supporting the hypothesis. However, if we find an selves above average rather odd number, we know that the hypothesis is false. Next, suppose we turn over the card than below average. 30a criterion to 45 percent on which might to ratejudge themselves driving ability, as average. suggesting The proportion that they chose of those criteria rat- on ingwhich themselves they indeed as below judge average themselves are fewer to be in above number, average. typically In a10 sense, percent this or proves less. the with the b. This card provides us with no evidence to judge the validity of the hypoth- Notably,point: people for undergraduates, prefer to regard the responsesthemselves for as above above average average, tend if to possible. be at the Virtuallyhigh esis, since the hypothesis says nothing about cards featuring consonants. Now con- endnobody of the likes range. being Some below people average. fault the wording in the question for not specifying sider the card with a 2 on it. If we turn it over, we might find a vowel. This would be a criterion on which to judge driving ability, suggesting that they chose criteria on consistent with the hypothesis. Alternatively, we might find a consonant. That would whichConfirmation they indeed judgeBias themselves to be above average. In a sense, this proves the be irrelevant to the hypothesis. Hence, this card offers no potential for falsification. point:People people who overlookprefer to regardinformation themselves that disconfirms as above average, positions if possible.which they Virtually are evalu- Last, suppose that we turn over the card with the 3 on it. If we find a vowel, we know nobodyating or likes views being which below they average. hold in favor of information that confirms those positions that the hypothesis is false. A consonant provides no information to support or falsify confirmation bias or views are said to exhibit confirmation bias.10 the hypothesis. Thus, the only two cards that offer the potential for falsification are People attach too much im- Confirmation Bias the a and the 3. However, most people choose a and 2, or a alone. Notice that while a portance to information that PeopleDescription who overlook information that disconfirms positions which they are evalu- allows for both confirmation and falsification, 2 allows for confirmation only.11 supports positions which ating or views which they hold in favor of information that confirms those positions they are evaluating or views People often spend too much time searching for reasons to support why a position confirmation bias confirmation bias.10 Illusion of Control which they hold relative orthey views are are evaluating said to exhibit or a view they hold are correct and too little time searching for Peopleto information attach too that much runs im- reasons that might lead them to conclude that the position or their views are incorrect. When a person makes a decision, the outcome typically depends on a combination portance to information that Description counter to those positions Confirmation bias pertains to the manner in which people either seek informa- of luck and skill. Psychologists have concluded people have an exaggerated view supportsor views. positions which they are evaluating or views Peopletion or often make spend use of too the much information time searching at their for disposal. reasons Psychologists to support why have a position concluded of how much control they exert over outcomes. The associated bias is known as the which they hold relative theythat are many evaluating people orare a vulnerableview they holdto confirmation are correct and bias. too This little conclusion time searching is based for on illusion of control illusion of control. to information that runs reasonsthe study that of might hypothesis lead them evaluation to conclude tasks that such the positionas the following. or their views are incorrect. People overestimate the counter to those positions Confirmation bias pertains to the manner in which people either seek informa- extent to which they can Description or views. control events. tionDiagnostic or make use Question of the information at their disposal. Psychologists have concluded The more control a person has over the outcome, the less the influence of chance thatImagine many thatpeople you are are vulnerable presented towith confirmation four cards bias.placed This flat conclusion on a table isin basedfront ofon you. and the more the influence of skill. theThere study is ofa letterhypothesis appearing evaluation on one tasks side such of the as cardthe following. and a number on the other side of the card. You see the following on the four cards: a, b, 2, and 3. Diagnostic Question12 Diagnostic Question Imagine that you agree to participate in a baseball pool. The pool works as follows. Imagine that you are presented with four cards placed flat on a table in front of you. Lying in front of you are two identical piles of baseball cards, with each pile con- There is a letter appearing on onea side of b the card 2 and a number 3 on the other side of taining 227 cards. The face of each card displays the picture of a different baseball the card. You see the following on the four cards: a, b, 2, and 3. player. The organizer of the baseball pool asks you to look through the pile, select one card, and show it to her. After you have done so, the organizer looks through Suppose you are askeda to test bthe following2 hypothesis3 about these four cards: the second (duplicate) pile, finds the twin of the card you selected, and deposits the “Any card having a vowel on one side has an even number on the other side.” Imagine twin into a brown cardboard carton. In order to participate, you pay $1 to the pool

Suppose you are asked to test the following hypothesis about these four cards: “Any card having a vowel on one side has an even number on the other side.” Imagine

she77208_ch01_001-031.indd 8 she77208_ch01_001-031.indd2/7/17 5:25 PM 9 2/7/17 5:25 PM

she77208_ch01_001-031.indd 8 2/7/17 5:25 PM 8 Chapter One Behavioral Foundations 9

Diagnostic Question that you are asked to select those cards, and only those cards, that will determine Relative to all the people in the class, how would you rate yourself as a driver? (1) Above whether the hypothesis is true. That is, please select the minimum number of cards average? (2) Average? or (3) Below average? Here average is defined as the median. that will enable you to determine whether or not the hypothesisBehavioral is true. Foundations Of the four 9 cards, which would you turn over to verify the hypothesis? Discussion that you are asked to select those cards, and only those cards, that will determine Discussion By definition, the median lies exactly in the middle, with the population equally di- whether the hypothesis is true. That is, please select the minimum number of cards vided on either side. The point of the last question is that very few people rate them- In thisthat card will task, enable most you people to determine turn over whether the card or with not thethe ahypothesis, and some is turn true. over Of the cardfour selves below average. Instead, almost everyone rates themselves as either above withcards, the 2 whichas well. would The proportion you turn over turning to verify over justthe hypothesis?the a and the 2 tends to vary between better than average average or average. This notion of overconfidence is sometimes called the better 30 and 70 percent. Typically less than a third choose to turn over just thea and the 3. effect than average effect. Yet,Discussion turning over only the a and 3 turns out to be the correct answer. This is because People tend to rate them- In a typical class, 45 to 55 percent might rate themselves as above average and the efficientIn this card way task, of testingmost people the validity turn over of the cardhypothesis with the is a ,to and follow some the turn “innocent over the untilcard selves above average rather provenwith guilty” the 2 as maxim well. The and proportion turn over turningonly the over cards just that the amight and the falsify 2 tends the to hypothesis. vary between than below average. 30 to 45 percent might rate themselves as average. The proportion of those rat- ing themselves as below average are fewer in number, typically 10 percent or less. Consider30 and 70 in percent. turn the Typically falsification less thanpotential a third of choose each card. to turn Suppose over just we the turna and over the the 3. Notably, for undergraduates, the responses for above average tend to be at the high cardYet, featuring turning the over a. Weonly will the finda and either 3 turns an outeven to numberbe the correct or an oddanswer. number. This Ifis because we find end of the range. Some people fault the wording in the question for not specifying an eventhe efficient number, way we of have testing evidence the validity supporting of the hypothesisthe hypothesis. is to follow However, the “innocent if we find until an a criterion on which to judge driving ability, suggesting that they chose criteria on oddproven number, guilty” we know maxim that and the turn hypothesis over only is the false. cards Next, that supposemight falsify we turn the hypothesis.over the card which they indeed judge themselves to be above average. In a sense, this proves the with theConsider b. This incard turn provides the falsification us with nopotential evidence of eachto judge card. the Suppose validity we of turn the hypothover the- point: people prefer to regard themselves as above average, if possible. Virtually esis,card since featuring the hypothesis the a. We says will findnothing either about an even cards number featuring or an consonants. odd number. Now If we con find- nobody likes being below average. sideran the even card number, with a we 2 onhave it. evidenceIf we turn supporting it over, we the might hypothesis. find a However,vowel. This if we would find bean consistentodd number, with thewe knowhypothesis. that the Alternatively, hypothesis is wefalse. might Next, find suppose a consonant. we turn overThat the would card Confirmation Bias be irrelevantwith the b .to This the card hypothesis. provides Hence,us with thisno evidence card offers to judge no potential the validity for offalsification. the hypoth- People who overlook information that disconfirms positions which they are evalu- Last,esis, suppose since thatthe hypothesiswe turn over says the nothing card with about the cards 3 on it.featuring If we find consonants. a vowel, Nowwe know con- ating or views which they hold in favor of information that confirms those positions thatsider the hypothesis the card with is false.a 2 on A it. consonant If we turn provides it over, we no might information find a vowel. to support This orwould falsify be confirmation bias or views are said to exhibit confirmation bias.10 the consistenthypothesis. with Thus, the hypothesis.the only two Alternatively, cards that offer we might the potential find a consonant. for falsification That would are People attach too much im- the bea and irrelevant the 3. However,to the hypothesis. most people Hence, choose this acard and offers 2, or ano alone. potential Notice for thatfalsification. while a portance to information that Description allowsLast, for suppose both confirmation that we turn andover falsification, the card with 2 the allows 3 on it.for If confirmation we find a vowel, only. we11 know supports positions which People often spend too much time searching for reasons to support why a position that the hypothesis is false. A consonant provides no information to support or falsify they are evaluating or views Illusion of Control which they hold relative they are evaluating or a view they hold are correct and too little time searching for the hypothesis. Thus, the only two cards that offer the potential for falsification are to information that runs reasons that might lead them to conclude that the position or their views are incorrect. Whenthe aa personand the makes3. However, a decision, most people the outcome choose atypically and 2, or depends a alone. Noticeon a combination that while a 11 counter to those positions Confirmation bias pertains to the manner in which people either seek informa- of luckallows and for skill. both Psychologistsconfirmation and have falsification, concluded 2 allowspeople for have confirmation an exaggerated only. view or views. tion or make use of the information at their disposal. Psychologists have concluded of howIllusion much ofcontrol Control they exert over outcomes. The associated bias is known as the that many people are vulnerable to confirmation bias. This conclusion is based on illusion of control illusion of control. the study of hypothesis evaluation tasks such as the following. People overestimate the When a person makes a decision, the outcome typically depends on a combination extent to which they can Descriptionof luck and skill. Psychologists have concluded people have an exaggerated view control events. Diagnostic Question Theof more how controlmuch control a person they has exert over over the outcomes. outcome, The the associated less the influence bias is known of chance as the Imagine that you are presented with four cards placed flat on a table in front of you. illusion of control and illusionthe more of thecontrol. influence of skill. There is a letter appearing on one side of the card and a number on the other side of People overestimate the extent to which they can DiagnosticDescription Question12 the card. You see the following on the four cards: a, b, 2, and 3. control events. ImagineThe more that youcontrol agree a personto participate has over in the a baseball outcome, pool. the Theless thepool influence works as of follows. chance Lyingand in the front more of the you influence are two ofidentical skill. piles of baseball cards, with each pile con- a b 2 3 tainingDiagnostic 227 cards. Question The face12 of each card displays the picture of a different baseball player. The organizer of the baseball pool asks you to look through the pile, select oneImagine card, and that show you agreeit to her. to participate After you in have a baseball done so,pool. the The organizer pool works looks as throughfollows. Suppose you are asked to test the following hypothesis about these four cards: the Lyingsecond in (duplicate) front of you pile, are finds two identicalthe twin pilesof the of card baseball you selected,cards, with and each deposits pile con- the “Any card having a vowel on one side has an even number on the other side.” Imagine twintaining into a 227brown cards. cardboard The face carton. of each In card order displays to participate, the picture you of paya different $1 to thebaseball pool player. The organizer of the baseball pool asks you to look through the pile, select one card, and show it to her. After you have done so, the organizer looks through the second (duplicate) pile, finds the twin of the card you selected, and deposits the twin into a brown cardboard carton. In order to participate, you pay $1 to the pool

she77208_ch01_001-031.indd 8 she77208_ch01_001-031.indd2/7/17 5:25 PM 9 2/7/17 5:25 PM

she77208_ch01_001-031.indd 9 2/7/17 5:25 PM 10 Chapter One Behavioral Foundations 11

organizer. Because you were the first participant approached, when the pool orga- 1.3 HEURISTICS nizer approaches the next participant, she will do so with two identical pools con- 10 Chapter One taining 226 cards, not 227 cards. After all the cards have been sold, the organizer Representativeness then plans to draw exactly one card from the brown cardboard carton. The owner of In asking about the extent to which an object or idea fits a stereotype, people are organizer.the winning Because card willyou receivewere the a first$50 participantprize. approached, when the pool orga- asking how representative that object or idea is for the class to which it belongs. nizerSuppose approaches that theall nextthe cards participant, have been she willsold, do but so the with drawing two identical has yet pools to take con- place. representativeness Psychologists refer to the underlying principle as representativeness. tainingThe pool 226 organizer cards, not approaches 227 cards. Afteryou to all say the that cards someone have been who sold, really the wanted organizer to par- People make judgments thenticipate plans cannot, to draw because exactly oneall thecard cards from havethe brown been cardboardsold. She carton.asks you The how owner much of you based on stereotypic think- Description thewould winning be willing card will to acceptreceive in a $50exchange prize. for the card you drew. What is the minimum ing, asking how represen- People often make judgments and predictions by relying on heuristics that make amountSuppose you that would all the ask cards for tohave give been up sold,your card?but the drawing has yet to take place. tative an object or idea is for the class to which it use of analogues and stereotypes. Psychologists have concluded that people place The pool organizer approaches you to say that someone who really wanted to par- belongs. too much reliance on representativeness and suggest that representativeness-based ticipateDiscussion cannot, because all the cards have been sold. She asks you how much you thinking can result in systematic errors. One such error is known as the conjunction wouldWhen be the willing study tousing accept the in preceding exchange questionfor the card was you first drew. run, What only ishalf the the minimum participants conjunction fallacy fallacy and was studied through the use of the following question.13 amountwere allowed you would to choose ask for their to give own up cards your fromcard? the pile of 227 cards. Each person who People misjudge the prob- ability that several events was not allowed to choose a card was instead handed a card by the pool organizer. occur simultaneously Diagnostic Question DiscussionApart from that change, everything else was the same for the two groups of partici- relative to the probabilities Imagine that you hear about a 31-year-old woman named Linda from people who Whenpants, the including study using the thevaluation preceding question question at thewas end. first run, only half the participants of those events occurring wereIs allowed it important to choose that their people own becards able from to choosethe pile oftheir 227 own cards. cards Each instead person whoof being separately. know her quite well. They tell you that she is single, outspoken, and very bright. washanded not allowed their cards? to choose Does a it card affect was their instead valuations? handed aIs card a pool by participantthe pool organizer. more likely When she was a student, she was deeply concerned with issues of social justice. Apartto win from the that $50 change, just because everything he orelse she was selected the same the for card the twoinstead groups of ofbeing partici- handed Linda’s friends neglect to tell you about her current interests and career. The ques- pants,the card? including Of course the valuation not; the question winner ofat thethe end.pool is determined entirely by a chance tion about Linda in the behavioral questionnaire provided you with several alterna- drawing.Is it important The odds that of people winning be $50able from to choose a single their card own in thiscards pool instead are 1of in being 227. The tives, which included the following: handedexpected their payoff cards? is Does $0.22. it affect And participantstheir valuations? have Is been a pool asked participant to pay more$1 per likely card no 1. Linda is a feminist, active in the women’s movement. tomatter win the who $50 selects just becausethe card. he or she selected the card instead of being handed 2. Linda is a psychiatric social worker. the card? Of course not; the winner of the pool is determined entirely by a chance In the actual study, among participants who selected their own cards, the typi- 3. Linda is a bank teller. drawing.cal response The odds was of$8.67. winning In the $50 alternative from a single version, card wherein this poolthe organizer are 1 in 227. selected The the expectedcard, the payoff typical is response$0.22. And was participants $1.96. have been asked to pay $1 per card no 4. Linda is a bank teller and is active in the women’s movement. matterWhat who accounts selects the for card. the difference? Psychologists have concluded that the illusion The diagnostic question asks you to rank these possibilities about Linda from top to ofIn control the actual leads study, people among to place participants a higher who value selected on their their cards own when cards, they the selecttypi- the bottom by assigning 1 to what you regard as the most likely possibility. calcards response than waswhen $8.67. the organizerIn the alternative selects version,their cards. where That the is,organizer people selected seem to the act as card,if they the cantypical control response the oddswas $1.96.of winning the pool by selecting the card themselves Discussion ratherWhat than accounts letting for somebody the difference? else doPsychologists it. However have the concludedodds are thethat same, the illusion no matter In answering this question, people often put activity in the women’s movement and ofwho control selects leads the people card. Beingto place able a higher to control value the on odds their iscards an illusion. when they select the psychiatric social worker at the top of their lists. The situation that people judge to cardsA thansubsidiary when thefinding organizer concerns selects overvaluation. their cards. ThatFirst, is, people people agree seem to to pay act $1 as for a be least likely is bank teller. ifrisky they expectedcan control payoff the odds amounting of winning to $0.22. the pool Second, by selecting they place the acard minimum themselves value on How do people arrive at these assessments? Psychologists suggest that they ratherthe card than well letting in excesssomebody of $1, else let do alone it. However $0.22. Once the odds they are own the that same, card, no itmatter increases consider each situation as a category and ask how representative Linda is of whodramatically selects the in card. value. Being able to control the odds is an illusion. the category. Since the description of Linda is highly representative of their AFor subsidiary the previous finding diagnostic concerns questions, overvaluation. the typical First, people results agree when to administered pay $1 for a to a impression of someone active in the women’s movement, they judge that Linda riskygiven expected class of payoff students amounting tend to conform to $0.22. to Second, the original they place experimental a minimum findings value onalmost is highly likely to be in the women’s movement. Similarly, since Linda does not theall card of the well time. in excess However, of $1, the let samealone $0.22.statement Once does they not own apply that card,to the it baseballincreases pool seem to fit the stereotype of a bank teller, they judge that Linda is unlikely to be dramaticallyquestion, where in value. it is not unusual for the group choosing their own cards to provide a a bank teller. valuationFor the previous that is lower diagnostic than the questions, group that the wastypical handed results their when cards. administered In addition, to athere Because heuristics are shortcuts, they may also lead their users astray. For givencan be class considerable of students variation tend to conform in valuations, to the original some of experimental which are much findings less almost than those example, consider two of the choices for Linda, item 3 (Linda is a bank teller) allreported of the time.in the However, original experiment.the same statement That this does is the not case apply suggests to the thebaseball need poolfor more and item 4 (Linda is a bank teller and is active in the women’s movement). As question,caution whenwhere applyingit is not unusual the concept for the of group illusion choosing of control. their own cards to provide a Exhibit 1-2 demonstrates, feminist bank tellers are members of the category of bank valuation that is lower than the group that was handed their cards. In addition, there can be considerable variation in valuations, some of which are much less than those reported in the original experiment. That this is the case suggests the need for more caution when applying the concept of illusion of control.

she77208_ch01_001-031.indd 10 she77208_ch01_001-031.indd2/7/17 5:25 PM 11 2/7/17 5:25 PM

she77208_ch01_001-031.indd 10 2/7/17 5:25 PM 10 Chapter One Behavioral Foundations 11 organizer. Because you were the first participant approached, when the pool orga- 1.3 HEURISTICS nizer approaches the next participant, she will do so with two identical pools con- taining 226 cards, not 227 cards. After all the cards have been sold, the organizer Representativeness Behavioral Foundations 11 then plans to draw exactly one card from the brown cardboard carton. The owner of In asking about the extent to which an object or idea fits a stereotype, people are the winning card will receive a $50 prize. 1.3 HEURISTICSasking how representative that object or idea is for the class to which it belongs. Suppose that all the cards have been sold, but the drawing has yet to take place. representativeness Psychologists refer to the underlying principle as representativeness. The pool organizer approaches you to say that someone who really wanted to par- People make judgments Representativeness ticipate cannot, because all the cards have been sold. She asks you how much you based on stereotypic think- DescriptionIn asking about the extent to which an object or idea fits a stereotype, people are would be willing to accept in exchange for the card you drew. What is the minimum ing, asking how represen- Peopleasking often how make representative judgments that and object predictions or idea byis forrelying the class on heuristics to which thatit belongs. make amount you would ask for to give up your card? tative an object or idea is for therepresentativeness class to which it use Psychologistsof analogues referand stereotypes.to the underlying Psychologists principle ashave representativeness. concluded that people place Discussion belongs.People make judgments too much reliance on representativeness and suggest that representativeness-based based on stereotypic think-thinkingDescription can result in systematic errors. One such error is known as the conjunction When the study using the preceding question was first run, only half the participants conjunctioning, asking fallacy how represen- fallacyPeople and often was studiedmake judgments through the and use predictions of the following by relying question. on heuristics13 that make were allowed to choose their own cards from the pile of 227 cards. Each person who Peopletative misjudge an object the prob-or idea is abilityfor that the several class to events which it use of analogues and stereotypes. Psychologists have concluded that people place was not allowed to choose a card was instead handed a card by the pool organizer. occur belongs.simultaneously Diagnostictoo much Question reliance on representativeness and suggest that representativeness-based Apart from that change, everything else was the same for the two groups of partici- relative to the probabilities thinking can result in systematic errors. One such error is known as the conjunction pants, including the valuation question at the end. conjunction fallacy Imagine that you hear about a 31-year-old woman named Linda from people who of those events occurring fallacy and was studied through the use of the following question.13 Is it important that people be able to choose their own cards instead of being separately.People misjudge the prob- know her quite well. They tell you that she is single, outspoken, and very bright. ability that several events When she was a student, she was deeply concerned with issues of social justice. handed their cards? Does it affect their valuations? Is a pool participant more likely occur simultaneously Diagnostic Question to win the $50 just because he or she selected the card instead of being handed Linda’s friends neglect to tell you about her current interests and career. The ques- relative to the probabilities tionImagine about Linda that you in the hear behavioral about a 31-year-old questionnaire woman provided named you Linda with from several people alterna- who the card? Of course not; the winner of the pool is determined entirely by a chance of those events occurring know her quite well. They tell you that she is single, outspoken, and very bright. drawing. The odds of winning $50 from a single card in this pool are 1 in 227. The separately. tives, which included the following: When she was a student, she was deeply concerned with issues of social justice. expected payoff is $0.22. And participants have been asked to pay $1 per card no 1. LindaLinda’s is friendsa feminist, neglect active to tellin theyou women’s about her movement. current interests and career. The ques- matter who selects the card. 2. Lindation about is a psychiatricLinda in the social behavioral worker. questionnaire provided you with several alterna- In the actual study, among participants who selected their own cards, the typi- 3. Lindatives, whichis a bank included teller. the following: cal response was $8.67. In the alternative version, where the organizer selected the card, the typical response was $1.96. 4. Linda1. Linda is a is bank a feminist, teller and active is active in the inwomen’s the women’s movement. movement. What accounts for the difference? Psychologists have concluded that the illusion The2. diagnostic Linda is a question psychiatric asks social you toworker. rank these possibilities about Linda from top to of control leads people to place a higher value on their cards when they select the bottom3. Linda by assigning is a bank 1 teller. to what you regard as the most likely possibility. cards than when the organizer selects their cards. That is, people seem to act as 4. Linda is a bank teller and is active in the women’s movement. if they can control the odds of winning the pool by selecting the card themselves Discussion rather than letting somebody else do it. However the odds are the same, no matter In answeringThe diagnostic this question,question askspeople you often to rank put these activity possibilities in the women’s about Linda movement from top and to who selects the card. Being able to control the odds is an illusion. psychiatricbottom by social assigning worker 1 toat whatthe top you of regard their lists.as the The most situation likely possibility. that people judge to A subsidiary finding concerns overvaluation. First, people agree to pay $1 for a be least likely is bank teller. Discussion risky expected payoff amounting to $0.22. Second, they place a minimum value on How do people arrive at these assessments? Psychologists suggest that they the card well in excess of $1, let alone $0.22. Once they own that card, it increases considerIn answering each situation this question, as a people category often and put askactivity how in representative the women’s movement Linda is and of dramatically in value. the psychiatric category. Sincesocial worker the description at the top of of their Linda lists. is The highly situation representative that people ofjudge their to For the previous diagnostic questions, the typical results when administered to a impressionbe least likelyof someone is bank activeteller. in the women’s movement, they judge that Linda given class of students tend to conform to the original experimental findings almost is highlyHow likely do people to be inarrive the women’sat these assessments? movement. Similarly, Psychologists since suggest Linda doesthat they not all of the time. However, the same statement does not apply to the baseball pool seemconsider to fit the each stereotype situation of as a abank category teller, and they ask judge how that representative Linda is unlikely Linda to is be of question, where it is not unusual for the group choosing their own cards to provide a a bankthe teller. category. Since the description of Linda is highly representative of their valuation that is lower than the group that was handed their cards. In addition, there Becauseimpression heuristics of someone are active shortcuts, in the they women’s may alsomovement, lead their they usersjudge astray.that Linda For can be considerable variation in valuations, some of which are much less than those example,is highly consider likely to two be ofin the women’s choices formovement. Linda, item Similarly, 3 (Linda since is Linda a bank does teller) not reported in the original experiment. That this is the case suggests the need for more andseem item to 4 fit (Linda the stereotype is a bank of teller a bank and teller, is active they judge in the that women’s Linda is movement). unlikely to Asbe caution when applying the concept of illusion of control. Exhibita bank 1-2 teller. demonstrates, feminist bank tellers are members of the category of bank Because heuristics are shortcuts, they may also lead their users astray. For example, consider two of the choices for Linda, item 3 (Linda is a bank teller) and item 4 (Linda is a bank teller and is active in the women’s movement). As Exhibit 1-2 demonstrates, feminist bank tellers are members of the category of bank

she77208_ch01_001-031.indd 10 she77208_ch01_001-031.indd2/7/17 5:25 PM 11 2/7/17 5:25 PM

she77208_ch01_001-031.indd 11 2/7/17 5:25 PM 12 Chapter One Behavioral Foundations 13

EXHIBIT 1-2 Discussion Venn Diagram for Most people think about the risks associated with the above four sources by try- 12Three Chapter Choices One in Linda ing to recall events associated with each, and basing their judgments of relative Problem frequency on the ease with which such events come to mind. Memories record both EXHIBIT 1-2 Bank tellers personal experiences and information obtained through the media. Women active Venn Diagram for active in the Bank Tellers in the women’s In August 2005, Hurricane Katrina struck the gulf states, and damaged the levee sys- Three Choices in Linda women’s movement tem that protected New Orleans. The resulting flood destroyed the city and was the main Problem movement media story day after day. The resulting death toll from Katrina was more than 1,500. Be- Bank tellers fore Katrina, the death toll from hurricanes and floods had been much lower. In 2004, four Women active active in the successive hurricanes struck Florida, in what was viewed as a highly unusual event. The Bank Tellers in the women’s women’s movement resulting death toll from all four storms was about 100, quite a high number at the time. movement The media reports attacks by sharks on swimmers and surfers, often with a level of drama that rivals the reporting of many floods and hurricanes. Stories about rip currents receive occasional coverage in the media, usually when someone has drowned as a result. Before Hurricane Katrina, most people ranked shark attacks or floods as posing the tellers. Therefore, it cannot be more probable for Linda to be a feminist bank teller most danger. Few people chose rip currents. Yet, according to the U.S. Lifesaving As- than just a bank teller. This would violate one of the laws of probability. Yet, most sociation, riptides pose the most danger. Using data reported for the years 1994–2014, people respond to the Linda question by assigning a higher likelihood to item 4 the United States Lifesaving Association reported that rip currents are the primary (feminist bank teller) than they do to item 3 (bank teller).14 It is typical for between cause of distress leading to rescue by lifeguards at surf beaches, comprising more tellers.67 and Therefore, 80 percent it ofcannot a class be tomore provide probable responses for Linda exhibiting to be a feministthe conjunction bank teller fallacy than 80 percent of incidents. The Association reports that between 2010 and 2014, thanfor thisjust question.a bank teller. This would violate one of the laws of probability. Yet, most there were on average approximately 72,700 rescues per year, and 111 deaths from peopleThe respond event that to theLinda Linda is both question a bank by teller assigning and is a active higher in likelihood the women’s to itemmovement 4 14 drowning. The National Oceanic and Atmospheric Administration (NOAA) reports (feministis an example bank teller) of the than conjunction they do to of item two 3 events. (bank teller).In this respect, It is typical the typicalfor between response that between 2004 and 2014, the average number of deaths per year from floods was 67to and the 80Linda percent question of a class constitutes to provide a conjunction responses exhibiting fallacy. the conjunction fallacy for this question. 75 and from hurricanes it was 108. Although the percentage of people ranking rip AvailabilityThe event that Linda is both a bank teller and is active in the women’s movement currents is somewhat variable from class to class, they are typically rated first about 50 percent less often than the top rated item, be it hurricanes or floods. isPsychologists an example of have the conjunction concluded thatof two people events. tend In to this attach respect, more the weight typical to response information tothat the isLinda more question readily availableconstitutes than a conjunction to information fallacy that. is less readily available. They Anchoring and Adjustment call the propensity to overrely on information that is readily available, especially Availability A number that people have in their minds can serve to anchor their judgments just availability from memory, the availability heuristic. People overweight informa- Psychologists have concluded that people tend to attach more weight to information as a dropped anchor keeps a boat from drifting too far. Psychologists have con- anchoring and anchoring and adjustment. tion that is readily available thatDescription is more readily available than to information that is less readily available. They cluded that people are susceptible to a bias known as adjustment and intuitive relative to in- call the propensity to overrely on information that is readily available, especially People typically rely on their own experiences and memories when forming judg- People form an estimate by availabilityformation that is less salient availability Description and more abstract, thereby fromments memory, of risk. the In other words, heuristic. salience matters. This conclusion is based on studies beginning with an initial People overweight informa- When forming judgments, people have a tendency to become anchored on numbers biasing judgments. that pose questions such as the following. number and adjusting to tion that is readily available Description reflect new information or in their heads and do not make sufficient adjustments relative to the anchor. The and intuitive relative to in- PeopleDiagnostic typically Question rely on their own experiences and memories when forming judg- circumstances. However, following two-part question has been used to study anchoring. formation that is less salient they tend to make insuf- and more abstract, thereby mentsConsider of risk. the In danger other words,of death salience or injury matters. stemming This conclusion from four issources, based on all studies involving that pose questions such as the following. ficient adjustments relative Diagnostic Question biasing judgments. water: to that number, thereby leading to anchoring bias. Record the last three digits of your home phone number. Now add 400 to the last Diagnostic1. Shark attacks Question three digits of your home phone number. Call the sum X. Consider2. Hurricanes the danger of death or injury stemming from four sources, all involving 1. Without looking up the answer anywhere, do you think that Attila the Hun was de- water:3. Rip currents feated in Europe at the Battle of the Catalaunian Plains before or after the year X? 4. Floods 1. Shark attacks 2. Without looking up the answer anywhere, provide your best guess about the actual 2.Which Hurricanes item in this list involves the most danger to people? year that Attila the Hun was defeated at the Battle of the Catalaunian Plains. 3. Rip currents 4. Floods Which item in this list involves the most danger to people?

she77208_ch01_001-031.indd 12 she77208_ch01_001-031.indd2/7/17 5:25 PM 13 2/7/17 5:25 PM

she77208_ch01_001-031.indd 12 2/7/17 5:25 PM 12 Chapter One Behavioral Foundations 13

EXHIBIT 1-2 Discussion Venn Diagram for Most people think about the risks associated with the above four sources by try- Three Choices in Linda ing to recall events associated with each, and basing their judgmentsBehavioral Foundations of relative 13 Problem frequency on the ease with which such events come to mind. Memories record both Bank tellers personalDiscussion experiences and information obtained through the media. Women active active in the Bank Tellers in the women’s InMost August people 2005, think Hurricane about the Katrina risks struckassociated the gulf with states, the aboveand damaged four sources the levee by try-sys- women’s movement teming that to protected recall events New Orleans. associated The with resulting each, flood and basingdestroyed their the judgments city and was of therelative main movement mediafrequency story day on after the easeday. Thewith resulting which such death events toll from come Katrina to mind. was Memories more than record 1,500. both Be- forepersonal Katrina, experiencesthe death toll and from information hurricanes obtainedand floods through had been the much media. lower. In 2004, four successiveIn August hurricanes 2005, struckHurricane Florida, Katrina in what struck was the viewed gulf states, as a andhighly damaged unusual the event. levee Thesys- resultingtem that death protected toll from New all Orleans. four storms The resulting was about flood 100, destroyed quite a high the city number and was at thethe time.main Themedia media story reports day after attacks day. The by resulting sharks ondeath swimmers toll from andKatrina surfers, was more often than with 1,500. a level Be- of dramafore Katrina, that rivalsthe death the toll reporting from hurricanes of many and floods floods had and been hurricanes. much lower. Stories In 2004, about four rip successivecurrents receive hurricanes occasional struck Florida, coverage in what in the was media, viewed usuallyas a highly when unusual someone event. hasThe drownedresulting as adeath result. toll from all four storms was about 100, quite a high number at the time. BeforeThe Hurricane media reports Katrina, attacks most by peoplesharks rankedon swimmers shark attacks and surfers, or floods often as with posing a level the tellers. Therefore, it cannot be more probable for Linda to be a feminist bank teller mostof danger. drama Few that people rivals thechose reporting rip currents. of many Yet, floodsaccording and to hurricanes. the U.S. Lifesaving Stories about As- than just a bank teller. This would violate one of the laws of probability. Yet, most sociation,rip currents riptides receive pose theoccasional most danger. coverage Using in datathe media, reported usually for the when years someone 1994–2014, has people respond to the Linda question by assigning a higher likelihood to item 4 the drownedUnited States as a result. Lifesaving Association reported that rip currents are the primary (feminist bank teller) than they do to item 3 (bank teller).14 It is typical for between cause Beforeof distress Hurricane leading Katrina, to rescue most by people lifeguards ranked atshark surf attacks beaches, or floods comprising as posing more the 67 and 80 percent of a class to provide responses exhibiting the conjunction fallacy thanmost 80 percentdanger. Fewof incidents. people chose The rip Association currents. Yet, reports according that betweento the U.S. 2010 Lifesaving and 2014, As- for this question. theresociation, were on riptides average pose approximately the most danger. 72,700 Using rescues data reported per year, for and the years111 deaths 1994–2014, from The event that Linda is both a bank teller and is active in the women’s movement drowning.the United The States National Lifesaving Oceanic Association and Atmospheric reported Administration that rip currents (NOAA) are the primaryreports is an example of the conjunction of two events. In this respect, the typical response thatcause between of distress2004 and leading 2014, to the rescue average by numberlifeguards of atdeaths surf beaches,per year comprisingfrom floods more was to the Linda question constitutes a conjunction fallacy. 75 andthan from 80 percent hurricanes of incidents. it was 108.The AlthoughAssociation the reports percentage that between of people 2010 ranking and 2014, rip Availability currentsthere iswere somewhat on average variable approximately from class 72,700 to class, rescues they perare year,typically and rated111 deaths first about from 50 percentdrowning. less The often National than the Oceanic top rated and item, Atmospheric be it hurricanes Administration or floods. (NOAA) reports Psychologists have concluded that people tend to attach more weight to information that between 2004 and 2014, the average number of deaths per year from floods was that is more readily available than to information that is less readily available. They Anchoring75 and from and hurricanes Adjustment it was 108. Although the percentage of people ranking rip call the propensity to overrely on information that is readily available, especially currents is somewhat variable from class to class, they are typically rated first about availability availability A number that people have in their minds can serve to anchor their judgments just from memory, the heuristic. 50 percent less often than the top rated item, be it hurricanes or floods. People overweight informa- as a dropped anchor keeps a boat from drifting too far. Psychologists have con- tion that is readily available Description anchoring and cluded that people are susceptible to a bias known as anchoring and adjustment. adjustment Anchoring and Adjustment and intuitive relative to in- People typically rely on their own experiences and memories when forming judg- formation that is less salient People form an estimate by DescriptionA number that people have in their minds can serve to anchor their judgments just ments of risk. In other words, salience matters. This conclusion is based on studies and more abstract, thereby beginning with an initial Whenas aforming dropped judgments, anchor keeps people a boat have from a tendency drifting totoo become far. Psychologists anchored on have numbers con- biasing judgments. that pose questions such as the following. number and adjusting to anchoring and in theircluded heads that peopleand do arenot susceptible make sufficient to a bias knownadjustments as anchoring relative andto the adjustment. anchor. The reflectadjustment new information or Diagnostic Question circumstances. However, following two-part question has been used to study anchoring. People form an estimate by they tend to make insuf- Description Consider the danger of death or injury stemming from four sources, all involving beginning with an initial ficient adjustments relative number and adjusting to DiagnosticWhen forming Question judgments, people have a tendency to become anchored on numbers water: to that number, thereby reflect new information or in their heads and do not make sufficient adjustments relative to the anchor. The leading to anchoring bias. Record the last three digits of your home phone number. Now add 400 to the last 1. Shark attacks circumstances. However, threefollowing digits of two-part your home question phone has number. been used Call to the study sum anchoring. X. 2. Hurricanes they tend to make insuf- ficient adjustments relative 1. WithoutDiagnostic looking Question up the answer anywhere, do you think that Attila the Hun was de- 3. Rip currents to that number, thereby featedRecord in the Europe last three at the digits Battle of of your the homeCatalaunian phone Plainsnumber. before Now or add after 400 the to yearthe lastX? 4. Floods leading to anchoring bias. 2. Withoutthree digits looking of your up homethe answer phone anywhere, number. Call provide the sum your X best. guess about the actual Which item in this list involves the most danger to people? year that Attila the Hun was defeated at the Battle of the Catalaunian Plains. 1. Without looking up the answer anywhere, do you think that Attila the Hun was de- feated in Europe at the Battle of the Catalaunian Plains before or after the year X? 2. Without looking up the answer anywhere, provide your best guess about the actual year that Attila the Hun was defeated at the Battle of the Catalaunian Plains.

she77208_ch01_001-031.indd 12 she77208_ch01_001-031.indd2/7/17 5:25 PM 13 2/7/17 5:25 PM

she77208_ch01_001-031.indd 13 2/7/17 5:25 PM 14 Chapter One Behavioral Foundations 15

Discussion When psychologists conducted an experiment based on the above diagnostic Most people do not know their history well enough to remember that Attila the Hun question, in which real jars with real beans were used, they found that a majority of 14 Chapter One was defeated at the Battle of the Catalaunian Plains in 451 CE. Because most people subjects selected Jar 2 over Jar 1, even though Jar 1 featured a higher probability of do not know the date, they have to estimate it based on whatever recollections they winning the $10. Moreover, many people stated that although they recognized that Discussionhave. Mean predictions are highly variable across different classes, typically lying the probability of winning was higher with Jar 1, they found themselves emotion- Mostbetween people 415 do andnot know1100. their Mean history responses well enough for Europeans to remember tend that to beAttila much the closerHun to ally drawn to choose Jar 2. In essence, people underweighted the denominator when was451 defeated than non-Europeans. at the Battle of the Catalaunian Plains in 451 CE. Because most people implicitly computing the odds of winning depending on which jar they selected. do notConsider know the the date, first they of the have two to partsestimate of theit based preceding on whatever question. recollections The first theyquestion Psychologists named the phenomenon “ratio bias” and concluded that the visual have.is intended Mean predictions to provide area context highly variablefor the way across they different think about classes, the typically second question.lying stimulus of seeing a higher absolute number of beans in Jar 2 than Jar 1 was the betweenThe issue 415 here and is1100. whether Mean people responses arrive for at Europeans their prediction tend to ofbe themuch year closer of Attila’s to driving factor. Jar 2 generated a higher level of positive affect than Jar 1. 451defeat than by non-Europeans. beginning with X and adjusting up or down. Specifically, do they anchor The role of affect in choice can be illustrated by another experiment. In this experi- onConsider X but not the adjust first ofsufficiently? the two parts of the preceding question. The first question ment, experienced forensic psychologists and psychiatrists were asked to recommend is intendedThere isto noprovide rational a context reason for to the expect way they that think people’s about predictions the second should question. in any whether to release a mental patient, Mr. Jones, who might be violent within six months Theway issue be correlated here is whether with the people last three arrive digits at their of theirprediction phone of numbers. the year Yet of Attila’sin practice, of being discharged from the hospital. Notably, the subjects based their recommenda- defeatpeople’s by beginningresponses with are soX correlated,and adjusting with up theor down. sign of Specifically, the correlation do they being anchor positive. tions on information provided by other clinical experts. When the information in ques- onAt X the but same not adjust time, sufficiently?correlations are often statistically insignificant unless class sizes tion was described as “20 out of every 100 patients similar to Mr. Jones are estimated to areThere very islarge, no rationalwell over reason a hundred. to expect that people’s predictions should in any commit an act of violence,” they were twice as likely to recommend against discharge way be correlated with the last three digits of their phone numbers. Yet in practice, than when the information was described as “patients similar to Mr. Jones are estimated people’sAffect responses Heuristic are so correlated, with the sign of the correlation being positive. to have a 20 percent chance of committing an act of violence.” The psychologists con- 15 AtDescription the same time, correlations are often statistically insignificant unless class sizes cluded that the first description generated much more negative affect than the second. are very large, well over a hundred. affect When psychologists use the word affect, they are referring to the specific quality of Interacting Phenomena An emotional feeling. Affectgoodness Heuristic or badness. This quality is typically experienced as a feeling state, possibly Description consciously but not always, with goodness being experienced positively and bad- Descriptionness being experienced negatively. People’s affective responses occur rapidly and Psychologists have discovered that one behavioral phenomenon can affect another. affect Whenautomatically, psychologists and userelate the to word what affect, is called they their are referring“affect pool.” to the Thisspecific pool quality consists of of a In this respect, consider the extent to which excessive optimism is related to the An emotional feeling. goodness“library” or of badness. positive This and quality negative is typically mental markersexperienced that aspeople a feeling associate, state, possibly either con- following four issues: consciouslysciously or butunconsciously, not always, withwith goodnessthe images being stored experienced in their memories. positively Whenand bad people- 1. The perceived controllability of the event. ness being experienced negatively. People’s affective responses occur rapidly and engage in the mental shortcut of basing their decisions on the contents of their affect 2. The desirability of the event. automatically,pools instead andof weighing relate to what the pros is called and constheir of“affect various pool.” alternatives, This pool orconsists retrieving of a rel- 3. The perceived probability of the event, reflecting previous personal experience— affect heuristic “library”evant information of positive fromand negative their memories, mental markers they are that said people to rely associate, on the affect either heuristic con- . effectively the degree to which the person is familiar with the event. Basing decisions primarily sciouslyIn this regard,or unconsciously, psychologists with note the imagesthat retrieving stored in relevant their memories. information When from people memory on intuition, instinct, and engageand then in the processing mental shortcut this information of basing their relies decisions on other on heuristicsthe contents such of theiras availability affect 4. The degree to which people view themselves as representative of the type of gut feeling. poolsand representativeness.instead of weighing theIn contrast,pros and consthe affective of various heuristic alternatives, is more or retrieving of a “quick rel- and persons to whom the event occurs. affect heuristic evantdirty” information shortcut. from their memories, they are said to rely on the affect heuristic. Diagnostic Question Basing decisions primarily In this regard, psychologists note that retrieving relevant information from memory on intuition, instinct, and andDiagnostic then processing Question this information relies on other heuristics such as availability The question pertaining to excessive optimism involves a series of possible life gut feeling. events. Please assess each of these events on the specified criteria. andImagine representativeness. that you have In an contrast, opportunity the affectiveto win $10 heuristic by randomly is more drawing of a “quick a white and bean dirty”from shortcut.one of two jars, denoted Jar 1 and Jar 2. Jar 1 contains 10 beans, of which one A. For each of the life events, assign a controllability category number. The list of is white and the rest are black. Jar 2 contains 100 beans, of which seven are white Diagnostic Question choices includes: and the rest are black. You can choose the jar from which the drawing will take 1. There is nothing one can do that will change the likelihood that the event will Imagineplace. Which that you jar have would an opportunity you choose, to Jar win 1 $10or Jar by 2?randomly drawing a white bean from one of two jars, denoted Jar 1 and Jar 2. Jar 1 contains 10 beans, of which one take place. isDiscussion white and the rest are black. Jar 2 contains 100 beans, of which seven are white 2. There are things one can do to have a moderate effect on the chances that the andJar the2 contains rest are moreblack. white You beanscan choose than Jar the 1. jar This from fact which turns theout drawingto be important will take in the event will occur. place.way peopleWhich makejar would choices you choose,in this decision Jar 1 or Jartask. 2? 3. The event is completely controllable. Discussion Jar 2 contains more white beans than Jar 1. This fact turns out to be important in the way people make choices in this decision task.

she77208_ch01_001-031.indd 14 she77208_ch01_001-031.indd2/7/17 5:25 PM 15 2/7/17 5:25 PM

she77208_ch01_001-031.indd 14 2/7/17 5:25 PM 14 Chapter One Behavioral Foundations 15

Discussion When psychologists conducted an experiment based on the above diagnostic Most people do not know their history well enough to remember that Attila the Hun question, in which real jars with real beans were used, they found that a majority of was defeated at the Battle of the Catalaunian Plains in 451 CE. Because most people subjects selected Jar 2 over Jar 1, even though Jar 1 featured aBehavioral higher probability Foundations of15 do not know the date, they have to estimate it based on whatever recollections they winning the $10. Moreover, many people stated that although they recognized that have. Mean predictions are highly variable across different classes, typically lying the probabilityWhen psychologists of winning conductedwas higher an with experiment Jar 1, they based found on themselves the above diagnosticemotion- between 415 and 1100. Mean responses for Europeans tend to be much closer to allyquestion, drawn to in choose which Jar real 2. jars In essence, with real people beans wereunderweighted used, they foundthe denominator that a majority when of 451 than non-Europeans. implicitlysubjects computing selected Jar the 2 overodds Jar of 1, winning even though depending Jar 1 featured on which a higher jar they probability selected. of Consider the first of the two parts of the preceding question. The first question Psychologistswinning the named $10. Moreover, the phenomenon many people “ratio stated bias” that and although concluded they thatrecognized the visual that is intended to provide a context for the way they think about the second question. stimulusthe probability of seeing of a winninghigher absolute was higher number with Jarof beans1, they in found Jar 2 themselves than Jar 1 emotion-was the The issue here is whether people arrive at their prediction of the year of Attila’s drivingally drawnfactor. to Jar choose 2 generated Jar 2. In a essence,higher level people of underweightedpositive affect thethan denominator Jar 1. when defeat by beginning with X and adjusting up or down. Specifically, do they anchor Theimplicitly role of computingaffect in choice the odds can beof winningillustrated depending by another on experiment. which jar theyIn this selected. experi- on X but not adjust sufficiently? ment,Psychologists experienced named forensic the psychologists phenomenon and “ratio psychiatrists bias” and wereconcluded asked thatto recommend the visual There is no rational reason to expect that people’s predictions should in any whetherstimulus to release of seeing a mental a higher patient, absolute Mr. Jones, number who of mightbeans bein violentJar 2 than within Jar six1 was months the way be correlated with the last three digits of their phone numbers. Yet in practice, of beingdriving discharged factor. Jar from 2 generated the hospital. a higher Notably, level ofthe positive subjects affect based than their Jar recommenda 1. - people’s responses are so correlated, with the sign of the correlation being positive. tions onThe information role of affect provided in choice by canother be clinical illustrated experts. by another When experiment. the information In this in experi ques-- At the same time, correlations are often statistically insignificant unless class sizes tionment, was described experienced as “20forensic out of psychologists every 100 patients and psychiatrists similar to Mr.were Jones asked are to estimatedrecommend to are very large, well over a hundred. commitwhether an actto releaseof violence,” a mental they patient, were Mr. twice Jones, as likelywho might to recommend be violent againstwithin six discharge months thanof when being the discharged information from was the described hospital. asNotably, “patients the similarsubjects to based Mr. Jones their recommendaare estimated- Affect Heuristic to havetions a on 20 information percent chance provided of committing by other clinical an act experts. of violence.” When Thethe information psychologists in quescon-- 15 Description cludedtion that was the described first description as “20 out generated of every 100 much patients more similar negative to Mr.affect Jones than are the estimated second. to commit an act of violence,” they were twice as likely to recommend against discharge affect affect, When psychologists use the word they are referring to the specific quality of Interactingthan when the Phenomena information was described as “patients similar to Mr. Jones are estimated An emotional feeling. goodness or badness. This quality is typically experienced as a feeling state, possibly Descriptionto have a 20 percent chance of committing an act of violence.” The psychologists con- consciously but not always, with goodness being experienced positively and bad- cluded that the first description generated much more negative affect than the second.15 ness being experienced negatively. People’s affective responses occur rapidly and Psychologists have discovered that one behavioral phenomenon can affect another. automatically, and relate to what is called their “affect pool.” This pool consists of a In thisInteracting respect, consider Phenomena the extent to which excessive optimism is related to the following four issues: “library” of positive and negative mental markers that people associate, either con- Description sciously or unconsciously, with the images stored in their memories. When people 1. ThePsychologists perceived havecontrollability discovered of that the one event. behavioral phenomenon can affect another. engage in the mental shortcut of basing their decisions on the contents of their affect 2. The desirability of the event. pools instead of weighing the pros and cons of various alternatives, or retrieving rel- In this respect, consider the extent to which excessive optimism is related to the 3. The perceived probability of the event, reflecting previous personal experience— affect heuristic evant information from their memories, they are said to rely on the affect heuristic. following four issues: effectively the degree to which the person is familiar with the event. Basing decisions primarily In this regard, psychologists note that retrieving relevant information from memory 1. The perceived controllability of the event. on intuition, instinct, and 4. The degree to which people view themselves as representative of the type of and then processing this information relies on other heuristics such as availability 2. The desirability of the event. gut feeling. persons to whom the event occurs. and representativeness. In contrast, the affective heuristic is more of a “quick and 3. The perceived probability of the event, reflecting previous personal experience— dirty” shortcut. Diagnosticeffectively Question the degree to which the person is familiar with the event. Diagnostic Question The4. question The degree pertaining to which to people excessive view optimism themselves involves as representative a series of of possible the type life of events. Please assess each of these events on the specified criteria. Imagine that you have an opportunity to win $10 by randomly drawing a white bean persons to whom the event occurs. from one of two jars, denoted Jar 1 and Jar 2. Jar 1 contains 10 beans, of which one A. Diagnostic For each of Questionthe life events, assign a controllability category number. The list of is white and the rest are black. Jar 2 contains 100 beans, of which seven are white Thechoices question includes: pertaining to excessive optimism involves a series of possible life and the rest are black. You can choose the jar from which the drawing will take events.1. There Please is nothing assess one each can of dothese that events will change on the specifiedthe likelihood criteria. that the event will place. Which jar would you choose, Jar 1 or Jar 2? A. take For eachplace. of the life events, assign a controllability category number. The list of Discussion 2. Therechoices are includes: things one can do to have a moderate effect on the chances that the event will occur. Jar 2 contains more white beans than Jar 1. This fact turns out to be important in the 1. There is nothing one can do that will change the likelihood that the event will way people make choices in this decision task. 3. Thetake event place. is completely controllable. 2. There are things one can do to have a moderate effect on the chances that the event will occur. 3. The event is completely controllable.

she77208_ch01_001-031.indd 14 she77208_ch01_001-031.indd2/7/17 5:25 PM 15 2/7/17 5:25 PM

she77208_ch01_001-031.indd 15 2/7/17 5:25 PM 16 Chapter One Behavioral Foundations 17

B. For each of the life events, assign a desirability number on a scale of 1 to 9 the lowest amount you would have to win in this risky alternative in order to accept where 1 means extremely undesirable, 5 is neutral, and 9 means extremely the risk? 16 Chapter One desirable. Discussion C. For each of the life events, assign a category number for familiarity, where the B. Forcategories each of therange life from events, “The assign event a desirabilityhas not happened number toon anyone a scale Iof know,” 1 to 9 and Psychological studies find that people’s responses to this question suggest that they where“The event 1 means has extremelyhappened undesirable,to me more than 5 is once.” neutral, and 9 means extremely experience a loss roughly 2.5 times as intensely as a gain of the same magnitude. loss aversion loss aversion. D.desirable. For each of the life events, assign a category number for mental imaging, where This phenomenon is called In this regard, the way a person answers Psychologically, people the preceding diagnostic question provides an indication of how loss averse he or C. Forthe each categories of the lifeare events, assign a category number for familiarity, where the experience a loss more categories range from “The event has not happened to anyone I know,” and she is. For example, we can say that a person whose response to the question is 1. No particular person with a high chance comes to mind. acutely than a gain of the “The event has happened to me more than once.” same magnitude. $1,250 has a coefficient of loss aversion of 2.5, since $1,250 is 2.5 times larger than the $500 he or she might lose in the event a tail comes up. Notably, loss aversion D. For2. each When of Ithe think life aboutevents, the assign event, a category a clear picturenumber comesfor mental to mind imaging, of a whereparticular leads people to be conservative about taking risk the categoriestype of person are to whom it is likely to happen. The psychological concept of framing is synonymous with “description,” as Discussion1. No particular person with a high chance comes to mind. in the phrase “framing of a decision task.” Suppose you were told that you 2. When I think about the event, a clear picture comes to mind of a particular would be paid $500 to participate in an experiment in which you would have the The generaltype of personfindings to whom about itexcessive is likely to optimism happen. is that its magnitude is related to opportunity to take a 50–50 risk where you might lose $500 or win an unspeci - other psychological variables. In this regard, one of the most important issues for fied amount. The experimenter will place the amount you might gain on a piece corporate finance involves the connection between control and excessive optimism. Discussion of paper face down in front of you, and will then ask you the diagnostic question The more that people report they are in control of a situation, the greater their opti- above. Once you have provided your answer, the experimenter will turn over the Themism general about findings how the situationabout excessive will turn optimism out. is that its magnitude is related to piece of paper, and if your answer is less than or equal to the amount of the gain otherWishful psychological thinking variables. also affects In this optimism. regard, Peopleone of theare mostmore important optimistic issues about for events shown, you will face the risk. Otherwise, the experiment will terminate at that corporatethat are desirablefinance involves than events the connectionthat are undesirable. between control and excessive optimism. point, and you will collect your participation fee. If you do face the risk, then at The Availabilitymore that people through report personal they are experience in control alsoof a situation,affects optimism. the greater The their more opti- famil- the end of the experiment, you will receive the sum of the participation fee and mismiar people about howfeel thewith situation a situation, will turnthe out.greater their optimism about how the situation the outcome of the risk. willWishful turn out.thinking People also are affects more optimism. optimistic People about are events more that optimistic have happened about events to them Will you answer the diagnostic question differently when it is part of an experi- thatthan are about desirable events than that events have thatnot happenedare undesirable. to anyone that they know. ment with a $500 participation fee than not? When there is a participation fee, will Availability through personal experience also affects optimism. The more famil- Representativeness also affects optimism. When people have a clear picture of you ignore it when answering the question because you receive it regardless of how iar people feel with a situation, the greater their optimism about how the situation the type of person to whom the event is likely to happen, and see themselves as you answer the diagnostic question? Or will you instead think about the diagnostic will turn out. People are more optimistic about events that have happened to them resembling that person, they are inclined to be especially optimistic. This phenom- question in net winning terms, seeing yourself as either winning $0 or winning than about events that have not happened to anyone that they know. enon is accentuated, the more likely people view the event as being. some amount plus $500? Representativeness also affects optimism. When people have a clear picture of People who ignore the $500 participation fee when answering the diagnostic the type of person to whom the event is likely to happen, and see themselves as question continue to frame the outcome associated with a tail as a $500 loss 1.4 FRAMING EFFECTSresembling that person, they are inclined to be especially optimistic. This phenom- and for most, losses loom larger than gains. Those who instead frame the out- enon is accentuated, the more likely people view the event as being. framing effect Framing effects pertain to the manner in which people’s choices are influenced by come associated with a tail as a $0 gain do not perceive themselves as facing a A person’s decisions are the descriptions of the decision tasks they face. This section uses a series of ques- fourfold pattern potential loss, and so loss aversion is not operative for them in this particular 1.4influenced FRAMING by the manner EFFECTStions to discuss two types of framing effects. When probabilities of situation. in which the setting for the nonzero extreme outcomes In the diagnostic question, loss aversion leads people to ask for a higher gain in framingdecision iseffect described. Framing effects are moderate, people are Loss Aversion pertain to the manner in which people’s choices are influenced by risk averse in the domain order to take a risk in which they might lose money. Those who frame the choice A person’s decisions are the descriptions of the decision tasks they face. This section uses a series of ques- of gains, and risk seeking as being only in the domain of gains would be inclined to ask for a zero gain in the influenced by the manner tionsDescription to discuss two types of framing effects. in the domain of losses. event of a head than in the original version of the diagnostic question, netting either in which the setting for the When probabilities of decision is described. People feel a loss more intensely than a gain of the same magnitude. zero or $500. Loss Aversion extreme nonzero outcomes Diagnostic Question are small, people are risk The Fourfold Risk Pattern Description seeking in the domain of Imagine a 50–50 risk involving a coin toss where you lose $500 if the coin toss gains, and risk averse in the In contrast to loss aversion, which applies to risks involving a mixture of Peoplecomes feel up tails,a loss but more win intensely a different than amount a gain of if the samecoin tossmagnitude. comes up heads. What is domain of losses. possible gains and losses, the fourfold pattern applies to risks whose possible Diagnostic Question Imagine a 50–50 risk involving a coin toss where you lose $500 if the coin toss comes up tails, but win a different amount if the coin toss comes up heads. What is

she77208_ch01_001-031.indd 16 she77208_ch01_001-031.indd2/7/17 5:25 PM 17 2/7/17 5:25 PM

she77208_ch01_001-031.indd 16 2/7/17 5:25 PM 16 Chapter One Behavioral Foundations 17

B. For each of the life events, assign a desirability number on a scale of 1 to 9 the lowest amount you would have to win in this risky alternative in order to accept where 1 means extremely undesirable, 5 is neutral, and 9 means extremely the risk? desirable. Behavioral Foundations 17 Discussion C. For each of the life events, assign a category number for familiarity, where the categories range from “The event has not happened to anyone I know,” and Psychologicalthe lowest amount studies youfind would that people’s have to winresponses in this riskyto this alternative question suggestin order thatto accept they “The event has happened to me more than once.” experiencethe risk? a loss roughly 2.5 times as intensely as a gain of the same magnitude. loss aversion loss aversion. D. For each of the life events, assign a category number for mental imaging, where This phenomenon is called In this regard, the way a person answers Psychologically, people Discussion the categories are the preceding diagnostic question provides an indication of how loss averse he or experience a loss more she Psychologicalis. For example, studies we findcan thatsay people’sthat a person responses whose to this response question to suggest the question that they is 1. No particular person with a high chance comes to mind. acutely than a gain of the same magnitude. $1,250experience has a coefficient a loss roughly of loss 2.5 aversion times as of intensely 2.5, since as $1,250a gain ofis 2.5the timessame largermagnitude. than 2. When I think about the event, a clear picture comes to mind of a particular loss aversion the This$500 phenomenon he or she might is called lose loss in the aversion. event a In tail this comes regard, up. the Notably, way a person loss aversion answers type of person to whom it is likely to happen. Psychologically, people leadsthe people preceding to be diagnostic conservative question about provides taking risk an indication of how loss averse he or experience a loss more Theshe is.psychological For example, concept we can ofsay framing that a personis synonymous whose response with “description,” to the question as is acutely than a gain of the Discussion same magnitude. in the$1,250 phrase has a “framing coefficient of of aloss decision aversion task.” of 2.5, Suppose since $1,250 you is were 2.5 times told larger that youthan wouldthe be$500 paid he $500or she to might participate lose in in the an event experiment a tail comes in which up. Notably, you would loss have aversion the The general findings about excessive optimism is that its magnitude is related to opportunityleads people to taketo be a conservative 50–50 risk whereabout taking you might risk lose $500 or win an unspeci - other psychological variables. In this regard, one of the most important issues for fied amount.The psychological The experimenter concept will of framing place the is amountsynonymous you mightwith “description,” gain on a piece as corporate finance involves the connection between control and excessive optimism. of paperin the face phrase down “framing in front of you, a decision and will task.” then Supposeask you the you diagnostic were told question that you The more that people report they are in control of a situation, the greater their opti- above.would Once be paidyou have$500 providedto participate your in answer, an experiment the experimenter in which you will would turn overhave the mism about how the situation will turn out. pieceopportunity of paper, andto take if your a 50–50 answer risk is where less than you mightor equal lose to $500 the amount or win anof theunspeci gain - Wishful thinking also affects optimism. People are more optimistic about events shown,fied amount.you will The face experimenter the risk. Otherwise, will place the the experiment amount you will might terminate gain on aat piece that that are desirable than events that are undesirable. point,of paperand you face will down collect in front your of participationyou, and will thenfee. Ifask you you do the face diagnostic the risk, question then at Availability through personal experience also affects optimism. The more famil- the above.end of Oncethe experiment, you have provided you will your receive answer, the the sum experimenter of the participation will turn overfee and the iar people feel with a situation, the greater their optimism about how the situation the pieceoutcome of paper, of the and risk. if your answer is less than or equal to the amount of the gain will turn out. People are more optimistic about events that have happened to them Willshown, you you answer will theface diagnostic the risk. Otherwise, question differently the experiment when itwill is partterminate of an experi-at that than about events that have not happened to anyone that they know. mentpoint, with and a $500 you participationwill collect your fee thanparticipation not? When fee. there If you is doa participation face the risk, fee, then will at Representativeness also affects optimism. When people have a clear picture of youthe ignore end itof when the experiment, answering theyou question will receive because the sumyou ofreceive the participation it regardless fee of howand the type of person to whom the event is likely to happen, and see themselves as youthe answer outcome the diagnostic of the risk. question? Or will you instead think about the diagnostic resembling that person, they are inclined to be especially optimistic. This phenom- questionWill in you net answer winning the terms,diagnostic seeing question yourself differently as either when winning it is part $0 of or an winning experi- enon is accentuated, the more likely people view the event as being. somement amount with aplus $500 $500? participation fee than not? When there is a participation fee, will Peopleyou ignore who it ignore when answering the $500 participationthe question because fee when you answeringreceive it regardless the diagnostic of how 1.4 FRAMING EFFECTS questionyou answer continue the diagnostic to frame question?the outcome Or will associated you instead with think a tail about as thea $500 diagnostic loss andquestion for most, in losses net winning loom larger terms, than seeing gains. yourself Those as eitherwho instead winning frame $0 or the winning out- framing effect Framing effects pertain to the manner in which people’s choices are influenced by comesome associated amount plus with $500? a tail as a $0 gain do not perceive themselves as facing a People who ignore the $500 participation fee when answering the diagnostic A person’s decisions are the descriptions of the decision tasks they face. This section uses a series of ques- fourfold pattern potential loss, and so loss aversion is not operative for them in this particular question continue to frame the outcome associated with a tail as a $500 loss influenced by the manner tions to discuss two types of framing effects. When probabilities of situation. in which the setting for the nonzero extreme outcomes Inand the for diagnostic most, losses question, loom losslarger aversion than gains. leads Those people who to askinstead for a frame higher the gain out in- decision is described. are moderate, people are Loss Aversion ordercome to take associated a risk inwith which a tail they as amight $0 gain lose do money. not perceive Those themselves who frame as the facing choice a risk aversefourfold in the pattern domain potential loss, and so loss aversion is not operative for them in this particular of gains, and risk seeking as being only in the domain of gains would be inclined to ask for a zero gain in the When probabilities of Description in the domain of losses. eventsituation. of a head than in the original version of the diagnostic question, netting either nonzero extreme outcomes When probabilities of In the diagnostic question, loss aversion leads people to ask for a higher gain in People feel a loss more intensely than a gain of the same magnitude. are moderate, people are zero or $500. extreme nonzero outcomes risk averse in the domain order to take a risk in which they might lose money. Those who frame the choice are small, people are risk Diagnostic Question of gains, and risk seeking Theas Fourfoldbeing only inRisk the domainPattern of gains would be inclined to ask for a zero gain in the seeking in the domain of in the domain of losses. event of a head than in the original version of the diagnostic question, netting either Imagine a 50–50 risk involving a coin toss where you lose $500 if the coin toss gains, and risk averse in the In contrast to loss aversion, which applies to risks involving a mixture of When probabilities of domain of losses. zero or $500. fourfold pattern comes up tails, but win a different amount if the coin toss comes up heads. What is extreme nonzero outcomes possible gains and losses, the applies to risks whose possible are small, people are risk The Fourfold Risk Pattern seeking in the domain of gains, and risk averse in the In contrast to loss aversion, which applies to risks involving a mixture of domain of losses. possible gains and losses, the fourfold pattern applies to risks whose possible

she77208_ch01_001-031.indd 16 she77208_ch01_001-031.indd2/7/17 5:25 PM 17 2/7/17 5:25 PM

she77208_ch01_001-031.indd 17 2/7/17 5:25 PM 18 Chapter One Behavioral Foundations 19

outcomes are either only gains or only losses. Below you will find an explanation Diagnostic Question of the fourfold pattern through four examples. Together the four examples illustrate Suppose that you face a choice between two risks. Which of the two would you 18 Chapter One how people’s willingness to take risk can depend on the circumstances in which choose? they find themselves, rather than being the same in all circumstances. outcomes are either only gains or only losses. Below you will find an explanation E. 90% chance of losing $2,000. ofDescription: the fourfold pattern Pattern through One four examples. Together the four examples illustrate 10% chance of losing $0. howFor people’srisks in willingnesswhich the probabilitiesto take risk canattached depend to on nonzero the circumstances gains are not in whichtoo small, F. 45% chance of losing $4,000. theypeople find are themselves, inclined torather be risk-averse. than being the same in all circumstances. 55% chance of losing $0. Description:Diagnostic QuestionPattern One Discussion ForSuppose risks inthat which you theface probabilities a choice between attached two to nonzerorisks. Which gains ofare the not two too wouldsmall, you The probabilities attached to nonzero losses in both E and F are moderate, not small. peoplechoose? are inclined to be risk-averse. Moreover, both risks in the above decision task feature an expected loss of $1,800. However, F is riskier than E. Most people choose F over E when facing this choice, A. 90% chance of winning $2,000. Diagnostic Question which is consistent with risk-seeking behavior. 10% chance of zero. Suppose that you face a choice between two risks. Which of the two would you Description: Pattern Four choose?B. 45% chance of winning $4,000. 55% chance of zero. For risks in which the probabilities attached to nonzero losses are quite small, people A. 90% chance of winning $2,000. are inclined to be risk-averse. 10% chance of zero. Discussion Diagnostic Question B.The 45% probabilities chance of winning attached $4,000. to nonzero outcomes in both A and B are moderate, not Suppose that you face a choice between two risks. Which of the two would you choose? small. 55% Moreover,chance of zero. both risks in the above decision task feature an expected gain of $1,800. However, B is riskier than A. Most people choose A over B when facing G. Lose $2,000 with probability .002. Discussion this choice, which is consistent with risk-averse behavior. Lose 0 with probability .998. The probabilities attached to nonzero outcomes in both A and B are moderate, not H. Lose $4,000 with probability .001. small.Description: Moreover, Pattern both risks Two in the above decision task feature an expected gain of $1,800.For risks However, in which B the is riskierprobabilities than A. attached Most people to nonzero choose gains A over are Bquite when small, facing people Lose $0 with probability .999. thisare choice, inclined which to be is risk-seeking. consistent with risk-averse behavior. Discussion Description:Diagnostic QuestionPattern Two The probabilities attached to nonzero losses in both G and H are small. Moreover, both risks in the above decision task feature an expected loss of $4. However, H ForSuppose risks in that which you the face probabilities a choice attachedbetween to two nonzero risks. gains Which are quiteof the small, two peoplewould you is riskier than G. Most people choose G over H when facing this choice, which is arechoose? inclined to be risk-seeking. consistent with risk-averse behavior. DiagnosticC. $2,000 Questionwith probability .002. Exhibit 1-3 summarizes the fourfold pattern. Suppose 0 with that probability you face a .998. choice between two risks. Which of the two would you Framing Pitfalls choose?D. $4,000 with probability .001. In a behavioral context, a “decision frame” is synonymous with the description of a C. $2,0000 with with probability probability .999. .002. decision task. For example, consider the following decision task: Discussion 0 with probability .998. Suppose that you face the following pair of concurrent decisions, with the out- come of your first decision being determined tomorrow afternoon and the outcome D.The $4,000 probabilities with probability attached .001. to nonzero outcomes in both C and D are small. More- over,0 with both probability risks in the .999. above decision task feature an expected gain of $4. However, D is riskier than C. Most people choose D over C when facing this choice, which is EXHIBIT 1-3 Discussion Fourfold Pattern Associated with Decision Tasks in Which the Risks consistent with risk-seeking behavior. Involve Either Gains Only or Losses Only The probabilities attached to nonzero outcomes in both C and D are small. More- over,Description: both risks Patternin the above Three decision task feature an expected gain of $4. However, Gains Losses DFor is riskier risks inthan which C. Most the peopleprobabilities choose attachedD over C towhen nonzero facing losses this choice, are not which too issmall, Probabilities moderate Risk averse Risk seeking Probabilities small Risk seeking Risk averse consistentpeople are with inclined risk-seeking to be risk-seeking. behavior. Description: Pattern Three For risks in which the probabilities attached to nonzero losses are not too small, people are inclined to be risk-seeking.

she77208_ch01_001-031.indd 18 she77208_ch01_001-031.indd2/7/17 5:25 PM 19 2/7/17 5:25 PM

she77208_ch01_001-031.indd 18 2/7/17 5:25 PM 18 Chapter One Behavioral Foundations 19 outcomes are either only gains or only losses. Below you will find an explanation Diagnostic Question of the fourfold pattern through four examples. Together the four examples illustrate Suppose that you face a choice between two risks. Which of the two would you how people’s willingness to take risk can depend on the circumstances in which choose? Behavioral Foundations 19 they find themselves, rather than being the same in all circumstances. E. Diagnostic 90% chance Question of losing $2,000. Description: Pattern One 10%Suppose chance that of you losing face $0. a choice between two risks. Which of the two would you For risks in which the probabilities attached to nonzero gains are not too small, F. choose? 45% chance of losing $4,000. people are inclined to be risk-averse. 55% chance of losing $0. E. 90% chance of losing $2,000. Diagnostic Question Discussion 10% chance of losing $0. Suppose that you face a choice between two risks. Which of the two would you TheF. probabilities 45% chance attached of losing to $4,000. nonzero losses in both E and F are moderate, not small. choose? Moreover, 55% both chance risks of inlosing the above$0. decision task feature an expected loss of $1,800. However, F is riskier than E. Most people choose F over E when facing this choice, A. 90% chance of winning $2,000. whichDiscussion is consistent with risk-seeking behavior. 10% chance of zero. Description:The probabilities Pattern attached Four to nonzero losses in both E and F are moderate, not small. B. 45% chance of winning $4,000. Moreover, both risks in the above decision task feature an expected loss of $1,800. 55% chance of zero. For However,risks in which F is riskier the probabilities than E. Most attached people tochoose nonzero F over losses E when are quite facing small, this choice,people are whichinclined is consistentto be risk-averse. with risk-seeking behavior. Discussion DiagnosticDescription: Question Pattern Four The probabilities attached to nonzero outcomes in both A and B are moderate, not Suppose that you face a choice between two risks. Which of the two would you choose? small. Moreover, both risks in the above decision task feature an expected gain of For risks in which the probabilities attached to nonzero losses are quite small, people $1,800. However, B is riskier than A. Most people choose A over B when facing G. are Lose inclined $2,000 to with be risk-averse. probability .002. this choice, which is consistent with risk-averse behavior. Diagnostic Lose 0 with Question probability .998. Description: Pattern Two H. Suppose Lose $4,000 that you with face probability a choice between .001. two risks. Which of the two would you choose? For risks in which the probabilities attached to nonzero gains are quite small, people G. Lose Lose $0 with$2,000 probability with probability .999. .002. are inclined to be risk-seeking. Discussion Lose 0 with probability .998. Diagnostic Question TheH. probabilities Lose $4,000 attached with probability to nonzero .001. losses in both G and H are small. Moreover, both risks Lose in $0 the with above probability decision .999. task feature an expected loss of $4. However, H Suppose that you face a choice between two risks. Which of the two would you is riskier than G. Most people choose G over H when facing this choice, which is choose? consistentDiscussion with risk-averse behavior. C. $2,000 with probability .002. ExhibitThe probabilities 1-3 summarizes attached the to fourfoldnonzero pattern.losses in both G and H are small. Moreover,

0 with probability .998. both risks in the above decision task feature an expected loss of $4. However, H Framingis riskier Pitfalls than G. Most people choose G over H when facing this choice, which is D. $4,000 with probability .001. In aconsistent behavioral with context, risk-averse a “decision behavior. frame” is synonymous with the description of a 0 with probability .999. decisionExhibit task. 1-3For summarizesexample, consider the fourfold the following pattern. decision task: Discussion SupposeFraming that Pitfalls you face the following pair of concurrent decisions, with the out- come of your first decision being determined tomorrow afternoon and the outcome The probabilities attached to nonzero outcomes in both C and D are small. More- In a behavioral context, a “decision frame” is synonymous with the description of a over, both risks in the above decision task feature an expected gain of $4. However, decision task. For example, consider the following decision task: D is riskier than C. Most people choose D over C when facing this choice, which is EXHIBITSuppose 1-3 Fourfoldthat you Patternface the Associated following with pair Decisionof concurrent Tasks indecisions, Which the with Risks the out- consistent with risk-seeking behavior. come of yourInvolve first decision Either Gains being Only determined or Losses tomorrow Only afternoon and the outcome Description: Pattern Three Gains Losses For risks in which the probabilities attached to nonzero losses are not too small, ProbabilitiesEXHIBIT 1-3moderate Fourfold Pattern AssociatedRisk with averse Decision Tasks in Which theRisk Risks seeking Probabilities small Risk seeking Risk averse people are inclined to be risk-seeking. Involve Either Gains Only or Losses Only Gains Losses Probabilities moderate Risk averse Risk seeking Probabilities small Risk seeking Risk averse

she77208_ch01_001-031.indd 18 she77208_ch01_001-031.indd2/7/17 5:25 PM 19 2/7/17 5:25 PM

she77208_ch01_001-031.indd 19 2/7/17 5:25 PM 20 Chapter One Behavioral Foundations 21

of your second decision being determined tomorrow evening. Examine both deci- EXHIBIT 1-4 Alternative Frame for the Concurrent Choice Problem sions, and then make your choices, keeping in mind that the outcomes are one-shot Probability Outcome Probability Outcome 20 Chapter One deals. I 100% 2,400 I and K 100% –$5,100 First decision: Choose between J 75% 0 I and L 25% $2,400 of your second decision being determined tomorrow evening. Examine both deci- 25% 10,000 75% –$7,600 I. A sure gain of $2,400. sions, and then make your choices, keeping in mind that the outcomes are one-shot K 100% –7,500 J and K 75% –$7,500 deals.J. A 25 percent chance to gain $10,000 and a 75 percent chance to gain nothing. 25% $2,500 L 25% 0 J and L 18.75% $0 FirstSecond decision: decision: Choose Choose between between 75% –10,000 56.25% –$10,000 I.K. A A sure sure gain loss of of $2,400. $7,500. 6.25% $10,000 18.75% $0 J.L. A A 25 75 percent percent chance chance to togain lose $10,000 $10,000 and and a 75 a percent25 percent chance chance to gain to losenothing. nothing. SecondThe frame decision: used Chooseto describe between this decision task presents the choice as consisting of twoK. concurrentA sure loss ofdecisions, $7,500. one in the domain of gains and the second in the domain of losses. When people are presented with the task framed in this way, they tend L. A 75 percent chance to lose $10,000 and a 25 percent chance to lose nothing. keep in mind that they are different concepts. Aversion to a sure loss entails the to choose the combination “I and L.” In the context of the fourfold pattern, it is willingness to accept risk in order to avert a sure loss. Loss aversion entails the easyThe to frame understand used to describewhy. When this consideringdecision task the presents first part the choiceof the astask, consisting they perceive of reluctance to accept risk when both gains and losses are possible, because psycho- twothemselves concurrent to decisions,be choosing one in in the the domain domain of of gains, gains withand the the second probability in the of domain a nonzero logically losses loom larger than gains. ofgain losses. not When being people small, are and presented as a result with tend the totask exhibit framed risk-averse in this way, behavior. they tend When toconsidering choose the thecombination second part “I ofand the L.” task, In thethey context perceive of themselvesthe fourfold topattern, be choosing it is in easythe domainto understand of losses, why. with When the considering probability the of firsta nonzero part of loss the nottask, being they small,perceive and as Prospect Theory themselvesa result tend to beto choosingexhibit risk-seeking in the domain behavior. of gains, with the probability of a nonzero prospect theory Prospect theory is a formal framework that presents a set of organizing principles gainNotice not being that small,the combination and as a result of separate tend to decisions exhibit risk-aversein the concurrent behavior. decision When task A general psychological to explain the framing issues described above. The theory was developed by psy- consideringinvolves mixed the second gains andpart losses.of the task, In this they regard, perceive the themselvescombination to beI and choosing L effectively in approach that describes the chologists Daniel Kahneman and Amos Tversky, for which Kahneman was awarded way people make choices 16 theadds domain $2,400 of tolosses, the outcome with the ofprobability L. Therefore, of a nonzero I and L lossis a notrisk being featuring small, a 25and percent as among risky alternatives. an Economics Nobel Prize in 2002. In making the award, the Nobel committee a probabilityresult tend toof exhibit incurring risk-seeking a $2,400 behavior. gain and a 75 percent probability of incurring a explicitly cited prospect theory, thereby suggesting that they would have conferred $7,600Notice loss. that theHowever, combination because of separateof the way decisions the decision in the concurrenttask is framed, decision most task people the award on Tversky, had he not died several years before. involvesdo not thinkmixed aboutgains andit this losses. way. In As this a regard,result, the combinationfourfold pattern I and effectively L effectively drives According to prospect theory, people act as if their incentives are to make choices addstheir $2,400 choices, to thewith outcome loss aversion of L. Therefore, playing no I androle Ldespite is a risk the featuring presence a of25 mixedpercent gains as though using a criterion similar to expected cash flow. You can think of con- probabilityand losses. of incurring a $2,400 gain and a 75 percent probability of incurring a structing the formula for this criterion by beginning with the formula for expected $7,600There loss. is However,a deeper framingbecause issueof the here way asthe well. decision The taskcombination is framed, J mostand K people is a better cash flow and then substituting alternatives for cash flows and probabilities. In the dochoice not think than aboutI and itL. this The way. combination As a result, J and the Kfourfold offers patterna 25 percent effectively probability drives of a substitution, the alternative to cash flow is the mental sensation an individual ex- their$2,500 choices, gain withand aloss 75 aversionpercent probability playing no roleof a despite$7,500 the loss. presence Notice ofthat mixed J and gains K offers periences from anticipating the gain or loss associated with that cash flow. Replace andthe losses. same probabilities as I and L, but $100 more regardless of whether the outcome each probability with the mental sensation associated with anticipating that prob- isThere a gain is or a adeeper loss: $2,500framing versus issue here$2,400 as well.and −$7,500 The combination versus −$7,600. J and K isNevertheless, a better ability, along with the relative ranking of its associated cash flow. For the sake of choicebecause than of Iframing, and L. The I and combination L emerges J theand most K offers popular a 25 choice percent combination probability ofand a few reference, call the resulting formula “prospect theory expected value.” $2,500people gain choose and Ja and 75 percentK. Exhibit probability 1-4 summarizes of a $7,500 the loss. structure Notice of that all four J and combinations. K offers psychophysics A single psychological principle called psychophysics defines the character the sameThe choiceprobabilities of L over as I andK in L, the but concurrent $100 more choice regardless problem of whether reflects the more outcome than the The diminishing marginal of these mental sensations. According to psychophysics, mental sensations in- istendency a gain or toa loss: be risk-seeking $2,500 versus in $2,400the domain and −$7,500 of losses. versus It also −$7,600. reflects Nevertheless, a feature known impact associated with crease with the magnitude of both cash flows and probabilities, but less than because of framing, I and L emerges the most popular choice combination and few successive increases in as “aversion to a sure loss.” In this case, choosing K represents accepting the sure stimuli. proportionally. aversion to a sure loss peopleloss, and choose choosing J and K. L Exhibitrepresents 1-4 summarizesthe aversion the to structure a sure loss.of all Peoplefour combinations. who are averse The psychophysics principle provides prospect theory with the power to ex - People choose to accept an toThe a sure choice loss of are L actuallyover K in willing the concurrent to take actuarially choice problem unfair reflects risks, hopingmore than to beatthe the plain the fourfold pattern that includes aversion to a sure loss. In this regard, the actuarially unfair risk in an tendencyodds, in toorder be risk-seeking to avoid sure in losses.the domain of losses. It also reflects a feature known mental sensations associated with anticipating a sequence of successive gains attempt to avoid a sure loss. as “aversionAversion to to a asure sure loss.” loss Inis thisan issue case, that choosing arises K at represents several points accepting in this the book. sure Al- increase, but at a decreasing rate. A similar statement applies to successive aversion to a sure loss aversion to a sure loss. loss,though and bothchoosing “loss L aversion”represents andthe “aversion to a sure loss” People seem wholike aresimilar averse terms, losses. Likewise, the mental sensations from anticipating successive increases People choose to accept an to a sure loss are actually willing to take actuarially unfair risks, hoping to beat the actuarially unfair risk in an odds, in order to avoid sure losses. attempt to avoid a sure loss. Aversion to a sure loss is an issue that arises at several points in this book. Al- though both “loss aversion” and “aversion to a sure loss” seem like similar terms,

she77208_ch01_001-031.indd 20 she77208_ch01_001-031.indd2/7/17 5:25 PM 21 2/7/17 5:25 PM

she77208_ch01_001-031.indd 20 2/7/17 5:25 PM 20 Chapter One Behavioral Foundations 21 of your second decision being determined tomorrow evening. Examine both deci- EXHIBIT 1-4 Alternative Frame for the Concurrent Choice Problem sions, and then make your choices, keeping in mind that the outcomes are one-shot Probability Outcome Probability Outcome deals. Behavioral Foundations 21 I 100% 2,400 I and K 100% –$5,100 First decision: J 75% 0 I and L 25% $2,400 Choose between EXHIBIT 1-4 25% Alternative10,000 Frame for the Concurrent Choice Problem75% –$7,600 I. A sure gain of $2,400. K 100%Probability–7,500 Outcome J and K Probability75% Outcome–$7,500 25% $2,500 J. A 25 percent chance to gain $10,000 and a 75 percent chance to gain nothing. I 100% 2,400 I and K 100% –$5,100 Second decision: L J 25%75% 0 0 J andI and L L 18.75%25% $2,400$0 Choose between 75% 10,000 56.25% $10,000 25% – 10,000 75% – –$7,600 6.25% $10,000 K. A sure loss of $7,500. K 100% 7,500 J and K 75% $7,500 – 18.75% – $0 L. A 75 percent chance to lose $10,000 and a 25 percent chance to lose nothing. 25% $2,500 L 25% 0 J and L 18.75% $0 The frame used to describe this decision task presents the choice as consisting of 75% –10,000 56.25% –$10,000 two concurrent decisions, one in the domain of gains and the second in the domain 6.25% $10,000 of losses. When people are presented with the task framed in this way, they tend keep in mind that they are different concepts. Aversion to a18.75% sure loss entails$0 the to choose the combination “I and L.” In the context of the fourfold pattern, it is willingness to accept risk in order to avert a sure loss. Loss aversion entails the easy to understand why. When considering the first part of the task, they perceive reluctance to accept risk when both gains and losses are possible, because psycho- themselves to be choosing in the domain of gains, with the probability of a nonzero logically losses loom larger than gains. gain not being small, and as a result tend to exhibit risk-averse behavior. When keep in mind that they are different concepts. Aversion to a sure loss entails the considering the second part of the task, they perceive themselves to be choosing in willingness to accept risk in order to avert a sure loss. Loss aversion entails the the domain of losses, with the probability of a nonzero loss not being small, and as Prospectreluctance Theory to accept risk when both gains and losses are possible, because psycho- a result tend to exhibit risk-seeking behavior. prospect theory Prospectlogically theory losses is loom a formal larger framework than gains. that presents a set of organizing principles Notice that the combination of separate decisions in the concurrent decision task A general psychological to explain the framing issues described above. The theory was developed by psy- involves mixed gains and losses. In this regard, the combination I and L effectively approach that describes the chologists Daniel Kahneman and Amos Tversky, for which Kahneman was awarded way people make choices Prospect Theory 16 adds $2,400 to the outcome of L. Therefore, I and L is a risk featuring a 25 percent among risky alternatives. an Economics Nobel Prize in 2002. In making the award, the Nobel committee probability of incurring a $2,400 gain and a 75 percent probability of incurring a prospect theory explicitlyProspect cited theory prospect is a formaltheory, framework thereby suggesting that presents that a they set of would organizing have conferredprinciples $7,600 loss. However, because of the way the decision task is framed, most people A general psychological the toaward explain on Tversky,the framing had issues he not described died several above. years The before. theory was developed by psy- approach that describes the do not think about it this way. As a result, the fourfold pattern effectively drives Accordingchologists Danielto prospect Kahneman theory, and people Amos act Tversky, as if their for incentives which Kahneman are to make was awardedchoices way people make choices 16 their choices, with loss aversion playing no role despite the presence of mixed gains among risky alternatives. as thoughan Economics using a Nobel criterion Prize similar in 2002. to expected In making cash the flow. award, You the can Nobel think committee of con- and losses. structingexplicitly the citedformula prospect for this theory, criterion thereby by beginningsuggesting withthat theythe formulawould have for conferredexpected There is a deeper framing issue here as well. The combination J and K is a better cashthe flow award and on then Tversky, substituting had he alternativesnot died several for cash years flows before. and probabilities. In the choice than I and L. The combination J and K offers a 25 percent probability of a substitution,According the toalternative prospect theory,to cash people flow isact the as ifmental their incentives sensation are an to individual make choices ex- $2,500 gain and a 75 percent probability of a $7,500 loss. Notice that J and K offers periencesas though from using anticipating a criterion the similar gain or to loss expected associated cash with flow. that You cash can flow. think Replace of con- the same probabilities as I and L, but $100 more regardless of whether the outcome eachstructing probability the formulawith the for mental this criterion sensation by associatedbeginning withwith the anticipating formula for that expected prob- is a gain or a loss: $2,500 versus $2,400 and −$7,500 versus −$7,600. Nevertheless, ability,cash along flow andwith then the substitutingrelative ranking alternatives of its associatedfor cash flows cash and flow. probabilities. For the sake In theof because of framing, I and L emerges the most popular choice combination and few reference,substitution, call the the resulting alternative formula to cash “prospect flow is thetheory mental expected sensation value.” an individual ex- people choose J and K. Exhibit 1-4 summarizes the structure of all four combinations. psychophysics Aperiences single psychological from anticipating principle the gain called or loss psychophysics associated with thatdefines cash theflow. character Replace The choice of L over K in the concurrent choice problem reflects more than the The diminishing marginal of theseeach probabilitymental sensations. with the mentalAccording sensation to psychophysics, associated with mental anticipating sensations that prob- in- tendency to be risk-seeking in the domain of losses. It also reflects a feature known impact associated with creaseability, with along the withmagnitude the relative of both ranking cash of flows its associated and probabilities, cash flow. Forbut theless sake than of successive increases in reference, call the resulting formula “prospect theory expected value.” as “aversion to a sure loss.” In this case, choosing K represents accepting the sure stimuli. proportionally. aversion to a sure loss loss, and choosing L represents the aversion to a sure loss. People who are averse psychophysics TheA psychophysics single psychological principle principle provides called prospect psychophysics theory with defines the power the character to ex - People choose to accept an The diminishing marginal of these mental sensations. According to psychophysics, mental sensations in- to a sure loss are actually willing to take actuarially unfair risks, hoping to beat the impact associated with plain the fourfold pattern that includes aversion to a sure loss. In this regard, the actuarially unfair risk in an odds, in order to avoid sure losses. mentalcrease sensations with the associatedmagnitude withof both anticipating cash flows a andsequence probabilities, of successive but less gains than attempt to avoid a sure loss. successive increases in Aversion to a sure loss is an issue that arises at several points in this book. Al- stimuli. increase,proportionally. but at a decreasing rate. A similar statement applies to successive though both “loss aversion” and “aversion to a sure loss” seem like similar terms, losses.The Likewise, psychophysics the mental principle sensations provides from prospect anticipating theory successivewith the power increases to ex - plain the fourfold pattern that includes aversion to a sure loss. In this regard, the mental sensations associated with anticipating a sequence of successive gains increase, but at a decreasing rate. A similar statement applies to successive losses. Likewise, the mental sensations from anticipating successive increases

she77208_ch01_001-031.indd 20 she77208_ch01_001-031.indd2/7/17 5:25 PM 21 2/7/17 5:25 PM

she77208_ch01_001-031.indd 21 2/7/17 5:25 PM 22 Chapter One Behavioral Foundations 23

in the probability of an extreme event increases with the likelihood with which One way of adapting these ideas from individual choice to firm choice is to consider the event occurs, but at a decreasing rate. The latter statement holds up to a March-Shapira firms as having two potential reference points, called focal points, one for survival and framework 22 Chapter One point, in that the mental sensations from anticipating successive decreases in the the other for aspiration. At any one time, only the first focal point will be operative, A model with two reference probability of a highly likely event decreases with the probability of the event, (focal) points, one low depending on the extent to which a firm experiences financial distress along with the inbut the again probability at a decreasing of an extreme rate. event increases with the likelihood with which (survival) and one high personality traits of its executives and board members. According to the theory, known the Prospectevent occurs, theory but explains at a decreasing loss aversion rate. The by treatinglatter statement the mental holds sensations up to a as- (aspiration), with only as the March-Shapira framework, a distressed firm takes low risk but increases that point,sociated in that with the the mental anticipation sensations of fromlosses anticipating to be larger successive than the decreasescorresponding in the sen - one point operative at any risk as the level of distress declines.18 For a non-distressed firm, the theory postulates probabilitysations associated of a highly with likely the eventanticipation decreases of gains.with the According probability to ofprospect the event, theory, one time, and risk appetite that when a firm is operating below aspiration, and therefore experiencing a shortfall, reference point reference increasing with the absolute butan againindividual at a decreasing frames choices rate. in terms of gains and losses relative to a distance between the current it takes on relatively high risk. However, as the shortfall shrinks the amount of risk Benchmark used to measure point.Prospect Cash theory flows explains above the loss reference aversion point by treating are framed the mentalas gains sensations while cash as flows- situation and the reference declines. When the shortfall turns into a surplus, risk falls discontinuously at the junc- gains and losses. sociatedbelow the with reference the anticipation point are of framed losses toas belosses. larger As than the the earlier corresponding discussion sen empha- - point. ture, but then increases with the magnitude of the surplus. sationssized, associatedthere can bewith ambiguity the anticipation attached of togains. how According people set to reference prospect points,theory, and reference point anhow individual they frame frames the choices consequences in terms of gainsthe decisions and losses they relative make. to Notably,a reference people Benchmark used to measure point.can frame Cash theirflows decisions above the narrowly reference or point broadly, are framed which as in gains turn whilecan impact cash flows the deci - 1.5 MITIGATING PITFALLS gains and losses. belowsions thethey reference make. A point good are example framed asis losses.the earlier As the discussion earlier discussion about alternatives empha - I, narrow framing sized,J, K, thereand L. can People be ambiguity who engage attached in narrowto how framingpeople set and reference treat risks points, in isolationand Biases such as excessive optimism and overconfidence illustrate faulty judgments. Treating a risk in isolation. howare theyprone frame to choose the consequences I and L, whereas of the thosedecisions who they frame make. risks Notably, more broadly people are The framing effect associated with the concurrent choice problem (involving choices caninclined frame totheir choose decisions the superior narrowly combination or broadly, which of J and in turn K. can impact the deci - I, J, K, and L) illustrates faulty decision making. Faulty judgments and faulty de- sionsProspect they make. theory A goodexpanded example our understandingis the earlier discussion of how attitude about toalternatives risk depends I, on cisions are part of the normal human experience. Because by nature people are narrow framing J,circumstances. K, and L. People Before who prospect engage in theory, narrow there framing was a and tendency treat risks to treat in isolationrisk attitude imperfect, expect mistakes. At the same time, there is an expression about not let- Treating a risk in isolation. areas proneunivariate, to choose and toI andmeasure L, whereas it using those a question who frame such risks as the more following: broadly Imagine are ting the perfect be the enemy of the good, which in this case means not letting the inclinedthat you to face choose a choice the superior about whethercombination to stay of Jin and your K. current job, which involves impossibility of perfection stand in the way of seeking ways to mitigate mistakes. noProspect risk to yourtheory wealth, expanded or to our take understanding a new job, which of how features attitude ato risk. risk Thedepends risk onis that The most common behavioral terms for mitigating mistakes are “nudges,” circumstances.with a probability Before of prospect 50 percent, theory, your there future was wealth a tendency will toeither treat double risk attitude or be cut debiasing “debiasing,” and “cognitive repairs.” These vary from small-scale interventions asby univariate, x percent. and Think to measure about what it using the value a question of x wouldsuch as have the tofollowing: be so that Imagine you would A process for mitigating called nudges,19 to large-scale interventions such as shifts in organizational cul- thatbe indifferentyou face a choicebetween about the whethertwo options. to stay Notably in your the current variable job, 1/xwhich – 1 involves can be used bias. ture. Although discussion of large-scale interventions is mostly confined to Chapter noto risk measure to your a conceptwealth, orcalled to take the a “coefficientnew job, which of relativefeatures risk a risk. aversion.” The risk Theis that higher nudge 10, throughout this book readers will find tips about corporate nudges. These tips withthe valuea probability of x, the of more 50 percent, tolerance your a personfuture wealth shows willto bearing either doublerisk. or be cut A form of weak interven- are set out as thematic boxes bearing the title “Corporate Nudges,” and structured to by xFor percent. a more Think formal about discussion what the valueof prospect of x would theory, have see to beAdditional so that you Resources would to tion for helping people to identify the error or bias, briefly explain why it happens, describe how it happens, beChapter indifferent 1, which between is available the two online.options. Notably the variable 1/x – 1 can be used avoid psychological pitfalls and offer a nudge for doing something about it. and make better choices. to measure a concept called the “coefficient of relative risk aversion.” The higher The purposeful design of decision frames to facilitate effective decision making theAspiration value of x, thePoints more tolerance a person shows to bearing risk. is called “choice architecture.” Good choice architects understand that there are ThereFor a are more natural formal reference discussion points, of prospect such as theory,zero for see earnings Additional per Resourcesshare. Reference to numerous challenges in getting people to change their behavior. Chapterpoints can1, which also isbe available aspirational, online. such as the 52-week high for a stock price. SP/A The rest of this section describes some of the general issues of which good choice Aspirationtheory is a psychologically Points based theory of choice emphasizing that while gains and architects need to be cognizant. Many people get set in their ways and resist change, a losses can be defined relative to one reference point, an individual might also aspire tendency known as “status quo bias.” Many profess through their System 2 that they Thereto achieve are natural a different reference reference points, point, such whichas zero is for usually earnings higher. per share.This aspiration Reference point wish to change, but find that they lack self-control in that their System 1 is resistant to pointsseparates can alsosuccess be aspirational,from failure, such as opposed as the 52-week to gain fromhigh forloss. a17 stock price. SP/A going along. Many are excessively optimistic, but because not all are so for the same theoryLike is aprospect psychologically theory, SP/Abased theory theory hasof choice a psychophysics emphasizing structure that while for gains its treatment and reason, choice architects might need to take a nuanced approach. For some, the main SP/A theory lossesof probabilities. can be defined However, relative unlike to one prospectreference theory,point, an SP/A individual theory might interprets also aspire the men - reason for excessive optimism is emotional in that some people simply feel hopeful. For A psychological theory to achieve a different reference point, which is usually higher. This aspiration point tal associations in terms of emotions, most notably fear and17 hope. Fear leads to the others the main reasons have more to do with specific factors such as people perceiving of risk taking focused on separatesoverweighting success of from rare failure, unfavorable as opposed events, to whilegain from hope loss. leads to the overweighting of themselves as in control, engaging in wishful thinking, being familiar with the decision security to address fear, rareLike favorable prospect events. theory, According SP/A theory to hasSP/A a psychophysics theory, people structure act as if for their its incentivestreatment are task, or perceiving themselves as representative of successful people. Knowing the SP/Apotential theory to address hope, of probabilities. However, unlike prospect theory, SP/A theory interprets the men- and the need for success in to make choices by balancing the perceived probability of being successful against source of the excessive optimism can sometimes help choice architects devise ways A psychological theory respect to aspiration. tal“behavioral” associations expected in terms value,of emotions, of which most prospect notably theory fear and expected hope. Fearvalue leads is an to example. the to provide feedback to decision makers in order to help them be aware of their biases. of risk taking focused on overweighting of rare unfavorable events, while hope leads to the overweighting of security to address fear, potential to address hope, rare favorable events. According to SP/A theory, people act as if their incentives are and the need for success in to make choices by balancing the perceived probability of being successful against respect to aspiration. “behavioral” expected value, of which prospect theory expected value is an example.

