Literature

Question 1

What is the article Weber & Camerer “the disposition effect in securities trading” about?

Answer 1

• It is about the disposition effect: tendency to sell assets that have gained value (“winners”) and sell stocks that have recently lost value (“losses”). Disposition effect can be explained by 2 factors of prospect theory:
  
  1. People value gains and losses relative to a reference point = reference effect
  2. Tendency to face risk when faced with losses and avoid risk when a certain gain is possible = reflection effect.
• Weber & Camerer designed an experiment to see if subjects would exhibit disposition effects.
• Results: contrary to Bayesian optimization, subjects did tend to sell winners and keep losers. But, if the shares were automatically sold after each period, the disposition effect was greatly reduced.

Question 2

What were the hypotheses of Weber & Camerer in “The disposition effect in securities trading: an experimental analysis”?

Answer 2

• The reference point used was the original purchase price, because if the reference point is the current price, there is no disposition effect. 2 periods.
• Hypotheses made for markets in which prices are independent of investor behavior (perfect competition).
• Main hypothesis:
  
  • Number of shares sold will be smaller for losing assets than for winning assets.
• H1: Purchase price reference point:
  
  • Subjects sell more shares when the sale price is above the purchase price than when the sale price is below the purchase price
• H2: Last period reference point:
  
  • Subjects sell more shares when the sale price is above the last period price than when the sale price is below the last period price.
• H3: Automatic selling because selling a stock requires a deliberate action:
  
  • Disposition effects are smaller when assets are automaticcaly sold than when selling is deliberate.
• H4: trading volume is positively correlated with the size of price changes:
  
  • A larger price will change the salience of a stock and increase trading volume.

Question 3

What was the experimental design in Weber and Camerer “The disposition effect in securities trading: an experimental analysis”?

Answer 3

• Portfolio decisions before each of 14 periods, sell and buy 6 risky assets at announced prices, prices generated by a random process and not determined by trading actions of the subjects because they were interested in isolating the disposition effect.
• Subjects got 10 000, not short sell or borrow. 2 types of sessions: session I: deliberate selling, holdings of the shares equal to last period. Session II: automatic selling: all shares immediately sold but subjects could buy back.
• Price of each asset would rise or fall, different chances:
  
  • ++:65%, +:55%, –=35% etc.
  • Subjects did not know which asset had which label (A-F), but they knew distribution.
  • Independent from that was the size of the price rise, the expected value of price change for a randomly chosen stock was zero.
• Subjects had to infer distribution from past data. A Bayesian subject would continually update her probabilities, based on observed price movements. The share with the most price increases is the most likely to have the trend: ++.

Question 4

Why did they choose that experimental design in Weber & Camerer (1998), The disposition effect in securities trading: an experimental analysis?

Answer 4

• Design had an important advantage: since the share that has risen most frequently is most likely the ++ share, the share investors should be least eager to sell.
• Similarily, the most frequent loser is most likely –, the investors should be eager to sell it.
• So: disposition effect is clearly a mistake.

Question 5

What were the results in Weber and Camerer (1998), The disposition effect in securities trading: an experimental analysis?

Answer 5

1. H1: nearly 60% of the shares sold were winners and 40% were losers. No substantial difference LIFO or FIFO.
2. H2: price of the last period is adopted as a reference point.
   
   • If this is true: more shares sold after (GG,LG) than (LL,GL).
   • Result: twice as many units where sold when the price rose in the last period. But in experiment II, with automatic selling, the effect almost completely dissappears.
   • So the disposition effect can be traced to a reluctance to sell deliberately, rather than an eagerment to hold to losing shares.
3. H3: Automatic selling does wipe out the disposition effect. Disposition coefficient alpha is zero if there is no disposition effect and positive if there is. Alpha 1 is not statistically equal to alpha II.
4. H4: price change is correltaed with volume traded = cannot be refected.

Both mean reversion and disposition effect existed here.