she77208_ch01_001-031.indd 22 she77208_ch01_001-031.indd2/7/17 5:25 PM 23 2/7/17 5:25 PM

she77208_ch01_001-031.indd 22 2/7/17 5:25 PM 22 Chapter One Behavioral Foundations 23 in the probability of an extreme event increases with the likelihood with which One way of adapting these ideas from individual choice to firm choice is to consider the event occurs, but at a decreasing rate. The latter statement holds up to a March-Shapira firms as having two potential reference points, called focal points, one for survival and framework point, in that the mental sensations from anticipating successive decreases in the the other for aspiration. At any one time, only the first focal pointBehavioral will Foundations be operative, 23 A model with two reference probability of a highly likely event decreases with the probability of the event, (focal) points, one low depending on the extent to which a firm experiences financial distress along with the but again at a decreasing rate. (survival) and one high personalityOne way traits of of adapting its executives these ideas and fromboard individual members. choice According to firm to choice the theory, is to consider known Prospect theory explains loss aversion by treating the mental sensations as- (aspiration),March-Shapira with only as thefirms March-Shapira as having two potentialframework reference, a distressed points, calledfirm takes focal lowpoints, risk one but for increases survival that and framework 18 sociated with the anticipation of losses to be larger than the corresponding sen - one point operative at any riskthe as theother level for ofaspiration. distress declines.At any one Fortime, a non-distressedonly the first focal firm, point the theorywill be postulates operative, one time,A model and riskwith appetite two reference sations associated with the anticipation of gains. According to prospect theory, (focal) points, one low thatdepending when a firm on isthe operating extent to belowwhich aspiration,a firm experiences and therefore financial experiencing distress along a shortfall, with the reference point reference increasing with the absolute an individual frames choices in terms of gains and losses relative to a distance(survival) between and the one current high it takespersonality on relatively traits of high its executives risk. However, and board as themembers. shortfall According shrinks tothe the amount theory, ofknown risk Benchmark used to measure point. Cash flows above the reference point are framed as gains while cash flows situation(aspiration), and the reference with only declines.as the WhenMarch-Shapira the shortfall framework turns into, a distressedsurplus, risk firm falls takes discontinuously low risk but increases at the junc that- gains and losses. below the reference point are framed as losses. As the earlier discussion empha - point.one point operative at any ture,risk but as then the increaseslevel of distress with the declines. magnitude18 For ofa non-distressed the surplus. firm, the theory postulates sized, there can be ambiguity attached to how people set reference points, and one time, and risk appetite that when a firm is operating below aspiration, and therefore experiencing a shortfall, increasing with the absolute how they frame the consequences of the decisions they make. Notably, people distance between the current it takes on relatively high risk. However, as the shortfall shrinks the amount of risk can frame their decisions narrowly or broadly, which in turn can impact the deci - 1.5 situation MITIGATING and the reference PITFALLSdeclines. When the shortfall turns into a surplus, risk falls discontinuously at the junc- sions they make. A good example is the earlier discussion about alternatives I, point. ture, but then increases with the magnitude of the surplus. narrow framing J, K, and L. People who engage in narrow framing and treat risks in isolation Biases such as excessive optimism and overconfidence illustrate faulty judgments. Treating a risk in isolation. are prone to choose I and L, whereas those who frame risks more broadly are The framing effect associated with the concurrent choice problem (involving choices inclined to choose the superior combination of J and K. 1.5 MITIGATINGI, J, K,PITFALLS and L) illustrates faulty decision making. Faulty judgments and faulty de- Prospect theory expanded our understanding of how attitude to risk depends on cisions are part of the normal human experience. Because by nature people are circumstances. Before prospect theory, there was a tendency to treat risk attitude imperfect,Biases suchexpect as mistakes.excessive optimismAt the same and time, overconfidence there is an expressionillustrate faulty about judgments. not let- as univariate, and to measure it using a question such as the following: Imagine tingThe the framingperfect effectbe the associatedenemy of thewith good,the concurrent which in choicethis case problem means (involving not letting choices the that you face a choice about whether to stay in your current job, which involves impossibilityI, J, K, and of L) perfection illustrates stand faulty in decisionthe way making.of seeking Faulty ways judgments to mitigate and mistakes. faulty de- no risk to your wealth, or to take a new job, which features a risk. The risk is that Thecisions most are common part of the behavioral normal human terms experience. for mitigating Because mistakes by nature are people “nudges,” are with a probability of 50 percent, your future wealth will either double or be cut debiasing “debiasing,imperfect,” expectand “cognitive mistakes. repairs.”At the same These time, vary there from is an small-scale expression interventionsabout not let- by x percent. Think about what the value of x would have to be so that you would A process for mitigating calledting nudges, the perfect19 to be large-scale the enemy interventionsof the good, which such inas thisshifts case in meansorganizational not letting cul- the be indifferent between the two options. Notably the variable 1/x – 1 can be used bias. ture. impossibility Although discussion of perfection of large-scalestand in the interventions way of seeking is waysmostly to confinedmitigate mistakes.to Chapter to measure a concept called the “coefficient of relative risk aversion.” The higher nudge 10, throughoutThe most this common book readers behavioral will termsfind tips for about mitigating corporate mistakes nudges. are These “nudges,” tips the value of x, the more tolerance a person shows to bearing risk. A formdebiasing of weak interven- are “setdebiasing, out as thematic” and “cognitive boxes bearing repairs.” the title These “Corporate vary from Nudges,” small-scale and structuredinterventions to 19 For a more formal discussion of prospect theory, see Additional Resources to tion forA processhelping forpeople mitigating to identifycalled the nudges, error or bias,to large-scale briefly explain interventions why it suchhappens, as shifts describe in organizational how it happens, cul- bias. Chapter 1, which is available online. avoid psychological pitfalls andture. offer Although a nudge discussionfor doing something of large-scale about interventions it. is mostly confined to Chapter and make better choices. nudge The10, throughoutpurposeful thisdesign book of readersdecision will frames find tipsto facilitate about corporate effective nudges. decision These making tips Aspiration Points A form of weak interven- is calledare set “choiceout as thematic architecture.” boxes bearing Good the choice title “Corporate architects understandNudges,” and that structured there are to There are natural reference points, such as zero for earnings per share. Reference tion for helping people to numerousidentify challenges the error or in bias, getting briefly people explain to change why it their happens, behavior. describe how it happens, avoid psychological pitfalls and offer a nudge for doing something about it. points can also be aspirational, such as the 52-week high for a stock price. SP/A and make better choices. The rest of this section describes some of the general issues of which good choice theory is a psychologically based theory of choice emphasizing that while gains and architectsThe need purposeful to be cognizant. design of decisionMany people frames get to set facilitate in their effectiveways and decision resist change, making a losses can be defined relative to one reference point, an individual might also aspire tendencyis called known “choice as “status architecture.” quo bias.” Good Many choice profess architects through understand their System that 2 therethat they are to achieve a different reference point, which is usually higher. This aspiration point wishnumerous to change, challenges but find thatin getting they lack people self-control to change in their that behavior.their System 1 is resistant to separates success from failure, as opposed to gain from loss.17 going along.The rest Many of this are section excessively describes optimistic, some of butthe becausegeneral issuesnot all of are which so for good the choice same Like prospect theory, SP/A theory has a psychophysics structure for its treatment reason,architects choice need architects to be cognizant. might need Many to take people a nuanced get set in approach. their ways For and some, resist thechange, main a SP/A theory of probabilities. However, unlike prospect theory, SP/A theory interprets the men- reasontendency for excessive known optimismas “status isquo emotional bias.” Many in that profess some through people simplytheir System feel hopeful. 2 that theyFor A psychological theory tal associations in terms of emotions, most notably fear and hope. Fear leads to the otherswish the to main change, reasons but find have that more they to lack do with self-control specific in factors that their such System as people 1 is perceivingresistant to of risk taking focused on overweighting of rare unfavorable events, while hope leads to the overweighting of themselvesgoing along. as in Manycontrol, are engaging excessively in wishful optimistic, thinking, but because being familiarnot all are with so thefor thedecision same security to address fear, rare favorable events. According to SP/A theory, people act as if their incentives are task,reason, or perceiving choice architects themselves might as need representative to take a nuanced of successful approach. people. For some, Knowing the main the potential to address hope, reason for excessive optimism is emotional in that some people simply feel hopeful. For and the need for success in to make choices by balancing the perceived probability of being successful against source of the excessive optimism can sometimes help choice architects devise ways respect to aspiration. “behavioral” expected value, of which prospect theory expected value is an example. to provideothers the feedback main reasons to decision have moremakers to doin orderwith specific to help factorsthem be such aware as people of their perceiving biases. themselves as in control, engaging in wishful thinking, being familiar with the decision task, or perceiving themselves as representative of successful people. Knowing the source of the excessive optimism can sometimes help choice architects devise ways to provide feedback to decision makers in order to help them be aware of their biases.

she77208_ch01_001-031.indd 22 she77208_ch01_001-031.indd2/7/17 5:25 PM 23 2/7/17 5:25 PM

she77208_ch01_001-031.indd 23 2/7/17 5:25 PM 24 Chapter One Behavioral Foundations 25

Choice architects understand the challenges that arise because people sometimes systems and they make the choices without needing to confront the stronger make poor decisions as a result of their attention being drawn to factors of second- temptation of having to choose between fruit and pastry with both options in 24 Chapter One ary importance rather than factors of primary importance. The factors that attract front of them. and keep our attention relate to brain chemistry—for example, the specific brain re- In some cases, choice architects will find it better to leave well enough alone, gionChoice known architects as the understandposterior cingulate the challenges cortex. that In arisethis regard,because people people for sometimes whom these despite the pitfalls and imperfections. For example, take the Linda problem makeregions poor exhibit decisions higher as a densityresult of aretheir better attention at maintaining being drawn focus.to factors Therefore, of second- choice discussed in Section 1.3, and the fact that most people form judgments in this aryarchitects importance need rather to understand than factors that of interventionsprimary importance. that work The for factors some that will attract not work problem by relying on representativeness. Rather than relying on their subjec- andfor keepall. our attention relate to brain chemistry—for example, the specific brain re- tive impressions about the alternatives to be ranked in the Linda problem, people gionBrain known structure as the posterior is particularly cingulate germane cortex. In when this regard, it comes people to risk for whomtaking, these a criti - could instead do research to identify the relative frequencies associated with these regionscal aspect exhibit of corporate higher density finance. are betterNeuroscientists at maintaining have focus. identified Therefore, heightened choice ac - Bayes rule choices and then apply Bayes’ rule to arrive at a ranking. Bayes’ rule stipulates architectstivity in needthe brain to understand region known that interventions as the nucleus that workaccumbens, for some which will not is involvedwork The relationship linking that for two events D and F, the conditional probability of F given D, P(F|D), can forin all.positive emotional states such as feelings of “excitement,” as increasing the the probabilities of two be computed as P(F) × P(D|F)/P(D). Most people find it extremely difficult to events F and D expressed salienceBrain structure of potential is particularly gains and withgermane it subsequent when it comes risk taking.to risk taking,Analogously, a criti -they as P(F|D) = P(F) × P(D|F) carry out this procedure. If representativeness-based thinking leads them to do calhave aspect identified of corporate heightened finance. activity Neuroscientists in the brain have region identified known heightened as the anterior ac - /P(D), where P(F) is called a decent job of rank ordering most of the possibilities, then despite succumbing tivityinsula, in thewhich brain is regionassociated known with as negativethe nucleus emotional accumbens, states which such is as involved feelings of the prior and P(F|D) is to the conjunction fallacy, they might be better off relying on the heuristic rather in“anxiety,” positive emotional as increasing states thesuch salience as feelings of potentialof “excitement,” losses as and increasing subsequent the risk called the posterior. than poorly executing a Bayesian procedure. For them, less information might be salienceavoidance. of potential20 gains and with it subsequent risk taking. Analogously, they better than more. testosterone haveLevels identified of steroid heightened hormones activity such in as the testosterone brain region and known cortisol as thealso anterior impact risk To recap, choice architects can offer decision-making improvements by keep- A steroid hormone as- insula,taking. which Prior successis associated tends withto induce negative increases emotional in testosterone states such and as decreases feelings ofin cor- ing in mind a set of factors. These factors relate to psychological incentive issues sociated with risk taking, “anxiety,”tisol, both as of increasingwhich are associated the salience with of higher potential risk losses taking. and21 Elevated subsequent testosterone risk including the elements that activate neurological systems and attention, awareness aggression, and sexuality. avoidance.is associated20 with success and feeling like a winner, physically manifested in high of how people respond to different frames, an appreciation of defaults, provisions testosteronecortisol powerLevels poses of steroid such ashormones raised arms. such 22as The testosterone neurotransmitter and cortisol dopamine also impactis a critical risk in- for giving decision makers feedback, expecting mistakes instead of perfection, and A steroid hormonehormone as- gener- taking.gredient Prior in thesuccess neuro-reward tends to induce system. increases In this inregard, testosterone good choiceand decreases architects in cor-will ap- promoting frames for dealing with complex choices that take psychological phe- sociatedated during with prolonged risk taking, tisol,preciate both thatof which strings are of associated successes with that higherlead to risk heightened taking.21 levelsElevated of testosteronetestosterone and nomena into account. aggression,periods of stress. and sexuality. isdopamine associated often with resultsuccess in andextreme feeling overconfidence. like a winner, physically manifested in high cortisol powerPerhaps poses thesuch best as raisedknown arms. psychology22 The neurotransmitter experiment in self-control dopamine is is a the critical “marshmal- in- A steroid hormone gener- gredientlow test” in theinvestigating neuro-reward how system. children In thisare regard,successfully good choiceable to architects face temptation will ap- and ated during prolonged preciate that strings of successes that lead to heightened levels of testosterone 23and periods of stress. delay gratification in order to obtain two marshmallows instead of one. Brain Summary Traditional corporate finance textbooks focus more on what corporate managers dopaminechemistry often also result plays in a extremerole in self-control. overconfidence. In this case, specific regions of the part should do rather than what they actually do. The behavioral approach seeks to in- prefrontal cortex ofPerhaps the brain the associatedbest known with psychology executive experiment function, in knownself-control as the is theprefrontal “marshmal- cortex, vestigate what they actually do, why they do it, and to offer suggestions for how The front part of the hu- lowplay test” critical investigating roles, one how being children the ventromedial are successfully prefrontal able to cortexface temptation (vmPFC) andand the they might do a better job when it is possible. 23 man brain associated with delayother gratificationbeing the dorsolateral in order to prefrontal obtain two cortex marshmallows (DLpfc), so instead named of for one. where Brain they are This chapter introduces 10 psychological phenomena that are hardwired. Psy- higher-order thinking and chemistry also plays a role in self-control. In this case, specific regions of the part executive function. located in the brain. Activity within the vmPFC is associated with pleasure, such chologists suggest that we can think of our brains as having two systems: an in- prefrontal cortex ofas the the brain belief associated one is drinking with executive expensive function, wine. When known most as thepeople prefrontal face a decisioncortex, be- tuitive system they call System 1, which is fast, and a more deliberative system The front part of the hu- playtween critical giving roles, in to one temptation being the and ventromedial delaying gratification, prefrontal cortex their (vmPFC) vmPFC regionsand the will they call System 2, which is slow. System 1 activities tend to be automatic, and its man brain associated with other being the dorsolateral prefrontal cortex (DLpfc), so named for where they are higher-order thinking and become active. Researchers have found that those able to resist the temptation will calculations feel effortless, as when we are driving a car or running to catch a ball. executive function. locatedhave activated in the brain. and Activityengaged withinDLpfc the regions vmPFC as iswell, associated whereas with those pleasure, unable such to resist System 2 activities feel more like work, as in the preparation of cash flow forecasts asthe the temptation belief one willis drinking not.24 expensive wine. When most people face a decision be- for the purpose of undertaking discounted cash flow analysis. tweenFood giving choices in to temptation are often and good delaying ways gratification, to explain interventions. their vmPFC regions Suppose will that Like everyone else, managers make professional decisions using the processes becomeaccording active. to Researchers Systems 2, have a person found that entering those able a lunch to resist cafeteria the temptation professes will that of both System 1 and System 2. Some decisions are very sophisticated, and require have activated and engaged DLpfc regions as well, whereas those unable to resist she would be better 24off choosing fruit for dessert instead of a pastry. Choice significant System 2 activity. Much of professional education is aimed at enhancing thearchitects temptation have will learned not. that by structuring the choice so that the fruit is at - System 2 skills. As impressive as people can be, few—if any—are perfect. Each of tractivelyFood choices displayed are often early goodon, and ways the to pastry explain placed interventions. further along, Suppose more that people us, to some degree, is vulnerable to some or all of the 10 phenomena at the heart of accordingwill choose to fruit Systems over 2,pastry. a person The enteringattractive a fruit lunch display cafeteria activates professes their that reward this chapter. she would be better off choosing fruit for dessert instead of a pastry. Choice architects have learned that by structuring the choice so that the fruit is at - tractively displayed early on, and the pastry placed further along, more people will choose fruit over pastry. The attractive fruit display activates their reward

she77208_ch01_001-031.indd 24 she77208_ch01_001-031.indd2/7/17 5:25 PM 25 2/7/17 5:25 PM

she77208_ch01_001-031.indd 24 2/7/17 5:25 PM 24 Chapter One Behavioral Foundations 25

Choice architects understand the challenges that arise because people sometimes systems and they make the choices without needing to confront the stronger make poor decisions as a result of their attention being drawn to factors of second- temptation of having to choose between fruit and pastry with both options in ary importance rather than factors of primary importance. The factors that attract front of them. Behavioral Foundations 25 and keep our attention relate to brain chemistry—for example, the specific brain re- In some cases, choice architects will find it better to leave well enough alone, gion known as the posterior cingulate cortex. In this regard, people for whom these despitesystems the and pitfalls they andmake imperfections. the choices without For example, needing taketo confront the Linda the problemstronger regions exhibit higher density are better at maintaining focus. Therefore, choice discussedtemptation in Section of having 1.3, to and choose the factbetween that mostfruit andpeople pastry form with judgments both options in this in architects need to understand that interventions that work for some will not work problemfront ofby them. relying on representativeness. Rather than relying on their subjec- for all. tive impressionsIn some cases, about choice the alternatives architects will to befind ranked it better in theto leave Linda well problem, enough people alone, Brain structure is particularly germane when it comes to risk taking, a criti - coulddespite instead the do pitfalls research and to imperfections.identify the relative For example,frequencies take associated the Linda with problem these cal aspect of corporate finance. Neuroscientists have identified heightened ac - Bayes rule choicesdiscussed and then in Section apply Bayes’1.3, and rule the tofact arrive that mostat a ranking. people form Bayes’ judgments rule stipulates in this tivity in the brain region known as the nucleus accumbens, which is involved The relationship linking thatproblem for two eventsby relying D and on F,representativeness. the conditional probability Rather than of relyingF given on D, theirP(F|D), subjec can- in positive emotional states such as feelings of “excitement,” as increasing the the probabilities of two be computedtive impressions as P(F) about × P(D|F)/P(D). the alternatives Most to be people ranked find in the it extremelyLinda problem, difficult people to events F and D expressed salience of potential gains and with it subsequent risk taking. Analogously, they as P(F|D) = P(F) × P(D|F) carrycould out instead this procedure. do research If torepresentativeness-based identify the relative frequencies thinking associated leads them with to these do have identified heightened activity in the brain region known as the anterior /P(D),Bayes where ruleP(F) is called a decentchoices job and of thenrank applyordering Bayes’ most rule of tothe arrive possibilities, at a ranking. then Bayes’despite rule succumbing stipulates insula, which is associated with negative emotional states such as feelings of the priorThe and relationship P(F|D) is linking to thethat conjunction for two events fallacy, D and they F, the might conditional be better probability off relying of onF given the heuristic D, P(F|D), rather can “anxiety,” as increasing the salience of potential losses and subsequent risk calledthe the probabilities posterior. of two thanbe poorly computed executing as P(F) a Bayesian× P(D|F)/P(D). procedure. Most For people them, find less it informationextremely difficult might beto 20 events F and D expressed avoidance. as P(F|D) = P(F) × P(D|F) bettercarry than out more. this procedure. If representativeness-based thinking leads them to do testosterone Levels of steroid hormones such as testosterone and cortisol also impact risk /P(D), where P(F) is called Toa decent recap, job choice of rank architects ordering can most offer of decision-makingthe possibilities, thenimprovements despite succumbing by keep- A steroid hormone as- taking. Prior success tends to induce increases in testosterone and decreases in cor- the prior and P(F|D) is ing toin themind conjunction a set of factors. fallacy, These they might factors be relatebetter tooff psychological relying on the incentive heuristic issuesrather sociated with risk taking, tisol, both of which are associated with higher risk taking.21 Elevated testosterone called the posterior. includingthan poorly the elements executing that a Bayesian activate neurologicalprocedure. For systems them, lessand informationattention, awareness might be aggression, and sexuality. is associated with success and feeling like a winner, physically manifested in high of howbetter people than more.respond to different frames, an appreciation of defaults, provisions cortisol power poses such as raised arms.22 The neurotransmitter dopamine is a critical in- for givingTo recap, decision choice makers architects feedback, can expectingoffer decision-making mistakes instead improvements of perfection, by keep- and A steroid hormone gener- gredient in the neuro-reward system. In this regard, good choice architects will ap- promotinging in mind frames a set for of dealing factors. withThese complex factors relatechoices to psychologicalthat take psychological incentive issues phe- ated during prolonged preciate that strings of successes that lead to heightened levels of testosterone and nomenaincluding into account.the elements that activate neurological systems and attention, awareness periods of stress. dopamine often result in extreme overconfidence. of how people respond to different frames, an appreciation of defaults, provisions Perhaps the best known psychology experiment in self-control is the “marshmal- for giving decision makers feedback, expecting mistakes instead of perfection, and low test” investigating how children are successfully able to face temptation and promoting frames for dealing with complex choices that take psychological phe- delay gratification in order to obtain two marshmallows instead of one.23 Brain Summary Traditionalnomena intocorporate account. finance textbooks focus more on what corporate managers chemistry also plays a role in self-control. In this case, specific regions of the part should do rather than what they actually do. The behavioral approach seeks to in- prefrontal cortex of the brain associated with executive function, known as the prefrontal cortex, vestigate what they actually do, why they do it, and to offer suggestions for how The front part of the hu- play critical roles, one being the ventromedial prefrontal cortex (vmPFC) and the they might do a better job when it is possible. man brain associated with other being the dorsolateral prefrontal cortex (DLpfc), so named for where they are Summary ThisTraditional chapter corporate introduces finance 10 psychological textbooks focus phenomena more on thatwhat are corporate hardwired. managers Psy- higher-order thinking and should do rather than what they actually do. The behavioral approach seeks to in- executive function. located in the brain. Activity within the vmPFC is associated with pleasure, such chologists suggest that we can think of our brains as having two systems: an in- as the belief one is drinking expensive wine. When most people face a decision be- tuitivevestigate system what they they call actually System do, 1, whywhich they is dofast, it, andand ato moreoffer deliberativesuggestions forsystem how tween giving in to temptation and delaying gratification, their vmPFC regions will theythey call might System do 2,a betterwhich job is whenslow. itSystem is possible. 1 activities tend to be automatic, and its become active. Researchers have found that those able to resist the temptation will calculationsThis chapter feel effortless, introduces as 10when psychological we are driving phenomena a car or thatrunning are hardwired.to catch a ball.Psy- have activated and engaged DLpfc regions as well, whereas those unable to resist Systemchologists 2 activities suggest feel that more we likecan work,think ofas inour the brains preparation as having of cashtwo systems: flow forecasts an in- the temptation will not.24 for thetuitive purpose system of theyundertaking call System discounted 1, which cash is fast,flow and analysis. a more deliberative system Food choices are often good ways to explain interventions. Suppose that Likethey everyonecall System else, 2, which managers is slow. make System professional 1 activities decisions tend to beusing automatic, the processes and its according to Systems 2, a person entering a lunch cafeteria professes that of bothcalculations System feel1 and effortless, System 2.as Somewhen wedecisions are driving are very a car sophisticated, or running to catchand require a ball. she would be better off choosing fruit for dessert instead of a pastry. Choice significantSystem System2 activities 2 activity. feel more Much like ofwork, professional as in the preparationeducation is of aimed cash flow at enhancing forecasts architects have learned that by structuring the choice so that the fruit is at - Systemfor the 2 skills.purpose As of impressive undertaking as discountedpeople can cash be, few—ifflow analysis. any—are perfect. Each of tractively displayed early on, and the pastry placed further along, more people us, to someLike everyonedegree, is else, vulnerable managers to somemake orprofessional all of the 10decisions phenomena using at the the processes heart of will choose fruit over pastry. The attractive fruit display activates their reward thisof chapter. both System 1 and System 2. Some decisions are very sophisticated, and require significant System 2 activity. Much of professional education is aimed at enhancing System 2 skills. As impressive as people can be, few—if any—are perfect. Each of us, to some degree, is vulnerable to some or all of the 10 phenomena at the heart of this chapter.

she77208_ch01_001-031.indd 24 she77208_ch01_001-031.indd2/7/17 5:25 PM 25 2/7/17 5:25 PM

she77208_ch01_001-031.indd 25 2/7/17 5:25 PM 26 Chapter One Behavioral Foundations 27

The 10 phenomena comprise three categories: biases, heuristics, and framing Explore the Web www.nobelprize.org/mediaplayer/?id=531 effects. Biases and heuristics mostly pertain to imperfect judgments. Framing ef- Nobel prize lecture by Daniel Kahneman, 2002. 26 Chapter One fects pertain to choices among alternatives, and some framing effects are associated www.psych-it.com.au/Psychlopedia/article.asp?id=238 with imperfect decisions. In this regard, one of the major lessons to be learned from Discussing optimism bias. studyingThe 10 framingphenomena effects comprise is that three people’s categories: psychological biases, heuristics,attitude toward and framing risk taking www.psychologytoday.com/blog/the-art-thinking-clearly/201306/the effects.is often Biases circumstantial. and heuristics mostly pertain to imperfect judgments. Framing ef- -overconfidence-effect fectsIn pertain a sense, to choices the route among to alternatives, better decisions and some is to framing develop effects a greater are associated reliance on Discussing overconfidence bias. withSystem imperfect 2 for decisions.judgment Inand this decision regard, onemaking. of the However, major lessons keep to in be mind learned that from System www.npr.org/2016/06/24/483112809/www.npr.org/programs/ted-radio studying2 requires framing much effects more effortis that thanpeople’s System psychological 1, and people attitude typically toward lack risk thetaking mental -hour/483080945/nudge isresources often circumstantial. and perhaps the capabilities necessary for System 2 perfection. More- Explaining the concept of nudge. over,In a managers sense, the might route compute to better net decisions present isvalue, to develop but if net a greaterpresent reliancevalue and on their Systemintuition 2 for point judgment in different and decision directions, making. managers However, might keep be in moremind pronethat System to follow 2their requires intuition. much more effort than System 1, and people typically lack the mental Chapter 1. Imagine a decision task in which you are to choose between two alternatives that involve resourcesFiguring and out perhaps how to the make capabilities better decisions necessary typically for System requires 2 perfection. working Morewithin- the blindly drawing a single chip from one of two urns, labeled A and B respectively. Both over,system, managers meaning might the compute joint interaction net present of value, System but 1 if and net Systempresent value2. Working and their within Questions urns contain colored balls. The proportion of the different colors is described below. intuitionthe system point requires in different knowledge directions, of choicemanagers architecture. might be more In some prone circumstances, to follow Also described below are the payoffs from drawing a ball of a specific color from each of the two urns. Your task is to decide from which urn you would draw a ball, urn A or theirpeople intuition. can make better decisions by reframing their decision tasks. Remember the urn B? adage:Figuring the outway how to solveto make an better important decisions problem typically is to requires ask the workingright question. within the Some - system,times others meaning will the be joint able interactionto recommend of System architectural 1 and System changes 2. thatWorking nudge within “us” into thebetter system decisions. requires Other knowledge times, “we” of choice might architecture. be able to ask In themselves some circumstances, questions that Urn A Urn B peoplepromote can self-nudging. make better decisions by reframing their decision tasks. Remember the Probability Outcome Probability Outcome adage: the way to solve an important problem is to ask the right question. Some - Yellow 2% –$15 2% –$15 times others will be able to recommend architectural changes that nudge “us” into Blue 1% –$15 1% –$10 better decisions. Other times, “we” might be able to ask themselves questions that Green 1% $30 1% $45 promote self-nudging. Red 6% $45 6% $45 Additional Barberis, N. C., “Thirty Years of Prospect Theory in Economics: A Review and Assess- White 90% $0 90% $0 Journal of Economic Perspectives Behavioral ment,” 27, no. 1 (Winter 2013), pp. 173–196. Gilovich, T., D. Griffin, and D. Kahneman, Heuristics and Biases: The Psychology of 2. Imagine yourself at a conference center, where there are two groups meeting in adjoin- Readings Intuitive Judgment (Cambridge, UK: Cambridge University Press, 2002). ing rooms, one a group of lawyers and the other a group of engineers. In fact, the room Additional Barberis,Kahneman, N. C., D. “Thirty and A. YearsTversky of Prospect“Prospect Theory Theory: in Economics:An Analysis A of Review Decision and MakingAssess- under with lawyers contains 30 lawyers and the room with engineers contains 70 engineers. Journal of Economic Perspectives Behavioral ment,”Risk,” Econometrica 5, no. 2 (1979), pp.27, 263–291. no. 1 (Winter 2013), pp. 173–196. Someone named Jack is attending a meeting in one of these two rooms. Here is what Heuristics and Biases: The Psychology of we are told about Jack: he is a 45-year-old man who is married and has four children. Gilovich,Kahneman, T., D.D., Griffin, Thinking, and FastD. Kahneman, and Slow (New York: Farrar, Straus, and Giroux, 2011). Readings Intuitive Judgment (Cambridge, UK: Cambridge University Press, 2002). He is generally conservative, careful, and ambitious. He shows no interest in political Kahneman, D. and A. Tversky “Prospect Theory: An Analysis of Decision Making under and social issues and spends most of his free time on his many hobbies, which include Risk,” Econometrica 5, no. 2 (1979), pp. 263–291. home carpentry, sailing, and mathematical puzzles. What probability would you assign Key Terms affect, 14 debiasing, 23 overconfidence, 6 to Jack being an engineer? Kahneman, D., Thinking,14 Fast and Slow (New York: Farrar,6 Straus, and Giroux, 2011).24 affect heuristic, excessive optimism, prefrontal cortex, 3. Suppose that you are offered the opportunity to accept a risk involving the toss of a fair 13 17 21 anchoring and adjustment, fourfold pattern, prospect theory, coin, in which you will win $450 if heads comes up and lose $450 if tails comes up. availability, 12 frame, 3 psychophysics, 21 a. Would you accept this opportunity or reject it? Key Terms affect,aversion 14 to a sure loss, 20 debiasing,framing 23 effect, 16 overconfidence,reference point, 6 22 affectBayes heuristic, rule, 25 14 excessiveheuristic, optimism, 3 6 prefrontalrepresentativeness, cortex, 24 11 b. If the stake size was changed to $225 from $450, would you accept the opportunity anchoringbetter than and average adjustment, 13 fourfoldillusion pattern, of control, 17 9 prospectSP/A theory, theory, 21 22 or reject it? availability,effect, 8 12 frame,loss 3aversion, 17 psychophysics,System 1, 3 21 4. Consider a trivia test consisting of 10 questions for you to answer from memory alone. aversionbias, 4 to a sure loss, 20 framingMarch-Shapira effect, 16 frame work, referenceSystem point, 2, 4 22 In addition to giving your best guess, consider a range: a low guess and a high guess so Bayesconfirmation rule, 25 bias, 8 heuristic,23 3 representativeness,testosterone, 24 11 that you feel 90 percent confident that the right answer will lie between your low guess betterconjunction than average fallacy, 11 illusionnarrow of framing,control, 9 22 SP/A theory, 22 and your high guess. Try not to make the range between your low guess and high guess cortisol,effect, 8 24 lossnudge, aversion, 23 17 System 1, 3 too narrow. Otherwise, you will appear overconfident. At the same time, try not to make bias, 4 March-Shapira frame work, System 2, 4 confirmation bias, 8 23 testosterone, 24 conjunction fallacy, 11 narrow framing, 22 cortisol, 24 nudge, 23 www.mhhe.com/shefrin2e she77208_ch01_001-031.indd 26 she77208_ch01_001-031.indd2/7/17 5:25 PM 27 2/7/17 5:25 PM

she77208_ch01_001-031.indd 26 2/7/17 5:25 PM www.mhhe.com/shefrin2e 2/7/17 5:25 PM 2/7/17 5:25 PM 27 27 $0 $0 $45 $45 –$15 –$10 $45 $45 –$15 –$10 Urn B Behavioral Foundations Urn B Behavioral Foundations 1% 1% 2% 6% 1% 1% 90% 2% 6% 90% $0 $45 $30 –$15 –$15 $0 $45 $30 –$15 –$15 Urn A Urn A 1% 1% 2% 6% 90% 1% 1% 2% 6% 90% Probability Outcome Probability Outcome Probability Outcome Probability Outcome

Yellow Blue Green Red White If the stake size was changed to $225 from $450, would you accept the opportunity or reject it?

Yellow Blue Green Red White Imagine yourself at a conference center, where there are two groups meeting in adjoin- in meeting groups two are there where center, conference a at yourself Imagine room the fact, In engineers. of group a other the and lawyers of group a one rooms, ing with lawyers contains 30 lawyers and the room with engineers contains 70 engineers. Someone named Jack is attending a meeting in one of these two rooms. Here is what we are told about Jack: he is a 45-year-old man who is married and has four children. He is generally conservative, careful, and ambitious. He shows no interest in political include which hobbies, many his on time free his of most spends and issues social and assign you would probability What puzzles. mathematical and sailing, carpentry, home to Jack being an engineer? Suppose that you are offered the opportunity to accept a risk involving the toss of a fair a of toss the involving risk a accept to opportunity the offered are you that Suppose coin, in which you will win $450 if heads comes up and lose $450 if tails comes up. Imagine a decision task in which you are to choose between two alternatives that involve that alternatives two between choose to are you which in task decision a Imagine urns, labeled A and B respectively. Both blindly drawing a single chip from one of two urns contain colored balls. The proportion of the different colors is described Also described below below. are the payoffs from drawing a ball of a specific color from each of the two urns. Your task is to decide from which urn you would draw a ball, urn A or urn B? Consider a trivia test consisting of 10 questions for you to answer from memory alone. so guess high a and guess low a range: a consider guess, best your giving to addition In that you feel 90 percent confident that the right answer will lie between your low guess guess high and guess low your between range the make to not Try guess. high your and make to not try time, same the At overconfident. appear will you Otherwise, narrow. too If the stake size was changed to $225 from $450, would you accept the opportunity or reject it? a. Would you accept this opportunity or reject it? b. 2. Imagine yourself at a conference center, where there are two groups meeting in adjoin- in meeting groups two are there where center, conference a at yourself Imagine room the fact, In engineers. of group a other the and lawyers of group a one rooms, ing with lawyers contains 30 lawyers and the room with engineers contains 70 engineers. Someone named Jack is attending a meeting in one of these two rooms. Here is what we are told about Jack: he is a 45-year-old man who is married and has four children. He is generally conservative, careful, and ambitious. He shows no interest in political include which hobbies, many his on time free his of most spends and issues social and assign you would probability What puzzles. mathematical and sailing, carpentry, home to Jack being an engineer? Suppose that you are offered the opportunity to accept a risk involving the toss of a fair a of toss the involving risk a accept to opportunity the offered are you that Suppose coin, in which you will win $450 if heads comes up and lose $450 if tails comes up. Consider a trivia test consisting of 10 questions for you to answer from memory alone. so guess high a and guess low a range: a consider guess, best your giving to addition In that you feel 90 percent confident that the right answer will lie between your low guess guess high and guess low your between range the make to not Try guess. high your and make to not try time, same the At overconfident. appear will you Otherwise, narrow. too Imagine a decision task in which you are to choose between two alternatives that involve that alternatives two between choose to are you which in task decision a Imagine and B respectively. Both chip from one of two urns, labeled A blindly drawing a single urns contain colored balls. The proportion of the different colors is described below. Also described below are the payoffs from drawing a ball of a specific color from each of the two urns. Your task is to decide from which urn you would draw a ball, urn A or urn B? 3. 1. 4. www.nobelprize.org/mediaplayer/?id=531 Daniel Kahneman, 2002. Nobel prize lecture by www.psych-it.com.au/Psychlopedia/article.asp?id=238 bias. Discussing optimism www.psychologytoday.com/blog/the-art-thinking-clearly/201306/the -overconfidence-effect bias. Discussing overconfidence www.npr.org/2016/06/24/483112809/www.npr.org/programs/ted-radio -hour/483080945/nudge of nudge. Explaining the concept a. Would you accept this opportunity or reject it? b. 2. 3. 4. 1. www.nobelprize.org/mediaplayer/?id=531 2002. lecture by Daniel Kahneman, Nobel prize www.psych-it.com.au/Psychlopedia/article.asp?id=238 bias. Discussing optimism www.psychologytoday.com/blog/the-art-thinking-clearly/201306/the -overconfidence-effect bias. Discussing overconfidence www.npr.org/2016/06/24/483112809/www.npr.org/programs/ted-radio -hour/483080945/nudge of nudge. Explaining the concept

Chapter Questions Explore the Web

Chapter Questions Explore the Web Explore

she77208_ch01_001-031.indd 27

she77208_ch01_001-031.indd 272/7/17 5:25 PM - - overconfidence, 6 prefrontal cortex, 24 prospect theory, 21 psychophysics, 21 reference point, 22 representativeness, 11 SP/A theory, 22 System 1, 3 System 2, 4 testosterone, 24 23 debiasing, 23 excessive optimism, 6 fourfold pattern, 17 frame, 3 framing effect, 16 heuristic, 3 illusion of control, 9 loss aversion, 17 work, March-Shapira frame narrow framing, 22 nudge, 23 The 10 phenomena comprise three categories: biases, heuristics, and framing In a sense, the route to better decisions is to develop a greater reliance on the within working requires typically decisions better make to how out Figuring effect, 8 Barberis, N. C., “Thirty Years of Prospect Theory in Economics: A Review and Assess- Barberis, N. C., “Thirty Years of Prospect 27, no. 1 (Winter 2013), pp. 173–196. ment,” Journal of Economic Perspectives Heuristics and Biases: The Psychology of Gilovich, T., D. Griffin, and D. Kahneman, 2002). Intuitive Judgment (Cambridge, UK: Cambridge University Press, Theory: An Analysis of Decision Making under Kahneman, D. and A. Tversky “Prospect Risk,” Econometrica 5, no. 2 (1979), pp. 263–291. Kahneman, D., Thinking, Fast and Slow (New York: Farrar, Straus, and Giroux, 2011). affect, 14 affect heuristic, 14 anchoring and adjustment, 13 availability, 12 aversion to a sure loss, 20 Bayes rule, 25 better than average bias, 4 confirmation bias, 8 conjunction fallacy, 11 cortisol, 24 effects. effects. Biases and heuristics mostly pertain to imperfect judgments. Framing associated ef- are effects framing some and alternatives, among choices pertain to fects from learned be to lessons major the of one regard, this In decisions. imperfect with studying framing effects is that psychological people’s attitude toward risk taking is often circumstantial. in keep mind that System judgment and decision 2 making. for However, System 2 requires much more effort than System 1, and people typically lack the mental resources and perhaps the capabilities necessary for System 2 perfection. More their and value present net if but value, present net compute might managers over, intuition point in different directions, managers might be more prone to follow their intuition. meaning system, the joint within interaction of 1 System and 2. System Working the system requires knowledge of the Remember tasks. choice decision their reframing architecture. by decisions In better make can some people circumstances, an to importantadage: thesolve way problem is to ask the right Some question. into “us” nudge that changes architectural recommend to able be will others times that questions themselves ask to able be might “we” times, Other decisions. better self-nudging. promote Additional Behavioral Readings Key Terms 26 Chapter One she77208_ch01_001-031.indd 26 she77208_ch01_001-031.indd 28 she77208_ch01_001-031.indd 28 www.mhhe.com/shefrin2e ChapterOne 28 ChapterOne 28 Considerthefollowingthreequestionsaboutbookbag. 5. 6. Considerthefollowingthreequestionsaboutbookbag. 5. 6. guess. . Inmiles,howlongistheDanubeRiver? h. HowmanybookswereintheU.S.LibraryofCongressatendyear2000? g. InwhatyeardidLeonardodaVincipainttheMonaLisa? b. . Whatistheairdistance,inmiles,fromParis,FrancetoSydney,Australia? d. random bymeansofacointoss. and 700 red chips. You cannot see inside any of the bags. One of the bags is selected at bags chips black 300 contain Forty-five bags 55 other chips. The chips. red 300 poker and chips black 1,000 700 contain contains which of each bags, book 100 Imagine correct answersyouprovidedoesnotliebetweenyourlowguessandhighguess. underconfident. If you are well-calibrated, you should expect that only one out of the 10 appear you make will This wide. too guess high and guess low your between range the their high school GPA scores. The three high school GPA scores are 2.2, 3.0, and 3.0, 2.2, are scores GPA school high three The scores. GPA school high their you to predict the college GPA scores of three graduating students, based solely on GPA ofthosesamestudentswas3.08(standarddeviation0.40). Thisquestionasks collegemean The 0.36). was deviation (standard 3.44 was graduated and freshmen as entered who students of GPA school high mean the Historically, . in located university a at undergraduates for data some are Below 4. and 0 between liescores GPA levels. GPA school high their upon based students graduating some of(GPA) average point grade the predict to attempting is university a that Suppose . c. In addition to giving your best probability estimate in question b, consider a range: a bag. selected the from replacement, with drawn, are chips 12 that imagine b. Now pre- contains bag selected the that event the to assign you would probability a. What . Howmanybonesareinthehumanbody? e. Howmanyindependentcountrieswerethereattheendofyear2000? c. Whatistheaverageweightofanadultbluewhale,inpounds? a. guess. . HowmanycombatantswerekilledinWorldWarI? f. . HowmanytransistorsareinIntel’sPentiumIIIprocessor? j. Inmilesperhour,howfastdoestheEarthspinatequator? i. . Inmiles,howlongistheDanubeRiver? h. HowmanybookswereintheU.S.LibraryofCongressatendyear2000? g. InwhatyeardidLeonardodaVincipainttheMonaLisa? b. . Whatistheairdistance,inmiles,fromParis,FrancetoSydney,Australia? d. random bymeansofacointoss. and 700 red chips. You cannot see inside any of the bags. One of the bags is selected at chips bags black 300 contain Forty-five bags 55 other The chips. chips. red 300 poker and chips black 700 1,000 contain contains which of each bags, book 100 Imagine correct answersyouprovidedoesnotliebetweenyourlowguessandhighguess. underconfident. If you are well-calibrated, you should expect that only one out of the 10 appear you make will This wide. too guess high and guess low your between range the their high school GPA scores. The three high school GPA scores are 2.2, 3.0, and 3.0, 2.2, are scores GPA school high three The scores. GPA school high their you to predict the college GPA scores of three graduating students, based solely on GPA ofthosesamestudentswas3.08(standarddeviation0.40). Thisquestionasks collegemean The 0.36). was deviation (standard 3.44 was graduated and freshmen as entered who students of GPA school high mean the Historically, California. in located university a at undergraduates for data some are Below 4. and 0 between liescores GPA levels. GPA school high their upon based students graduating some of(GPA) average point grade the predict to attempting is university a that Suppose . c. In addition to giving your best probability estimate in question b, consider a range: a bag. selected the from replacement, with drawn, are chips 12 that imagine b. Now pre- contains bag selected the that event the to assign you would probability a. What . Howmanybonesareinthehumanbody? e. Howmanyindependentcountrieswerethereattheendofyear2000? c. Whatistheaverageweightofanadultbluewhale,inpounds? a. . HowmanycombatantswerekilledinWorldWarI? f. After each question, write down three numbers, your best guess, low guess, and high and guess, low guess, best your numbers, three down write question, each After . HowmanytransistorsareinIntel’sPentiumIIIprocessor? j. Inmilesperhour,howfastdoestheEarthspinatequator? i. range betweenyourlowestimateandhighonetime inten. the outside lie to probability true the expect should you well-calibrated, are you If low estimate and high estimate too wide. This will make you appear underconfident. will appear overconfident. At the same time, try not to make the range between your you Otherwise, narrow. too estimate high and estimate low your between range the make to not Try estimate. high your and estimate low your between lie will answer right the that confident percent 90 feel you that so estimate high a and estimate low dominantly blackchips? predominantly blackchips?Ifso,whatnewprobabilitywouldyouassign? contains bag selected the that probability your revise to chips of drawing the about information new the use you Would reds. 4 and blacks 8 produce draws 12 These After each question, write down three numbers, your best guess, low guess, and high and guess, low guess, best your numbers, three down write question, each After range betweenyourlowestimateandhighonetime inten. the outside lie to probability true the expect should you well-calibrated, are you If low estimate and high estimate too wide. This will make you appear underconfident. will appear overconfident. At the same time, try not to make the range between your you Otherwise, narrow. too estimate high and estimate low your between range the make to not Try estimate. high your and estimate low your between lie will answer right the that confident percent 90 feel you that so estimate high a and estimate low dominantly blackchips? predominantly blackchips?Ifso,whatnewprobabilitywouldyouassign? contains bag selected the that probability your revise to chips of drawing the about information new the use you Would reds. 4 and blacks 8 produce draws 12 These 25 25 2/7/17 5:25PM she77208_ch01_001-031.indd 29 2/7/17 5:25PM

Inconnectionwithchapterquestion2,answerthefollowingquestions. 8. 10. 7. 9.

color from each of the two urns. Your task is to decide from which urn you would you urn draw aball,urnAorB. which from decide to is task Your urns. two the of each from color scribed below. Also described below are the payoffs from drawing a ball of a specific de is colors different the of proportion The balls. colored contain urns Both tively. that alternatives respec B two and A labeled urns, two between of one from chip single a choose drawing blindly involve to are you which in task decision a Imagine researcher in a park with a lot of pedestrians invited passersby to examine an un an examine to passersby invited pedestrians of lot a with park a in researcher the pingpongballshadsamenumber. bag had a number between 1 and 100 written on it. However, this was not true, as all from a bag, keeping their eyes closed. The researcher indicated that each ball in the ally came up with their estimates, the researcher asked them to pull a ping pong ball actu participants Before paid. had researcher the price the estimate and shop, wine opened bottle of wine, which the researcher stated he had just purchased at a nearby program BBC the of episode An this question,donotchangeanyofyourpreviousanswers. directly relating to the binomial distribution, and if so which one or ones? In answering tion in finance? Do any of the problems in the problem set for this chapter strike you as the probability of drawing a success is success a drawing of probability the xe hs fnto, IO.IT fr h bnma poaiiy itiuin This distribution. probability drawing of probability the binomial computes function the for BINOM.DIST, function, a has Excel explanation ofyourthoughtprocess. brief a with along students, three the of each for predictions your down Write 3.8. . Didyouaskthesequestionsofyourselfwhenansweringquestion2,yesorno? d. Inyouropinion,whatproportionoflawyersmatchJack’sdescription? c. Inyouropinion,whatproportionofengineersmatchJack’sdescription? b. a. phenomenon mightexplainthisbehaviorpattern. second group guessed much higher, between $40 and $100. Discuss what the psychological while $30, and $10 between be to cost the estimated group first The price. wine’s the of estimate her or his for participant each asked researcher the ball, the on number the seeing After 65. number the with ball pong ping a drew group second the while 10, groups of participants. The first group of people drew a ping pong ball with the number Yellow

Green Red White tains 30engineers? con- engineers with room the and lawyers 70 has lawyers with room the that stated instead question the if engineer an being Jack to assign you would probability What h eprmn ws odce tie o to ifrn dy, ih w different two with days, different two on twice, conducted was experiment The Probability 90% 6% 3% 1% Urn A Outcome featured the following experiment. A experiment. following the featured Horizons –$15 $30 $45 $0 Do you see a role for the binomial distribu- binomial the for role a see you Do p. successes in successes x Probability Probability 90% 2% 7% 1% independent trials when trials independent n Behavioral Foundations Urn B Outcome $45 –$10 –$15 $0 29 - - - - 2/7/17 5:25PM www.mhhe.com/shefrin2e 2/7/17 5:25 PM 2/7/17 5:25 PM ------29 29 $0 $0 –$15 –$10 $45 –$15 –$10 $45 Outcome Outcome Urn B Urn B Behavioral Foundations n independent trials when Behavioral Foundations n independent trials when 1% 7% 2% 90% 1% 7% 2% 90% Probability Probability x successes in x successes in p. Do you see a role for the binomial distribu- $0 $45 p. Do you see a role for the binomial distribu- $30 –$15 $0 Horizons featured the following experiment. A $45 $30 Outcome –$15 Horizons featured the following experiment. A Outcome Urn A Urn A 1% 3% 6% 90% 1% 3% 6% 90% Probability Probability The experiment was conducted twice, on two different days, with two different What probability would you assign to Jack being an engineer if the question instead stated that the room with lawyers has 70 lawyers and the room with engineers con- tains 30 engineers? Green Red White

Yellow groups of participants. The first group of people drew a ping pong ball with the number the with ball pong ping a drew people of group first The participants. of groups 10, while the second group drew a ping pong ball with the number 65. After seeing the number on the ball, the researcher asked each participant for his or her estimate of the wine’s price. The first group estimated the cost to be between $10 and $30, while psychological what the Discuss $100. and $40 between higher, much guessed group second phenomenon might explain this behavior pattern. a. b. In your opinion, what proportion of engineers match Jack’s description? c. In your opinion, what proportion of lawyers match Jack’s description? d. or no? Did you ask these questions of yourself when answering question 2, yes 3.8. Write down your predictions for each of the three students, along with a brief process. explanation of your thought Excel has a function, BINOM.DIST, for the function computes the binomial probability of drawing probability distribution. This the probability of drawing a success is tion in finance? Do any of the problems in the problem set for this chapter strike you as you strike chapter this for set problem the in problems the of any Do finance? in tion answering In ones? or one which so if and distribution, binomial the to relating directly this question, do not change any of your previous answers. An episode of the BBC program researcher in a park with a lot of pedestrians invited passersby to examine nearby a at an purchased just had un he stated researcher the which wine, of bottle opened wine shop, and estimate the price the researcher had paid. Before participants actu ball pong ping a pull to them asked researcher the estimates, their with up came ally the in ball each that indicated researcher The closed. eyes their keeping bag, a from all as true, not was this However, it. on written 100 and 1 between number a had bag the same number. the ping pong balls had tively. Both urns contain colored balls. The proportion of the different colors is de specific a of ball a drawing from payoffs the are below described Also below. scribed color from each of the two urns. Your task is to decide from which draw a ball, urn A or urn B. urn you would Imagine a decision task in which you are to involve blindly drawing a single choose chip from one of two between urns, labeled A and B two respec alternatives that The experiment was conducted twice, on two different days, with two different What probability would you assign to Jack being an engineer if the question instead stated that the room with lawyers has 70 lawyers and the room with engineers con- tains 30 engineers? White Red Green

Yellow

9. 7. groups of participants. The first group of people drew a ping pong ball with the number the with ball pong ping a drew people of group first The participants. of groups 10, while the second group drew a ping pong ball with the number 65. After seeing the number on the ball, the researcher asked each participant for his or her estimate of the wine’s price. The first group estimated the cost to be between $10 and $30, while psychological the what Discuss $100. and $40 between higher, much guessed group second pattern. phenomenon might explain this behavior a. b. description? In your opinion, what proportion of engineers match Jack’s c. In your opinion, what proportion of lawyers match Jack’s description? d. 2, yes or no? Did you ask these questions of yourself when answering question 3.8. Write down your predictions for each of the three students, along with a brief of your thought process. explanation Excel has a function, BINOM.DIST, for the function computes binomial the probability of drawing probability distribution. This the probability of drawing a success is tion in finance? Do any of the problems in the problem set for this chapter strike you as you strike chapter this for set problem the in problems the of any Do finance? in tion answering In ones? or one which so if and distribution, binomial the to relating directly this question, do not change any of your previous answers. opened bottle of wine, which the researcher stated he had just purchased at a nearby nearby a at purchased just had he stated researcher the which wine, of bottle opened wine shop, and estimate the price the researcher had paid. Before participants actu ball pong ping a pull to them asked researcher the estimates, their with up came ally the in ball each that indicated researcher The closed. eyes their keeping bag, a from all as true, not was this However, it. on written 100 and 1 between number a had bag the same number. the ping pong balls had researcher in a park with a lot of pedestrians invited passersby to examine an un An episode of the BBC program Imagine a decision task in which you are to involve blindly drawing choose a single chip from one of between two urns, labeled A and two B respec alternatives that tively. Both urns contain colored balls. The proportion of the different colors is de specific a of ball a drawing from payoffs the are below described Also below. scribed color from each of the two urns. Your task is to decide from which draw a ball, urn A or urn B. urn you would 10.

8. In connection with chapter question 2, answer the following questions.

9. 7. 10. 8. question 2, answer the following questions. In connection with chapter

she77208_ch01_001-031.indd 29 2/7/17 5:25 PM she77208_ch01_001-031.indd 29 25 These 12 draws produce 8 blacks and 4 reds. Would you use the new information about the drawing of chips to revise your probability that the selected bag contains probability would you assign? predominantly black chips? If so, what new dominantly black chips? low estimate and a high estimate so that you feel 90 percent confident that the right answer will lie between your low estimate and your high estimate. Try not to make the range between your low estimate and high estimate too narrow. Otherwise, you your between range the make to not try time, same the At overconfident. appear will underconfident. appear you make will This wide. too estimate high and estimate low If you are well-calibrated, you should expect the true probability to lie outside the high estimate one time in ten. range between your low estimate and your After each question, write down three numbers, your best guess, low guess, and high i. In miles per hour, how fast does the Earth spin at the equator? j. How many transistors are in Intel’s Pentium III processor? f. War I? How many combatants were killed in World a. blue whale, in pounds? What is the average weight of an adult c. there at the end of the year 2000? How many independent countries were e. How many bones are in the human body? a. What probability would you assign to the event that the selected bag contains pre- b. Now imagine that 12 chips are drawn, with replacement, from the selected bag. a range: a consider b, question in estimate probability best your giving to addition c. In Suppose that a university is attempting to predict the grade point average (GPA) of Suppose that a university is attempting to predict the grade point average (GPA) of some graduating students based upon their high school GPA levels. GPA scores lie between 0 and 4. Below are some data for undergraduates at a university located in California. Historically, the mean high school GPA of students who entered as freshmen and graduated was 3.44 (standard deviation was 0.36). The mean college GPA of those same students was 3.08 (standard deviation 0.40). This question asks on solely based students, graduating three of scores GPA college the predict to you their high school GPA scores. The three high school GPA scores are 2.2, 3.0, and the range between your low guess and high guess too wide. This will make you appear 10 the of out one only that expect should you well-calibrated, are you If underconfident. high guess. low guess and your not lie between your you provide does correct answers Imagine 100 book bags, each of which contains contain 1,000 700 black chips and poker 300 red chips. chips. The other 55 bags Forty-five contain 300 black bags chips at selected is bags the of One bags. the of any inside see cannot You chips. red 700 and random by means of a coin toss. d. Paris, France to Sydney, Australia? What is the air distance, in miles, from b. the Mona Lisa? In what year did Leonardo da Vinci paint g. of Congress at the end of the year 2000? How many books were in the U.S. Library h. In miles, how long is the Danube River? guess. 6. 5. the book bag. Consider the following three questions about 28 Chapter One she77208_ch01_001-031.indd 28 she77208_ch01_001-031.indd 30 she77208_ch01_001-031.indd 30 www.mhhe.com/shefrin2e sule summary of the main points in coverage by the whereby $3.1 billion rose to $5.25 billion. Below is a cap- costoverrun a and canal, faultsthe earthquake beneath of identification the narrow,too was that design lock new a from error formargin insufficient tugboats, stable less to ships guiding for locomotives stable from shift a canal, the in ships large support to water insufficient concrete, between 2009 and 2016. new the ship passedthrough locks inJune2016. container first the and deadline, 2014 August the missed However,canal. original the consortium of anniversarythe set a goal of completing the project in time for the 100-year experienced engineering and construction in firms the world. most the of one Bechtel, included Notably,bidders other the billion, wasnext $3.1 lowestapproximately lower the billion bid. than $1 bid, Their Italy. and Belgium, Panama, from firms included members consortium’s The Vallehermoso. Sacyr firm Spanish the by led was that Canal el por Unidos soliciting bidsfor project. the (RFP), proposal for request a out put they 2007 In ships. container larger the handle would that locksnew build to canal would become obsolete, the ACP made the decision traverseto waterway.big too the its fear of that out Partly were ships container largest the where point the grew,to through traffic the Panama Canal. Notably, ship sizes also volumeof the it greatly,grewworld with shipping of and end of1999. the at did they This waterway. the of command assuming (ACP) Authority Canal Panama the in resulted and Panama States United the between treaty a but century, twenti- eth the of remainder the for waterway States the United operated The 1914. August in completed was and years 10 took project The Canal. Panama the built States In one of the world’s greatest engineering feats, the United Panama Canal Minicase ChapterOne 30 sule summary of the main points in coverage by the whereby $3.1 billion rose to $5.25 billion. Below is a cap- costoverrun a and canal, faultsthe earthquake beneath of identification the narrow,too was that design lock new a from error formargin insufficient tugboats, stable less to ships guiding for locomotives stable from shift a canal, the in ships large support to water insufficient concrete, ship passed through the new the ship passedthrough locks inJune2016. container first the and deadline, 2014 August the missed However,canal. original the consortium of anniversarythe set a goal of completing the project in time for the 100-year experienced engineering and construction in firms the world. most the of one Bechtel, included Notably,bidders other the billion, wasnextlowestapproximately lower the billion bid. $3.1 than $1 bid, Their Italy. and Belgium, Panama, from firms included members consortium’s The Vallehermoso. Sacyr firm Spanish the by led was that Canal el por Unidos soliciting bidsfor project. the (RFP), proposal for request a out put they 2007 In ships. container larger the handle would that locksnew build to canal would become obsolete, the ACP made the decision traverseto waterway.big too the its fear of that out Partly were ships container largest the where point the grew,to through traffic the Panama Canal. Notably, ship sizes also volumeof the it greatly,grewworld with shipping of and end of1999. the at did they This waterway. the of command assuming (ACP) Authority Canal Panama the in resulted and Panama States United the between treaty a but century, twenti- eth the of remainder the for waterway States the operated United The 1914. August in completed was and years 10 took project The Canal. Panama the built States In one of the world’s greatest engineering feats, the United Panama Canal Minicase ChapterOne 30 between 2009 and 2016. www.mhhe.com/shefrin2e Additional ResourcesandMaterialsforChapter1AreAvailableat www.mhhe.com/shefrin2e Additional ResourcesandMaterialsforChapter1AreAvailableat The In accepting the consortium’s bid in July 2009, the ACP Grupo called consortium a from came bid lowest The volume the century, twentieth the of course the Over The In accepting the consortium’s bid in July 2009, the ACP Grupo called consortium a from came bid lowest The volume the century, twentieth the of course the Over documented a series of problemsYorkNew of series a Timesdocumented documented a series of problemsYorkNew of series a Timesdocumented 26 26 The problems involved leaks in The problems involved leaks in Times. Times. 06 ruh cniin cue te ee o te ae to lake the of level the caused conditions drought 2016 in solution, clever a While basins. water-savinglarge in water conserving by problem the address could they that about a shortfall of fresh water, the project leaders decided Bank World the and Bank Development Inter-American the both from 2008 in warnings Despite shelved. being income Panamanians, and political protests led to the idea ever, building new dams wouldHow- have displaced many low-issue. the address would that dams of series a for planned and water more require would project expansion the ACP took control of the canal, they recognized that an Panama’stime water.of the most Fromdrinking supplies also lake The other. the to canal the of side one from go they as This ships large support to enough deep sets. be to has lake two the between lake artificial a with 50-mile-long side, Atlantic the on one and side Pacific the on concreteabout the lasting for acentury. “confident” as itself ACP the described trial-and-error, as problems fixing to approach project’s the characterizing Although practices. pouring substandard as well as ents ingredi- concrete of mix wrong the used having project the of criticism his for situation the monitored who pert steel reinforcing bars. The reinforcingbars. steel additional of insertion the requiring rushed cracks, water the through and walls, lock the lining concrete the in gregate. the During testing ag- phase in 2015, cracks emerged import to planned had bid, its preparing that in Bechtel, reason that for was It concrete. making questionable for of quality being as regarded generally and canal, the of side Pacific the excavatedfrom was aggregate, as est bidder. Moreover, the raw material for concrete, known concrete was next of the 71 percentthat smaller than low- for budget consortium’s the insurer, consortium’s the by from the firm analysisHill International, which was confidential commissioned 2010 a to According concrete. the bars, and the consortium chose low-quality ingredients for 06 ruh cniin cue te ee o te ae to lake the of level the caused conditions drought 2016 in solution, clever a While basins. water-savinglarge in water conserving by problem the address could they that about a shortfall of fresh water, the project leaders decided Bank World the and Bank Development Inter-American the both from 2008 in warnings Despite shelved. being income Panamanians, and political protests led to the idea ever, building new dams would haveHow- displaced many low-issue. the address would that dams of series a for planned and water more require would project expansion the ACP took control of the canal, they recognized that an Panama’stime water.of the most Fromdrinking supplies also lake The other. the to canal the of side one from go they as ships large This support to enough deep be sets. to has lake two the between lake artificial a 50-mile-long with side, Atlantic the on one and side Pacific the on concreteabout the lasting for acentury. “confident” as itself ACP the described trial-and-error, as problems fixing to approach project’s the characterizing Although practices. pouring substandard as well as ents ingredi- concrete of mix wrong the used having project the of criticism his for situation the monitored who pert steel reinforcing bars. The reinforcingbars. steel additional of insertion the requiring cracks, rushed the water through and walls, lock the lining concrete the in gregate. the During testing phase in ag- 2015, cracks emerged import to planned had bid, its preparing that in Bechtel, reason that for was It concrete. making questionable for quality of being as regarded generally and canal, the of side Pacific the excavatedfrom was aggregate, as est bidder. Moreover, the raw material for concrete, known concrete was next of the 71 percentthat smaller than low- for budget consortium’s the insurer, consortium’s the by from the firm Hill analysisInternational, which was commissioned confidential 2010 a to According concrete. the bars, and the consortium chose low-quality ingredients for h Pnm Cnl etrs w st o lcs one locks, of sets two features Canal Panama The The locks are made from concrete and reinforcing steel h Pnm Cnl etrs w st o lcs one locks, of sets two features Canal Panama The The locks are made from concrete and reinforcing steel quoted one concrete ex-concrete one quotedTimes quoted one concrete ex-concrete one quotedTimes 2/7/17 5:25PM she77208_ch01_001-031.indd 31 2/7/17 5:25PM locks. had insultedhisrepublic. geologists the that sayingby responded president manian Pana- then The believed.previously was than risk quake Panama indicated that Panama faced a much higher earth- forACP the workinganother and for U.S. the in Embassy ships, the especially onwindydays. alongside move to boats tug the for error for approximately 100 feet long. That leaves very little margin are instead 1,400 feet long and 180 theyHowever, feet wide. wide. Tugboatsfeet are 200 and long feet 1,528 about be to have would locks new the wide, feet 160 and long feet 1,200 approximately are ships container largest the Because through. going ships the than wider feet 40 and cluded that the new locks should be at least 328 feet longer New YorkTimes New the that stated ACP the at administration and gineering Berendrecht. in one the and canal Panama’s between differences tant impor- ignored addition in and process, planning the in workers canal include not did ACP the Panama, in tives forward much of the time. According to union representa- go to have will yet forward, than backward going stable more are theyMoreover, weather. turbulent during ships largeto providing stability of capable as notmotives, and loco- the than stable less areversatile, tugboats While the ships. large guide to tugboats uses Belgium which (Berendrecht), in system lock a on based The made was tugboats. decision with locomotives the replace to decided ACP the weather, bad in especially time, over effective very havebeen these Although locks. the of out and into large ships would have to lighten loads their considerably. that advisory an issued ACP the result a as and decline,

In June 2016, the first large first In June 2016, the ship went new the through working geologist one by reports 2007 November In In 2003 a feasibility study requested by the ACP con- ACP the by requested study feasibility a 2003 In ships guide to locomotives uses Canal original The 27 At the time, the executive vice-president of en of executivevice-president the time, the At coverage was out of date and addressed and date of out wascoverage - Case AnalysisQuestions . h US Am Crs f nier mitis locks maintains Engineers of Corps Army U.S. The 3. The 2. in introduced phenomena psychological 10 the Of 1. phase butwere resolved. construction the during onlyoccurred challengeshad that cal phenomenadescribedinChapter1. nothing psychologi- the of any reflects article Times the and more, locks the about did it what only saying but mentions little more than this. Discuss whether, by balanced, finely are and tons hundred canal several original weigh the in gates the that out points article generally the concrete is not heavily loaded. heavily not is concrete the generally and strength compressive gain to continues it because years hundred a last might concrete standard, below is case. One of these consultants points out that even if it mini- the in described issues of kinds the analyze to consultants engages and waterways, U.S. on dams and ship presentsanypotentialconflictsofinterest. relation- a such whether Discuss manufacturer. boat canal tug- the representing firm law the joined administrator the of son the tugboats, the manufactures that firm Spanish the to contract million $158 a awarded and givereasonstosupportyouranswer. minicase, the to apply ones which identify 1, Chapter repair or replace. In mentioning lock gates, the gates, lock mentioning In replace. or repair maintenance problems, and can take several months to omy. Michigan and posed a major threat to the U.S. econ Marie, Ste. Sault in locks an the at surfaced problems, which issue long-term create to tend to compression tension from changes associated The decades. of course the over hour an twice or once close and open nals is not so much the concrete, but the steel gates that notes that the problem with locks and dams on such ca- article notes that three days after the ACP the after days three that notes article Times 29 In this respect, lock gates can become serious become can gates lock respect, this In Behavioral Foundations 28 He also He Times 31 - 2/7/17 5:25PM www.mhhe.com/shefrin2e 2/7/17 5:25 PM 2/7/17 5:25 PM - - 31 31 Times Times He also He also 28 28 Behavioral Foundations Behavioral Foundations In this respect, lock gates can become serious 29 Times article notes that three days after the ACP In this respect, lock gates can become serious 29 Times article notes that three days after the ACP notes that the problem with locks and dams on such ca- such on dams and locks with problem the that notes that gates steel the but concrete, the much so not is nals open and close once or twice an hour over the course of decades. The associated changes from tension compression to tend to create long-term issue which problems, surfaced at the locks an in Sault Ste. Marie, econ U.S. the to threat major a posed and Michigan omy. maintenance problems, and can take several months to months several take can and problems, maintenance repair or replace. In mentioning lock gates, the Chapter 1, identify which ones apply to the minicase, your answer. and give reasons to support awarded a $158 million contract to the Spanish firm that manufactures the tugboats, the son of the administrator joined the law firm representing the tug- canal boat manufacturer. Discuss whether such a relation- ship presents any potential conflicts of interest. and dams on U.S. waterways, and engages consultants to analyze the kinds of issues described in the mini- it if even that out points consultants these of One case. is below standard, concrete might last a hundred years because it continues to gain compressive strength and generally the concrete is not heavily loaded. article points out that the gates in the weigh original several canal hundred tons and are finely balanced, by whether, Discuss this. than more little mentions but saying only what it did about the locks more, the Times article and reflects any of the psychologi- nothing cal phenomena described in Chapter 1. notes that the problem with locks and dams on such ca- such on dams and locks with problem the that notes that gates steel the but concrete, the much so not is nals open and close once or twice an hour over the course of decades. The associated changes from tension compression to tend to create long-term issue which problems, surfaced at the an locks in Sault Ste. Marie, econ U.S. the to threat major a posed and Michigan omy. maintenance problems, and can take several months to months several take can and problems, maintenance repair or replace. In mentioning lock gates, the Chapter 1, identify which ones apply to the minicase, your answer. and give reasons to support awarded a $158 million contract to the Spanish firm that manufactures the tugboats, the son of the administrator joined the law firm representing the tug- canal boat manufacturer. Discuss whether such a relation- conflicts of interest. ship presents any potential and dams on U.S. waterways, and engages consultants to analyze the kinds of issues described in the mini- it if even that out points consultants these of One case. is below standard, concrete might last a hundred years because it continues to gain compressive strength and generally the concrete is not heavily loaded. article points out that the gates in the weigh original several canal hundred tons and are finely balanced, by whether, Discuss this. than more little mentions but saying only what it did about the locks more, and the Times article reflects any of the psychologi- nothing cal phenomena described in Chapter 1. only challenges that had challenges occurredonly during the construction resolved. phase but were 1. Of the 10 psychological phenomena introduced in 2. The 3. The U.S. Army Corps of Engineers maintains locks Case Analysis Questions only challenges that had challenges occurredonly during the construction resolved. phase but were 1. Of the 10 psychological phenomena introduced in 2. The 3. The U.S. Army Corps of Engineers maintains locks Case Analysis Questions Case Analysis - - coverage was out of date and addressed coverage was out of date and addressed At the time, the vice-president executive of en 27 The original Canal uses locomotives to guide ships In 2003 a feasibility study requested by the ACP con- In November 2007 reports by one geologist working through the new went ship the2016, June In first large At the time, the vice-president executive of en

27

into into and out of the locks. Although these been have very effective over time, especially in bad weather, the ACP decided to replace the locomotives with decision tugboats. was made The based on a lock system (Berendrecht), in which uses Belgium tugboats to guide large ships. the While tugboats versatile, are less stable than the loco- and motives, not as capable of stability providing to large ships during turbulent weather. Moreover, they are more stable going backward than forward, yet will have to go representa- union to According time. the of much forward tives in Panama, the ACP did not include canal workers in the planning process, and in addition ignored impor- tant differences between Panama’s canal and the one in Berendrecht. decline, and as a result the ACP issued an advisory that considerably. theirloads lighten to have would ships large gineering and administration at the ACP stated that the New Times York for the ACP and another working for the U.S. Embassy in the U.S. for and another working the ACP for earth-higher much a faced Panama that indicated Panama quake risk than was previously believed. The then Pana- manian president responded by saying that the geologists had insulted his republic. locks. cluded that the new locks should be at least 328 feet longer longer feet 328 least at be should locks new the that cluded and 40 feet wider than the ships going through. Because the largest container ships are approximately 1,200 feet long and 160 feet wide, the new locks would have to be about 1,528 feet long and 200 are Tugboats feet wide. wide. feet 180 However, they and long feet 1,400 instead are margin little very leaves That long. feet 100 approximately for error for the tug boats to move alongside on windy days. especially the ships,

The original Canal uses locomotives to guide ships In 2003 a feasibility study requested by the ACP con- In November 2007 reports by one geologist working through the new went ship the2016, June In first large

decline, and as a result the ACP issued an advisory that considerably. theirloads lighten to have would ships large into and out of the locks. Although these been have very effective over time, especially in bad weather, the ACP decided to replace the locomotives with decision tugboats. was made The based on a lock system in (Berendrecht), which Belgium uses tugboats to guide large ships. the While tugboats versatile, are less stable than the loco- and motives, not as capable of stability providing to large ships during turbulent weather. Moreover, they are more stable going backward than forward, yet will have to go representa- union to According time. the of much forward tives in Panama, the ACP did not include canal workers in the planning process, and in addition ignored impor- tant differences between Panama’s canal and the one in Berendrecht. gineering and administration at the ACP stated that the New Times York cluded that the new locks should be at least 328 feet longer longer feet 328 least at be should locks new the that cluded and 40 feet wider than the ships going through. Because the largest container ships are approximately 1,200 feet long and 160 feet wide, the new locks would have to be about 1,528 feet long and 200 are feet Tugboats wide. wide. feet However, they 180 and long feet 1,400 instead are margin little very leaves That long. feet 100 approximately for error for the tug boats to move alongside on windy days. especially the ships, Embassy in the U.S. for and another working the ACP for earth-higher much a faced Panama that indicated Panama quake risk than was previously believed. The then Pana- manian president responded by saying that the geologists had insulted his republic. locks.

she77208_ch01_001-031.indd 31 2/7/17 5:25 PM she77208_ch01_001-031.indd 31 Times quoted one concrete ex- The locks are made from concrete and reinforcing steel steel reinforcing and concrete from made are locks The The Panama Canal features two sets of locks, one bars, and the consortium chose low-quality ingredients for for ingredients low-quality chose consortiumthe and bars, the concrete. According to a 2010 confidential commissioned was which International, analysis Hill firm the from by the consortium’s insurer, the consortium’s budget for low- than smaller thatpercent 71 the of next was concrete known concrete, for material raw the Moreover, bidder. est as aggregate, was from excavated the Pacific side of the canal, and generally regarded as being of quality for questionable making concrete. It was for that reason Bechtel, in that preparing its bid, had planned to import emerged cracks 2015, ag- in phase testing Duringthe gregate. in the concrete lining the lock walls, and through water the rushed cracks, requiring the insertion of additional steel bars. reinforcing The pert who monitored the situation for his criticism of the project having used the wrong mix of concrete ingredi- ents as well as substandard pouring practices. Although characterizing the project’s approach to fixing problems as trial-and-error, described the ACP itself as “confident” a century. for lasting about the concrete on the Pacific side and one on the Atlantic side, with50-mile-long a artificial lake between the two lake has to sets. be deep enough to support This large ships as they go from one side of the canal to the other. The lake also supplies drinkingFrom most the of water. time Panama’s an that recognized they canal, the of control took ACP the expansion project would require more water and planned for a series of dams that would address the issue. low- many displaced How- have would dams new building ever, idea the to led protests political and Panamanians, income being shelved. Despite warnings in 2008 from both the Inter-American Development Bank and the World Bank decided leaders project the water, fresh of shortfall a about that they could address the problem by conserving water in large water-saving basins. While a clever solution, in 2016 drought conditions caused the level of the lake to Times. The problems involved leaks in leaks involved problems The 26 documented Times a series of New York problems Over Over the course of the twentieth century, the volume The lowest bid came from a consortium called Grupo ACP the 2009, July in bid consortium’s the accepting In The Additional Resources and Materials for Chapter 1 Are Available at 1 Are Available for Chapter and Materials Resources Additional www.mhhe.com/shefrin2e between 2009 and 2016. and 2009 between 30 30 Chapter One Chapter One Minicase Panama Canal United the feats, engineering greatest world’s the of one In States built the Panama Canal. The project took 10 years and was completed in August 1914. The United operated the States waterway for the remainder of the eth twenti- century, but a treaty between the United States Panama and resulted in the Panama Canal Authority (ACP) assuming command of the waterway. This they did at the end of 1999. and of shipping with world grew greatly, it the of volume also sizes ship Notably, Canal. Panama thetraffic through to grew, the point where the largest container ships were Partly out thatof fear its thetoo big waterway. to traverse decision the made ACP the obsolete, become would canal to build new locks that would handle the larger container ships. In 2007 they put out a request for proposal (RFP), the project. soliciting bids for Unidos por el Canal that was led by the Spanish firm Sacyr Vallehermoso. The consortium’s members included firms from Panama, Belgium, and Italy. Their bid, $1 than $3.1 bid. billion the lower approximately lowest next was billion, the other bidders Notably, included Bechtel, one of the most world. the firmsin construction and engineering experienced 100-year the for time in project the completing of goal a set theanniversary of consortiumthe original canal. However, missed the August 2014 deadline, and the first container in June 2016. locks ship passed through the new concrete, concrete, insufficient water to support large ships in the canal, a shift from stable locomotives for guiding ships to less stable tugboats, insufficient margin for error from a new lock design that was too narrow, the identification of beneathearthquake thefaults canal, and a cost overrun cap- a is Below billion. $5.25 to rose billion $3.1 whereby the by coverage in points main the of summary sule she77208_ch01_001-031.indd 30 Introduction to Behavioral Analysis 33 Chapter Two flows. According to DCF analysis, projects should be undertaken if the present value of the expected future cash flows exceeds the initial investment required. In the same vein, funding decisions should be made in order to maximize the value of the firm, based on DCF analysis. For example, the use of debt provides a tax shield, which managers should balance or trade off against potential bankruptcy costs. Chapter Two In theory, managers serve the interests of investors, the owners of the firm. In practice, the interests of managers and investors might not be perfectly aligned. Con- flicts of interest are known as agency conflicts, in that investors are principals whose Introduction to interests are being served by managers acting as their agents. In general, agency con- flicts may prevent the firm’s managers from making decisions that result in full value maximization. In these situations, the attendant costs are known as agency costs. Behavioral Analysis For organizational purposes, imagine that managers seek to maximize an objec- tive function measuring performance. This objective function is a sum of terms such TheIntroduction main objective of this chapter is for to students to learn to structure ques - as the present value of the firm’s after-tax cash flows from operations and invest- tions as the first step in applying the 10 foundational psychological phenomena ment, the value associated with its financing strategy, and net benefits associated to behavioral corporate issues. Because these phenomena pertain to common with managerial interests such as compensation and perquisites. The summation of humanBehavioral traits, they affect both managers Analysis and investors. Certainly, managers need all three components is often called adjusted present value (APV). Notably, in the to understand how these phenomena impact their own judgments and decisions; traditional setting, markets are efficient so that managers and investors agree on Thehowever, main they objective also need of thisto understand chapter is the for decisions students of to other learn managers to structure as well ques as- how valuation reflects available information. tionsthe investing as the first public step whose in applying trading the activities 10 foundational establish psychologicalmarket prices. phenomena Managers’ decision environment is simplest when these components are inde- Afterto behavioral completing corporate this chapter issues. students Because will be able these to phenomenastructure questions pertain as tothey common engage pendent of each other, so that decisions about operations, financing, and compen- inhuman behavioral traits, analysis they affect to: both managers and investors. Certainly, managers need sation feature no interactions. When this is the case, we can consider decisions to understand how these phenomena impact their own judgments and decisions; about capital budgeting and capital structure separately, with decisions about in- 1. Identify the key biases that lead managers to make faulty financial judgments about however, they also need to understand the decisions of other managers as well as vestment made to maximize net present value (NPV), and decisions about capital risky alternatives. the investing public whose trading activities establish market prices. structure made to trade off the benefits of tax shields against the costs associated 2. Explain why reliance on heuristics and susceptibility to framing effects render After completing this chapter students will be able to structure questions as they engage with financial distress. However, in practice, there will be interactions. For exam- managers vulnerable to making faulty decisions that reduce firm value. in behavioral analysis to: ple, financing costs might be too high to justify a project whose net present value is 3. Recognize that investors are susceptible to the same biases as managers and otherwise positive, or interest payments associated with high debt levels might be 1. that Identify mispricing the key stemming biases that from lead investor managers errors to makecan cause faulty managers financial judgments to make faulty about used to provide managerial discipline in an attempt to counteract the potential for decisionsrisky alternatives. that reduce firm value. agency conflicts. Information asymmetries between managers and investors can 2. Explain why reliance on heuristics and susceptibility to framing effects render also complicate matters, as investors seek to make inferences about what managers managers vulnerable to making faulty decisions that reduce firm value. know based on their actions. 2.1 TRADITIONAL3. RecognizeTREATMENT that investors OF are CORPORATE susceptible to the same biases as managers and that mispricing stemming from investor errors can cause managers to make faulty FINANCIAL DECISIONSdecisions that reduce firm value. 2.2 BEHAVIORAL TREATMENT OF CORPORATE FINANCIAL DECISIONS Firms raise funds by borrowing on lines of credit, issuing commercial paper, selling 2.1 TRADITIONALcorporate TREATMENT bonds, issuing OFnew CORPORATEequity, and generating operating cash flows. Firms use The traditional material taught in corporate finance courses offers powerful tech- funds when they undertake projects, acquire fixed assets, build inventory, pay divi- niques that in theory help managers to make value-maximizing decisions for their FINANCIAL DECISIONSdends, engage in mergers and acquisitions, and deal with legal and regulatory issues. firms. Yet, in practice, psychological pitfalls hamper managers in applying these The traditional value-maximizing approach is based on discounted cash flow techniques correctly. The purpose of this text is to make students aware of potential Firms(DCF) raise analysis. funds Virtually by borrowing all the on decisions lines of justcredit, mentioned issuing commercialinvolve risk. paper, Because selling the psychological pitfalls as they arise in the various decision tasks that financial man- corporatecash flows bonds, are uncertain, issuing new expected equity, orand mean generating values operatingare used tocash assess flows. value. Firms Risk use agers confront, and to offer advice on how to mitigate the impact of these pitfalls. fundsenters wheninto the they analysis undertake through projects, the discount acquire ratefixed that assets, is applied build toinventory, the expected pay cashdivi- The behavioral pitfalls discussed in this text are not unique to managers, but are 32 dends, engage in mergers and acquisitions, and deal with legal and regulatory issues. prevalent across the general population, impacting investors as well as managers. The traditional value-maximizing approach is based on discounted cash flow (DCF) analysis. Virtually all the decisions just mentioned involve risk. Because the cash flows are uncertain, expected or mean values are used to assess value. Risk enters into the analysis through the discount rate that is applied to the expected cash 32 she77208_ch02_032-057.indd 32 2/7/17 6:20 PM

she77208_ch02_032-057.indd 33 2/7/17 6:20 PM

she77208_ch02_032-057.indd 32 2/7/17 6:20 PM Introduction to Behavioral Analysis 33 Chapter Two flows. According to DCF analysis, projects should be undertaken if the present value of the expected future cash flows exceeds the initial investment required. In the same vein, funding decisions should be made in orderIntroduction to maximize to Behavioral the Analysis value of33 the firm, based on DCF analysis. For example, the use of debt provides a tax shield, whichflows. managers According should to balance DCF analysis, or trade projects off against should potential be undertaken bankruptcy if the costs. present Chapter Two Invalue theory, of the managers expected serve future the cash interests flows exceedsof investors, the initial the ownersinvestment of therequired. firm. In practice,the same the interestsvein, funding of managers decisions and should investors be made might in ordernot be to perfectly maximize aligned. the value Con of- flictsthe of firm, interest based are on known DCF analysis.as agency For conflicts, example, in the that use investors of debt areprovides principals a tax whoseshield, Introduction to interestswhich are managers being served should by balance managers or trade acting off as against their agents. potential In bankruptcygeneral, agency costs. con- flicts mayIn theory, prevent managers the firm’s serve managers the interests from making of investors, decisions the thatowners result of inthe full firm. value In maximization.practice, the In interests these situations, of managers the and attendant investors costs might are not known be perfectly as agency aligned. costs. Con- Behavioral Analysis Forflicts organizational of interest are purposes,known as agencyimagine conflicts, that managers in that investorsseek to maximize are principals an objec-whose tiveinterests function are measuring being served performance. by managers This acting objective as their function agents. isIn ageneral, sum of agencyterms such con- TheIntroduction main objective of this chapter is for to students to learn to structure ques - as theflicts present may prevent value ofthe the firm’s firm’s managers after-tax from cash making flows decisions from operations that result andin full invest- value tions as the first step in applying the 10 foundational psychological phenomena ment,maximization. the value associated In these situations, with its financingthe attendant strategy, costs are and known net benefits as agency associated costs. to behavioral corporate issues. Because these phenomena pertain to common with managerialFor organizational interests purposes, such as compensationimagine that managers and perquisites. seek to maximizeThe summation an objec- of Behavioralhuman traits, they affect both managers Analysis and investors. Certainly, managers need all threetive function components measuring is often performance. called adjusted This objective present functionvalue (APV). is a sum Notably, of terms in such the to understand how these phenomena impact their own judgments and decisions; traditionalas the present setting, value markets of the are firm’s efficient after-tax so that cash managers flows from and operations investors and agree invest- on Thehowever, main they objective also need of thisto understand chapter is the for decisions students of to other learn managers to structure as well ques as- howment, valuation the value reflects associated available with information. its financing strategy, and net benefits associated tionsthe investing as the first public step whose in applying trading the activities 10 foundational establish psychologicalmarket prices. phenomena Managers’with managerial decision interests environment such as compensation is simplest when and perquisites. these components The summation are inde of- Afterto behavioral completing corporate this chapter issues. students Because will be able these to phenomenastructure questions pertain as tothey common engage pendentall three of each components other, so is thatoften decisions called adjusted about operations,present value financing, (APV). Notably, and compen in the- inhuman behavioral traits, analysis they affect to: both managers and investors. Certainly, managers need sationtraditional feature setting,no interactions. markets are When efficient this isso thethat case, managers we can and consider investors decisions agree on to understand how these phenomena impact their own judgments and decisions; abouthow capital valuation budgeting reflects and available capital information. structure separately, with decisions about in- 1. Identify the key biases that lead managers to make faulty financial judgments about however, they also need to understand the decisions of other managers as well as vestmentManagers’ made to decision maximize environment net present is valuesimplest (NPV), when and these decisions components about are capital inde- risky alternatives. the investing public whose trading activities establish market prices. structurependent made of each to trade other, off so thethat benefits decisions of about tax shields operations, against financing, the costs and associated compen- 2. Explain why reliance on heuristics and susceptibility to framing effects render After completing this chapter students will be able to structure questions as they engage withsation financial feature distress. no interactions. However, Whenin practice, this is there the case,will bewe interactions. can consider For decisions exam- managers vulnerable to making faulty decisions that reduce firm value. in behavioral analysis to: ple,about financing capital costs budgeting might be and too capital high to structure justify aseparately, project whose with netdecisions present about value in is- 3. Recognize that investors are susceptible to the same biases as managers and otherwisevestment positive, made to or maximize interest payments net present associated value (NPV), with and high decisions debt levels about might capital be 1. that Identify mispricing the key stemming biases that from lead investor managers errors to makecan cause faulty managers financial judgments to make faulty about usedstructure to provide made managerial to trade off discipline the benefits in an of attempttax shields to counteract against the the costs potential associated for decisionsrisky alternatives. that reduce firm value. agencywith conflicts.financial distress.Information However, asymmetries in practice, between there will managers be interactions. and investors For exam can- 2. Explain why reliance on heuristics and susceptibility to framing effects render alsople, complicate financing matters, costs might as investors be too high seek to tojustify make a inferencesproject whose about net whatpresent managers value is managers vulnerable to making faulty decisions that reduce firm value. knowotherwise based on positive, their actions. or interest payments associated with high debt levels might be 2.1 TRADITIONAL3. RecognizeTREATMENT that investors OF are CORPORATE susceptible to the same biases as managers and used to provide managerial discipline in an attempt to counteract the potential for that mispricing stemming from investor errors can cause managers to make faulty agency conflicts. Information asymmetries between managers and investors can FINANCIAL DECISIONSdecisions that reduce firm value. 2.2 BEHAVIORALalso TREATMENT complicate matters, OF as investors CORPORATE seek to make inferences about what managers FINANCIAL DECISIONSknow based on their actions. Firms raise funds by borrowing on lines of credit, issuing commercial paper, selling 2.1 TRADITIONALcorporate TREATMENT bonds, issuing OFnew CORPORATEequity, and generating operating cash flows. Firms use 2.2 BEHAVIORALThe traditional TREATMENT material taught OF in CORPORATEcorporate finance courses offers powerful tech- funds when they undertake projects, acquire fixed assets, build inventory, pay divi- niques that in theory help managers to make value-maximizing decisions for their FINANCIAL DECISIONSdends, engage in mergers and acquisitions, and deal with legal and regulatory issues. FINANCIALfirms. DECISIONS Yet, in practice, psychological pitfalls hamper managers in applying these The traditional value-maximizing approach is based on discounted cash flow techniques correctly. The purpose of this text is to make students aware of potential Firms(DCF) raise analysis. funds Virtually by borrowing all the on decisions lines of justcredit, mentioned issuing commercialinvolve risk. paper, Because selling the psychologicalThe traditional pitfalls material as they taught arise in in corporate the various finance decision courses tasks offers that financialpowerful man-tech- corporatecash flows bonds, are uncertain, issuing new expected equity, orand mean generating values operatingare used tocash assess flows. value. Firms Risk use agersniques confront, that in and theory to offer help advice managers on how to make to mitigate value-maximizing the impact decisionsof these pitfalls. for their fundsenters wheninto the they analysis undertake through projects, the discount acquire ratefixed that assets, is applied build toinventory, the expected pay cashdivi- Thefirms. behavioral Yet, in practice, pitfalls discussedpsychological in this pitfalls text arehamper not unique managers to managers, in applying but these are 32 dends, engage in mergers and acquisitions, and deal with legal and regulatory issues. prevalenttechniques across correctly. the general The purposepopulation, of this impacting text is to makeinvestors students as well aware as ofmanagers. potential The traditional value-maximizing approach is based on discounted cash flow psychological pitfalls as they arise in the various decision tasks that financial man- (DCF) analysis. Virtually all the decisions just mentioned involve risk. Because the agers confront, and to offer advice on how to mitigate the impact of these pitfalls. cash flows are uncertain, expected or mean values are used to assess value. Risk The behavioral pitfalls discussed in this text are not unique to managers, but are enters into the analysis through the discount rate that is applied to the expected cash prevalent across the general population, impacting investors as well as managers. 32 she77208_ch02_032-057.indd 32 2/7/17 6:20 PM she77208_ch02_032-057.indd 33 2/7/17 6:20 PM

she77208_ch02_032-057.indd 32 2/7/17 6:20she77208_ch02_032-057.indd PM 33 2/7/17 6:20 PM 34 Chapter Two Introduction to Behavioral Analysis 35

Many of these pitfalls were identified by psychologists Daniel Kahneman and the from investors. The distinction is important in that the two sources of cost typically late Amos Tversky. Their work was recognized with a Nobel Prize in Economics, require very different remedies. Remedies for agency conflicts tend to emphasize 34 Chapter Two awarded to Kahneman in 2002, jointly with economist Vernon Smith, who pioneered the manipulation of incentives. Remedies for behavioral pitfalls tend to emphasize the use of experimental techniques to study markets. training and process. ManyAlthough of these executivespitfalls were are identified prone to byexperiencing psychologists the Daniel same psychologicalKahneman and phenom- the For organizational purposes, imagine that in a behavioral environment, manag- lateena Amos as the Tversky. general population,Their work wasthe evidencerecognized indicates with a Nobelthat for Prize some in phenomena Economics, they ers seek to maximize a performance objective, just as they do in a traditional set- awardeddo so with to Kahneman greater intensity. in 2002, jointly In particular, with economist chief executivesVernon Smith, exhibit who morepioneered disposi- ting. However, unlike the traditional setting, psychological pitfalls might prevent thetional use ofoptimism experimental and less techniques risk aversion to study than markets. the general population, features which managers from achieving perfect maximization—for example because of behaviors areAlthough accentuated executives among are CEOs prone whoto experiencing are taller and the youngersame psychological than their counterparts,phenom- described in Exhibit 2-1. enaand as who the generalare also population, male. CEOs the also evidence exhibit indicates greater thatdispositional for some phenomenaoptimism than they their Moreover, because psychological pitfalls can also impact the judgments of inves- doCFOs, so with which greater is a traitintensity. that manyIn particular, CFOs have chief noticed. executives1 exhibit more disposi- tors, managers might judge value differently from the markets even when all infor- tionalBecause optimism psychologically and less risk aversioninduced mistakesthan the general can be, population,and often are, features very expensive,which mation is common. When psychological pitfalls distort market prices, we say that arestudying accentuated behavioral among corporate CEOs who finance are taller is vital.and younger Exhibit than 2-1 their provides counterparts, a thumbnail Sentiment prices reflect sentiment. When managers perceive sentiment in the market, they andsketch who of are the also main male. points CEOs in also the exhibitchapter. greater The chapter dispositional discusses optimism how the than 10 their founda- Distortion in market prices might view themselves as having opportunities to engage in behaviors that reflect CFOs,tional which psychological is a trait thatphenomena many CFOs introduced have noticed. in Chapter1 1 lead managers to make caused by psychological catering and market timing. Catering entails actions that exploit sentiment in order phenomena faultyBecause decisions, psychologically thereby reducing induced mistakesthe values can of be, their and firms. often are, very expensive, to increase a firm’s stock price. Market timing entails actions that take advantage of studyingLike agency behavioral conflicts, corporate behavioral finance phenomena is vital. Exhibit also cause 2-1 provides managers a thumbnailto take actions Catering mispriced securities, such as issuing new equity when stocks are overvalued. sketchthat are of thedetrimental main points to thein theinterests chapter. of The shareholders. chapter discusses However, how behavioral the 10 founda- costs are Actions that exploit senti- Just as we speak of traditional APV, we can speak of behavioral APV. Behavioral tionalthe result psychological of managers’ phenomena mistakes, introduced not the result in Chapter of managers 1 lead having managers different to make interests ment to increase the market APV incorporates the impacts of psychological phenomena that affect managers price of securities. faulty decisions, thereby reducing the values of their firms. directly, as well as the effects that stem from catering and market timing. Like agency conflicts, behavioral phenomena also cause managers to take actions Market timing The remainder of the chapter provides two examples that serve as vehicles in Actions that take advantage EXHIBIT 2 1 that are detrimental to the interests of shareholders. However, behavioral costs are helping students learn to apply psychological concepts when analyzing corporate - How Psychologythe result Affects of managers’ Financial Decisionsmistakes, not the result of managers having different interests of securities being mis- financial issues. The first example pertains to , known mostly for Psychological priced at different points in time. manufacturing hardware such as servers, and to a lesser extent software such as the Phenomenon Example of Faulty Financial Decision Resulting Outcome for Firm programming language , which transmits web pages over the Internet. The sec- 1. Biases ond example pertains to Merck, known for developing and selling a wide a variety EXHIBIT 2-1 How Psychology Affects Financial Decisions Excessive optimism Delay cost cutting during a business recession Lower profits of pharmaceutical products. PsychologicalOverconfidence Make inferior acquisitions when cash-rich Reduce firm value because risk underestimated In working through the examples, a major objective is to structure questions that Phenomenon Example of Faulty Financial Decision Resulting Outcome for Firm Confirmation bias Ignore information that is counter to current Lower profits from delayed reaction to changing generate discussion around the issues described in Exhibit 2-1. The example involv- 1. Biases viewpoint environment ing Sun Microsystems is used to focus on biases and heuristics, while the example Illusion of control Overestimate own degree of control Incur higher costs than necessary Excessive optimism Delay cost cutting during a business recession Lower profits involving Merck is used to focus on framing effects. Overconfidence2. Heuristics Make inferior acquisitions when cash-rich Reduce firm value because risk underestimated Confirmation bias Ignore information that is counter to current Lower profits from delayed reaction to changing Representativeness Choose wrong projects based on biased forecasts Reduce firm value because net present value viewpoint environment (NPV) not maximized Illusion of control Overestimate own degree of control Incur higher costs than necessary 2.3 ANALYZING BIASES Availability Choose wrong projects based on biased forecasts Reduce firm value because of misjudged 2. Heuristics priorities and risks Anchoring Become fixated on a number and adjust Reduce firm value because of biased growth The discussion about Sun Microsystems in the nearby Behavioral Pitfalls box is Representativeness Choose wrong projects based on biased forecasts Reduce firm value because net present value insufficiently forecasts based on an article that appeared in Businessweek. For the sake of exposition, the (NPV) not maximized Affect Rely on instincts instead of formal valuation Reduce firm value because negative NPV Availability Choose wrong projects based on biased forecasts Reduce firm value because of misjudged discussion below takes the characterizations described in the article as accurate, and analysis projects adopted priorities and risks then proceeds to analyze them. Anchoring Become fixated on a number and adjust Reduce firm value because of biased growth As you read the discussion about Sun in the Behavioral Pitfalls box, try to iden- 3. Framing effects insufficiently forecasts tify two sets of concepts, first the psychological attributes of Sun’s chief executive AffectLoss aversion RelyLosses on instincts loom larger instead than of gainsformal of valuation the same size ReduceForegone firm value tax shield because benefits negative because NPV of officer (CEO) Scott McNealy, and second the decisions made by Sun’s managers. analysis projects adopted aversion to debt The manner in which psychological attributes impact managers’ decisions about Fourfold pattern Throw good money after bad in losing projects Reduce firm value because of a negative NPV investment and financing lies at the core of behavioral corporate finance. Although 3. Framing effects decision most of the examples in this chapter involve decisions made by the CEO, the Loss aversion Losses loom larger than gains of the same size Foregone tax shield benefits because of aversion to debt Fourfold pattern Throw good money after bad in losing projects Reduce firm value because of a negative NPV decision

she77208_ch02_032-057.indd 34 she77208_ch02_032-057.indd2/7/17 6:20 PM 35 2/7/17 6:20 PM

she77208_ch02_032-057.indd 34 2/7/17 6:20 PM 34 Chapter Two Introduction to Behavioral Analysis 35

Many of these pitfalls were identified by psychologists Daniel Kahneman and the from investors. The distinction is important in that the two sources of cost typically late Amos Tversky. Their work was recognized with a Nobel Prize in Economics, require very different remedies. Remedies for agency conflicts tend to emphasize awarded to Kahneman in 2002, jointly with economist Vernon Smith, who pioneered the manipulation of incentives. Remedies for behavioralIntroduction pitfalls to Behavioraltend to emphasizeAnalysis 35 the use of experimental techniques to study markets. training and process. Although executives are prone to experiencing the same psychological phenom- Forfrom organizational investors. The purposes, distinction imagine is important that in thata behavioral the two sources environment, of cost typicallymanag- ena as the general population, the evidence indicates that for some phenomena they ers requireseek to verymaximize different a performance remedies. Remedies objective, for justagency as they conflicts do in tend a traditional to emphasize set- do so with greater intensity. In particular, chief executives exhibit more disposi- ting.the However, manipulation unlike of theincentives. traditional Remedies setting, for psychological behavioral pitfalls pitfalls tend might to emphasize prevent tional optimism and less risk aversion than the general population, features which managerstraining from and process.achieving perfect maximization—for example because of behaviors are accentuated among CEOs who are taller and younger than their counterparts, describedFor inorganizational Exhibit 2-1. purposes, imagine that in a behavioral environment, manag- and who are also male. CEOs also exhibit greater dispositional optimism than their Moreover,ers seek to because maximize psychological a performance pitfalls objective, can also just impact as they the do judgments in a traditional of inves- set- CFOs, which is a trait that many CFOs have noticed.1 tors,ting. managers However, might unlike judge the value traditional differently setting, from psychological the markets pitfalls even when might all prevent infor- Because psychologically induced mistakes can be, and often are, very expensive, mationmanagers is common. from achieving When psychological perfect maximization—for pitfalls distort example market because prices, ofwe behaviors say that studying behavioral corporate finance is vital. Exhibit 2-1 provides a thumbnail Sentiment pricesdescribed reflect insentiment Exhibit 2-1.. When managers perceive sentiment in the market, they sketch of the main points in the chapter. The chapter discusses how the 10 founda- Distortion in market prices might Moreover,view themselves because as psychological having opportunities pitfalls can to also engage impact in thebehaviors judgments that of reflect inves- tional psychological phenomena introduced in Chapter 1 lead managers to make caused by psychological cateringtors, managersand market might timing. judge Catering value differently entails actions from the that markets exploit even sentiment when allin orderinfor- phenomena faulty decisions, thereby reducing the values of their firms. to increasemation isa firm’scommon. stock When price. psychological Market timing pitfalls entails distort actions market that prices, take advantage we say that of Like agency conflicts, behavioral phenomena also cause managers to take actions CateringSentiment mispricedprices reflectsecurities, sentiment such as. Whenissuing managers new equity perceive when sentimentstocks are inovervalued. the market, they that are detrimental to the interests of shareholders. However, behavioral costs are ActionsDistortion that exploit in market senti- prices Justmight as weview speak themselves of traditional as having APV, opportunities we can speak to engageof behavioral in behaviors APV. Behavioralthat reflect the result of managers’ mistakes, not the result of managers having different interests ment tocaused increase by psychological the market APVcatering incorporates and market the timing.impacts Catering of psychological entails actions phenomena that exploit that sentiment affect managers in order price ofphenomena securities. directly,to increase as well a firm’sas the stockeffects price. that Market stem from timing catering entails and actions market that timing. take advantage of MarketCatering timing Themispriced remainder securities, of the such chapter as issuing provides new twoequity examples when stocks that areserve overvalued. as vehicles in ActionsActions that take that advantageexploit senti- EXHIBIT 2 1 helpingJust students as we speak learn of to traditional apply psychological APV, we can concepts speak of behavioralwhen analyzing APV. Behavioralcorporate - How Psychology Affects Financial Decisions of securitiesment to being increase mis- the marketfinancial APV incorporatesissues. The first the exampleimpacts ofpertains psychological to Sun Microsystems, phenomena that known affect mostly managers for price of securities. Psychological priced at different points directly, as well as the effects that stem from catering and market timing. in time. manufacturing hardware such as servers, and to a lesser extent software such as the Phenomenon Example of Faulty Financial Decision Resulting Outcome for Firm Market timing programmingThe remainder language of Java,the chapter which transmitsprovides two web examples pages over that the serve Internet. as vehicles The sec- in 1. Biases Actions that take advantageondhelping example students pertains learn to Merck, to apply known psychological for developing concepts and when selling analyzing a wide corporatea variety of securities being mis- financial issues. The first example pertains to Sun Microsystems, known mostly for Excessive optimism Delay cost cutting during a business recession Lower profits priced at different points of pharmaceutical products. Overconfidence Make inferior acquisitions when cash-rich Reduce firm value because risk underestimated in time. Inmanufacturing working through hardware the examples, such as servers, a major and objective to a lesser is toextent structure software questions such as that the Confirmation bias Ignore information that is counter to current Lower profits from delayed reaction to changing generateprogramming discussion language around Java, the issues which describedtransmits webin Exhibit pages over2-1. Thethe Internet. example The involv- sec- viewpoint environment ing ondSun exampleMicrosystems pertains is toused Merck, to focus known on forbiases developing and heuristics, and selling while a wide the examplea variety Illusion of control Overestimate own degree of control Incur higher costs than necessary involvingof pharmaceutical Merck is used products. to focus on framing effects. 2. Heuristics In working through the examples, a major objective is to structure questions that Representativeness Choose wrong projects based on biased forecasts Reduce firm value because net present value generate discussion around the issues described in Exhibit 2-1. The example involv- (NPV) not maximized 2.3 ANALYZING ingBIASES Sun Microsystems is used to focus on biases and heuristics, while the example Availability Choose wrong projects based on biased forecasts Reduce firm value because of misjudged involving Merck is used to focus on framing effects. priorities and risks Anchoring Become fixated on a number and adjust Reduce firm value because of biased growth The discussion about Sun Microsystems in the nearby Behavioral Pitfalls box is insufficiently forecasts based on an article that appeared in Businessweek. For the sake of exposition, the Affect Rely on instincts instead of formal valuation Reduce firm value because negative NPV 2.3 ANALYZINGdiscussion BIASES below takes the characterizations described in the article as accurate, and analysis projects adopted then proceeds to analyze them. AsThe you discussion read the aboutdiscussion Sun Microsystemsabout Sun in thein theBehavioral nearby Behavioral Pitfalls box, Pitfalls try to boxiden is- 3. Framing effects tify basedtwo sets on ofan concepts,article that first appeared the psychological in Businessweek attributes. For the of sakeSun’s of chief exposition, executive the Loss aversion Losses loom larger than gains of the same size Foregone tax shield benefits because of officerdiscussion (CEO) below Scott takesMcNealy, the characterizations and second the described decisions in made the article by Sun’s as accurate, managers. and aversion to debt Thethen manner proceeds in which to analyze psychological them. attributes impact managers’ decisions about Fourfold pattern Throw good money after bad in losing projects Reduce firm value because of a negative NPV investmentAs you and read financing the discussion lies at theabout core Sun of inbehavioral the Behavioral corporate Pitfalls finance. box, try Although to iden- decision mosttify of two the sets examples of concepts, in this first chapter the psychological involve decisions attributes made of Sun’s by chiefthe CEO, executive the officer (CEO) Scott McNealy, and second the decisions made by Sun’s managers. The manner in which psychological attributes impact managers’ decisions about investment and financing lies at the core of behavioral corporate finance. Although most of the examples in this chapter involve decisions made by the CEO, the

she77208_ch02_032-057.indd 34 she77208_ch02_032-057.indd2/7/17 6:20 PM 35 2/7/17 6:20 PM

she77208_ch02_032-057.indd 35 2/7/17 6:20 PM 36 Chapter Two bchba_tt Behavioral Pitfalls: Scott McNealy and Sun Microsystems Introduction to Behavioral Analysis 37 36 Chapter Two bchba_tt Behavioral Pitfalls: Scott McNealy and Sun Microsystems Delayed Cost Cutting and Value Loss The cover story in the July 26, 2004, issue of Business- new projects. In justifying that decision, McNealy stated: There is evidence suggesting that when press coverage describes an executive as “op- 36week Chapter is about Two Scott McNealy, the chief executive officer of “The Internet is still wildly underhyped, underutilized, and bchba_tt timistic,” that the executive being discussed is excessively optimistic. In this regard, high-technologyBehavioral firm Sun Pitfalls: Microsystems. Sun Scott is known McNealy as underimplemented. and Sun Microsystems I think we’re looking at the largest Businessweek “optimistic” aThe leading cover manufacturer story in the of July servers 26, 2004, and for issue having of Business- invented equipmentnew projects. business In justifying in the history that decision, of anything. McNealy The growth stated: the magazine article specifically included among the list theweek Internet is about software Scott McNealy, programming the chief language executive Java. officer The of opportunities“The Internet areis still stunning.” wildly underhyped, underutilized, and of attributes to describe Sun’s CEO Scott McNealy. Therefore, a natural question to Businessweekhigh-technology article firm Sunuses Microsystems. the following Sunadjectives is known to de- as underimplemented.Cisco Systems is Ithe think leading we’re producer looking of at router the largest prod- ask is whether excessive optimism affected McNealy’s decisions as a manager. For The cover story in the July 26, 2004, issue of Business- new projects. In justifying that decision, McNealy stated: scribea leading Scott manufacturer McNealy: optimistic, of servers smart, and foracerbic, having cocky, invented and uctsequipment used onbusiness the Internet. in the history On March of anything. 8, 2001, The Cisco growth an- example, is there evidence that excessive optimism led him to delay cost-cutting mea- week is about Scott McNealy, the chief executive officer of “The Internet is still wildly underhyped, underutilized, and combative.the Internet These software are all programming psychological language traits that influenced Java. The nouncedopportunities that arebecause stunning.” the economic downturn looked like sures, which resulted in steep losses for Sun? Notably, the general evidence suggests high-technology firm Sun Microsystems. Sun is known as underimplemented. I think we’re looking at the largest McNealy’sBusinessweek business article decisions. uses the following adjectives to de- it wouldCisco last Systems much longeris the leadingthan expected, producer it was of router going toprod- lay a leading manufacturer of servers and for having invented equipment business in the history of anything. The growth that people who report that economic conditions will improve are twice as optimistic scribeMcNealy Scott McNealy: was one optimistic, of Sun Microsystem’s smart, acerbic, founders, cocky, and be- offucts 18 used percent on the of its Internet. workforce. On MarchSome of 8, Sun’s 2001, executives Cisco an- 2 the Internet software programming language Java. The opportunities are stunning.” as those who report that economic conditions will either stay the same or get worse. camecombative. its CEO These in 1984, are all and psychological had a history traits of being that influenced willing to wantednounced to that follow because suit. Onethe economic stated: “When downturn times looked are hard, like Businessweek article uses the following adjectives to de- Cisco Systems is the leading producer of router prod- Was McNealy excessively optimistic when he predicted that the recession of 2001 takeMcNealy’s major businessrisks in order decisions. to avoid disaster. During the 1980s, you’veit would got last tomuch shoot longer activities than expected, that aren’t it makingwas going money.” to lay scribe Scott McNealy: optimistic, smart, acerbic, cocky, and ucts used on the Internet. On March 8, 2001, Cisco an- would be brief, albeit with extreme changes (sharp edges)? The National Bureau againstMcNealy the advice was one of his of Sunexecutives, Microsystem’s he decided founders, to substi- be- Chiefoff 18 Operatingpercent of Officer its workforce. Ed Zander Some proposed of Sun’s major executives cuts. combative. These are all psychological traits that influenced nounced that because the economic downturn looked like of Economic Research (NBER) is officially responsible for determining when eco- tutecame Sun’s its CEO own in microprocessors 1984, and had a for history those of manufactured being willing byto However,wanted to McNealy follow suit. refused One stated: to do so, “When and times Zander are subse- hard, McNealy’s business decisions. it would last much longer than expected, it was going to lay Motorola.take major That risks decision in order turnedto avoid out disaster. well for During Sun. the 1980s, quentlyyou’ve gotresigned. to shoot activities that aren’t making money.” nomic recessions begin and end in the United States. According to the NBER, the McNealy was one of Sun Microsystem’s founders, be- off 18 percent of its workforce. Some of Sun’s executives againstDuring the the advice 1990s, of his Sun’s executives, competitors he decided produced to serverssubsti- ChiefIn an Operating effort to cut Officer costs, Ed Sun’s Zander customers proposed sought major low-end cuts. U.S. economy entered a recession in March 2001, and the recession lasted until came its CEO in 1984, and had a history of being willing to wanted to follow suit. One stated: “When times are hard, thattute usedSun’s Microsoft’sown microprocessors Windows operating for those systems. manufactured McNealy by servers.However, McNealy McNealy initially refused dismissed to do so, their and concerns. Zander subse- Later, November 2001, a period of nine months that was neither brief nor atypical. take major risks in order to avoid disaster. During the 1980s, you’ve got to shoot activities that aren’t making money.” insteadMotorola. chose That todecision invest turnedin servers out thatwell ranfor Sun.Sun’s own soft- Sunquently spent resigned. $2 billion to acquire Cobalt, a manufacturer of The slowdown was longer for firms engaging in business investment. Sun against the advice of his executives, he decided to substi- Chief Operating Officer Ed Zander proposed major cuts. wareDuring Solaris. the That 1990s, decision Sun’s also competitors turned out produced well, as serversSolaris low-costIn an effortservers. to cutHowever, costs, Sun’s after customersthe acquisition, sought Sun low-end chose Microsystems’ customers are other firms. When those firms purchase Sun’s servers, they tute Sun’s own microprocessors for those manufactured by However, McNealy refused to do so, and Zander subse- camethat used to be Microsoft’s highly praised Windows for its operating speed, reliability, systems. and McNealy secu- toservers. limit Cobalt’s McNealy budget, initially and dismissed McNealy later their admitted concerns. that Later, the Motorola. That decision turned out well for Sun. quently resigned. are engaging in business investment. During the 2001 recession, investment fell for six rity.instead Sun’s chose sales, to profits, invest and in serverscash holdings that ran soared, Sun’s andown it softdra- acquisitionSun spent $2was billion a mistake. to acquire Cobalt, a manufacturer of During the 1990s, Sun’s competitors produced servers In an effort to cut costs, Sun’s customers sought low-end consecutive quarters, twice as long as the period real gross domestic product (GDP) fell. maticallyware Solaris. increased That decision its spending also turnedon a great out well,many as research Solaris low-costThe economic servers. However,downturn afterwas muchthe acquisition, longer than Sun McNealy chose that used Microsoft’s Windows operating systems. McNealy servers. McNealy initially dismissed their concerns. Later, Scott McNealy’s prediction that the recession of 2001 would be brief was ex- andcame development to be highly praised projects. for Duringits speed, the reliability, technology and secu stock- hadto limit forecast. Cobalt’s In budget,the next and three McNealy years, later Sun admittedlost a third that of the its instead chose to invest in servers that ran Sun’s own soft- Sun spent $2 billion to acquire Cobalt, a manufacturer of cessively optimistic. Was his prediction necessarily biased? In other words, might marketrity. Sun’s bubble sales, profits, in the lateand cash 1990s, holdings which soared, peaked and in Marchit dra- marketacquisition share, was its a sales mistake. fell by 48 percent, and its stock price ware Solaris. That decision also turned out well, as Solaris low-cost servers. However, after the acquisition, Sun chose Scott McNealy’s prediction about the recession of 2001 have been reasonable, 2000,matically Sun’s increased price-to-earnings its spending ratio on (P/E) a great soared many to 119.research fell Thefrom economic $64 in 2000 downturn to about was $4much in 2004.longer Itsthan net McNealy income came to be highly praised for its speed, reliability, and secu- to limit Cobalt’s budget, and McNealy later admitted that the andSun’s development fortunes changed projects. with During the onset the technology of an economic stock turnedhad forecast. negative In inthe 2002, next withthree a years,loss of Sun $587 lost million. a third Losses of its even though it was wrong after the fact? rity. Sun’s sales, profits, and cash holdings soared, and it dra- acquisition was a mistake. recessionmarket bubble in 2001. in theWall late Street 1990s, analysts which called peaked for Sun in Marchto cut inmarket 2003 share, and 2004 its sales totalled fell by $3.4 48 billionpercent, and and $388 its stock million price re- Between World War II and 2000, the average length of a U.S. recession was matically increased its spending on a great many research The economic downturn was much longer than McNealy costs.2000, However,Sun’s price-to-earnings McNealy was ratio optimistic (P/E) soared that the to recession 119. spectively,fell from $64 the inresult 2000 of toboth about reduced $4 in demand 2004. Itsfor netits comput income- 11.6 months. During those recessions, real gross domestic investment tended to and development projects. During the technology stock had forecast. In the next three years, Sun lost a third of its wouldSun’s be fortunesshort-lived. changed In 2001, with during the onseta conference of an economic call with ersturned from negative major corporate in 2002, customerswith a loss and of $587 greater million. competition Losses fall for 12 months. The recession before 2001 occurred in 1990–1991 and featured market bubble in the late 1990s, which peaked in March market share, its sales fell by 48 percent, and its stock price analysts,recession hein 2001. said aboutWall Street business analysts cycles: called “We for don’t Sun to have cut fromin 2003 IBM, and Hewlett-Packard 2004 totalled (HP),$3.4 andbillion Dell. and $388 million re- two consecutive quarters of negative growth in real GDP. During the recession of costs.2000, However,Sun’s price-to-earnings McNealy was ratio optimistic (P/E) soared that the to recession 119. spectively,fell from $64 the inresult 2000 of toboth about reduced $4 in demand 2004. Itsfor netits comput income- rolling waves. We seem to have real edges.” Instead of Source: J. Kerstetter and P. Burrows, “Sun: A CEO’s Last Stand,” 1990–1991, real gross domestic investment fell for 12 months. wouldSun’s be fortunesshort-lived. changed In 2001, with during the onseta conference of an economic call with ersturned from negative major corporate in 2002, customerswith a loss and of $587 greater million. competition Losses cutting costs during the recession, Sun invested heavily in Businessweek, July 26, 2004, Bloomberg. What are we to conclude? There is nothing in the historical record to suggest analysts,recession hein 2001. said aboutWall Street business analysts cycles: called “We for don’t Sun to have cut fromin 2003 IBM, and Hewlett-Packard 2004 totalled (HP),$3.4 andbillion Dell. and $388 million re- costs. However, McNealy was optimistic that the recession spectively, the result of both reduced demand for its comput- that recessions would be brief affairs. The more reasonable conclusion might be rolling waves. We seem to have real edges.” Instead of Source: J. Kerstetter and P. Burrows, “Sun: A CEO’s Last Stand,” cuttingwould be costs short-lived. during the In 2001,recession, during Sun a conference invested heavily call with in ersBusinessweek, from major July corporate 26, 2004, customers Bloomberg. and greater competition that Scott McNealy’s prediction was biased in the direction of optimism and that his analysts, he said about business cycles: “We don’t have from IBM, Hewlett-Packard (HP), and Dell. optimism led him to delay cost cutting, thereby destroying value at his firm. rolling waves. We seem to havepsychological real edges.” traits Instead described of affectSource: managers J. Kerstetter at and all P. levels Burrows, of “Sun: the firm.A CEO’s Indeed, Last Stand,” they cutting costs during the recession,affect Sunmost invested people, heavily be they in managersBusinessweek, or not. July 26, 2004, Bloomberg. Sentiment: Excessively Optimistic Sun Stockholders Recall from Chapter 1 that a bias is a predisposition toward error. This section psychological traits described affect managers at all levels of the firm. Indeed, they During the stock market bubble between January 1997 and June 2000, irrational discusses whether four specific biases were manifest within Sun’s processes, with affect most people, be they managers or not. exuberance drove up the prices of both the S&P 500 and Sun’s stock. Exhibit 2-2 the four biases being excessive optimism, overconfidence, confirmation bias, and Recall from Chapter 1 that a bias is a predisposition toward error. This section displays the market value of Sun’s equity between 1986 and 2003. An investor who thepsychological illusion of traitscontrol. described affect managers at all levels of the firm. Indeed, they discusses whether four specific biases were manifest within Sun’s processes, with held the S&P 500 during this period would have seen his or her investment almost affect most people, be they managers or not. the four biases being excessive optimism, overconfidence, confirmation bias, and double. An investor who held Sun stock during this period would have seen his or ExcessiveRecall from Optimism Chapter 1 that a bias is a predisposition toward error. This section the illusion of control. her stock increase by more than fourteen fold. No firm the size of Sun has histori- discussesWhen people whether exhibit four excessive specific optimism, biases were they manifest overestimate within how Sun’s frequently processes, they with will cally merited a price-to-earnings ratio (P/E) over 100. In March 2000, at the height theexperienceExcessive four biases favorable Optimism being outcomesexcessive andoptimism, underestimate overconfidence, how frequently confirmation they will bias, experi and- of the bubble, Sun’s P/E reached 119, reflecting the sentiment of the time.3 ence unfavorable outcomes. theWhen illusion people of exhibit control. excessive optimism, they overestimate how frequently they will It seems plausible that the spectacular rise in Sun’s stock price encouraged some 36 experienceExcessive favorable Optimism outcomes and underestimate how frequently they will experi- of the excessive optimism experienced by Sun’s managers. In this text, considerable ence unfavorable outcomes. When people exhibit excessive optimism, they overestimate how frequently they will 36 experience favorable outcomes and underestimate how frequently they will experi- ence unfavorable outcomes. 36 she77208_ch02_032-057.indd 36 2/7/17 6:20 PM

she77208_ch02_032-057.indd 37 2/7/17 6:20 PM she77208_ch02_032-057.indd 36 2/7/17 6:20 PM

she77208_ch02_032-057.indd 36 2/7/17 6:20 PM 36 Chapter Two bchba_tt Behavioral Pitfalls: Scott McNealy and Sun Microsystems Introduction to Behavioral Analysis 37 36 Chapter Two bchba_tt Behavioral Pitfalls: Scott McNealy and Sun Microsystems Delayed Cost Cutting and Value Loss The cover story in the July 26, 2004, issue of Business- new projects. In justifying that decision, McNealy stated: There is evidence suggesting that when press coverage describes an executive as “op- 36week Chapter is about Two Scott McNealy, the chief executive officer of “The Internet is still wildly underhyped, underutilized, and Introduction to Behavioral Analysis 37 bchba_tt timistic,” that the executive being discussed is excessively optimistic. In this regard, high-technology firm Sun Microsystems. Sun Scott is known McNealy as underimplemented. and Sun Microsystems I think we’re looking at the largest TheBehavioral cover story in the July Pitfalls: 26, 2004, issue of Business- new projects. In justifying that decision, McNealy stated: the Businessweek magazine article specifically included “optimistic” among the list a leading manufacturer of servers and for having invented equipment business in the history of anything. The growth Delayed Cost Cutting and Value Loss theweek Internet is about software Scott McNealy, programming the chief language executive Java. officer The of opportunities“The Internet areis still stunning.” wildly underhyped, underutilized, and of attributes to describe Sun’s CEO Scott McNealy. Therefore, a natural question to Businessweekhigh-technology article firm Sunuses Microsystems. the following Sunadjectives is known to de- as underimplemented.Cisco Systems is Ithe think leading we’re producer looking of at router the largest prod- ask Thereis whether is evidence excessive suggesting optimism that whenaffected press McNealy’s coverage describesdecisions an as executive a manager. as “opFor- The cover story in the July 26, 2004, issue of Business- new projects. In justifying that decision, McNealy stated: scribea leading Scott manufacturer McNealy: optimistic, of servers smart, and foracerbic, having cocky, invented and uctsequipment used onbusiness the Internet. in the history On March of anything. 8, 2001, The Cisco growth an- example,timistic,” is there that evidencethe executive that beingexcessive discussed optimism is excessively led him to optimistic. delay cost-cutting In this regard, mea- week is about Scott McNealy, the chief executive officer of “The Internet is still wildly underhyped, underutilized, and combative.the Internet These software are all programming psychological language traits that influenced Java. The nouncedopportunities that arebecause stunning.” the economic downturn looked like sures, whichBusinessweek resulted in steep losses for Sun? Notably, the “optimistic”general evidence suggests high-technology firm Sun Microsystems. Sun is known as underimplemented. I think we’re looking at the largest the magazine article specifically included among the list McNealy’sBusinessweek business article decisions. uses the following adjectives to de- it wouldCisco last Systems much longeris the leadingthan expected, producer it was of router going toprod- lay a leading manufacturer of servers and for having invented equipment business in the history of anything. The growth thatof people attributes who to report describe that Sun’seconomic CEO conditions Scott McNealy. will improve Therefore, are a twice natural as questionoptimistic to scribeMcNealy Scott McNealy: was one optimistic, of Sun Microsystem’s smart, acerbic, founders, cocky, and be- offucts 18 used percent on the of its Internet. workforce. On MarchSome of 8, Sun’s 2001, executives Cisco an- 2 the Internet software programming language Java. The opportunities are stunning.” as thoseask is who whether report excessive that economic optimism conditions affected willMcNealy’s either stay decisions the same as a or manager. get worse. For camecombative. its CEO These in 1984, are all and psychological had a history traits of being that influenced willing to wantednounced to that follow because suit. Onethe economic stated: “When downturn times looked are hard, like Businessweek article uses the following adjectives to de- Cisco Systems is the leading producer of router prod- Wasexample, McNealy is there excessively evidence thatoptimistic excessive when optimism he predicted led him that to delay the recession cost-cutting of 2001 mea- takeMcNealy’s major businessrisks in order decisions. to avoid disaster. During the 1980s, you’veit would got last tomuch shoot longer activities than expected, that aren’t it makingwas going money.” to lay scribe Scott McNealy: optimistic, smart, acerbic, cocky, and ucts used on the Internet. On March 8, 2001, Cisco an- wouldsures, be whichbrief, resultedalbeit with in steep extreme losses changes for Sun? (sharpNotably, edges)? the general The evidence National suggestsBureau againstMcNealy the advice was one of his of Sunexecutives, Microsystem’s he decided founders, to substi- be- Chiefoff 18 Operatingpercent of Officer its workforce. Ed Zander Some proposed of Sun’s major executives cuts. combative. These are all psychological traits that influenced nounced that because the economic downturn looked like of Economicthat people Research who report (NBER) that economic is officially conditions responsible will improve for determining are twice as when optimistic eco- tutecame Sun’s its CEO own in microprocessors 1984, and had a for history those of manufactured being willing byto However,wanted to McNealy follow suit. refused One stated: to do so, “When and times Zander are subse- hard, McNealy’s business decisions. it would last much longer than expected, it was going to lay 2 Motorola.take major That risks decision in order turnedto avoid out disaster. well for During Sun. the 1980s, quentlyyou’ve gotresigned. to shoot activities that aren’t making money.” nomicas those recessions who report begin that and economic end in the conditions United willStates. either According stay the sameto the or NBER, get worse. the McNealy was one of Sun Microsystem’s founders, be- off 18 percent of its workforce. Some of Sun’s executives againstDuring the the advice 1990s, of his Sun’s executives, competitors he decided produced to serverssubsti- ChiefIn an Operating effort to cut Officer costs, Ed Sun’s Zander customers proposed sought major low-end cuts. U.S. economyWas McNealy entered excessively a recession optimistic in March when 2001,he predicted and the that recession the recession lasted of 2001 until came its CEO in 1984, and had a history of being willing to wanted to follow suit. One stated: “When times are hard, thattute usedSun’s Microsoft’sown microprocessors Windows operating for those systems. manufactured McNealy by servers.However, McNealy McNealy initially refused dismissed to do so, their and concerns. Zander subse- Later, Novemberwould be2001, brief, a periodalbeit withof nine extreme months changes that was (sharp neither edges)? brief Thenor atypical. National Bureau take major risks in order to avoid disaster. During the 1980s, you’ve got to shoot activities that aren’t making money.” insteadMotorola. chose That todecision invest turnedin servers out thatwell ranfor Sun.Sun’s own soft- Sunquently spent resigned. $2 billion to acquire Cobalt, a manufacturer of Theof Economic slowdown Research was longer (NBER) for is officially firms engaging responsible in businessfor determining investment. when eco Sun- against the advice of his executives, he decided to substi- Chief Operating Officer Ed Zander proposed major cuts. wareDuring Solaris. the That 1990s, decision Sun’s also competitors turned out produced well, as serversSolaris low-costIn an effortservers. to cutHowever, costs, Sun’s after customersthe acquisition, sought Sun low-end chose Microsystems’nomic recessions customers begin are and other end firms. in the When United those States. firms According purchase Sun’sto the servers, NBER, they the tute Sun’s own microprocessors for those manufactured by However, McNealy refused to do so, and Zander subse- camethat used to be Microsoft’s highly praised Windows for its operating speed, reliability, systems. and McNealy secu- toservers. limit Cobalt’s McNealy budget, initially and dismissed McNealy later their admitted concerns. that Later, the Motorola. That decision turned out well for Sun. quently resigned. are U.S.engaging economy in business entered investment. a recession During in March the 2001 2001, recession, and the recessioninvestment lasted fell for until six rity.instead Sun’s chose sales, to profits, invest and in serverscash holdings that ran soared, Sun’s andown it softdra- acquisitionSun spent $2was billion a mistake. to acquire Cobalt, a manufacturer of During the 1990s, Sun’s competitors produced servers In an effort to cut costs, Sun’s customers sought low-end consecutiveNovember quarters, 2001, atwice period as oflong nine as monthsthe period that real was gross neither domestic brief nor product atypical. (GDP) fell. maticallyware Solaris. increased That decision its spending also turnedon a great out well,many as research Solaris low-costThe economic servers. However,downturn afterwas muchthe acquisition, longer than Sun McNealy chose that used Microsoft’s Windows operating systems. McNealy servers. McNealy initially dismissed their concerns. Later, ScottThe McNealy’s slowdown prediction was longer that for the firms recession engaging of 2001 in businesswould be investment. brief was ex Sun- andcame development to be highly praised projects. for Duringits speed, the reliability, technology and secu stock- hadto limit forecast. Cobalt’s In budget,the next and three McNealy years, later Sun admittedlost a third that of the its instead chose to invest in servers that ran Sun’s own soft- Sun spent $2 billion to acquire Cobalt, a manufacturer of cessively Microsystems’ optimistic. customers Was his are prediction other firms. necessarily When those biased?firms purchase In other Sun’s words, servers, might they marketrity. Sun’s bubble sales, profits, in the lateand cash 1990s, holdings which soared, peaked and in Marchit dra- marketacquisition share, was its a sales mistake. fell by 48 percent, and its stock price ware Solaris. That decision also turned out well, as Solaris low-cost servers. However, after the acquisition, Sun chose Scottare McNealy’sengaging in predictionbusiness investment. about the During recession the 2001 of 2001recession, have investment been reasonable, fell for six 2000,matically Sun’s increased price-to-earnings its spending ratio on (P/E) a great soared many to 119.research fell Thefrom economic $64 in 2000 downturn to about was $4much in 2004.longer Itsthan net McNealy income came to be highly praised for its speed, reliability, and secu- to limit Cobalt’s budget, and McNealy later admitted that the consecutive quarters, twice as long as the period real gross domestic product (GDP) fell. andSun’s development fortunes changed projects. with During the onset the technology of an economic stock turnedhad forecast. negative In inthe 2002, next withthree a years,loss of Sun $587 lost million. a third Losses of its even though it was wrong after the fact? rity. Sun’s sales, profits, and cash holdings soared, and it dra- acquisition was a mistake. recessionmarket bubble in 2001. in theWall late Street 1990s, analysts which called peaked for Sun in Marchto cut inmarket 2003 share, and 2004 its sales totalled fell by $3.4 48 billionpercent, and and $388 its stock million price re- BetweenScott McNealy’s World War prediction II and 2000, that the recession average lengthof 2001 of would a U.S. be recession brief was wasex- matically increased its spending on a great many research The economic downturn was much longer than McNealy costs.2000, However,Sun’s price-to-earnings McNealy was ratio optimistic (P/E) soared that the to recession 119. spectively,fell from $64 the inresult 2000 of toboth about reduced $4 in demand 2004. Itsfor netits comput income- 11.6 cessively months. optimistic. During those Was his recessions, prediction real necessarily gross domestic biased? investmentIn other words, tended might to and development projects. During the technology stock had forecast. In the next three years, Sun lost a third of its wouldSun’s be fortunesshort-lived. changed In 2001, with during the onseta conference of an economic call with ersturned from negative major corporate in 2002, customerswith a loss and of $587 greater million. competition Losses fall Scottfor 12 McNealy’s months. The prediction recession about before the 2001 recession occurred of 2001in 1990–1991 have been and reasonable, featured market bubble in the late 1990s, which peaked in March market share, its sales fell by 48 percent, and its stock price analysts,recession hein 2001. said aboutWall Street business analysts cycles: called “We for don’t Sun to have cut fromin 2003 IBM, and Hewlett-Packard 2004 totalled (HP),$3.4 andbillion Dell. and $388 million re- twoeven consecutive though it quarters was wrong of negativeafter the fact?growth in real GDP. During the recession of 2000, Sun’s price-to-earnings ratio (P/E) soared to 119. fell from $64 in 2000 to about $4 in 2004. Its net income rollingcosts. However, waves. We McNealy seem was to have optimistic real edges.” that the Insteadrecession of spectively, the result of both reduced demand for its comput- Between World War II and 2000, the average length of a U.S. recession was Sun’s fortunes changed with the onset of an economic turnedSource: negative J. Kerstetter in 2002,and P. Burrows,with a loss “Sun: of A$587 CEO’s million. Last Stand,” Losses 1990–1991, real gross domestic investment fell for 12 months. cuttingwould be costs short-lived. during the In 2001,recession, during Sun a conference invested heavily call with in ersBusinessweek, from major July corporate 26, 2004, customers Bloomberg. and greater competition recession in 2001. Wall Street analysts called for Sun to cut in 2003 and 2004 totalled $3.4 billion and $388 million re- What11.6 months.are we to During conclude? those There recessions, is nothing real gross in the domestic historical investment record to tended suggest to analysts, he said about business cycles: “We don’t have from IBM, Hewlett-Packard (HP), and Dell. fall for 12 months. The recession before 2001 occurred in 1990–1991 and featured rollingcosts. However, waves. We McNealy seem was to have optimistic real edges.” that the Insteadrecession of spectively, the result of both reduced demand for its comput- that recessions would be brief affairs. The more reasonable conclusion might be Source: J. Kerstetter and P. Burrows, “Sun: A CEO’s Last Stand,” two consecutive quarters of negative growth in real GDP. During the recession of cuttingwould be costs short-lived. during the In 2001,recession, during Sun a conference invested heavily call with in ersBusinessweek, from major July corporate 26, 2004, customers Bloomberg. and greater competition that Scott McNealy’s prediction was biased in the direction of optimism and that his analysts, he said about business cycles: “We don’t have from IBM, Hewlett-Packard (HP), and Dell. optimism1990–1991, led him real to gross delay domestic cost cutting, investment thereby fell destroying for 12 months. value at his firm. rolling waves. We seem to havepsychological real edges.” traits Instead described of affectSource: managers J. Kerstetter at and all P. levels Burrows, of “Sun: the firm.A CEO’s Indeed, Last Stand,” they What are we to conclude? There is nothing in the historical record to suggest cutting costs during the recession,affect Sunmost invested people, heavily be they in managersBusinessweek, or not. July 26, 2004, Bloomberg. Sentiment:that recessions Excessively would Optimisticbe brief affairs. Sun TheStockholders more reasonable conclusion might be Recall from Chapter 1 that a bias is a predisposition toward error. This section psychological traits described affect managers at all levels of the firm. Indeed, they Duringthat Scott the stock McNealy’s market prediction bubble betweenwas biased January in the direction 1997 and of Juneoptimism 2000, and irrational that his discusses whether four specific biases were manifest within Sun’s processes, with affect most people, be they managers or not. exuberanceoptimism drove led him up tothe delay prices cost of cutting, both the thereby S&P destroying500 and Sun’s value stock. at his firm.Exhibit 2-2 the four biases being excessive optimism, overconfidence, confirmation bias, and Recall from Chapter 1 that a bias is a predisposition toward error. This section displays the market value of Sun’s equity between 1986 and 2003. An investor who psychologicalthe illusion of traitscontrol. described affect managers at all levels of the firm. Indeed, they Sentiment: Excessively Optimistic Sun Stockholders discusses whether four specific biases were manifest within Sun’s processes, with held the S&P 500 during this period would have seen his or her investment almost affect most people, be they managers or not. During the stock market bubble between January 1997 and June 2000, irrational the four biases being excessive optimism, overconfidence, confirmation bias, and double. An investor who held Sun stock during this period would have seen his or ExcessiveRecall from Optimism Chapter 1 that a bias is a predisposition toward error. This section exuberance drove up the prices of both the S&P 500 and Sun’s stock. Exhibit 2-2 the illusion of control. her stock increase by more than fourteen fold. No firm the size of Sun has histori- discussesWhen people whether exhibit four excessive specific optimism, biases were they manifest overestimate within how Sun’s frequently processes, they with will callydisplays merited the a marketprice-to-earnings value of Sun’s ratio equity (P/E) between over 100. 1986 In Marchand 2003. 2000, An atinvestor the height who theexperienceExcessive four biases favorable Optimism being outcomesexcessive andoptimism, underestimate overconfidence, how frequently confirmation they will bias, experi and- of theheld bubble, the S&P Sun’s 500 P/E during reached this period 119, reflecting would have the seen sentiment his or ofher the investment time.3 almost ence unfavorable outcomes. double. An investor who held Sun stock during this period would have seen his or Whenthe illusion people of exhibit control. excessive optimism, they overestimate how frequently they will It seems plausible that the spectacular rise in Sun’s stock price encouraged some her stock increase by more than fourteen fold. No firm the size of Sun has histori- 36 experience favorable outcomes and underestimate how frequently they will experi- of the excessive optimism experienced by Sun’s managers. In this text, considerable Excessive Optimism cally merited a price-to-earnings ratio (P/E) over 100. In March 2000, at the height ence unfavorable outcomes. When people exhibit excessive optimism, they overestimate how frequently they will of the bubble, Sun’s P/E reached 119, reflecting the sentiment of the time.3 36 experience favorable outcomes and underestimate how frequently they will experi- It seems plausible that the spectacular rise in Sun’s stock price encouraged some ence unfavorable outcomes. of the excessive optimism experienced by Sun’s managers. In this text, considerable 36 she77208_ch02_032-057.indd 36 2/7/17 6:20 PM she77208_ch02_032-057.indd 37 2/7/17 6:20 PM she77208_ch02_032-057.indd 36 2/7/17 6:20 PM

she77208_ch02_032-057.indd 36 2/7/17 6:20she77208_ch02_032-057.indd PM 37 2/7/17 6:20 PM 38 Chapter Two Introduction to Behavioral Analysis 39

EXHIBIT 2-2 $250,000,000 EXHIBIT 2-3 3% Sun Microsystems’ GDP Growth Rates 38Market Chapter Capitalization Two During U.S. Recessions 2% 1986–2003 $200,000,000 (Inflation Adjusted) EXHIBITSource: Center 2-2 for Research in $250,000,000 Source: Federal Reserve Bank of 1% SunSecurity Microsystems’ Prices. St. Louis, Economic Data (FRED), $150,000,000 research.stlouisfed .org/fred2 Market Capitalization /categories/22. 0% 1986–2003 $200,000,000 $100,000,000 Source: Center for Research in Dollars ($000) Security Prices. −1% $150,000,000

$50,000,000 −2% $100,000,000 Dollars ($000) −3% $0 −3 −2 −1 012345678 $50,000,000 Quarter May-86May-87May-88May-89May-90May-91May-92May-93May-94May-95May-96May-97May-98May-99May-00May-01May-02May-03 Date 1974 1980 1982 1990 2001 $0 space is devoted to explaining why blindly trusting market prices can lead even the best-intentionedMay-86May-87May-88May-89 managersMay-90May-91May-92 to May-93 makeMay-94 May-95 faultyMay-96 May-97 decisionsMay-98May-99 aboutMay-00May-01 investmentMay-02May-03 policy, Overconfidence about Knowledge financing, and acquisitions. Date Was Scott McNealy overconfident about his knowledge of U.S. business cycles? spaceOverconfidence is devoted to explaining why blindly trusting market prices can lead even He was certainly confident that the 2001 recession would be sharp-edged, by which the In general, best-intentioned overconfidence managers relates to make to difficult faulty decisions tasks and about abilities. investment Overconfidence policy, is he meant that it would feature a sharp downturn followed by a sharp upturn. Was it? financing,a bias that and pertains acquisitions. to how well people understand their own abilities and the limits Exhibit 2-3 contrasts the manner in which GDP growth behaved in the quarters of their knowledge. People who are overconfident about their abilities think they just before, during, and after a recession for all recessions between 1974 and 2001. Overconfidenceare better than they actually are. People who are overconfident about their level of Quarter 0 marks the onset of recession. The heavy line in the exhibit depicts the Inknowledge general, overconfidence think they know relates more to difficult than they tasks actually and abilities. know. Overconfidence This overconfidence is recession of 2001. You can decide for yourself whether the recession in 2001 was a doesbias thatnot pertainsnecessarily to how mean well that people these understand people are their ignorant own abilities or incompetent. and the limits It just markedly brief or sharp-edged, relative to its predecessors. ofmeans their knowledge.that in their People own eyes who they are areoverconfident smarter and about better their than abilities is actually think the they case. are better than they actually are. People who are overconfident about their level of Overconfidence about Ability knowledge think they know more than they actually know. This overconfidence CONCEPT Five cards are placed in front of you as shown. All cards are either green-backed or red- doesAmong not necessarilythe attributes mean used that in thesethe Businessweek people are ignorant article toor describeincompetent. Scott It McNealy just PREVIEW backed. Cards 1 and 5 are face down, and the other three are face up. cocky smart. meansare that andin their own The eyes appearance they are smarter of these and adjectives better than naturally is actually leads the to case. the ques- Question 2.1 tion of whether Scott McNealy is overconfident, as cockiness is a symptom of over- Card 1 Card 2 Card 3 Card 4 Card 5 Overconfidence about Ability confidence. Overconfident people can certainly be smart, just not quite as smart as Amongthey think the attributesthey are. usedMoreover, in the peopleBusinessweek can learn article to be to overconfident,describe Scott McNealyas a result of cocky smart.4 arepast successes. and McNealy The appearance had achieved of these major adjectives successes naturally in the leads past, to particularly the ques- in ♥ tion of whether Scott McNealy is overconfident, as cockiness is a symptom of over- his decision to support Sun’s own software over Microsoft’s software. Red ☺ ☺ Four Green confidence.Overconfident Overconfident managers people tend canto make certainly poor be decisions smart, just about not quiteboth investmentsas smart as and they think they are. Moreover, people can learn to be overconfident, as a result of Back Joker Joker of Back mergers and acquisitions,4 especially if their firms are cash-rich. Sun’s increased spend- pasting successes.on research McNealyand development had achieved in 2000 major and itssuccesses acquisition in the of Cobaltpast, particularly are cases in in point. his decision to support Sun’s own software over Microsoft’s software. Hearts Both resulted in dramatic reductions in Sun’s market value. In March 2004, analysts is- ♥ suedOverconfident very negative managers reports tendabout to Sun. make McNealy poor decisions responded about by bothpointing investments out that andthe firm mergerswas hardly and acquisitions,likely to become especially distressed, if their given firms that are cash-rich.it held more Sun’s than increased $5 billion spend in -cash. ing on research and development in 2000 and its acquisition of Cobalt are cases in point. Both resulted in dramatic reductions in Sun’s market value. In March 2004, analysts is- sued very negative reports about Sun. McNealy responded by pointing out that the firm was hardly likely to become distressed, given that it held more than $5 billion in cash.

she77208_ch02_032-057.indd 38 she77208_ch02_032-057.indd2/7/17 6:20 PM 39 2/7/17 6:20 PM

she77208_ch02_032-057.indd 38 2/7/17 6:20 PM 38 Chapter Two Introduction to Behavioral Analysis 39

EXHIBIT 2-2 $250,000,000 EXHIBIT 2-3 3% Sun Microsystems’ GDP Growth Rates Market Capitalization During U.S. Recessions 2% Introduction to Behavioral Analysis 39 1986–2003 $200,000,000 (Inflation Adjusted) 3% Source: Center for Research in Source:EXHIBIT Federal Reserve 2-3 Bank of 1% Security Prices. St. Louis,GDP Economic Growth Data (FRED),Rates $150,000,000 research.stlouisfed .org/fred2 During U.S. Recessions 2% /categories/22. 0% (Inflation Adjusted) $100,000,000 Dollars ($000) Source: Federal Reserve Bank of 1% St. Louis, Economic Data (FRED), −1% research.stlouisfed .org/fred2 /categories/22. 0% $50,000,000 −2%

−1% −3% $0 −3 −2 −1 012345678 Quarter May-86May-87May-88May-89May-90May-91May-92May-93May-94May-95May-96May-97May-98May-99May-00May-01May-02May-03 −2% Date 1974 1980 1982 1990 2001 −3% −3 −2 −1 012345678 space is devoted to explaining why blindly trusting market prices can lead even Quarter the best-intentioned managers to make faulty decisions about investment policy, Overconfidence about Knowledge financing, and acquisitions. 1974 1980 1982 1990 2001 Was Scott McNealy overconfident about his knowledge of U.S. business cycles? Overconfidence He was certainly confident that the 2001 recession would be sharp-edged, by which In general, overconfidence relates to difficult tasks and abilities. Overconfidence is he meant that it would feature a sharp downturn followed by a sharp upturn. Was it? Overconfidence about Knowledge a bias that pertains to how well people understand their own abilities and the limits Exhibit 2-3 contrasts the manner in which GDP growth behaved in the quarters of their knowledge. People who are overconfident about their abilities think they justWas before, Scott during, McNealy and afteroverconfident a recession about for hisall recessionsknowledge betweenof U.S. business1974 and cycles? 2001. are better than they actually are. People who are overconfident about their level of QuarterHe was 0 markscertainly the confident onset of that recession. the 2001 The recession heavy wouldline in be the sharp-edged, exhibit depicts by which the knowledge think they know more than they actually know. This overconfidence recessionhe meant of that2001. it wouldYou can feature decide a sharp for yourselfdownturn whether followed the by recession a sharp upturn. in 2001 Was was it? does not necessarily mean that these people are ignorant or incompetent. It just markedlyExhibit brief 2-3 or contrastssharp-edged, the manner relative in to which its predecessors. GDP growth behaved in the quarters means that in their own eyes they are smarter and better than is actually the case. just before, during, and after a recession for all recessions between 1974 and 2001. Quarter 0 marks the onset of recession. The heavy line in the exhibit depicts the Overconfidence about Ability recession of 2001. You can decide for yourself whether the recession in 2001 was CONCEPT Five cards are placed in front of you as shown. All cards are either green-backed or red- markedly brief or sharp-edged, relative to its predecessors. Among the attributes used in the Businessweek article to describe Scott McNealy PREVIEW backed. Cards 1 and 5 are face down, and the other three are face up. cocky smart. are and The appearance of these adjectives naturally leads to the ques- Question 2.1 tion of whether Scott McNealy is overconfident, as cockiness is a symptom of over- Card 1 Card 2 Card 3 Card 4 Card 5 confidence. Overconfident people can certainly be smart, just not quite as smart as CONCEPT Five cards are placed in front of you as shown. All cards are either green-backed or red- they think they are. Moreover, people can learn to be overconfident, as a result of PREVIEW backed. Cards 1 and 5 are face down, and the other three are face up. 4 past successes. McNealy had achieved major successes in the past, particularly in Question 2.1 ♥ Card 1 Card 2 Card 3 Card 4 Card 5 his decision to support Sun’s own software over Microsoft’s software. Red ☺ ☺ Four Green Overconfident managers tend to make poor decisions about both investments and mergers and acquisitions, especially if their firms are cash-rich. Sun’s increased spend- Back Joker Joker of Back ♥ ing on research and development in 2000 and its acquisition of Cobalt are cases in point. Hearts ☺ ☺ Four Both resulted in dramatic reductions in Sun’s market value. In March 2004, analysts is- Red ♥ Green sued very negative reports about Sun. McNealy responded by pointing out that the firm Joker Joker of was hardly likely to become distressed, given that it held more than $5 billion in cash. Back Back Hearts ♥

she77208_ch02_032-057.indd 38 she77208_ch02_032-057.indd2/7/17 6:20 PM 39 2/7/17 6:20 PM

she77208_ch02_032-057.indd 39 2/7/17 6:20 PM 40 Chapter Two Introduction to Behavioral Analysis 41

the gap. In a short meeting, McNealy ordered that Sun would not feature “Intel Inside.” Suppose you are asked to test the following hypothesis about these five cards: “Each card that has a green back on one side has a joker on the other side.” In particular, select Seven years later, Intel chips were twice as fast as those produced by Sun. In retrospect, 40 Chapter Two those cards and only those cards that will determine whether the statement is true. That McNealy describes his decision to not use Intel chips as one of his biggest regrets. is, select the minimum number of cards that will enable you to determine whether or not the statement is true. Of the five cards, which would you turn over? Suppose you are asked to test the following hypothesis about these five cards: “Each 2.4 ANALYZING HEURISTICS card that has a green back on one side has a joker on the other side.” In particular, select those cards and only those cards that will determine whether the statement is true. That is, select the minimum number of cards that will enable you to determine whether or not A heuristic is a rule of thumb. This section discusses representativeness, availability, theConfirmation statement is true. Bias Of the five cards, which would you turn over? anchoring and adjustment, and affect. People who overlook information that disconfirms their views in favor of informa- Representativeness tion that confirms their views are said to exhibit confirmation bias. People exhibiting People often make judgments and predictions by relying on heuristics that make confirmation bias often only hear what they want to hear. They spend too much time Confirmation Bias use of analogues and stereotypes. Psychologists refer to the underlying principle as searching for reasons to support why their views are right and too little time searching representativeness. In asking about the extent to which an object or idea fits a ste- Peoplefor reasons who overlookthat might information lead them thatto conclude disconfirms that theirtheir viewsviews inare favor wrong. of informa- reotype, people are asking how representative that object or idea is for the class to tion Inthat Concept confirms preview their views question are said 2.1, to asking exhibit for confirmation cards displaying bias. People jokers exhibiting to be turned which it belongs. In general, people place too much reliance on representativeness, confirmationover reflects bias confirmation often only hearbias. what The theyprinciple want tois thehear. same They as spend in the too diagnostic much time ques - and representativeness-based thinking can result in bias. searchingtion for confirmationfor reasons to supportbias in Chapterwhy their 1. views Only are the right green and backed too little card time and searching the four of forhearts reasons can that provide might evidence lead them that to concludedisconfirms that thetheir hypothesis. views are wrong. The Internet Represents the Overall Economy In Concept preview question 2.1, asking for cards displaying jokers to be turned Relatively speaking, Internet firms have a short history. During this history, the overTurning reflects a confirmationBlind Eye bias. The principle is the same as in the diagnostic ques- growth rate in sales of the typical or representative Internet firm featured extreme tionIs there for confirmation anything in thebias Businessweek in Chapter 1. magazineOnly the green article backed to suggest card thatand Scottthe four McNealy of movements. Scott McNealy is quoted as having said in early 2001: “The Internet is heartsexhibited can provideconfirmation evidence bias? that The disconfirms article mentions the hypothesis. that in late 2000, executives at Sun still wildly underhyped, underutilized, and underimplemented. I think we’re looking learned that the revenues of industry leader Cisco Systems were declining dramatically at the largest equipment business in the history of anything. The growth opportuni- Turningand began a Blindto suggest Eye that a cost-cutting program be put in place at Sun. In March 2001, ties are stunning.” IsCisco there Systemsanything laidin the off Businessweek 18 percent of magazine its workforce. article However,to suggest thisthat informationScott McNealy did not The Businessweek article reported that Scott McNealy believed that the Internet exhibitedconfirm confirmationMcNealy’s view bias? about The article recessions mentions being that short. in late Despite 2000, theexecutives recommendations at Sun had fundamentally changed the U.S. economy and that the Internet was critical to learnedof his upper-levelthat the revenues executives, of industry McNealy leader refusedCisco Systems to approve were any declining cost cutting dramatically at Sun. a great many firms. Is there reason to believe that Scott McNealy relied on repre- and began to suggest that a cost-cutting program be put in place at Sun. In March 2001, sentativeness? Representativeness-based reasoning might have led him to conclude CiscoIllusion Systems of laidControl off 18 percent of its workforce. However, this information did not that because of the growing importance of the Internet in the economy, and be- confirm McNealy’s view about recessions being short. Despite the recommendations cause Internet firms experience brief sharp swings in business conditions, the U.S. ofWhen his upper-level a person makesexecutives, a decision, McNealy the refused outcome to approve typically any depends cost cutting on aat combination Sun. economy as a whole would experience brief sharp swings rather than rolling waves. of luck and skill. Those who have an exaggerated view of how much control they Illusionexert over of outcomes Control exhibit the illusion of control. As was discussed in Chapter 1, Availability psychological studies have found that an increase in perceived control leads to an People exhibit the availability heuristic when they rely on information that is more Whenincrease a person in excessive makes a optimism. decision, the outcome typically depends on a combination of luck and skill. Those who have an exaggerated view of how much control they readily available than information that is less readily available. exertNot overMade outcomes Here exhibit the illusion of control. As was discussed in Chapter 1, Out of Sight, Out of Mind psychological studies have found that an increase in perceived control leads to an increaseDid Scott in excessiveMcNealy optimism.exhibit the illusion of control? Although not mentioned earlier, Sun played a principal role in a large law suit against rival firm Microsoft that the Businessweek article describes a decision that Sun’s managers had to make in received national attention. The dispute had extended out several years and had become quite personal between Scott McNealy and Microsoft’s founder Bill Gates. Not1997: Made whether Here to use their own microchips for Sun servers or to use Intel’s chips. In 1997, Sun could purchase Intel’s chips for 30 percent less than what it cost to As such, it was in the forefront of McNealy’s mind and highly salient. Did the dis- Didproduce Scott McNealyits own comparable exhibit the illusionchips. of control? Although not mentioned earlier, pute lead him to exhibit availability bias? Businessweek the Despite the desire article of describessome Sun aexecutives decision thatto buy Sun’s Intel managers chips instead had toof makemaking in their Sun’s upper-level executives communicated their concerns that the Microsoft suit 1997:own, whetherScott McNealy to use their felt thatown Sun’s microchips chip design for Sun group servers exerted or to enough use Intel’s control chips. to close had distracted McNealy from focusing on the needs expressed by Sun’s customers. In 1997, Sun could purchase Intel’s chips for 30 percent less than what it cost to produce its own comparable chips. Despite the desire of some Sun executives to buy Intel chips instead of making their own, Scott McNealy felt that Sun’s chip design group exerted enough control to close

she77208_ch02_032-057.indd 40 she77208_ch02_032-057.indd2/7/17 6:20 PM 41 2/7/17 6:20 PM

she77208_ch02_032-057.indd 40 2/7/17 6:20 PM 40 Chapter Two Introduction to Behavioral Analysis 41

the gap. In a short meeting, McNealy ordered that Sun would not feature “Intel Inside.” Suppose you are asked to test the following hypothesis about these five cards: “Each card that has a green back on one side has a joker on the other side.” In particular, select Seven years later, Intel chips were twice as fast as those produced by Sun. In retrospect, Introduction to Behavioral Analysis 41 those cards and only those cards that will determine whether the statement is true. That McNealy describes his decision to not use Intel chips as one of his biggest regrets. is, select the minimum number of cards that will enable you to determine whether or not the statement is true. Of the five cards, which would you turn over? the gap. In a short meeting, McNealy ordered that Sun would not feature “Intel Inside.” 2.4 ANALYZING HEURISTICSSeven years later, Intel chips were twice as fast as those produced by Sun. In retrospect, McNealy describes his decision to not use Intel chips as one of his biggest regrets. A heuristic is a rule of thumb. This section discusses representativeness, availability, anchoring and adjustment, and affect. Confirmation Bias 2.4 ANALYZING HEURISTICS People who overlook information that disconfirms their views in favor of informa- Representativeness tion that confirms their views are said to exhibit confirmation bias. People exhibiting PeopleA heuristic often make is a rule judgments of thumb. and This predictions section discusses by relying representativeness, on heuristics availability, that make confirmation bias often only hear what they want to hear. They spend too much time use anchoringof analogues and andadjustment, stereotypes. and affect. Psychologists refer to the underlying principle as searching for reasons to support why their views are right and too little time searching representativeness. In asking about the extent to which an object or idea fits a ste- for reasons that might lead them to conclude that their views are wrong. reotype,Representativeness people are asking how representative that object or idea is for the class to In Concept preview question 2.1, asking for cards displaying jokers to be turned whichPeople it belongs. often make In general, judgments people and place predictions too much by reliance relying onon heuristicsrepresentativeness, that make over reflects confirmation bias. The principle is the same as in the diagnostic ques- anduse representativeness-based of analogues and stereotypes. thinking Psychologists can result in refer bias. to the underlying principle as tion for confirmation bias in Chapter 1. Only the green backed card and the four of representativeness. In asking about the extent to which an object or idea fits a ste- hearts can provide evidence that disconfirms the hypothesis. Thereotype, Internet people Represents are asking the how Overall representative Economy that object or idea is for the class to Relativelywhich it speaking,belongs. In Internet general, firms people have place a too short much history. reliance During on representativeness, this history, the Turning a Blind Eye growthand representativeness-basedrate in sales of the typical thinking or representative can result in bias.Internet firm featured extreme Is there anything in the Businessweek magazine article to suggest that Scott McNealy movements. Scott McNealy is quoted as having said in early 2001: “The Internet is The Internet Represents the Overall Economy exhibited confirmation bias? The article mentions that in late 2000, executives at Sun still wildly underhyped, underutilized, and underimplemented. I think we’re looking learned that the revenues of industry leader Cisco Systems were declining dramatically at theRelatively largest equipment speaking, Internetbusiness firms in the have history a short of anything. history. DuringThe growth this history,opportuni the- and began to suggest that a cost-cutting program be put in place at Sun. In March 2001, ties growthare stunning.” rate in sales of the typical or representative Internet firm featured extreme Cisco Systems laid off 18 percent of its workforce. However, this information did not Themovements. Businessweek Scott McNealyarticle reported is quoted that as Scotthaving McNealy said in early believed 2001: that “The the Internet Internet is confirm McNealy’s view about recessions being short. Despite the recommendations hadstill fundamentally wildly underhyped, changed underutilized, the U.S. economy and underimplemented. and that the Internet I think was we’re critical looking to of his upper-level executives, McNealy refused to approve any cost cutting at Sun. a greatat the many largest firms. equipment Is there business reason into thebelieve history that of Scottanything. McNealy The growth relied opportunion repre-- sentativeness?ties are stunning.” Representativeness-based reasoning might have led him to conclude Businessweek Illusion of Control that becauseThe of the growing article importance reported that of Scott the InternetMcNealy in believed the economy, that the andInternet be- causehad Internet fundamentally firms experience changed the brief U.S. sharp economy swings and in that business the Internet conditions, was critical the U.S. to When a person makes a decision, the outcome typically depends on a combination economya great as many a whole firms. would Is there experience reason tobrief believe sharp that swings Scott rather McNealy than relied rolling on waves. repre- of luck and skill. Those who have an exaggerated view of how much control they sentativeness? Representativeness-based reasoning might have led him to conclude exert over outcomes exhibit the illusion of control. As was discussed in Chapter 1, Availabilitythat because of the growing importance of the Internet in the economy, and be- psychological studies have found that an increase in perceived control leads to an Peoplecause exhibit Internet the firms availability experience heuristic brief sharpwhen swingsthey rely in businesson information conditions, that theis more U.S. increase in excessive optimism. readilyeconomy available as a wholethan information would experience that is briefless readilysharp swings available. rather than rolling waves. Not Made Here OutAvailability of Sight, Out of Mind Did Scott McNealy exhibit the illusion of control? Although not mentioned earlier, SunPeople played exhibit a principal the availability role in aheuristic large law when suit they against rely on rival information firm Microsoft that is more that the Businessweek article describes a decision that Sun’s managers had to make in receivedreadily national available attention. than information The dispute that is had less extended readily available. out several years and had 1997: whether to use their own microchips for Sun servers or to use Intel’s chips. becomeOut ofquite Sight, personal Out of between Mind Scott McNealy and Microsoft’s founder Bill Gates. In 1997, Sun could purchase Intel’s chips for 30 percent less than what it cost to As such, it was in the forefront of McNealy’s mind and highly salient. Did the dis- produce its own comparable chips. puteSun lead played him to a exhibit principal availability role in a bias? large law suit against rival firm Microsoft that Despite the desire of some Sun executives to buy Intel chips instead of making their Sun’sreceived upper-level national executives attention. Thecommunicated dispute had their extended concerns out that several the Microsoft years and suit had own, Scott McNealy felt that Sun’s chip design group exerted enough control to close had becomedistracted quite McNealy personal from between focusing Scott onMcNealy the needs and expressedMicrosoft’s by founder Sun’s customers.Bill Gates. As such, it was in the forefront of McNealy’s mind and highly salient. Did the dis- pute lead him to exhibit availability bias? Sun’s upper-level executives communicated their concerns that the Microsoft suit had distracted McNealy from focusing on the needs expressed by Sun’s customers.