Question 6

What is Odean (1998), “Are Investors Reluctant to Realize Their Losses” about,

Answer 6

• Odean tests the disposition effect (Shefrin and Statman), the tendency of investors to hold losing investments too long and sell winning investments too soon, by analyzing trading records for 10,000 accounts at a large discount brokerage house.
• Their behavior demonstrate a strong preference for realizing winners rather than losers. Their behavior does not seem to be motivated by the desire to rebalance portfolios or to avoid the higher trading costs of low priced stocks. Nor is it justified by subsequent portfolio performance. For taxable investments, it is suboptimal and leads to lower after-tax returns. Tax-motivated selling is most evident in December.

Question 7

What is the methodology in Odean (1998), Are Investors Reluctant to Realize Their Losses?

Answer 7

• The study tests whether investors sell their winnings too soon and hold losers too long. It also investigates tax-motivated trading in December.
• Reference point: purchase date. 97 000 transactions: compare the selling price for each stock and determine whether it has been sold for a gain or a loss.
  
  • Each stock that is in that portfolio but is not sold is considered a paper gain or loss. Comparing the high or low to the average purchase price. It both its daily high and low are above the average purchase price, then it is counted as a paper gain.
• PGR = proportion of Gains realized and PLR = proportion of losses realized.
  
  • PGR=1/2
  • PLR=1/4
• Any test for the disposition effect = joint hypothesis that people sell gains more readily than losses and of the specification of the reference point.
• Primary finding: investors are reluctant to sell their losers and prefer to sell winners.
• Null hypothesis: PGR<plr: this is rejected.>
  <li>In December: some investors do engage in tax-motivated selling in December. </li><li>The investors does not seem to appear to be motivated by a desire to rebalance portfolios or reluctance to incure trading costs. Nor is it justified by subsequent portfolio performance.
  <ul>
  <li>It leads to lower returns, particularily for taxable accounts. </li>
  </ul>
  </li></plr:>

Question 8

What is Bernatzi and Thaler (1995), “Myopic Loss Aversion and the Equity Premium Puzzle” about?

Answer 8

• The equity premium puzzle refers to the empirical fact that stocks have outperformed bonds over the last century by a large margin. 2 explanations based on behavioral concepts:
  
  1. Loss averse investors: more sensitive to losses than to gains
  2. Evaluate their portfolios frequently
• Leads to myopic loss aversion.
• Using simulations, we find that the size of the equity premium is consistent with the parameters of prospect theory if investors evaluated their portfolios annually.
• Difference returns stocks and fixed income securities = 7%, if only risk aversion, it would be 30 –> why is the equity premium so large or why is anyone willing to hold bonds?

Question 9

What is mental accounting in Bernatzi and Thaler (1995), Myopic Loss Aversion and the Equity Premium Puzzle?

Answer 9

• Mental accounting refers to the implicit methods individuals use to code and evaluate financial outomes: transactions, investments and gambles.

Question 10

How often are portfolios evaluated in Bernatzi and Thaler (1995), Myopic Loss Aversion and the Equity Premium Puzzle?

Answer 10

• They use samples from the historical (1926-1990). They rank the returns from best to worst –> it is possible to compute the prospective utility of the given asset for the specified holding period.
• Nominal is preferred to real, because that is the way returns are usually given, bonds a better substitute than T-bills.
• The point where the curves cross is the evaluation period at which the stocks and bonds are equally attractive. This is 13 months for nominal and for real 10-11 months. One year is highly plausible.
• Most frequent allocation between stocks and bonds is 50-50, with the average allocation to stocks below 50%. Stocks become more attractive as the evaluation period increases.
• Pension funds: the CFO is not expected to remain in this job forever, so a shorter horizon. Therefore there is a conflict of interest between the pension fund manager and the stockholders. Agency costs produce myopic loss aversion.

Question 11

What is the conclusion on Bernatzi and Thaler (1995), Myopic Loss Aversion and the Equity Premium Puzzle?

Answer 11

• Equity premium puzzle is a puzzle within the standard expected utility-maximizing paradigm. Mehra and Prescott: impossible to reconcile the high rates of return on stocks with the very low risk-free rate.
• Solution: combine high sensitivity to losses with a prudent tendency to frequently monitor one’s wealth. So people demand a large premium to accept return variability.
• Myopic loss aversion is a possible solution to Mehra’s and Prescotts puzzle.

Question 12

What is Gneezy, Kapteyn & Potters (2003), Evaluation Periods and Asset Prices in a Market Experiment about?

Answer 12

• Gneezy and Potters test whether the frequency of feedback information about the performance of an investment portfolio and the flexibility with which the investor can change the portfolio influence her risk attitude in markets.
• In line with myopic loss aversion (Bernatzi and Thaler): more information and more flexbility results in less risk taking. Market prices of risky assets are significantly higher if feedback frequencey and decisions flexibility are reduced.
• This result supports the findings from individual decision making, and shows that market interactions do not eliminate such behavior or its consequences for prices.
• Eg. Bank Hapoalim: Israel’s largest mutual funds manager announced change to its information policy: only every 3 months would it sent information.
• According to myopic loss aversion:
  
  • Mypoic refers to myopia, inappropriate treatment of time dimension: bad news one day is treated as bad news longer period.
• There has been some empirical evidence, but all individual decision-making, so this does not mean that the market performs like this or that the market will violate expected utility theory. Small number of rational agents might be enough to make the outcome rational.

Question 13

What is the goal of Gneezy, Kapteyn & Potters (2003), Evaluation Periods and Asset Prices in a Market Experiment?

Answer 13

• Test whether the effects of MLA show up in a competitive environment. They set up experimental markets in which traders adjust their portfolios by buying ad selling risky financial assets.
  
  • High-frequency treatment: investors commit for 1 period.
  • Low-frequency treatment: commit for 3 periods and only informed about the returns after the 3 periods.
• Finding: Prices of the risky asset in the low-frequence are significantly higher than in the high-frequencey experiments. Investors are more willing to invest in risky assets if they evaluate the consequences in a more time-aggregated way. This has a positive effect on prices.
  
  • There are natural intervals over which decisions will be made and evaluated.

Question 14

What is the experimental design in Gneezy, Kapteyn & Potters (2003), Evaluation Periods and Asset Prices in a Market Experiment?

Answer 14

• Market which 8 participants can trade units of a risky asset in a sequence of 15 trading periods. Each unit of the asset is a lottery ticket, which at the end of a trading period pays 150 cents with a probability of 1/3, 0 otherwise.
• Trader is endowed with 200 cents and 3 units of the asset.
  
  • High-frequency: opens and able to trade every period.
  • Low-frequency: only open in blocks of 3. The holdings are fixed, if you buy in period 1, also in 2 and 3. In 4 you can make choices again. Traders are informed abou the results simultaneously.
• Information feedback and frequency of portfolio adjustment is lower, participants in the treatment are expected the evaluate in a more aggregated way.
• Undergraduate students were used, double auction rules. Traders could submit bids to buy and offers to sell. All traders were informed about all bids and offers.
• Result was used by drawing a disk out of a box: black meant all assets = 0, one of the 2 white balls = 150.

Question 15

What is the result in Gneezy, Kapteyn & Potters (2003), Evaluation Periods and Asset Prices in a Market Experiment?

Answer 15

• Prices are volatile in the beginning, but stabilize fairly quickly.
• Clear treatment effect in the direction predicted by MLA. In all rounds, average transaction prices are lower in treatment H (49.3) than in treatment L (58,4). The average number of trades per round per trader is almost identical to each treatment. Manipulation only affected the price level and not the willingness to trade. Average range of final number of assets is similar across the treatments.
• The average price of the asset in treatment L is above its expected value of 50 = subjects are risk seeking. Overpricing is quite common in experimental markets. Could possible be the endowment effect = traders more reluctant to sell than they would be on strict evaluation of financial gains and losses.
• Possible: utility of gambling = upward effect on pricing, people like having the asset. Overconfidence if possible, putting too much weight on probability that the asset will have a positive value. House money effect, people more willing to gamble when they have earned money on prior gambles, but can here not explain since the average realized asset value was a bit higher in H (luck).