she77208_ch02_032-057.indd 40 she77208_ch02_032-057.indd2/7/17 6:20 PM 41 2/7/17 6:20 PM

she77208_ch02_032-057.indd 41 2/7/17 6:20 PM 42 Chapter Two Introduction to Behavioral Analysis 43

Customers themselves echoed this concern. They had been asking for low-end servers greater or equal to zero, the fair value of that acquisition is at least as high as the in order to cut costs during the downturn. With Microsoft on his mind, McNealy paid number of dollars invested making the acquisition. 42 Chapter Two little attention to customers’ requests. In April 2004, Sun and Microsoft announced Lehman explained that the managers at Sun did not say to themselves: “For that Sun had dropped its suit, in exchange for a cash settlement and technology every dollar we invest, we’re going to get at least an NPV of one dollar in return.” Customersagreement. themselves In July 2004, echoed Sun this was concern. offering They low-end had been servers asking and for McNealy low-end serverswas claim - (Lehman likely misspoke here, having used NPV when he meant PV.) He went on to ining order that to Sun cut wascosts more during focused the downturn. than any With of its Microsoft competitors. on his mind, McNealy paid say that instead of relying on DCF, the managers at Sun asked how an “acquisition little attention to customers’ requests. In April 2004, Sun and Microsoft announced will enhance our overall capabilities, and how that enhancement will contribute to thatAnchoring Sun had dropped and Adjustment its suit, in exchange for a cash settlement and technology our overall market value.” agreement.People often In July develop 2004, quickSun was estimates offering low-end by beginning servers with and McNealy an initial was number claim - with Intuition is important, make no mistake about it. Experience is valuable, and ingwhich that Sunthey was are morefamiliar focused and thanthen anyadjusting of its competitors. that number to reflect new information firms pay for experienced managers. Indeed, the emotions that managers feel are a or circumstances. Just as a dropped anchor keeps a boat from drifting too far, the manifestation of their minds making associations with the memories of past expe- Anchoringinitial numbers and with Adjustment which managers begin can serve to anchor their judgments. riences. However, experience is not a substitute for careful analysis. Acquisitions PeopleThe heuristic often develop has come quick to be estimates called anchoring by beginning and with adjustment. an initial The number attendant with bias that feel right might well feature negative NPV. Think back to Sun’s experience whichis known they asare anchoring familiar and bias. then When adjusting forming that judgments,number to reflect people new have information a tendency to with Cobalt. orbecome circumstances. anchored Just on asnumbers a dropped in their anchor heads keeps and a do boat not from make drifting sufficient too adjustmentsfar, the initialrelative numbers to the anchor.with which managers begin can serve to anchor their judgments. The heuristic has come to be called anchoring and adjustment. The attendant bias 2.5 VALUE DESTRUCTION AND SUN’S ENDGAME isAnchored known as anchoringto Growth bias. When forming judgments, people have a tendency to becomeWere Sun’s anchored managers on numbers anchored in their on theirheads past and growthdo not make rates? sufficient During its adjustments most success- Businessweek’s July 2004 cover story on Sun Microsystems described how the relativeful period, to the the anchor. turn of the century, Sun’s earnings growth rate reached 50 percent security analysts who followed Sun reacted when the firm failed to reduce its per quarter, faster than competitors Microsoft, Intel, and Dell. That rate was not sus- costs after the recession of 2001. Their comments, along with media coverage Anchoredtainable on to aGrowth permanent basis. In forming forecasts going forward, the question is such as the Businessweek article, constituted clear warnings that Sun needed to Werehow Sun’sto adjust managers relative anchored to the 50 on percent. their past If Sun’sgrowth executives rates? During became its most anchored success- on the change its business strategy. ful50 period, percent, the then turn even of the if century,they adjusted Sun’s theirearnings forecasts growth of ratefuture reached growth 50 downwards,percent Subsequently, Sun’s losses continued to mount, with net income still negative perthey quarter, would faster be psychologically than competitors disposed Microsoft, to Intel,adjust and insufficiently. Dell. That rate That was is not the sus- nature in 2005 at –$107 million. In February 2006, Michael Lehman rejoined the firm as tainableof anchoring on a permanent bias. In thisbasis. case, In forming anchoring forecasts would going have forward,contributed the questionto excessively is CFO, and analysts expressed the hope that he would institute cost-cutting measures. howoptimistic to adjust forecasts relative toof the growth. 50 percent. If Sun’s executives became anchored on the Two months later, the firm reported a fiscal third-quarter loss of $217 million, which 50 percent, then even if they adjusted their forecasts of future growth downwards, included $87 million in costs associated with two acquisitions and $57 million in theyAffect would Heuristic be psychologically disposed to adjust insufficiently. That is the nature costs from stock-based compensation. Shortly thereafter, Scott McNealy resigned as ofMost anchoring people bias. base Intheir this decisions case, anchoring on what would feels right have to contributed them emotionally. to excessively Psycholo- CEO and was replaced by Sun’s chief operating officer Jonathan Schwartz. Notably, optimisticgists use forecaststhe technical of growth. term affect to mean emotional feeling, and they use the term McNealy remained as chairman and also stayed on as an employee of the firm. In taking over as CEO, Schwartz announced that he wanted to hold the line on Affectaffect heuristicHeuristic to describe behavior that places heavy reliance on intuition or “gut feeling.” As with other heuristics, the affect heuristic involves mental shortcuts that costs and focus aggressively on growth opportunities rather than reduce headcount. Mostcan predisposepeople base managers their decisions to bias. on what feels right to them emotionally. Psycholo- Nevertheless, in June 2006 Sun announced that over the subsequent six months it gists use the technical term affect to mean emotional feeling, and they use the term would lay off 4,000 to 5,000 employees, comprising 11 to 13 percent of its work affectAcquisitions heuristic toThat describe Feel behaviorRight but that Destroy places Valueheavy reliance on intuition or “gut force. Schwartz then began to reshape Sun’s strategy by shifting its focus from hard- feeling.”Michael As Lehman with other joined heuristics, Sun’s board the affect of directors heuristic in involves 2002. Before mental thatshortcuts he was that Sun’s ware to software, while maintaining control of all the major aspects of its technol- canchief predispose financial managers officer (CFO).to bias. In 2000, Lehman described Sun’s decision process ogy, both hardware and software. for making acquisitions. He stated: “Now, in determining the price we are willing Exhibit 2-4 displays the trajectory of Sun’s sales and net income for the pe- Acquisitions That Feel Right but Destroy Value to pay for such acquisitions, we are not nearly as formal as the corporate finance riod 1995 through 2009. In 2007 and 2008, Sun’s net income turned positive at Michaeltextbooks Lehman suggest joined we Sun’sperhaps board ought of directorsto be. Our in 2002.approach Before to acquisitionthat he was Sun’spricing is $473 million and $403 million, respectively. However, by June 2009 its net income chiefmore financial intuitive.” officer5 (CFO). In 2000, Lehman described Sun’s decision process was again negative, at –$2.23 billion, and Sun announced that it planned to reduce for makingRemember acquisitions. that intuition He stated: is based “Now, on affect.in determining In contrast, the pricethe formal we are textbook willing ap- the number of its employees by 6,000, from 33,000. A month earlier, in May 2009, toproach pay for to such corporate acquisitions, decision we making are not nearlyis based as onformal the netas the present corporate value finance (NPV) of Sun agreed to be acquired by software firm Oracle for $7.4 billion. Oracle had a textbooksthe acquisition suggest being we perhaps positive, ought or at to worst be. Our zero. approach When the to acquisitionNPV of an pricingacquisition is is history of making major reductions in headcount at firms it acquired, and analysts more intuitive.”5 Remember that intuition is based on affect. In contrast, the formal textbook ap- proach to corporate decision making is based on the net present value (NPV) of the acquisition being positive, or at worst zero. When the NPV of an acquisition is

she77208_ch02_032-057.indd 42 she77208_ch02_032-057.indd2/7/17 6:20 PM 43 2/7/17 6:20 PM

she77208_ch02_032-057.indd 42 2/7/17 6:20 PM 42 Chapter Two Introduction to Behavioral Analysis 43

Customers themselves echoed this concern. They had been asking for low-end servers greater or equal to zero, the fair value of that acquisition is at least as high as the in order to cut costs during the downturn. With Microsoft on his mind, McNealy paid number of dollars invested making the acquisition. little attention to customers’ requests. In April 2004, Sun and Microsoft announced Lehman explained that the managers at Sun didIntroduction not say to Behavioral themselves: Analysis “For 43 that Sun had dropped its suit, in exchange for a cash settlement and technology every dollar we invest, we’re going to get at least an NPV of one dollar in return.” agreement. In July 2004, Sun was offering low-end servers and McNealy was claim- (Lehmangreater likely or equal misspoke to zero, here, the havingfair value used of NPVthat acquisitionwhen he meant is at PV.)least Heas highwent as on the to ing that Sun was more focused than any of its competitors. say numberthat instead of dollars of relying invested on DCF,making the the managers acquisition. at Sun asked how an “acquisition will enhanceLehman our explained overall capabilities, that the managers and how at Sunthat didenhancement not say to will themselves: contribute “For to Anchoring and Adjustment our everyoverall dollar market we value.”invest, we’re going to get at least an NPV of one dollar in return.” People often develop quick estimates by beginning with an initial number with Intuition(Lehman islikely important, misspoke make here, no having mistake used aboutNPV when it. Experience he meant PV.) is valuable, He went onand to which they are familiar and then adjusting that number to reflect new information firmssay pay that for instead experienced of relying managers. on DCF, Indeed,the managers the emotions at Sun asked that managershow an “acquisition feel are a or circumstances. Just as a dropped anchor keeps a boat from drifting too far, the manifestationwill enhance of ourtheir overall minds capabilities, making associations and how that with enhancement the memories will of contribute past expe to- initial numbers with which managers begin can serve to anchor their judgments. riences.our overall However, market experience value.” is not a substitute for careful analysis. Acquisitions The heuristic has come to be called anchoring and adjustment. The attendant bias that feelIntuition right mightis important, well feature make negativeno mistake NPV. about Think it. Experience back to Sun’s is valuable, experience and is known as anchoring bias. When forming judgments, people have a tendency to withfirms Cobalt. pay for experienced managers. Indeed, the emotions that managers feel are a become anchored on numbers in their heads and do not make sufficient adjustments manifestation of their minds making associations with the memories of past expe- relative to the anchor. riences. However, experience is not a substitute for careful analysis. Acquisitions 2.5 VALUE DESTRUCTIONthat feel right ANDmight well SUN’S feature ENDGAMEnegative NPV. Think back to Sun’s experience Anchored to Growth with Cobalt. Were Sun’s managers anchored on their past growth rates? During its most success- Businessweek’s July 2004 cover story on Sun Microsystems described how the ful period, the turn of the century, Sun’s earnings growth rate reached 50 percent security analysts who followed Sun reacted when the firm failed to reduce its per quarter, faster than competitors Microsoft, Intel, and Dell. That rate was not sus- 2.5 VALUE DESTRUCTIONcosts after the recession AND of SUN’S 2001. Their ENDGAME comments, along with media coverage tainable on a permanent basis. In forming forecasts going forward, the question is such as the Businessweek article, constituted clear warnings that Sun needed to how to adjust relative to the 50 percent. If Sun’s executives became anchored on the changeBusinessweek’s its business Julystrategy. 2004 cover story on Sun Microsystems described how the 50 percent, then even if they adjusted their forecasts of future growth downwards, Subsequently, security analysts Sun’s who losses followed continued Sun reacted to mount, when with the netfirm income failed stillto reduce negative its they would be psychologically disposed to adjust insufficiently. That is the nature in 2005costs at after –$107 the million.recession In ofFebruary 2001. Their 2006, comments, Michael Lehman along with rejoined media the coverage firm as of anchoring bias. In this case, anchoring would have contributed to excessively CFO,such and as analysts the Businessweek expressed thearticle, hope constituted that he would clear institute warnings cost-cutting that Sun measures.needed to optimistic forecasts of growth. Twochange months its later, business the firm strategy. reported a fiscal third-quarter loss of $217 million, which includedSubsequently, $87 million Sun’s in costs losses associated continued with to mount,two acquisitions with net income and $57 still million negative in Affect Heuristic costsin from2005 stock-based at –$107 million. compensation. In February Shortly 2006, thereafter, Michael Lehman Scott McNealy rejoined resigned the firm asas Most people base their decisions on what feels right to them emotionally. Psycholo- CEOCFO, and andwas analysts replaced expressed by Sun’s the chief hope operating that he wouldofficer institute Jonathan cost-cutting Schwartz. measures. Notably, gists use the technical term affect to mean emotional feeling, and they use the term McNealyTwo months remained later, as the chairman firm reported and also a fiscal stayed third-quarter on as an employeeloss of $217 of million, the firm. which affect heuristic to describe behavior that places heavy reliance on intuition or “gut Inincluded taking $87over million as CEO, in Schwartzcosts associated announced with twothat acquisitionshe wanted to and hold $57 the million line on in feeling.” As with other heuristics, the affect heuristic involves mental shortcuts that costscosts and from focus stock-based aggressively compensation. on growth opportunitiesShortly thereafter, rather Scott than McNealy reduce headcount.resigned as can predispose managers to bias. Nevertheless,CEO and was in Junereplaced 2006 by SunSun’s announced chief operating that overofficer the Jonathan subsequent Schwartz. six months Notably, it wouldMcNealy lay off remained 4,000 to as 5,000 chairman employees, and also comprising stayed on as 11 an to employee 13 percent of the of firm.its work Acquisitions That Feel Right but Destroy Value force. SchwartzIn taking overthen asbegan CEO, to Schwartzreshape Sun’s announced strategy that by he shifting wanted its to focus hold fromthe line hard- on Michael Lehman joined Sun’s board of directors in 2002. Before that he was Sun’s warecosts to software, and focus while aggressively maintaining on growth control opportunities of all the majorrather thanaspects reduce of its headcount. technol- chief financial officer (CFO). In 2000, Lehman described Sun’s decision process ogy,Nevertheless, both hardware in andJune software. 2006 Sun announced that over the subsequent six months it for making acquisitions. He stated: “Now, in determining the price we are willing Exhibitwould lay 2-4 off displays 4,000 to the 5,000 trajectory employees, of Sun’s comprising sales and11 to net 13 income percent forof its the work pe- to pay for such acquisitions, we are not nearly as formal as the corporate finance riodforce. 1995 Schwartz through then2009. began In 2007 to reshape and 2008,Sun’s strategySun’s net by shiftingincome itsturned focus positive from hard- at textbooks suggest we perhaps ought to be. Our approach to acquisition pricing is $473 millionware to software, and $403 while million, maintaining respectively. control However, of all the by major June aspects 2009 itsof netits technol-income more intuitive.”5 wasogy, again both negative, hardware at and–$2.23 software. billion, and Sun announced that it planned to reduce Remember that intuition is based on affect. In contrast, the formal textbook ap- the numberExhibit of 2-4its employees displays the by trajectory 6,000, from of Sun’s33,000. sales A month and net earlier, income in May for the 2009, pe- proach to corporate decision making is based on the net present value (NPV) of Sunriod agreed 1995 to through be acquired 2009. by In software2007 and firm 2008, Oracle Sun’s for net $7.4 income billion. turned Oracle positive had ata the acquisition being positive, or at worst zero. When the NPV of an acquisition is history$473 million of making and major $403 reductions million, respectively. in headcount However, at firms by it June acquired, 2009 its and net analysts income was again negative, at –$2.23 billion, and Sun announced that it planned to reduce the number of its employees by 6,000, from 33,000. A month earlier, in May 2009, Sun agreed to be acquired by software firm Oracle for $7.4 billion. Oracle had a history of making major reductions in headcount at firms it acquired, and analysts

she77208_ch02_032-057.indd 42 she77208_ch02_032-057.indd2/7/17 6:20 PM 43 2/7/17 6:20 PM

she77208_ch02_032-057.indd 43 2/7/17 6:20 PM Introduction to Behavioral Analysis 45 bchba_tt 44 Chapter Two Behavioral Pitfalls: Judy Lewent and Merck Introduction to Behavioral Analysis 45 bchba_tt EXHIBIT 2-4 $20,000 Behavioral Pitfalls: Judy Lewent and Merck Sun Microsystems Sales Judy Lewent joined Merck in 1980, and became its CFO At the same time, there were dark financial clouds on the hori- 44and ChapterNet Income, Two 10 years later. In doing so she became the first woman in the zon. Pharmaceutical productsIntroduction typically to Behavioral receive patent Analysis protec 45- $15,000 bchba_tt 1995–2009 Sales (Net) BehavioralJudyUnited Lewent States joinedto serve Merck as CFOPitfalls: in 1980,of a major and becamecorporation.Judy its Lewent CFOAnd andAttion the forMerck same a period time, of there 17 years. were By dark the financial end of 2001,clouds all on five the of hori the- 10 yearsMerck was later. undeniably In doing aso major she becamecorporation. the first During woman the 1980sin the zon.major Pharmaceutical drugs mentioned products were duetypically to go receive off patent. patent protec- EXHIBITSource: Compustat. 2-4 $20,000 and 1990s, it was recognized as America’s Most Admired The pharmaceutical industry is also regulated. In the $10,000 United States to serve as CFO of a major corporation. And tion for a period of 17 years. By the end of 2001, all five of the Sun Microsystems Sales Company in Fortune magazine’s annual survey, winning that United States, the Food and Drug Administration (FDA) must and Net Income, JudyMerck Lewent was undeniably joined Merck a major in corporation. 1980, and becameDuring the its 1980s CFO Atmajor the drugssame time,mentioned there werewere darkdue financialto go off cloudspatent. on the hori- $15,000 honor seven times, more frequently than any other firm. approve all new drugs as being safe and effective. To estab- Sales (Net) 10 yearsand 1990s, later. it In was doing recognized so she became as America’s the first Most woman Admired in the zon.The Pharmaceutical pharmaceutical products industry typically is also receive regulated. patent protec In the- 1995–2009 $5,000 In 1972, Lewent completed her Master of Science in lish safety and efficacy, pharmaceutical firms collect data for Net Income (Loss) UnitedCompany States in Fortune to serve magazine’s as CFO of annual a major survey, corporation. winning Andthat tionUnited for States,a period the of Food17 years. and By Drug the Administration end of 2001, all (FDA) five ofmust the Source: Compustat. honorbusiness seven at times, the Sloan more frequently School, Massachusetts than any other firm. Institute approvesubmission all newto the drugs FDA, as using being clinical safe and trials effective. to test new To estab- drugs $10,000 Merck was undeniably a major corporation. During the 1980s major drugs mentioned were due to go off patent. of TechnologyIn 1972, Lewent (MIT), completed studying under her Master some of of Science the major in lishin small safety samples and efficacy, involving pharmaceutical both animals firms and people.collect data for $0 and 1990s, it was recognized as America’s Most Admired The pharmaceutical industry is also regulated. In the businessacademics at in thefinance: Sloan Fischer School, Black, Massachusetts Myron Scholes, Institute and submissionIn May 1999, to the the FDA, FDA using approved clinical Vioxx trials to to treat test pain.new drugsHow- StewartCompany Myers. in Fortune Before magazine’s joining Merck,annual survey, she worked winning as that an ever,United it States,asked Merckthe Food to andconduct Drug aAdministration new large postapproval (FDA) must $5,000 of Technology (MIT), studying under some of the major in small samples involving both animals and people. Net Income (Loss) analysthonor seven at several times, financial more frequently institutions than and any in other 1976 firm. moved trialapprove in order all new to extend drugs as its being study safe of the and side effective. effect profileTo estab- of −$5,000 academics in finance: Fischer Black, Myron Scholes, and In May 1999, the FDA approved Vioxx to treat pain. How- to pharmaceuticalIn 1972, Lewent firm completed Pfizer. her Master of Science in Vioxxlish safety in respect and efficacy, to stomach pharmaceutical ulcers. The firms issue collect was data impor for- businessStewart Myers. at the Before Sloan joining School, Merck, Massachusetts she worked Institute as an submissionever, it asked to theMerck FDA, to usingconduct clinical a new trials large to test postapproval new drugs $0Jan-95 Jan-97 Jan-99 Jan-01 Jan-03 Jan-05 Jan-07 Jan-09 analystIn 1994 at severalLewent financialwas one institutionsof the most and respected in 1976 CFOsmoved in trialtant, in in order that to existing extend painkillers its study of that the were side effect already profile on the of theof Technology United States. (MIT), That studying year the under Harvard some Business of the Review major market,in small samplessuch as aspirininvolving and both naproxen animals (for and example, people. Aleve), academicsto pharmaceutical in finance: firm FischerPfizer. Black, Myron Scholes, and VioxxIn May in respect 1999, the to FDAstomach approved ulcers. Vioxx The to issue treat was pain. impor How- publishedIn 1994 an Lewent interview was with one her, of whichthe most they respected entitled “Scientific CFOs in tant,caused in thatstomach existing irritation. painkillers Merck’s that earlier were studies already showed on the −$5,000 Stewart Myers. Before joining Merck, she worked as an ever, it asked Merck to conduct a new large postapproval Managementthe United States. at Merck: That An year Interview the Harvard with CFO Business Judy Lewent.” Review market,that Vioxx such did as not aspirin irritate and people’s naproxen stomachs. (for example, Aleve), predicted that it would do so at Sun. Most of Sun’s executives, including McNealy However,analyst at in several 2004, CFOfinancial magazine institutions ran an articleand in entitled 1976 moved “What trialBecause in order aspirinto extend reduces its study blood of clotting,the side it effect reduces profile the inof- Jan-95 Jan-97 Jan-99 Jan-01 Jan-03 Jan-05 Jan-07 Jan-09 published an interview with her, which they entitled “Scientific caused stomach irritation. Merck’s earlier studies showed and Schwartz, resigned after the acquisition became final. Willto pharmaceutical Judy Do?” asking firm whether Pfizer. she would be able to keep her cidenceVioxx in ofrespect heart attacksto stomach and strokes.ulcers. TheVioxx issue does was not, impor a fact- ManagementIn 1994 Lewent at Merck: was Anone Interview of the most with CFOrespected Judy Lewent.”CFOs in tant,that Vioxx in that did existing not irritate painkillers people’s that stomachs. were already on the A year after Oracle acquired Sun, Fortune magazine interviewed Scott McNealy, However,job. What inhappened? 2004, CFO magazine ran an article entitled “What of whichBecause Merck’s aspirin scientists reduces and blood executives clotting, wereit reduces well aware.the in- the The United short States. answer That to the year last the question Harvard is that Business in September Review market,Surprisingly, such as Merck’saspirin and postapproval naproxen (for study example, appeared Aleve), to asking him what had gone wrong at Sun during his overall tenure as well as the publishedWill Judy Do?” an interview asking whether with her, she which would they be entitled able to “Scientific keep her causedcidence stomachof heart attacksirritation. and Merck’s strokes. earlier Vioxx studies does not, showed a fact 6 2004 Merck recalled its blockbuster drug Vioxx, a drug that show that Vioxx actually caused heart attacks and strokes. predictedfirm’s final that years it would as an do independent so at Sun. Most company. of Sun’s McNealy executives, joked including that his McNealy interviewer Managementjob. What happened? at Merck: An Interview with CFO Judy Lewent.” thatof which Vioxx Merck’s did not scientists irritate people’s and executives stomachs. were well aware. and Schwartz, resigned after the acquisition became final. hadThe contributed short answer 8.7 percent to the last of the question firm’s globalis that revenuesin September that However,Surprisingly, the firm’s Merck’s executives postapproval resisted that study interpretation appeared and to didn’t have enough video storage to record all his mistakes. He then made two spe- year.However, The inlong 2004, answer CFO magazineis more complicated ran an article and entitled requires “What a investedBecause heavily aspirin in promotingreduces blood the drug.clotting, Then, it reduces in September the in- cificA year points after about Oracle what acquired he might Sun, have Fortune done magazine differently. interviewed Scott McNealy, Will2004 Judy Merck Do?” recalled asking whetherits blockbuster she would drug be Vioxx, able toa drugkeep that her cidenceshow that of Vioxxheart attacks actually and caused strokes. heart Vioxx attacks does and not, strokes. a fact hadshort contributed history of the 8.7 firm.percent of the firm’s global revenues that However,2004 a new the separate firm’s executives Merck study resisted of colon that interpretationpolyps also found and askingFirst, him McNealy what had stated gone thatwrong during at Sun the during technology his overall bubble, tenure Sun’s as stockwell asprice the was job.According What happened? to Fortune magazine’s annual reputation sur- thatof which Vioxx Merck’s caused scientistsheart attacks and andexecutives strokes. 7were Merck’s well manag aware.- 6 year. The long answer is more complicated and requires a invested heavily in promoting the drug. Then, in September firm’sabout final 10 times years itsas anrevenue, independent which company. for a hardware McNealy company joked that was his unsustainable, interviewer he vey,The Merck short was answer one ofto America’sthe last question most admired is that in companies. September ers Surprisingly,belatedly recalled Merck’s the postapprovaldrug and prepared study for appeared the lawsuits to didn’t have enough video storage to record all his mistakes. He then made two spe- 2004short historyMerck ofrecalled the firm. its blockbuster drug Vioxx, a drug that show2004 a that new Vioxx separate actually Merck caused study heartof colon attacks polyps and also strokes. found suggested. Although he pointed out that Sun had monetized the valuation very well, In fact,According for the sevento Fortune straight magazine’s years between annual 1986reputation and 1992, sur- thatto come. Vioxx Between caused heart September attacks 27and and strokes. November7 Merck’s 5, itsmanag stock- cifiche saidpoints that about perhaps what he mightmight have have done tried differently. to talk the analysts into not running up Merckhad contributed ranked as 8.7 America’s percent ofmost the firm’sadmired global company. revenues Merck that However,price fell from the firm’s $44.46 executives to $26.21. resisted that interpretation and year.vey, Merck The long was answerone of America’sis more complicated most admired and companies. requires a investeders belatedly heavily recalled in promoting the drug theand drug. prepared Then, for in theSeptember lawsuits Sun’sFirst, stock McNealy to the stated extent that that during they did.the technology bubble, Sun’s stock price was gained its exalted status by producing a series of block- shortIn fact, history for the of seven the firm. straight years between 1986 and 1992, 2004to come. a new Between separate September Merck study 27 andof colon November polyps 5,also its foundstock buster drugs, such as Vasotec for the treatment of blood A. Mathews and B. Martinez, “E-Mails Suggest Early aboutSecond, 10 times McNealy its revenue, noted which that hefor mighta hardware have hiredcompany a chief was operatingunsustainable, officer he dur- Merck ranked as America’s most admired company. Merck priceSources: fell from $44.46 to $26.21. 7 According to Fortune magazine’s annual reputation sur- thatVioxx Vioxx Worries—As caused Evidence heart attacks of Heart and Risk strokes. Rose, Merck Merck’s Officials manag - suggested.ing his tenure Although as CEO, he pointed rather outthan that undertaking Sun had monetized the responsibility the valuation himself very well,at a time gainedpressure, its Prinivil exalted for status cardiac by medication, producing aPepcid series and of block-Prilo- secvey, forMerck ulcers, was and one Mevacor of America’s to lower most cholesterol. admired companies. ersPlayed belatedly Hardball; recalled One Internal the drug Message: and prepared‘Dodge!’” Thefor theWall lawsuits Street hein saidhis lifethat when perhaps he hadhe might four childrenhave tried with to talkwhom the he analysts also wanted into not to runningspend his up time. buster drugs, such as Vasotec for the treatment of blood Journal,Sources: November A. Mathews 2, and2004; B. Martinez,“Scientific “E-Mails Management Suggest at Merck:Early In fact,The foryear the 1999 seven was straight auspicious years for between Merck. That 1986 year and the 1992, five to come. Between September 27 and November 5, its stock Sun’sHe noted stock that to the he extent did spend that theytime did. with his children and that he was thrilled with how Merckpressure, ranked Prinivil as forAmerica’s cardiac most medication, admired Pepcid company. and Merck Prilo- priceAnVioxx Interview fellWorries—As from with $44.46 CFO Evidence Judy to $26.21. Lewent,”of Heart RiskHarvard Rose, Business Merck Officials Review, secdrugs for mentioned ulcers, and earned Mevacor Merck to lower$4.38 cholesterol.billion in U.S. sales and Jan–Feb,Played Hardball; 1994; Kate One O’Sullivan, Internal Message: “What Will ‘Dodge!’” Judy Do?” The CFO Wall Street thingsSecond, had McNealy turned out noted for himthat personally.he might have hired a chief operating officer dur- gained its exalted status by producing a series of block- Journal, November 2, 2004; “Scientific Management at Merck: royalties,The year at a1999 time was when auspicious its total netfor Merck.sales were That $32.7year the billion. five Magazine, December 3, 2004. ing hisThere tenure are asseveral CEO, issuesrather thanthat undertakingMcNealy might the responsibility have mentioned himself in his at ainterview, time buster drugs, such as Vasotec for the treatment of blood Sources:An Interview A. Mathews with CFO and Judy B. Lewent,”Martinez, Harvard “E-Mails BusinessSuggest EarlyReview, drugs mentioned earned Merck $4.38 billion in U.S. sales and inbut his did life not.when A he discussion had four childrenof these with issues whom appears he also in wantedthe Additional to spend Resourceshis time. to pressure, Prinivil for cardiac medication, Pepcid and Prilo- VioxxJan–Feb, Worries—As 1994; Kate Evidence O’Sullivan, of Heart “What Risk Will Rose, Judy Do?”Merck CFO Officials He noted that he did spend time with his children and that he was thrilled with how secroyalties, for ulcers, at a time and whenMevacor its total to lower net sales cholesterol. were $32.7 billion. PlayedMagazine, Hardball; December One Internal3, 2004. Message: ‘Dodge!’” The Wall Street Chapter 2. Journal, November 2, 2004; “Scientific Management at Merck: The year 1999 was auspicious for Merck. That year the five things had turned out for him personally. CONCEPT Only one in 10,000 chemicals thatAn are Interview investigated with CFO for Judy their Lewent,” drug Harvardpotential Business actually Review, turns There are several issues that McNealy might have mentioned in his interview, drugsPREVIEW mentioned earned Merckout $4.38 to be billion successful in U.S. salesin the and sense ofJan–Feb, becoming 1994; aKate prescription O’Sullivan, “Whatdrug. WillMoreover, Judy Do?” completing CFO royalties, at a time when its total net sales were $32.7 billion. Magazine, December 3, 2004. 2.6 ANALYZINGbut FRAMING did not. A discussion EFFECTS of these issues appears in the Additional Resources to CONCEPTQuestion 2.2 Onlythe process one in 10,000often takes chemicals at least that a decade. are investigated Suppose for that their a pharmaceutical drug potential actuallyfirm explores turns Chapter 2. PREVIEW out1,000 to chemicals,be successful and in that the the sense success of becoming probability a prescription for one chemical drug. Moreover, is independent completing of the others. What do you think the probability is that the firm will successfully produce at least To discuss how framing effects might impact corporate financial decision making, Question 2.2 the process often takes at least a decade. Suppose that a pharmaceutical firm explores 1,000two prescription chemicals, drugs?and that the success probability for one chemical is independent of the 2.6 ANALYZING considerFRAMING the history EFFECTS of Merck, a firm quite different from Sun, which is discussed in CONCEPT Only one in 10,000 chemicals that are investigated for their drug potential actually turns the nearby Behavioral Pitfalls box. Merck is one of the best-known pharmaceutical PREVIEW outothers. to be What successful do you inthink the thesense probability of becoming is that a theprescription firm will successfully drug. Moreover, produce completing at least two prescription drugs? firms in the world. Question 2.2 the process often takes at least a decade. Suppose that a pharmaceutical firm explores45 To discuss how framing effects might impact corporate financial decision making, 1,000 chemicals, and that the success probability for one chemical is independent of the consider the history of Merck, a firm quite different from Sun, which is discussed in others. What do you think the probability is that the firm will successfully produce at least45 the nearby Behavioral Pitfalls box. Merck is one of the best-known pharmaceutical two prescription drugs? firms in the world. 45 she77208_ch02_032-057.indd 45 2/7/17 6:20 PM she77208_ch02_032-057.indd 44 2/7/17 6:20 PM she77208_ch02_032-057.indd 45 2/7/17 6:20 PM

she77208_ch02_032-057.indd 44 2/7/17 6:20she77208_ch02_032-057.indd PM 45 2/7/17 6:20 PM Introduction to Behavioral Analysis 45 bchba_tt 44 Chapter Two Behavioral Pitfalls: Judy Lewent and Merck Introduction to Behavioral Analysis 45 bchba_tt EXHIBIT 2-4 $20,000 Behavioral Pitfalls: Judy Lewent and Merck Sun Microsystems Sales Judy Lewent joined Merck in 1980, and became its CFO At the same time, there were dark financial clouds on the hori- and Net Income, 10 years later. In doing so she became the first woman in the zon. Pharmaceutical productsIntroduction typically to Behavioral receive patent Analysis protec 45- $15,000 bchba_tt 1995–2009 Sales (Net) BehavioralJudyUnited Lewent States joinedto serve Merck as CFOPitfalls: in 1980,of a major and becamecorporation.Judy its Lewent CFOAnd andAttion the forMerck same a period time, of there 17 years. were By dark the financial end of 2001,clouds all on five the of hori the- 10 yearsMerck was later. undeniably In doing aso major she becamecorporation. the first During woman the 1980sin the zon.major Pharmaceutical drugs mentioned products were duetypically to go receive off patent. patent protec- Source: Compustat. and 1990s, it was recognized as America’s Most Admired The pharmaceutical industry is also regulated. In the $10,000 United States to serve as CFO of a major corporation. And tion for a period of 17 years. By the end of 2001, all five of the MerckCompany was in undeniably Fortune magazine’s a major corporation. annual survey, During winning the 1980s that majorUnited drugs States, mentioned the Food wereand Drug due toAdministration go off patent. (FDA) must honorJudy Lewentseven times, joined more Merck frequently in 1980, than and any became other firm. its CFO approveAt the same all newtime, drugs there aswere being dark safe financial and effective. clouds on To the estab- hori- 10 yearsand 1990s, later. it In was doing recognized so she became as America’s the first Most woman Admired in the zon.The Pharmaceutical pharmaceutical products industry typically is also receive regulated. patent protec In the- $5,000 In 1972, Lewent completed her Master of Science in lish safety and efficacy, pharmaceutical firms collect data for Net Income (Loss) Company in Fortune magazine’s annual survey, winning that United States, the Food and Drug Administration (FDA) must businessUnited States at theto serve Sloan as School,CFO of a Massachusetts major corporation. Institute And submissiontion for a period to the of FDA, 17 years. using By clinical the end trials of 2001, to test all new five drugsof the Merckhonor sevenwas undeniably times, more a major frequently corporation. than any During other the firm. 1980s majorapprove drugs all new mentioned drugs as were being due safe to goand off effective. patent. To estab- of TechnologyIn 1972, Lewent (MIT), completed studying under her Master some of of Science the major in lishin small safety samples and efficacy, involving pharmaceutical both animals firms and people.collect data for $0 academicsand 1990s, in it wasfinance: recognized Fischer asBlack, America’s Myron MostScholes, Admired and InThe May pharmaceutical 1999, the FDA approved industry is Vioxx also to regulated. treat pain. In How the- Companybusiness in at Fortune the Sloan magazine’s School, annual Massachusetts survey, winning Institute that Unitedsubmission States, to the FDA,Food usingand Drug clinical Administration trials to test (FDA)new drugs must ofStewart Technology Myers. (MIT), Before studying joining Merck,under some she worked of the asmajor an inever, small it asked samples Merck involving to conduct both animals a new largeand people. postapproval analysthonor seven at several times, financial more frequently institutions than and any in other 1976 firm. moved trialapprove in order all new to extend drugs as its being study safe of the and side effective. effect profileTo estab- of −$5,000 academics in finance: Fischer Black, Myron Scholes, and In May 1999, the FDA approved Vioxx to treat pain. How- to pharmaceuticalIn 1972, Lewent firm completed Pfizer. her Master of Science in Vioxxlish safety in respect and efficacy, to stomach pharmaceutical ulcers. The firms issue collect was data impor for- businessStewart Myers. at the Before Sloan joining School, Merck, Massachusetts she worked Institute as an submissionever, it asked to theMerck FDA, to usingconduct clinical a new trials large to test postapproval new drugs Jan-95 Jan-97 Jan-99 Jan-01 Jan-03 Jan-05 Jan-07 Jan-09 analystIn 1994 at severalLewent financialwas one institutionsof the most and respected in 1976 CFOsmoved in trialtant, in in order that to existing extend painkillers its study of that the were side effect already profile on the of theof Technology United States. (MIT), That studying year the under Harvard some Business of the Review major market,in small samplessuch as aspirininvolving and both naproxen animals (for and example, people. Aleve), academicsto pharmaceutical in finance: firm FischerPfizer. Black, Myron Scholes, and VioxxIn May in respect 1999, the to FDAstomach approved ulcers. Vioxx The to issue treat was pain. impor How- publishedIn 1994 an Lewent interview was with one her, of whichthe most they respected entitled “Scientific CFOs in tant,caused in thatstomach existing irritation. painkillers Merck’s that earlier were studies already showed on the ManagementStewart Myers. at Merck: Before An joining Interview Merck, with sheCFO worked Judy Lewent.” as an thatever, Vioxx it asked did Mercknot irritate to conduct people’s a stomachs.new large postapproval analystthe United at several States. financial That year institutions the Harvard and Business in 1976 moved Review trialmarket, in order such to as extend aspirin its and study naproxen of the (forside example, effect profile Aleve), of predicted that it would do so at Sun. Most of Sun’s executives, including McNealy publishedHowever, inan 2004, interview CFO withmagazine her, which ran an they article entitled entitled “Scientific “What causedBecause stomach aspirin irritation. reduces Merck’s blood clotting, earlier itstudies reduces showed the in- and Schwartz, resigned after the acquisition became final. Willto pharmaceutical Judy Do?” asking firm whether Pfizer. she would be able to keep her cidenceVioxx in ofrespect heart attacksto stomach and strokes.ulcers. TheVioxx issue does was not, impor a fact- ManagementIn 1994 Lewent at Merck: was Anone Interview of the most with CFOrespected Judy Lewent.”CFOs in tant,that Vioxx in that did existing not irritate painkillers people’s that stomachs. were already on the A year after Oracle acquired Sun, Fortune magazine interviewed Scott McNealy, However,job. What inhappened? 2004, CFO magazine ran an article entitled “What of whichBecause Merck’s aspirin scientists reduces and blood executives clotting, wereit reduces well aware.the in- the The United short States. answer That to the year last the question Harvard is that Business in September Review market,Surprisingly, such as Merck’saspirin and postapproval naproxen (for study example, appeared Aleve), to asking him what had gone wrong at Sun during his overall tenure as well as the publishedWill Judy Do?” an interview asking whether with her, she which would they be entitled able to “Scientific keep her causedcidence stomachof heart attacksirritation. and Merck’s strokes. earlier Vioxx studies does not, showed a fact 6 2004 Merck recalled its blockbuster drug Vioxx, a drug that show that Vioxx actually caused heart attacks and strokes. firm’s final years as an independent company. McNealy joked that his interviewer Managementjob. What happened? at Merck: An Interview with CFO Judy Lewent.” thatof which Vioxx Merck’s did not scientists irritate people’s and executives stomachs. were well aware. hadThe contributed short answer 8.7 percent to the last of the question firm’s globalis that revenuesin September that However,Surprisingly, the firm’s Merck’s executives postapproval resisted that study interpretation appeared and to didn’t have enough video storage to record all his mistakes. He then made two spe- year.However, The inlong 2004, answer CFO magazineis more complicated ran an article and entitled requires “What a investedBecause heavily aspirin in promotingreduces blood the drug.clotting, Then, it reduces in September the in- cific points about what he might have done differently. 2004Will Judy Merck Do?” recalled asking whetherits blockbuster she would drug be Vioxx, able toa drugkeep that her cidenceshow that of Vioxxheart attacks actually and caused strokes. heart Vioxx attacks does and not, strokes. a fact hadshort contributed history of the 8.7 firm.percent of the firm’s global revenues that However,2004 a new the separate firm’s executives Merck study resisted of colon that interpretationpolyps also found and First, McNealy stated that during the technology bubble, Sun’s stock price was job. What happened? of which Merck’s scientists and executives 7were well aware. year.According The long to answer Fortune is magazine’smore complicated annual reputationand requires sur- a investedthat Vioxx heavily caused in heart promoting attacks the and drug. strokes. Then, Merck’s in September manag- about 10 times its revenue, which for a hardware company was unsustainable, he vey,The Merck short was answer one ofto America’sthe last question most admired is that in companies. September ers Surprisingly,belatedly recalled Merck’s the postapprovaldrug and prepared study for appeared the lawsuits to 2004short historyMerck ofrecalled the firm. its blockbuster drug Vioxx, a drug that show2004 a that new Vioxx separate actually Merck caused study heartof colon attacks polyps and also strokes. found suggested. Although he pointed out that Sun had monetized the valuation very well, In fact,According for the sevento Fortune straight magazine’s years between annual 1986reputation and 1992, sur- thatto come. Vioxx Between caused heart September attacks 27and and strokes. November7 Merck’s 5, itsmanag stock- he said that perhaps he might have tried to talk the analysts into not running up hadMerck contributed ranked as 8.7 America’s percent ofmost the firm’sadmired global company. revenues Merck that However,price fell from the firm’s $44.46 executives to $26.21. resisted that interpretation and year.vey, Merck The long was answerone of America’sis more complicated most admired and companies. requires a investeders belatedly heavily recalled in promoting the drug theand drug. prepared Then, for in theSeptember lawsuits Sun’s stock to the extent that they did. gained its exalted status by producing a series of block- shortIn fact, history for the of seven the firm. straight years between 1986 and 1992, 2004to come. a new Between separate September Merck study 27 andof colon November polyps 5,also its foundstock buster drugs, such as Vasotec for the treatment of blood A. Mathews and B. Martinez, “E-Mails Suggest Early Second, McNealy noted that he might have hired a chief operating officer dur- Merck ranked as America’s most admired company. Merck priceSources: fell from $44.46 to $26.21. 7 According to Fortune magazine’s annual reputation sur- thatVioxx Vioxx Worries—As caused Evidence heart attacks of Heart and Risk strokes. Rose, Merck Merck’s Officials manag - ing his tenure as CEO, rather than undertaking the responsibility himself at a time gainedpressure, its Prinivil exalted for status cardiac by medication, producing aPepcid series and of block-Prilo- secvey, forMerck ulcers, was and one Mevacor of America’s to lower most cholesterol. admired companies. ersPlayed belatedly Hardball; recalled One Internal the drug Message: and prepared‘Dodge!’” Thefor theWall lawsuits Street in his life when he had four children with whom he also wanted to spend his time. buster drugs, such as Vasotec for the treatment of blood Journal,Sources: November A. Mathews 2, and2004; B. Martinez,“Scientific “E-Mails Management Suggest at Merck:Early In fact,The foryear the 1999 seven was straight auspicious years for between Merck. That 1986 year and the 1992, five to come. Between September 27 and November 5, its stock He noted that he did spend time with his children and that he was thrilled with how Merckpressure, ranked Prinivil as forAmerica’s cardiac most medication, admired Pepcid company. and Merck Prilo- priceAnVioxx Interview fellWorries—As from with $44.46 CFO Evidence Judy to $26.21. Lewent,”of Heart RiskHarvard Rose, Business Merck Officials Review, secdrugs for mentioned ulcers, and earned Mevacor Merck to lower$4.38 cholesterol.billion in U.S. sales and Jan–Feb,Played Hardball; 1994; Kate One O’Sullivan, Internal Message: “What Will ‘Dodge!’” Judy Do?” The CFO Wall Street things had turned out for him personally. gained its exalted status by producing a series of block- Journal, November 2, 2004; “Scientific Management at Merck: royalties,The year at a1999 time was when auspicious its total netfor Merck.sales were That $32.7year the billion. five Magazine, December 3, 2004. There are several issues that McNealy might have mentioned in his interview, buster drugs, such as Vasotec for the treatment of blood Sources:An Interview A. Mathews with CFO and Judy B. Lewent,”Martinez, Harvard “E-Mails BusinessSuggest EarlyReview, drugs mentioned earned Merck $4.38 billion in U.S. sales and but did not. A discussion of these issues appears in the Additional Resources to pressure, Prinivil for cardiac medication, Pepcid and Prilo- VioxxJan–Feb, Worries—As 1994; Kate Evidence O’Sullivan, of Heart “What Risk Will Rose, Judy Do?”Merck CFO Officials secroyalties, for ulcers, at a time and whenMevacor its total to lower net sales cholesterol. were $32.7 billion. PlayedMagazine, Hardball; December One Internal3, 2004. Message: ‘Dodge!’” The Wall Street Chapter 2. Journal, November 2, 2004; “Scientific Management at Merck: The year 1999 was auspicious for Merck. That year the five CONCEPT Only one in 10,000 chemicals thatAn are Interview investigated with CFO for Judy their Lewent,” drug Harvardpotential Business actually Review, turns drugsPREVIEW mentioned earned Merckout $4.38 to be billion successful in U.S. salesin the and sense ofJan–Feb, becoming 1994; aKate prescription O’Sullivan, “Whatdrug. WillMoreover, Judy Do?” completing CFO royalties, at a time when its total net sales were $32.7 billion. Magazine, December 3, 2004. 2.6 ANALYZING FRAMING EFFECTS CONCEPTQuestion 2.2 Onlythe process one in 10,000often takes chemicals at least that a decade. are investigated Suppose for that their a pharmaceutical drug potential actuallyfirm explores turns PREVIEW out1,000 to chemicals,be successful and in that the the sense success of becoming probability a prescription for one chemical drug. Moreover, is independent completing of the To discuss how framing effects might impact corporate financial decision making, theothers. process What often do you takes think at the least probability a decade. is Supposethat the firm that will a pharmaceuticalsuccessfully produce firm explores at least Question 2.2 two prescription drugs? consider the history of Merck, a firm quite different from Sun, which is discussed in CONCEPT Only1,000 one chemicals, in 10,000 and chemicals that the success that are probabilityinvestigated for for one their chemical drug potential is independent actually of turns the the nearby Behavioral Pitfalls box. Merck is one of the best-known pharmaceutical PREVIEW outothers. to be What successful do you inthink the thesense probability of becoming is that a theprescription firm will successfully drug. Moreover, produce completing at least two prescription drugs? firms in the world. Question 2.2 the process often takes at least a decade. Suppose that a pharmaceutical firm explores45 1,000 chemicals, and that the success probability for one chemical is independent of the others. What do you think the probability is that the firm will successfully produce at least45 two prescription drugs?

45 she77208_ch02_032-057.indd 45 2/7/17 6:20 PM she77208_ch02_032-057.indd 44 2/7/17 6:20 PM she77208_ch02_032-057.indd 45 2/7/17 6:20 PM

she77208_ch02_032-057.indd 45 2/7/17 6:20 PM 46 Chapter Two Introduction to Behavioral Analysis 47

EXHIBIT 2-5 2.5 60 The “one in 10,000 probability” in Concept Preview Question 2.2 is a statistic Judy Merck’s Capital Lewent frequently mentioned. Applying the binomial probability distribution to this ques- 46 Chapter Two Structure and ROE tion produces an answer of 9.5 percent. As was mentioned in the Behavioral Pitfalls box 1983–2014 50 about Judy Lewent, one thing that set her apart from other CFOs was her application of 2.0 scientific methodology to make judgments of, and decisions about, risk. In this regard, she Source: Compustat. The “one in 10,000 probability” in Concept Preview Question 2.2 is a statistic Judy developed Monte Carlo simulation models to assess Merck’s long-term risks, and used Lewent frequently mentioned. Applying the binomial probability distribution to this ques- 40 real option techniques to manage those risks. Although these techniques are described in tion produces an answer of 9.5 percent. As was mentioned in the Behavioral Pitfalls box corporate finance textbooks, they are not well used in practice. Lewent’s model became 1.5 about Judy Lewent, one thing that set her apart from other CFOs was her application of the basis of Merck’s research planning framework, seeking to capture all its risks and scientific methodology to make judgments of, and decisions about, risk. In this regard, she long-term development timetables to show how they play out over a 20-year time horizon. 30 developed Monte Carlo simulation models to assess Merck’s long-term risks, and used In theory, decisions about capital structure—the mix of debt and equity on a firm’s real option techniques to manage those risks. Although these techniques are described in 1.0 balance sheet—reflect judgments about risk. In respect to capital structure, Lewent corporate finance textbooks, they are not well used in practice. Lewent’s model became 20 thestated basis the of following Merck’s research in 2012: “This planning is certainly framework, true seeking in the research to capture pharmaceutical all its risks and indus- long-termtry, which development carries by definition timetables significant to show how risk. they And play when out overthe risksa 20-year become time realhorizon. and not justIn theoretical,theory, decisions that’s aboutwhen capitalthe value structure—the of a prudent mix capital of debt structure and equity becomes on a evident.firm’s A 0.5 capital structure that enables the company to continue to pay dividends and fund R&D 10 balance sheet—reflect judgments about risk. In respect to capital structure, Lewent8 statedcan mean the following the difference in 2012: between “This is certainly weathering true thein the storm research or sinking pharmaceutical into oblivion.” indus- In re- try,gard which to dividends, carries by indefinition 2006 she significant stated: risk.“In the And fourth when quarter the risks of become2004, we real had and a callnot with 0 justanalysts. theoretical, I won’t that’s say when the most the valueimportant, of a prudent but one capital of the structure most important becomes questions evident. Aasked 0.0 9 capitalwas, ‘Is structure the dividend that enables secure?’ the And company I could to [answer continue with] to pay an unequivocaldividends and ‘yes.’” fund R&D 1982 1985 1988 1991 1994 1997 2000 2003 2006 2009 2012 can meanWith the the precedingdifference commentsbetween weathering as a backdrop, the storm consider or sinking two framinginto oblivion.” issues8 Inat re-Merck, Total Assets/Common Equity Total Debt/Total Assets gardone to relating dividends, to its in decisions 2006 she about stated: capital “In the structure fourth quarter and the of other2004, relating we had toa callits decisions with analysts.about Vioxx. I won’t Given say the Lewent’s most important, remarks, butit is oneclear of that the shemost perceives important these questions issues asked as being Return on Common Equity Times Interest Earned was,interconnected. ‘Is the dividend secure?’ And I could [answer with] an unequivocal ‘yes.’”9 With the preceding comments as a backdrop, consider two framing issues at Merck, stemming from excessive debt. Firms having large amounts of tangible assets that one relating to its decisions about capital structure and the other relating to its decisions can be sold to service debt, if need be, have lower expected distress costs than firms aboutLoss Vioxx. Aversion Given Lewent’s remarks, it is clear that she perceives these issues as being whose assets are mostly intangible. Of course, Merck would have significant intan- interconnected. Loss aversion leads people to behave in risk-averse fashion when facing alternatives gible assets stemming from research and development. Those assets largely reside in that feature the possibility of both gain and loss. the brains of its scientists and managers. As a result, Merck’s executives would have to LossTraditional Aversion textbooks in corporate finance teach that debt can be used to shield be concerned that the firm’s scientists and managers might react to financial distress investors from corporate taxes. Yet many firms appear to take on less debt than text- by leaving the firm, taking with them a large portion of the firm’s intangible assets. Lossbook aversion theory leadssuggests, people thereby to behave leaving in risk-averse tax shield fashion dollars when on the facing table. alternatives10 A concern about losing intangible assets might reasonably lead the firm to shy away that Considerfeature the Merck’s possibility debt of policy.both gain Has and Merck loss. been averse to debt? Exhibit 2-5 dis- from taking on too much debt. playsTraditional the time textbooks paths of in three corporate variables finance between teach that1983 debt and can 2014: be usedMerck’s to shield leverage Such a concern is not hypothetical. After Merck withdrew Vioxx from the mar- investorsmeasured from as totalcorporate assets taxes. to common Yet many equity, firms its appear times to interest take on earned,less debt its than total text- liabili- ket, the price of its stock fell by more than a third over the next two months. That 10 bookties totheory total suggests,assets, and thereby its return leaving on commontax shield equity. dollars All on ofthe these table. variables are based drop made Merck a strong takeover candidate, and its directors began to worry onConsider book value, Merck’s not marketdebt policy. values. Has Merck been averse to debt? Exhibit 2-5 dis- about talented managers leaving the firm. In response, Merck filed a new compen- playsExhibit the time 2-5 paths shows of threethat Merck’s variables total between debt 1983has typically and 2014: been Merck’s below leverage 50 percent sation plan for top executives with the Securities and Exchange Commission. The measuredof its assets as total during assets the to time common period equity, displayed. its times Financial interest textbooksearned, its suggesttotal liabili- that de- new plan featured one-time severance payments of up to triple executives’ salary tiescisions to total about assets, debt and be its made return using on common market equity.values Allrather of these than variables debt values. are based Consider and bonus, along with other perquisites like health benefits, should Merck be ac- onthe book period value, before not market2004 when values. Merck withdrew Vioxx from the market. During this quired and the executives lose their jobs. time,Exhibit the 2-5market shows value that of Merck’s Merck’s total equity debt averaged has typically six timesbeen belowthe book 50 percentvalue of its This change in compensation suggests that there are alternatives to addressing ofequity its assets during during the period.the time Therefore, period displayed. in market Financial terms Merck’s textbooks debt suggest was a thatmuch de- lower brain drain besides the assumption of lower debt. Moreover, it is important to ask cisionsfraction about of its debt assets be madethan shownusing marketin Exhibit values 2-5. rather than debt values. Consider whether in the late 1980s and throughout the 1990s the risk of distress was reason- the periodIn the beforetraditional 2004 trade-off when Merck theory withdrew approach Vioxx to debt from policy, the market. managers During balance this the able or instead was overblown. Between 1988 and 2001, Merck had a high tax bill, time,benefits the marketof additional value taxof Merck’s shields againstequity averagedthe costs sixof possibletimes the future book financial value of distressits held a large amount in cash, and had several blockbuster drugs. Its high cash holdings equity during the period. Therefore, in market terms Merck’s debt was a much lower fraction of its assets than shown in Exhibit 2-5. In the traditional trade-off theory approach to debt policy, managers balance the benefits of additional tax shields against the costs of possible future financial distress

she77208_ch02_032-057.indd 46 she77208_ch02_032-057.indd2/7/17 6:20 PM 47 2/7/17 6:20 PM

she77208_ch02_032-057.indd 46 2/7/17 6:20 PM 46 Chapter Two Introduction to Behavioral Analysis 47

EXHIBIT 2-5 2.5 60 The “one in 10,000 probability” in Concept Preview Question 2.2 is a statistic Judy Merck’s Capital Lewent frequently mentioned. Applying the binomial probability distribution to this ques- Structure and ROE Introduction to Behavioral Analysis 47 tion produces an answer of 9.5 percent. As was mentioned in the Behavioral Pitfalls box 1983–2014 50 about Judy Lewent, one thing that set her apart from other CFOs was her application of 2.0 scientific methodology to make judgments of, and decisions about, risk. In this regard, she Source:EXHIBIT Compustat. 2-5 2.5 60 developed Monte Carlo simulation models to assess Merck’s long-term risks, and used Merck’s Capital 40 real option techniques to manage those risks. Although these techniques are described in Structure and ROE corporate finance textbooks, they are not well used in practice. Lewent’s model became 1983–2014 1.5 50 2.0 the basis of Merck’s research planning framework, seeking to capture all its risks and Source: Compustat. long-term development timetables to show how they play out over a 20-year time horizon. 30 In theory, decisions about capital structure—the mix of debt and equity on a firm’s 40 1.0 balance sheet—reflect judgments about risk. In respect to capital structure, Lewent 1.5 stated the following in 2012: “This is certainly true in the research pharmaceutical indus- 20 try, which carries by definition significant risk. And when the risks become real and not 30 just theoretical, that’s when the value of a prudent capital structure becomes evident. A 0.5 capital structure that enables the company to continue to pay dividends and fund R&D 1.0 10 can mean the difference between weathering the storm or sinking into oblivion.”8 In re- 20 gard to dividends, in 2006 she stated: “In the fourth quarter of 2004, we had a call with analysts. I won’t say the most important, but one of the most important questions asked 0.0 0 9 0.5 was, ‘Is the dividend secure?’ And I could [answer with] an unequivocal ‘yes.’” 1982 1985 1988 1991 1994 1997 2000 2003 2006 2009 2012 10 With the preceding comments as a backdrop, consider two framing issues at Merck, one relating to its decisions about capital structure and the other relating to its decisions Total Assets/Common Equity Total Debt/Total Assets about Vioxx. Given Lewent’s remarks, it is clear that she perceives these issues as being 0.0 Return on Common Equity Times Interest Earned 0 interconnected. 1982 1985 1988 1991 1994 1997 2000 2003 2006 2009 2012 stemming from excessive debt. Firms having large amounts of tangible assets that Total Assets/Common Equity Total Debt/Total Assets can be sold to service debt, if need be, have lower expected distress costs than firms Return on Common Equity Times Interest Earned Loss Aversion whose assets are mostly intangible. Of course, Merck would have significant intan- Loss aversion leads people to behave in risk-averse fashion when facing alternatives gible assets stemming from research and development. Those assets largely reside in that feature the possibility of both gain and loss. the brainsstemming of its from scientists excessive and managers.debt. Firms As having a result, large Merck’s amounts executives of tangible would assets have that to Traditional textbooks in corporate finance teach that debt can be used to shield be concernedcan be sold that to servicethe firm’s debt, scientists if need be, and have managers lower expectedmight react distress to financial costs than distress firms investors from corporate taxes. Yet many firms appear to take on less debt than text- by leavingwhose assetsthe firm, are mostlytaking intangible.with them Ofa large course, portion Merck of wouldthe firm’s have intangiblesignificant assets. intan- book theory suggests, thereby leaving tax shield dollars on the table.10 A concerngible assets about stemming losing intangible from research assets and might development. reasonably Those lead assets the firm largely to shy reside away in Consider Merck’s debt policy. Has Merck been averse to debt? Exhibit 2-5 dis- fromthe taking brains on of tooits scientists much debt. and managers. As a result, Merck’s executives would have to plays the time paths of three variables between 1983 and 2014: Merck’s leverage Suchbe concerned a concern that is thenot firm’s hypothetical. scientists After and managers Merck withdrew might react Vioxx to financial from the distress mar- measured as total assets to common equity, its times interest earned, its total liabili- ket,by the leaving price ofthe its firm, stock taking fell withby more them than a large a third portion over of the the next firm’s two intangible months. assets. That ties to total assets, and its return on common equity. All of these variables are based dropA madeconcern Merck about a losing strong intangible takeover assets candidate, might reasonably and its directors lead the beganfirm to toshy worry away on book value, not market values. aboutfrom talented taking managerson too much leaving debt. the firm. In response, Merck filed a new compen- Exhibit 2-5 shows that Merck’s total debt has typically been below 50 percent sation Suchplan afor concern top executives is not hypothetical. with the Securities After Merck and withdrew Exchange Vioxx Commission. from the mar-The of its assets during the time period displayed. Financial textbooks suggest that de- newket, plan the featured price of one-timeits stock fellseverance by more payments than a third of upover to the triple next executives’ two months. salary That cisions about debt be made using market values rather than debt values. Consider anddrop bonus, made along Merck with a other strong perquisites takeover candidate,like health and benefits, its directors should began Merck to be worry ac- the period before 2004 when Merck withdrew Vioxx from the market. During this quiredabout and talented the executives managers lose leaving their thejobs. firm. In response, Merck filed a new compen- time, the market value of Merck’s equity averaged six times the book value of its Thissation change plan for in topcompensation executives withsuggests the Securities that there and are Exchange alternatives Commission. to addressing The equity during the period. Therefore, in market terms Merck’s debt was a much lower brainnew drain plan besides featured the one-time assumption severance of lower payments debt. Moreover,of up to triple it is executives’important tosalary ask fraction of its assets than shown in Exhibit 2-5. whetherand bonus, in the alonglate 1980s with andother throughout perquisites the like 1990s health the benefits, risk of shoulddistress Merck was reason be ac-- In the traditional trade-off theory approach to debt policy, managers balance the ablequired or instead and the was executives overblown. lose Between their jobs. 1988 and 2001, Merck had a high tax bill, benefits of additional tax shields against the costs of possible future financial distress held a Thislarge changeamount in in compensationcash, and had suggestsseveral blockbuster that there aredrugs. alternatives Its high cashto addressing holdings brain drain besides the assumption of lower debt. Moreover, it is important to ask whether in the late 1980s and throughout the 1990s the risk of distress was reason- able or instead was overblown. Between 1988 and 2001, Merck had a high tax bill, held a large amount in cash, and had several blockbuster drugs. Its high cash holdings

she77208_ch02_032-057.indd 46 she77208_ch02_032-057.indd2/7/17 6:20 PM 47 2/7/17 6:20 PM

she77208_ch02_032-057.indd 47 2/7/17 6:20 PM 48 Chapter Two Introduction to Behavioral Analysis 49

(10 to 20 percent of sales) and blockbuster drugs with many years remaining of pat- accustomed. It is plausible to suggest that if the firm’s profits declined, these execu- ent protection provided for low risk as far as bankruptcy was concerned. tives would have difficulty adjusting psychologically from their days of past glory. As 48 Chapter Two Several editions of the corporate finance textbook co authored by Judy Lewent’s a result, they might have used past profitability as an aspirational reference point and corporate finance professor Stewart Myers, along with Richard Brealey and Franklin framed the lower forecasted profit situation as a loss relative to that reference point. (10Allen, to 20 analyze percent Merck’sof sales) decisionand blockbuster about capitaldrugs with structure. many years The tenthremaining edition of patfocuses- After the FDA approved Vioxx in May 1999, Merck’s executives were hoping enton protection the year 2008,provided when for lowMerck’s risk as market-to-book far as bankruptcy ratio was (for concerned. equity) was about 3.6. that its sales would help them to avoid entering the domain of psychological losses, TheSeveral authors editions examine of the the corporate impact finance of a $1 textbook billion co dollar authored recapitalization by Judy Lewent’s in which meaning lower profits. However, the results from its postapproval clinical study, corporateMerck borrows finance professor$1 billion Stewart and uses Myers, the proceedsalong with to Richard repurchase Brealey shares. and Franklin They argue code named VIGOR (for Vioxx Gastrointestinal Outcomes Research), indicating Allen,that doing analyze so Merck’sraises Merck’s decision intrinsic about capital value structure. by $350 Themillion, tenth andedition conclude: focuses “It’s that Vioxx might actually cause heart attacks and strokes, presented the executives onhard the toyear believe 2008, that when Merck’s Merck’s financial market-to-book managers ratio are simply(for equity) missing was the about boat.” 3.6.11 with a dilemma involving two alternatives. The By authors taking examine on more the debt impact in the of alate $1 1980s billion and dollar early recapitalization 1990s, Merck in could which have The first alternative, called second line therapy, was to promote Vioxx for a small Merckreduced borrows a tax $1bill billion that amounted and uses the to proceedsroughly 60to repurchasepercent of shares.its cash They holdings. argue As market, namely people who needed pain relief, whose sensitive stomachs did not that Exhibit 2-5 doing so shows, raises MerckMerck’s generally intrinsic maintained value by $350 its leverage million, between and conclude: 1982 and “It’s 1997, tolerate drugs like aspirin and naproxen, and who were not prone to heart problems. hardeven to as believe its return that onMerck’s equity financial rose and managers stabilized. are simply missing the boat.”11 This alternative was akin to accepting a sure loss (meaning, lower profits). The sec- ByBrealey, taking Myers,on more and debt Allen in the examine late 1980s two andpossible early explanations 1990s, Merck for could Merck’s have con- ond alternative, called first line therapy, was to try and beat the odds, hoping that the reducedservative a taxdebt bill policy, that amountedthe first involving to roughly the 60 combination percent of itsof corporatecash holdings. and personal As negative findings from VIGOR would not carry over to the general population. The Exhibit 2-5taxes, and shows, the second Merck involving generally bankruptcymaintained its costs. leverage They between find that 1982 the and inclusion 1997, of second alternative involved promoting Vioxx as a general painkiller for the whole evenpersonal as its returntaxes doeson equity reduce rose the and advantage stabilized. to debt somewhat, but does not elimi- market, downplaying the cardiac side effect profile, and hoping for blockbuster sales. nateBrealey, it. As Myers, for bankruptcy, and Allen theyexamine state two the possiblefollowing explanations about Merck: for Merck’s“It could con- borrow People who view themselves in the domain of losses are much more prone to servativeenough todebt save policy, tens ofthe millions first involving of dollars the withoutcombination raising of acorporate whisker andof concern personal about accept risks than people who view themselves in the domain of gains. Because of taxes,possible and financialthe second distress involving . . . bankruptcyMerck illustrates costs. Theyan odd find fact that about the inclusionreal-life capitalof the looming patent expiration on its older drugs, Merck executives felt the pressure personalstructures. taxes The does most reduce profitable the advantage companies to borrowdebt somewhat, the least.” but12 does not elimi- to make Vioxx a major success. They chose the risky alternative. nateThis it. As last for statement bankruptcy, naturally they state leads the to thefollowing following about question: Merck: Is “It the could underlying borrow issue Researchers at , , and the Cleveland Clinic enoughpsychological to save tens and ofrelated millions to referenceof dollars pointswithout and raising loss aaversion? whisker ofTo concern answer about this ques - wrote scientific articles that raised concerns about Vioxx’s safety. Merck challenged possibletion, suppose financial that distress the operative . . . Merck reference illustrates point foran odda firm fact is aspirationalabout real-life in thecapital sense of these concerns and continued to promote Vioxx as safe. In February 2001, the FDA structures.the March-Shapira The most modelprofitable described companies in Chapter borrow 1, the with least.” the 12firm perceiving its current issued a letter to Merck’s CEO at the time, Ray Gilmartin, chastising the firm for positionThis last as statement a success, naturally relative toleads its aspiration. to the following In this question: case, loss Is aversion the underlying becomes issue central deceptive promotional practices. In August 2004, a researcher from the FDA’s drug- psychologicaland tends to andinduce related a low-risk/low to reference leveragepoints and strategy. loss aversion? However, To ifanswer the operative this ques -refer- safety office presented data that showed that higher doses of Vioxx correlated with tion,ence suppose point for that a thefirm operative is aspirational, reference but point the for firm a firm views is aspirational achieving aspirationin the sense as of chal - a tripled risk of a heart attack or sudden cardiac death. Merck responded by issuing thelenging, March-Shapira then risk model seeking described in the domainin Chapter of losses1, with rather the firm than perceiving loss aversion its current becomes a press release reiterating its confidence in the safety and efficacy of its drug. positioncentral, as in a whichsuccess, case relative the firmto its wouldaspiration. choose In this a high-risk/high case, loss aversion leverage becomes strategy. central Between 2000 and 2004, twenty-eight million people took Vioxx. After Merck and Ittends is also to induce possible a low-risk/low that Lewent’s leverage models strategy. showed However, the risks if ofthe financial operative distress refer- to finally withdrew Vioxx from the market, the British medical journal The Lancet encebe considerably point for a firm higher is aspirational, than those outsidebut the firmthe firm views perceived achieving them aspiration to be. asEven chal if- that wrote that the firm should have made that decision several years earlier. In December lenging,were the then case, risk that seeking would in thenot domain explain of the losses general rather pattern than loss that aversion the most becomes profitable 2004, the best estimates were that the lawsuits against Merck would total $18 billion. central,companies in which borrow case the the least, firm wouldwhereas choose a reference a high-risk/high point-based leverage argument strategy. would. In February 2005, by a vote of 17 to 15, members of an FDA advisory panel rec- It is also possible that Lewent’s models showed the risks of financial distress to ommended that Vioxx was safe enough to be sold in the United States, as long as its beAversion considerably to higher a Sure than Loss those outside the firm perceived them to be. Even if that package label carried strict warnings about the risk of heart attacks and strokes. The wereRecall the from case, Chapter that would 1 that not aversion explain tothe a generalsure loss pattern can lead that people the most to chooseprofitable a risky vote led Merck to inform the FDA that it was considering requesting permission to companiesalternative borrow instead the of least, accepting whereas a sure a reference loss, even point-based when the argument odds are would.against them. In reintroduce Vioxx. However, in the end it chose not to do so. Aversiondoing so, they to a choose Sure riskLoss hoping to beat the odds, in an effort to break even. In 1999, the executives at Merck were watching the clock tick down, as five of their Lawsuits and Aftermath Recallmost successfulfrom Chapter products 1 that wereaversion due to goa sure off patentloss can in lead2000 people and 2001. to choose One possible a risky sce- In August 2005, a Texas jury rendered its verdict on the first of the Vioxx trials, find- alternativenario was insteadthat the of firm accepting would bea sure unable loss, to even find when new productsthe odds toare replace against the them. lost In profits ing Merck liable for the death of 59-year-old marathon runner Robert Ernst who had doingfrom so,the theyones choose going offrisk patent. hoping Merck to beat continued the odds, toin bean oneeffort of toAmerica’s break even. most admired taken the drug. The jury award of $253 million jolted Merck executives, and Merck’s companiesIn 1999, the throughout executives the at Merck1990s, werea state watching of affairs the to clock which tick its down, executives as five hadof their become market capitalization fell by 7.7 percent after the announcement of the jury’s verdict. most successful products were due to go off patent in 2000 and 2001. One possible sce- nario was that the firm would be unable to find new products to replace the lost profits from the ones going off patent. Merck continued to be one of America’s most admired companies throughout the 1990s, a state of affairs to which its executives had become

she77208_ch02_032-057.indd 48 she77208_ch02_032-057.indd2/7/17 6:20 PM 49 2/7/17 6:20 PM

she77208_ch02_032-057.indd 48 2/7/17 6:20 PM 48 Chapter Two Introduction to Behavioral Analysis 49

(10 to 20 percent of sales) and blockbuster drugs with many years remaining of pat- accustomed. It is plausible to suggest that if the firm’s profits declined, these execu- ent protection provided for low risk as far as bankruptcy was concerned. tives would have difficulty adjusting psychologically from their days of past glory. As Several editions of the corporate finance textbook co authored by Judy Lewent’s a result, they might have used past profitability as an aspirationalIntroduction to Behavioralreference Analysis point and49 corporate finance professor Stewart Myers, along with Richard Brealey and Franklin framed the lower forecasted profit situation as a loss relative to that reference point. Allen, analyze Merck’s decision about capital structure. The tenth edition focuses Afteraccustomed. the FDA It isapproved plausible Vioxx to suggest in May that 1999,if the firm’sMerck’s profits executives declined, were these hoping execu- on the year 2008, when Merck’s market-to-book ratio (for equity) was about 3.6. thattives its sales would would have helpdifficulty them adjusting to avoid psychologicallyentering the domain from their of psychological days of past glory. losses, As The authors examine the impact of a $1 billion dollar recapitalization in which meaninga result, lower they profits.might have However, used past the profitability results from as anits aspirational postapproval reference clinical point study, and Merck borrows $1 billion and uses the proceeds to repurchase shares. They argue codeframed named the VIGOR lower forecasted (for Vioxx profit Gastrointestinal situation as a loss Outcomes relative to Research), that reference indicating point. that doing so raises Merck’s intrinsic value by $350 million, and conclude: “It’s that VioxxAfter might the FDA actually approved cause Vioxx heart inattacks May 1999,and strokes, Merck’s presented executives the were executives hoping hard to believe that Merck’s financial managers are simply missing the boat.”11 withthat a dilemma its sales wouldinvolving help two them alternatives. to avoid entering the domain of psychological losses, By taking on more debt in the late 1980s and early 1990s, Merck could have Themeaning first alternative,lower profits. called However, second the line results therapy, from was its to postapproval promote Vioxx clinical for a study,small reduced a tax bill that amounted to roughly 60 percent of its cash holdings. As market,code namelynamed VIGORpeople who(for Vioxxneeded Gastrointestinal pain relief, whose Outcomes sensitive Research), stomachs indicating did not Exhibit 2-5 shows, Merck generally maintained its leverage between 1982 and 1997, toleratethat Vioxxdrugs likemight aspirin actually and cause naproxen, heart attacksand who and were strokes, not prone presented to heart the executivesproblems. even as its return on equity rose and stabilized. Thiswith alternative a dilemma was involving akin to accepting two alternatives. a sure loss (meaning, lower profits). The sec- Brealey, Myers, and Allen examine two possible explanations for Merck’s con- ond alternative,The first alternative, called first called line therapy, second linewas therapy,to try and was beat to promotethe odds, Vioxx hoping for that a small the servative debt policy, the first involving the combination of corporate and personal negativemarket, findings namely from people VIGOR who neededwould notpain carry relief, over whose to the sensitive general stomachs population. did The not taxes, and the second involving bankruptcy costs. They find that the inclusion of secondtolerate alternative drugs like involved aspirin promotingand naproxen, Vioxx and aswho a generalwere not painkiller prone to heart for the problems. whole personal taxes does reduce the advantage to debt somewhat, but does not elimi- market,This downplayingalternative was the akin cardiac to accepting side effect a sure profile, loss (meaning,and hoping lower for blockbuster profits). The sales. sec- nate it. As for bankruptcy, they state the following about Merck: “It could borrow Peopleond alternative, who view called themselves first line intherapy, the domain was to tryof lossesand beat are the much odds, more hoping prone that theto enough to save tens of millions of dollars without raising a whisker of concern about acceptnegative risks findingsthan people from who VIGOR view would themselves not carry in overthe domainto the general of gains. population. Because The of possible financial distress . . . Merck illustrates an odd fact about real-life capital the secondlooming alternative patent expiration involved onpromoting its older Vioxx drugs, as Merck a general executives painkiller felt forthe the pressure whole structures. The most profitable companies borrow the least.”12 to makemarket, Vioxx downplaying a major success.the cardiac They side chose effect theprofile, risky and alternative. hoping for blockbuster sales. This last statement naturally leads to the following question: Is the underlying issue ResearchersPeople who at Stanford view themselves University, in theHarvard domain University, of losses andare themuch Cleveland more prone Clinic to psychological and related to reference points and loss aversion? To answer this ques- wroteaccept scientific risks thanarticles people that whoraised view concerns themselves about in Vioxx’s the domain safety. of Merck gains. challengedBecause of tion, suppose that the operative reference point for a firm is aspirational in the sense of thesethe concerns looming andpatent continued expiration to promoteon its older Vioxx drugs, as Mercksafe. In executives February felt2001, the thepressure FDA the March-Shapira model described in Chapter 1, with the firm perceiving its current issuedto make a letter Vioxx to Merck’s a major success.CEO at theThey time, chose Ray the Gilmartin, risky alternative. chastising the firm for position as a success, relative to its aspiration. In this case, loss aversion becomes central deceptiveResearchers promotional at Stanford practices. University, In August Harvard 2004, University, a researcher and from the theCleveland FDA’s Clinicdrug- and tends to induce a low-risk/low leverage strategy. However, if the operative refer- safetywrote office scientific presented articles data that that raised showed concerns that higher about Vioxx’sdoses of safety. Vioxx Merck correlated challenged with ence point for a firm is aspirational, but the firm views achieving aspiration as chal- a tripledthese riskconcerns of a heartand continued attack or tosudden promote cardiac Vioxx death. as safe. Merck In February responded 2001, by the issuing FDA lenging, then risk seeking in the domain of losses rather than loss aversion becomes a pressissued release a letter reiterating to Merck’s its CEOconfidence at the intime, the Ray safety Gilmartin, and efficacy chastising of its thedrug. firm for central, in which case the firm would choose a high-risk/high leverage strategy. Betweendeceptive 2000 promotional and 2004, practices. twenty-eight In August million 2004, people a researcher took fromVioxx. the After FDA’s Merck drug- It is also possible that Lewent’s models showed the risks of financial distress to finallysafety withdrew office presented Vioxx from data that the market,showed that the higher British doses medical of Vioxx journal correlated The Lancet with be considerably higher than those outside the firm perceived them to be. Even if that wrotea tripled that the risk firm of shoulda heart haveattack made or sudden that decision cardiac death.several Merck years respondedearlier. In Decemberby issuing were the case, that would not explain the general pattern that the most profitable 2004,a press the best release estimates reiterating were its that confidence the lawsuits in the against safety Merck and efficacy would totalof its $18 drug. billion. companies borrow the least, whereas a reference point-based argument would. In FebruaryBetween 2005,2000 andby a2004, vote oftwenty-eight 17 to 15, members million people of an FDAtook advisoryVioxx. After panel Merck rec- ommended finally withdrew that Vioxx Vioxx was safe from enough the market, to be thesold British in the United medical States, journal as The long Lancet as its Aversion to a Sure Loss packagewrote label that the carried firm shouldstrict warnings have made about that decisionthe risk ofseveral heart years attacks earlier. and strokes.In December The Recall from Chapter 1 that aversion to a sure loss can lead people to choose a risky vote2004, led Merck the best to estimates inform the were FDA that that the lawsuitsit was considering against Merck requesting would total permission $18 billion. to alternative instead of accepting a sure loss, even when the odds are against them. In reintroduceIn February Vioxx. 2005, However, by a vote in the of end17 to it 15, chose members not to ofdo an so. FDA advisory panel rec- doing so, they choose risk hoping to beat the odds, in an effort to break even. ommended that Vioxx was safe enough to be sold in the United States, as long as its In 1999, the executives at Merck were watching the clock tick down, as five of their Lawsuitspackage labeland carriedAftermath strict warnings about the risk of heart attacks and strokes. The most successful products were due to go off patent in 2000 and 2001. One possible sce- In Augustvote led 2005, Merck a Texas to inform jury the rendered FDA that its verdictit was considering on the first ofrequesting the Vioxx permission trials, find to- nario was that the firm would be unable to find new products to replace the lost profits ing reintroduceMerck liable Vioxx. for the However, death of in59-year-old the end it chose marathon not to runner do so. Robert Ernst who had from the ones going off patent. Merck continued to be one of America’s most admired takenLawsuits the drug. Theand jury Aftermath award of $253 million jolted Merck executives, and Merck’s companies throughout the 1990s, a state of affairs to which its executives had become market capitalization fell by 7.7 percent after the announcement of the jury’s verdict. In August 2005, a Texas jury rendered its verdict on the first of the Vioxx trials, find- ing Merck liable for the death of 59-year-old marathon runner Robert Ernst who had taken the drug. The jury award of $253 million jolted Merck executives, and Merck’s market capitalization fell by 7.7 percent after the announcement of the jury’s verdict.

she77208_ch02_032-057.indd 48 she77208_ch02_032-057.indd2/7/17 6:20 PM 49 2/7/17 6:20 PM

she77208_ch02_032-057.indd 49 2/7/17 6:20 PM 50 Chapter Two Introduction to Behavioral Analysis 51

Jury members indicated that they were particularly troubled that Merck execu- In 2011, Lewent was inducted into the Hall of Fame of Financial Executives Inter- tives had knowledge of Vioxx’s negative side-effect profile before they placed the national (FEI), the professional association of financial managers. For her reflections 50 Chapter Two drug on the market, but took actions to conceal that knowledge from the public. on being inducted, see her comments in Additional Resources for Chapter 2. Although Texas law requires the amount of the actual award to be much less than $253Jury million, members and indicated Merck plannedthat they towere appeal particularly the verdict, troubled the outcome that Merck of theexecu- trial set tivesan important had knowledge precedent of Vioxx’s for the negativeremaining side-effect Vioxx suits, profile which before numbered they placed 4,200 the at the 2.7 DEBIASING AND NUDGES drugtime. on On the August market, 22, but 2005, took Theactions Wall to Streetconceal Journal that knowledge reported thatfrom after the public.the verdict Althoughwas announced, Texas law analysts’ requires estimate the amount of the of valuethe actual of Merck’s award to legal be much liability less increased than The psychological phenomena described in this chapter are systematic and persistent. $253to $30 million, billion, and raising Merck questions planned toabout appeal Merck’s the verdict, long-term the outcome ability toof survive.the trial set Although they vary in incidence and degree from person to person, by and large the an importantSome analysts precedent estimated for the Merck’sremaining legal Vioxx liability suits, which to be evennumbered more, 4,200 at $50 at billion.the average responses are similar across different groups of people. What can be done to time.However, On August in the 22, end, 2005, Merck’s The Wall chief Street counsel Journal Kenneth reported Frazier that after was the able verdict to craft reduce these errors and biases? How can people debias? wasa strategy announced, that analysts’proved farestimate less expensive of the value for of the Merck’s firm. legalThat liabilitystrategy increased centered on Debiasing turns out to be an enormous challenge. Psychologists have repeatedly toframing, $30 billion, the raisingability questionsof Merck’s about lawyers Merck’s to long-termbe able to ability frame to the survive. scientific issues demonstrated that recognizing our errors and biases does not lead us to change forSome juries analysts in ways estimated jury members Merck’s found legal liabilityintuitive. to Notably,be even more, Merck at $50won billion. a series of our behavior automatically. The psychology that underlies errors and biases is However,important in Vioxx the end, cases, Merck’s and in chief 2007 counsel reached Kenneth a $4.85 Frazierbillion wassettlement able to with craft most remarkably resistant to change.13 That is not to say that people cannot learn to avoid a ofstrategy the remaining that proved plaintiffs. far less expensive Two years for after the havingfirm. That withdrawn strategy Vioxx,centered Merck’s on mistakes. People can learn. However, people learn slowly, and the task of debiasing framing,stock price the reboundedability of Merck’s to $46, lawyersand the tofirm be wasable beating to frame analysts’ the scientific consensus issues earn - requires great effort. forings juries estimates. in ways jury members found intuitive. Notably, Merck won a series of Debiasing is more difficult in some situations than in others. Situations where importantIn 2005 Vioxx CEO cases, and chairman and in 2007 Gilmartin reached retired a $4.85 from billion Merck, settlement and was with replaced most by people receive quick, clear feedback about the results of their actions are more con- oflong-time the remaining Merck plaintiffs. employee Two Richard years Clark, after who having had withdrawn been CEO Vioxx, of Merck’s Merck’s Medco ducive to debiasing efforts than situations where the feedback is slow, and outcomes stockHealth price Solutions. rebounded As tocan $46, be andseen the from firm the was ROE beating trajectory analysts’ in Exhibitconsensus 2-4, earn Merck’s- are influenced by many factors. Unfortunately, many of the tasks in corporate fi- ingsprofitability estimates. limped along thereafter. In 2013, its total returns lagged behind almost nance feature long lags between the time decisions are made and the time the out- allIn of 2005 its competitors CEO and chairman over one-five-, Gilmartin and retired ten-year from horizons. Merck, Inand 2006 was andreplaced 2007 byMerck comes occur, as well as sources of risk that can obscure the underlying biases. long-timelaunched Merck five new employee drugs that Richard drove Clark, its earnings who had for been the nextCEO several of Merck’s years. Medco However, Throughout this textbook, you will encounter tips and suggestions for mitigat - Healthafter 2007 Solutions. it did Asnot canlaunch be seen a drug from for the which ROE annual trajectory sales in were Exhibit more 2-4, than Merck’s $1 billion. ing bias. These tips and suggestions appear in boxes labeled Corporate Nudges. profitabilityMoreover, limped Merck along encountered thereafter. Inanother 2013, itsimage-tarnishing total returns lagged problem behind when almost it was As you read these tips, remember to keep your expectations modest. Debiasing allcriticized, of its competitors along with over itsone-five-, partner and Schering-Plough, ten-year horizons. of In having 2006 and avoided 2007 Merckanalyzing is a process that rarely produces quick results. The tips and suggestions produce launchedthe results five of new a trialdrugs of that the drove cholesterol its earnings drug for Vytorin. the next severalEventually, years. theHowever, results of progress, but rarely miracles. In respect to corporate finance, managers make afterclinical 2007 trialsit did not showed launch that a drug Vytorin for which was annual no more sales effective were more than than an $1 inexpensive billion. errors both individually and in groups. Mitigating bias at both levels requires generic.Moreover, The Merck number encountered of prescriptions another of image-tarnishing Vytorin fell from problem 22 million when init was2007 to criticized,fewer than along 4.6 millionwith its inpartner 2012. Schering-Plough, In addition, Merck of havingpaid $688 avoided million analyzing to settle a thelawsuit results with of ashareholders trial of the cholesterolwho alleged drug that Vytorin. the firm Eventually, had intentionally the results withheld of clinicalunfavorable trials clinical showed trial that data. Vytorin was no more effective than an inexpensive Corporate Nudges generic.In January The number 2011, Kennethof prescriptions Frazier ofreplaced Vytorin Clark fell fromas CEO. 22 millionIn 2012, in Merck 2007 toexperi - fewerenced than more 4.6 bad million within 2012. one ofIn itsaddition, heart drugs Merck and paid another $688 for million treating to osteoporosis.settle a lawsuitFrazier with hired shareholders a new head ofwho research alleged and that development the firm had to reshapeintentionally the firm’s withheld focus. Errors or biases: Accepting an actuarially unfair risk. Why does it happen? Aversion to a sure loss, aspirational risk seeking. unfavorableIn the aftermath clinical trialof the data. Vioxx incident, Brealey, Myers, and Allen did not change How does it happen? When faced with the prospect of a sure loss, people try to beat theirIn January perspective 2011, on Kenneth Merck’s Frazier capital replaced structure. Clark They as CEO. acknowledged In 2012, Merck that the experi firm- had enced more bad news with one of its heart drugs and another for treating osteoporosis. the odds. set aside $5 billion for legal costs and settlements, but pointed out that the firm’s credit What can be done about it? First, view decision tasks broadly, rather than narrowly, re- Frazier hired a new head of research and development to reshape the firm’s focus. rating was not damaged, its cash flow remained sufficient to fund its investments membering that over the course of a lifetime, risks are faced repeatedly. Because of the ( includingIn the aftermath research of and the development),Vioxx incident, and Brealey, it had Myers, maintained and Allen its regular did not dividend. change law of averages, accepting an actuarially unfair risk as a policy is likely to produce infe- theirLewent perspective resigned on Merck’s from Merckcapital instructure. 2007. AlthoughThey acknowledged she had beenthat thea strong firm had candi- rior results over the long term. Second, reframe by resetting reference points in order to setdate aside to $5be billionthe firm’s for legal next costs CEO, and replacing settlements, Gilmartin, but pointed the outboard that instead the firm’s chose credit Clark. accept losses and treat sunk costs as sunk. Try using stock phrases such as “that’s water ratingMoreover, was not Clark damaged, installed its an cash executive flow remained named sufficientPeter Loescher to fund as its global investments marketing under the bridge” and “don’t cry over spilled milk” as helpful reminders. ( includingchief and research designated and him,development), not Lewent, and as it hadhis numbermaintained two its executive. regular dividend. Lewent resigned from Merck in 2007. Although she had been a strong candi- date to be the firm’s next CEO, replacing Gilmartin, the board instead chose Clark. Moreover, Clark installed an executive named Peter Loescher as global marketing chief and designated him, not Lewent, as his number two executive.

she77208_ch02_032-057.indd 50 she77208_ch02_032-057.indd2/7/17 6:20 PM 51 2/7/17 6:20 PM

she77208_ch02_032-057.indd 50 2/7/17 6:20 PM 50 Chapter Two Introduction to Behavioral Analysis 51

Jury members indicated that they were particularly troubled that Merck execu- In 2011, Lewent was inducted into the Hall of Fame of Financial Executives Inter- tives had knowledge of Vioxx’s negative side-effect profile before they placed the national (FEI), the professional association of financial managers. For her reflections drug on the market, but took actions to conceal that knowledge from the public. on being inducted, see her comments in Additional ResourcesIntroduction for to Behavioral Chapter Analysis2. 51 Although Texas law requires the amount of the actual award to be much less than $253 million, and Merck planned to appeal the verdict, the outcome of the trial set In 2011, Lewent was inducted into the Hall of Fame of Financial Executives Inter- an important precedent for the remaining Vioxx suits, which numbered 4,200 at the 2.7 DEBIASING ANDnational NUDGES (FEI), the professional association of financial managers. For her reflections time. On August 22, 2005, reported that after the verdict on being inducted, see her comments in Additional Resources for Chapter 2. was announced, analysts’ estimate of the value of Merck’s legal liability increased The psychological phenomena described in this chapter are systematic and persistent. to $30 billion, raising questions about Merck’s long-term ability to survive. Although they vary in incidence and degree from person to person, by and large the Some analysts estimated Merck’s legal liability to be even more, at $50 billion. 2.7 DEBIASINGaverage AND responses NUDGES are similar across different groups of people. What can be done to However, in the end, Merck’s chief counsel Kenneth Frazier was able to craft reduce these errors and biases? How can people debias? a strategy that proved far less expensive for the firm. That strategy centered on DebiasingThe psychological turns out phenomena to be an enormous described inchallenge. this chapter Psychologists are systematic have and repeatedly persistent. framing, the ability of Merck’s lawyers to be able to frame the scientific issues demonstratedAlthough they that vary recognizing in incidence our and errors degree and from biases person does to notperson, lead by us and to large change the for juries in ways jury members found intuitive. Notably, Merck won a series of our average behavior responses automatically. are similar The across psychology different groups that underlies of people. errors What andcan be biases done isto important Vioxx cases, and in 2007 reached a $4.85 billion settlement with most remarkablyreduce these resistant errors to and change. biases?13 ThatHow iscan not people to say debias? that people cannot learn to avoid of the remaining plaintiffs. Two years after having withdrawn Vioxx, Merck’s mistakes.Debiasing People turns can learn. out to However, be an enormous people challenge. learn slowly, Psychologists and the task have of repeatedlydebiasing stock price rebounded to $46, and the firm was beating analysts’ consensus earn- requiresdemonstrated great effort. that recognizing our errors and biases does not lead us to change ings estimates. Debiasingour behavior is more automatically. difficult in The some psychology situations that than underlies in others. errors Situations and biases where is In 2005 CEO and chairman Gilmartin retired from Merck, and was replaced by people remarkably receive resistantquick, clear to change. feedback13 That about is notthe toresults say that of theirpeople actions cannot are learn more to avoidcon- long-time Merck employee Richard Clark, who had been CEO of Merck’s Medco ducivemistakes. to debiasing People effortscan learn. than However, situations people where learn the feedbackslowly, and is theslow, task and of outcomes debiasing Health Solutions. As can be seen from the ROE trajectory in Exhibit 2-4, Merck’s are requiresinfluenced great by effort. many factors. Unfortunately, many of the tasks in corporate fi- profitability limped along thereafter. In 2013, its total returns lagged behind almost nance Debiasingfeature long is morelags between difficult thein sometime decisionssituations arethan made in others. and theSituations time the where out- all of its competitors over one-five-, and ten-year horizons. In 2006 and 2007 Merck comespeople occur, receive as well quick, as sourcesclear feedback of risk about that can the obscureresults of the their underlying actions are biases. more con- launched five new drugs that drove its earnings for the next several years. However, Throughoutducive to debiasing this textbook, efforts than you situations will encounter where tipsthe feedback and suggestions is slow, and for outcomes mitigat - after 2007 it did not launch a drug for which annual sales were more than $1 billion. ing arebias. influenced These tips by andmany suggestions factors. Unfortunately, appear in boxes many labeled of the tasksCorporate in corporate Nudges. fi- Moreover, Merck encountered another image-tarnishing problem when it was As nanceyou read feature these long tips, lags remember between theto keeptime decisionsyour expectations are made andmodest. the time Debiasing the out- criticized, along with its partner Schering-Plough, of having avoided analyzing is a comesprocess occur, that asrarely well producesas sources quick of risk results. that can The obscure tips andthe underlyingsuggestions biases. produce the results of a trial of the cholesterol drug Vytorin. Eventually, the results of progress,Throughout but rarely this miracles. textbook, Inyou respect will encounter to corporate tips and finance, suggestions managers for mitigat make- clinical trials showed that Vytorin was no more effective than an inexpensive errorsing bothbias. individuallyThese tips and and suggestions in groups. appear Mitigating in boxes bias labeled at both Corporate levels requiresNudges. generic. The number of prescriptions of Vytorin fell from 22 million in 2007 to As you read these tips, remember to keep your expectations modest. Debiasing fewer than 4.6 million in 2012. In addition, Merck paid $688 million to settle a is a process that rarely produces quick results. The tips and suggestions produce lawsuit with shareholders who alleged that the firm had intentionally withheld progress, but rarely miracles. In respect to corporate finance, managers make unfavorable clinical trial data. Corporateerrors both individually Nudges and in groups. Mitigating bias at both levels requires In January 2011, Kenneth Frazier replaced Clark as CEO. In 2012, Merck experi- enced more bad news with one of its heart drugs and another for treating osteoporosis. Frazier hired a new head of research and development to reshape the firm’s focus. Errors or biases: Accepting an actuarially unfair risk. Why does it happen? Aversion to a sure loss, aspirational risk seeking. In the aftermath of the Vioxx incident, Brealey, Myers, and Allen did not change HowCorporate does it happen? When Nudges faced with the prospect of a sure loss, people try to beat their perspective on Merck’s capital structure. They acknowledged that the firm had the odds. set aside $5 billion for legal costs and settlements, but pointed out that the firm’s credit WhatErrors can orbe biases: done about Accepting it? First, an actuarially view decision unfair tasks risk. broadly, rather than narrowly, re- rating was not damaged, its cash flow remained sufficient to fund its investments memberingWhy does that it happen?over the courseAversion of to a alifetime, sure loss, risks aspirational are faced risk repeatedly. seeking. Because of the ( including research and development), and it had maintained its regular dividend. law Howof averages, does it happen?accepting When an actuarially faced with unfair the prospect risk as a of policy a sure is loss,likely people to produce try to infe-beat Lewent resigned from Merck in 2007. Although she had been a strong candi- rior theresults odds. over the long term. Second, reframe by resetting reference points in order to date to be the firm’s next CEO, replacing Gilmartin, the board instead chose Clark. acceptWhat losses can beand done treat about sunk costsit? First, as viewsunk. decision Try using tasks stock broadly, phrases rather such than as “that’s narrowly, water re- Moreover, Clark installed an executive named Peter Loescher as global marketing undermembering the bridge” that and over “don’t the course cry over of aspilled lifetime, milk” risks as are helpful faced reminders. repeatedly. Because of the chief and designated him, not Lewent, as his number two executive. law of averages, accepting an actuarially unfair risk as a policy is likely to produce infe- rior results over the long term. Second, reframe by resetting reference points in order to accept losses and treat sunk costs as sunk. Try using stock phrases such as “that’s water under the bridge” and “don’t cry over spilled milk” as helpful reminders.

she77208_ch02_032-057.indd 50 she77208_ch02_032-057.indd2/7/17 6:20 PM 51 2/7/17 6:20 PM

she77208_ch02_032-057.indd 51 2/7/17 6:20 PM 52 Chapter Two Introduction to Behavioral Analysis 53

the use of explicit procedures and the exercise of discipline. Debiasing requires Additional Baker, M. and J. Wurgler, “Behavioral Corporate Finance: A Current Survey.” Handbook of the Economics of Finance. Vol. 2. Edited by G. M. Constantinides, M. Harris, and R. M. major effort. Behavioral 52 Chapter Two The Corporate Nudges box at the bottom of page 51 contains an example of debi- Stulz. Handbooks in Economics. New York, NY: Elsevier, 2012. asing that builds on the earlier discussion about aversion to a sure loss. The example Readings Bernstein, P., Against the Gods: The Remarkable Story of Risk. New York: John Wiley and theprovides use of someexplicit advice procedures on steps and that the managers exercise canof discipline. take to mitigate Debiasing their requiresvulnerability Sons, 1996. majorto faulty effort. decisions that reduce value. Belsky, G. and T. Gilovich, Why Smart People Make Big Money Mistakes and How to TheFinally, Corporate a cautionary Nudges box note: at theAbsolute bottom perfectionof page 51 containsis rarely anattainable, example ofand debi some- - Correct Them: Lessons from the New Science of Behavioral Economics. New York: Simon asingtimes that it isbuilds best onto theleave earlier well discussion enough alone. about aversionIndeed, tomoderate a sure loss. degrees The example of disposi - & Schuster, 2000. providestional optimism some advice and on overconfidence steps that managers can canbe goodtake tothings. mitigate Economists their vulnerability report that tomoderate faulty decisions optimists that displayreduce value. financial habits that tend to be prudent. However, it Key Terms Catering, 35 Market timing, 35 Sentiment, 35 isFinally, possible a tocautionary have too note:much Absolute of a good perfection thing: extreme is rarely optimists attainable, display and somebehaviors- timesthat areit is generally best to leave unwise. well14 enough Certainly, alone. some Indeed, measure moderate of excessive degrees of optimism disposi- and tionaloverconfidence optimism and is essential overconfidence for leadership. can be However,good things. leaders Economists for whom report these that biases Explore the Web http://www.forbes.com/2009/04/06/sun-microsystems-enterprise-technology-enterprise -tech-sun.html moderateare extreme optimists place displaytheir organizations financial habits at risk that in tend respect to be to prudent.value destruction, However, itif not This web site contains an analysis of Sun Microsystems’s business strategy by Andy issurvival. possible to have too much of a good thing: extreme optimists display behaviors Greenberg and Brian Caulfield, “Sun’s Six Biggest Mistakes,” April 7, 2009. that are generally unwise.14 Certainly, some measure of excessive optimism and http://money.cnn.com/video/technology/2011/02/28/ctd_mcnealy_sun_oracle.fortune/ overconfidence is essential for leadership. However, leaders for whom these biases This web site contains an interview by Fortune magazine editor Adam Lashinsky with Summary areHeuristics, extreme placebiases, their and organizations framing effects at risk impede in respect managers to value from destruction, making the if bestnot use Scott McNealy. survival.of the traditional tools of corporate finance, causing them to make faulty decisions http://www.cancer.net.nz/Vioxx_warning_letter_misleading_marketing.html that destroy value. This chapter illustrates types of questions to ask when applying This web site contains the warning letter that the FDA sent to Merck in 2001. psychological concepts to analyze behavioral issues in corporate finance. http://mitsloanexperts.mit.edu/mit-sloan-alumna-judy-lewent-on-the-future-of-finance/ Summary Heuristics,The main biases, biases and framing discussed effects in the impede chapter managers are excessive from making optimism, the best overcon- use This web site contains remarks made by Judy Lewent, in “MIT Sloan Alumna Judy Lewent offidence, the traditional confirmation tools of bias,corporate and thefinance, illusion causing of control. them to Managersmake faulty are decisions inclined to on the Future of Finance,” January 9, 2012. thatchoose destroy negative value. net-present-valueThis chapter illustrates projects types because of questions they are to askexcessively when applying optimistic http://knowledge.wharton.upenn.edu/article/mercks-judy-lewent-once-again-talking psychologicalabout the future concepts prospects to analyze of their behavioral firms, overconfidentissues in corporate about finance. the risks they face, -about-the-future/ discountThe main information biases discussed that does in not the support chapter their are views, excessive and optimism,exaggerate overcon-the extent of This web site contains the article “Merck’s Judy Lewent: Once Again, ‘Talking About the fidence,control confirmationthey wield over bias, final and outcomes. the illusion of control. Managers are inclined to Future,’” November 1, 2006. chooseThe negative main heuristics net-present-value discussed projects relate to because representativeness, they are excessively availability, optimistic anchoring http://www.world-nuclear.org/info/safety-and-security/safety-of-plants/fukushima aboutand adjustment, the future prospectsand affect. of Managers their firms, are overconfident prone to make about faulty the decisions risks they about face, uses of -accident/ discountfunds because information they place that does too much not support reliance their on stereotypicviews, and exaggeratethinking when the formingextent of judg - This web site contains an analysis by the World Nuclear Association, “Fukushima Accident,” controlments, they attach wield too muchover final emphasis outcomes. to information that is readily available, become overly updated October 2015. fixatedThe main on specific heuristics numbers discussed in their relate analyses, to representativeness, and place too muchavailability, reliance anchoring on intuition. and Theadjustment, main topics and affect. discussed Managers in the are section prone toon make framing faulty effects decisions are loss about aversion uses of and Chapter 1. Does the chapter present any information that would lead you to conclude that managers funds because they place too much reliance on stereotypic thinking when forming judg- the four-fold risk pattern, especially aspiration-based risk behavior and aversion to Questions at Sun Microsystems were averse to a sure loss? Discuss. ments,a sure attach loss. Managers’too much emphasis attitudes to towardinformation risk typicallythat is readily vary available, with the become circumstances overly in 2. During its most profitable years in the late 1990s, Sun did not carry any debt. In 1999, fixatedwhich on they specific find numbersthemselves. in their Circumstances analyses, and canplace vary too accordingmuch reliance to theon intuition.firm’s posi- Sun paid $138.6 million in taxes. Compare the situation at Sun with the situation at tionThe relative main topics to its discussedaspirations in andthe sectionfinancial on position. framing effectsNotably, are managers loss aversion are inclinedand Merck in respect to their debt policies. theto four-foldmake faulty risk decisionspattern, especially about investment aspiration-based policy riskand behaviorfinancing and because aversion they to are 3. Discuss Scott McNealy’s comments about sentiment in his interview with Fortune a undulysure loss. sensitive Managers’ about attitudes potential toward future risk losses typically and vary find with it difficult the circumstances to accept inlosses magazine. In your answer, describe any valuation metric Scott McNealy mentions. Do whichthat have they alreadyfind themselves. occurred. Circumstances can vary according to the firm’s posi- McNealy’s comments suggest that Sun engaged either in catering behavior or market tion Managersrelative to needits aspirations to be aware and that financial psychological position. phenomena Notably, managers also cause are investor inclined errors timing during the technology bubble? tothat make can faulty result decisions in the mispricing about investment of the securities policy and issued financing by their because firms. they Mispricing are 4. Discuss the comment Scott McNealy made in his interview about not having hired a undulyraises issuessensitive about about catering potential behavior future andlosses market and findtiming. it difficult At the same to accept time, losses managers chief operating officer. thatneed have to bealready concerned occurred. with their own vunerability to bias. Avoiding bias, or debias- 5. You can find Sun’s statement of cash flows on the book web site. Do these statements ing,Managers is a major need challenge to be aware that that generally psychological requires phenomena a sophisticated, also cause disciplined investor errorsapproach. suggest that Sun engaged in market timing during the technology bubble? that can result in the mispricing of the securities issued by their firms. Mispricing raises issues about catering behavior and market timing. At the same time, managers need to be concerned with their own vunerability to bias. Avoiding bias, or debias- ing, is a major challenge that generally requires a sophisticated, disciplined approach.

she77208_ch02_032-057.indd 52 she77208_ch02_032-057.indd2/7/17 6:20 PM 53 2/7/17 6:20 PM

she77208_ch02_032-057.indd 52 2/7/17 6:20 PM www.mhhe.com/shefrin2e 2/7/17 6:20 PM 2/7/17 6:20 PM

5353 5353 Fortune Fortune Sentiment, 35 Sentiment, 35 Introduction to Behavioral Analysis to Behavioral Analysis Introduction Introduction Introduction to Behavioral Analysis to Behavioral Analysis Introduction Introduction Market timing, 35 Market timing, 35 at Sun Microsystems were averse to a sure loss? Discuss. Sun paid $138.6 million in taxes. Compare the situation at Merck in respect to their debt policies. Sun with the situation at magazine. In your answer, describe any valuation metric Scott McNealy mentions. Do McNealy’s comments suggest that Sun engaged either in catering behavior or market timing during the technology bubble? chief operating officer. suggest that Sun engaged in market timing during the technology bubble? 1. Does the chapter present any information that would lead you to conclude that managers managers that conclude to you lead would 1. that information any present chapter the Does 2. During its most profitable years in the late 1990s, Sun did not carry any debt. In 1999, 3. Discuss Scott McNealy’s comments about sentiment in his interview with 4. Discuss the comment Scott McNealy made in his interview about not having hired a 5. You can find Sun’s statement of cash flows on the book web site. Do these statements at Sun Microsystems were averse to a sure loss? Discuss. at Sun Microsystems were averse to a sure Sun paid $138.6 million in taxes. Compare the situation at Sun Merck in respect to their debt policies. with the situation at magazine. In your answer, describe any valuation metric Scott McNealy mentions. Do McNealy’s comments suggest that Sun engaged either in catering behavior or market timing during the technology bubble? chief operating officer. suggest that Sun engaged in market timing during the technology bubble? http://www.forbes.com/2009/04/06/sun-microsystems-enterprise-technology-enterprise -tech-sun.html business strategy by Andy This web site contains an analysis of Sun Microsystems’s Biggest Mistakes,” April 7, 2009. Greenberg and Brian Caulfield, “Sun’s Six http://money.cnn.com/video/technology/2011/02/28/ctd_mcnealy_sun_oracle.fortune/ magazine editor Adam Lashinsky with This web site contains an interview by Fortune Scott McNealy. http://www.cancer.net.nz/Vioxx_warning_letter_misleading_marketing.html the FDA sent to Merck in 2001. This web site contains the warning letter that http://mitsloanexperts.mit.edu/mit-sloan-alumna-judy-lewent-on-the-future-of-finance/ Lewent, in “MIT Sloan Alumna Judy LewentThis web site contains remarks made by Judy on the Future of Finance,” January 9, 2012. http://knowledge.wharton.upenn.edu/article/mercks-judy-lewent-once-again-talking -about-the-future/ Judy Lewent: Once Again, ‘Talking About the This web site contains the article “Merck’s Future,’” November 1, 2006. http://www.world-nuclear.org/info/safety-and-security/safety-of-plants/fukushima -accident/ Nuclear Association, “Fukushima Accident,”This web site contains an analysis by the World updated October 2015. Baker, M. and J. Wurgler, “Behavioral Corporate Finance: A Current Survey.” Handbook “Behavioral Corporate Finance: A Current Baker, M. and J. Wurgler, M. Harris, and R. M. Vol. 2. Edited by G. M. Constantinides, of the Economics of Finance. New York, NY: Elsevier, 2012. Stulz. Handbooks in Economics. and Remarkable Story of Risk. New York: John Wiley Bernstein, P., Against the Gods: The Sons, 1996. Big Money Mistakes and How to Why Smart People Make Belsky, G. and T. Gilovich, New York: Simon from the New Science of Behavioral Economics. Correct Them: Lessons & Schuster, 2000. Catering, 35 1. Does the chapter present any information that would lead you to conclude that managers that conclude to you lead would 1. that information any present chapter the Does 2. During its most profitable years in the late 1990s, Sun did not carry any debt. In 1999, 3. Discuss Scott McNealy’s comments about sentiment in his interview with 4. Discuss the comment Scott McNealy made in his interview about not having hired a 5. You can find Sun’s statement of cash flows on the book web site. Do these statements http://www.forbes.com/2009/04/06/sun-microsystems-enterprise-technology-enterprise Baker, M. and J. Wurgler, “Behavioral Corporate Finance: A Current Survey.” Handbook A Current Survey.” Corporate Finance: and J. Wurgler, “Behavioral Baker, M. Harris, and R. M. Constantinides, M. 2. Edited by G. M. of Finance. Vol. of the Economics 2012. New York, NY: Elsevier, in Economics. Stulz. Handbooks and New York: John Wiley Remarkable Story of Risk. Bernstein, P., Against the Gods: The Sons, 1996. Big Money Mistakes and How to Why Smart People Make Belsky, G. and T. Gilovich, New York: Simon from the New Science of Behavioral Economics. Correct Them: Lessons & Schuster, 2000. Catering, 35 -tech-sun.html strategy by Andy an analysis of Sun Microsystems’s business This web site contains Biggest Mistakes,” April 7, 2009. Greenberg and Brian Caulfield, “Sun’s Six http://money.cnn.com/video/technology/2011/02/28/ctd_mcnealy_sun_oracle.fortune/ magazine editor Adam Lashinsky with This web site contains an interview by Fortune Scott McNealy. http://www.cancer.net.nz/Vioxx_warning_letter_misleading_marketing.html the FDA sent to Merck in 2001. This web site contains the warning letter that http://mitsloanexperts.mit.edu/mit-sloan-alumna-judy-lewent-on-the-future-of-finance/ Lewent, in “MIT Sloan Alumna Judy LewentThis web site contains remarks made by Judy on the Future of Finance,” January 9, 2012. http://knowledge.wharton.upenn.edu/article/mercks-judy-lewent-once-again-talking -about-the-future/ Judy Lewent: Once Again, ‘Talking About the This web site contains the article “Merck’s Future,’” November 1, 2006. http://www.world-nuclear.org/info/safety-and-security/safety-of-plants/fukushima -accident/ Nuclear Association, “Fukushima Accident,”This web site contains an analysis by the World updated October 2015.

Chapter Questions Explore the Web Key Terms Additional Behavioral Readings

Chapter Questions Key Terms Explore the Web Additional Additional Behavioral Readings

she77208_ch02_032-057.indd 53 2/7/17 6:20 PM she77208_ch02_032-057.indd 53

- - - - - errors

Certainly, some measure of excessive optimism and 14 The main biases discussed in the chapter are excessive optimism, overcon- The main heuristics discussed relate to representativeness, availability, anchoring and aversion loss are effects framing on section the in discussed topics main The investor cause also phenomena psychological that aware be to need Managers The Corporate at box the Nudges bottom of 51 page contains of an debi example a Finally, cautionary note: Absolute perfection is rarely attainable, and some Heuristics, biases, and framing effects impede managers from making the best use of the traditional tools of corporate finance, causing them to make faulty decisions that destroy value. This chapter illustrates types of questions to ask when applying issues in corporate finance. psychological concepts to analyze behavioral fidence, confirmation bias, and the illusion of control. Managers choose negative are net-present-value projects because they optimistic inclined are excessively to about the future prospects of their firms, overconfident about the risks of extent the they exaggerate and face, supportviews, their not does that information discount final outcomes. wield over they control and are adjustment, Managers and prone decisions to affect. about faulty make uses of judg forming when thinking stereotypic on reliance much too place they because funds overly become available, readily is that information to emphasis much too attach ments, intuition. on reliance much too place and specificfixatedon their in analyses, numbers risk the pattern, four-fold especially aspiration-based risk to and behavior aversion in circumstances the with vary typically risk toward attitudes Managers’ loss. sure a which they find themselves. Circumstances can vary according to the firm’s posi- inclined are managers Notably, position. financial and aspirations its to relative tion to make faulty investment policy decisions and about financing because they are unduly sensitive about potential future losses and find it difficult to accept losses occurred. already that have the use of explicit procedures and the of exercise discipline. Debiasing requires effort. major asing that builds on the earlier discussion to about a aversion sure loss. The example provides some advice on steps that managers can take to mitigate their vulnerability decisions that value. reduce faulty to times it is best to leave well enough alone. Indeed, moderate degrees of disposi tional optimism and overconfidence can be good things. Economists report that moderate optimists display financial habits that tend to be prudent. However, it is possible too to much of have a good thing: optimists extreme behaviors display that are generally unwise. that can result in the mispricing of the securities issued by their firms.managers time, same the MispricingAt timing. market and behavior catering about issues raises need to be concerned with their own vunerability to bias, bias. or Avoiding debias approach. disciplined sophisticated, a requires generally that challenge major a is ing, overconfidence is essential for leadership. However, leaders for whom these biases biases these whom for leaders However, leadership. for essential is overconfidence are extreme place their organizations at risk in respect to value destruction, if not survival. Summary 52 Chapter Two she77208_ch02_032-057.indd 52 she77208_ch02_032-057.indd 54 she77208_ch02_032-057.indd 54 www.mhhe.com/shefrin2e ChapterTwo 54 ChapterTwo 54 10. 7. Did Merck’s managers exhibit confirmation bias in their assessment of Vioxx? Discuss 6. 9. 8. 10. 7. Did Merck’s managers exhibit confirmation bias in their assessment of Vioxx? Discuss 6. 9. 8. the firm’s history. However, in August the in time other any at as productive as was pipeline research Merck’s that stated she later, of their top five drugs would achieve the upper end of their prediction ranges. One month four range, prediction the of end lower the to closer be would sales Vioxx although that say to prediction her qualified she June In billion. $3.5 and $3 between be would year the for sales Vioxx that predicted Lewent Judy CFO Merck’s 2001, of beginning the At this question. included an article describing the results of the VIGOR study. VIGOR the of results the describing article an included Medicine of Journal article pointed out that the November 2000 issue Vioxx. of about decisions made executives Merck’s in which manner 2004, 2, November On fluke, withVioxxactuallybeingnoriskierthannaproxenwhenitcomestoheartattacks. is the case. From the vantage point of 1999, discuss whether the study outcome was just a ers believe that naproxen does not reduce the incidence of heart attacks. Suppose that this heart attacks. Of those receiving naproxen, four had heart attacks. Many medical research had 20 Vioxx, receiving study those Of the naproxen. received half in other the subjects and Vioxx, the received Half low. was attack heart a experiencing of risk whose jects sub only include to chose Merck Notably, subjects. 8,000 used study VIGOR Merck’s in Exhibit2-5,discusswhethertheeventsjustdescribedreflectanybehavioralbiases. displayed equity on return of graph the of light In billion. $2.5 were sales Vioxx 2002 In billion. $2.3 were sales Vioxx 2001 In slow. to began Vioxx of sales thereafter, ately Immedi stroke. and attack heart of risk increased an to Vioxx linking study a published interest)intheprecedingdiscussion? ofconflicts general (meaning issues conflict agency and issues behavioral any detect consulting contracts or research grants from Merck and employees of Merck. Can you received who academics were article the of authors The article. the in presented data accurately article the of text the that ensuring on efforts its centrated not have all the details that were available to the FDA. He stated that his journal con England Journal, Gregory Curfman, told ing a heart attack was anything but low. In retrospect, the executive editor of having a heart attack. Notably, Merck had excluded subjects whose risk of experienc ofrisk high at be to appear not did who patients among attacks heart of incidence the Medicine of Journal England order tocompeteagainstcompetitor Pfizer’sdrugCelebrex. inmarket the on drug the placed instead firm the However, application. drug new its withdrawn have could and 1999 in approval FDA won it before attacks heart and strokes to relating Vioxx by posed risks the of aware was Merck that alleges drawn from the settlement Merck had negotiated some years earlier. The complaint with had investors These Funds. Transamerica and BV Management Investment complaint filed in 2015 against Merck by the Kuwait Investment Authority, Aegon so, Vioxx might still be done on the Merck market. The had second position that is articulated is by a position legal this of thrust The risks.” the about honest being about problem prolonged serious, “a labeling Science deficient Translational Merck’s Scripps called Diego’s Institute, San of director Topol, Eric ads. and ing packag drug on language prominent with drug) the of use prolonged (after stroke is position first The orattack heart of risk enhanced the about patients warned have should Merck that Vioxx. of handling Merck’s about positions two Compare the firm’s history. However, in August the in time other any at as productive as was pipeline research Merck’s that stated she later, of their top five drugs would achieve the upper end of their prediction ranges. One month four range, prediction the of end lower the to closer be would sales Vioxx although that say to prediction her qualified she June In billion. $3.5 and $3 between be would year the for sales Vioxx that predicted Lewent Judy CFO Merck’s 2001, of beginning the At this question. included an article describing the results of the VIGOR study. VIGOR the of results the describing article an included Medicine of Journal article pointed out that the November 2000 issue Vioxx. of about decisions made executives Merck’s in which manner 2004, 2, November On fluke, withVioxxactuallybeingnoriskierthannaproxenwhenitcomestoheartattacks. is the case. From the vantage point of 1999, discuss whether the study outcome was just a ers believe that naproxen does not reduce the incidence of heart attacks. Suppose that this heart attacks. Of those receiving naproxen, four had heart attacks. Many medical research had 20 Vioxx, receiving those study Of naproxen. the received half in other the and subjects Vioxx, received the Half low. was attack heart a experiencing of risk whose jects sub only include to chose Merck Notably, subjects. 8,000 used study VIGOR Merck’s in Exhibit2-5,discusswhethertheeventsjustdescribedreflectanybehavioralbiases. displayed equity on return of graph the of light In billion. $2.5 were sales Vioxx 2002 In billion. $2.3 were sales Vioxx 2001 In slow. to began Vioxx of sales thereafter, ately Immedi stroke. and attack heart of risk increased an to Vioxx linking study a published centrated its efforts on ensuring that the text of the article accurately article the of text the that ensuring on efforts its centrated not have all the details that were available to the FDA. He stated that his journal con England Journal, Gregory Curfman, told ing a heart attack was anything but low. In retrospect, the executive editor of having a heart attack. Notably, Merck had excluded subjects whose risk of experienc ofrisk high at be to appear not did who patients among attacks heart of incidence the Medicine of Journal England interest)intheprecedingdiscussion? ofconflicts general (meaning issues conflict agency and issues behavioral any detect consulting contracts or research grants from Merck and employees of Merck. Can you received who academics were article the of authors The article. the in presented data order tocompeteagainstcompetitorPfizer’sdrugCelebrex. inmarket the on drug the placed instead firm the However, application. drug new its withdrawn have could and 1999 in approval FDA won it before attacks heart and strokes to relating Vioxx by posed risks the of aware was Merck that alleges drawn from the settlement Merck had negotiated some years earlier. The complaint with had investors These Funds. Transamerica and BV Management Investment complaint filed in 2015 against Merck by the Kuwait Investment Authority, Aegon so, Vioxx might still be on done the market. Merck The second had position is that articulated by is a legalposition this of thrust The risks.” the about honest being about problem prolonged serious, “a labeling Science deficient Translational Merck’s Scripps called Institute, Diego’s San of director Topol, Eric ads. and ing packag drug on language prominent with drug) the of use prolonged (after stroke is position first The orattack heart of risk enhanced the about patients warned have should Merck that Vioxx. of handling Merck’s about positions two Compare The Wall Street Journal Street Wall The The Wall Street Journal Street Wall The article stated that Vioxx did not significantly increase significantly not did Vioxx that stated article article stated that Vioxx did not significantly increase significantly not did Vioxx that stated article Journal of the American Medical Association The Wall Street Journal Journal of the American Medical Association The Wall Street Journal published an article describing the describing article an published published an article describing the describing article an published The New England Journal The New England Journal that his journal did The Wall Street Wall The

represented the represented that his journal did The Wall Street Wall The

represented the represented The New The The New The New The The New ------

------

2/7/17 6:20PM she77208_ch02_032-057.indd 55 2/7/17 6:20PM

that that was expected to occur only once in 10,000 years. The event seismic major a for plan to was standard the built, Safetyand Industrial Agency (NISA). Nuclear Japan’s by regulated was and Electric, General by the built was by (TEPCO), Company operated Power Electric Tokyo and 1971 in commissioned was plant the regard, this In operation. in was it when period quent subse- the in and plant Daiichi for Fukushima the stage building planning the in both assessed, were tsunamis systems subsequently failed. cooling core emergency the and generators backup the swamped it struck, earthquake the after hour an arrived tsunami the However,when overheating. from cores tor tors being used to supply power in order to keep the reac shut down automatically, as planned, with backup genera reactors operating Daiichi’s Fukushima earthquake, the Nuclear MeltdownatFukushimaDaiichi Minicase in the vicinity of $300 billion. $300 of vicinity the be in to disaster the from damages total estimated govern ment Japanese The least for. at unaccounted with people 2,500 15,000 exceeded deaths of number the penetrated inlandforasmuch6miles(10kilometers). in Japan. Tsunami waves reached 128 feet (39 meters) and damage major caused that tsunami powerful a generated it and country, the in recorded ever largest the was Japan. This of coast northeast the off occurred scale Richter the on 9 magnitude of earthquake an 2011, 11, March On but vulnerable to the ensuing tsunami. ensuing the to vulnerable but ation, the reactors proved to have been robust seismically, of very bad luck. According to the World Nuclear Associ case a was tsunami and earthquake of combination 2011 the and Japan, in risks recognized well are tsunamis and Earthquakes rods. fuel nuclear its of some of meltdown the experienced which plant, power Daiichi Fukushima alternatives such asoil. nuclear reactors until 2015, switching to more expensive its of all down shut Japan consequences. long-term ing www.mhhe.com/shefrin 2e Additional ResourcesandMaterialsforChapter2AreAvailableat During the time that the Fukushima Daiichi plant was plant Daiichi Fukushima the that time the During and earthquakes with associated risks how Consider According to the country’s National Police Agency,Police National country’s the to According Included in the damage was the destruction of the the of destruction the was damage the in Included 16 15 This event had ongo- had event This 17 In response to response In - - - - tively, one in 1983 and the other in 1993. Subsequent 1993. in other the and 1983 in tively,one 45 feet (15 meters) and over of90 feet heights (30 meters) respec- with tsunamis generated 7.7 magnitude of averageon years.Twoevery 12 once ring earthquakes of magnitude 7.7 to 8.4 on the Richter scale, and occur earthquakesby generated meters), feet(10 30 ceeding the region around Japan with maximum amplitudes ex havetsunamis eight previouscentury recordedbeen in the In meters). (20 feet 60 meters), exceeded having six (10 with feet 30 exceeded amplitude the maximum which for 1498, since occurred have that events 12aroundincludes and Japan in tsunamis historical of earthquakes that occur under the ocean. A compilation major some with associated only are they as quakes, and warning shook buildings200milesaway inTokyo. tsunami a and triggered also collapses, it outages; building power casualties, caused The quake 7.5. 2005 and 7.2 between magnitudes with of all island Honshū, Japanese the of coast east the off Miyagi near In 1936, 1978, and 2005, three major earthquakes occurred 2015. and 2000 years the between occurred the number same about Notably, greater. or 6.3 magnitude of twentieth century the in occurred docu quakes 20 well than events More mented. seismic destructive major and basis near-daily a on occurring tremors minor with year, every of 10 feet (3.1meters). coast of Chile, which generated a tsunami having a height decisions was an earthquake that occurred in 1960 was off the these for that basis The slope level. sea above a meters) (10 of 30 feet top the at plant located main were the buildings and level sea above meters) (4 feet 12 located were buildings seawaterintake the which in sign region ofanuclear powerthe plant. in tsunamis and earthquakes historic and prehistoric on earthquakes and tsunamis should include of the feasible collection of risks data the assessing for practice best that stipulates (IAEA) Agency Energy Atomic International - earth than frequently less much occur Tsumamis earthquakes 1,500 approximately experiences Japan de- plant a on agreed NISA and TEPCO Originally, Introduction toBehavioralAnalysis 55 - - - 2/7/17 6:20PM www.mhhe.com/shefrin2e 2/7/17 6:20 PM 2/7/17 6:20 PM ------5555 5555 Introduction to Behavioral Analysis to Behavioral Analysis Introduction Introduction Introduction to Behavioral Analysis to Behavioral Analysis Introduction Introduction Originally, Originally, TEPCO and NISA agreed on a plant de- Japan experiences approximately 1,500 earthquakes Tsumamis occur much less frequently than earth- Originally, Originally, TEPCO and NISA agreed on a plant de- Japan experiences approximately 1,500 earthquakes Tsumamis occur much less frequently than earth- International Atomic Energy Agency (IAEA) stipulates that best practice for assessing the data of riskscollection feasible the include of should tsunamis and earthquakes on prehistoric and historic earthquakes and tsunamis in plant. thepower of a nuclear region sign in which the intake seawater buildings were located 12 feet (4 meters) above sea level buildings and the were main located plant at the top 30 feet of (10 meters) above a sea level. The slope basis for that these the off 1960 in occurredthat was earthquake an was decisions height a having tsunami a generated which Chile, of coast (3.1 meters). of 10 feet every year, with minor tremors occurring on a near-daily basis and major destructive seismic mented. More events than well 20 quakes docu occurred in the century twentieth of magnitude 6.3 or greater. Notably, about same number occurred the between the years 2000 and 2015. occurred earthquakes major three 2005, and 1978, 1936, In near Miyagi off the east coast of the Japanese Honshū, island all of with magnitudes between 7.2 and 2005 7.5. quake The caused casualties, power building outages; it collapses, also triggered and a tsunami in Tokyo. shook buildings 200 miles away warning and quakes, as they are only associated with some major compilation A ocean. the under occur thatearthquakes of historical tsunamis in and Japan around includes 12 events that have occurred since 1498, for which maximum the amplitude exceeded 30 feet with six (10 having exceeded 60 meters), feet (20 meters). In the been in recorded centuryprevious eight tsunamis have ex amplitudes maximum with Japan around region the ceeding 30 (10 feet meters), generated by earthquakes occur and scale, Richter the on 8.4 to 7.7 magnitude of earthquakes ring once 12 every Two years. on average of magnitude 7.7 generated tsunamis - respec with meters) (30 heights feet 90 of over and meters) (15 feet 45 one tively, in 1983 and the other in 1993. Subsequent International Atomic Energy Agency (IAEA) stipulates that best practice for assessing the data risksof collection feasible the of include should tsunamis and earthquakes on prehistoric and historic earthquakes and tsunamis in plant. the of a nuclear power region sign in which the intake seawater buildings were located 12 feet (4 meters) above sea level and buildings the were main located plant at the top 30 feet of (10 meters) a above sea level. slope The basis that for these the off was 1960 in occurredthat earthquake an was decisions height a having tsunami a generated which Chile, of coast (3.1 meters). of 10 feet every year, with minor tremors occurring on a near-daily basis and major destructive seismic mented. More events than well 20 quakes docu occurred in the century twentieth of magnitude 6.3 or greater. Notably, about same number the occurred between the years 2000 and 2015. occurred earthquakes major three 2005, and 1978, 1936, In near Miyagi off the east coast of the Japanese Honshū, island all of with magnitudes between 7.2 and 2005 7.5. quake The caused casualties, power building outages; it collapses, also triggered and a tsunami in Tokyo. shook buildings 200 miles away warning and quakes, as they are only associated with some major compilation A ocean. the under occur thatearthquakes of historical tsunamis in and Japan around includes 12 events that have occurred since 1498, for which maximum the amplitude exceeded 30 feet with (10 six having exceeded meters), 60 feet (20 meters). In the been in recorded centuryprevious eight tsunamis have ex amplitudes maximum with Japan around region the ceeding 30 (10 feet meters), generated by earthquakes occur and scale, Richter the on 8.4 to 7.7 magnitude of earthquakes ring once 12 every Two on years. average of magnitude 7.7 generated tsunamis with- respec meters) (30 heights feet 90 of over and meters) (15 feet 45 one tively, in 1983 and the other in 1993. Subsequent ------In response to 17 In response to 17 This event had ongo- 15 This event had ongo- 15 16 16 Included in the damage was the destruction of the According According to the country’s National Police Agency, Consider how risks associated with earthquakes and During the time that the Fukushima Daiichi plant was

Additional Resources and Materials for Chapter 2 Are Available at Chapter 2 Are Available at and Materials for Additional Resources 2e www.mhhe.com/shefrin ing long-term consequences. Japan shut down all of its expensive more to switching 2015, until reactors nuclear as oil. such alternatives Fukushima Daiichi power plant, which experienced the meltdown of some of its nuclear fuel rods. Earthquakes and tsunamis are well recognized risks in Japan, and the 2011 combination of earthquake and tsunami was a case Associ Nuclear World the to According luck. bad very of ation, the reactors proved to have been robust seismically, seismically, robust been have to proved reactors the ation, but vulnerable to the ensuing tsunami. ment estimated total damages from the disaster to in the be vicinity of $300 billion. On March 11, 2011, an earthquake of magnitude 9 on the Richter scale occurred off the northeast coast of This Japan. was the largest ever recorded in the country, and it generated a powerful tsunami that caused major damage and meters) (39 feet 128 reached waves Tsunami Japan. in kilometers). penetrated inland for as much as 6 miles (10 the number of deaths exceeded 2,500 15,000 people with unaccounted at for. least The Japanese govern Minicase at Fukushima Daiichi Nuclear Meltdown tsunamis were assessed, both in the planning building the stage Fukushima for Daiichi plant and in the subse- quent period when it was in operation. In this regard, the plant was commissioned in 1971 Tokyo and Electric Power operated Company (TEPCO), was by built by the General Electric, and was regulated by Japan’s Nuclear (NISA). Agency and Industrial Safety built, the standard was to plan for a major seismic event The years. 10,000 in once only occur to expected thatwas the earthquake, Fukushima Daiichi’s operating reactors genera backup with planned, as automatically, down shut reac the keep to order in power supply to used being tors tor cores from overheating. when However, the tsunami arrived an hour after the earthquake struck, it swamped the backup generators and the emergency core cooling failed. subsequently systems

Included in the damage was the destruction of the According According to the country’s National Police Agency, Consider how risks associated with earthquakes and During the time that the Fukushima Daiichi plant was Additional Resources and Materials for Chapter 2 Are Available at 2 Are Available for Chapter and Materials Resources Additional 2e www.mhhe.com/shefrin ing long-term consequences. Japan shut down all of its expensive more to switching 2015, until reactors nuclear as oil. such alternatives Fukushima Daiichi power plant, which experienced the meltdown of some of its nuclear fuel rods. Earthquakes and tsunamis are well recognized risks in Japan, and the 2011 combination of earthquake and tsunami was a case Associ Nuclear World the to According luck. bad very of ation, the reactors proved to have been robust seismically, seismically, robust been have to proved reactors the ation, but vulnerable to the ensuing tsunami. On March 11, 2011, an earthquake of magnitude 9 on the Richter scale occurred off the northeast coast of This Japan. was the largest ever recorded in the country, and it generated a powerful tsunami that caused major damage and meters) (39 feet 128 reached waves Tsunami Japan. in as much as 6 miles (10 kilometers). penetrated inland for the number of deaths exceeded 15,000 2,500 people with unaccounted at for. least The Japanese govern ment estimated total damages from the disaster to in be the vicinity of $300 billion. Nuclear Meltdown at Fukushima Daiichi Nuclear Meltdown Minicase the earthquake, Fukushima Daiichi’s operating reactors genera backup with planned, as automatically, down shut reac the keep to order in power supply to used being tors tor cores from overheating. when However, the tsunami arrived an hour after the earthquake struck, it swamped the backup generators and the emergency core cooling failed. subsequently systems tsunamis were assessed, both in the planning building stage the Fukushima for Daiichi plant and in the subse- quent period when it was in operation. In this regard, the plant was commissioned in 1971 and Tokyo Electric Power operated Company (TEPCO), by was built the by General Electric, and was regulated by Japan’s Nuclear (NISA). Agency and Industrial Safety built, the standard was to plan for a major seismic event The years. 10,000 in once only occur to expected thatwas

she77208_ch02_032-057.indd 55 2/7/17 6:20 PM she77208_ch02_032-057.indd 55

------The New New The The New represented the

The Wall Street that his journal did did journal his that The New England Journal Journal England New The published an article describing the The Wall Street Journal Street Wall The Journal of the American Medical Association Medical American the of Journal article stated that Vioxx did not significantly increase The Wall Street Journal Compare two positions about Merck’s handling of Vioxx. that Merck should have warned patients about the enhanced risk of heart attack or The first position is stroke (after prolonged use of the drug) with prominent language on drug packag ing and ads. Eric Topol, director of San Diego’s Institute, called Scripps Merck’s Translational deficient Science labeling “a serious, prolonged problem about being honest about the risks.” The thrust of this position legal a is by articulated that is position had second The Merck market. the done on be still might Vioxx so, Aegon Authority, Investment Kuwait the by Merck against 2015 in filed complaint Investment Management BV and Transamerica Funds. These investors had with complaint The earlier. years some negotiated had Merck settlement the from drawn alleges that Merck was aware of the risks posed by Vioxx relating to strokes and heart attacks before it won FDA approval in 1999 and could have withdrawn its new drug application. However, the firm instead placed the drug on the market in order to compete against competitor Pfizer’s drug Celebrex. data presented in the article. The authors of the article were academics who received you Can Merck. of employees and Merck from grants research or contracts consulting detect any behavioral issues and agency conflict issues (meaning general conflicts of interest) in the preceding discussion? England Journal of Medicine the incidence of heart attacks among patients who did not appear to be at high risk of experienc of risk whose subjects excluded had Merck Notably, attack. heart a having of editor executive the retrospect, In low. but anything was attack heart a ing told Curfman, Gregory Journal, England con journal his that stated He FDA. the to available were that details the all have not centrated its efforts on ensuring that the text of the article accurately published a study linking Vioxx to an increased risk of heart attack and stroke. Immedi ately thereafter, sales of Vioxx began to slow. In 2001 Vioxx sales were $2.3 billion. In 2002 Vioxx sales were $2.5 billion. In light of the graph of return on equity displayed any behavioral biases. whether the events just described reflect in Exhibit 2-5, discuss Merck’s VIGOR study used 8,000 subjects. Notably, Merck chose to include only sub jects whose risk of experiencing a heart attack was low. Half the received Vioxx, subjects and the other in half received the naproxen. Of study those receiving Vioxx, 20 had research medical Many attacks. heart had four naproxen, receiving those Of attacks. heart thisthat Suppose attacks. heart of incidence the reduce not does naproxen that believe ers ajust was outcome study the whether discuss 1999, of point vantage the From case. the is than naproxen when it comes to heart attacks. fluke, with Vioxx actually being no riskier On November 2, 2004, manner in which Merck’s executives made decisions about of Vioxx. issue 2000 November the that out pointed article Journal of Medicine included an article describing the results of the VIGOR study. this question. At the beginning of 2001, Merck’s CFO Judy Lewent predicted that Vioxx sales for the year would be between $3 and $3.5 billion. In June she qualified her prediction to say that although Vioxx sales would be closer to the lower end of the prediction range, four month One ranges. prediction their of end upper the achieve would drugs five top their of later, she stated that Merck’s research pipeline was as productive as at any other time in the August in However, history. firm’s the 8. 9. 6. Discuss Vioxx? of assessment their in bias confirmation exhibit managers Merck’s Did 7. 10. 54 54 Chapter Two Chapter Two she77208_ch02_032-057.indd 54 she77208_ch02_032-057.indd 56 she77208_ch02_032-057.indd 56 www.mhhe.com/shefrin2e sloped area on which the nuclear power plants rested. plants power nuclear the which on area sloped the reached it when behave would tsunami a how mean- ing “runup,” determine to methodology a veloped in hazard prediction.” overconfidence against warning sobering “a as outcome the characterizes report the 9, was earthquake actual the of magnitude the that Given years. 30 within region the strike would earthquake 7.5 magnitude a that probability percent 99 a is there that judgment its repeated ernment, gov- Japanese the by funded is which Promotion, search Fukushima Fukushima Daiichi meltdown. the with connection in decisions and judgments lyzedthe of points that are summarized below. are summarized of pointsthat the Fukushima Daiichi plant. Daiichi Fukushima the to relevantwas that incident an experienced and so, did for cooling. However, a plant in France located at Blayais Chernobyl was located on a sea coast and using seawater Unlike Fukushima Daiichi, neither Three Mile Island nor land in 1979 and the other at Chernobyl, Ukraine in 1986. downs at two nuclear power plants, one at Three Mile Is Richter 8.3onthe scale. measuring earthquake an of result a as 869, year the in occurred havingepisode recent most the with so, or years sand thou every once occur tsunamis massive that gesting layerssug discoveredsedimentary research academic ChapterTwo 56 approximately 1in10million. 1 million. Future plants will likely be built to a standard of In years. 100,000 in wereoperation in in plants bestoperating1 ofat the 2015, 1 is utilities U.S. for requirement threshold frequency.The damage core 1-in-10,000-year a Regulatory Commission specifies that reactor designs meet stricter safety standards across the globe. The U.S. Nuclear were not sufficiently protected from flooding. equipment emergency contained which rooms the that low,also too werebut dikes the that onlyfoundnot ities Thereafter,investigationan byconducted French author site. the at reactors all down shut to prudent was it that prove dangerous. The decision makers at Blayais decided nearbyestuary. systempumping breakdown can A the in heat from the reactor rods to a heat “sink,” in this case the carry to needs system pumping the meltdown, reactor a preventto order In station. pumping a flooded and dikes plant’s the overflowedsurge storm meter) foot-high(2.5 eight- an 1999 In dikes.surges,by storm from protected sloped area on which the nuclear power plants rested. plants power nuclear the which on area sloped the reached it when behave would tsunami a how mean- ing “runup,” determine to methodology a veloped in hazard prediction.” overconfidence against warning sobering “a as outcome the characterizes report the 9, was earthquake actual the of magnitude the that Given years. 30 within region the strike would earthquake 7.5 magnitude a that probability percent 99 a is there that judgment its repeated ernment, gov- Japanese the by funded is which Promotion, search Fukushima Fukushima Daiichi meltdown. the with connection in decisions and judgments lyzedthe of points that are summarized below. are summarized of pointsthat academic research discovered sedimentary layerssug discoveredsedimentary research academic ChapterTwo 56 the Fukushima Daiichi plant. Daiichi Fukushima the to relevantwas that incident an experienced and so, did for cooling. However, a plant in France located at Blayais Chernobyl was located on a sea coast and using seawater Unlike Fukushima Daiichi, neither Three Mile Island nor land in 1979 and the other at Chernobyl, Ukraine in 1986. downs at two nuclear power plants, one at Three Mile Is Richter 8.3onthe scale. measuring earthquake an of result a as 869, year the in occurred havingepisode recent most the with so, or years sand thou every once occur tsunamis massive that gesting approximately 1in10million. 1 million. Future plants will likely be built to a standard of In years. 100,000 in wereoperation in in plants bestoperating1 ofat the 2015, 1 is utilities U.S. for requirement threshold frequency.The damage core 1-in-10,000-year a Regulatory Commission specifies that reactor designs meet stricter safety standards across the globe. The U.S. Nuclear were not sufficiently protected from flooding. equipment emergency contained which rooms the that low,also too werebut dikes the that onlyfoundnot ities Thereafter,investigationan byconducted French author site. the at reactors all down shut to prudent was it that prove dangerous. The decision makers at Blayais decided nearbyestuary. systempumping breakdown can A the in heat from the reactor rods to a heat “sink,” in this case the carry to needs system pumping the meltdown, reactor a preventto order In station. pumping a flooded and dikes plant’s the overflowedsurge storm meter) foot-high(2.5 eight- an 1999 In dikes.surges,by storm from protected n 02 te aa Scey f ii Egnes de Engineers Civil of Society Japan the 2002, In In January 2011, the Headquarters for Earthquake Re- for Earthquake Headquarters the 2011, January In uuhm Dicis xsec cicdd ih melt- with coincided existence Daiichi’s Fukushima A report published by the Carnegie Endowment ana- Endowment Carnegie the by published report A of institution the prompted incidents nuclear These n 02 te aa Scey f ii Egnes de Engineers Civil of Society Japan the 2002, In In January 2011, the Headquarters for Earthquake Re- for Earthquake Headquarters the 2011, January In uuhm Dicis xsec cicdd ih melt- with coincided existence Daiichi’s Fukushima A report published by the Carnegie Endowment ana- Endowment Carnegie the by published report A of institution the prompted incidents nuclear These 18 19 18 19 The plant at Blayais wasBlayais at plant The The report makes a series The plant at Blayais wasBlayais at plant The The report makes a series ------one of its plants located approximately 100 miles south south miles 100 approximately located plants its of one struck tsunami The improvements. such out carrying of Japan Atomic Power Company (JAPC), was in the process suchfact topointing the actions, utility,Japanese one that walls to protect against asevere tsunami. sea- and dikes built having not and plant; the and plies sup power emergency between connections watertight established having not site; plant the on ground higher emergency movedto sources poweremergency other havingand generators diesel not compartments; or ings build watertight bunkered, dedicated, in pumps cooling forhavingnotemergency installed power equipmentand TEPCO faults report The plant. Daiichi Fukushima the of lifetime operating four-decade the during cumulated learned from the body of international knowledge that ac perts suggest that TEPCO did not put into practice lessons ex These program. powerJapan’sknowledgeas nuclear of well as IAEA, the at and Japan of outside programs perts who had many years of exexperience - in nuclear power safety and executives by made decisions TEPCO’s 2011, four days7, before disaster. the March on did it which results, its of NISA ing inform- before years three waited TEPCO Moreover, so. did never it However, Engineers. Civil of Society Japan the with further issue investigateto the planned therefore and reliable, be to simulation the consider not did it that indicated TEPCO time, the At 869. year the in occurred that earthquake the on based were simulations Fukushima sociated the as- to The underestimated. severely been hazard had plant Daiichi tsunami the that suggested tsunamistruck. until the successfully, part most the for operating, begin did tors genera- diesel emergency The earthquake. the stroyedby de- were all and Japan’sgrid, to lines power external six As it happened, the power plant at Fukushima Daiichi had power wouldfrom the grid be restored within 30 minutes. electrical in loss any that assumption implicit prevalent the characterized executivewho Japanese a of comments the describes report The supply.power country’s the of tsunamireachedthe slopedarea. the when occur would that runup the of height the mating underesti- foranalysis TEPCO’s faults report Carnegie from 10 feet feet (5.7 meters from 3.1 meters). However, 18 the to Daiichi Fukushima the at tsunami increase design-basis to TEPCO prompted methodology That one of its plants located approximately 100 miles south south miles 100 approximately located plants its of one struck tsunami The improvements. such out carrying of Japan Atomic Power Company (JAPC), was in the process suchfact topointing the actions, utility,Japanese one that walls to protect against asevere tsunami. sea- and dikes built having not and plant; the and plies sup power emergency between connections watertight established having not site; plant the on ground higher emergency to moved sources poweremergency other and havinggenerators diesel not compartments; or ings build watertight bunkered, dedicated, in pumps cooling forhavingnotemergency installed power equipmentand TEPCO faults report The plant. Daiichi Fukushima the of lifetime operating four-decade the during cumulated learned from the body of international knowledge that ac perts suggest that TEPCO did not put into practice lessons ex These program. powerJapan’sknowledge nuclear of as well as IAEA, the at and Japan of outside programs perts who had many years of experience exin - nuclear power safety and executives by made decisions TEPCO’s 2011, four days before disaster.7, the March on did it which results, its of NISA ing inform- before years three waited TEPCO Moreover, so. did never it However, Engineers. Civil of Society Japan the with further issue investigateto the planned therefore and reliable, be to simulation the consider not did it that indicated TEPCO time, the At 869. year the in occurred that earthquake the on based were simulations Fukushima sociated the as- The to underestimated. severely been had hazard plant Daiichi tsunami the that suggested tsunamistruck. until the successfully, part most the for operating, begin did tors genera- diesel emergency The earthquake. the stroyedby de- were all and Japan’sgrid, to lines power external six As it happened, the power plant at Fukushima Daiichi had power wouldfrom the grid be restored within 30 minutes. electrical in loss any that assumption implicit prevalent the characterized executivewho Japanese a of comments the describes report The supply.power country’s the of tsunamireachedthe slopedarea. the when occur would that runup the of height the mating underesti- foranalysis TEPCO’s faults report Carnegie from 10 feet feet(5.7 meters from 3.1 meters). However, 18 the to Daiichi Fukushima at the tsunami increase design-basis to TEPCO prompted methodology That The Carnegie report emphasizes the value of taking taking of value the emphasizes report Carnegie The about judgments summarizes report Carnegie The that modeling computer performed TEPCO 2008, In Within Japan there is great confidence in the reliability The Carnegie report emphasizes the value of taking taking of value the emphasizes report Carnegie The about judgments summarizes report Carnegie The that modeling computer performed TEPCO 2008, In Within Japan there is great confidence in the reliability ------2/7/17 6:20PM she77208_ch02_032-057.indd 57 2/7/17 6:20PM ae bgn epn it dmgd eco buildings reactor damaged into seeping began water ment of floors reactor buildings after the accident, ground into plant facilities. Because cracks developed in the base- groundwater of incursions from threat the increasedwhich ocean, the to plant the from distance the shorten to side awaycut theyhill- plant, a the built TEPCO When lenge. confidence they safety hadinthe confidence power oftheir plants. the populations local to communicate to order in accident an eventof the in liability unlimited to firms their expose to chose owners plant Japanese that namely confidence,” “excessive illustrate to example another provides report onlyThe less. or minutes guide 30 of safetyinterruptions power described 1990 the in blackouts about dis cussion the that notes report The accident.” severe a suffer never would program power nuclear Japan’s that makers decision by overconfidence “supreme illustrate to serves accident at Fukushima the experts who stated that Carnegie report communicates the views of some Japanese have assessments played norole inactualdecisionmaking. probabilistic concept, the examined have makers decision Japanese some while And events. external severe of threat the against protected are plants in 10,000” or “1 in a million,” in order to demonstrate that do not require probabilistic safety assessments, such as “1 public discussionormediacoverage abouttsunami safety. in interestwas little meltdown,Daiichithere Fukushima the bias against openly to discussingworst-case Prior scenarios. cultural Japanese general a of presence the noted regulators tial danger posedby tsunamisandotherexternalevents. poten- the recognize slowerto been has country the trast, con- In structures. engineering civil its of all indeed and plants, power nuclear its for requirements technical bust ro- and firm has country the result a activity.As seismic in Japan is highly informed about dangers associated with and perceptions about tsunamis. In this regard, the culture earthquakes about perceptionsbetween asymmetry the to pertains comment first The suggestions. and comments offer does it assessment, to risk TEPCO’s contributed in shortcomings that factors various the of prioritization planned, preventing a bad situation from becoming worse. where pipes had been made watertight, they functioned as in two pits housed flooding pumpthat the rooms. In areas againstseawallprotect a to build to and pump watertight rooms its make to steps taken had JAPC Fukushima. of

h acdn rsle i a eomu cenp chal- cleanup enormous an in resulted accident The The characterization. to pertains comment fourth The regulations Japanese most that is comment third The The second comment is that both Japanese executives and conclusive no offers report Carnegie the Although Daiichi - -

aito-oioig ru caatrzd h ie wall ice play.”strategy as“aHailMary the characterized group radiation-monitoring independent an from researcher a regard, this In steel. or concrete from built walls as such solutions conventional over advantages few very offered and effective, pletely expensive,wascom- more muchwallnotsolution ice the that suggested Some issues. cleanup complex of myriad crewsthe cleanup address could that yearsfive leastso at for protection provide to intended was and million, $320 of cost a at government central Japanese the by built was wall The barrier. a as serve would that wall ice mafrost per- artificial an create to order in ground of area large a from crews cleaning upreactor fuel.In2016,TEPCO began to freeze TEPCO prevented has water accumulated the Moreover, contamination. ocean limit to site, on ers radioactivelargeinto collected be to had has and contain- became water The problems. of kinds all creating and Case AnalysisQuestions Given a standard of 1 in 10,000 years, use the binomial 2. Identify the psychological phenomena in the minicase. 1. lative reactor-years. lative cumu- 15,000 in meltdowns” “core more or three be will there that probability the compute to distribution in 33countries. operation power nuclear commercial of reactor-years cumulative 15,000 over in occurred have to cidents ac major only the are These radioactivity. of release some allowing containment, the tested severely third intense fire without provision for containment, and the an involved next the con anyone, to harm was without tained One Daiichi. Fukushima and Island, Chernobyl, Mile power—Three nuclear civil of history the in accidents reactor major three been have there at supply 2015, to of As electricity. total their of one-quarter least energy nuclear on relied countries 13 2014, In 2012. in production electricity world’s the of percent 10.9 provided plants power Nuclear tries. coun 15 in construction under are plants nuclear new 67 and generation electricity for reactors nuclear 438 As of July 2015, 30 countries worldwide are operating design ofanuclearreactorsuchasFukushimaDaiichi. the planning when assessments safety probabilistic ing avoid of implications the discuss answer, your of part As minicase. the in role their describing then and ena, important. Begin your answer by defining the phenom- least to important most from phenomena the Prioritize when thestandardis1in10million? Introduction toBehavioralAnalysis 21 How does your answer change answer your does How 20 57 - - - -

2/7/17 6:20PM www.mhhe.com/shefrin2e 2/7/17 6:20 PM

2/7/17 6:20 PM

------5757 5757 20 20 How does your answer change How does your answer change 21 Introduction to Behavioral Analysis to Behavioral Analysis Introduction Introduction 21 Introduction to Behavioral Analysis to Behavioral Analysis Introduction Introduction when the standard is 1 in 10 million? Prioritize the phenomena from most important to least phenom- the defining by answer your Begin important. ena, and then describing their role in the minicase. As part of your answer, discuss the implications of avoid ing probabilistic safety assessments when planning the Daiichi. design of a nuclear reactor such as Fukushima operating are worldwide countries 30 2015, July of As 438 nuclear reactors for electricity generation and 67 new nuclear plants are under construction in 15 coun tries. Nuclear power plants provided 10.9 percent of the world’s electricity production in 2012. In 2014, 13 countries relied on nuclear energy least one-quarter of their total electricity. As of 2015, to supply at there have been three major reactor accidents in the history of civil nuclear power—Three Mile Chernobyl, Island, and Fukushima Daiichi. One tained without was harm to anyone, con the next involved an the and containment, for provision without fire intense third severely tested the containment, allowing some release of radioactivity. These are the only major ac cidents to have occurred in over 15,000 cumulative reactor-years of commercial nuclear power operation in 33 countries. distribution to compute the probability that there will be three or more “core meltdowns” in 15,000 cumu- lative reactor-years. when the standard is 1 in 10 million? Prioritize the phenomena from most important to least phenom- the defining by answer your Begin important. ena, and then describing their role in the minicase. As part of your answer, discuss the implications of avoid ing probabilistic safety assessments when planning the Daiichi. design of a nuclear reactor such as Fukushima operating are worldwide countries 30 2015, July of As 438 nuclear reactors for electricity generation and 67 new nuclear plants are under construction in 15 coun tries. Nuclear power plants provided 10.9 percent of the world’s electricity production in 2012. In 2014, 13 countries relied on nuclear energy least one-quarter of their total electricity. As of to 2015, supply at there have been three major reactor accidents in the history of civil nuclear power—Three Mile Chernobyl, Island, and Fukushima Daiichi. One tained without was harm to anyone, con the next involved an the and containment, for provision without fire intense third severely tested the containment, allowing some release of radioactivity. These are the only major ac cidents to have occurred in over 15,000 cumulative reactor-years of commercial nuclear power operation in 33 countries. distribution to compute the probability that there will be three or more “core meltdowns” in 15,000 cumu- lative reactor-years. 1. minicase. the in phenomena psychological the Identify 2. binomial the use years, 10,000 in 1 of standard a Given Case Analysis Questions and creating all kinds of problems. The water became contain- and has had to be collected into large radioactive ers on site, to limit ocean contamination. Moreover, the accumulated water has prevented TEPCO freeze to began fuel. In 2016, TEPCO cleaning up reactor crews from a large area of ground in order to create an artificial per- mafrost ice wall that would serve as a barrier. The wall was built by the Japanese central government at a cost of $320 million, and was intended to provide protection for at so least five years thatcould address cleanup thecrews myriad of complex cleanup issues. Some suggested that the ice solution not wall much more com- was expensive, pletely effective, and offered very few advantages over conventional solutions such as walls built from concrete or steel. In this regard, a researcher from an independent radiation-monitoring group characterized the as “a Hail Marystrategy play.” ice wall 1. minicase. the in phenomena psychological the Identify 2. binomial the use years, 10,000 in 1 of standard a Given Case Analysis Questions and creating all kinds of problems. The water became contain- and has had to be collected into large radioactive ers on site, to limit ocean contamination. Moreover, the accumulated water has prevented TEPCO freeze to began fuel. In 2016, TEPCO cleaning up reactor crews from a large area of ground in order to create an artificial per- mafrost ice wall that would serve as a barrier. The wall was built by the Japanese central government at a cost of $320 million, and was intended to provide protection for at so least five years thatcould address cleanup thecrews myriad of complex cleanup issues. Some suggested that the ice solution not wall much more com- was expensive, pletely effective, and offered very few advantages over conventional solutions such as walls built from concrete or steel. In this regard, a researcher from an independent radiation-monitoring group characterized the as “a Hail Marystrategy play.” ice wall

- -

- - Daiichi Daiichi Daiichi Daiichi The accident resulted in an enormous cleanup chal- Although the Carnegie report offers no conclusive and executives Japanese both that is comment second The The third comment is that most Japanese regulations The fourth comment pertains to characterization. The

cussion about blackouts in the 1990 described power interruptions safety of 30 minutes guide or less. The only report provides another example to illustrate “excessive confidence,” namely that Japanese plant owners chose to expose their firms to unlimited liability in the of event an accident in order to communicate to local populations the plants. of their power confidence had in the safety they serves serves to illustrate “supreme overconfidence by decision makers that Japan’s nuclear power program would never suffer a severe accident.” The report notes that the dis prioritization of the various factors that shortcomings in contributed TEPCO’s risk to assessment, it does offer comments and suggestions. The first comment pertains to the asymmetry between perceptions about earthquakes culture the regard, this In tsunamis. about perceptions and withassociated dangers about informed highly is Japan in seismic As activity. a result the country has firm and ro- bust technical requirements for its nuclear power plants, and indeed all of its civil engineering structures. In con- trast, the country has been to slower recognize the poten- tsunamis and other external events. posed by tial danger regulators noted the presence of a general Japanese cultural scenarios. Prior discussing worst-case to openly bias against the Fukushima thereDaiichi meltdown, little was interest in about tsunami safety. public discussion or media coverage “1 as such assessments, safety probabilistic require not do that demonstrate to order in million,” a in “1 or 10,000” in plants are protected against the threat of severe external events. And while some Japanese decision makers have examined the concept, probabilistic in actual decision making. no role played assessments have Japanese some of views the communicates report Carnegie statedthatexpertswho the Fukushimaat accident of Fukushima. JAPC had taken steps to make its rooms watertight and pump to build to a protect seawall against areas In rooms. the thatpump floodinghoused pits two in as functioned they watertight, made been had pipes where worse. becoming from situation bad a preventing planned, water water began seeping into damaged reactor buildings lenge. lenge. When TEPCO built the a plant, hill- they cut away side to shorten the distance from the plant to the ocean, which increased the threat from incursions of groundwater base- the in developed cracks Because facilities. plant into ground accident, the after buildings reactor floorsof ment

The accident resulted in an enormous cleanup chal- Although the Carnegie report offers no conclusive and executives Japanese both that is comment second The The third comment is that most Japanese regulations The fourth comment pertains to characterization. The

serves serves to illustrate “supreme overconfidence by decision makers that Japan’s nuclear power program would never suffer a severe accident.” The report notes that the cussion dis about blackouts in the 1990 described power interruptions safety of 30 guide minutes or less. The only report provides another example to illustrate “excessive confidence,” namely that Japanese plant owners chose to expose their firms to unlimited liability in the of event an accident in order to communicate to local populations the plants. of their power confidence had in the safety they of Fukushima. JAPC had taken steps to make its rooms watertight pump and to build to a protect seawall against areas In rooms. the thatpump floodinghoused pits two in as functioned they watertight, made been had pipes where worse. becoming from situation bad a preventing planned, prioritization of the various factors that shortcomings in contributed TEPCO’s risk to assessment, it does offer comments and suggestions. The first comment pertains to the asymmetry between perceptions about earthquakes culture the regard, this In tsunamis. about perceptions and withassociated dangers about informed highly is Japan in seismic As activity. a result the country has firm and ro- bust technical requirements for its nuclear power plants, and indeed all of its civil engineering structures. In con- trast, the country has been to slower recognize the poten- events. tsunamis and other external posed by tial danger regulators noted the presence of a general Japanese cultural scenarios. Prior discussing worst-case to openly bias against the Fukushima thereDaiichi meltdown, little was interest in about tsunami safety. public discussion or media coverage “1 as such assessments, safety probabilistic require not do that demonstrate to order in million,” a in “1 or 10,000” in plants are protected against the threat of severe external events. And while some Japanese decision makers have examined the concept, probabilistic in actual decision making. no role played assessments have Japanese some of views the communicates report Carnegie statedthatexpertswho the Fukushimaat accident lenge. lenge. When TEPCO built the a plant, hill- they cut away side to shorten the distance from the plant to the ocean, which increased the threat from incursions of groundwater base- the in developed cracks Because facilities. plant into ground accident, the after buildings reactor floorsof ment water began seeping into damaged reactor buildings

she77208_ch02_032-057.indd 57 2/7/17 6:20 PM she77208_ch02_032-057.indd 57 - - - - Within Japan there is great confidence in the reliability reliability the in confidence great is there Japan Within In 2008, TEPCO performed computer modeling that The Carnegie report summarizes judgments about The Carnegie report emphasizes the value of taking That methodology prompted TEPCO to design-basis increase tsunami the at Fukushima Daiichi to the 18 However, meters). 3.1 from meters (5.7 feet feet 10 from Carnegie report faults TEPCO’s analysis for - underesti mating the height of the runup that would occur when the sloped area. the tsunami reached of the country’s power supply. The report describes the comments of a Japanese who executive characterized the prevalent implicit assumption that any loss in electrical minutes. within30 restored be thegrid from would power had Daiichi Fukushima at plant power the happened, it As six external power lines to grid,Japan’s and all were de- by stroyed the earthquake. The emergency diesel genera- tors did begin operating, for the most part successfully, until the tsunami struck. suggested that the tsunami Daiichi plant hazard had been severely underestimated. to The as- the sociated Fukushima simulations were based on the earthquake that occurred in the year 869. At the time, TEPCO indicated that it did not consider the simulation to be reliable, and therefore planned the to investigate issue further with the Japan Society of Civil Engineers. However, it never did so. Moreover, TEPCO waited three years before inform- ing NISA of its results, which it did on March the7, disaster. before days four 2011, TEPCO’s decisions made by executives and safety power nuclear - in ex experience of years many had who perts programs outside of Japan and at the IAEA, as well as of nuclear knowledge Japan’s power program. These ex lessons practice into put not did TEPCO that suggest perts ac that knowledge international of body the from learned cumulated during the four-decade operating lifetime of the Fukushima Daiichi plant. The report faults TEPCO and equipment power installed emergency not having for cooling pumps in dedicated, bunkered, watertight build ings or compartments; not diesel generators having and other emergency power sources moved to emergency higher ground on the plant site; not having established watertight connections between emergency power sup plies and the plant; and not having built dikes and sea- tsunami. a severe against protect to walls that one Japanese utility, actions, pointing theto fact such process the in was (JAPC), Company Power Atomic Japan of carrying out such improvements. The tsunami struck one of its plants located approximately 100 miles south - - - - - The report makes a series a makes report The The plant at Blayais was 19 18 These nuclear incidents prompted the institution of A report published by the Carnegie Endowment ana- Fukushima Daiichi’s existence coincided with melt- In January 2011, the Headquarters Earthquake for Re- In 2002, the Japan Society of Civil Engineers de protected protected from storm by surges, dikes. In 1999 an eight- (2.5 foot-high meter) storm surge overflowed the plant’s dikes and flooded a pumping station. In order to prevent a reactor meltdown, the pumping system needs to carry the case this in “sink,” heat a to rods reactor the from heat in A the can breakdown pumping system estuary. nearby decided Blayais at makers decision The dangerous. prove that it was prudent to shut down all reactors at the site. author French conducted by an investigation Thereafter, ities not found only that the dikes but were too also low, that the rooms which contained emergency equipment flooding. from protected sufficiently not were Nuclear U.S. The globe. the across standards safety stricter meet designs reactor that specifies Commission Regulatory a 1-in-10,000-year core damage The frequency. threshold requirement for U.S. utilities is 1 2015, the of at best operating 1 plants inin operation were in 100,000 years. In ofstandard a to built be likely will plants Future million. 1 1 in 10 million. approximately 56 56 Chapter Two Chapter Two academic research sedimentarydiscovered sug layers gesting that massive tsunamis occur once every thou sand years or so, with the most recent episode having occurred in the year 869, as a result of an earthquake measuring scale. 8.3 on the Richter Is Mile Three at one plants, power nuclear two at downs 1986. in Ukraine Chernobyl, at other the and 1979 in land nor Island Mile Three neither Daiichi, Fukushima Unlike seawater using and coast sea a on located was Chernobyl Blayais at located France in plant a However, cooling. for did so, and experienced an incident that was relevant to the Fukushima Daiichi plant. of points that summarized are below. lyzed thelyzed judgments and decisions in connection with the meltdown. Daiichi Fukushima search search Promotion, which is funded by the Japanese gov- ernment, repeated its judgment that there is a 99 percent probability that a magnitude 7.5 earthquake would strike the region within 30 years. Given that the magnitude of the actual earthquake was 9, the report characterizes the outcome as “a sobering warning against overconfidence prediction.” in hazard veloped veloped a methodology to determine “runup,” ing mean- how a tsunami would behave when it reached the sloped area on which the nuclear power plants rested. she77208_ch02_032-057.indd 56 Valuation 59