Question 16

what is the conclusion on Gneezy, Kapteyn & Potters (2003), Evaluation Periods and Asset Prices in a Market Experiment?

Answer 16

• Main question: whether the frequence of information feedback and flexibility of portfolio adjustment affect asset prices. Result: More information feedback and more flexibility reduce the price of the risky asset.
  
  • In line with individual decision making, intertemporal framing effects matter, not just for individual decision making, but also in market settings.
  • Direction of the price effect is in line with MLA.

Question 17

What is Haigh & List (2005), Do Professional Traders Exhibit Myopic Loss Aversion? An Experimental Analysis about?

Answer 17

• As MLA has been experimently proven with students, they want to show if traders from CBOT also have this effect.
• Surprisingly, they find that traders exhibit behaviour consistent with MLA to a greater extent than students.
• Because ordinary students don’t have that big of an effect on markets, it is important to check whether professionals show this behavior. They use 54 professional traders and options pit traders.
  
  • Important: market prices of risky assets might be significantly higher if feedback frequency and decision flexibility are reduced. Institutions have the ability to influence prices.

Question 18

What is the experimental design of Haigh & List (2005), Do Professional Traders Exhibit Myopic Loss Aversion? An Experimental Analysis?

Answer 18

• 2x2 experimental design: professional traders and undergraduate students as a control group.
  
  • Treatment F = frequent feedback
  • Treatment I = infrequent feedback. Blocks of 3 homogenous and information at the end of the 3 periods.
• 9 rounds with 100 units per round, where they get to choose how much of the endowment they can use for the lottery: 1/3 winning 350 and 2/3 losing. No communication and no seeing of other people’s payoffs.

Question 19

Results of Haigh & List (2005), Do Professional Traders Exhibit Myopic Loss Aversion? An Experimental AnalysisK

Answer 19

• Professional traders exhibit behavior consistent with MLA to a greater extent than undergraduate students.
• Both traders and students exhibit investment behavior in line with MLA, traders exhibit this behavior to a greater extent than students. They show this more than any other subject pool that has been evaluated.
• So expected utility theory does not model professional trader’s behavior well and so providing less freedom to adjust might redue the likelihood that a sell-off enssues after a minor setback. If market information becomes more readily available at a lower cost one might expect it to be used more often, hence affecting behavior over riskier assets and therefore relative prices.

Question 20

What is the article Trautmann, Vieider & Wakker (2008), Causes of Ambiguity Aversion: Known Versus Unknown Preferences?

Answer 20

• Ambiguity aversion appears to have suble psychological causes. Curley, Yates and Ambrams found that the fear of negative evoluation by other (FNE) increases ambiguity aversion.
• This paper introduces a design in which preferences can be private information of individuals so FNE can be avoided entirely to control for FNE and other social factors to check to what extent ambiguity aversion is driven by such social factors.
• Ambiguity aversion completely disappears if FNE is eliminated.
• Uncertainty = risk vs. vague probabilities = ambiguity.
• Ambiguity aversion increases with the perception that others are more competent and more knowledgeable. If people choose an ambiguous option and receive a bad outcome, then they fear criticism by others.
  
  • When a risky choice is made, it is simply bad luck and therefore no negative evaluation.
  • Fear of negative evaluation = FNE.

Question 21

What is the general experimental design in Trautmann, Vieider & Wakker (2008), Causes of Ambiguity Aversion: Known Versus Unknown Preferences and why is it good for this?

Answer 21

• Introduce a design where preferences between outcomes are the subjects’ private infromation that cannot be known to the experimenter or other people unless the subjects explicity reveal it = completely control for presence or absence of FNE.
• Stimuli: 2 DVDs: equally popular but most individuals have strong preference that is unknown to others. Subjects choose between a risky and an ambiguous prospect to win. With preferences unknown, the subject cannot be judged negatively.
  
  • Then ambiguity aversion disappears completely.
  • But: letting subjects announce their preferences beforehand –> ambiguity aversion back as strongly as before.

Question 22

How did the experiment go in Trautmann, Vieider & Wakker (2008), Causes of Ambiguity Aversion: Known Versus Unknown Preferences?

Answer 22

• Subjects would always win one DVD, they were not told the price and they could earn up to 80cents more. Most subjects had a strong preference, but no difference in social desirability and no difference by gender = preferences were unpredictable.
• Choice: Endowment of 10 and could keep the rest
  
  • Risky lottery: 9,80
  • Ambiguous prospect, which was 50cents cheaper. 9,20
• 2 groups:
  
  • Known preferences: write down which movie they wanted
  • Unknown preferences: not to tell the experimenter.
• Questionnaires ex post: valuation difference = subject’s maximum willingness to pay to exchange her less preferred for her more perferred = showed that people had strong preferences.
• Lottery mechanism:
  
  • Risky: 3XXX and 3O’s: role a die and see if they win
  • Ambiguous: they don’t know how many X’s or O’s on back of the cards.
  • Always got 1 DVD

Question 23

What was the result in Trautmann, Vieider & Wakker (2008), Causes of Ambiguity Aversion: Known Versus Unknown Preferences?

Answer 23

• Known preferences: more than half of the subjects choose the risky prospect = ambiguity aversion.
• Unknown: no ambiguity aversion, significantly less than half of the subjects choose the risky prospect. Statistical significant difference. 30-percentage point reduction of ambiguity averse choices. If you’re ambiguity neutral, the fact that the ambiguous is 50c cheaper, you should choose that one.
  
  • FNE comes from ex-post behavior: if they had chosen the ambiguous prospect, afterwards they claimed that they were succesfull much more than people who picked the risky option. So that suggest that losing after playing the ambiguous prospect is more embarrassing than after playing a 50-50 prospect. So such ex-post justifications are motivated primarily by social evalutions

Question 24

What is the article Sarin & Weber (1993), Effects of Ambiguity in Market Experiments about?

Answer 24

• Ambiguity aversion: they want to evaluate the effect of ambiguity on individual decisions and the resulting market price in market settings: whether ambiguity effects persist in the face of market incentives and feedback.
• 2 different market organizations: sealed bid auction and double oral auction.
  
  • Individual bids and market prices for lotteries with ambiguous probabilities are consistently lower than the corresponding bids and market prices for equivalent lotteries with well defined probabilities.
  • Aversion to ambiguity does not dissappear in market settings.

Question 25

What was the experimental design in Sarin & Weber (1993), Effects of Ambiguity in Market Experiments?

Answer 25

• 3 studies involving 14 experiments.
  
  • Objective in the first 8: are the individual bids and market prices of ambiguous and unambiguous assets different in sealed-bid and double oral auctions.
  • 9-10: would the results be sustained when there are experienced executives. Only sealed-bid auctions
  • 11-14: Robust to a larger number of trading periods. 16 periods for sealed bid and .8 periods for double-oral.
• Students of university and executives of JP Morgan. Everyone had basic economic and statistics knowledge.
• Types of assets:
  
  • Red: paid 100DM with a chance p and 0 (1-p) = unambiguous. If 0.5=p, a ball out of an urn would be drawn 50/50 split. For 0.05: 1/20 balls.
  • Green: 100DM with a chance of anywhere between 0% and a 100% but a mean of p*. An urn was filled with 20 yellow and white balls but subjects did not know proportion, decided by a random number table = ambiguous asset.
• Bid price for an ambiguous asset must be at least as much as the bid price for an unambiguous asset since the subject could choose yellow or white ball. Neutral: uniform distributions around the mean.

Question 26

What was the market procedure in Sarin & Weber (1993), Effects of Ambiguity in Market Experiments?

Answer 26

• Double oral auctions: convergence to competitive equilibrium is much faster. Verbal bids to buy a certificate and sell = offer price
• Sealed-bid: 4 certificates were sold each period and they were sold to the 4 highest bidders at a price equal to the 5th highest. So incentive to their highest willingness to pay (Vickrey) = bid their true reservation price.
• Endowment was more than the maximum possible number of certificates.
  *

Question 27

What were the results in Sarin & Weber (1993), Effects of Ambiguity in Market Experiments?