Chapter Three The intrinsic value of a firm’s equity, VE, can be obtained in several ways. The first is by subtracting the intrinsic value of the firm’s debt from its intrinsic value VF. The second is by using the dividend discount method (DDM), which involves computing the present value of the firm’s expected dividend stream, discounted at the required return on equity kE. Chapter Three A third way to compute VE is through the formula E1∕kE + PVGO, where E1 de- notes expected earnings next period and PVGO represents the firm’s present value of growth opportunities.1 A firm has positive growth opportunities when the ex- Valuation pected return on its future projects exceeds the required return on those projects. Effectively, PVGO is the net present value of a firm’s current and future projects. The main objective of this chapter is for students to demonstrate that they can iden- A fourth way to compute VE is through the formula BE + PV(expected EVA stream). tify how heuristics and framing affect the way managers and analysts value firms. Here BE is a firm’s current book value of equity and EVA stands for economic value AfterValuation completing this chapter students will be able to: added. Economic value added for the firm is the difference between net operating 1. Explain why some financial executives and analysts rely on valuation heuristics profit after tax (NOPAT, effectively earnings before interest and tax minus tax), and a instead of textbook techniques that emphasize intrinsic value. charge to capital defined as the product of the cost of capital and book value of assets. The main objective of this chapter is for students to demonstrate that they can iden- Notably, the PVGO approach provides a formula for a firm’s intrinsic forward 2. Describe the main heuristics that financial executives and analysts use to compute −1 tify how heuristics and framing affect the way managers and analysts value firms. P∕E ratio V∕E1, namely [kE × (1 − PVGO∕VE)] . This formula implies that intrin- value. After completing this chapter students will be able to: sic P∕E is decreasing in the firm’s required return on equity kE and increasing in the 3. Identify the biases that arise in connection with the use of valuation heuristics. proportion of its intrinsic value that stems from its growth opportunities. Similarly, 1. Explain why some financial executives and analysts rely on valuation heuristics 4. Identify the biases that arise in connection with the use of traditional textbook the economic value added approach provides a formula for a firm’s intrinsic market- instead of textbook techniques that emphasize intrinsic value. techniques that emphasize intrinsic value. to-book ratio VE∕BE, namely [1 + PV(EVA stream)∕BE]. This formula implies that 2. Describe the main heuristics that financial executives and analysts use to compute when a firm expects that in the future it will earn exactly its cost of capital, then its value. intrinsic market-to-book ratio will be unity. 3. Identify the biases that arise in connection with the use of valuation heuristics. valuation by Valuation by comparables focuses on the relationship between the market 3.1 TRADITIONAL4. APPROACH Identify the biases thatTO arise VALUATION in connection with the use of traditional textbook comparables value of a firm’s equity, and the characteristics of that firm relative to similar firms techniques that emphasize intrinsic value. Valuation based on to see if the relationships are comparable. For example, in valuing a specific firm’s intrinsic valuation valua- The two most prominent approaches to valuation are and comparison of a specific stock, financial analysts might use the valuation formula P = P∕E × E, where P∕E tion by comparables. The former is based on discounted cash flow (DCF) analysis, firm’s characteristics with is their judgment of the appropriate value of forward-looking P0∕E1 and E1 is their while the latter is mostly based on ratios such as price-to-earnings and price-to- the characteristics of comparable firms estimate of the firm’s earnings over the next year. What estimate of P∕E to use in 3.1 TRADITIONALsales. APPROACH TO VALUATION the analysis, and how much confidence to have in this estimate, might depend on The most general DCF approach to the intrinsic valuation of a firm involves its how the firm’s “trailing P∕E” (written as P0 ∕E0 and defined as current price divided Thefuture two free most cash prominent flows. The approaches annual free to cashvaluation flow toare a intrinsicfirm is the valuation cash the and firm valua- gen- by current earnings), compares to the trailing P∕E ratios of similar firms. FCF tionerates by in comparables a year that is. The available former to is be based paid onto discountedits investors. cash Free flow cash (DCF) flow (analysis,) is In addition, analysts tend to repeat the analysis for other ratios such as price-to- whilegenerated the fromlatter ais firm’s mostly operating based on activities, ratios such net asof investmentsprice-to-earnings in fixed and assets price-to- and sales, price-to-book (also known as market-to-book), price-to-cash flow, and PEG sales.changes in working capital. The standard textbook formula for free cash flow is defined as P∕E divided by the estimated growth rate of a firm. Doing so provides a EBITThe(1 most− tax general rate) + DCF depreciation approach − to investment the intrinsic − changevaluation in ofworking a firm capital involves. The its V range of possible values for a firm’s stock, leading each analyst to make a judgment futureintrinsic free value cash flows.F of a firm The isannual the present free cash value flow of itsto expecteda firm is thefree cash cash the flow firm stream, gen- call about where in the associated range to place his or her final valuation estimate. eratesdiscounted in a yearat its that cost is of available capital ktoW ,be plus paid the to value its investors. of its current Free cashcash. flow It is common(FCF) is generatedto divide the from free a firm’scash flow operating into two activities, horizons, net respectively of investments called in thefixed intermediate assets and changeshorizon andin working the terminal capital. horizon. The standard During textbookthe terminal formula horizon, for freewhich cash begins flow inis 3.2 TARGET PRICE HEURISTICS EBITperiod(1 T − + tax 1, ratefree) cash + depreciation flow is expected − investment to grow − at change a constant in working rate g, withcapital present. The value PVT beingV given by the formula FCFT+1∕(kW − g). A fuller discussion of free intrinsic value F of a firm is the present value of its expected free cash flow stream, The formulas underlying present value analysis have been rigorously derived as part cash flows can be found in the Additionalk Resources to Chapter 3, which is posted discounted at its cost of capital W, plus the value of its current cash. It is common of a well-established theory. Nevertheless, in practice, equity valuation strongly on the book web site. to divide the free cash flow into two horizons, respectively called the intermediate reflects the use of heuristics and therefore vulnerability to biases. In this regard, 58 horizon and the terminal horizon. During the terminal horizon, which begins in valuation by comparables is a heuristic technique: it is based on formulas that are period T + 1, free cash flow is expected to grow at a constant rate g, with present value PVT being given by the formula FCFT+1∕(kW − g). A fuller discussion of free cash flows can be found in the Additional Resources to Chapter 3, which is posted on the book web site. 58 she77208_ch03_058-084.indd 58 2/7/17 5:26 PM

she77208_ch03_058-084.indd 59 2/7/17 5:26 PM

she77208_ch03_058-084.indd 58 2/7/17 5:26 PM Valuation 59

Chapter Three The intrinsic value of a firm’s equity, VE, can be obtained in several ways. The first is by subtracting the intrinsic value of the firm’s debt from its intrinsic value VF. The second is by using the dividend discount method (DDM), whichValuation involves 59 computing the present value of the firm’s expected dividend stream, discounted at the requiredThe intrinsic return valueon equity of a firm’s kE. equity, VE, can be obtained in several ways. The first Chapter Three Ais thirdby subtracting way to compute the intrinsic VE is value through of the the firm’s formula debt E from1∕kE its + intrinsicPVGO, wherevalue V EF.1 de- notes expectedThe second earnings is by using next periodthe dividend and PVGO discount represents method the(DDM), firm’s which present involves value Valuation of growthcomputing opportunities. the present1 valueA firm of thehas firm’spositive expected growth dividend opportunities stream, when discounted the ex- at pectedthe requiredreturn on return its future on equity projects kE. exceeds the required return on those projects. Effectively,A third PVGO way to is compute the net present VE is through value ofthe a formula firm’s current E1∕kE + and PVGO future, where projects. E1 de- The main objective of this chapter is for students to demonstrate that they can iden- Anotes fourth expected way to computeearnings VnextE is periodthrough and the PVGO formula represents BE + PV the(expected firm’s EVApresent stream value). tify how heuristics and framing affect the way managers and analysts value firms. Hereof BE growth is a firm’s opportunities. current 1book A firm value has of positive equity andgrowth EVA opportunitiesstands for economic when the value ex- After completing this chapter students will be able to: added.pected Economic return on value its futureadded projects for the exceedsfirm is thethe differencerequired return between on those net operating projects. Valuation PVGO 1. Explain why some financial executives and analysts rely on valuation heuristics profitEffectively, after tax (NOPAT, is theeffectively net present earnings value beforeof a firm’s interest current and andtax minusfuture tax),projects. and a V BE PV expected EVA stream instead of textbook techniques that emphasize intrinsic value. chargeA to fourth capital way defined to compute as the productE is through of the the cost formula of capital + and( book value of assets.). The main objective of this chapter is for students to demonstrate that they can iden- Notably,Here BE theis a PVGOfirm’s current approach book provides value of aequity formula and forEVA a standsfirm’s for intrinsic economic forward value 2. Describe the main heuristics that financial executives and analysts use to compute −1 tify how heuristics and framing affect the way managers and analysts value firms. P∕Eadded. ratio VEconomic∕E1, namely value [k Eadded × (1 −for PVGO the firm∕VE is)] the. This difference formula between implies net that operating intrin- value. After completing this chapter students will be able to: sic P∕Eprofit is after decreasing tax (NOPAT, in the effectively firm’s required earnings return before on interestequity kandE and tax increasing minus tax), in and the a 3. Identify the biases that arise in connection with the use of valuation heuristics. proportioncharge to of capital its intrinsic defined value as the that product stems of from the cost its ofgrowth capital opportunities. and book value Similarly, of assets. 1. Explain why some financial executives and analysts rely on valuation heuristics 4. Identify the biases that arise in connection with the use of traditional textbook the economicNotably, value the PVGO added approach provides a formula for a firm’sfirm’s intrinsicintrinsic market-forward instead of textbook techniques that emphasize intrinsic value. −1 techniques that emphasize intrinsic value. to-bookP∕E ratio VV∕EE∕1BE, namely, namely [kE [ ×1 +(1 PV− PVGO(EVA stream∕VE)] )∕. ThisBE]. formulaThis formula implies implies that intrin- that 2. Describe the main heuristics that financial executives and analysts use to compute whensic a P∕E firm is expects decreasing that in in the the firm’s future required it will earnreturn exactly on equity its costkE and of increasingcapital, then in theits value. intrinsicproportion market-to-book of its intrinsic ratio value will that be unity.stems from its growth opportunities. Similarly, 3. Identify the biases that arise in connection with the use of valuation heuristics. valuation by Valuationthe economic by value comparables added approach focuses provides on the a formula relationship for a firm’s between intrinsic the market- market 3.1 TRADITIONAL4. APPROACH Identify the biases thatTO arise VALUATION in connection with the use of traditional textbook comparables valueto-book of a firm’s ratio Vequity,E∕BE, andnamely the [characteristics1 + PV(EVA stream of that)∕ BEfirm]. Thisrelative formula to similar implies firms that techniques that emphasize intrinsic value. Valuation based on to seewhen if the a firm relationships expects that are in comparable. the future it willFor example,earn exactly in valuingits cost of a specificcapital, then firm’s its intrinsic valuation valua- The two most prominent approaches to valuation are and comparison of a specific stock,intrinsic financial market-to-book analysts might ratio use will the be valuation unity. formula P = P∕E × E, where P∕E tion by comparables. The former is based on discounted cash flow (DCF) analysis, firm’svaluation characteristics by with Valuation by comparables the characteristics of is their judgment of the appropriate focuses value of on forward-looking the relationship P 0 between∕E1 and theE1 is market their while the latter is mostly based on ratios such as price-to-earnings and price-to- comparables comparable firms estimatevalue ofof thea firm’s firm’s equity, earnings and theover characteristics the next year. of Whatthat firm estimate relative of to P ∕similarE to use firms in 3.1 TRADITIONALsales. APPROACH TO VALUATION Valuation based on the toanalysis, see if the and relationships how much are confidence comparable. to haveFor example, in this estimate, in valuing might a specific depend firm’s on The most general DCF approach to the intrinsic valuation of a firm involves its comparison of a specific P P E E P E firm’s characteristics with howstock, the firm’s financial “trailing analysts P∕E” might (written use the as valuationP0 ∕E0 and formula defined as= current∕ × price, where divided ∕ Thefuture two free most cash prominent flows. The approaches annual free to cashvaluation flow toare a intrinsicfirm is the valuation cash the and firm valua- gen- by currentis their judgmentearnings), of compares the appropriate to the trailingvalue of P∕E forward-looking ratios of similar P0∕ Efirms.1 and E1 is their FCF the characteristics of tionerates by in comparables a year that is. The available former to is be based paid onto discountedits investors. cash Free flow cash (DCF) flow (analysis,) is comparable firms Inestimate addition, of theanalysts firm’s tend earnings to repeat over thethe analysisnext year. for What other estimate ratios such of P ∕asE price-to-to use in whilegenerated the fromlatter ais firm’s mostly operating based on activities, ratios such net asof investmentsprice-to-earnings in fixed and assets price-to- and sales,the price-to-book analysis, and how(also much known confidence as market-to-book), to have in this price-to-cash estimate, might flow, depend and PEG on changes in working capital. The standard textbook formula for free cash flow is sales. definedhow theas P∕E firm’s divided “trailing by P∕E”the estimated (written as growth P0 ∕E0 rate and ofdefined a firm. as Doingcurrent so price provides divided a EBITThe(1 most− tax general rate) + DCF depreciation approach − to investment the intrinsic − changevaluation in ofworking a firm capital involves. The its by current earnings), compares to the trailing P∕E ratios of similar firms. V range of possible values for a firm’s stock, leading each analyst to make a judgment futureintrinsic free value cash flows.F of a firm The isannual the present free cash value flow of itsto expecteda firm is thefree cash cash the flow firm stream, gen- call aboutIn addition, where in analysts the associated tend to repeatrange theto place analysis his foror herother final ratios valuation such as estimate. price-to- eratesdiscounted in a yearat its that cost is of available capital ktoW ,be plus paid the to value its investors. of its current Free cashcash. flow It is common(FCF) is sales, price-to-book (also known as market-to-book), price-to-cash flow, and PEG togenerated divide the from free a firm’scash flow operating into two activities, horizons, net respectively of investments called in thefixed intermediate assets and defined as P∕E divided by the estimated growth rate of a firm. Doing so provides a horizonchanges andin working the terminal capital. horizon. The standard During textbookthe terminal formula horizon, for freewhich cash begins flow inis 3.2 TARGET PRICE rangeHEURISTICS of possible values for a firm’s stock, leading each analyst to make a judgment periodEBIT(1 T − + tax 1, ratefree) cash + depreciation flow is expected − investment to grow − at change a constant in working rate g, withcapital present. The call about where in the associated range to place his or her final valuation estimate. value PVT beingV given by the formula FCFT+1∕(kW − g). A fuller discussion of free intrinsic value F of a firm is the present value of its expected free cash flow stream, The formulas underlying present value analysis have been rigorously derived as part cash flows can be found in the Additionalk Resources to Chapter 3, which is posted discounted at its cost of capital W, plus the value of its current cash. It is common of a well-established theory. Nevertheless, in practice, equity valuation strongly on the book web site. to divide the free cash flow into two horizons, respectively called the intermediate 3.2 TARGET PRICEreflects HEURISTICSthe use of heuristics and therefore vulnerability to biases. In this regard, 58 horizon and the terminal horizon. During the terminal horizon, which begins in valuation by comparables is a heuristic technique: it is based on formulas that are period T + 1, free cash flow is expected to grow at a constant rate g, with present The formulas underlying present value analysis have been rigorously derived as part value PVT being given by the formula FCFT+1∕(kW − g). A fuller discussion of free of a well-established theory. Nevertheless, in practice, equity valuation strongly cash flows can be found in the Additional Resources to Chapter 3, which is posted reflects the use of heuristics and therefore vulnerability to biases. In this regard, on the book web site. valuation by comparables is a heuristic technique: it is based on formulas that are 58 she77208_ch03_058-084.indd 58 2/7/17 5:26 PM she77208_ch03_058-084.indd 59 2/7/17 5:26 PM

she77208_ch03_058-084.indd 58 2/7/17 5:26she77208_ch03_058-084.indd PM 59 2/7/17 5:26 PM 60 Chapter Three Valuation 61

true by definition (tautologies), relies on judgments of similarity that depend on 3.3 A CFO’S RELIANCE ON VALUATION HEURISTICS how representative a firm is in the class of its comparables, and reconciles possibly 60 Chapter Three diverse valuation estimates associated with different ratios by using instinct (gut In order to make the ideas about valuation heuristics and biases concrete, this chap- feel or affect). Recall the discussion about representativeness and the affect heuris- ter uses the case of the major online auction firm eBay as a vehicle for discussion. truetic inby Chapters definition 1 (tautologies),and 2. relies on judgments of similarity that depend on eBay’s users use its site to buy and sell items such as antiques, art, jewelry, books, howBoth representative analysts and a firm financial is in the executives class of its rely comparables, on a combination and reconciles of discounted possibly cash collectibles, and automotive parts. 2 diverseflow (DCF) valuation and estimatesvaluation associated by comparables. with different This chapterratios by focuses using instincton the heuristic(gut The initial public offering (IPO) for eBay took place in September 1998. In 2002 feelnature or affect). of valuation Recall themethodologies discussion about and representativenesson the biases associated and the withaffect their heuris- imple- eBay acquired the Internet payment provider PayPal in a $1.4 billion stock deal, ticmentation in Chapters in practice.1 and 2. In this respect, much of the chapter focuses on security ana- thereby providing its users with a safe and reliable way to make their online purchases. lysts’Both estimates analysts andof future financial values executives known relyas “price on a combination targets.” of discounted cash By the end of 2002 eBay was one of the most successful Internet companies in the 2 flowThe (DCF) general and evidencevaluation indicatesby comparables. that analysts This are chapter prone focusesto a series on theof specific heuristic biases world. A $100 investment in eBay stock at the time of its IPO would have grown to naturewhen ofsetting valuation price methodologiestargets. They overweigh and on the past biases sales associatedgrowth and with underweigh their imple- the level more than $640 by May 2003. In contrast, the return to both the S&P 500 and the mentationof past earnings. in practice. They In thistend respect, to overvalue much offirms the chapterthat report focuses losses, on securityparticularly ana- when Nasdaq Composite index during the same period was close to zero, if not negative. lysts’those estimates losses are of extreme.future values They known tend asto “priceundervalue targets.” firms that have large cash flows On May 20, 2003, eBay’s stock closed at $97.75, having risen by more than moreThe thangeneral firms evidence that have indicates large that accruals, analysts and are they prone are to pronea series to ofmisvalue specific firmsbiases with 44 percent from the beginning of the year. That increase gave eBay a market valu- 3 whenhigh settingprofit pricemargins. targets. The They next overweigh part of the past discussion sales growth specifies and underweigh the equations the level used in ation higher than the market valuations of both General Motors and McDonald’s. ofthree past particularearnings. priceThey targettend to heuristics overvalue associated firms that with report valuation losses, byparticularly comparables. when On that day the consensus analyst estimate of its earnings per share (EPS) for the those losses are extreme. They tend to undervalue firms that have large cash flows coming four quarters was $1.56. That produced a forward P∕E for eBay of 62.7.5 moreP∕E than Heuristic firms that have large accruals, and they are prone to misvalue firms with highValuation profit margins.based on3 Thethe P∕Enext ratiopart ofinvolves the discussion the product specifies of two the terms: equations a P∕E used ratio in and threean earnings particular estimate. price target The heuristics P∕E ratio associated is a forward with P∕E, valuation having by thecomparables. form P0∕E1, where CONCEPT After Internet stock prices had generally soared in May 2003, the San Jose Mercury P0 denotes current price and E1 denotes the forecast of earnings per share for the next PREVIEW News asked whether the rapid increase was due to solid fundamentals or 1999-style P E Heuristic 4 P/E heuristic year.∕ Security analysts rely on the P∕E heuristic more than any other technique. Question 3.1 bandwagon investing. Investors were asking the same question. If you were a financial An approach to valuation ValuationThe valuation based on identitythe P∕E isratio given involves by P0 the= P product0∕E1 × ofE1 two. In orderterms: to a forecastP∕E ratio the and price executive at eBay, what combination of traditional and heuristic valuation techniques based on multiplying a anP 1earnings a year hence,estimate. apply The theP∕E relationship ratio is a forward P1 = P∕E,P1∕E having2 × E2 the. Here form the P0 ∕challengeE1, where is to would you use to answer this question? P/E ratio and an earnings P0 denotes current price and E1 denotes the forecast of earnings per share for the next forecast. arrive at sensible forecasts of the P∕E ratio that will apply a year hence and earnings P/E heuristic year.per shareSecurity two analysts years hence. rely on the P∕E heuristic more than any other technique. 4 An approach to valuation The valuation identity is given by P = P ∕E × E . In order to forecast the price 0 0 1 1 Consider next how the financial executives at eBay actually approached valua- based on multiplying a PPEG a year Heuristic hence, apply the relationship P = P ∕E × E . Here the challenge is to 1 1 1 2 2 tion at the time. See the Behavior Pitfalls box “Attaching a Value to eBay.” 6,7 P/E ratio and an earnings arrive at sensible forecasts of the P∕E ratio that will apply a year hence and earnings forecast. A firm’s PEG ratio is defined as its P∕E ratio divided by its expected earnings growth Both eBay’s chief financial officer and vice president for corporate finance were per share two years hence. rate per year (actually 100 × expected earnings growth rate). The premise underlying highly educated and experienced. Despite all the emphasis in corporate finance text- PEGthe use Heuristic of PEG-based valuation is that the stocks of high-growth firms merit higher books on discounted cash flow analysis, many financial executives rely on heuristics P∕E ratios than the stocks of low-growth firms. To compute the value of a firm’s instead of the fundamental valuation techniques taught in textbooks. Heuristics are A firm’s PEG ratio is defined as its P∕E ratio divided by its expected earnings growth stock based on its PEG ratio, multiply the following three terms: its PEG ratio, an simpler to use: P∕E, PEG, and price-to-sales require very few variables and involve rate per year (actually 100 × expected earnings growth rate). The premise underlying estimate of future earnings per share, and an estimate of expected earnings growth. simple formulas. As such they have intuitive appeal, the basis for the affect heuristic. the use of PEG-based valuation is that theP stocks of high-growthE G firmsG merit higher PEG heuristic The valuation identity is given by 0 = PEG × 1 × , where is 100 × growth The discussion in the nearby box vividly illustrates this point. The DCF-based anal- An approach to valuation P∕E ratios than the stocks of low-growthPEG firms. heuristic To compute the value of a firm’s P rate. As with the P∕E heuristic, the can be used to forecast price 1 yses taught in textbooks require far more detail than the heuristic techniques, involve based on multiplying a PEG stocka year based hence. on itsDoing PEG soratio, requires multiply forecasts the following of the PEGthree ratioterms: that its PEGwill applyratio, ana year ratio, an earnings forecast, more complex formulas, and are far less intuitive. estimatehence, earningsof future earningsper share per two share, years and hence, an estimate and the of growth expected rate earnings that will growth. apply over PEGand a heuristicforecast of the growth The valuation identity is given by P = PEG × E × G, where G is 100 × growth rate. the forecast period. 0 1 An approach to valuation rate. As with the P∕E heuristic, the PEG heuristic can be used to forecast price P1 3.4 HOW ANALYSTS VALUE FIRMS: AN ILLUSTRATIVE EXAMPLE basedprice-to-sales on multiplying heuristic a PEG a Price-to-Salesyear hence. Doing Heuristic so requires forecasts of the PEG ratio that will apply a year ratio,An approach an earnings to valuation forecast, and a forecast of the growth hence,The price-to-sales earnings per share heuristic two years has hence,the same and structurethe growth as rate the that P∕E will heuristic, apply over except Analyst Mary Meeker based on multiplying a the forecast period. rate.price-to-sales ratio and a that future sales are substituted for future earnings. The valuation identity is given The Behavioral Pitfalls box on page 62 illustrates how the financial executives at P P S S S price-to-salessales forecast. heuristic Price-to-Salesby 0 = 0 ∕ 1 × Heuristic 1, where stands for sales. eBay approached the value of their firm. Essentially, they relied on the PEG heuristic An approach to valuation price-to-sales heuristic based on multiplying a The has the same structure as the P∕E heuristic, except price-to-sales ratio and a that future sales are substituted for future earnings. The valuation identity is given sales forecast. by P0 = P0 ∕S1 × S1, where S stands for sales.

she77208_ch03_058-084.indd 60 she77208_ch03_058-084.indd2/7/17 5:26 PM 61 2/7/17 5:26 PM

she77208_ch03_058-084.indd 60 2/7/17 5:26 PM 60 Chapter Three Valuation 61 true by definition (tautologies), relies on judgments of similarity that depend on 3.3 A CFO’S RELIANCE ON VALUATION HEURISTICS how representative a firm is in the class of its comparables, and reconciles possibly diverse valuation estimates associated with different ratios by using instinct (gut In order to make the ideas about valuation heuristics and biases concrete,Valuation this chap- 61 feel or affect). Recall the discussion about representativeness and the affect heuris- ter uses the case of the major online auction firm eBay as a vehicle for discussion. tic in Chapters 1 and 2. 3.3 A CFO’S RELIANCEeBay’s users useON its VALUATION site to buy and sell HEURISTICS items such as antiques, art, jewelry, books, Both analysts and financial executives rely on a combination of discounted cash collectibles, and automotive parts. 2 flow (DCF) and valuation by comparables. This chapter focuses on the heuristic TheIn order initial to publicmake theoffering ideas about(IPO) valuationfor eBay heuristicstook place and in biasesSeptember concrete, 1998. this In chap-2002 nature of valuation methodologies and on the biases associated with their imple- eBayter acquireduses the case the Internetof the major payment online provider auction firm PayPal eBay in as a $1.4a vehicle billion for stockdiscussion. deal, mentation in practice. In this respect, much of the chapter focuses on security ana- therebyeBay’s providing users use its itsusers site with to buy a safe and and sell reliable items such way as to antiques,make their art, online jewelry, purchases. books, lysts’ estimates of future values known as “price targets.” By collectibles,the end of 2002 and automotiveeBay was oneparts. of the most successful Internet companies in the The general evidence indicates that analysts are prone to a series of specific biases world. TheA $100 initial investment public offering in eBay (IPO) stock for ateBay the tooktime placeof its in IPO September would have 1998. grown In 2002 to when setting price targets. They overweigh past sales growth and underweigh the level moreeBay than acquired $640 by the May Internet 2003. paymentIn contrast, provider the return PayPal to inboth a $1.4 the S&P billion 500 stock and deal, the of past earnings. They tend to overvalue firms that report losses, particularly when Nasdaqthereby Composite providing index its users during with the a safe same and period reliable was way close to make to zero, their if online not negative. purchases. those losses are extreme. They tend to undervalue firms that have large cash flows OnBy Maythe end 20, of 2003, 2002 eBay eBay’s was stock one closedof the most at $97.75, successful having Internet risen companies by more in than the more than firms that have large accruals, and they are prone to misvalue firms with 44 percentworld. A from $100 the investment beginning in ofeBay the stock year. at That the timeincrease of its gave IPO eBay would a havemarket grown valu- to 3 high profit margins. The next part of the discussion specifies the equations used in ationmore higher than than $640 the by marketMay 2003. valuations In contrast, of both the Generalreturn to Motorsboth the and S&P McDonald’s. 500 and the three particular price target heuristics associated with valuation by comparables. On Nasdaqthat day Composite the consensus index analystduring theestimate same period of its wasearnings close perto zero, share if not(EPS) negative. for the 5 P E Heuristic comingOn four May quarters 20, 2003, was eBay’s$1.56. That stock produced closed at a $97.75, forward having P∕E for risen eBay by of more 62.7. than ∕ 44 percent from the beginning of the year. That increase gave eBay a market valu- Valuation based on the P∕E ratio involves the product of two terms: a P∕E ratio and ation higher than the market valuations of both General Motors and McDonald’s. an earnings estimate. The P∕E ratio is a forward P∕E, having the form P0∕E1, where CONCEPT AfterOn Internet that day stock the pricesconsensus had analystgenerally estimate soared of in itsMay earnings 2003, theper Sanshare Jose (EPS) Mercury for the P0 denotes current price and E1 denotes the forecast of earnings per share for the next PREVIEW Newscoming asked four whether quarters the wasrapid $1.56. increase That was produced due to a solid forward fundamentals P∕E for eBay or 1999-style of 62.7.5 4 P/E heuristic year. Security analysts rely on the P∕E heuristic more than any other technique. Question 3.1 bandwagon investing. Investors were asking the same question. If you were a financial An approach to valuation The valuation identity is given by P0 = P0∕E1 × E1. In order to forecast the price executive at eBay, what combination of traditional and heuristic valuation techniques based on multiplying a P1 a year hence, apply the relationship P1 = P1∕E2 × E2. Here the challenge is to CONCEPT wouldAfter you Internet use to stockanswer prices this question?had generally soared in May 2003, the San Jose Mercury P/E ratio and an earnings News asked whether the rapid increase was due to solid fundamentals or 1999-style forecast. arrive at sensible forecasts of the P∕E ratio that will apply a year hence and earnings PREVIEW per share two years hence. Question 3.1 bandwagon investing. Investors were asking the same question. If you were a financial executive at eBay, what combination of traditional and heuristic valuation techniques PEG Heuristic Considerwould you next use tohow answer the thisfinancial question? executives at eBay actually approached valua- tion at the time. See the Behavior Pitfalls box “Attaching a Value to eBay.” 6,7 A firm’s PEG ratio is defined as its P∕E ratio divided by its expected earnings growth Both eBay’s chief financial officer and vice president for corporate finance were rate per year (actually 100 × expected earnings growth rate). The premise underlying highly educated and experienced. Despite all the emphasis in corporate finance text- the use of PEG-based valuation is that the stocks of high-growth firms merit higher books Consideron discounted next howcash theflow financial analysis, executives many financial at eBay executives actually approachedrely on heuristics valua- 6,7 P∕E ratios than the stocks of low-growth firms. To compute the value of a firm’s insteadtion ofat thethe time.fundamental See the Behaviorvaluation Pitfalls techniques box “Attaching taught in textbooks.a Value to eBay.”Heuristics are stock based on its PEG ratio, multiply the following three terms: its PEG ratio, an simplerBoth to use: eBay’s P∕E, chief PEG, financial and price-to-sales officer and vice require president very fewfor corporate variables finance and involve were estimate of future earnings per share, and an estimate of expected earnings growth. simplehighly formulas. educated As and such experienced. they have intuitive Despite allappeal, the emphasis the basis in for corporate the affect finance heuristic. text- P E G G PEG heuristic The valuation identity is given by 0 = PEG × 1 × , where is 100 × growth Thebooks discussion on discounted in the nearby cash flow box analysis,vividly illustrates many financial this point. executives The DCF-based rely on heuristics anal- An approach to valuation PEG heuristic P rate. As with the P∕E heuristic, the can be used to forecast price 1 ysesinstead taught of in thetextbooks fundamental require valuation far more techniques detail than taught the heuristic in textbooks. techniques, Heuristics involve are based on multiplying a PEG a year hence. Doing so requires forecasts of the PEG ratio that will apply a year simpler to use: P∕E, PEG, and price-to-sales require very few variables and involve ratio, an earnings forecast, more complex formulas, and are far less intuitive. and a forecast of the growth hence, earnings per share two years hence, and the growth rate that will apply over simple formulas. As such they have intuitive appeal, the basis for the affect heuristic. rate. the forecast period. The discussion in the nearby box vividly illustrates this point. The DCF-based anal- 3.4 HOW ANALYSTSyses VALUE taught in textbooks FIRMS: require AN far ILLUSTRATIVE more detail than the heuristic EXAMPLE techniques, involve price-to-sales heuristic Price-to-Sales Heuristic more complex formulas, and are far less intuitive. An approach to valuation price-to-sales heuristic Analyst Mary Meeker based on multiplying a The has the same structure as the P∕E heuristic, except price-to-sales ratio and a that future sales are substituted for future earnings. The valuation identity is given 3.4 HOW ANALYSTSThe Behavioral VALUE Pitfalls FIRMS: box on ANpage ILLUSTRATIVE62 illustrates how the EXAMPLE financial executives at sales forecast. by P0 = P0 ∕S1 × S1, where S stands for sales. eBay approached the value of their firm. Essentially, they relied on the PEG heuristic Analyst Mary Meeker The Behavioral Pitfalls box on page 62 illustrates how the financial executives at eBay approached the value of their firm. Essentially, they relied on the PEG heuristic

she77208_ch03_058-084.indd 60 she77208_ch03_058-084.indd2/7/17 5:26 PM 61 2/7/17 5:26 PM

she77208_ch03_058-084.indd 61 2/7/17 5:26 PM Behavioral Pitfalls: Attaching a Value to eBay bchba_tt Behavioral Pitfalls: Wal-Marting of the Web bchba_tt Attaching a Value to eBay Behavioral Pitfalls: Wal-Marting of the Web OnBehavioral April 1, 2003, Rajiv Dutta, Pitfalls: eBay’s chief financial officer was 1.56, which was actually lower than Wal-Mart’s PEG of During the late 1990s Mary Meeker was a security analyst Securities and Exchange Commission launched a major in- (CFO) addressed a chapter meeting of Financial Executives 1.62. By the PEG metric, eBay was actually cheaper than at Morgan Stanley and was one of the highest-paid analysts vestigation. Among those investigated was Mary Meeker. In bchba_tt International (FEI). During his presentation he discussedAttaching the a ValueWal-Mart, to eBaya point that eBay’s CFO emphasized. Duringon Wall the Street, late having1990s Maryearned Meeker approximately was a security $15Wal-Marting million analyst in ofSecurities2003, the the Web andattorney Exchange general’s Commission office and launched the SEC a criticizedmajor in- BehavioralOn April 1, 2003, Rajiv Dutta, Pitfalls: eBay’s chief financial officer was 1.56, which was actually lower than Wal-Mart’s PEG of Behavioral Pitfalls: manner in which eBay’s executives respond to investors In May 2003 Mark Rubash, eBay’s vice president for at1999 Morgan and $23 Stanley million and in was 2000. one At of the the time,highest-paid she placed analysts very vestigation.her valuations Among as excessive, those investigated but they did was not Mary bring Meeker. charges In who(CFO) suggest addressed that thea chapter firm’s stock meeting is overvalued. of Financial He Executives indicated corporate1.62. By the finance PEG metric, and investor eBay was relations, actually explicitly cheaper stated than high valuations on some Internet firms, and as a result came against her. International (FEI). During his presentation he discussed the Wal-Mart, a point that eBay’s CFO emphasized. on Wall Street, having earned approximately $15 million in 2003, the attorney general’s office and the SEC criticized that eBay’s executives take skeptical investors through the that the firm’s managers did not track return on equity and 1999to be andcalled $23 Queen million of in the 2000. Internet. At the time, she placed very herOn valuations February as 23,excessive, 2004, Thebut theyWall didStreet not Journalbring charges ran a Onmanner April in1, 2003, which Rajiv eBay’s Dutta, executives eBay’s respondchief financial to investors officer wasIn 1.56, May which 2003 was Mark actually Rubash, lower eBay’s than viceWal-Mart’s president PEG forof During the late 1990s Mary Meeker was a security analyst Securities and Exchange Commission launched a major in- following PEG-based analysis. were uncertain about the value of their firm’s cost of capital. highIn valuations 1998, Time on Magazine some Internet interviewed firms, and Mary as aMeeker, result came ask- againstfeature her.titled “Ah, the 1990s” updating the fates of key (CFO)who suggest addressed that thea chapter firm’s stock meeting is overvalued. of Financial He Executives indicated 1.62.corporate By the finance PEG metric, and investor eBay was relations, actually explicitly cheaper stated than at Morgan Stanley and was one of the highest-paid analysts vestigation. Among those investigated was Mary Meeker. In In March 2003, eBay’s P/E ratio was 79, which some in- Based on conversations with investors and perhaps ana- toing be her called to justify Queen the of high the valuationsInternet. of Internet companies WallOn Street February personalities 23, 2004, fromThe Wall the 1990sStreet stockJournal market ran a Internationalthat eBay’s executives (FEI). During take his skeptical presentation investors he discussed through the Wal-Mart,that the firm’s a point managers that eBay’s did CFOnot track emphasized. return on equity and on Wall Street, having earned approximately $15 million in 2003, the attorney general’s office and the SEC criticized vestors took to imply that its stock price was too high. How- lysts, they believed their cost of capital to be in the range of thatIn prevailed 1998, Time at Magazine the time. Sheinterviewed began byMary stating Meeker, that ask- the featurebubble. titled Regarding “Ah, the Mary 1990s” Meeker, updating the article the fates notes of thatkey mannerfollowing in PEG-based which eBay’s analysis. executives respond to investors wereIn uncertain May 2003 about Mark the Rubash, value of eBay’s their firm’s vice cost president of capital. for 1999 and $23 million in 2000. At the time, she placed very her valuations as excessive, but they did not bring charges ever, Dutta suggested that a valuation based on P/E alone 11 to 12 percent. However, they claimed not to compute their ingdevelopment her to justify of the commercialhigh valuations Internet of Internet was “the companies biggest Wallshe was Street recently personalities named fromMorgan the Stanley’s 1990s stock coleader market of whoIn suggest March 2003, that the eBay’s firm’s P/Estock ratio is overvalued. was 79, which He indicatedsome in- corporateBased on financeconversations and investor with investors relations, and explicitly perhaps stated ana- high valuations on some Internet firms, and as a result came against her. failed to factor in earnings growth. In this respect, he sug- cost of capital, say, by using a textbook framework such thatnew prevailedtechnology at cycle the time.ever.” She began by stating that the bubble.tech-sector Regarding research. Mary In August Meeker, 2004, the Meeker article told notes News that- thatvestors eBay’s took executives to imply that take its skepticalstock price investors was too through high. How- the thatlysts, the they firm’s believed managers their costdid notof capital track returnto be inon the equity range and of to be called Queen of the Internet. On February 23, 2004, The Wall Street Journal ran a gested comparing the stock of eBay to the stock of the large as the capital asset pricing model (CAPM), or by factoring developmentContinuing, of she the pointedcommercial out thatInternet just aswas the “the traditional biggest sheweek was and recently CBS.MarketWatch.Com named Morgan that Stanley’s she was coleader not going of followingever, Dutta PEG-based suggested analysis. that a valuation based on P/E alone were11 to 12 uncertain percent. aboutHowever, the valuethey claimed of their firm’snot to costcompute of capital. their In 1998, Time Magazine interviewed Mary Meeker, ask- feature titled “Ah, the 1990s” updating the fates of key retail firm Wal-Mart, and using the PEG ratio as the valuation their historical return on equity into their valuation analy- newretailer technology Wal-Mart cycle came ever.” to dominate the retail sector, web- tech-sectorto be hiding research. out any In more. August The 2004, press Meeker duly notedtold News that- failedIn Marchto factor 2003, in earnings eBay’s P/E growth. ratio wasIn this 79, respect, which some he sug- in- Basedcost of on capital, conversations say, by using with ainvestors textbook and framework perhaps such ana- ing her to justify the high valuations of Internet companies Wall Street personalities from the 1990s stock market metric. According to Fortune magazine, Wal-Mart was the sis. Given the high and accelerating growth rates they were basedContinuing, counterparts she pointedwould emerge out that and just dominateas the traditional Internet weekher stock and picksCBS.MarketWatch.Com were outperforming that the she market. was not In going2010, vestorsgested comparingtook to imply the that stock its of stock eBay price to the was stock too of high. the How-large lysts,as the they capital believed asset their pricing cost model of capital (CAPM), to be or in theby factoringrange of that prevailed at the time. She began by stating that the bubble. Regarding Mary Meeker, the article notes that most admired company in America during 2003. witnessing during 2003, eBay’s executives stated, frankly, retailercommerce. Wal-Mart Mary came Meeker to dominate described the the retail phenomenon sector, web- as toMeeker be hiding left Morgan out any Stanley more. and The moved press dulyto California noted that to ever,retail firmDutta Wal-Mart, suggested and that using a valuationthe PEG ratio based as onthe P/E alone valuation 11their to 12 historical percent. returnHowever, on equitythey claimed into their not to valuation compute analy- their development of the commercial Internet was “the biggest she was recently named Morgan Stanley’s coleader of On May 20, 2003, eBay’s forward-looking P/E ratio was that they had very little idea about the intrinsic value of their basedthe “Wal-Marting counterparts of thewould Web.” emerge and dominate Internet herjoin stock the venture picks were capital outperforming firm Kleiner the Perkins market. Caufield In 2010, & failedmetric. to According factor in earningsto Fortune growth. magazine, In this Wal-Mart respect, washe sug- the costsis. Given of capital, the high say, and by accelerating using a textbook growth framework rates they were such new technology cycle ever.” tech-sector research. In August 2004, Meeker told News- 66.7, while Wal-Mart’s P/E ratio was 22.7. On a P/E basis, firm’s stock. commerce.Wal-Mart Maryhad built Meeker its business described by thebeating phenomenon its competi- as MeekerByers (KPCB) left Morgan as a partner. Stanley In theand ensuing moved yearsto California she came to gestedmost admired comparing company the stock in America of eBay duringto the stock2003. of the large aswitnessing the capital during asset 2003, pricing eBay’s model executives (CAPM), orstated, by factoring frankly, Continuing, she pointed out that just as the traditional week and CBS.MarketWatch.Com that she was not going eBay appeared to be over twice as expensive as Wal-Mart. To assess the above discussion in hindsight, a brief thetion “Wal-Martingin regard to ofconvenience, the Web.” product selection, and low jointo be the recognized venture as capital one of firm the Kleiner world’s Perkinsleading Caufieldexperts on & retailOn firm May Wal-Mart, 20, 2003, and eBay’s using forward-lookingthe PEG ratio as P/Ethe ratiovaluation was theirthat they historical had very return little on idea equity about into the theirintrinsic valuation value of analy- their retailer Wal-Mart came to dominate the retail sector, web- to be hiding out any more. The press duly noted that However, Dutta pointed out that analysts were expecting chronology of eBay for the period 2004 through 2014 ap- prices.Wal-Mart She argued had built that its the business Internet by provided beating theits competi-opportu- Byersdigital (KPCB)technology. as a partner. In the ensuing years she came metric.66.7, while According Wal-Mart’s to Fortune P/E ratio magazine, was 22.7. Wal-Mart On a P/E was basis, the sis.firm’s Given stock. the high and accelerating growth rates they were based counterparts would emerge and dominate Internet her stock picks were outperforming the market. In 2010, eBay to grow by 42.5 percent, while they were only expect- pears later in the chapter, with further detail provided in tionnity forin regardother firms to convenience, to follow Wal-Mart, product but selection, on the web. and low to be recognized as one of the world’s leading experts on mosteBay admiredappeared company to be over in Americatwice as duringexpensive 2003. as Wal-Mart. witnessingTo assess during the 2003, above eBay’s discussion executives in hindsight, stated, afrankly, brief commerce. Mary Meeker described the phenomenon as Meeker left Morgan Stanley and moved to California to ing Wal-Mart to grow by 14 percent. As a result, eBay’s PEG the Additional Resources to Chapter 3. prices.In 2000, She arguedafter a thatperiod the of Internet excessive provided optimism, the opportu-if not ir- digital technology. However,On May Dutta 20, 2003, pointed eBay’s out thatforward-looking analysts were P/E expectingratio was thatchronology they had of very eBay little for idea the periodabout the 2004 intrinsic through value 2014 of their ap- the “Wal-Marting of the Web.” joinSources: the M. venture Meeker capital and M. Murray firm Kleiner Buechner, Perkins “Q&A: Morgan’s Caufield & nityrational for otherexuberance, firms to thefollow prices Wal-Mart, of many but Internet on the web.stocks fell Mary Meeker: Look for the Net’s ‘Top Dogs’,” Time, April 1, 1998; 66.7,eBay whileto grow Wal-Mart’s by 42.5 percent, P/E ratio while was they 22.7. were On only a P/E expect- basis, firm’spears stock.later in the chapter, with further detail provided in Wal-Mart had built its business by beating its competi- Byers (KPCB) as a partner. In the ensuing years she came dramatically.In 2000, afterThis aevent period has of been excessive described optimism, as the if notburst- ir- E. S. Browning, “Ah, the 1990,” The Wall Street Journal, February 23, eBaying Wal-Mart appeared to growto be byover 14 twicepercent. as expensiveAs a result, as eBay’s Wal-Mart. PEG theTo Additional assess theResources above discussionto Chapter in3. hindsight, a brief tion in regard to convenience, product selection, and low toSources: be recognized M. Meeker as and one M. ofMurray the world’sBuechner, leading “Q&A: Morgan’sexperts on rationaling of the exuberance, dot-com bubble. the prices In May of many2002 Internet Fortune stocksmagazine fell 2004; F. Barnako, “Mary Meeker Is Still a Believer,” CBS.MarketWatch However, Dutta pointed out that analysts were expecting chronology of eBay for the period 2004 through 2014 ap- prices. She argued that the Internet provided the opportu- digital.com,Mary Meeker:August technology. 23, Look 2004; for the C. GasparinoNet’s ‘Top andDogs’,” S. Craig, Time, “Meeker April 1, 1998; Won’t dramatically.placed Mary ThisMeeker’s event picture has been on theirdescribed cover asand the asked burst- if eBay to grow by 42.5 percent,to while assess they whether were only or expect- not their stockpears price later seemedin the chapter, to be fairlywith further priced. detail However, provided they in nity for other firms to follow Wal-Mart, but on the web. FaceE. S. Browning, Securities-Fraud “Ah, the Charges,” 1990,” The The Wall Wall Street Street Journal, Journal, February April 3,23, inginvestors of the could dot-com ever bubble. trust Wall In MayStreet 2002 again. Fortune magazine 2004; F. Barnako, “Mary Meeker Is Still a Believer,” CBS.MarketWatch ing Wal-Mart to grow by 14 percent.did little As analysisa result, eBay’s beyond PEG that. the Additional Resources to Chapter 3. In 2000, after a period of excessive optimism, if not ir- 2003; C. Gasparino, “Climbing Back Up; Ready to Reign Again, placedIn the Mary wake Meeker’s of the collapsepicture on in theirstock cover prices, and regulators asked if Sources:Wall.com, Street’s August M. FallenMeeker23, 2004; ‘Queen and C. M. Gasparino of Murray the Net’ Buechner, and Tells S. Newsweek:Craig, “Q&A: “Meeker Morgan’s ‘I’m Won’tnot What about security analysts? One of their main tasks is to assess value. Do rational exuberance, the prices of many Internet stocks fell Face Securities-Fraud Charges,” The Wall Street Journal, April 3, to assess whether or not their stock price seemed to be fairly priced. However, they investorsfrom New could York State’sever trust attorney Wall Street general’s again. office and the U.S. hidingMary Meeker: out anymore,’” Look for Newsweek, the Net’s ‘Top August 30, Dogs’,” Time,2004. April 1, 1998; they rely on DCF? Do they use heuristics, and if so which ones? In the Spring of dramatically. This event has been described as the burst- E.2003; S. Browning, C. Gasparino, “Ah, the “Climbing 1990,” The Back Wall Up; Street Ready Journal, to Reign February Again, 23, did little analysis beyond that. In the wake of the collapse in stock prices, regulators 2003, 24 analysts followed eBay. Among these, the most well known was Mary ing of the dot-com bubble. In May 2002 Fortune magazine 2004;Wall Street’s F. Barnako, Fallen “Mary ‘Queen Meeker of the Is StillNet’ a Tells Believer,” Newsweek: CBS.MarketWatch ‘I’m not What about security analysts? One of their main tasks is to assess value. Do from New York State’s attorney general’s office and the U.S. hiding out anymore,’” Newsweek, August 30, 2004. placed Mary Meeker’s picture on their cover and asked if .com, August 23, 2004; C. Gasparino and S. Craig, “Meeker Won’t toMeeker assess from whether the firmor not Morgan their stock Stanley. price She seemed led a toteam be fairlyof analysts, priced. hereafter However, called they Face Securities-Fraud Charges,” The Wall Street Journal, April 3, they rely on DCF? Do they use heuristics, and if so which ones? In the Spring of investors could ever trust Wall Street again. 2003,didthe littleMorgan 24 analysts analysis Stanley beyond followed team, that. which eBay. conductedAmong these, an analysis the most of welleBay known using awas series Mary of 2003; C. Gasparino, “Climbing Back Up; Ready to Reign Again, In the wake of the collapserespectively. in stock prices, They regulators then averaged Wall these Street’s together Fallen ‘Queen with of their the Net’ DCF Tells valuation Newsweek: to‘I’m arrive not Meeker valuationWhat from abouttechniques, the security firm eachMorgan analysts? of which Stanley. One gave She of a their leddifferent a main team price tasks of analysts, target is to (also assess hereafter known value. called as Do a from New York State’s attorneyat general’sa price target office andof $106 the U.S. per share.hiding8 out anymore,’” Newsweek, August 30, 2004. thetheytarget Morgan rely price) on StanleyDCF?for eBay. Do team, Herthey whichApril 2003 use heuristics, conducted report and an offers ifanalysis so awhich rich of set eBayones? of insights usingIn the a Spring intoseries how of Exhibit 3-2 describes the assumptions and computations that underlie the average valuation2003,analysts 24 analysts apply techniques, these followed various each of eBay.techniques. which Among gave Before a these, different describing the pricemost targetwellthe details, known(also knownrefer was toMary as the a respectively. They then averaged these together with their DCF valuation to arrive target values displayed in Exhibit 3-1.8 In Exhibit 3-2, an E after a year (for example, targetMeekerBehavioral price) from Pitfalls for the eBay. firm box MorganHer “Wal-Marting April 2003 Stanley. of reportShe the led Web”offers a team fora rich ofsome analysts, set background of insights hereafter intoinforma- called how at a price target of $106 per share. 2004E) designates that the numbers are estimates, not actual values. In all cases price analyststhetion Morganon Mary apply Stanley Meeker. these team,various which techniques. conducted Before an analysisdescribing of theeBay details, using refera series to the of Exhibit 3-2 describes the assumptions and computations that underlie the average targets were computed for year-end 2004 and then discounted back to mid-2004. Behavioral valuation techniques, Pitfalls box each “Wal-Marting of which gave of thea different Web” for price some target background (also known informa- as a targetrespectively. values displayedThey then in averaged Exhibit 3-1.these In together Exhibit with3-2, antheir E afterDCF a valuation year (for toexample, arrive The Morgan Stanley Team’s Mid-2004 Price Target for eBay To arrive at their price target for December 8 2004, the team used a P∕E value of 40 tiontarget on price) Mary for Meeker. eBay. Her April 2003 report offers a rich set of insights into how 2004E)at a price designates target of that$106 the per numbers share. are estimates, not actual values. In all cases price analystsAt the time apply the theseMorgan various Stanley techniques. team released Before their describing April 2003 the report,details, eBay’s refer to price the targetsfor Exhibit2005, were the 3-2 midpoint computed describes of forthe 47 year-endassumptionsand 33 shown 2004 and in and thecomputations then table. discounted For sales,that underlie backthey used to the mid-2004. the average gross BehavioralwasThe $89.22. Morgan Pitfalls The Stanley team’s box “Wal-Martingtask Team’s was to developMid-2004 of the aWeb” price Price fortarget some Target for backgroundeBay for over eBay the informa- subse- targetToglobal arrive valuessales at activity theirdisplayed price on inalltarget ExhibiteBay for web December 3-1. sites, In Exhibit called 2004, gross3-2, the an teammerchandise E after used a ayear P∕Esales (for value (GMS). example, of 40 Attionquent the on 12time Mary months. the Meeker. Morgan In undertaking Stanley team this releasedtask, they their used April all three2003 report,valuation eBay’s heuristics price 2004E)for As2005, can designates the be midpointseen thatin Exhibit ofthe 47 numbers and 3-1, 33 the areshown Morgan estimates, in the Stanley table. not actual Forteam sales, values. used they a InDCF usedall casesapproach, the grossprice described earlier, as well as a discounted cash flow (DCF) computation. targetsglobalapplied sales wereto eBay’s activity computed free on cash all for eBay flows. year-end web This sites, 2004 computation called and thengross is discounted merchandise the only one back sales of tothe (GMS). mid-2004. four that wasThe $89.22. Morgan The Stanley team’s task Team’s was to developMid-2004 a price Price target Target for eBay for over eBay the subse- quentExhibit 12 months. 3-1 provides In undertaking a summary this of thetask, components they used ofall the three Morgan valuation Stanley heuristics team’s Topurports Asarrive can to atbe derive their seen price thein Exhibit price target target 3-1,for December inthe terms Morgan of 2004, intrinsic Stanley the teamorteam fundamental used used a a P∕E DCF value. value approach, of 40 describedAtprice the target time earlier, theanalysis. Morgan as well For Stanley as each a discounted ofteam the released first cash three theirflow heuristics, April(DCF) 2003 computation. the report, team eBay’scomputed price a forapplied A2005, firm to the eBay’s generates midpoint free positive cashof 47 flows. and free 33 This cashshown computation flows in the when table. is its Forthe after-tax sales,only one they cash of used the flows fourthe fromgross that waslow,Exhibit $89.22.high, 3-1and The provides intermediate team’s a task summary guess,was to of whichdevelop the components they a price called target of downside, the for Morgan eBay upside, over Stanley the and team’ssubse- base globalpurports operations sales to arederiveactivity positive the on price all and eBay target it doesweb in sites,terms not spend called of intrinsic allgross those merchandiseor fundamental after-tax flowssales value. (GMS). acquiring 62 pricequent target12 months. analysis. In undertakingFor each of this the task,first theythree used heuristics, all three the valuation team computed heuristics a AsA firmcan be generates seen in positiveExhibit 3-1, free the cash Morgan flows Stanley when its team after-tax used a cash DCF flows approach, from63 low,described high, earlier,and intermediate as well as aguess, discounted which cash they flow called (DCF) downside, computation. upside, and base applied operations to eBay’s are positive free cash and flows. it does This not computation spend all those is the after-tax only one flows of the acquiring four that Exhibit 3-1 provides a summary of the components of the Morgan Stanley team’s purports to derive the price target in terms of intrinsic or fundamental value. 63 62 price target analysis. For each of the first three heuristics, the team computed a A firm generates positive free cash flows when its after-tax cash flows from low, high, and intermediate guess, which they called downside, upside, and base operations are positive and it does not spend all those after-tax flows acquiring 62 63 she77208_ch03_058-084.indd 62 2/7/17 5:26she77208_ch03_058-084.indd PM 63 2/7/17 5:26 PM

she77208_ch03_058-084.indd 63 2/7/17 5:26 PM she77208_ch03_058-084.indd 62 2/7/17 5:26 PM

she77208_ch03_058-084.indd 62 2/7/17 5:26she77208_ch03_058-084.indd PM 63 2/7/17 5:26 PM Behavioral Pitfalls: Attaching a Value to eBay bchba_tt Behavioral Pitfalls: Wal-Marting of the Web bchba_tt Attaching a Value to eBay Behavioral Pitfalls: Wal-Marting of the Web OnBehavioral April 1, 2003, Rajiv Dutta, Pitfalls: eBay’s chief financial officer was 1.56, which was actually lower than Wal-Mart’s PEG of During the late 1990s Mary Meeker was a security analyst Securities and Exchange Commission launched a major in- (CFO) addressed a chapter meeting of Financial Executives 1.62. By the PEG metric, eBay was actually cheaper than at Morgan Stanley and was one of the highest-paid analysts vestigation. Among those investigated was Mary Meeker. In bchba_tt International (FEI). During his presentation he discussedAttaching the a ValueWal-Mart, to eBaya point that eBay’s CFO emphasized. Duringon Wall the Street, late having1990s Maryearned Meeker approximately was a security $15Wal-Marting million analyst in ofSecurities2003, the the Web andattorney Exchange general’s Commission office and launched the SEC a criticizedmajor in- BehavioralOn April 1, 2003, Rajiv Dutta, Pitfalls: eBay’s chief financial officer was 1.56, which was actually lower than Wal-Mart’s PEG of Behavioral Pitfalls: manner in which eBay’s executives respond to investors In May 2003 Mark Rubash, eBay’s vice president for at1999 Morgan and $23 Stanley million and in was 2000. one At of the the time,highest-paid she placed analysts very vestigation.her valuations Among as excessive, those investigated but they did was not Mary bring Meeker. charges In who(CFO) suggest addressed that thea chapter firm’s stock meeting is overvalued. of Financial He Executives indicated corporate1.62. By the finance PEG metric, and investor eBay was relations, actually explicitly cheaper stated than high valuations on some Internet firms, and as a result came against her. International (FEI). During his presentation he discussed the Wal-Mart, a point that eBay’s CFO emphasized. on Wall Street, having earned approximately $15 million in 2003, the attorney general’s office and the SEC criticized that eBay’s executives take skeptical investors through the that the firm’s managers did not track return on equity and 1999to be andcalled $23 Queen million of in the 2000. Internet. At the time, she placed very herOn valuations February as 23,excessive, 2004, Thebut theyWall didStreet not Journalbring charges ran a Onmanner April in1, 2003, which Rajiv eBay’s Dutta, executives eBay’s respondchief financial to investors officer wasIn 1.56, May which 2003 was Mark actually Rubash, lower eBay’s than viceWal-Mart’s president PEG forof During the late 1990s Mary Meeker was a security analyst Securities and Exchange Commission launched a major in- following PEG-based analysis. were uncertain about the value of their firm’s cost of capital. highIn valuations 1998, Time on Magazine some Internet interviewed firms, and Mary as aMeeker, result came ask- featureagainst her.titled “Ah, the 1990s” updating the fates of key (CFO)who suggest addressed that thea chapter firm’s stock meeting is overvalued. of Financial He Executives indicated 1.62.corporate By the finance PEG metric, and investor eBay was relations, actually explicitly cheaper stated than at Morgan Stanley and was one of the highest-paid analysts vestigation. Among those investigated was Mary Meeker. In In March 2003, eBay’s P/E ratio was 79, which some in- Based on conversations with investors and perhaps ana- ingto be her called to justify Queen the of high the valuationsInternet. of Internet companies WallOn Street February personalities 23, 2004, fromThe Wall the 1990sStreet stockJournal market ran a Internationalthat eBay’s executives (FEI). During take his skeptical presentation investors he discussed through the Wal-Mart,that the firm’s a point managers that eBay’s did CFOnot track emphasized. return on equity and on Wall Street, having earned approximately $15 million in 2003, the attorney general’s office and the SEC criticized vestors took to imply that its stock price was too high. How- lysts, they believed their cost of capital to be in the range of thatIn prevailed 1998, Time at Magazine the time. Sheinterviewed began byMary stating Meeker, that ask- the featurebubble. titled Regarding “Ah, the Mary 1990s” Meeker, updating the article the fates notes of thatkey mannerfollowing in PEG-based which eBay’s analysis. executives respond to investors wereIn uncertain May 2003 about Mark the Rubash, value of eBay’s their firm’s vice cost president of capital. for 1999 and $23 million in 2000. At the time, she placed very her valuations as excessive, but they did not bring charges ever, Dutta suggested that a valuation based on P/E alone 11 to 12 percent. However, they claimed not to compute their ingdevelopment her to justify of the commercialhigh valuations Internet of Internet was “the companies biggest Wallshe was Street recently personalities named fromMorgan the Stanley’s 1990s stock coleader market of whoIn suggest March 2003, that the eBay’s firm’s P/Estock ratio is overvalued. was 79, which He indicatedsome in- corporateBased on financeconversations and investor with investors relations, and explicitly perhaps stated ana- high valuations on some Internet firms, and as a result came against her. failed to factor in earnings growth. In this respect, he sug- cost of capital, say, by using a textbook framework such thatnew prevailedtechnology at cycle the time.ever.” She began by stating that the bubble.tech-sector Regarding research. Mary In August Meeker, 2004, the Meeker article told notes News that- thatvestors eBay’s took executives to imply that take its skepticalstock price investors was too through high. How- the thatlysts, the they firm’s believed managers their costdid notof capital track returnto be inon the equity range and of to be called Queen of the Internet. On February 23, 2004, The Wall Street Journal ran a gested comparing the stock of eBay to the stock of the large as the capital asset pricing model (CAPM), or by factoring developmentContinuing, of she the pointedcommercial out thatInternet just aswas the “the traditional biggest sheweek was and recently CBS.MarketWatch.Com named Morgan that Stanley’s she was coleader not going of followingever, Dutta PEG-based suggested analysis. that a valuation based on P/E alone were11 to 12 uncertain percent. aboutHowever, the valuethey claimed of their firm’snot to costcompute of capital. their In 1998, Time Magazine interviewed Mary Meeker, ask- feature titled “Ah, the 1990s” updating the fates of key retail firm Wal-Mart, and using the PEG ratio as the valuation their historical return on equity into their valuation analy- newretailer technology Wal-Mart cycle came ever.” to dominate the retail sector, web- tech-sectorto be hiding research. out any In more. August The 2004, press Meeker duly notedtold News that- failedIn Marchto factor 2003, in earnings eBay’s P/E growth. ratio wasIn this 79, respect, which some he sug- in- Basedcost of on capital, conversations say, by using with ainvestors textbook and framework perhaps such ana- ing her to justify the high valuations of Internet companies Wall Street personalities from the 1990s stock market metric. According to Fortune magazine, Wal-Mart was the sis. Given the high and accelerating growth rates they were basedContinuing, counterparts she pointedwould emerge out that and just dominateas the traditional Internet weekher stock and picksCBS.MarketWatch.Com were outperforming that the she market. was not In going2010, vestorsgested comparingtook to imply the that stock its of stock eBay price to the was stock too of high. the How-large lysts,as the they capital believed asset their pricing cost model of capital (CAPM), to be or in theby factoringrange of that prevailed at the time. She began by stating that the bubble. Regarding Mary Meeker, the article notes that most admired company in America during 2003. witnessing during 2003, eBay’s executives stated, frankly, retailercommerce. Wal-Mart Mary came Meeker to dominate described the the retail phenomenon sector, web- as toMeeker be hiding left Morgan out any Stanley more. and The moved press dulyto California noted that to ever,retail firmDutta Wal-Mart, suggested and that using a valuationthe PEG ratio based as onthe P/E alone valuation 11their to 12 historical percent. returnHowever, on equitythey claimed into their not to valuation compute analy- their development of the commercial Internet was “the biggest she was recently named Morgan Stanley’s coleader of On May 20, 2003, eBay’s forward-looking P/E ratio was that they had very little idea about the intrinsic value of their basedthe “Wal-Marting counterparts of thewould Web.” emerge and dominate Internet herjoin stock the venture picks were capital outperforming firm Kleiner the Perkins market. Caufield In 2010, & failedmetric. to According factor in earningsto Fortune growth. magazine, In this Wal-Mart respect, washe sug- the costsis. Given of capital, the high say, and by accelerating using a textbook growth framework rates they were such new technology cycle ever.” tech-sector research. In August 2004, Meeker told News- 66.7, while Wal-Mart’s P/E ratio was 22.7. On a P/E basis, firm’s stock. commerce.Wal-Mart Maryhad built Meeker its business described by thebeating phenomenon its competi- as MeekerByers (KPCB) left Morgan as a partner. Stanley In theand ensuing moved yearsto California she came to gestedmost admired comparing company the stock in America of eBay duringto the stock2003. of the large aswitnessing the capital during asset 2003, pricing eBay’s model executives (CAPM), orstated, by factoring frankly, Continuing, she pointed out that just as the traditional week and CBS.MarketWatch.Com that she was not going eBay appeared to be over twice as expensive as Wal-Mart. To assess the above discussion in hindsight, a brief thetion “Wal-Martingin regard to ofconvenience, the Web.” product selection, and low jointo be the recognized venture as capital one of firm the Kleiner world’s Perkinsleading Caufieldexperts on & retailOn firm May Wal-Mart, 20, 2003, and eBay’s using forward-lookingthe PEG ratio as P/Ethe ratiovaluation was theirthat they historical had very return little on idea equity about into the theirintrinsic valuation value of analy- their retailer Wal-Mart came to dominate the retail sector, web- to be hiding out any more. The press duly noted that However, Dutta pointed out that analysts were expecting chronology of eBay for the period 2004 through 2014 ap- prices.Wal-Mart She argued had built that its the business Internet by provided beating theits competi-opportu- Byersdigital (KPCB)technology. as a partner. In the ensuing years she came metric.66.7, while According Wal-Mart’s to Fortune P/E ratio magazine, was 22.7. Wal-Mart On a P/E was basis, the sis.firm’s Given stock. the high and accelerating growth rates they were based counterparts would emerge and dominate Internet her stock picks were outperforming the market. In 2010, eBay to grow by 42.5 percent, while they were only expect- pears later in the chapter, with further detail provided in tionnity forin regardother firms to convenience, to follow Wal-Mart, product but selection, on the web. and low to be recognized as one of the world’s leading experts on mosteBay admiredappeared company to be over in Americatwice as duringexpensive 2003. as Wal-Mart. witnessingTo assess during the 2003, above eBay’s discussion executives in hindsight, stated, afrankly, brief commerce. Mary Meeker described the phenomenon as Meeker left Morgan Stanley and moved to California to ing Wal-Mart to grow by 14 percent. As a result, eBay’s PEG the Additional Resources to Chapter 3. prices.In 2000, She arguedafter a thatperiod the of Internet excessive provided optimism, the opportu-if not ir- digital technology. However,On May Dutta 20, 2003, pointed eBay’s out thatforward-looking analysts were P/E expectingratio was thatchronology they had of very eBay little for idea the periodabout the 2004 intrinsic through value 2014 of their ap- the “Wal-Marting of the Web.” joinSources: the M. venture Meeker capital and M. Murray firm Kleiner Buechner, Perkins “Q&A: Morgan’s Caufield & nityrational for otherexuberance, firms to thefollow prices Wal-Mart, of many but Internet on the web.stocks fell Mary Meeker: Look for the Net’s ‘Top Dogs’,” Time, April 1, 1998; 66.7,eBay whileto grow Wal-Mart’s by 42.5 percent, P/E ratio while was they 22.7. were On only a P/E expect- basis, firm’spears stock.later in the chapter, with further detail provided in Wal-Mart had built its business by beating its competi- Byers (KPCB) as a partner. In the ensuing years she came dramatically.In 2000, afterThis aevent period has of been excessive described optimism, as the if notburst- ir- E. S. Browning, “Ah, the 1990,” The Wall Street Journal, February 23, eBaying Wal-Mart appeared to growto be byover 14 twicepercent. as expensiveAs a result, as eBay’s Wal-Mart. PEG theTo Additional assess theResources above discussionto Chapter in3. hindsight, a brief tion in regard to convenience, product selection, and low toSources: be recognized M. Meeker as and one M. ofMurray the world’sBuechner, leading “Q&A: Morgan’sexperts on rationaling of the exuberance, dot-com bubble. the prices In May of many2002 Internet Fortune stocksmagazine fell 2004; F. Barnako, “Mary Meeker Is Still a Believer,” CBS.MarketWatch However, Dutta pointed out that analysts were expecting chronology of eBay for the period 2004 through 2014 ap- prices. She argued that the Internet provided the opportu- digitalMary.com, Meeker:August technology. 23, Look 2004; for the C. GasparinoNet’s ‘Top andDogs’,” S. Craig, Time, “Meeker April 1, 1998; Won’t dramatically.placed Mary ThisMeeker’s event picture has been on theirdescribed cover asand the asked burst- if eBay to grow by 42.5 percent,to while assess they whether were only or expect- not their stockpears price later seemedin the chapter, to be fairlywith further priced. detail However, provided they in nity for other firms to follow Wal-Mart, but on the web. E.Face S. Browning, Securities-Fraud “Ah, the Charges,” 1990,” The The Wall Wall Street Street Journal, Journal, February April 3,23, inginvestors of the could dot-com ever bubble. trust Wall In MayStreet 2002 again. Fortune magazine 2004; F. Barnako, “Mary Meeker Is Still a Believer,” CBS.MarketWatch ing Wal-Mart to grow by 14 percent.did little As analysisa result, eBay’s beyond PEG that. the Additional Resources to Chapter 3. In 2000, after a period of excessive optimism, if not ir- 2003; C. Gasparino, “Climbing Back Up; Ready to Reign Again, placedIn the Mary wake Meeker’s of the collapsepicture on in theirstock cover prices, and regulators asked if Sources:Wall.com, Street’s August M. FallenMeeker23, 2004; ‘Queen and C. M. Gasparino of Murray the Net’ Buechner, and Tells S. Newsweek:Craig, “Q&A: “Meeker Morgan’s ‘I’m Won’tnot What about security analysts? One of their main tasks is to assess value. Do rational exuberance, the prices of many Internet stocks fell Face Securities-Fraud Charges,” The Wall Street Journal, April 3, to assess whether or not their stock price seemed to be fairly priced. However, they investorsfrom New could York State’sever trust attorney Wall Street general’s again. office and the U.S. Maryhiding Meeker: out anymore,’” Look for Newsweek, the Net’s ‘Top August 30, Dogs’,” Time,2004. April 1, 1998; they rely on DCF? Do they use heuristics, and if so which ones? In the Spring of dramatically. This event has been described as the burst- E.2003; S. Browning, C. Gasparino, “Ah, the “Climbing 1990,” The Back Wall Up; Street Ready Journal, to Reign February Again, 23, did little analysis beyond that. In the wake of the collapse in stock prices, regulators 2003, 24 analysts followed eBay. Among these, the most well known was Mary ing of the dot-com bubble. In May 2002 Fortune magazine 2004;Wall Street’s F. Barnako, Fallen “Mary ‘Queen Meeker of the Is StillNet’ a Tells Believer,” Newsweek: CBS.MarketWatch ‘I’m not What about security analysts? One of their main tasks is to assess value. Do from New York State’s attorney general’s office and the U.S. hiding out anymore,’” Newsweek, August 30, 2004. placed Mary Meeker’s picture on their cover and asked if .com, August 23, 2004; C. Gasparino and S. Craig, “Meeker Won’t toMeeker assess from whether the firmor not Morgan their stock Stanley. price She seemed led a toteam be fairlyof analysts, priced. hereafter However, called they Face Securities-Fraud Charges,” The Wall Street Journal, April 3, they rely on DCF? Do they use heuristics, and if so which ones? In the Spring of investors could ever trust Wall Street again. 2003,didthe littleMorgan 24 analysts analysis Stanley beyond followed team, that. which eBay. conductedAmong these, an analysis the most of welleBay known using awas series Mary of 2003; C. Gasparino, “Climbing Back Up; Ready to Reign Again, In the wake of the collapserespectively. in stock prices, They regulators then averaged Wall these Street’s together Fallen ‘Queen with of their the Net’ DCF Tells valuation Newsweek: to‘I’m arrive not Meeker valuationWhat from abouttechniques, the security firm eachMorgan analysts? of which Stanley. One gave She of a their leddifferent a main team price tasks of analysts, target is to (also assess hereafter known value. called as Do a from New York State’s attorneyat general’sa price target office andof $106 the U.S. per share.hiding8 out anymore,’” Newsweek, August 30, 2004. thetheytarget Morgan rely price) on StanleyDCF?for eBay. Do team, Herthey whichApril 2003 use heuristics, conducted report and an offers ifanalysis so awhich rich of set eBayones? of insights usingIn the a Spring intoseries how of Exhibit 3-2 describes the assumptions and computations that underlie the average valuation2003,analysts 24 analysts apply techniques, these followed various each of eBay.techniques. which Among gave Before a these, different describing the pricemost targetwellthe details, known(also knownrefer was toMary as the a respectively. They then averaged these together with their DCF valuation to arrive target values displayed in Exhibit 3-1.8 In Exhibit 3-2, an E after a year (for example, targetMeekerBehavioral price) from Pitfalls for the eBay. firm box MorganHer “Wal-Marting April 2003 Stanley. of reportShe the led Web”offers a team fora rich ofsome analysts, set background of insights hereafter intoinforma- called how at a price target of $106 per share. 2004E) designates that the numbers are estimates, not actual values. In all cases price analyststhetion Morganon Mary apply Stanley Meeker. these team,various which techniques. conducted Before an analysisdescribing of theeBay details, using refera series to the of Exhibit 3-2 describes the assumptions and computations that underlie the average targets were computed for year-end 2004 and then discounted back to mid-2004. Behavioral valuation techniques, Pitfalls box each “Wal-Marting of which gave of thea different Web” for price some target background (also known informa- as a targetrespectively. values displayedThey then in averaged Exhibit 3-1.these In together Exhibit with3-2, antheir E afterDCF a valuation year (for toexample, arrive The Morgan Stanley Team’s Mid-2004 Price Target for eBay To arrive at their price target for December 8 2004, the team used a P∕E value of 40 tiontarget on price) Mary for Meeker. eBay. Her April 2003 report offers a rich set of insights into how 2004E)at a price designates target of that$106 the per numbers share. are estimates, not actual values. In all cases price analystsAt the time apply the theseMorgan various Stanley techniques. team released Before their describing April 2003 the report,details, eBay’s refer to price the targetsfor Exhibit2005, were the 3-2 midpoint computed describes of forthe 47 year-endassumptionsand 33 shown 2004 and in and thecomputations then table. discounted For sales,that underlie backthey used to the mid-2004. the average gross BehavioralwasThe $89.22. Morgan Pitfalls The Stanley team’s box “Wal-Martingtask Team’s was to developMid-2004 of the aWeb” price Price fortarget some Target for backgroundeBay for over eBay the informa- subse- targetToglobal arrive valuessales at activity theirdisplayed price on inalltarget ExhibiteBay for web December 3-1. sites, In Exhibit called 2004, gross3-2, the an teammerchandise E after used a ayear P∕Esales (for value (GMS). example, of 40 Attionquent the on 12time Mary months. the Meeker. Morgan In undertaking Stanley team this releasedtask, they their used April all three2003 report,valuation eBay’s heuristics price 2004E)for As2005, can designates the be midpointseen thatin Exhibit ofthe 47 numbers and 3-1, 33 the areshown Morgan estimates, in the Stanley table. not actual Forteam sales, values. used they a InDCF usedall casesapproach, the grossprice described earlier, as well as a discounted cash flow (DCF) computation. targetsglobalapplied sales wereto eBay’s activity computed free on cash all for eBay flows. year-end web This sites, 2004 computation called and thengross is discounted merchandise the only one back sales of tothe (GMS). mid-2004. four that wasThe $89.22. Morgan The Stanley team’s task Team’s was to developMid-2004 a price Price target Target for eBay for over eBay the subse- quentExhibit 12 months. 3-1 provides In undertaking a summary this of thetask, components they used ofall the three Morgan valuation Stanley heuristics team’s Topurports Asarrive can to atbe derive their seen price thein Exhibit price target target 3-1,for December inthe terms Morgan of 2004, intrinsic Stanley the teamorteam fundamental used used a a P∕E DCF value. value approach, of 40 describedAtprice the target time earlier, theanalysis. Morgan as well For Stanley as each a discounted ofteam the released first cash three theirflow heuristics, April(DCF) 2003 computation. the report, team eBay’scomputed price a forapplied A2005, firm to the eBay’s generates midpoint free positive cashof 47 flows. and free 33 This cashshown computation flows in the when table. is its Forthe after-tax sales,only one they cash of used the flows fourthe fromgross that waslow,Exhibit $89.22.high, 3-1and The provides intermediate team’s a task summary guess,was to of whichdevelop the components they a price called target of downside, the for Morgan eBay upside, over Stanley the and team’ssubse- base globalpurports operations sales to arederiveactivity positive the on price all and eBay target it doesweb in sites,terms not spend called of intrinsic allgross those merchandiseor fundamental after-tax flowssales value. (GMS). acquiring 62 pricequent target12 months. analysis. In undertakingFor each of this the task,first theythree used heuristics, all three the valuation team computed heuristics a AsA firmcan be generates seen in positiveExhibit 3-1, free the cash Morgan flows Stanley when its team after-tax used a cash DCF flows approach, from63 low,described high, earlier,and intermediate as well as aguess, discounted which cash they flow called (DCF) downside, computation. upside, and base applied operations to eBay’s are positive free cash and flows. it does This not computation spend all those is the after-tax only one flows of the acquiring four that Exhibit 3-1 provides a summary of the components of the Morgan Stanley team’s purports to derive the price target in terms of intrinsic or fundamental value. 63 62 price target analysis. For each of the first three heuristics, the team computed a A firm generates positive free cash flows when its after-tax cash flows from low, high, and intermediate guess, which they called downside, upside, and base operations are positive and it does not spend all those after-tax flows acquiring 62 63 she77208_ch03_058-084.indd 62 2/7/17 5:26she77208_ch03_058-084.indd PM 63 2/7/17 5:26 PM

she77208_ch03_058-084.indd 63 2/7/17 5:26 PM she77208_ch03_058-084.indd 62 2/7/17 5:26 PM

she77208_ch03_058-084.indd 62 2/7/17 5:26she77208_ch03_058-084.indd PM 63 2/7/17 5:26 PM 64 Chapter Three Valuation 65

EXHIBIT 3-1 Technique Downside Upside Base EXHIBIT 3-3 Free Cash Flow Computation in Mary Meeker’s Price Target Analysis of eBay Summary of Components P∕E $74 $111 $ 84 Note: Values given are in $ thousands. 64in Mary Chapter Meeker’s Three Price PEG 45 91 68 Source: Mary Meeker and Brian Pitz, “CQ1 Results: Tales of a Growth Machine,” Morgan Stanley Analyst Report on eBay, April 23, 2003. Target Analysis of eBay Price-to-sales 97 210 154 Discounted cash flow 117 2000 2001 2002 2003E 2004E 2005E 2006E 2007E 2008E 2009E 2010E 2011E on EXHIBITSource: Mary 3-1Meeker and Technique Downside Upside Base Average $104 SummaryBrian Pitz, “CQ1 of Results:Components Tales Revenue 431,424 748,821 1,214,100 2,050,492 2,829,679 3,763,473 4,907,751 6,287,602 7,857,002 9,582,843 11,400,983 of a Growth Machine,” Morgan P∕UpsideE adjusted average $74 $111 $ 84 $106 EBITDA 84,072 229,438 444,614 723,735 1,005,276 1,401,682 1,853,230 2,426,297 3,110,990 3,985,085 4,982,032 inStanley Mary Analyst Meeker’s Report on Price eBay, PEG 45 91 68 Taxes — — — — — 140,168 370,646 849,204 1,088,846 1,394,780 1,743,711 April 23, 2003. Target Analysis of eBay Price-to-sales 97 210 154 Change in 47,582 41,091 (39,232) 26,792 (6,849) 58,695 (9,709) 52,096 (7,375) 53,278 (4,105) Source: Mary Meeker and Discounted cash flow 117 working Brian Pitz, “CQ1 Results: Tales Average $104 capital of a Growth Machine,” Morgan Upside adjusted average $106 Capital 49,753 57,420 138,670 188,908 190,000 190,000 190,000 190,000 190,000 190,000 190,000 Stanley Analyst Report on eBay, expenditures April 23, 2003.EXHIBIT 3-2 Valuation Methodologies in the Morgan Stanley Team’s Price Target Analysis of eBay Free cash flow (13,263) 130,927 345,176 508,035 822,125 1,012,819 1,302,292 1,334,997 1,839,519 2,347,027 3,052,426 65,321,907

Source: Mary Meeker and Brian Pitz, “CQ1 Results: Tales of a Growth Machine,” Morgan Stanley Analyst Report on eBay, April 23, 2003. Price Target—Dec 31, 2003 Year Present value eBay free cash flows $36,478,759 EXHIBIT 3-2 Valuation Methodologies in the Morgan Stanley Team’s Price Target Analysis of eBay Less debt $79,592 2002 2003E 2004E 2005E Plus cash $2,280,857 Source: Mary Meeker and Brian Pitz,EPS “CQ1 at Results: 32% Talesgrowth of a Growth Machine,” Morgan $0.86 Stanley Analyst$1.14 Report on $1.50eBay, April 23, 2003. $1.98 eBay’s full value $38,680,023 EPS at 38% growth $0.86 $1.19 $1.64 $2.26 Shares outstanding (‘000) 330,259 GMS/share at 38% growth $39.53 $54.55Year $75.27 $103.88 Discount rate 12% P/E 2002103 2003E 63 2004E 47 2005E 33 Future growth rate 7% DCF per share value $117 EPS at 32% growthPrice-to- $0.86 $1.14 $1.50GMS $1.98 Price Price P/E PEG Sales Sales per Target Target EPS at 38% growth $0.86 $1.19 $1.64 $2.26 Note that we use a Hurdle Rate rather than the Weighted Average Cost of Capital (WACC) as the discount rate for our DCF analysis. We believe that a Hurdle Rate is more appropri- Method RatioGMS/share Ratio at 38% growth Ratio $39.53 EPS $54.55 Growth $75.27 Share $103.88 Dec 2004 (Discounted) ate because it more accurately reflects what we view as company-specific risks. We calculate the Cost of Equity at 9.5 percent based on a risk-free rate of 3.9 percent taken from the P/E 103 63 47 33 10-year U.S. Government Bond Yield as of 10/31/02; a risk premium of 4.0 percent from Morgan Stanley Economics Research estimates of expected equity risk premium; and beta of P/E 40 $2.26 $90 $84 1.38 from Bloomberg relative to the S&P 500 as of 10/31/02. Note that eBay’s debt is immaterial. PEG 1.5 Price-to- 1.50 32 GMS Price 72 Price 68 Price-to-sales P/E PEG Sales1.5 2.26 Sales$103.88 per Target 156 Target 154 Method Ratio Ratio Ratio EPS Growth Share Dec 2004 (Discounted) P/E 40 $2.26 $90 $84 In their April 2003 report, the Morgan Stanley team forecast free cash flows PEG 1.5 1.50 32 72 68 through the end of 2010. They used a terminal value to capture the free cash working capital and new fixed assets. In this respect, working capital is to be un- Price-to-sales 1.5 2.26 $103.88 156 154 flows that would occur after 2010 and discounted these at a rate of 12 percent.9 In derstood to mean “operational working capital,” namely the change in net working Exhibit 3-3, they obtained their $65.3 billion terminal value in 2011 by assuming capital for items only associated with cash flow from operations but not cash flows that free cash flows would grow at the rate of 7 percent from 2011 on, and then val- from financing or cash and marketable securities. Therefore, annual free cash flow ued the future expected free cash flow stream beginning in 2011 using the constant- workingcomprises capital cash and flow new generated fixed assets. by theIn thisfirm respect, within workingthe year capital that is isavailable to be un- to be growth-rate perpetuity formula. Specifically, they forecast that free cash flows in derstoodpaid to theto mean firm’s “operational investors. Thisworking available capital,” cash namely flow consiststhe change of incash net flow working actually 2011 would be $3,266,096 = 1.07 × $3,052,426. They then applied the perpetuity capitalpaid to for investors items only together associated with withthe increase cash flow in fromthe firm’s operations cash and but marketablenot cash flows securi- formula: fromties. financingIn this respect, or cash the and firm’s marketable investors securities. are its debtholders Therefore, annual and shareholders, free cash flow so that comprises cash flow generated by the firm within the year that is available to be cash paid to investors consists of interest, repayment of principal net of new debt, $3,266,096 paiddividends, to the firm’s and share investors. repurchases This available net of newcash stockflow consistsissues. of cash flow actually PV = = $65,321,907 0.12–0.07 paidExhibit to investors 3-3 togetherdisplays with the theDCF increase valuation in the table firm’s from cash the and Morgan marketable Stanley securi- team’s ties.April In this2003 respect, report. the The firm’s heart investors of their are DCF its debtholders analysis consists and shareholders, of the forecasts so that for The Morgan Stanley team computed the present value of the free cash flow casheBay’s paid revenues to investors and consistsits earnings of interest, before repaymentinterest, taxes, of principal depreciation, net of andnew amortiza-debt, stream by summing the discounted free cash flow values for every year between dividends,tion (EBITDA). and share repurchases net of new stock issues. 2004E and the terminal year. As shown in Exhibit 3-3, that sum came to $36.5 bil- ExhibitThe free 3-3 cash flows displays the for DCF eBay valuation are displayed table fromin Exhibit the Morgan 3-3. The Stanley present team’s value of lion. Then they added eBay’s $2.3 billion cash holdings to arrive at the value of the Aprilthe firm’s 2003 expectedreport. The free heart cash of flows, their plus DCF the analysis value of consists its current of the cash forecasts and marketable for firm as a whole and subtracted the value of eBay’s $79 million in debt to arrive at the eBay’ssecurities, revenues comprise and its the earnings value ofbefore the entire interest, firm. taxes, depreciation, and amortiza- value of eBay’s equity. The net result was $38 billion, which the team called eBay’s tion (EBITDA). The free cash flows for eBay are displayed in Exhibit 3-3. The present value of the firm’s expected free cash flows, plus the value of its current cash and marketable securities, comprise the value of the entire firm.

she77208_ch03_058-084.indd 64 she77208_ch03_058-084.indd2/7/17 5:26 PM 65 2/7/17 5:26 PM

she77208_ch03_058-084.indd 64 2/7/17 5:26 PM 64 Chapter Three Valuation 65

EXHIBIT 3-1 Technique Downside Upside Base EXHIBIT 3-3 Free Cash Flow Computation in Mary Meeker’s Price Target Analysis of eBay Summary of Components P∕E $74 $111 $ 84 Note: Values given are in $ thousands. in Mary Meeker’s Price PEG 45 91 68 Source: Mary Meeker and Brian Pitz, “CQ1 Results: Tales of a Growth Machine,” Morgan Stanley Analyst Report on eBay, April 23, 2003. Valuation 65 Target Analysis of eBay Price-to-sales 97 210 154 Discounted cash flow 117 2000 2001 2002 2003E 2004E 2005E 2006E 2007E 2008E 2009E 2010E 2011E on Source: Mary Meeker and EXHIBIT 3-3 Free Cash Flow Computation in Mary Meeker’s Price Target Analysis of eBay Brian Pitz, “CQ1 Results: Tales Average $104 Revenue 431,424 748,821 1,214,100 2,050,492 2,829,679 3,763,473 4,907,751 6,287,602 7,857,002 9,582,843 11,400,983 of a Growth Machine,” Morgan Upside adjusted average $106 EBITDANote: Values given 84,072 are in $ 229,438 thousands. 444,614 723,735 1,005,276 1,401,682 1,853,230 2,426,297 3,110,990 3,985,085 4,982,032 Stanley Analyst Report on eBay, TaxesSource: Mary Meeker— and Brian — Pitz, “CQ1 — Results: Tales — of a Growth — Machine,” 140,168 Morgan Stanley370,646 Analyst Report 849,204 on eBay, 1,088,846April 23, 2003. 1,394,780 1,743,711 April 23, 2003. Change in 47,582 41,091 (39,232) 26,792 (6,849) 58,695 (9,709) 52,096 (7,375) 53,278 (4,105) working 2000 2001 2002 2003E 2004E 2005E 2006E 2007E 2008E 2009E 2010E 2011E on capitalRevenue 431,424 748,821 1,214,100 2,050,492 2,829,679 3,763,473 4,907,751 6,287,602 7,857,002 9,582,843 11,400,983 CapitalEBITDA 49,753 84,072 57,420 229,438 138,670 444,614 188,908 723,735 190,000 1,005,276 190,000 1,401,682 190,000 1,853,230 190,000 2,426,297 190,000 3,110,990 3,985,085190,000 4,982,032190,000 expenditures EXHIBIT 3-2 Taxes — — — — — 140,168 370,646 849,204 1,088,846 1,394,780 1,743,711 Valuation Methodologies in the Morgan Stanley Team’s Price Target Analysis of eBay Free cashChange flow in (13,263) 47,582 130,927 41,091 345,176 (39,232) 508,035 26,792 822,125 (6,849) 1,012,819 58,695 1,302,292 (9,709) 1,334,997 52,096 1,839,519 (7,375) 2,347,027 53,278 3,052,426(4,105) 65,321,907 Source: Mary Meeker and Brian Pitz, “CQ1 Results: Tales of a Growth Machine,” Morgan Stanley Analyst Report on eBay, April 23, 2003. working capital Price Target—Dec 31, 2003 Year Capital 49,753 57,420 138,670 188,908Present 190,000 value eBay free190,000 cash flows 190,000 $36,478,759 190,000 190,000 190,000 190,000 expenditures Less debt $79,592 Free cash flow (13,263) 130,927 345,176 508,035 822,125 1,012,819 1,302,292 1,334,997 1,839,519 2,347,027 3,052,426 65,321,907 2002 2003E 2004E 2005E Plus cash $2,280,857 EPS at 32% growth $0.86 $1.14 $1.50 $1.98 eBay’sPrice full Target—Dec value 31, 2003 $38,680,023 EPS at 38% growth $0.86 $1.19 $1.64 $2.26 Shares outstanding (‘000) 330,259 GMS/share at 38% growth $39.53 $54.55 $75.27 $103.88 DiscountPresent rate value eBay free cash flows $36,478,75912% P/E 103 63 47 33 FutureLess growth debt rate $79,5927% DCFPlus per cashshare value $2,280,857$117 Price-to- GMS Price Price eBay’s full value $38,680,023 P/E PEG Sales Sales per Target Target Note that we use a Hurdle Rate rather than the Weighted Average CostShares of Capital outstanding (WACC) as(‘000) the discount rate for our330,259 DCF analysis. We believe that a Hurdle Rate is more appropri- Method Ratio Ratio Ratio EPS Growth Share Dec 2004 (Discounted) ate because it more accurately reflects what we view as company-specificDiscount risks. Werate calculate the Cost of Equity at 9.5 percent12% based on a risk-free rate of 3.9 percent taken from the 10-year U.S. Government Bond Yield as of 10/31/02; a risk premium ofFuture 4.0 percent growth from rate Morgan Stanley Economics Research7% estimates of expected equity risk premium; and beta of P/E 40 $2.26 $90 $84 1.38 from Bloomberg relative to the S&P 500 as of 10/31/02. Note that eBay’s debt is immaterial. PEG 1.5 1.50 32 72 68 DCF per share value $117 Price-to-sales 1.5 2.26 $103.88 156 154 Note that we use a Hurdle Rate rather than the Weighted Average Cost of Capital (WACC) as the discount rate for our DCF analysis. We believe that a Hurdle Rate is more appropri- ate because it more accurately reflects what we view as company-specific risks. We calculate the Cost of Equity at 9.5 percent based on a risk-free rate of 3.9 percent taken from the 10-year U.S. Government Bond Yield as of 10/31/02; a risk premium of 4.0 percent from Morgan Stanley Economics Research estimates of expected equity risk premium; and beta of 1.38 from Bloomberg relative to the S&PIn 500 theiras of 10/31/02. April Note 2003 that eBay’s report, debt is immaterial. the Morgan Stanley team forecast free cash flows through the end of 2010. They used a terminal value to capture the free cash working capital and new fixed assets. In this respect, working capital is to be un- flows that would occur after 2010 and discounted these at a rate of 12 percent.9 In derstood to mean “operational working capital,” namely the change in net working Exhibit 3-3, they obtained their $65.3 billion terminal value in 2011 by assuming capital for items only associated with cash flow from operations but not cash flows that freeIn cash their flows April would 2003 report,grow at the the Morgan rate of 7 Stanley percent team from forecast 2011 on, free and cash then flows val- from financing or cash and marketable securities. Therefore, annual free cash flow uedthrough the future the expected end of 2010.free cash They flow stream used a terminal beginning value in 2011 to capture using thethe constant- free cash comprises cash flow generated by the firm within the year that is available to be 9 growth-rateflows that perpetuity would occur formula. after 2010 Specifically, and discounted they forecast these at thata rate free of 12cash percent. flows Inin paid to the firm’s investors. This available cash flow consists of cash flow actually 2011Exhibit would 3-3, be $3,266,096they obtained = their1.07 $65.3× $3,052,426. billion terminal They then value applied in 2011 the by perpetuity assuming paid to investors together with the increase in the firm’s cash and marketable securi- formula:that free cash flows would grow at the rate of 7 percent from 2011 on, and then val- ties. In this respect, the firm’s investors are its debtholders and shareholders, so that ued the future expected free cash flow stream beginning in 2011 using the constant- cash paid to investors consists of interest, repayment of principal net of new debt, growth-rate perpetuity formula.$3,266,096 Specifically, they forecast that free cash flows in dividends, and share repurchases net of new stock issues. PV = = $65,321,907 2011 would be $3,266,096 = 1.070.12–0.07 × $3,052,426. They then applied the perpetuity Exhibit 3-3 displays the DCF valuation table from the Morgan Stanley team’s formula: April 2003 report. The heart of their DCF analysis consists of the forecasts for The Morgan Stanley team computed the present value of the free cash flow eBay’s revenues and its earnings before interest, taxes, depreciation, and amortiza- stream by summing the discounted$3,266,096 free cash flow values for every year between PV = = $65,321,907 tion (EBITDA). 2004E and the terminal year. As shown0.12–0.07 in Exhibit 3-3, that sum came to $36.5 bil- The free cash flows for eBay are displayed in Exhibit 3-3. The present value of lion. Then they added eBay’s $2.3 billion cash holdings to arrive at the value of the the firm’s expected free cash flows, plus the value of its current cash and marketable firm asThe a whole Morgan and Stanleysubtracted team the computed value of eBay’s the present $79 million value in of debt the freeto arrive cash at flow the securities, comprise the value of the entire firm. valuestream of eBay’s by summing equity. Thethe discountednet result was free $38 cash billion, flow valueswhich forthe everyteam calledyear between eBay’s 2004E and the terminal year. As shown in Exhibit 3-3, that sum came to $36.5 bil- lion. Then they added eBay’s $2.3 billion cash holdings to arrive at the value of the firm as a whole and subtracted the value of eBay’s $79 million in debt to arrive at the value of eBay’s equity. The net result was $38 billion, which the team called eBay’s

she77208_ch03_058-084.indd 64 she77208_ch03_058-084.indd2/7/17 5:26 PM 65 2/7/17 5:26 PM

she77208_ch03_058-084.indd 65 2/7/17 5:26 PM 66 Chapter Three Valuation 67

“full value.” Dividing full value by eBay’s shares outstanding (from Exhibit 3-3) EXHIBIT 3-4 200% 10 resulted in a target value of about $117 per share at year-end 2003. In using this Year Over Year Growth 180% 66 Chapter Three Rates in Morgan Stanley amount, they did not take the future value of the $117 to mid-2004. 160% 2003 Report on eBay, 140% “full value.” Dividing full value by eBay’s shares outstanding (from Exhibit 3-3) for Revenue, EBITDA, 3.5 VALUATION HEURISTICS AND BIASES: 10 and Free Cash Flow 120% resulted in a target value of about $117 per share at year-end 2003. In using this IN FORESIGHTamount, they did not take the future value of the $117 to mid-2004. Source of data: Morgan Stanley 100% Report on eBay, April 2003. 80% This section uses the Morgan Stanley eBay report and eBay CFO Rajiv Dutta’s 60% 3.5 VALUATION valuationHEURISTICS perspective AND about BIASES: valuing the company as vehicles to discuss identifying 40% IN FORESIGHTvaluation heuristics and biases. Keep in mind that the discussion is intended to be illustrative, and it is important to avoid overgeneralizing on the basis of a single ex- 20% Thisample. section The drawinguses the ofMorgan broad Stanleyconclusions eBay needs report to and be basedeBay onCFO fairly Rajiv large Dutta’s samples: 0%

valuationbecause perspectiveof luck, good about judgments valuing thein foresightcompany ascan vehicles look bad to discussin hindsight, identifying and vice 2000 2001 2002 2003E 2004E 2005E 2006E 2007E 2008E 2009E 2010E valuationversa. heuristics and biases. Keep in mind that the discussion is intended to be illustrative,For this and reason, it is important this section to avoid focuses overgeneralizing on two approaches, on the basis one offoresight-focused a single ex- Revenue Y/Y Growth EBITDA Y/Y Growth FCF Y/Y Growth ample.and the The other drawing hindsight-focused. of broad conclusions The foresight-focused needs to be based onapproach fairly large seeks samples: to identify becausebiases basedof luck, on goodinformation judgments available in foresight at the cantime look the judgmentsbad in hindsight, were made, and vice and ig- versa.noring what we now know in hindsight. The hindsight-focused approach examines thoseFor thisjudgments reason, inthis light section of how focuses actual on events two approaches, played out. oneBoth foresight-focused approaches provide In the Morgan Stanley report, the 30 percent assumption pertains to P∕E, PEG, anddifferent the other insights, hindsight-focused. and offer different The foresight-focused lessons about how approach to improve seeks tothe identify process of and P∕S. However, the free cash flow analysis also involves longer-term forecasts. biasesvaluation. based on information available at the time the judgments were made, and ig- Exhibit 3-4 illustrates the growth rates of three critical inputs, namely revenues, noring what we now know in hindsight. The hindsight-focused approach examines EBITDA, and free cash flows. Over the forecast period from year-end 2003 through thoseOptimism judgments Bias: in light In ofForesight how actual events played out. Both approaches provide 2010, revenue growth declines from 38 percent to 19 percent, EBITDA growth differentThe Behavioral insights, Pitfallsand offer box different “Attaching lessons a aboutValue how to eBay” to improve describes the process how in of 2003 declines from 39 percent to 25 percent, and free cash flow growth declines from valuation.eBay’s financial executives thought about the value of their firm. They were un- 62 percent to 30 percent. certain as to how to conduct an evaluation, engaged in no discounted cash flow Plausibly, the short-term 30 percent forecasts might have been optimistic, when Optimismanalysis, and Bias: viewed In theForesight market price as justifiable on the basis of PEG. In contrast, made. In this regard, a reasonable question to pose might be whether the very rapid TheMary Behavioral Meeker andPitfalls her analystbox “Attaching team at Morgana Value Stanleyto eBay” had describes a methodology how in 2003for valu- growth rates for revenue and EBITDA in the period 2000 through 2002 served as eBay’sing eBay financial featuring executives discounted thought cash about flow theanalysis value along of their with firm. a series They of were price un- target anchors for the forecasts in 2003 through 2010? certainheuristics, as to one how of to which conduct involved an evaluation, PEG. This engaged section in discusses no discounted the presence cash flow of bias More importantly, from an intrinsic valuation point of view, the short-term analysis,in the valuation and viewed judgments the market described price as justifiable in the previous on the basis section. of PEG. In thisIn contrast, regard, the growth rates are much less important than the growth rate associated with terminal Maryfocus Meeker on eBay and simply her analyst provides team a at vehicle Morgan for Stanley making had the a methodology discussion concrete. for valu- The value. The free cash flows from 2004 through 2010 only comprise 19 percent of the inggeneral eBay featuringideas apply discounted widely. cashNotably, flow theanalysis Morgan along Stanley with a analysts’series of pricefocus target on DCF, $36.5 billion value that Morgan Stanley analysts assigned to all of eBay’s future free heuristics,P∕E, PEG, one and of P∕Swhich is involvedcommon PEG.in the This analyst section community. discusses the presence of bias cash flows. Terminal value comprised the lion’s share, namely 81 percent. in theRecall valuation the discussion judgments of described optimism in bias the from previous Chapter section. 1. On In April this regard, 23, 2003, the The Looking back at Exhibit 3-3, we can see that free cash flows were expected focusWall onStreet eBay Journal simply providespublished a vehiclean article for makingsuggesting the discussionthat the analysts concrete. following The to grow at 7 percent during the terminal horizon beginning in 2011. That is, the generaleBay were ideas excessively apply widely. optimistic Notably, about the Morgan eBay’s Stanley future revenueanalysts’ stream.focus on The DCF, article Morgan Stanley analysts forecasted that beginning in 2011, eBay’s free cash flows P∕E,pointed PEG, out and that P∕S the is Morgancommon Stanley in the analyst team’s community. forecast for the period ending in 2010 would grow at 7 percent forever. According to World Bank Statistics,11 GDP growth wasRecall the themost discussion optimistic. of optimismNick Wingfield, bias from the Chapter author 1. of On The April Wall 23, Street2003, TheJournal in 2003 was approximately 4 percent and the GDP deflator was about 2 percent. The Wallarticle, Street laid Journal out several published possible an articlescenarios suggesting that might that supportthe analysts the optimistic following rev- two together total 6 percent, which is a full percentage point less than the 7 percent eBayenue were forecasts, excessively and concluded optimistic thatabout 30 eBay’s percent future revenue revenue growth stream. forecasts The article for eBay forecast for eBay. The difference is small for a single year, but compounding out pointedwere unduly out that optimistic. the Morgan Notably, Stanley thisteam’s judgment forecast applies for the periodto the assumptionsending in 2010 stated to eternity is another matter. If eBay were to grow 1 percent faster than the U.S. The Wall Street Journal wasin Exhibits the most 3-2 optimistic. and 3-3. Nick Wingfield, the author of economy, it would eventually become the U.S. economy. article, laid out several possible scenarios that might support the optimistic rev- enue forecasts, and concluded that 30 percent revenue growth forecasts for eBay were unduly optimistic. Notably, this judgment applies to the assumptions stated in Exhibits 3-2 and 3-3.

she77208_ch03_058-084.indd 66 she77208_ch03_058-084.indd2/7/17 5:26 PM 67 2/7/17 5:26 PM

she77208_ch03_058-084.indd 66 2/7/17 5:26 PM 66 Chapter Three Valuation 67

“full value.” Dividing full value by eBay’s shares outstanding (from Exhibit 3-3) EXHIBIT 3-4 200% 10 resulted in a target value of about $117 per share at year-end 2003. In using this Year Over Year Growth 180% Rates in Morgan Stanley Valuation 67 amount, they did not take the future value of the $117 to mid-2004. 160% 2003 Report on eBay, 140% for Revenue,EXHIBIT EBITDA, 3-4 200% 3.5 VALUATION HEURISTICS AND BIASES: and Free Cash Flow 120% Year Over Year Growth 180% IN FORESIGHT Source Ratesof data: Morganin Morgan Stanley Stanley 100% Report on eBay, April 2003. 160% 2003 Report on eBay, 80% for Revenue, EBITDA, 140% This section uses the Morgan Stanley eBay report and eBay CFO Rajiv Dutta’s 60% valuation perspective about valuing the company as vehicles to discuss identifying and Free Cash Flow 120% 40% valuation heuristics and biases. Keep in mind that the discussion is intended to be Source of data: Morgan Stanley 100% Report on eBay, April 2003. 20% illustrative, and it is important to avoid overgeneralizing on the basis of a single ex- 80% 0% ample. The drawing of broad conclusions needs to be based on fairly large samples: 60% because of luck, good judgments in foresight can look bad in hindsight, and vice 2000 2001 2002 40% 2003E 2004E 2005E 2006E 2007E 2008E 2009E 2010E versa. For this reason, this section focuses on two approaches, one foresight-focused Revenue20% Y/Y Growth EBITDA Y/Y Growth FCF Y/Y Growth and the other hindsight-focused. The foresight-focused approach seeks to identify 0% biases based on information available at the time the judgments were made, and ig- 2000 2001 2002 2003E 2004E 2005E 2006E 2007E 2008E 2009E 2010E noring what we now know in hindsight. The hindsight-focused approach examines those judgments in light of how actual events played out. Both approaches provide In theRevenue Morgan Y/Y StanleyGrowth report,EBITDA the Y/Y 30 Growth percent assumptionFCF Y/Y Growth pertains to P∕E, PEG, different insights, and offer different lessons about how to improve the process of and P∕S. However, the free cash flow analysis also involves longer-term forecasts. valuation. Exhibit 3-4 illustrates the growth rates of three critical inputs, namely revenues, EBITDA, and free cash flows. Over the forecast period from year-end 2003 through Optimism Bias: In Foresight 2010, In revenue the Morgan growth Stanley declines report, from the 3830 percentpercent assumption to 19 percent, pertains EBITDA to P∕E, growth PEG, The Behavioral Pitfalls box “Attaching a Value to eBay” describes how in 2003 declinesand P∕S. from However, 39 percent the tofree 25 cash percent, flow analysisand free also cash involves flow growth longer-term declines forecasts. from eBay’s financial executives thought about the value of their firm. They were un- 62 percentExhibit to3-4 30 illustrates percent. the growth rates of three critical inputs, namely revenues, certain as to how to conduct an evaluation, engaged in no discounted cash flow Plausibly,EBITDA, theand short-termfree cash flows. 30 percent Over the forecasts forecast might period have from been year-end optimistic, 2003 through when analysis, and viewed the market price as justifiable on the basis of PEG. In contrast, made.2010, In this revenue regard, growth a reasonable declines question from 38 to percent pose might to 19 be percent, whether EBITDA the very growth rapid Mary Meeker and her analyst team at Morgan Stanley had a methodology for valu- growthdeclines rates from for revenue39 percent and to EBITDA 25 percent, in theand period free cash 2000 flow through growth 2002 declines served from as ing eBay featuring discounted cash flow analysis along with a series of price target anchors62 percent for the to forecasts 30 percent. in 2003 through 2010? heuristics, one of which involved PEG. This section discusses the presence of bias MorePlausibly, importantly, the short-term from an 30 intrinsic percent forecasts valuation might point have of view,been optimistic, the short-term when in the valuation judgments described in the previous section. In this regard, the growthmade. rates In thisare regard,much less a reasonable important question than the to growth pose might rate associatedbe whether withthe very terminal rapid focus on eBay simply provides a vehicle for making the discussion concrete. The value.growth The freerates cash for revenue flows from and 2004EBITDA through in the 2010 period only 2000 comprise through 19 2002 percent served of the as general ideas apply widely. Notably, the Morgan Stanley analysts’ focus on DCF, $36.5anchors billion for value the forecasts that Morgan in 2003 Stanley through analysts 2010? assigned to all of eBay’s future free P∕E, PEG, and P∕S is common in the analyst community. cash flows.More Terminal importantly, value from comprised an intrinsic the lion’s valuation share, point namely of view,81 percent. the short-term Recall the discussion of optimism bias from Chapter 1. On April 23, 2003, The Lookinggrowth rates back are at much Exhibit less 3-3,important we can than see the that growth free rate cash associated flows were with expected terminal Wall Street Journal published an article suggesting that the analysts following to growvalue. at The 7 percent free cash during flows fromthe terminal 2004 through horizon 2010 beginning only comprise in 2011. 19 percent That is, of thethe eBay were excessively optimistic about eBay’s future revenue stream. The article Morgan$36.5 Stanley billion valueanalysts that forecasted Morgan Stanley that beginning analysts assigned in 2011, to eBay’s all of eBay’s free cash future flows free pointed out that the Morgan Stanley team’s forecast for the period ending in 2010 wouldcash grow flows. at 7Terminal percent valueforever. comprised According the to lion’s World share, Bank namely Statistics, 81 percent.11 GDP growth was the most optimistic. Nick Wingfield, the author of The Wall Street Journal in 2003Looking was approximately back at Exhibit 4 percent 3-3, weand can the seeGDP that deflator free cash was about flows 2 were percent. expected The article, laid out several possible scenarios that might support the optimistic rev- twoto together grow at total 7 percent 6 percent, during which the isterminal a full percentagehorizon beginning point less in than2011. the That 7 percent is, the Morgan Stanley analysts forecasted that beginning in 2011, eBay’s free cash flows enue forecasts, and concluded that 30 percent revenue growth forecasts for eBay forecast for eBay. The difference is small for a single year, but compounding11 out were unduly optimistic. Notably, this judgment applies to the assumptions stated to eternitywould grow is another at 7 percent matter. forever. If eBay According were to to grow World 1 Bank percent Statistics, faster than GDP the growth U.S. in Exhibits 3-2 and 3-3. economy,in 2003 it was would approximately eventually 4become percent theand U.S. the GDP economy. deflator was about 2 percent. The two together total 6 percent, which is a full percentage point less than the 7 percent forecast for eBay. The difference is small for a single year, but compounding out to eternity is another matter. If eBay were to grow 1 percent faster than the U.S. economy, it would eventually become the U.S. economy.

she77208_ch03_058-084.indd 66 she77208_ch03_058-084.indd2/7/17 5:26 PM 67 2/7/17 5:26 PM

she77208_ch03_058-084.indd 67 2/7/17 5:26 PM 68 Chapter Three Valuation 69

There are deeper issues associated with the question of whether the forecast in- Either lower free cash flows or a lower growth rate, or both, will result in a lower puts feature bias. Think about the competitive dynamic in which currently successful DCF valuation for eBay. The result is bias, namely excessive optimism, resulting 68 Chapter Three firms find themselves when they have a competitive advantage. Successful firms from biased inputs used in the DCF analysis. The specific form of optimism per- attract competitors seeking to reduce, if not eliminate, that competitive advantage. tains to the implicit assumption that there are positive growth opportunities during OnThere this arepoint, deeper in 2014 issues eBay associated CEO Devinwith the Wenig question characterized of whether eBay’sthe forecast competitive in- growth opportunities the terminal horizon, a feature called “growth opportunities bias.” putslandscape feature bias.as being Think “a about more thecompetitive competitive market,” dynamic adding: in which “We currently are not successful going to have bias What would eBay’s price per share be if we maintain its investment policy into firmsthe field find tothemselves [ourselves].” when12 they have a competitive advantage. Successful firms Unwarranted assumption of the terminal horizon, but require that PVGO be zero after 2010? To answer that positive growth opportuni- g attractThe competitors Morgan Stanley seeking report to reduce, mentions if not Amazon eliminate, and that Google competitive as being advantage. two of eBay’s ties during the terminal question, consider a growth rate that would be consistent with zero PVGO. To Onmain this competitors. point, in 2014 During eBay theCEO time Devin a firm Wenig enjoys characterized a competitive eBay’s advantage, competitive its proj- horizon. find such a growth rate, suppose that the fraction of eBay’s after-tax EBIT going landscapeects feature as beinga higher “a more rate ofcompetitive return than market,” the competitive adding: “We rate are of not return. going Notably to have it is to net CapEx and working capital during the terminal horizon is the average value thethe field competitive to [ourselves].” rate of12 return that serves as the discount rate associated with posi- during the period 2006–2010. That average turns out to be 6.3%, which given the tiveThe net Morgan present Stanley value reportprojects mentions and positive Amazon PVGO. and Google However, as being competitive two of eBay’s advantage equation g∕k = 6.3%. With k assumed to be 12%, g must be 0.76%. If we redo the mainis usually competitors. temporary During and the disappears time a firm in the enjoys long a runcompetitive along with advantage, positive itsPVGO. proj- The computation in Exhibit 3-3 with this value of g, we obtain a price per share of $65. ectsterminal feature horizon a higher is rate the of long return term. than Therefore, the competitive only inrate very of return.unusual Notably circumstances it is thewould competitive we expect rate positive of return PVGO that serves during as the the terminal discount horizon. rate associated with posi- Biases Associated with P E, PEG, and PVGO: In Foresight tive Unlessnet present they value are magicians,projects and firms positive have PVGO. to invest However, enough, competitive both in fixed advantage assets and ∕ isworking usually temporarycapital, in orderand disappears to grow. Take in the another long run look along at Exhibit with positive 3-3 to see PVGO. what Thethe Mor - terminalgan Stanley horizon group is thewas long forecasting term. Therefore, in respect onlyto eBay’s in very capital unusual expenditures circumstances (CapEx). wouldNotice we that expect during positive the period PVGO 2004 during through the terminal 2010, CapEx horizon. is flat at $190 million every CONCEPT In traditional textbook valuation analysis, what variables determine whether the stock of year,Unless even they as areeBay’s magicians, forecasted firms growth have tois investsoaring, enough, as displayed both in infixed Exhibit assets 3-4. and This PREVIEW a firm has a low intrinsic P∕E ratio or a high P∕E ratio? Textbook theory stipulates that intrinsic P/E is a function of two variables, required return working capital, in order to grow. Take another look at Exhibit 3-3 to see what the Mor- Question 3.2 suggests that the forecasted CapEx is too low to support the forecasted growth. and the fraction of a firm’s value that derives from growth opportunities PVGO. When a firm gan ThereStanley is group a particular was forecasting condition in respectthat must to eBay’s hold if capital PVGO expenditures is to be zero (CapEx). in the ter- has zero growth opportunities, meaning its PVGO is equal to zero, then its intrinsic forward Noticeminal that horizon. during Asthe isperiod discussed 2004 through in the Additional2010, CapEx Resources is flat at $190 for million this chapter, every the P∕E is 1∕kE, where kE is the required return on equity. However, as PVGO becomes positive, year, even as eBay’s forecasted growth is soaring, as displayed in Exhibit 3-4. This condition stipulates that the fraction of after-tax EBIT that is allocated to the sum P∕E rises above 1∕kE. As PVGO approaches P0, P0∕E1 rises to infinity. suggestsof CapEx that (net the of forecasted replacing CapEx depreciated is too low assets) to support and the the change forecasted in net growth. working capital mustThere equal is a theparticular ratio g ∕conditionk, where gthat is themust growth hold ifrate PVGO and kis is to the be discountzero in the rate. ter- Since minalMorgan horizon. Stanley As assumed is discussed that g in would the Additional be 7 percent Resources and k would for this be chapter,12 percent, the g∕k conditionis 58.3 percent. stipulates Is thatthere the any fraction hint that of after-taxeBay is planningEBIT that to is invest allocated 58.3 to percent the sum of its Notably, in April 2003, the Morgan Stanley team titled their report on eBay “Tales ofafter-tax CapEx (netEBIT of replacingon net CapEx depreciated and working assets) capital?and the change in net working capital of a Growth Machine” in order to emphasize growth as the justification for their price mustThe equal sum the of ratio net gCapEx∕k, where and g workingis the growth capital rate is and just k theis the difference discount rate.between Since after- target of $106, with its associated P∕E being in the vicinity of 50. Notably, that value Morgantax EBIT Stanley and free assumed cash thatflow. g Wewould can be use 7 percent this fact and to kcompute would be the 12 time percent, series g∕ ofk the of P∕E is indirect, obtained by taking the future value of the price target at year-end isratio 58.3 “net percent. CapEx Is thereand working any hint capital that eBay to after-tax is planning EBIT” to invest for the 58.3 period percent 2004 of through its 2004 divided by forecasted EPS for 2005 given 38 percent growth. after-tax2010. The EBIT ratio on averagesnet CapEx 7.8 and percent working during capital? the period, and declines from 10.1 per- Other analysts covering eBay used even higher values than the Morgan Stanley centThe in sum 2004 of tonet 3.1 CapEx percent and inworking 2010. capitalIt reaches is just a maximum the difference of 13 between percent after- in 2005, team for P∕E. For example, a Prudential report on eBay, dated May 12, 2003, used a taxwhich EBIT is andstill free a far cash cry flow.from We58.3 can percent. use this fact to compute the time series of the P∕E ratio of 75.13 The Prudential analysts assessed the value of eBay stock at $108, ratioThere “net CapEx are at andleast working three conclusions capital to after-tax that one EBIT” might for draw the periodat this 2004point. through The first is not that different from Morgan Stanley’s $106. However, their valuations involved 2010.that eBay’sThe ratio intellectual averages 7.8capital percent and during brand theis so period, powerful and anddeclines persistent from 10.1that per-the firm an EPS forecast of $1.43 for fiscal year 2003 and an EPS estimate of $2.15 for 2004, centwill in be 2004 able to to 3.1 maintain percent a incompetitive 2010. It reaches advantage a maximum in the terminal of 13 percent period in with 2005, mini- implying a growth rate of 50 percent, much higher than the corresponding forecasts whichmal investment. is still a far Givencry from the 58.3 competitive percent. dynamic in high technology, this possibility of the Morgan Stanley team. isThere unrealistic. are at leastThe secondthree conclusions is that eBay’s that oneinvestment might draw levels, at this if continued point. The into first the is ter- As a general matter, analysts rely heavily on the P∕E heuristic to arrive at price thatminal eBay’s horizon, intellectual are far capital too low and to brandsupport is soa growth powerful rate and of persistent7 percent. that If we the stick firm with targets, but their reports rarely, if ever, use techniques that emphasize growth op- will7 percent, be able thento maintain the level a competitiveof free cash advantageflows during in the the terminal terminal period horizon with will mini- have to portunities (in the sense of PVGO), despite the prominence of growth opportunity- malbe muchinvestment. lower Given than thethe Morgancompetitive Stanley dynamic team in assumes. high technology, Third, if this we possibility stick with the based formulas in finance textbooks. This is not to say that analysts ignore growth; islevel unrealistic. of investment, The second then is eBay’s that eBay’s growth investment rate during levels, the if terminal continued horizon into the will ter- have they most certainly do not. However, they are inclined to focus on growth in EPS minalto be horizon, much lower are far than too 7 low percent. to support a growth rate of 7 percent. If we stick with rather than growth in future NPV that forms the basis for PVGO. 7 percent, then the level of free cash flows during the terminal horizon will have to be much lower than the Morgan Stanley team assumes. Third, if we stick with the level of investment, then eBay’s growth rate during the terminal horizon will have to be much lower than 7 percent.

she77208_ch03_058-084.indd 68 she77208_ch03_058-084.indd2/7/17 5:26 PM 69 2/7/17 5:26 PM

she77208_ch03_058-084.indd 68 2/7/17 5:26 PM 68 Chapter Three Valuation 69

There are deeper issues associated with the question of whether the forecast in- Either lower free cash flows or a lower growth rate, or both, will result in a lower puts feature bias. Think about the competitive dynamic in which currently successful DCF valuation for eBay. The result is bias, namely excessive optimism, resulting firms find themselves when they have a competitive advantage. Successful firms from biased inputs used in the DCF analysis. The specific form of optimismValuation per- 69 attract competitors seeking to reduce, if not eliminate, that competitive advantage. tains to the implicit assumption that there are positive growth opportunities during On this point, in 2014 eBay CEO Devin Wenig characterized eBay’s competitive growth opportunities the terminalEither horizon,lower free a cashfeature flows called or a “growthlower growth opportunities rate, or both, bias.” will result in a lower landscape as being “a more competitive market,” adding: “We are not going to have bias WhatDCF wouldvaluation eBay’s for eBay. price The per resultshare isbe bias, if we namely maintain excessive its investment optimism, policy resulting into the field to [ourselves].”12 Unwarranted assumption of the fromterminal biased horizon, inputs butused require in the DCFthat PVGOanalysis. be The zero specific after 2010?form of To optimism answer thatper- positive growth opportuni- g The Morgan Stanley report mentions Amazon and Google as being two of eBay’s ties during the terminal question,tains to consider the implicit a growth assumption rate that that there would are bepositive consistent growth with opportunities zero PVGO. during To main competitors. During the time a firm enjoys a competitive advantage, its proj- horizon.growth opportunities findthe such terminal a growth horizon, rate, a suppose feature called that the “growth fraction opportunities of eBay’s after-tax bias.” EBIT going ects feature a higher rate of return than the competitive rate of return. Notably it is bias to net WhatCapEx would and workingeBay’s price capital per during share be the if terminalwe maintain horizon its investment is the average policy value into the competitive rate of return that serves as the discount rate associated with posi- Unwarranted assumption ofduring the theterminal period horizon, 2006–2010. but require That averagethat PVGO turns be out zero to after be 6.3%, 2010? which To answer given thatthe positive growth opportuni- g k k g g tive net present value projects and positive PVGO. However, competitive advantage ties during the terminal equationquestion, ∕ consider= 6.3%. aWith growth assumed rate that to bewould 12%, be consistentmust be 0.76%. with zero If we PVGO. redo theTo is usually temporary and disappears in the long run along with positive PVGO. The horizon. computationfind such ina growthExhibit rate, 3-3 withsuppose this that value the of fraction g, we obtain of eBay’s a price after-tax per share EBIT of going $65. terminal horizon is the long term. Therefore, only in very unusual circumstances to net CapEx and working capital during the terminal horizon is the average value would we expect positive PVGO during the terminal horizon. Biasesduring Associated the period 2006–2010. with P ThatE, PEG, average and turns PVGO: out to beIn 6.3%,Foresight which given the Unless they are magicians, firms have to invest enough, both in fixed assets and equation g∕k = 6.3%. With k assumed∕ to be 12%, g must be 0.76%. If we redo the working capital, in order to grow. Take another look at Exhibit 3-3 to see what the Mor- computation in Exhibit 3-3 with this value of g, we obtain a price per share of $65. gan Stanley group was forecasting in respect to eBay’s capital expenditures (CapEx). CONCEPT In traditional textbook valuation analysis, what variables determine whether the stock of Notice that during the period 2004 through 2010, CapEx is flat at $190 million every Biases Associated with P E, PEG, and PVGO: In Foresight year, even as eBay’s forecasted growth is soaring, as displayed in Exhibit 3-4. This PREVIEW a firm has a low intrinsic P∕E ratio or a∕ high P∕E ratio? Question 3.2 Textbook theory stipulates that intrinsic P/E is a function of two variables, required return suggests that the forecasted CapEx is too low to support the forecasted growth. and the fraction of a firm’s value that derives from growth opportunities PVGO. When a firm There is a particular condition that must hold if PVGO is to be zero in the ter- CONCEPT has Inzero traditional growth textbookopportunities, valuation meaning analysis, its PVGO what variablesis equal to determine zero, then whether its intrinsic the stockforward of minal horizon. As is discussed in the Additional Resources for this chapter, the P E is 1 k , where k is the required return on equity. However, as PVGO becomes positive, PREVIEW ∕ a firm∕ Ehas a low Eintrinsic P∕E ratio or a high P∕E ratio? condition stipulates that the fraction of after-tax EBIT that is allocated to the sum P E rises above 1 k . As PVGO approaches P , P E rises to infinity. Question 3.2 ∕ Textbook theory∕ E stipulates that intrinsic P/E0 0is∕ a 1function of two variables, required return of CapEx (net of replacing depreciated assets) and the change in net working capital and the fraction of a firm’s value that derives from growth opportunities PVGO. When a firm must equal the ratio g∕k, where g is the growth rate and k is the discount rate. Since has zero growth opportunities, meaning its PVGO is equal to zero, then its intrinsic forward g k g k Morgan Stanley assumed that would be 7 percent and would be 12 percent, ∕ P∕E is 1∕kE, where kE is the required return on equity. However, as PVGO becomes positive, is 58.3 percent. Is there any hint that eBay is planning to invest 58.3 percent of its Notably,P∕E rises inabove April 1∕ k2003,E. As PVGO the Morgan approaches Stanley P0, P 0team∕E1 rises titled to infinity.their report on eBay “Tales after-tax EBIT on net CapEx and working capital? of a Growth Machine” in order to emphasize growth as the justification for their price The sum of net CapEx and working capital is just the difference between after- target of $106, with its associated P∕E being in the vicinity of 50. Notably, that value tax EBIT and free cash flow. We can use this fact to compute the time series of the of P∕E is indirect, obtained by taking the future value of the price target at year-end ratio “net CapEx and working capital to after-tax EBIT” for the period 2004 through 2004 dividedNotably, by in forecasted April 2003, EPS the forMorgan 2005 Stanley given 38 team percent titled growth.their report on eBay “Tales 2010. The ratio averages 7.8 percent during the period, and declines from 10.1 per- Otherof a Growth analysts Machine” covering in eBayorder toused emphasize even higher growth values as the than justification the Morgan for their Stanley price cent in 2004 to 3.1 percent in 2010. It reaches a maximum of 13 percent in 2005, teamtarget for P∕E. of $106, For withexample, its associated a Prudential P∕E reportbeing in on the eBay, vicinity dated of May50. Notably, 12, 2003, that used value a which is still a far cry from 58.3 percent. P∕Eof ratio P∕E of is 75.indirect,13 The obtained Prudential by takinganalysts the assessed future value the valueof the of price eBay target stock at atyear-end $108, There are at least three conclusions that one might draw at this point. The first is not 2004that differentdivided by from forecasted Morgan EPS Stanley’s for 2005 $106. given However,38 percent their growth. valuations involved that eBay’s intellectual capital and brand is so powerful and persistent that the firm an EPSOther forecast analysts of $1.43 covering for fiscal eBay year used 2003 even and higher an EPS values estimate than the of Morgan$2.15 for Stanley 2004, implyingteam for a growth P∕E. For rate example, of 50 percent, a Prudential much report higher on than eBay, the dated corresponding May 12, 2003, forecasts used a will be able to maintain a competitive advantage in the terminal period with mini- 13 mal investment. Given the competitive dynamic in high technology, this possibility of theP∕E Morgan ratio of Stanley 75. The team. Prudential analysts assessed the value of eBay stock at $108, is unrealistic. The second is that eBay’s investment levels, if continued into the ter- Asnot a that general different matter, from analysts Morgan rely Stanley’s heavily $106. on the However, P∕E heuristic their valuations to arrive involvedat price minal horizon, are far too low to support a growth rate of 7 percent. If we stick with targets,an EPS but forecast their reports of $1.43 rarely, for fiscal if ever, year use 2003 techniques and an EPS that estimate emphasize of $2.15 growth for 2004, op- 7 percent, then the level of free cash flows during the terminal horizon will have to portunitiesimplying (in a growththe sense rate of of PVGO), 50 percent, despite much the higher prominence than the ofcorresponding growth opportunity- forecasts be much lower than the Morgan Stanley team assumes. Third, if we stick with the basedof theformulas Morgan in Stanleyfinance team. textbooks. This is not to say that analysts ignore growth; level of investment, then eBay’s growth rate during the terminal horizon will have they mostAs a certainly general matter, do not. analysts However, rely they heavily are oninclined the P∕E to heuristicfocus on to growth arrive atin priceEPS to be much lower than 7 percent. rathertargets, than butgrowth their in reports future rarely,NPV thatif ever, forms use the techniques basis for thatPVGO. emphasize growth op- portunities (in the sense of PVGO), despite the prominence of growth opportunity- based formulas in finance textbooks. This is not to say that analysts ignore growth; they most certainly do not. However, they are inclined to focus on growth in EPS rather than growth in future NPV that forms the basis for PVGO.

she77208_ch03_058-084.indd 68 she77208_ch03_058-084.indd2/7/17 5:26 PM 69 2/7/17 5:26 PM

she77208_ch03_058-084.indd 69 2/7/17 5:26 PM 70 Chapter Three Valuation 71

Are analysts, and for that matter CFOs, inclined to form judgments of P∕E Biases Related to the 1∕n Heuristic: In Foresight in which they implicitly view high earnings growth as representative of high Consider the methodology used in Exhibit 3-1, whereby the four price target compo- 70 Chapter Three PVGO? In other words, do analysts and CFOs succumb to representativeness nents were equally weighted. Despite the enormous dispersion in values associated bias when forming judgments about P∕E? To be sure, reliance on a PEG-based with P∕E, PEG, price-to-sales, and DCF, the Morgan Stanley team simply averaged valuationAre analysts, heuristic and isfor consistent that matter with CFOs, the answerinclined being to form yes. judgments In this respect, of P∕E think the numbers, which, in their words, “combine to an average fair value of about $106.” inback which to Rajiv they implicitlyDutta’s argument view high in the earnings Behavioral growth Pitfalls as representative box “Attaching of high a Value 1∕n heuristic The 1∕n heuristic is a rule of thumb that assigns the same weight to each tech- PVGO?to eBay,” In asother well words, as the Morgando analysts Stanley and CFOsteam’s succumb use of a PEGto representativeness heuristic in arriving A decision rule based on nique component, as if they are all equally valid. In particular, textbook-favored biasat a when price formingtarget. judgments about P∕E? To be sure, reliance on a PEG-based using the simple average of DCF receives the same weight as the PEG heuristic, which has no basis in textbook a series of variables. valuationThe PEG heuristic heuristic is consistent specifically with states the answerthat P∕E being is proportional yes. In this to respect, earnings think growth. theory. Some psychologists call the 1∕n heuristic “tallying” and suggest that it is One Up on Wall Street, backNotably, to Rajiv legendary Dutta’s money argument manager in the PeterBehavioral Lynch, Pitfalls in his bookbox “Attaching a Value reasonable to use it in complex environments when the level of understanding is toexplicitly eBay,” as stated well as that the for Morgan companies Stanley whose team’s stocks use of are a PEG fairly heuristic valued, inP∕E arriving will equal poor. atthe a price growth target. rate. In contrast, standard investment textbooks warn about using PEG, Valuation is certainly complex, and perhaps the wide range of valuation esti - indicatingThe PEG heuristicthat it is specifically a Wall Street states rule that of P∕E thumb is proportional that has shortcomings. to earnings growth. A major mates associated with the Morgan Stanley team’s four techniques suggests that Notably,shortcoming legendary is that money the valuemanager of PEGPeter Lynch,for the inS&P his book500 has One fluctuated Up on Wall widely Street, over they have a poor understanding of the drivers of eBay’s stock market value. In explicitlytime, in statedthe range that 1for to companies 1.5. In addition, whose stocks keep inare mind fairly thatvalued, in theory, P∕E will a firmequal with this regard, the P∕E ratio associated with the Morgan Stanley team’s DCF-based k thezero growth PVGO rate. will In havecontrast, a P∕E standard ratio equal investment to 1∕ textbooksE no matter warn what about its growthusing PEG, rate. In- estimate of $117 was 56, suggesting strong growth opportunities. Given that indicatingdeed, its growththat it israte a Wallwill beStreet determined rule of thumbby the thatdividend has shortcomings. payout ratio andA major return on eBay’s stock was trading at $89.22 a share when the report was released, a price shortcomingequity, with is a thathigher the payoutvalue of lowering PEG for gthe but S&P not 500P∕E. has There fluctuated are many widely zero over PVGO target of $117 corresponds to an expected return of 31 percent. As 31 percent time,firms—most in the range children 1 to 1.5.cannot In addition,be above keepaverage—and in mind that therefore in theory, their a PEGfirm ratioswith are exceeds the required return of 12 percent, the implication is that eBay’s stock is k zeronot PVGOpegged will to a have specific a P∕E number ratio equal like 1.5.to 1∕ E no matter what its growth rate. In- intrinsically undervalued. deed,In its contrast growth to rate the will PEG be perspectivedetermined byin thewhich dividend P∕E is payout proportional ratio and to return growth on rate That being said, Exhibit 3-1 tells us that the Morgan Stanley team’s valuations equity,g, textbook with a theory higher tellspayout us loweringsomething g butquite not different P∕E. There about are intrinsic many zero P∕E, PVGO required directly based on P∕E and PEG imply that eBay’s stock was overvalued. The key firms—mostreturn, and children PVGO. cannot Recall be that above the average—and Morgan Stanley therefore team’s their discussion PEG ratios of are eBay assumption underlying the overvaluation was not the team’s forecast for earnings notused pegged a rate to ofa specific 12 percent number to discountlike 1.5. eBay’s future cash flows, thereby treating growth, but its assumption about P∕E, which the report predicted would decline k 12In percentcontrast asto the eBay’s PEG requiredperspective return in which E. The P∕E intrinsicis proportional forward to growth P∕E ratio rate of a from 63 to 33 between 2003 and 2005. g,firm textbook with theoryzero growth tells us opportunities something quite and different a required about return intrinsic of 12 P∕E, percent required is 8.33 In applying the 1∕n heuristic, the Morgan Stanley team assigned equal weight return,(1∕0.12). and At PVGO. the same Recall time, that the the Morgan Morgan Stanley Stanley team team’s used discussion a P∕E ratio of eBay of 40 in to the two overvaluation components and the two undervaluation components in usedtheir a P∕Erate valuationof 12 percent heuristic. to discount If this eBay’s P∕E were future intrinsic, cash flows, then athereby value oftreating 40 would k Exhibit 3-1. At the same time, the tone of the Morgan Stanley 2003 report did 12imply percent that aseBay eBay’s had significant required return PVGO. E. Textbook The intrinsic finance forward theory P∕E stipulates ratio of that a in not reflect the weighting and was very upbeat, calling many of eBay’s elements firm with zero growth opportunities and a required return of 12 percent is 8.33 order for a firm to have positive PVGO, its expected return on equity (ROE) must “outstanding.” The report noted that the number of “what’s good” items they (1∕0.12). At the same time, the Morgan Stanley team used a P∕E ratio of 40 in k exceed its required return. In contrast, firms for which expected ROE equals E analyzed vastly outnumbered the number of “what’s bad” items, and the analyst their P∕E valuation heuristic. If this P∕E were intrinsic, then a value of 40 would have zero growth opportunities. team’s recommendation for eBay stock was “overweight.” Moreover, the report imply that eBay had significant PVGO. Textbook finance theory stipulates that in Consider how, between the time eBay went public in 1998 and the time the indicated that all the positive trends gave the team “more confidence” in the firm’s order for a firm to have positive PVGO, its expected return on equity (ROE) must Morgan Stanley team issued its April 2003 report, the firm’s actual ROE com - long term outlook, leading them to have much more comfort creating long-term exceed its required return. In contrast, firms for which expected ROE equals kE pared to its required return. Between 1998 and 2002, eBay’s ROE rose from forecasts. have zero growth opportunities. 5 percent to a maximum of 10 percent, with an average ROE of 6.1 percent for A general issue associated with the 1∕n heuristic is that the judgments it gener- Consider how, between the time eBay went public in 1998 and the time the the period. ates are highly dependent on the composition of the choice menu. For example, Morgan Stanley team issued its April 2003 report, the firm’s actual ROE com - Even though its annual growth rate of its EBITDA exceeded 90 percent every when applied to portfolio selection in retirement plans, the 1∕n heuristic leads in- pared to its required return. Between 1998 and 2002, eBay’s ROE rose from year from 2000 through 2002, its ROE never rose to the required return of 12 per- vestors to overweigh equities when their choice menu features more equity funds 5 percent to a maximum of 10 percent, with an average ROE of 6.1 percent for cent used in the Morgan Stanley report—although this does not mean that looking than bond funds, and to underweigh equities when the reverse holds. theforward period. in mid-2003, eBay necessarily had zero growth opportunities and that a 1 n Even though its annual growth rate of its EBITDA exceeded 90 percent every The Morgan Stanley team’s reliance on the ∕ heuristic implies that they assign P∕E ratio of 40 for 2005 was unjustified. However, the history of ROE up to that a weight of 25 percent to eBay’s stock being driven by intrinsic value. The other yearpoint from does 2000 not throughsupport 2002,the intrinsic its ROE value never of rose P∕E to being the required 40. Therefore, return of any 12 justifica-per- cent used in the Morgan Stanley report—although this does not mean that looking 75 percent reflect mispricing, with techniques based on P∕E and PEG suggesting tion for intrinsic P∕E being that high would require identifying the magnitude of that eBay’s stock was overvalued, and price-to-sales suggesting it was undervalued. forwardgrowth inopportunities mid-2003, eBay explicitly necessarily and in hada systematic zero growth fashion. opportunities and that a P∕E ratio of 40 for 2005 was unjustified. However, the history of ROE up to that Notably, if the Morgan Stanley team were to drop one of the techniques, such as point does not support the intrinsic value of P∕E being 40. Therefore, any justifica- tion for intrinsic P∕E being that high would require identifying the magnitude of growth opportunities explicitly and in a systematic fashion.

she77208_ch03_058-084.indd 70 she77208_ch03_058-084.indd2/7/17 5:26 PM 71 2/7/17 5:26 PM

she77208_ch03_058-084.indd 70 2/7/17 5:26 PM 70 Chapter Three Valuation 71

Are analysts, and for that matter CFOs, inclined to form judgments of P∕E Biases Related to the 1∕n Heuristic: In Foresight in which they implicitly view high earnings growth as representative of high Consider the methodology used in Exhibit 3-1, whereby the four price target compo- PVGO? In other words, do analysts and CFOs succumb to representativeness nents were equally weighted. Despite the enormous dispersion in values Valuationassociated 71 bias when forming judgments about P∕E? To be sure, reliance on a PEG-based with P∕E, PEG, price-to-sales, and DCF, the Morgan Stanley team simply averaged valuation heuristic is consistent with the answer being yes. In this respect, think the numbers,Biases Related which, in totheir the words, 1∕n “combineHeuristic: to anIn average Foresight fair value of about $106.” back to Rajiv Dutta’s argument in the Behavioral Pitfalls box “Attaching a Value 1∕n heuristic TheConsider 1∕n heuristic the methodology is a rule used of thumb in Exhibit that 3-1, assigns whereby the samethe four weight price totarget each compo tech-- to eBay,” as well as the Morgan Stanley team’s use of a PEG heuristic in arriving A decision rule based on niquenents component, were equally as ifweighted. they are Despite all equally the enormous valid. In dispersion particular, in textbook-favoredvalues associated at a price target. using the simple average of DCF receives the same weight as the PEG heuristic, which has no basis in textbook a series of variables. with P∕E, PEG, price-to-sales, and DCF, the Morgan Stanley team simply averaged The PEG heuristic specifically states that P∕E is proportional to earnings growth. theory.the numbers,Some psychologists which, in their call words, the “combine1∕n heuristic to an “tallying” average fair and value suggest of about that $106.” it is One Up on Wall Street, Notably, legendary money manager Peter Lynch, in his book 1∕n heuristic reasonableThe 1∕ton useheuristic it in complexis a rule ofenvironments thumb that assigns when thethe levelsame ofweight understanding to each tech- is explicitly stated that for companies whose stocks are fairly valued, P∕E will equal A decision rule based on poor.nique component, as if they are all equally valid. In particular, textbook-favored the growth rate. In contrast, standard investment textbooks warn about using PEG, using the simple average of ValuationDCF receives is certainly the same complex,weight as theand PEG perhaps heuristic, the wide which range has no of basis valuation in textbook esti - a series of variables. indicating that it is a Wall Street rule of thumb that has shortcomings. A major matestheory. associated Some psychologists with the Morgan call the Stanley 1∕n heuristic team’s four“tallying” techniques and suggest suggests that thatit is shortcoming is that the value of PEG for the S&P 500 has fluctuated widely over theyreasonable have a poor to use understanding it in complex of environments the drivers ofwhen eBay’s the levelstock of market understanding value. In is time, in the range 1 to 1.5. In addition, keep in mind that in theory, a firm with thispoor. regard, the P∕E ratio associated with the Morgan Stanley team’s DCF-based k zero PVGO will have a P∕E ratio equal to 1∕ E no matter what its growth rate. In- estimateValuation of $117 is certainly was 56, suggestingcomplex, and strong perhaps growth the wide opportunities. range of valuation Given esti that- deed, its growth rate will be determined by the dividend payout ratio and return on eBay’smates stock associated was trading with atthe $89.22 Morgan a shareStanley when team’s the fourreport techniques was released, suggests a price that equity, with a higher payout lowering g but not P∕E. There are many zero PVGO targetthey of have $117 a correspondspoor understanding to an expectedof the drivers return of ofeBay’s 31 percent. stock market As 31 value. percent In firms—most children cannot be above average—and therefore their PEG ratios are exceedsthis regard, the required the P∕E return ratio ofassociated 12 percent, with the the implication Morgan Stanley is that team’s eBay’s DCF-based stock is not pegged to a specific number like 1.5. intrinsicallyestimate ofundervalued. $117 was 56, suggesting strong growth opportunities. Given that In contrast to the PEG perspective in which P∕E is proportional to growth rate ThateBay’s being stock said, was Exhibit trading 3-1at $89.22 tells us a sharethat the when Morgan the report Stanley was team’s released, valuations a price g, textbook theory tells us something quite different about intrinsic P∕E, required directlytarget based of $117 on P∕Ecorresponds and PEG to imply an expected that eBay’s return stock of 31 was percent. overvalued. As 31 Thepercent key return, and PVGO. Recall that the Morgan Stanley team’s discussion of eBay assumptionexceeds theunderlying required the return overvaluation of 12 percent, was the not implication the team’s is forecast that eBay’s for earnings stock is used a rate of 12 percent to discount eBay’s future cash flows, thereby treating growth,intrinsically but its assumptionundervalued. about P∕E, which the report predicted would decline k 12 percent as eBay’s required return E. The intrinsic forward P∕E ratio of a from 63That to 33being between said, Exhibit2003 and 3-1 2005. tells us that the Morgan Stanley team’s valuations firm with zero growth opportunities and a required return of 12 percent is 8.33 Indirectly applying based the on 1∕ P∕En heuristic, and PEG the imply Morgan that eBay’sStanley stock team was assigned overvalued. equal Theweight key (1∕0.12). At the same time, the Morgan Stanley team used a P∕E ratio of 40 in to theassumption two overvaluation underlying componentsthe overvaluation and the was two not undervaluation the team’s forecast components for earnings in their P∕E valuation heuristic. If this P∕E were intrinsic, then a value of 40 would Exhibitgrowth, 3-1. but At its the assumption same time, about the toneP∕E, ofwhich the Morganthe report Stanley predicted 2003 would report decline did imply that eBay had significant PVGO. Textbook finance theory stipulates that in not fromreflect 63 theto 33 weighting between 2003and wasand very2005. upbeat, calling many of eBay’s elements order for a firm to have positive PVGO, its expected return on equity (ROE) must “outstanding.”In applying The the report 1∕n heuristic, noted that the Morganthe number Stanley of “what’steam assigned good” equal items weight they k exceed its required return. In contrast, firms for which expected ROE equals E analyzedto the twovastly overvaluation outnumbered components the number and of the “what’s two undervaluation bad” items, and components the analyst in have zero growth opportunities. team’sExhibit recommendation 3-1. At the same for time,eBay thestock tone was of “overweight.”the Morgan Stanley Moreover, 2003 thereport report did Consider how, between the time eBay went public in 1998 and the time the indicatednot reflect that allthe the weighting positive and trends was gave very the upbeat, team calling“more manyconfidence” of eBay’s in the elements firm’s Morgan Stanley team issued its April 2003 report, the firm’s actual ROE com - long“outstanding.” term outlook, The leading report them noted to havethat the much number more ofcomfort “what’s creating good” itemslong-term they pared to its required return. Between 1998 and 2002, eBay’s ROE rose from forecasts.analyzed vastly outnumbered the number of “what’s bad” items, and the analyst 5 percent to a maximum of 10 percent, with an average ROE of 6.1 percent for Ateam’s general recommendation issue associated for with eBay the stock 1∕n washeuristic “overweight.” is that the Moreover, judgments the it gener-report the period. atesindicated are highly that dependent all the positive on the trends composition gave the teamof the “more choice confidence” menu. For in theexample, firm’s Even though its annual growth rate of its EBITDA exceeded 90 percent every whenlong applied term outlook,to portfolio leading selection them into retirementhave much plans,more comfortthe 1∕n creatingheuristic long-term leads in- year from 2000 through 2002, its ROE never rose to the required return of 12 per- vestors forecasts. to overweigh equities when their choice menu features more equity funds cent used in the Morgan Stanley report—although this does not mean that looking than bondA general funds, issue and toassociated underweigh with equities the 1∕n whenheuristic the isreverse that the holds. judgments it gener- forward in mid-2003, eBay necessarily had zero growth opportunities and that a Theates Morganare highly Stanley dependent team’s on reliance the composition on the 1∕n of heuristic the choice implies menu. that For they example, assign P∕E ratio of 40 for 2005 was unjustified. However, the history of ROE up to that a weightwhen appliedof 25 percent to portfolio to eBay’s selection stock in beingretirement driven plans, by intrinsicthe 1∕n heuristicvalue. The leads other in- point does not support the intrinsic value of P∕E being 40. Therefore, any justifica- 75 percentvestors toreflect overweigh mispricing, equities with when techniques their choice based menu on featuresP∕E and more PEG equity suggesting funds tion for intrinsic P∕E being that high would require identifying the magnitude of thatthan eBay’s bond stock funds, was and overvalued, to underweigh and price-to-salesequities when suggestingthe reverse holds.it was undervalued. 1 n growth opportunities explicitly and in a systematic fashion. Notably,The if Morgan the Morgan Stanley Stanley team’s team reliance were on tothe drop ∕ heuristicone of the implies techniques, that they such assign as a weight of 25 percent to eBay’s stock being driven by intrinsic value. The other 75 percent reflect mispricing, with techniques based on P∕E and PEG suggesting that eBay’s stock was overvalued, and price-to-sales suggesting it was undervalued. Notably, if the Morgan Stanley team were to drop one of the techniques, such as

she77208_ch03_058-084.indd 70 she77208_ch03_058-084.indd2/7/17 5:26 PM 71 2/7/17 5:26 PM

she77208_ch03_058-084.indd 71 2/7/17 5:26 PM 72 Chapter Three Valuation 73

the one based on PEG, the 1∕n heuristic would imply that the weight attached to EXHIBIT 3-5 $3.00 intrinsic value would increase to 33 percent from 25 percent. In this regard, the Cumulative Returns 72 Chapter Three Morgan Stanley team later dropped its use of the PEG-based technique—but that is to eBay and S&P 500 $2.50 getting ahead of ourselves. Between April 2003 and 1 n December 2014 theBiases one based Using on PEG, the theCAPM ∕ heuristic Heuristic: would In imply Foresight that the weight attached to $2.00 intrinsic value would increase to 33 percent from 25 percent. In this regard, the Source: Center for Research in MorganAs was Stanley noted in team Exhibit later 3-3,dropped the Morganits use of Stanley the PEG-based team used technique—but the CAPM to that come is up Security Prices. $1.50 gettingwith a ahead discount of ourselves. rate of 12 percent for its forecasted free cash flow stream. Notably, they arrived at the 12 percent rate by using a CAPM expected return of 9.5 percent, Biasesto which Using they added the CAPM2.5 percent Heuristic: reflecting In “company-specific Foresight risks.” Of course, tra- $1.00 Asditional was noted textbooks in Exhibit argue 3-3, that the in Morgan a CAPM Stanley world, team investors used the will CAPM not be to compensatedcome up withwith a higherdiscount expected rate of 12 returns percent for for bearing its forecasted non-market free cashrisk. flow stream. Notably, $0.50 theyIn arrived practice, at the the 12 estimates percent forrate the by risk using premium a CAPM are expected wide and return varied. of The9.5 percent, average text- tobook which estimate they added of the 2.5 market percent risk reflecting premium “company-specific has been 6.5 percent. risks.” However, Of course, the tra-range of ditionalvalues fortextbooks the market argue risk that premium in a CAPM across world, different investors finance will textbooks not be compensatedhas been wide—0 $0.00

withto 10 higher percent. expected And it returns has varied for bearing over time. non-market In 2003, risk.the range was 5 to 8 percent. Apr-30Feb-04Dec-04Oct-05Aug-06Jun-07Apr-08Feb-09Dec-09Oct-10Aug-11Jun-12Apr-13Feb-14Dec-14 InSurvey practice, evidence the estimates from for 2011 the riskfor premiumthe United are Stateswide and shows varied. that The analysts average havetext- the booklowest estimate estimates of the for market the market risk premium risk premium, has been at 6.5 5 percent, percent. followedHowever, bythe financial range of ex- eBay S&P 500 valuesecutives for theat 5.6market percent risk premiumand academics across different at 5.7 percent. finance However,textbooks hasthere been is morewide—0 agree- toment 10 percent. among And analysts it has aboutvaried theover value time. Inof 2003,the market the range risk was premium 5 to 8 percent. than among the stock 2-for-1. At the end of June 2004, eBay’s stock price reached $180 (on a pre- otherSurvey two evidence groups. Thefrom Morgan 2011 for Stanley the United estimate States of 4shows percent that fits analysts the general have thepattern split basis), well above the Morgan Stanley team’s price target of $106. Therefore, lowestfor analysts, estimates as forit lay the below market the risk range premium, of academic at 5 percent, estimates followed for by2003. financial ex- in hindsight, some might have felt justified in concluding that if anything the Mor- ecutivesEstimates at 5.6 forpercent beta areand wide academics and varied. at 5.714 percent. Historical However, betas, computed there is more over agree- fixed win- gan Stanley team was too pessimistic, not optimistic about eBay’s prospects.15 mentdows among of time analysts such as about 60 months, the value exhibit of the considerable market risk variation premium over than time among for the the same Turning to overconfidence, notice that Exhibit 3-1 displays the Morgan Stanley otherstock. two The groups. Morgan The Stanley Morgan team Stanley drew estimate their estimate of 4 percent of beta, fits 1.38,the general from Bloomberg.pattern team’s low to high ranges for three of the four valuation components, which ef- forA analysts,month later, as it Reuters lay below reported the range that of eBay’s academic beta estimates was 1.09. for A 2003. straight CAPM regres- fectively serve as confidence intervals. Averaging the low and high values yields a 14 sionEstimates of eBay’s for beta risk are premium wide and against varied. that Historical of the betas,S&P 500computed for the over prior fixed 24 win months- range from $72 to $137. Given that $180 lies well above the high end of the range, dowsyields of antime estimated such as 60 beta months, of 1.9. exhibit An estimate considerable based variation on 60 months over time yields for thean estimatedsame it seems reasonable to conclude that the Morgan Stanley team was overconfident in stock.beta ofThe 2.8, Morgan but the Stanley early portionteam drew of the their sample estimate occurred of beta, close 1.38, to from eBay’s Bloomberg. IPO. establishing its price target. A month later, Reuters reported that eBay’s beta was 1.09. A straight CAPM regres- As Exhibit 3-5 shows, eBay’s stock return was very high during the early period. sion of eBay’s risk premium against that of the S&P 500 for the prior 24 months At year-end 2004 the firm’s stock price closed (pre-split) at $232, with analysts’ 3.6 VALUATIONyields HEURISTICS an estimated beta AND of 1.9. AnBIASES: estimate based IN HINDSIGHTon 60 months yields an estimated consensus P∕E for eBay at 70, well above 47, which was the value for P∕E that the beta of 2.8, but the early portion of the sample occurred close to eBay’s IPO. Morgan Stanley team had forecast in their April 2003 report. To evaluate the Morgan Stanley team’s 2003 forecasts in hindsight, consider eBay’s At the same time, consider not just price targets for 2004 that were established in cumulative returns for the period May 2003 through December 2014. These are dis- 2003, but the long-term forecasts that provided the inputs for these price targets. In 3.6 VALUATION playedHEURISTICS in Exhibit 3-5, AND along BIASES: with the cumulative IN HINDSIGHT returns for the S&P 500. A detailed this regard, examine Exhibit 3-5 to see what happened from 2005 onward. Analysts’ chronology of eBay’s history for this period is provided in the Additional Resources optimistic recommendations about eBay changed on January 20, 2005, when the Tofor evaluate this chapter. the Morgan Stanley team’s 2003 forecasts in hindsight, consider eBay’s firm announced that it had missed analysts’ consensus earnings forecast by a penny cumulativeExhibit returns 3-5 shows for the that period a $1 May investment 2003 through in eBay December at the end 2014. of AprilThese 2003are dis- would a share. At the time, eBay CEO Meg Whitman indicated that future earnings would playedhave produced in Exhibit $2.41,3-5, along while with a $1the investmentcumulative returnsin the S&P for the 500 S&P would 500. haveA detailed produced be lower because of higher advertising costs and reinvestment. In response to the chronology$2.25. These of eBay’svalues historycorrespond for this to periodan annualized is provided geometric in the Additional return of 7.9Resources percent for announcement, eBay’s stock price fell by 20 percent to $81 per share. foreBay this and chapter. 7.2 percent for the S&P 500. Not surprisingly, eBay’s stock return volatil- Investors were largely disappointed with the performance of eBay’s stock until ityExhibit was much 3-5 showshigher that than a that$1 investment of the S&P in 500. eBay at the end of April 2003 would 2009, reflecting managerial decisions that had not turned out well. Thereafter, a change haveConsider produced the $2.41, price while target a $1 for investment mid-2004, in which the S&P the 500 Morgan would Stanley have produced price target in management and strategy led to better performance for the firm and its stock, as can $2.25.established These valuesin April correspond 2003. How to an did annualized that turn geometric out? In August return of 2003, 7.9 percent eBay splitfor its be seen in Exhibit 3-5 for the period late 2009 through December 2014. eBay and 7.2 percent for the S&P 500. Not surprisingly, eBay’s stock return volatil- ity was much higher than that of the S&P 500. Consider the price target for mid-2004, which the Morgan Stanley price target established in April 2003. How did that turn out? In August 2003, eBay split its

she77208_ch03_058-084.indd 72 she77208_ch03_058-084.indd2/7/17 5:26 PM 73 2/7/17 5:26 PM

she77208_ch03_058-084.indd 72 2/7/17 5:26 PM 72 Chapter Three Valuation 73 the one based on PEG, the 1∕n heuristic would imply that the weight attached to EXHIBIT 3-5 $3.00 intrinsic value would increase to 33 percent from 25 percent. In this regard, the Cumulative Returns Morgan Stanley team later dropped its use of the PEG-based technique—but that is to eBay and S&P 500 $2.50 Valuation 73 getting ahead of ourselves. Between April 2003 and DecemberEXHIBIT 2014 3-5 $3.00 Biases Using the CAPM Heuristic: In Foresight $2.00 Source:Cumulative Center for Research Returns in Security Prices. As was noted in Exhibit 3-3, the Morgan Stanley team used the CAPM to come up to eBay and S&P 500 $2.50 with a discount rate of 12 percent for its forecasted free cash flow stream. Notably, Between April 2003 and $1.50 December 2014 they arrived at the 12 percent rate by using a CAPM expected return of 9.5 percent, $2.00 to which they added 2.5 percent reflecting “company-specific risks.” Of course, tra- Source: Center for Research in $1.00 ditional textbooks argue that in a CAPM world, investors will not be compensated Security Prices. $1.50 with higher expected returns for bearing non-market risk. $0.50 In practice, the estimates for the risk premium are wide and varied. The average text- book estimate of the market risk premium has been 6.5 percent. However, the range of $1.00 values for the market risk premium across different finance textbooks has been wide—0 $0.00 to 10 percent. And it has varied over time. In 2003, the range was 5 to 8 percent. Apr-30$0.50Feb-04Dec-04Oct-05Aug-06Jun-07Apr-08Feb-09Dec-09Oct-10Aug-11Jun-12Apr-13Feb-14Dec-14 Survey evidence from 2011 for the United States shows that analysts have the lowest estimates for the market risk premium, at 5 percent, followed by financial ex- eBay S&P 500 $0.00 ecutives at 5.6 percent and academics at 5.7 percent. However, there is more agree- ment among analysts about the value of the market risk premium than among the stock Apr-302-for-1.Feb-04Dec-04 At Oct-05theAug-06 endJun-07 of Apr-08JuneFeb-09 2004,Dec-09Oct-10 eBay’sAug-11Jun-12 stockApr-13 priceFeb-14Dec-14 reached $180 (on a pre- other two groups. The Morgan Stanley estimate of 4 percent fits the general pattern split basis), well above the Morgan Stanley team’s price target of $106. Therefore, eBay S&P 500 for analysts, as it lay below the range of academic estimates for 2003. in hindsight, some might have felt justified in concluding that if anything the Mor- Estimates for beta are wide and varied.14 Historical betas, computed over fixed win- gan Stanley team was too pessimistic, not optimistic about eBay’s prospects.15 dows of time such as 60 months, exhibit considerable variation over time for the same Turningstock 2-for-1. to overconfidence, At the end of Junenotice 2004, that eBay’sExhibit stock 3-1 displaysprice reached the Morgan $180 (on Stanley a pre- stock. The Morgan Stanley team drew their estimate of beta, 1.38, from Bloomberg. team’ssplit low basis), to highwell aboveranges the for Morgan three of Stanley the four team’s valuation price target components, of $106. Therefore,which ef- A month later, Reuters reported that eBay’s beta was 1.09. A straight CAPM regres- fectivelyin hindsight, serve as some confidence might have intervals. felt justified Averaging in concluding the low andthat highif anything values theyields Mor- a sion of eBay’s risk premium against that of the S&P 500 for the prior 24 months rangegan from Stanley $72 team to $137. was tooGiven pessimistic, that $180 not lies optimistic well above about the eBay’s high end prospects. of the range,15 yields an estimated beta of 1.9. An estimate based on 60 months yields an estimated it seemsTurning reasonable to overconfidence, to conclude that notice the thatMorgan Exhibit Stanley 3-1 displays team was the overconfident Morgan Stanley in beta of 2.8, but the early portion of the sample occurred close to eBay’s IPO. establishingteam’s low its toprice high target. ranges for three of the four valuation components, which ef- Asfectively Exhibit serve 3-5 shows,as confidence eBay’s intervals. stock return Averaging was very the high low duringand high the values early yieldsperiod. a At year-endrange from 2004 $72 theto $137. firm’s Given stock that price $180 closed lies well (pre-split) above the at high$232, end with of theanalysts’ range, 3.6 VALUATION HEURISTICS AND BIASES: IN HINDSIGHT consensusit seems P∕E reasonable for eBay to atconclude 70, well that above the Morgan47, which Stanley was theteam value was foroverconfident P∕E that the in Morganestablishing Stanley its team price had target. forecast in their April 2003 report. To evaluate the Morgan Stanley team’s 2003 forecasts in hindsight, consider eBay’s At theAs Exhibitsame time, 3-5 considershows, eBay’s not just stock price return targets was forvery 2004 high that during were the established early period. in cumulative returns for the period May 2003 through December 2014. These are dis- 2003,At butyear-end the long-term 2004 the forecasts firm’s stock that priceprovided closed the (pre-split) inputs for at these $232, price with targets. analysts’ In played in Exhibit 3-5, along with the cumulative returns for the S&P 500. A detailed thisconsensus regard, examine P∕E for Exhibit eBay at 3-5 70, to well see above what happened47, which fromwas the 2005 value onward. for P∕E Analysts’ that the chronology of eBay’s history for this period is provided in the Additional Resources optimisticMorgan recommendations Stanley team had forecastabout eBay in their changed April 2003 on January report. 20, 2005, when the for this chapter. firm announcedAt the same that time, it had consider missed not analysts’ just price consensus targets for earnings 2004 that forecast were established by a penny in Exhibit 3-5 shows that a $1 investment in eBay at the end of April 2003 would a share.2003, At but the the time, long-term eBay CEO forecasts Meg that Whitman provided indicated the inputs that for future these earningsprice targets. would In have produced $2.41, while a $1 investment in the S&P 500 would have produced be lowerthis regard, because examine of higher Exhibit advertising 3-5 to see costswhat happenedand reinvestment. from 2005 In onward. response Analysts’ to the $2.25. These values correspond to an annualized geometric return of 7.9 percent for announcement,optimistic recommendations eBay’s stock price about fell eBayby 20 changed percent onto $81January per share.20, 2005, when the eBay and 7.2 percent for the S&P 500. Not surprisingly, eBay’s stock return volatil- Investorsfirm announced were largely that it had disappointed missed analysts’ with the consensus performance earnings of forecast eBay’s by stock a penny until ity was much higher than that of the S&P 500. 2009,a share. reflecting At the managerial time, eBay decisions CEO Meg that Whitman had not turnedindicated out that well. future Thereafter, earnings a change would Consider the price target for mid-2004, which the Morgan Stanley price target in managementbe lower because and strategy of higher led advertisingto better performance costs and reinvestment. for the firm and In responseits stock, asto canthe established in April 2003. How did that turn out? In August 2003, eBay split its be seenannouncement, in Exhibit 3-5 eBay’s for thestock period price late fell 2009 by 20 through percent December to $81 per 2014.share. Investors were largely disappointed with the performance of eBay’s stock until 2009, reflecting managerial decisions that had not turned out well. Thereafter, a change in management and strategy led to better performance for the firm and its stock, as can be seen in Exhibit 3-5 for the period late 2009 through December 2014.

she77208_ch03_058-084.indd 72 she77208_ch03_058-084.indd2/7/17 5:26 PM 73 2/7/17 5:26 PM

she77208_ch03_058-084.indd 73 2/7/17 5:26 PM 74 Chapter Three Valuation 75

EXHIBIT 3-6 Comparison of Morgan Stanley Forecasts from April 2003 to Actual Values for Period 2003 EXHIBIT 3-7 $4,500 Through 2010 Comparison of Morgan $4,000 Chapter Three 74Source of data: Top panel is from Morgan Stanley Report on eBay, April 2003. Bottom panel is from Compustat, accessed through WRDS. Stanley Team’s Forecast of Free Cash Flows with $3,500 Forecast ($ Thousands) 2003E 2004E 2005E 2006E 2007E 2008E 2009E 2010E Actual Free Cash Flows, $3,000 EXHIBIT 3-6 Comparison of Morgan Stanley Forecasts from April 2003 to Actual Values for Period 2003 2003–2014, $million Revenue Through 20102,050,492 2,829,679 3,763,473 4,907,751 6,287,602 7,857,002 9,582,843 11,400,983 $2,500 EBITDA 723,735 1,005,276 1,401,682 1,853,230 2,426,297 3,110,990 3,985,085 4,982,032 Source of data: Morgan Stanley Source of data: Top panel is from Morgan Stanley Report on eBay, April 2003. Bottom panel is from Compustat, accessed through WRDS. Report on eBay, April 2003, and $2,000 Taxes — — 140,168 370,646 849,204 1,088,846 1,394,780 1,743,711 Compustat accessed through $1,500 ForecastChange in ($ Working Thousands) Capital 2003E 26,792 2004E (6,849) 2005E 58,695 2006E (9,709) 2007E 52,096 2008E (7,375) 2009E 53,278 2010E (4,105) WRDS. RevenueCapital Expenditures 2,050,492 188,908 2,829,679 190,000 3,763,473 190,000 4,907,751 190,000 6,287,602 190,000 7,857,002 190,000 9,582,843 190,000 11,400,983 190,000 $1,000 EBITDAFree Cash Flow 723,735508,035 1,005,276 822,125 1,401,682 1,012,819 1,853,230 1,302,292 2,426,297 1,334,997 3,110,990 1,839,519 3,985,085 2,347,027 4,982,032 3,052,426 $500 Taxes — — 140,168 370,646 849,204 1,088,846 1,394,780 1,743,711 $0 Change in Working Capital 26,792 (6,849) 58,695 (9,709) 52,096 (7,375) 53,278 (4,105) 2002 CapitalActual Expenditures($ Thousands) 188,9082003 190,000 2004 190,000 2005 190,000 2006 190,000 2007 190,000 2008 190,000 2009 190,000 2010 2003E 2004E 2005E 2006E 2007E 2008E 2009E 2010E 2011E 2012E 2013E 2014E FreeRevenue Cash Flow 508,0352,165,096 822,1253,271,309 1,012,819 4,552,401 1,302,292 5,969,741 1,334,997 7,672,329 1,839,519 8,541,261 2,347,027 8,727,362 3,052,426 9,156,274 Morgan Stanley FCF actual using Morgan EBITDA 818,209 1,312,932 1,819,872 1,967,508 2,605,825 2,844,615 2,649,098 2,837,473 2003 FCF forecasts Stanley methodology Taxes 3,237 8,234 3,478 5,916 10,474 7,759 6,050 54 Actual ($ Thousands) 2003 2004 2005 2006 2007 2008 2009 2010 Change in Working Capital 29,432 190,588 4,242 (204,674) 172,318 147,623 (220,575) 899,102 RevenueCapital Expenditures 2,165,096 365,384 3,271,309 292,838 4,552,401 338,281 5,969,741 515,448 7,672,329 453,967 8,541,261 565,890 8,727,362 567,094 9,156,274 723,912 Exhibit 3-7 for the period beginning in 2011. First, actual FCF lay below forecasted EBITDA 818,209 1,312,932 1,819,872 1,967,508 2,605,825 2,844,615 2,649,098 2,837,473 FCF. Second, the growth rate of FCF after 2010 is highly variable, but does appear to Free Cash Flow 420,156 821,272 1,473,871 1,650,818 1,969,066 2,123,343 2,296,529 1,214,405 Taxes 3,237 8,234 3,478 5,916 10,474 7,759 6,050 54 be consistent with the forecasted growth rate after 2011. It just began at a lower level. Change in Working Capital 29,432 190,588 4,242 (204,674) 172,318 147,623 (220,575) 899,102 Third, by 2010 the Morgan Stanley team, with some but not all of the original Capital Expenditures 365,384 292,838 338,281 515,448 453,967 565,890 567,094 723,912 members, had changed parts of its approach. For one thing, it dropped the use of Free Cash Flow 420,156With this 821,272 history as 1,473,871 a backdrop, 1,650,818 how did 1,969,066 the Morgan 2,123,343 Stanley 2,296,529 team’s long-term 1,214,405 fore- the PEG heuristic. For another, it modified its views on the long-term growth rate cast turn out in respect to sales, EBITDA, CapEx, working capital, and free cash flow? of free cash flows, and lowered its forecast from 7 percent to 4.5 percent, with the Exhibit 3-6 displays the comparison. We can see that the Morgan Stanley team was terminal horizon beginning in 2020. This modification is highly significant, and reasonably accurate in their forecasts for revenue and EBITDA up through the end of raises the following question: In 2003, did the Morgan Stanley team exhibit bias by 2008With when this history the global as a backdrop,financial crisishow did occurred. the Morgan They Stanley were lessteam’s accurate long-term in their fore- fore- choosing a terminal horizon that began too early and featured too high a long-term castcasts turn of out taxes, in respect working to sales,capital, EBITDA, and capital CapEx, expenditures. working capital, Nevertheless, and free cash their flow? forecasts growth rate, not so much for the period around 2011, but for the period further into Exhibitof eBay’s 3-6 freedisplays cash theflows comparison. were, if anything, We can seetoo thatlow thethrough Morgan the Stanley end of 2009.team was Only in the future that would begin later, such as in 2020? Coming back to the discussion reasonably2010 did eBay’saccurate actual in their free forecasts cash flow for fallrevenue short and of theEBITDA Morgan up Stanley through team’s the end forecast. of of CAPM, a CAPM regression for the period May 2003 through December 2014 2008Given when thethe difficultyglobal financial of forecasting crisis occurred. the future They ofwere a firm less accuratelike eBay, in theirthe degree fore- to produces an estimate of beta that is 1.5, with an estimated alpha that is effectively castswhich of thetaxes, Morgan working Stanley capital, team’s and capital forecasts expenditures. for the intermediate Nevertheless, horizon their forecasts came close zero. In hindsight, the Morgan Stanley team’s 1.38 estimate of beta was a little low, ofto eBay’s the actual free cashvalues flows is noteworthy. were, if anything, As for too 2010 low when through free the cash end flow of 2009. was Onlywell belowin but within range of the subsequent actual value of 1.5 for the period. The team’s 2010its forecasted did eBay’s value,actual freethat cashmight flow have fall been short a of lagged the Morgan effect Stanley from the team’s financial forecast. crisis, estimate of 4 percent for the market risk premium was much lower than its historical asGiven revenues the difficultyand EBITDA of forecasting were both the significantly future of a lowerfirm likethan eBay, forecast. the degree to value and its value for the intermediate horizon. The annualized geometric return whichTurning the Morgan next toStanley the terminal team’s forecastsperiod, remember for the intermediate that more horizon than 80 came percent close of the for the S&P 500 was 7.2 percent, its annualized arithmetic return was 8.2 percent, tovalue the actual in the values 2003 isMorgan noteworthy. Stanley As forprice 2010 target when of free eBay cash came flow from was terminalwell below value, and its annualized arithmetic risk premium was 6.8 percent. itsand forecasted terminal value, value that is heavily might havedependent been a onlagged the forecasteffect from of freethe financialcash flow crisis, in 2010. asTherefore, revenues andif the EBITDA 2003 DCFwere bothcomputation significantly was lower intended than toforecast. provide an estimate of eBay’sTurning intrinsic next to value the terminal at the time, period, the remember magnitudes that of more free cashthan 80flow percent in 2011 of theand the 3.7 BIASES ASSOCIATED WITH FREE CASH FLOW FORMULA valuegrowth in therate 2003 of those Morgan cash Stanley flows from price 2011 target onward of eBay are came both from crucial. terminal value, and Exhibitterminal 3-7value contrasts is heavily forecasted dependent and on actual the forecast free cash of free flows cash for flow the in period 2010. 2002 The Morgan Stanley team used the textbook formula for free cash flow in its DCF Therefore,through 2014.if the Here2003 DCF 2002, computation which corresponds was intended to the to provide actual value,an estimate is displayed of analysis. It is important to understand the basis for this formula, and how it relates eBay’sas a starting intrinsic value value forat the both time, series. the magnitudes There are of three free important cash flow pointsin 2011 to and note the from to the value of a firm and the value of its equity. growth rate of those cash flows from 2011 onward are both crucial. Exhibit 3-7 contrasts forecasted and actual free cash flows for the period 2002 through 2014. Here 2002, which corresponds to the actual value, is displayed as a starting value for both series. There are three important points to note from

she77208_ch03_058-084.indd 74 she77208_ch03_058-084.indd2/7/17 5:26 PM 75 2/7/17 5:26 PM

she77208_ch03_058-084.indd 74 2/7/17 5:26 PM 74 Chapter Three Valuation 75

EXHIBIT 3-6 Comparison of Morgan Stanley Forecasts from April 2003 to Actual Values for Period 2003 EXHIBIT 3-7 $4,500 Through 2010 Comparison of Morgan $4,000

Source of data: Top panel is from Morgan Stanley Report on eBay, April 2003. Bottom panel is from Compustat, accessed through WRDS. Stanley Team’s Forecast Valuation 75 of Free Cash Flows with $3,500 Forecast ($ Thousands) 2003E 2004E 2005E 2006E 2007E 2008E 2009E 2010E ActualEXHIBIT Free Cash 3-7 Flows, $3,000$4,500 Revenue 2,050,492 2,829,679 3,763,473 4,907,751 6,287,602 7,857,002 9,582,843 11,400,983 2003–2014, $million Comparison of Morgan $2,500$4,000 EBITDA 723,735 1,005,276 1,401,682 1,853,230 2,426,297 3,110,990 3,985,085 4,982,032 Source Stanleyof data: Morgan Team’s Stanley Forecast Report on eBay, April 2003, and $2,000$3,500 Taxes — — 140,168 370,646 849,204 1,088,846 1,394,780 1,743,711 of Free Cash Flows with Compustat accessed through $1,500 Change in Working Capital 26,792 (6,849) 58,695 (9,709) 52,096 (7,375) 53,278 (4,105) WRDS.Actual Free Cash Flows, $3,000 Capital Expenditures 188,908 190,000 190,000 190,000 190,000 190,000 190,000 190,000 2003–2014, $million $1,000$2,500 Free Cash Flow 508,035 822,125 1,012,819 1,302,292 1,334,997 1,839,519 2,347,027 3,052,426 Source of data: Morgan Stanley $500 Report on eBay, April 2003, and $2,000 Compustat accessed through $0$1,500 WRDS. 2002 Actual ($ Thousands) 2003 2004 2005 2006 2007 2008 2009 2010 $1,000 2003E 2004E 2005E 2006E 2007E 2008E 2009E 2010E 2011E 2012E 2013E 2014E

Revenue 2,165,096 3,271,309 4,552,401 5,969,741 7,672,329 8,541,261 8,727,362 9,156,274 $500 Morgan Stanley FCF actual using Morgan EBITDA 818,209 1,312,932 1,819,872 1,967,508 2,605,825 2,844,615 2,649,098 2,837,473 $0 2003 FCF forecasts Stanley methodology Taxes 3,237 8,234 3,478 5,916 10,474 7,759 6,050 54 2002 Change in Working Capital 29,432 190,588 4,242 (204,674) 172,318 147,623 (220,575) 899,102 2003E 2004E 2005E 2006E 2007E 2008E 2009E 2010E 2011E 2012E 2013E 2014E Capital Expenditures 365,384 292,838 338,281 515,448 453,967 565,890 567,094 723,912 Exhibit 3-7 for theMorgan period Stanley beginning inFCF 2011. actual First, using Morgan actual FCF lay below forecasted FCF. Second, the growth2003 FCF rateforecasts of FCF afterStanley 2010 methodology is highly variable, but does appear to Free Cash Flow 420,156 821,272 1,473,871 1,650,818 1,969,066 2,123,343 2,296,529 1,214,405 be consistent with the forecasted growth rate after 2011. It just began at a lower level. Third, by 2010 the Morgan Stanley team, with some but not all of the original members,Exhibit had3-7 forchanged the period parts beginning of its approach. in 2011. For First, one actual thing, FCF it droppedlay below the forecasted use of With this history as a backdrop, how did the Morgan Stanley team’s long-term fore- the FCF.PEG Second,heuristic. the For growth another, rate ofit FCFmodified after 2010its views is highly on the variable, long-term but does growth appear rate to cast turn out in respect to sales, EBITDA, CapEx, working capital, and free cash flow? of freebe consistent cash flows, with and the lowered forecasted its growth forecast rate from after 7 2011.percent It just to 4.5began percent, at a lower with level. the Exhibit 3-6 displays the comparison. We can see that the Morgan Stanley team was terminalThird, horizon by 2010 beginning the Morgan in 2020. Stanley This team, modification with some is but highly not all significant, of the original and reasonably accurate in their forecasts for revenue and EBITDA up through the end of raisesmembers, the following had changed question: parts In of2003, its approach. did the Morgan For one Stanley thing, itteam dropped exhibit the bias use byof 2008 when the global financial crisis occurred. They were less accurate in their fore- choosingthe PEG a terminal heuristic. horizon For another, that began it modified too early its viewsand featured on the long-termtoo high a growth long-term rate casts of taxes, working capital, and capital expenditures. Nevertheless, their forecasts growthof free rate, cash not flows, so much and for lowered the period its forecast around from 2011, 7 percentbut for tothe 4.5 period percent, further with into the of eBay’s free cash flows were, if anything, too low through the end of 2009. Only in the terminalfuture that horizon would beginning begin later, in 2020.such asThis in 2020?modification Coming is backhighly to significant,the discussion and 2010 did eBay’s actual free cash flow fall short of the Morgan Stanley team’s forecast. of CAPM,raises the a followingCAPM regression question: Infor 2003, the period did the May Morgan 2003 Stanley through team December exhibit bias 2014 by Given the difficulty of forecasting the future of a firm like eBay, the degree to produceschoosing an estimatea terminal of horizon beta that that is began 1.5, with too earlyan estimated and featured alpha too that high is aeffectively long-term which the Morgan Stanley team’s forecasts for the intermediate horizon came close zero.growth In hindsight, rate, not the so muchMorgan for Stanleythe period team’s around 1.38 2011, estimate but for of the beta period was afurther little low, into to the actual values is noteworthy. As for 2010 when free cash flow was well below but thewithin future range that ofwould the subsequentbegin later, suchactual as value in 2020? of 1.5 Coming for the back period. to the The discussion team’s its forecasted value, that might have been a lagged effect from the financial crisis, estimateof CAPM, of 4 percent a CAPM for regression the market for risk the premium period May was much2003 throughlower than December its historical 2014 as revenues and EBITDA were both significantly lower than forecast. valueproduces and its an value estimate for the of betaintermediate that is 1.5, horizon. with an The estimated annualized alpha geometricthat is effectively return Turning next to the terminal period, remember that more than 80 percent of the for thezero. S&P In hindsight, 500 was the7.2 Morganpercent, Stanley its annualized team’s 1.38 arithmetic estimate return of beta was was 8.2 a little percent, low, value in the 2003 Morgan Stanley price target of eBay came from terminal value, andbut its annualizedwithin range arithmetic of the subsequent risk premium actual was value 6.8 of percent. 1.5 for the period. The team’s and terminal value is heavily dependent on the forecast of free cash flow in 2010. estimate of 4 percent for the market risk premium was much lower than its historical Therefore, if the 2003 DCF computation was intended to provide an estimate of value and its value for the intermediate horizon. The annualized geometric return eBay’s intrinsic value at the time, the magnitudes of free cash flow in 2011 and the 3.7 BIASES ASSOCIATEDfor the S&P WITH500 was 7.2FREE percent, CASH its annualized FLOW arithmetic FORMULA return was 8.2 percent, growth rate of those cash flows from 2011 onward are both crucial. and its annualized arithmetic risk premium was 6.8 percent. Exhibit 3-7 contrasts forecasted and actual free cash flows for the period 2002 The Morgan Stanley team used the textbook formula for free cash flow in its DCF through 2014. Here 2002, which corresponds to the actual value, is displayed analysis. It is important to understand the basis for this formula, and how it relates as a starting value for both series. There are three important points to note from 3.7 BIASES ASSOCIATEDto the value of a firm WITH and the FREE value CASHof its equity. FLOW FORMULA The Morgan Stanley team used the textbook formula for free cash flow in its DCF analysis. It is important to understand the basis for this formula, and how it relates to the value of a firm and the value of its equity.

she77208_ch03_058-084.indd 74 she77208_ch03_058-084.indd2/7/17 5:26 PM 75 2/7/17 5:26 PM

she77208_ch03_058-084.indd 75 2/7/17 5:26 PM 76 Chapter Three Valuation 77

3.8 AGENCY CONFLICTS Corporate Nudges 76 Chapter Three It is important to keep in mind that analysts face conflicts of interest. Analysts are Errors or biases: Misvaluing stocks. not necessarily objective evaluators, but instead are agents of financial organiza- Why does it happen? Managers and analysts rely on crude heuristics to measure tions that seek to do business with the firms being covered. These organizations Corporatefundamental value, Nudges use biased forecasts of future cash flows, and misframe free earn fees from firms’ raising external capital, and earn commissions from the trad- cash flows. ing of their securities. In respect to Morgan Stanley, that point is prominently noted How does it happen? Managers and analysts find heuristics based on P/E, PEG, and in a text box on the first page of the Morgan Stanley team’s report. Errors or biases: Misvaluing stocks. price-to-sales to be simpler and more intuitive than DCF. Moreover, managers and ana- Why does it happen? Managers and analysts rely on crude heuristics to measure The managers of firms prefer favorable coverage from analysts to unfavorable lysts are prone to using cash flows that reflect biases stemming from excessive optimism fundamental value, use biased forecasts of future cash flows, and misframe free coverage. Therefore, analysts whose firms seek to do business with the companies and misframing. cash flows. that analysts cover have an incentive to generate favorable reports. This agency What can be done about it? Managers and analysts can: How does it happen? Managers and analysts find heuristics based on P/E, PEG, and conflict may induce analysts to be excessively optimistic about their forecasts of price-to-sales• Apply a variety to be simpler of valuation and more techniques intuitive that than are DCF. firmly Moreover, grounded managers in textbook and ana-theory, in earnings and cash flows. The agency conflict might also induce analysts to view lysts additionare prone to to crude using ratio-basedcash flows that heuristics, reflect biases and weight stemming them from sensibly. excessive optimism valuation heuristics as alternative instruments that can provide numbers they want and• misframing. Ensure that terminal value based on free cash flow features zero PVGO, and if not, to deliver. In other words, excessive optimism and the reliance on heuristics may Whathave can abe reasonable done about explanation it? Managers for and why analysts not. can: have an agency conflict component as well as a behavioral component. • • Apply Develop a variety confidence of valuation intervals techniques for valuation that are firmly judgments grounded that in can textbook be calibrated theory, in over The general evidence is that agency issues explain some but not all of the bi- additiontime. to crude ratio-based heuristics, and weight them sensibly. ases in analysts’ price targets. An important indication of the impact of agency • • Ensure Avoid thatthe dogmaticterminal value use ofbased heuristics on free that cash tie flowP/E tofeatures earnings zero growth. PVGO, and if not, considerations is that the forecasted returns implied by analysts’ price targets are • have Prepare a reasonable free cash explanation flow forecasts for why based not. on cash flow from operations, cash flow from unjustifiably higher for firms that raise external financing than those that do not. • Developinvestment, confidence and interest intervals paid. for valuation judgments that can be calibrated over Price target-based forecasted returns are also positively related to trading volume, time. controlling for firm size—and yet, realized returns and volume are uncorrelated. • Avoid the dogmatic use of heuristics that tie P/E to earnings growth. • Prepare free cash flow forecasts based on cash flow from operations, cash flow from investment, and interest paid. The textbook formula for the free cash flow FCF of a firm is: Summary The main takeaway from this chapter is that heuristics and biases play a major role in valuation. The financial world is an imperfect place. This is not to say that FCF = EBIT(1 − tax rate) + Depreciation − CapEx − Change in Working Capital financial textbooks fail to offer rigorous techniques for doing valuation. Textbooks An alternative formula for free cash flow is based on items appearing in the do provide structured valuation processes, some of which are quite sophisticated. StatementThe textbook of Cash formula Flows. for theThe free alternative cash flow formula FCF of isa firmto add is: cash flow from opera- Nevertheless, in practice, a lot of valuation is heuristically driven. This chapter has tionsFCF and = EBIT(1 interest − paid, tax rate) and + then Depreciation to subtract − CapEx cash −flow Change from in investment Working Capital (where a concentrated on the case of a single firm, eBay, to illustrate the point. To be sure, positive entry for investment is interpreted as a cash outflow). it is important to refrain from overgeneralizing from a single case, and the chapter AnTechnically, alternative the formula most forimportant free cash issue flow relating is based to on the items two appearing formulas infor the com - minicase provides an opportunity to investigate valuation issues for another firm Statementputing free of Cashcash Flows.flow is The that alternative CapEx is formula only one is to component add cash flow of cashfrom flowopera- from and the judgments of other analysts. tionsinvestment. and interest Conceptually, paid, and then the mostto subtract important cash flowissue fromis that investment the textbook (where formula a Optimism, overconfidence, representativeness, and anchoring are psychological positivemight featureentry for bias investment and not isyield interpreted the amount as a cashof cash outflow). per year a firm generates that phenomena that arose several times in the discussion about eBay, and were manifest isTechnically, available to thebe paidmost to important its investors. issue Cash relating flow to available the two formulasto be paid for to cominvestors- in questions such as the following: Did the Morgan Stanley team’s assumed growth putingis the freesum cashof cash flow flow is thatactually CapEx paid is andonly cash one flow component added toof cash cash and flow marketable from rate for terminal value reflect excessive optimism? Was the fact that the actual stock investment.securities. Conceptually, A discussion the in themost Additional important Resourcesissue is that to the Chapter textbook 3 usesformula eBay’s price lay outside their forecast range imply that they were overconfident? Do ana- mightfinancial feature statements bias and tonot illustrate yield the that amount between of cash 2002 per and year 2014 a firm the generates present thatvalue of lysts generally view earnings growth as representative of growth opportunities in iscash available flow availableto be paid to to be its paidinvestors. to the Cash firm’s flow investors available was to roughly be paid one-thirdto investors of the the sense of PVGO? Do analysts’ free cash flow forecasts for the terminal period isvalue the sum associated of cash flowwith actually the free paid cash and flows cash computed flow added using to cash the and textbook marketable formula. implicitly assume positive growth opportunities? In arriving at a discount rate, do securities.In retrospect, A discussion investors inwho the relied Additional on the Resources textbook formula to Chapter for 3free uses cash eBay’s flow to analysts anchor on estimates based on the CAPM? Do CFOs rely on a PEG heuristic financialestimate statements the amount to ofillustrate cash eBay that generatedbetween 2002 that andwas 2014 available the present to be paidvalue to of them because they regard earnings growth as the basis for P∕E? Do analysts, financial ex- cashshould flow have available felt duped. to be paid to the firm’s investors was roughly one-third of the ecutives, and investors misframe free cash flows because they employ the textbook value associated with the free cash flows computed using the textbook formula. In retrospect, investors who relied on the textbook formula for free cash flow to estimate the amount of cash eBay generated that was available to be paid to them should have felt duped.

she77208_ch03_058-084.indd 76 she77208_ch03_058-084.indd2/7/17 5:26 PM 77 2/7/17 5:26 PM

she77208_ch03_058-084.indd 76 2/7/17 5:26 PM 76 Chapter Three Valuation 77

3.8 AGENCY CONFLICTS Corporate Nudges It is important to keep in mind that analysts face conflicts of interest. AnalystsValuation are77 Errors or biases: Misvaluing stocks. not necessarily objective evaluators, but instead are agents of financial organiza- Why does it happen? Managers and analysts rely on crude heuristics to measure 3.8 AGENCY CONFLICTStions that seek to do business with the firms being covered. These organizations fundamental value, use biased forecasts of future cash flows, and misframe free earn fees from firms’ raising external capital, and earn commissions from the trad- cash flows. ing Itof is their important securities. to keep In respect in mind to that Morgan analysts Stanley, face conflicts that point of isinterest. prominently Analysts noted are How does it happen? Managers and analysts find heuristics based on P/E, PEG, and in anot text necessarily box on the objective first page evaluators, of the Morgan but instead Stanley are team’s agents report. of financial organiza- price-to-sales to be simpler and more intuitive than DCF. Moreover, managers and ana- Thetions managers that seek ofto firmsdo business prefer withfavorable the firms coverage being fromcovered. analysts These to organizations unfavorable lysts are prone to using cash flows that reflect biases stemming from excessive optimism coverage.earn fees Therefore, from firms’ analysts raising whose external firms capital, seek andto do earn business commissions with the from companies the trad- and misframing. thating analysts of their cover securities. have In an respect incentive to Morgan to generate Stanley, favorable that point reports. is prominently This agency noted What can be done about it? Managers and analysts can: conflictin a text may box induce on the analysts first page to ofbe the excessively Morgan Stanley optimistic team’s about report. their forecasts of • Apply a variety of valuation techniques that are firmly grounded in textbook theory, in earningsThe and managers cash flows. of firms The prefer agency favorable conflict coverage might also from induce analysts analysts to unfavorable to view addition to crude ratio-based heuristics, and weight them sensibly. valuationcoverage. heuristics Therefore, as alternative analysts whose instruments firms seek that to can do businessprovide numberswith the companiesthey want • Ensure that terminal value based on free cash flow features zero PVGO, and if not, to deliver.that analysts In other cover words, have excessive an incentive optimism to generate and the favorable reliance reports. on heuristics This agency may have a reasonable explanation for why not. have conflict an agency may conflict induce analystscomponent to be as excessivelywell as a behavioral optimistic component. about their forecasts of • Develop confidence intervals for valuation judgments that can be calibrated over Theearnings general and evidence cash flows. is thatThe agencyagency issuesconflict explain might alsosome induce but not analysts all of tothe view bi- time. ases valuation in analysts’ heuristics price as targets. alternative An importantinstruments indication that can provide of the numbers impact of they agency want • Avoid the dogmatic use of heuristics that tie P/E to earnings growth. considerationsto deliver. In is other that thewords, forecasted excessive returns optimism implied and bythe analysts’reliance onprice heuristics targets may are • Prepare free cash flow forecasts based on cash flow from operations, cash flow from unjustifiablyhave an agency higher conflict for firms component that raise as wellexternal as a financingbehavioral thancomponent. those that do not. investment, and interest paid. Price target-basedThe general evidenceforecasted is returnsthat agency are also issues positively explain relatedsome but to nottrading all of volume, the bi- controllingases in analysts’for firm size—and price targets. yet, Anrealized important returns indication and volume of the are impact uncorrelated. of agency considerations is that the forecasted returns implied by analysts’ price targets are unjustifiably higher for firms that raise external financing than those that do not. Price target-based forecasted returns are also positively related to trading volume, The textbook formula for the free cash flow FCF of a firm is: Summary Thecontrolling main takeaway for firm from size—and this chapter yet, realized is that returns heuristics and volume and biases are uncorrelated. play a major role in valuation. The financial world is an imperfect place. This is not to say that FCF = EBIT(1 − tax rate) + Depreciation − CapEx − Change in Working Capital financial textbooks fail to offer rigorous techniques for doing valuation. Textbooks An alternative formula for free cash flow is based on items appearing in the do provide structured valuation processes, some of which are quite sophisticated. Statement of Cash Flows. The alternative formula is to add cash flow from opera- Summary Nevertheless,The main takeawayin practice, from a lot this of valuation chapter is is that heuristically heuristics driven. and biases This play chapter a major has tions and interest paid, and then to subtract cash flow from investment (where a concentratedrole in valuation. on the caseThe financialof a single world firm, is eBay,an imperfect to illustrate place. the This point. is not To to besay sure, that positive entry for investment is interpreted as a cash outflow). it isfinancial important textbooks to refrain fail from to offer overgeneralizing rigorous techniques from fora single doing case, valuation. and the Textbooks chapter Technically, the most important issue relating to the two formulas for com- minicasedo provide provides structured an opportunity valuation processes,to investigate some valuation of which issues are quite for sophisticated.another firm puting free cash flow is that CapEx is only one component of cash flow from andNevertheless, the judgments in of practice, other analysts. a lot of valuation is heuristically driven. This chapter has investment. Conceptually, the most important issue is that the textbook formula Optimism,concentrated overconfidence, on the case of arepresentativeness, single firm, eBay, andto illustrate anchoring the arepoint. psychological To be sure, might feature bias and not yield the amount of cash per year a firm generates that phenomenait is important that arose to refrain several from times overgeneralizing in the discussion from about a single eBay, case, and and were the manifest chapter is available to be paid to its investors. Cash flow available to be paid to investors in questionsminicase suchprovides as the an following: opportunity Did to theinvestigate Morgan valuationStanley team’s issues assumedfor another growth firm is the sum of cash flow actually paid and cash flow added to cash and marketable rateand for theterminal judgments value of reflect other excessiveanalysts. optimism? Was the fact that the actual stock securities. A discussion in the Additional Resources to Chapter 3 uses eBay’s price layOptimism, outside overconfidence,their forecast range representativeness, imply that they and were anchoring overconfident? are psychological Do ana- financial statements to illustrate that between 2002 and 2014 the present value of lystsphenomena generally thatview arose earnings several growth times inas the representative discussion about of growth eBay, and opportunities were manifest in cash flow available to be paid to the firm’s investors was roughly one-third of the the insense questions of PVGO? such as Do the analysts’ following: free Did cash the Morganflow forecasts Stanley for team’s the terminalassumed growthperiod value associated with the free cash flows computed using the textbook formula. implicitlyrate for assumeterminal positive value reflect growth excessive opportunities? optimism? In Was arriving the fact at athat discount the actual rate, stock do In retrospect, investors who relied on the textbook formula for free cash flow to analystsprice anchorlay outside on estimates their forecast based range on the imply CAPM? that Dothey CFOs were relyoverconfident? on a PEG heuristic Do ana- estimate the amount of cash eBay generated that was available to be paid to them becauselysts theygenerally regard view earnings earnings growth growth as the as basisrepresentative for P∕E? ofDo growth analysts, opportunities financial ex- in should have felt duped. ecutives,the sense and ofinvestors PVGO? misframe Do analysts’ free free cash cash flows flow because forecasts they for employ the terminal the textbook period implicitly assume positive growth opportunities? In arriving at a discount rate, do analysts anchor on estimates based on the CAPM? Do CFOs rely on a PEG heuristic because they regard earnings growth as the basis for P∕E? Do analysts, financial ex- ecutives, and investors misframe free cash flows because they employ the textbook

she77208_ch03_058-084.indd 76 she77208_ch03_058-084.indd2/7/17 5:26 PM 77 2/7/17 5:26 PM

she77208_ch03_058-084.indd 77 2/7/17 5:26 PM she77208_ch03_058-084.indd 78 she77208_ch03_058-084.indd 78 www.mhhe.com/shefrin2e Explore theWeb Key Terms Readings Behavioral Additional ChapterThree 78 Explore theWeb Key Terms Readings Behavioral Additional ChapterThree 78 growth opportunitiesbias, 1/n heuristic,71 Bhatia, S.,“TheHistoryofeBay,”TheTelegraphOnline,November17,2015. 2014. “eBay IncatCitiGlobalTechnologyConference−Final,”CQFDDisclosure,September4, (FAIR DISCLOSURE)WIRE,June8,2005. “eBay atBear,Stearns&Co.Inc.16thAnnualTechnologyConference−Final,”FD Journal, April23,2003. The Bullish ProjectionsAssumeGiganticRiseinOnlineShopping,”TheWallStreet Wingfield, N.,“StockofeBayActsLike.eBay,ButHowCanSharesStayLofty? School Press,2002. Rappaport, A.andM.Mauboussin,ExpectationsInvesting.Boston:HarvardBusiness Montier, J.,BehaviouralFinance.WestSussex,UK:Wiley,2002. Press, 2000. Martin, J.andW.Petty,ValueBasedManagement.Boston:HarvardBusinessSchool strategically, willpleasemanagers. atnumbersthat inorder to arrive the managers of the theyfirms cover. Hence, they might choose valuation heuristics analyses DCF-based las compared to the intextbooks. taught require fewvery price-to-sales and PEG, P∕E, involveand variables formu-simple fundamental valuation techniques taught in textbooks. Heuristics are simpler to use: questionsthese isyes. answer the eachto definitive,argumentsthat of be togood made be can difficult is it Although overlysimplistic? is definition that when flow cash free of definition Investors. The websiteforAetnahasseveral sectionsofinterest,especiallyWhoWeAreand www.aetna.com can viewsamplereports. The websiteforMorganStanleyhasaninterestingsectionwhere institutionalinvestors www.morganstanley.com gateway tothefirm’stransactionservices. The websiteforeBaycontainsfinancialinformationaboutthefirm, aswellservinga www.ebay.com Investors. The websiteforAetnahasseveralsectionsofinterest,especially WhoWeAreand www.aetna.com can viewsamplereports. The websiteforMorganStanleyhasaninterestingsectionwhere institutionalinvestors www.morganstanley.com gateway tothefirm’stransactionservices. The websiteforeBaycontainsfinancialinformationaboutthe firm,aswellservinga www.ebay.com growth opportunitiesbias, 1/n heuristic,71 Bhatia, S.,“TheHistoryofeBay,”TheTelegraphOnline,November17,2015. 2014. “eBay IncatCitiGlobalTechnologyConference−Final,”CQFDDisclosure,September4, (FAIR DISCLOSURE)WIRE,June8,2005. “eBay atBear,Stearns&Co.Inc.16thAnnualTechnologyConference−Final,”FD Journal, April23,2003. The Bullish ProjectionsAssumeGiganticRiseinOnlineShopping,”TheWallStreet Wingfield, N.,“StockofeBayActsLike.eBay,ButHowCanSharesStayLofty? School Press,2002. Rappaport, A.andM.Mauboussin,ExpectationsInvesting.Boston:HarvardBusiness Montier, J.,BehaviouralFinance.WestSussex,UK:Wiley,2002. Press, 2000. Martin, J.andW.Petty,ValueBasedManagement.Boston:HarvardBusinessSchool strategically, willpleasemanagers. atnumbersthat inorder to arrive the managers of the theyfirms cover. Hence, they might choose valuation heuristics analyses DCF-based las compared to the intextbooks. taught require fewvery price-to-sales and PEG, P∕E, involveand variables formu-simple fundamental valuation techniques taught in textbooks. Heuristics are simpler to use: questionsthese isyes. answer the eachto definitive,argumentsthat of be togood made be can difficult is it Although overlysimplistic? is definition that when flow cash free of definition 69 Moreover, analysts face agency conflicts. They need to manage relations with with relations manage to need They conflicts. agency face analystsMoreover, the of instead heuristics valuation on reliance the to contribute factors Several 69 Moreover, analysts face agency conflicts. They need to manage relations with with relations manage to need They conflicts. agency face analystsMoreover, the of instead heuristics valuation on reliance the to contribute factors Several price-to-sales heuristic,60 PEG heuristic,60 P/E heuristic,60 price-to-sales heuristic,60 PEG heuristic,60 P/E heuristic,60 valuation bycomparables, 59 valuation bycomparables, 59 2/7/17 5:26PM she77208_ch03_058-084.indd 79 2/7/17 5:26PM Questions Chapter

6. Consider the following excerpt from a Prudential report on Wal-Mart, dated May 13, May dated Wal-Mart, on report Prudential a from excerpt following the Consider 6. 5. of issue 2004, 13, February the in appeared that Tape” the of “Ahead column The 4. the in reflected were investments eBay’s that stated Dutta Rajiv CFO eBay 2005, In 3. DiscussthereasoningofPrudential analysts. 2. “Adjusted as EBITDA computed team Stanley Morgan the report, 2003 April its In 1. 7. 2003. havioral elements described in Chapter 1 might have affected Rashtchy’s long-term forecast? firm forecast that its revenue would increase by 58 percent in 2003. Which, if any, of the be the and percent, 62 of rate the at grown had revenue eBay’s year, previous the In 2010. and 2002 between rate growth annual compounded percent 30 a about be would growth annual Analyst Safa Rashtchy developed his 2010 forecast for eBay’s revenue by assuming that its expenditure. Discussthesecomments. cash free flow is of cash flow definition from operations minus typical capital expenditure, the not EBITDA that minus capital explains article The EBITDA. of instead flow cash free using firms value to prefer investors prudent that states Journal Street Wall of CapExwereaccurateinhindsight. Then discuss whether both Dutta’s perspective and the Morgan Stanley team’s forecasts perspective, in foresight, was reflected in the Morgan Stanley team’s April 2003 report. Dutta’s which to extent the Discuss sheet. balance its than rather statement income firm’s tions ofaP/E75tomultiply2003earnings,and502004earnings assump their that stated report Their PEG. to appeal by analysis their in values P/E the justifying eBay, for price target a established Prudential at analysts 2003, of spring the In income beenusedinstead.Discusstherelevanceofapproachtakeninreport. operating reported had been have would it than higher is report the in used EBITDA of value the result, a As statement. income the in down further income operating reported of components as position their from items line moves adjustment the is, That items.” removes income these items operating from operating expenses Adjusted and treats charges.’ them directly back as ‘below-the-line’ unusual ‘add with these adjusts then and cludes unusual items such as amortization of stock compensation in operating expenses in- “eBay follows: as adjustment the explained report The Income.” Operating ported “Re- to is question in adjustment The Addback.” Depreciation plus Income Operating PEG, and price-to-sales. Analyze the degree to which these assumptions are mutually mutually are assumptions these which to degree the Analyze price-to-sales. and PEG, P/E, with associated valuations the underlie that 3-2 Exhibit in assumptions the Consider to thecontentsofchapter. reference with excerpt, this in mentioned technique valuation the of merits the Discuss valuation target andour2003EPSestimateprice of$2.01yieldsa12-month of$56,upfrom $55. stock will continue to hover around a 55% the premium to the believemarket We multiple or 27.9 times. growth. Using this projected 13% for particularly premium, larger even an paying investors envision to us for difficult is It 2002. Mayachieved in $59.30, of high 52-week its to close also is stock The 59%. of premium averageretailer’sfive-yearhigh the from far not is This 500. S&P to rate . . growth EPS 5-yearcompany’s 13% the reflects adequately $2.01 of estimate EPS 2003 our times 28 stock’sbelieveof wevaluationWal-Mart the as on current rating Hold our We maintaining are 16.5 times forward-year earnings, equating to a P/E/G ratio of 1.3. with those of other growth companies. We note that the S&P 500 currently trades at approximately . equates to P/E/G ratios of 1.5 and 1.0, respectively, which we believe are reasonable, compared 16 Thereportstates: . Wal-Mart is currently trading at 28.2 times our 2003 EPS estimate of $2.01, a 57% premium Valuation The 79 - -

2/7/17 5:26PM www.mhhe.com/shefrin2e 2/7/17 5:26 PM

2/7/17 5:26 PM

- - 79 79 The The Valuation Valuation . Wal-Mart is currently trading at 28.2 times our 2003 EPS estimate of $2.01, a 57% premium 57% a $2.01, of estimate EPS 2003 our times 28.2 at trading currently is Wal-Mart . The report states: 16 . Wal-Mart is currently trading at 28.2 times our 2003 EPS estimate of $2.01, a 57% premium 57% a $2.01, of estimate EPS 2003 our times 28.2 at trading currently is Wal-Mart . The report states: We are maintaining We our Hold rating current on as the Wal-Mart we valuation of believe stock’s 28 times our 2003 EPS estimate of $2.01 adequately reflects the 13% company’s 5-year EPS growth . . rate to S&P 500. This is not far from the high five-year retailer’s average premium of 59%. The stock is also close to its 52-week high of $59.30, in achieved May 2002. It is difficult for us to envision investors paying an even larger premium, particularly for 13% projected this Using times. 27.9 growth. or multiple market We the to believe premium 55% thea around hover to continue will stock $55. of $56, up from of $2.01 yields a 12-month and our 2003 EPS estimate price target valuation . . . equates to P/E/G ratios of 1.5 and 1.0, respectively, which we believe are reasonable, compared compared reasonable, are believe we which respectively, 1.0, and 1.5 of ratios P/E/G to . . . equates approximately at trades currently 500 S&P the that note We companies. growth other of those with 1.3. of ratio P/E/G a to equating earnings, forward-year times 16.5 16 Discuss the merits of the valuation technique mentioned in this excerpt, with reference to the contents of the chapter. Consider the assumptions in Exhibit 3-2 that underlie the valuations associated with P/E, PEG, and price-to-sales. Analyze the degree to which these assumptions are mutually perspective, in foresight, was reflected in the Morgan Stanley team’s April 2003 report. 2003 April team’s Stanley Morgan the in reflected was foresight, in perspective, forecasts team’s Stanley Morgan the and perspective Dutta’s both whether discuss Then of CapEx were accurate in hindsight. Wall Street Journal states that prudent investors prefer to value firms using free cash flow instead of EBITDA. The article explains capital minus that EBITDA not expenditure, the capital minus typical operations from flow definition cash is of flow free cash expenditure. Discuss these comments. its that assuming by revenue eBay’s for forecast 2010 his developed Rashtchy Safa Analyst annual growth would be about a 30 percent compounded annual growth rate between 2002 and 2010. In the previous year, eBay’s revenue had grown at the rate of 62 percent, and the be the of any, if Which, 2003. in percent 58 by increase would revenue its that forecast firm forecast? long-term Rashtchy’s affected have might 1 Chapter in described elements havioral 2003. firm’s income statement rather than its balance sheet. Discuss the extent to which Dutta’s Operating Income plus Depreciation Addback.” The adjustment in question is to “Re- ported Operating Income.” The report explained the adjustment as follows: “eBay in- expenses operating in compensation stock of amortization as such items unusual cludes and then adjusts these with unusual ‘add ‘below-the-line’ as back directly them treats charges.’ and expenses Adjusted operating from operating items these income removes items.” That is, the adjustment moves line items from their position as components of reported operating income further down in the income statement. As a result, the value of EBITDA used in the report is higher than it would have been had reported operating taken in the report. Discuss the relevance of the approach income been used instead. In the spring of 2003, analysts at Prudential established a target price for eBay, justifying the P/E values in their analysis by appeal to PEG. Their report stated that their assump- and a P/E of 50 to multiply 2004 earnings tions of a P/E of 75 to multiply 2003 earnings, We are maintaining We our Hold rating current on as the Wal-Mart we valuation of believe stock’s 28 times our 2003 EPS estimate of $2.01 adequately reflects the 13% company’s 5-year EPS growth . . rate to S&P 500. This is not far from the high five-year retailer’s average premium of 59%. The stock is also close to its 52-week high of $59.30, in achieved May 2002. It is difficult for us to envision investors paying an even larger premium, particularly for 13% projected this Using growth. times. 27.9 or multiple We market believe the to premium the 55% a around hover to continue will stock $55. of $56, up from of $2.01 yields a 12-month price and our 2003 EPS estimate target valuation . . . equates to P/E/G ratios of 1.5 and 1.0, respectively, which we believe are reasonable, compared compared reasonable, are believe we which respectively, 1.0, and 1.5 of ratios P/E/G to . . . equates approximately at trades currently 500 S&P the that note We companies. growth other of those with 1.3. of ratio P/E/G a to equating earnings, forward-year times 16.5 7. 4. The column “Ahead of the Tape” that appeared in the February 13, 2004, issue of 5. 6. Consider the following excerpt from a Prudential report on Wal-Mart, dated May 13, Discuss the reasoning of the Prudential analysts. 3. In 2005, eBay CFO Rajiv Dutta stated that eBay’s investments were reflected in the 1. In its April 2003 report, the Morgan Stanley team computed EBITDA as “Adjusted 2. Discuss the merits of the valuation technique mentioned in this excerpt, with reference to the contents of the chapter. Consider the assumptions in Exhibit 3-2 that underlie the valuations associated with P/E, PEG, and price-to-sales. Analyze the degree to which these assumptions are mutually firm’s income statement rather than its balance sheet. Discuss the extent to which Dutta’s Operating Income plus Depreciation Addback.” The adjustment in question is to “Re- ported Operating Income.” The report explained the adjustment as follows: “eBay in- expenses operating in compensation stock of amortization as such items unusual cludes and then adjusts these with ‘add unusual ‘below-the-line’ as back directly them charges.’ treats and Adjusted expenses operating from operating items these income removes items.” That is, the adjustment moves line items from their position as components of reported operating income further down in the income statement. As a result, the value of EBITDA used in the report is higher than it would have been had reported operating taken in the report. Discuss the relevance of the approach income been used instead. In the spring of 2003, analysts at Prudential established a target price for eBay, justifying the P/E values in their analysis by appeal to PEG. Their report stated that their assump- multiply 2004 earnings multiply 2003 earnings, and a P/E of 50 to tions of a P/E of 75 to report. 2003 April team’s Stanley Morgan the in reflected was foresight, in perspective, forecasts team’s Stanley Morgan the and perspective Dutta’s both whether discuss Then of CapEx were accurate in hindsight. Wall Street Journal states that prudent investors prefer to value firms using free cash flow instead of EBITDA. The article explains capital minus that EBITDA not the expenditure, capital typical minus operations from definition flow cash of is flow free cash expenditure. Discuss these comments. itsthat assuming by revenue eBay’s for forecast 2010 his developed Rashtchy Safa Analyst annual growth would be about a 30 percent compounded annual growth rate between 2002 and 2010. In the previous year, eBay’s revenue had grown at the rate of 62 percent, and the be the of any, if Which, 2003. in percent 58 by increase would revenue its that forecast firm forecast? long-term Rashtchy’s affected have might 1 Chapter in described elements havioral 2003. 7. reasoning of the Prudential analysts. Discuss the 3. In 2005, eBay CFO Rajiv Dutta stated that eBay’s investments were reflected in the 1. In its April 2003 report, the Morgan Stanley team computed EBITDA as “Adjusted 2. 4. The column “Ahead of the Tape” that appeared in the February 13, 2004, issue of 5. 6. Consider the following excerpt from a Prudential report on Wal-Mart, dated May 13,

Chapter Questions

Chapter Chapter Questions

she77208_ch03_058-084.indd 79 2/7/17 5:26 PM she77208_ch03_058-084.indd 79 59 valuation by comparables, P/E heuristic, 60 PEG heuristic, 60 price-to-sales heuristic, 60 Several Several factors contribute to the reliance on valuation heuristics instead of the Moreover, analysts face agency conflicts. They need to manage relations with 69 definition of free cash flow when that definition is simplistic? overly Although it is difficult can be made good to be of thatarguments definitive, to each the answer is yes. these questions use: to simpler are Heuristics textbooks. in taught techniques valuation fundamental simple formu- variables and involve P∕E, PEG, and price-to-sales very few require taught in textbooks. the to las compared DCF-based analyses heuristics valuation choose might they Hence, cover. firmsthey the of managers the arrive to in order that at numbers will please managers. strategically, 1/n heuristic, 71 growth opportunities bias, www.ebay.com as a The web site for eBay contains financial information about the firm, as well as serving gateway to the firm’s transaction services. www.morganstanley.com The web site for Morgan Stanley has an interesting section where institutional investors can view sample reports. www.aetna.com The web site for Aetna has several sections of interest, especially Who We Are and Investors. Boston: Harvard Business School Value Based Management. Boston: Harvard Business Martin, J. and J. W. Petty, Press, 2000. Montier, J., Behavioural Finance. West Sussex, UK: Wiley, 2002. Boston: Harvard Business Rappaport, A. and M. Mauboussin, Expectations Investing. School Press, 2002. . . eBay, But How Can Shares Stay Lofty? Wingfield, N., “Stock of eBay Acts Like . Rise in Online Shopping,” The Wall Street The Bullish Projections Assume Gigantic Journal, April 23, 2003. Technology Conference − Final,” FD “eBay at Bear, Stearns & Co. Inc. 16th Annual (FAIR DISCLOSURE) WIRE, June 8, 2005. − Final,” CQ FD Disclosure, September 4,“eBay Inc at Citi Global Technology Conference 2014. November 17, 2015. Bhatia, S., “The History of eBay,” The Telegraph Online, Key Terms Explore the Web 78 Chapter Three Additional Behavioral Readings she77208_ch03_058-084.indd 78 she77208_ch03_058-084.indd 80 she77208_ch03_058-084.indd 80 www.mhhe.com/shefrin2e ChapterThree 80 ChapterThree 80 13. 12. 11. 10. 9. geometric annualized eBay’s eBay, on report 2003 Stanley Morgan the to respect In 8. 13. 12. 11. 10. 9. geometric annualized eBay’s eBay, on report 2003 Stanley Morgan the to respect In 8.

price target of $29 per share. per $29 of target forward one-year price a established 2015, 23, July dated eBay, on report Stanley Morgan The Then discussthemethodology MorganStanleyteamusedtoarriveattheirforecasts. growth rate of 1 percent. Check to see if the free cash flow computations were done correctly. terminal a used and WACC, its as described percent, at8 flows cash free eBay’s discounted report the that shows and site, web book the at provided is case base the for spreadsheet A percent). The report used $24 to $38 as its interval forecast range and $29 as its point forecast. percent), and a bear case involving a price per share of $24 (and implicit target return of −16 34 of return target implicit (and share per $38 of price a involving case bull a $34, of share cash flow analysis in the report contained three cases, a base case involving a target price per free discounted The percent. 2 of appreciation price target a implied noted report the which over timeinvaluationmethodology? changes these interpret you would How check.” “sanity a and “support” free as flows of cash values discounted on rely to claimed and P/E, on primarily targets based price they their 2015, In they flows. 2013, cash free In discounted used flows. only cash and P/S, free and P/E of dropped values discounted and P/S, P/E, on relying eBay, value to PEG using stopped had eBay covering analysts Stanley Morgan the 2010, In nature ofthisforecasterrorfromabehavioralperspective. the Discuss period. this during percent 16.7 of rate growth average an by grew actually flow cash Free percent. 5.7 by grow average eBay on would on flow cash report free 2010 that forecasted team’s Stanley Morgan the 2014, through 2010 period the For eBay maintainsaconstant52percentratioofdepreciationtoCapEx. to CapEx, and that this ratio is 52 percent. 52 is ratio this that and CapEx, to For this purpose, assume that over time, eBay maintains a constant ratio of depreciation zero. being PVGO with consistent is tables these in assumed rates growth terminal the made in 2010 and 2013. Use the analysis described in the chapter to investigate whether valuations and forecasts for but 3-3, Exhibit in table flow cash free the to counterparts In the section of the book web site for this chapter, you will find an Excel file containing and 2010. and 2002 between rate compounded percent 30 a at grow could eBay that assumption his ing reconsider- by reacted Rashtchy that reported Wingfield optimistic. unduly were eBay for forecasts growth revenue percent 30 his whether consider to him asked and Rashtchy Safa interviewed Wingfield, Nick article, the of author The report. Jaffray Piper Bancorp U.S. cle pointed out that the second most optimistic analyst was Safa Rashtchy, the author of the arti- The stream. revenue future eBay’s about optimistic excessively were eBay following On April 23, 2003, of a12percentdiscountrate? assumption team’s Stanley Morgan the judge you would how information, this of light return was 14.2 percent, and its annualized arithmetic risk premium was 12.7 percent. In return between May 2003 and December 2014 was 7.9 percent, its annualized arithmetic addition, discusswhattomakeofthewidedispersioninthreeassociatedvaluations. In applied. properly were equations heuristic the whether correspondingly and consistent, his pricetargetof$101pershare.DiscussRashtchy’sjudgmentsabouteBay. maintained but percent, 5.3 by 2005 for estimate earnings his lowered also He perfection. on eBay from “outperform” to “market perform,” stating that the stock had been priced for fell by 20 percent to $81 per share. Thereupon, Rashtchy downgraded his recommendation price stock eBay’s announcement, the to response In share. a penny a by forecast earnings

price target of $29 per share. per $29 of target forward price one-year a established 2015, 23, July dated eBay, on report Stanley Morgan The Then discussthemethodologyMorganStanleyteamusedtoarrive attheirforecasts. growth rate of 1 percent. Check to see if the free cash flow computations were done correctly. terminal a used and WACC, its as described percent, at8 flows cash free eBay’s discounted report the that shows and site, web book the at provided is case base the for spreadsheet A percent). The report used $24 to $38 as its interval forecast range and $29 as its point forecast. percent), and a bear case involving a price per share of $24 (and implicit target return of −16 34 of return target implicit (and share per $38 of price a involving case bull a $34, of share cash flow analysis in the report contained three cases, a base case involving a target price per free discounted The percent. 2 of appreciation price target a implied noted report the which over timeinvaluationmethodology? changes these interpret you would How check.” “sanity a and “support” as free flows cash of values discounted on rely to claimed and P/E, on primarily targets based price they their 2015, In they flows. cash 2013, free In discounted used flows. only and cash P/S, free and P/E of dropped values discounted and P/S, P/E, on relying eBay, value to PEG using stopped had eBay covering analysts Stanley Morgan the 2010, In nature ofthisforecasterrorfromabehavioralperspective. the Discuss period. this during percent 16.7 of rate growth average an by grew actually flow cash Free percent. 5.7 by grow average on eBay would on flow cash report free that 2010 forecasted team’s Stanley Morgan the 2014, through 2010 period the For eBay maintainsaconstant52percentratioofdepreciationtoCapEx. to CapEx, and that this ratio is 52 percent. 52 is ratio this that and CapEx, to For this purpose, assume that over time, eBay maintains a constant ratio of depreciation zero. being PVGO with consistent is tables these in assumed rates growth terminal the made in 2010 and 2013. Use the analysis described in the chapter to investigate whether valuations and forecasts for but 3-3, Exhibit in table flow cash free the to counterparts In the section of the book web site for this chapter, you will find an Excel file containing and 2010. and 2002 between rate compounded percent 30 a at grow could eBay that assumption his ing reconsider- by reacted Rashtchy that reported Wingfield optimistic. unduly were eBay for forecasts growth revenue percent 30 his whether consider to him asked and Rashtchy Safa interviewed Wingfield, Nick article, the of author The report. Jaffray Piper Bancorp U.S. cle pointed out that the second most optimistic analyst was Safa Rashtchy, the author of the arti- The stream. revenue future eBay’s about optimistic excessively were eBay following On April 23, 2003, of a12percentdiscountrate? assumption team’s Stanley Morgan the judge you would how information, this of light return was 14.2 percent, and its annualized arithmetic risk premium was 12.7 percent. In return between May 2003 and December 2014 was 7.9 percent, its annualized arithmetic addition, discusswhattomakeofthewidedispersioninthreeassociatedvaluations. In applied. properly were equations heuristic the whether correspondingly and consistent, his pricetargetof$101pershare.DiscussRashtchy’sjudgmentsabouteBay. maintained but percent, 5.3 by 2005 for estimate earnings his lowered also He perfection. on eBay from “outperform” to “market perform,” stating that the stock had been priced for fell by 20 percent to $81 per share. Thereupon, Rashtchy downgraded his recommendation price stock eBay’s announcement, the to response In share. a penny a by forecast earnings 17 On January 20, 2005, eBay announced that it had missed analysts’ consensus consensus analysts’ missed had it that announced eBay 2005, 20, January On 17 On January 20, 2005, eBay announced that it had missed analysts’ consensus consensus analysts’ missed had it that announced eBay 2005, 20, January On The Wall Street Journal The Wall Street Journal 19 At the time of the report, eBay’s stock price was $28.45, $28.45, was price stock eBay’s report, the of time the At 19 At the time of the report, eBay’s stock price was $28.45, $28.45, was price stock eBay’s report, the of time the At published an article suggesting that the analysts 18 published an article suggesting that the analysts For this purpose, assume that over time, over that assume purpose, this For 18 For this purpose, assume that over time, over that assume purpose, this For 2/7/17 5:26PM she77208_ch03_058-084.indd 81 2/7/17 5:26PM was 82.5 percent, compared to 82.4 percent 82.4 to compared percent, 82.5 wasratio, benefit used to pay patient medical costs, known as its WellPoint. For the quarter, the amount and of premiums AetnaGroup UnitedHealth as such competitors as well as business, insurance Aetna’scommercial in costs strained re- turn in which expenditures, care health their reduce to consumers led had recession great and crisis financial times. three forecast the raised already had but share, per $5.40 least wouldforecast at be adjustedearnings that firm the 2013, of beginning the At share. per $5.90 to $5.80 of range a to estimate earnings full-yearoperating its increased firm the result, a quarter.previousAs the in million 18.3 from 30, June of as million 22 to grown had membership cal medi- Total revenue. and membership in growth its ing reflect- percent, 17 by risen had earnings second-quarter mate to fall range inthe of$5.70to $5.85ashare. esti- earnings operating full-yearforwas guidance the firm’s time, the At members. D Part Medicare million 1.5 and members medical Aetna million 3.7 so, approximately added doing By care. health government-financed in presence its strengthen to so did It billion. $5.7 worth stockdeal and cash a in Inc. Care Health CoventryMCO ment services. manage- care health Medicaid as well as and plans, disability life, group pharmacy, dental, medical, include that firm offers a wide range of health care insurance products The country. every almost in residing members million a half approximately support benefits care health global Its company. global a is Aetna expatriates. and indi- viduals, groups, labor plans, government-sponsored units, employer governmental providers, care health plans, of health groups, composed being diverse, are members Its Ameri- membership. by third-largest (MCO) organization care the managed can it making people, million 44 approximately serves It firm. benefits care health diver- sified large a is Aetna Connecticut, Hartford, in Based Aetna Minicase www.mhhe.com/shefrin2e Additional ResourcesandMaterialsforChapter3AreAvailableat The weak economic conditions that followed the global its that announced Aetna 2013, July of end the At the of acquisition the completed Aetna 2013, May In total medical to push the numberdownward,to pushthe andnot up.” “workingwere projections financial the that stated He ing. operat exchangesbegin first the would when 2014 in pate partici might Aetna where states of number the for meant caution that what explained Guertin Shawn CFO Aetna’s will “I tell you we analysts:remain very cautious if not more cautious.”to said Bertolini regard, this In states. run the by were which of many exchanges, the in participating of profitability the affectwould which rates, and readiness Bertolini’s caution stemmed from concerns about exchange will 2014.” in EPS we operating and earnings operating both grow that when confident optimism “increasingly being his described he expressed Bertolini analysts, with call conference July a In Obamacare. so-called Act, Care Health Affordable and Protection Patient the by required new when 2014, exchangescare health as operating, werebegin expected to for prospects about optimistic tiously July 31. Even before its second quarter earnings announce JulyEven31. earnings before quarter second its on $64.17 at closed it and percent, 38 by increased stock men was as earnings, operating full-year for guidance its changeto previousfirm year, the the $1.31 whichled from a year earlier. Operating earnings increased to $1.52 a share $1.32 or million $457.6 from up million, $536 of earnings to corresponding share, a $1.49 of EPS reported Aetna but percent, 31 of increase an billion, $11.54 be to out turned (EPS) of $1.41 on revenue of $11.99 billion. Actual revenue share per earnings forwasReuters Thomson forecastfrom noting that “Aetna should have the fastest earnings growth earnings fastest“Aetna have the that should noting $72, to $68 from target price its increased Citigroup ment, announce the after Immediately $77. to price $74 his from target raised analyst Jefferies a July, in Earlier up ward. targets price their revising begun had analysts ment, tioned above. previous patterns of health-care usage.previous ofhealth-care patterns their resume to consumers induce would economy U.S. improving an awarenesswas that there time, same the At a year earlier and 81.9 percent during the previous quarter. From the beginning of 2013 until the end of July, Aetna’s earnings analyst consensus the quarter second its For Aetna’s CEO Mark Bertolini indicated that he wascau he that indicated Aetna’s Bertolini Mark CEO 20 Valuation 81 ------2/7/17 5:26PM www.mhhe.com/shefrin2e 2/7/17 5:26 PM 2/7/17 5:26 PM ------81 81 Valuation Valuation 20 20 Aetna’s CEO Mark Bertolini Aetna’s indicated that he cau was For For its second quarter the consensus analyst earnings Aetna’s July, of end the until 2013 of beginning the From Aetna’s CEO Mark Bertolini Aetna’s indicated that he cau was From the beginning of 2013 until the end of July, Aetna’s Aetna’s July, of end the until 2013 of beginning the From For For its second quarter the consensus analyst earnings a year earlier and 81.9 percent during the previous quarter. quarter. previous the during percent 81.9 and earlier year a At the same time, there that was awareness an improving U.S. economy would induce consumers to resume their patternsprevious of health-care usage. forecast from forecast Thomson Reuters was for earnings per share revenue Actual billion. $11.99 of revenue on $1.41 of (EPS) turned out to be $11.54 billion, an increase of 31 percent, but Aetna reported EPS of $1.49 a share, corresponding to earnings of $536 million, up from $457.6 million or $1.32 share a $1.52 to increased earnings Operating earlier. year a from led which $1.31 thethe year, firmprevious to change its guidance for full-year operating earnings, as was men tioned above. stock increased by 38 percent, and it closed at $64.17 on its second quarter before earnings31. Even July announce ment, analysts had begun revising their price targets ward. up Earlier in July, a Jefferies analyst target raised from his $74 to price $77. Immediately after the announce ment, Citigroup increased its price target from $68 to $72, noting should that the have “Aetna fastest earnings growth tiously tiously optimistic about prospects for to expected begin were operating, as health care exchanges 2014, when new required by the Patient Protection and Affordable Health Care Act, so-called Obamacare. In a July conference call with analysts, Bertolini expressed he described his being “increasingly optimism confident when that grow both operating earnings and operating EPS we in 2014.” will exchange about concerns from stemmed caution Bertolini’s readiness and rates, which would affect the profitability of participating in the exchanges, many of which by were the run states. In this regard, Bertolini said cautious.” to more not if cautious very remain we analysts: you tell will “I Aetna’s CFO Shawn Guertin explained what that caution meant for the number of states where Aetna might partici pate in 2014 when would thefirst exchanges begin operat ing. He stated that the financial projections were “working up.” and not push theto number downward, a year earlier and 81.9 percent during the previous quarter. quarter. previous the during percent 81.9 and earlier year a At the same time, there that was awareness an improving U.S. economy would induce consumers to resume their patterns previous of health-care usage. tioned above. stock increased by 38 percent, and it closed at $64.17 on its second quarter before earnings31. Even July announce ment, analysts had begun revising their price targets ward. up Earlier in July, a Jefferies analyst raised target from his $74 price to $77. Immediately after the announce ment, Citigroup increased its price target from $68 to $72, noting should that the have “Aetna fastest earnings growth forecast from forecast Thomson Reuters was for earnings per share revenue Actual billion. $11.99 of revenue on $1.41 of (EPS) turned out to be $11.54 billion, an increase of 31 percent, but Aetna reported EPS of $1.49 a share, corresponding to earnings of $536 million, up from $457.6 million or $1.32 share a $1.52 to increased earnings Operating earlier. year a from led which $1.31 thethe year, firmprevious to change its guidance for full-year operating earnings, as was men tiously tiously optimistic about prospects for to expected begin were operating, as health care exchanges 2014, when new required by the Patient Protection and Affordable Health Care Act, so-called Obamacare. In a July conference call with analysts, Bertolini expressed he described his being “increasingly optimism confident when that grow both operating earnings and operating we EPS in 2014.” will exchange about concerns from stemmed caution Bertolini’s readiness and rates, which would affect the profitability of participating in the exchanges, many of which were by the run states. In this regard, Bertolini said to cautious.” more not if cautious very remain analysts: we you tell “Iwill Aetna’s CFO Shawn Guertin explained what that caution meant for the number of states where Aetna might partici pate in 2014 when would thefirst exchanges begin operat ing. He stated that the financial projections were “working up.” and not push theto number downward, total medical total total medical total At At the end of July 2013, Aetna announced that its global the followed that conditions economic weak The In May 2013, Aetna completed the acquisition of the

Additional Resources and Materials for Chapter 3 Are Available at Chapter 3 Are Available at and Materials for Additional Resources www.mhhe.com/shefrin2e Minicase Aetna

second-quarter earnings had risen by 17 percent, reflect- ing its growth in membership and revenue. Total medi- cal membership had grown to 22 million as of June 30, from 18.3 million in the As previous quarter. a result, the firm increased its operating full-year earnings estimate to a range of $5.80 to $5.90 per share. At the beginning of 2013, the firmthat earningsadjusted be at forecast would least $5.40 per share, but had already raised the forecast three times. financial crisis and great recession had led consumers to reduce their health care expenditures, which in turn strained re- costs in commercial Aetna’s insurance business, as well as competitors such as UnitedHealth Aetna Group premiums of amount and the quarter, the For WellPoint. its as known costs, medical patient pay to used MCO MCO Coventry Health Care Inc. in a cash and deal stock worth $5.7 billion. It did so to strengthen its presence in government-financed health care. By added doing approximately so, 3.7 million Aetna medical members and 1.5 million Medicare Part D members. At the time, firm’s guidance the was for full-year operating earnings esti- $5.85 a share. of $5.70 to in the range fall mate to benefit ratio, was 82.5 percent, compared to 82.4 percent Based in Hartford, Connecticut, Aetna is a large sified diver- health care benefits firm. It serves approximately 44 million people, making it can managed care the organization (MCO) third-largest by membership. Ameri- Its members are diverse, being composed groups, health plans, health of care providers, governmental employer units, government-sponsored plans, labor groups, viduals, indi- and expatriates. Aetna is a global company. Its global health care benefits support approximately half a million members residing in almost every country. The products insurance care health of range wide a offers firm that include medical, dental, pharmacy, group life, disability plans, as and well as Medicaid health care manage- ment services.

In May 2013, Aetna completed the acquisition of the At the end of July 2013, Aetna announced that its global the followed that conditions economic weak The Additional Resources and Materials for Chapter 3 Are Available at 3 Are Available for Chapter and Materials Resources Additional www.mhhe.com/shefrin2e Aetna Minicase

Based in Hartford, Connecticut, Aetna is a large sified diver- health care benefits firm. It serves approximately 44 million people, making it can managed the care organization (MCO) third-largest by membership. Ameri- Its members are diverse, being composed groups, health of plans, health care providers, governmental employer units, government-sponsored plans, labor groups, viduals, indi- and expatriates. Aetna is a global company. Its global health care benefits support approximately half a million members residing in almost every country. The products insurance care health of range wide a offers firm that include medical, dental, pharmacy, group life, disability plans, and as well as Medicaid health care manage- ment services. MCO Coventry Health Care Inc. in a cash and deal stock worth $5.7 billion. It did so to strengthen its presence in government-financed health care. By doing added approximately so, 3.7 million Aetna medical members and 1.5 million Medicare Part D members. At the time, firm’s the guidance was for full-year operating earnings esti- a share. $5.85 of $5.70 to in the range fall mate to second-quarter earnings had risen by 17 percent, reflect- ing its growth in membership and revenue. Total medi- cal membership had grown to 22 million as of June 30, from 18.3 million in the As previous quarter. a result, the firm increased its operating full-year earnings estimate to a range of $5.80 to $5.90 per share. At the beginning of 2013, the firmthat earningsadjusted be at forecast would least $5.40 per share, but had already raised the forecast three times. financial crisis and great recession had led consumers to reduce their health care expenditures, which in turn re- strained costs in commercial Aetna’s insurance business, as well as competitors such as UnitedHealth Group Aetna premiums of and amount the quarter, the For WellPoint. its as known costs, medical patient pay to used was 82.5 percent, compared to 82.4 percent benefit ratio, was 82.5 percent, compared to 82.4 percent

she77208_ch03_058-084.indd 81 2/7/17 5:26 PM she77208_ch03_058-084.indd 81 For this purpose, assume that over time, 18 published an article suggesting that the analysts the that suggesting article an published At the time of the report, eBay’s stock price was $28.45, 19 The Wall Street Journal Street Wall The On January 20, 2005, eBay announced that it had missed analysts’ consensus 17 earnings forecast by a penny a share. In response to the announcement, eBay’s stock price recommendation his downgraded Rashtchy Thereupon, share. per $81 to percent 20 by fell forpriced been had stock the that stating perform,” “market to “outperform” from eBay on perfection. He also lowered his earnings estimate for 2005 by 5.3 percent, but maintained judgments about eBay. his price target of $101 per share. Discuss Rashtchy’s consistent, and correspondingly whether the heuristic equations were properly applied. In valuations. in the three associated the wide dispersion what to make of addition, discuss arithmetic annualized its percent, 7.9 was 2014 December and 2003 May between return In percent. 12.7 was premium risk arithmetic annualized its and percent, 14.2 was return light of this information, how would you judge the Morgan Stanley team’s assumption rate? of a 12 percent discount 2003, 23, April On following eBay were excessively optimistic about eBay’s future revenue stream. The arti- the of author the Rashtchy, Safa was analyst optimistic most second the that out pointed cle U.S. Bancorp Piper Jaffray report. The author of the article, Nick Wingfield, interviewed Safa Rashtchy and asked him to consider whether his 30 percent revenue growth forecasts for eBay were unduly optimistic. Wingfield reported that Rashtchy reacted by reconsider- ing his assumption that eBay could grow at a 30 percent compounded rate between 2002 and 2010. In the section of the book web site for this chapter, you will find an Excel file containing file Excel an find will you chapter, this for site web book the of section the In counterparts to the free cash flow table in Exhibit 3-3, but for forecasts and valuations whether investigate to chapter the in described analysis the Use 2013. and 2010 in made the terminal growth rates assumed in these tables is consistent with PVGO being zero. depreciation of ratio constant a maintains eBay time, over that assume purpose, this For to CapEx, and that this ratio is 52 percent. eBay maintains a constant 52 percent ratio of depreciation to CapEx. eBay maintains a constant 52 percent ratio For the period 2010 through 2014, the Morgan Stanley team’s forecasted 2010 that free report cash flow on would eBay on average grow by 5.7 percent. Free cash flow actually grew by an average growth rate of 16.7 percent during this period. Discuss the perspective. nature of this forecast error from a behavioral In 2010, the Morgan Stanley analysts covering eBay had stopped using PEG to value eBay, relying on P/E, P/S, and discounted values dropped of P/E and free P/S, cash and only flows. used discounted In free 2013, cash flows. they In 2015, their they price based targets primarily on P/E, and claimed to rely on discounted values of cash flows free as “support” and a “sanity check.” How would you interpret these changes over time in valuation methodology? which the report noted implied a target price appreciation of 2 percent. The discounted free perprice target a involving case base a cases, three contained report the in analysis flow cash share of $34, a bull case involving a price of $38 per share (and implicit target return of 34 −16of return target implicit (and $24 of share per price a involving case bear a and percent), forecast. point its as $29 and range forecast interval its as $38 to $24 used report The percent). A spreadsheet for the base case is provided at the book web site, and shows that the report discounted eBay’s free cash flows at8 percent, described as its WACC, and used a terminal correctly. done were computations flow cash free the if see to Check percent. 1 of rate growth Then discuss the methodology the Morgan Stanley team used to arrive at their forecasts. The Morgan Stanley report on eBay, dated July 23, 2015, established a one-year price forward target of $29 per share.

8. In respect to the Morgan Stanley 2003 report on eBay, eBay’s annualized geometric 9. 10. 11. 12. 13. 80 Chapter Three she77208_ch03_058-084.indd 80 she77208_ch03_058-084.indd 82 she77208_ch03_058-084.indd 82 www.mhhe.com/shefrin2e tc ws rdn a $17, efre sae te following, noting the that it was stated maintaining its price target of Jefferies $77 from July: $61.78, at trading was stock ated methodologies. Below are four examples. ticker is AET, were varied in their price targets and associ- ber closeafew days laterwas at$62.70. stock by price 1.9 percent to $60.63, stock’sbut the Octo- its in decline a was announcement earnings third-quarter $5.80 and $5.90 per share. The market response to Aetna’s betweenrangewould earnings adjusted 2013 its that ated reiter- firm The quarter. the in percent 57 increased had expense, largest its were which costs, care Health MA). year had hurt its Medicare business (Medicare Advantage, that cuts budgetfederal and products two from formance down third-quarter earnings. down earnings. third-quarter talked had firm the call, conference previous its in even that commented Aetna covering analyst Citigroup lead the prise, sur earnings negative the to respect In share. per $1.50 was that excluding items such as integration costs, its adjusted EPS or $1.38 per share on revenue of $13.04 billion. The firm noted third- million, $518.6 to bypercent increased 4 had earnings quarter its that announced Aetna 28, billion October $12.87 On revenue. approximately in on share per $1.52 was EPS exchanges, target andincreased itsprice from $68to $79. care reform health associated with and risks discounted the overly valuation Aetna’s current that stated Lynch Merrill America/ of Bank mid-September, In share. a $65 to $60 from target price its increased Fitzgerald Cantor August of the lowest multiple in the group.” the in multiple lowest the among the large-cap plans in 2014/2015, and yet it trades at ChapterThree 82 tc ws rdn a $17, efre sae te following, noting the that it was maintaining its stated price target of $77 Jefferies from July: $61.78, at trading was stock ated methodologies. Below are four examples. ticker is AET, were varied in their price targets and associ- ber closeafew days laterwas at$62.70. stock by price 1.9 percent to $60.63, stock’sbut the Octo- its in decline a was announcement earnings third-quarter $5.80 and $5.90 per share. The market response to Aetna’s betweenrangewould earnings adjusted 2013 its that ated reiter- firm The quarter. the in percent 57 increased had expense, largest its were which costs, care Health MA). year had hurt its Medicare business (Medicare Advantage, that cuts budgetfederal and products two from formance down third-quarter earnings. down earnings. third-quarter talked had firm the call, conference previous its in even that commented Aetna covering analyst Citigroup lead the prise, sur earnings negative the to respect In share. per $1.50 was that excluding items such as integration costs, its adjusted EPS or $1.38 per share on revenue of $13.04 billion. The firm noted million, third- $518.6 to bypercent increased 4 had earnings quarter its that announced Aetna 28, billion October $12.87 On revenue. approximately in on share per $1.52 was EPS exchanges, target andincreased itsprice from $68to $79. care reform health associated with and risks discounted the overly valuation Aetna’s current that stated Lynch Merrill America/ of Bank mid-September, In share. a $65 to $60 from target price its increased Fitzgerald Cantor August of ChapterThree 82 the lowest multiple in the group.” the in multiple lowest the among the large-cap plans in 2014/2015, and yet it trades at n t rpr dtd coe 2, hn Aetna’s when 29, October dated report its In Jefferies: stock whose Aetna, covering analysts of reports The n t anucmn, en epand ht or per- poor that explained Aetna announcement, its In third-quarter Aetna’s for forecast analyst consensus The (FCF), EV/EBITDA, P/E, PE/growth, P/CF, P/FCF, premium flow cash free (CF), flow cash EPS, EBITDA, cash (DCF), flow discounted stream, of revenue analyses rate, growth to, risk, restricted market not are but include, may which 3)Exchanges,gration, 4)2014MAheadwinds. inte- Coventry 2) trend, cost in spike 1) Risks: 11.1x. torical averagepeer broaderbelowwhich is his- the its and 10.7x of EPS, 2014E our 9.8x at trades AET EPS. 2014E our 12.4x (9.8x/8.4x 2014-15EEPS). stock MCO cheapest the is which AET, on rating Buy our 12/12 Investor Day will more bring 2014 clarity. We reiterate AET’s visibility. improves and floor growth EPS modest 2014 implies that a to guided company the but uneventful, were results 3Q 2015. in 15%+ EPS earnings increase and year grow next modestly can AET believe to continue We n t rpr dtd coe 2, hn Aetna’s when 29, October dated report its In Jefferies: stock whose Aetna, covering analysts of reports The n t anucmn, en epand ht or per- poor that explained Aetna announcement, its In third-quarter Aetna’s for forecast analyst consensus The (FCF), EV/EBITDA, P/E, PE/growth, P/CF, P/FCF, premium flow cash free (CF), flow cash EPS, EBITDA, (DCF), cash flow discounted stream, revenue of rate, analyses growth to, risk, restricted market not are but include, may which 3)Exchanges,gration, 4)2014MAheadwinds. inte- Coventry 2) trend, cost in spike 1) Risks: 11.1x. torical averagepeer broaderbelowwhich is his- the its and 10.7x of EPS, 2014E our 9.8x at trades AET EPS. 2014E our 12.4x (9.8x/8.4x 2014-15EEPS). stock MCO cheapest the is which AET, on rating Buy our 12/12 Investor Day will more bring 2014 clarity. We reiterate AET’s visibility. improves and floor growth EPS modest implies 2014 that a to guided company the but uneventful, were results 3Q 2015. in 15%+ EPS increase earnings and year next grow modestly can AET believe to continue We h pie agt ae ae o svrl methodologies several on based are targets price The of multiple P/E a reflects PT $77 Our Valuation/Risks: h pie agt ae ae o svrl methodologies several on based are targets price The of multiple P/E a reflects PT $77 Our Valuation/Risks: 21 21 During the first week first the During During the first week first the During - - stated the following: the stated when Aetnawhen following: the stated and Aetna, for $67.28 of target price one-year a had ValuEngine $64.75, at trading was stockAetna’s following: the stated when Aetna’s stock was trading at Swann$65.08, Leerink stated the following: the stated when Aetnawhen following: the stated and Aetna, for $67.28 of target price one-year a had ValuEngine $64.75, at trading was stockAetna’s following: the stated when Aetna’s stock was trading at Swann$65.08, Leerink In its report dated October 29, October dated report its In Fitzgerald: Cantor In its report dated November 19, when 19, November dated report its In ValuEngine: 19, November dated report its In Swann: Leerink Coventry Healthcare. of acquisition 5/7/13 the and years senior recent in management in changes transformative potentially the despite valuation, in-line traditional its with consistent group, the with parity at roughly trades AET estimate, 10.3x EPS 2014 our of valuation current its At $6.00. of estimate EPS 2014 our 10.8x to equates price targethistory. Our trading ownstock’s the and plans commercial large other of that with earnings estimated 2014 on valuation current its ing whichindividual book, accountsfor 3%ofrevenue. its increase materially not will plans) bronze and catastrophic on exchange(focused its business saysthat and carriers tional na other than exposed less is Aetna but business, MA vidual indi the for outlook 2014 the about cautious is management and ratios, loss higher reporting is rated), (experienced group is which of half company’sbook, MedicareThe $5.80–$5.90. of year’sguidance this with flat least at be to EPS 2014 pects ex company the and line, in mostly was it miss, slight a was 3Q:13 speaking, strictly Although, rating. HOLD and target We are maintaining our 2013–14 estimates, earnings $65 price onequity andreturn (ROE)returns, over next the 12months. average group P/E, sum of the netparts, asset value, dividend (discount)/ premium EV/EBITDA, (discount)/average group actual earnings actual per earnings share (EPS) for the previous four quarters AET,on For worth. wethis base true its at everythingtraded and efficient perfectly were market stock the if today at ing Fair Value indicates what we overvalued.believe the stock30.32% should be AET trad- makes This $49.68. at trading be should AET that believe we 2013, Nov.19, of as data able VALUATION PRICE: MARKET FAIR of 2014+2015estimated EPSviews .(SeeExhibit3-8.) average the 10.5x of P/E a contemplates $72 Targetof Price month 12–18 Our catalyst. important most the upside ings that value company intrinsic ofthe points to the . . model DCF a with approach P/E-based the combine We estimates. EPS 2015 and 2014 our of average the on valuation approachto multiple-based relativeP/E and Weabsolute an use In its report dated October 29, October dated report its In Fitzgerald: Cantor In its report dated November 19, when 19, November dated report its In ValuEngine: 19, November dated report its In Swann: Leerink Coventry Healthcare. of acquisition 5/7/13 the and years senior recent in management in changes transformative potentially the despite valuation, in-line traditional its with consistent group, the with parity at roughly trades AET estimate, EPS 10.3x 2014 our of valuation current its At $6.00. of estimate EPS 2014 our 10.8x to equates price targethistory. Our trading ownstock’s the and plans commercial large other of that with earnings estimated 2014 on valuation current its ing whichindividual book, accountsfor 3%ofrevenue. its increase materially not will plans) bronze and catastrophic on exchange(focused its business saysthat and carriers tional na other than exposed less is Aetna but business, MA vidual indi the for outlook 2014 the about cautious is management and ratios, loss higher reporting is rated), (experienced group is which of half company’sbook, Medicare The $5.80–$5.90. of year’sguidance this with flat least at be to EPS 2014 pects ex company the and line, in mostly was it miss, slight a was 3Q:13 speaking, strictly Although, rating. HOLD and target We are maintaining our 2013–14 estimates, earnings $65 price returns, and return onequity andreturn (ROE)returns, over next the 12months. average group P/E, sum of the netparts, asset value, dividend (discount)/ premium EV/EBITDA, (discount)/average group actual earnings actual per earnings share (EPS) for the previous four quarters AET,on For worth. wethis base true its at everythingtraded and efficient perfectly were market stock the if today at ing Fair Value indicates what we believeovervalued. the stock 30.32% should be trad- AET makes This $49.68. at trading be should AET that believe we 2013, Nov.19, of as data able VALUATION PRICE: MARKET FAIR of 2014+2015estimated EPSviews .(SeeExhibit3-8.) average the 10.5x of P/E a contemplates $72 Targetof Price month 12–18 Our catalyst. important most the upside ings value company that intrinsic ofthe points to the . model DCF a with approach P/E-based the combine We estimates. EPS 2015 and 2014 our of average the on valuation approachto multiple-based relativeP/E and Weabsolute an use We arrive at our $65 price target for AET by compar by AET for target price $65 our at arrive We Valuation has pulled back to 10x P/E with 2014E earn- 2014E with P/E 10x to back pulled has Valuation We arrive at our $65 price target for AET by compar by AET for target price $65 our at arrive We Valuation has pulled back to 10x P/E with 2014E earn- 2014E with P/E 10x to back pulled has Valuation ’ s stockCantor s $61.78, wasat trading ’ s stockCantor s $61.78, wasat trading ae o avail- on Based ae o avail- on Based

Fitzgerald

Fitzgerald - - - -

- - - -

2/7/17 5:26PM 2/7/17 5:26PM

www.mhhe.com/shefrin2e Source: Company Reports, Leerink Swann Research Estimates. Research Swann Leerink Reports, Company Source:

2/7/17 5:26 PM $107.44 Share Per Value

368 Shares Diluted

$39,507 Value Equity

+ $990 Cash ) (

– ($8,350) Debt ) (

$46,866 Value Enterprise

Calculation

Equity Value per Share Share per Value Equity

$46,866 Value Enterprise

Vo emnlVle $22,916 Value Terminal of PV

FCFs

$23,950 PV of Forecast Period Period Forecast of PV

PV of Cash Flows Cash of PV 222 411 348 266 255 239 228 202$22,916 $2,062 $2,228 $2,399 $2,555 $2,676 $3,458 $4,141 $2,292

Discount Factor @ 9.0% @ Factor Discount 099 .9 084 .5 063 .3 054 0.535 0.584 0.636 0.693 0.756 0.824 0.898 0.979 1

Years Out Years 03 . 23 . 43 . 63 7.3 6.3 5.3 4.3 3.3 2.3 1.3 0.3 0

ΔWC/ΔRevs. 1.%–50 1.%–50 1.%–50 1.%–50 –15.0% –15.0% –15.0% –15.0% –15.0% –15.0% –15.0% –15.0% –15.0%

CapEx .% .% .% .% .% .% .% .% 1.5% 1.5% 1.5% 1.4% 1.4% 1.4% 1.3% 1.3% 1.0%

D&A .% .% .% .% .% .% .% .% 1.5% 1.5% 1.5% 1.5% 1.5% 1.5% 1.5% 1.5% 1.3%

Misc. Operating Metrics Operating Misc.

Adjusted Free Cash Flow Cash Free Adjusted 222 421 382 329 331 340 352 354$284 $3,210 $42,804 $3,534 $3,502 $3,460 $3,381 $3,249 $3,852 $4,231 $2,292

Reform Haircut Reform % % 0 0 0 0 0 0 0 0% 0% 0% 0% 0% 0% 0% 0% 0% 0%

Free Cash Flow Cash Free 222 421 382 329 331 340 352 354$284 $3,210 $42,804 $3,534 $3,502 $3,460 $3,381 $3,249 $3,852 $4,231 $2,292

– ) Chg. in Working Cap. Working in Chg. ) ( 9 184 ,3 0 56 69 0 60 559 610 608 609 586 509 1,237 1,834 292

– ) CapEx ) ( 38 69 75 79 84 93 104 119 –1,175 –1,119 –1,034 –953 –874 –799 –735 –609 –338

+ ) D&A ) ( 40 73 86 87 96 97 108 119 $1,17 $1,119 $1,058 $997 $936 $877 $826 $703 $450

NOPAT $1,348 $1,675 $2,106 $1,888 $2,303 $2,523 $2,662 $2,732 $2,807 $2,871 $2,924 $2,924 $2,871 $2,807 $2,732 $2,662 $2,523 $2,303 $1,888 $2,106 $1,675 $1,348 $2,651

Taxes –767 –953 –1,198 –1,074 –1,311 –1,435 –1,515 –1,555 –1,597 –1,633 –1,664 –1,664 –1,633 –1,597 –1,555 –1,515 –1,435 –1,311 –1,074 –1,198 –953 –767 –1,508

% Margin % 6.30% 8.00% 10.10% 8.60% 7.70% 7.20% 7.10% 6.90% 6.60% 6.40% 6.20% 6.20% 6.40% 6.60% 6.90% 7.10% 7.20% 7.70% 8.60% 10.10% 8.00% 6.30% 5.30%

Operating Income Operating $2,115 $2,629 $3,304 $2,963 $3,614 $3,958 $4,177 $4,287 $4,404 $4,504 $4,588 $4,588 $4,504 $4,404 $4,287 $4,177 $3,958 $3,614 $2,963 $3,304 $2,629 $2,115 $4,159

Investment Income Investment $1,036 $1,056 $931 $917 $882 $1,021 $1,291 $1,440 $1,611 $1,841 $2,097 $2,097 $1,841 $1,611 $1,440 $1,291 $1,021 $882 $917 $931 $1,056 $1,036 $2,202

% Margin Margin % .% 48 73 59 58 .% 49 .% .% 38 .% 2.5% 3.3% 3.8% 4.2% 4.6% 4.9% 5.3% 5.8% 5.9% 7.3% 4.8% 3.2%

EBIT $1,079 $1,572 $2,373 $2,045 $2,732 $2,937 $2,886 $2,847 $2,793 $2,663 $2,491 $2,491 $2,663 $2,793 $2,847 $2,886 $2,937 $2,732 $2,045 $2,373 $1,572 $1,079 $1,958

– – % Growth % 0.9% 6.0% 35.3% 17.6% 6.2% 6.7% 6.5% 6.1% 5.8% 5.8% 6.1% 6.5% 6.7% 6.2% 17.6% 35.3% 6.0% 0.9% 2.0% 9.6% 5.0%

Operating Revenues $33,643 $32,962 $32,681 $34,628 $46,851 $55,099 $58,493 $62,402 $66,463 $70,516 $74,580 $74,580 $70,516 $66,463 $62,402 $58,493 $55,099 $46,851 $34,628 $32,681 $32,962 $33,643 Revenues Operating $78,309

2009 2010 2011 2012 2013E 2014E 2015E 2016E 2017E 2018E 2019E 2019E 2018E 2017E 2016E 2015E 2014E 2013E 2012 2011 2010 2009 Year Value

Terminal Terminal Terminal

Source: Company Reports, Leerink Swann Research Estimates. Research Swann Leerink Reports, Company Source:

AET DCF Valuation DCF AET

— 2015E — — — $59,672.00 $59,672.00 — — 70 9.3x $7.03 — —

2014E — — — — $56,118.00 $56,118.00 — — 63 10.3x $6.32 — —

2013E $9,507.0A $11,537.0A $12,994. 0A $13,694. 00 $47,754.00 $47,754.00 00 $13,694. 0A $12,994. $11,537.0A $9,507.0A 2013E 15A$.2 15A$.6$.7 11.1x $5.87 $1.36 $1.50A $1.52A $1.50A

2012A $8,864.00 $8,827.00 $8.899.00 $8.955.00 $35,545.00 $35,545.00 $8.955.00 $8.899.00 $8,827.00 $8,864.00 2012A 13 13 15 09 $.312.7x $5.13 $0.94 $1.55 $1.31 $1.34

e r1Q Yr Dec 2Q 3Q 4Q Ye Q2 Q4 YESP/E EPS FY 4Q 3Q 2Q 1Q FYRev Leerink Swann Free Cash Flow Table for Aetna for Table Flow Cash Free Swann Leerink EXHIBIT 3-8 3-8 EXHIBIT

she77208_ch03_058-084.indd 83 2/7/17 5:26 PM

- - - - Fitzgerald Fitzgerald

Based on avail- s stock was trading at was $61.78, s Cantor stock ’ Valuation Valuation has pulled back to 10x P/E with 2014E earn- We We arrive at our $65 price target for AET by compar FAIR FAIR MARKET PRICE: VALUATION able data as of 19, Nov. 2013, we believe that AET should be trading at $49.68. This makes AET trad- be should 30.32% stock the overvalued. believe we what indicates Value Fair ing at today if the stock market were perfectly efficient and traded everything at its true base worth.we this For on AET, quarters four previous the for (EPS) share earningsper actual We use an We absolute and P/E relative multiple-based toapproach valuation on the average of our 2014 and 2015 EPS estimates. We combine the P/E-based approach with a DCF model . . . thepoints to of the intrinsic that company value ings upside the most important catalyst. Our 12–18 month Price of Target $72 contemplates a P/E of 10.5x the average Exhibit 3-8.) . . . (See EPS views of 2014 + 2015 estimated We are maintaining our 2013–14 earnings estimates, $65 price$65 earnings estimates, 2013–14 our maintaining are We target and HOLD rating. Although, strictly speaking, 3Q:13 was a slight miss, it was mostly in line, and the company ex pects 2014 EPS to be at least flat with this guidance year’s of $5.80–$5.90. The Medicare book,company’s half of which is group (experienced rated), is reporting higher loss ratios, and management is cautious about the 2014 outlook for the indi vidual MA business, but Aetna is less exposed than other na tional carriers and thatsays its business exchange (focused on catastrophic and bronze plans) will not materially increase its 3% of revenue. accounts for individual book, which ing its current valuation on 2014 estimated earnings with that of other large commercial plans and the stock’s own trading Our history. target price equates to 10.8x our 2014 EPS estimate of $6.00. At its current valuation of our 2014 10.3x EPS estimate, AET trades roughly at parity with the group, consistent with its traditional in-line valuation, despite the potentially transformative changes in management in recent senior years and the 5/7/13 acquisition of Healthcare. Coventry (discount)/average group (discount)/average EV/EBITDA, premium (discount)/ dividend value, asset parts,net the of sum P/E, group average 12 months. the next over returns, (ROE) return and on equity Leerink Swann: In its report dated November 19, ValuEngine: In its report dated November 19, when Cantor Cantor Fitzgerald: In its report dated October 29, when Aetna’s stock was trading at $65.08, Leerink Swann Leerink$65.08, Swann at trading was stock Aetna’s when stated the following: Aetna’s stock was trading at $64.75, ValuEngine had a one-year price target of $67.28 for Aetna, and stated the following: when Aetna stated the following: - During the first week 21 Valuation/Risks: Valuation/Risks: Our $77 PT reflects a P/E multiple of The price targets are based on several methodologies We We continue to believe AET can modestly grow next year and earnings increase EPS 15%+ in 2015. 3Q results were uneventful, but the company guided to a that 2014 implies modest EPS growth floor and improves visibility. AET’s reiterate We clarity. 2014 bring more will Day Investor 12/12 our Buy rating on AET, which is the cheapest MCO stock (9.8x/8.4x 2014-15E EPS) . . . 12.4x our 2014E EPS. AET trades at 9.8x our 2014E EPS, of 10.7x and its thehis- is which below broader peer average torical 11.1x. Risks: 1) spike in cost trend, 2) Coventry inte- 4) 2014 MA headwinds . . . gration, 3) Exchanges, which may include, but are not market restricted risk, to, growth analyses rate, of revenue stream, discounted flow cash (DCF), EBITDA, EPS, cash flow (CF), free cash flow premium P/FCF, P/CF, PE/growth, P/E, EV/EBITDA, (FCF), The consensus analyst forecast for Aetna’s third-quarter In its announcement, Aetna explained that poor per- The reports of analysts covering Aetna, whose stock Jefferies: In its report dated October 29, when Aetna’s 82 Chapter Three at trades it yet and 2014/2015, in plans large-cap the among the lowest multiple in the group.” of August Cantor Fitzgerald increased its price target from $60 to $65 a share. In mid-September, Bank of America/ Merrill Lynch stated that current Aetna’s valuation overly thediscounted risks and withassociated health reform care $79. $68 to from its price and increased target exchanges, EPS was $1.52 per share on in approximately revenue. On $12.87 October billion 28, Aetna announced that its quarter earnings had 4 increased percent by to $518.6 third- million, notedfirm The billion. $13.04 of revenue on share per $1.38 or EPS adjusted its costs, integration as such items excluding that was $1.50 per share. In respect to the negative earnings sur prise, the lead Citigroup analyst covering Aetna commented that even in its previous conference call, the firm had talked third-quarter earnings.down formance formance from two products and federal budget cuts that Advantage, (Medicare business Medicare its hurt had year MA). Health care costs, which were its largest expense, had increased 57 percent in the quarter. The firm reiter- ated that its 2013 adjusted earnings would range between Aetna’s to response market The share. per $5.90 and $5.80 third-quarter earnings announcement was a decline in its Octo- thebut stock’s $60.63, to percent 1.9 price by stock at $62.70. later was days ber close a few associ- and targets price their in varied were AET, is ticker examples. four are Below ated methodologies. stock was trading at $61.78, July: from Jefferies $77 of target price stated its maintaining was it that the noting following, she77208_ch03_058-084.indd 82 she77208_ch03_058-084.indd 84 she77208_ch03_058-084.indd 84 www.mhhe.com/shefrin2e ChapterThree 84 ChapterThree 84 1. stock’sthis valuation today. stands alternative perspectives should give you a good idea about where two These size. similar of companies from and sector same the in stocks comparable five of today ratio marketaverage the (ii) and stock), the for available (or data is years there long 10 however past the over stock the of ratio market historical P/Sto assessitsfair to the wise, we resort value. today.Other- of as stock the for value fair a give to P/E the apply we positive, is EPS quarter 4 forward its but negative is growth forecastedEPS its If positive. are rate growth EPS forecasted and EPS quarter 4 trailing its both if assessment value fair a give to PEG the apply we stock, given each For and expected future EPS growth, while P/E is better than ratio P/S. price/earnings the both reflects it as informative most (price to trailing 4 quarter sales). Among the three, PEG is the ratio P/S and ratio), earnings quarter forward4 to (price P/E growth), EPS year next forecasted analyst consensus the by divided ratio, earnings quarter 4 trailing to (price PEG tios: forecasts are asreliable aspossible. return stock our that ensure to techniques tical/econometric statis- advanced most the apply then Weindustry. every for and horizon time each for model forecasting distinct a use reversals.We price Long-term and continuation, momentum Intermediate-term reversals, price Short-term exhibit: tently consis- price stock that tendencies important several things, other among capture, models forecasting Our publications. forecasting well-established variables derived from credible financial research studies and and models proprietary price both target include forecast ValuEngine’s in used ables vari- predictive The PRICES: TARGET FORECASTED . .. 30-year tothe 3.50% of Treasuryyield bond correlations and $6.16, of quarters fournext the for EPS forecasted $5.60, of 1. stock’sthis valuation today. stands alternative perspectives should give you a good idea about where two These size. similar of companies from and sector same the in stocks comparable five of today ratio marketaverage the (ii) and stock), the for available data (or is years there long however 10 past the over stock the of ratio market historical P/Sto assessitsfair to the wise, we resort value. today.Other- of as stock the for value fair a give to P/E the apply we positive, is EPS quarter 4 forward its but negative is growth forecastedEPS its If positive. are rate growth EPS forecasted and EPS quarter 4 trailing its both if assessment value fair a give to PEG the apply we stock, given each For and expected future EPS growth, while P/E is better than P/S. ratio price/earnings the both reflects it as informative most (price to trailing 4 quarter sales). Among the three, PEG is the ratio P/S and ratio), earnings quarter forward4 to (price P/E growth), EPS year next forecasted analyst consensus the by divided ratio, earnings quarter 4 trailing to (price PEG tios: forecasts are asreliable aspossible. return stock our that ensure to techniques tical/econometric statis- advanced most the apply then Weindustry. every for and horizon time each for model forecasting distinct a use reversals.We price Long-term and continuation, momentum Intermediate-term reversals, price Short-term exhibit: tently consis- price stock that tendencies important several things, other among capture, models forecasting Our publications. forecasting well-established variables derived from credible financial research studies and and models proprietary price both target include forecast ValuEngine’s in used ables vari- predictive The PRICES: TARGET FORECASTED . .. 30-year tothe 3.50% of Treasuryyield bond correlations and $6.16, of quarters fournext the for EPS forecasted $5.60, of for AET. forecasted the over growthrateand EPS nextquarters 4 the EPS quarter 4 recent with PEG historical the multiplying on based is assessment following The 0.55. is ratio sales price current AET’s EPS. forward quarter 4 its for $6.16 is estimate consensus analyst The quarters. 4 recent its in share per $5.60 earned AET’sAET average2.84. is PEG Valuation Based on AET’s Past PEG: Over pastthe 10 years, for AET. forecasted the over growthrateand EPS nextquarters 4 the EPS quarter 4 recent with PEG historical the multiplying on based is assessment following The 0.55. is ratio sales price current AET’s EPS. forward quarter 4 its for $6.16 is estimate consensus analyst The quarters. 4 recent its in share per $5.60 earned AET’sAET average2.84. is PEG Valuation Based on AET’s Past PEG: Over pastthe 10 years, To establish a valuation standard, we use both (i) the average ra- market three of one on based valuation a Wepresent To establish a valuation standard, we use both (i) the average ra- market three of one on based valuation a Wepresent E-ae arVle$159.75 2.84 PEG-Based FairValue Historical AveragePEG Fair Value E-ae arVle$159.75 2.84 PEG-Based FairValue Historical AveragePEG Fair Value Case AnalysisQuestions . o wud o ue eairl ocps o explain to concepts behavioral use you would How 5. Morgan the of discussion the from insights What 4. Morgan the of discussion the from insights What 3. Morgan the of discussion the from insights What 2. Morgan the of discussion the from insights What 1. Case AnalysisQuestions . o wud o ue eairl ocps o explain to concepts behavioral use you would How 5. Morgan the of discussion the from insights What 4. Morgan the of discussion the from insights What 3. Morgan the of discussion the from insights What 2. Morgan the of discussion the from insights What 1. 2. eeal wy ifrn aayt arv a different at arrive price targets? analysts different why generally on AetnabyValuEngine? report 2013 the to apply eBay on report 2003 Stanley on AetnabyLeerinkSwann? report 2013 the to apply eBay on report 2003 Stanley on AetnabyCantorFitzgerald? report 2013 the to apply eBay on report 2003 Stanley on AetnabyJefferies? report 2013 the to apply eBay on report 2003 Stanley .mhhe.com/shefrin2e. ments for Aetna is available on the book web site, forecasted EPSgrowth rate over next the 4quarters. the and EPS quarter 4 recent AET’s with today PEG age sessment is based on multiplying comparable stocks’ aver- followingas- The 0.55. is ratio sales price EPS. AET’scurrent 4-quarter-forward its for $6.16 is estimate sensus con- analyst The quarters. 4 recent its in share per $5.60 earned AET 1.66. is comparables these of average PEG rables are CNC, HNT, HUM, WCG and WLP. compa- The current AET’s PEG: Comparables’ on Based Valuation 2. eeal wy ifrn aayt arv a different at price targets? arrive analysts different why generally on AetnabyValuEngine? report 2013 the to apply eBay on report 2003 Stanley on AetnabyLeerinkSwann? report 2013 the to apply eBay on report 2003 Stanley on AetnabyCantorFitzgerald? report 2013 the to apply eBay on report 2003 Stanley on AetnabyJefferies? report 2013 the to apply eBay on report 2003 Stanley .mhhe.com/shefrin2e. ments for Aetna is available on the book web site, An Excel file containing historical financial state- financial historical containing file Excel An oprbePGBsdFi au $93.72 Comparable PEG-BasedFairValue Comparable StocksAgPEG Fair Value forecasted EPSgrowth rate over next the 4quarters. the and EPS quarter 4 recent AET’s with today PEG age sessment is based on multiplying comparable stocks’ aver- followingas- The 0.55. is ratio sales price EPS. AET’scurrent 4-quarter-forward its for $6.16 is estimate sensus con- analyst The quarters. 4 recent its in share per $5.60 earned AET 1.66. is comparables these of average PEG rables are CNC, HNT, HUM, WCG and WLP. The compa- current AET’s PEG: Comparables’ on Based Valuation An Excel file containing historical financial state- financial historical containing file Excel An oprbePGBsdFi au $93.72 Comparable PEG-BasedFairValue Comparable StocksAgPEG Fair Value WLP WCG HUM CurrentPEG HNT CNC Comparables Comparables’ PEG WLP WCG HUM CurrentPEG HNT CNC Comparables Comparables’ PEG 0.69 2.64 n/a n/a n/a 0.69 2.64 n/a n/a n/a 1.66 1.66 www www 2/7/17 5:26PM 2/7/17 5:26PM she77208_ch04_085-109.indd 85 she77208_ch04_085-109.indd 85 4.1 4.1 Chapter Four Chapter Four TRADITIONALTREATMENT OFCAPITAL BUDGETING TRADITIONALTREATMENT OFCAPITAL BUDGETING Capital Budgeting Capital Budgeting 2. 2. . Distinguishbetweentheremediesappropriatetoagency conflicts andtheremedies 4. 3. Distinguishbetweentheremediesappropriatetoagency conflicts andtheremedies 4. 3. taxes, andtransaction costs. risk, differentialproject reflect to capital of cost the adjust havingto and equity of These challenges involve having to rely on asset models pricing to estimate the cost tice, managers confront a series of challenges in deriving appropriate discount rates. average of its constituent components, where the weights are market values. In prac- weighted a as estimated is capital of cost equity.the theory,and In debt long-term taxes, andtransaction costs. risk, differentialproject reflect investorsfirm’sto that the expected hold return to capital the order require in capital, of of costcost the adjust havingto and equity of These challenges involve having to rely on asset models pricing to estimate the cost firm’s investors,wereittobeadopted. tice, managers confront a series of challenges in deriving appropriate discount rates. the for create would project the value fundamental incremental the measures NPV average of its constituent components, where the weights are market values. In prac- weighted a as estimated is capital of cost equity.the theory,and In for debt deciding whether or long-term not to adopt a project is net present value (NPV). In theory, recommend academics that criterion important most The analysis. the into investorsfirm’s porated that the expected hold return to the order require in capital, of cost for criteria incor- is different risk project which emphasizes in manner the budgeting including flows, cash project capital assessing to approach traditional The firm’s investors,wereittobeadopted. the for create would project the value fundamental incremental the measures NPV for deciding whether or not to adopt a project is net present value (NPV). In theory, recommend academics that criterion important most The analysis. the into porated for criteria incor- is different risk project which emphasizes in manner the budgeting including flows, cash project capital assessing to approach traditional The and termination. impact managers’ adversely forecasts of project effects cash flows framing and their decisions and about project adoption biases heuristics, which in manner the tify The main objective of this chapter is for students to demonstrate that they can iden- and termination. impact managers’ adversely forecasts of project effects cash flows framing and their decisions and about project adoption biases heuristics, which in manner the tify The main objective of this chapter is for students to demonstrate that they can iden- After completingthischapterstudentswillbeableto: After completingthischapterstudentswillbeableto: . Explainwhyexcessiveoptimismandoverconfidenceleadmanagers toadopt 1. Explainwhyexcessiveoptimismandoverconfidenceleadmanagers toadopt 1. appropriate tobehavioralbiases. projects. exclusively ontheaffectheuristicarepronetoselectlow-valueprojectsoverhigh-value Explain whymanagerswhoavoiddiscountedcashflowanalysisandinsteadrely appropriate tobehavioralbiases. leads managerstocontinuefailingprojectswhentheyshouldterminatethoseprojects. Explain whythecombinationofaversiontoasureloss,regret,andconfirmationbias projects. exclusively ontheaffectheuristicarepronetoselectlow-valueprojectsoverhigh-value Explain whymanagerswhoavoiddiscountedcashflowanalysisandinsteadrely leads managerstocontinuefailingprojectswhentheyshouldterminatethoseprojects. Explain whythecombinationofaversiontoasureloss,regret,andconfirmationbias The starting point for obtaining the project discount rate is typically the firm’s the typically is rate discount project the obtaining for point starting The firm’s the typically is rate discount project the obtaining for point starting The and latecompletiontimes. negative net-present-valueprojectsbecauseofcostoverruns,revenueshortfalls, and latecompletiontimes. negative net-present-valueprojectsbecauseofcostoverruns,revenueshortfalls, 85 85 2/7/17 5:26PM 2/7/17 5:26PM