Answer 27

• For p=0.5, the mean price averaged over all periods was lower for the ambiguous assets in each of the 10 experiments. It was lower in almost every period. The price of the ambiguous asset seemed to increase from the first to the last trading period
• For p=0.05: the mean price averaged over all periods was higher for the ambiguous asset, but the prices converged rapidly. But no systematic differences between the 2 assets prices.
• Behavior of experienced traders was not different than for the student subjects.
• Market prices for ambiguous assets are lower in both, but showed convergence in sealed-bid but not in double-oral auctions.
• Conclusion: Both assets had identical mean probabilities, but the individual bid prices and market prices for ambiguous asset were consistently below the corresponding individual bids and market price for an unambiguous asset when the probability of gain was 0.5. Regardless of the type of auction.
  
  • Why: ambiguous is more risky, more psychological discomfor, regret due to hindsight.
  • Market forces alone not enough to get rid of ambiguity on decisions.

Question 29

What is “Barber & Odean, 2001, Boys will be Boys: Gender, Overconfidence, and Common Stock Investment” about?

Answer 29

• Psychological research has shown that men are more overconfident than women, so the theory predicts that men will trade more excessively than women. They use account data from over 35,000 households from a large discount brokerage and document that men trade 45 percent more than women.
• Trading reduces men’s net returns by 2.65 percentage points per year as opposed to 1.72 percentage points for women.
• Overconfidence about knowledge and future prospects and abilities. Greater overconfidence leads to greater trading and to lower expected utility.
• 2 hypotheses:
  
  • Men trade more than women
  • Performance of men will be hurt more by excessive trading than women.
• Annual turnover rate for common stocks is nearly one and a half times that for women. Biggest difference between single men and single women.

Question 30

What was the data and methods in Barber & Odean, 2001, Boys will be Boys: Gender, Overconfidence, and Common Stock Investment?

Answer 30

• Common stock investments of 37,664 households –> gender identifified by the person who opened the brokerage account. Focus on common stocks, so no mutual funds etc.
• Women report having less investment experience than men: 47 percent for good or extensive, and 62.5% for men.
• Because married couples might influence decisions, biggest difference should be between single men and single women.
• Gross and net return of each household is calculated. Also the monthly turnover (women = 53% vs. 77%)/
• Possible theory could be of underperformance of men is inferior security selection.
• In summary:
  
  • Men who are more overconfident than women, trade more than women (as measured by monthly portfolio turnover).
  • Men lower their returns more through excessive trading than do women.
• Gender serves as a reasonable proxy for overconfidence, certainly in male-dominated fields, such as finance. Overconfident investors overestimate the precision of their information and the expected gains of trading.
• Other explanations: risk aversion, gambling, entertainment accounts.

Question 31

What is Glaser and Weber (2007), Overconfidence and Trading Volume about?

Answer 31

• Theoretical models predict that overconfident investors will trade more than rational investors. Test this hypothesis by correlating individual overconfidence scores with several measures of trading volume of individual investors.
• 3000 online broker investors = measure overconfidence (miscalibration, volatility estimates, better-than-average effect).
  
  • Investors who think that they are above average in terms of investment skills or past performance (but did not have above average performance) trade more.
  • Measure of miscalibration are unrelated to measures of trading volume.
• Annualized monthly turnover is incredibly high: DeBondt and Thaler: perhaps the single most embarrassing fact to the standard finance paradigm.

Question 32

What are the possible reasons for the amount of trading volume in Glaser and Weber (2007), Overconfidence and Trading Volume?

Answer 32

• Differences in opinion: cannot alone explain the high level of volume.
  
  • Differences in prior beliefs or due to differences in the way that investors interpret public information. “Agree to disagree”.
  • Overconfidence: assumes that investors overestimate the precision of information: “calibration literature”. Overconfident investors trade more than rational.

Question 33

What is the experimental design in Glaser and Weber (2007), Overconfidence and Trading Volume?

Answer 33

• German online broker and a link to an online questionnaire; Response rate of 6,98%.
• Following measures of trading volum:
  
  • Number of stock market transactions, number of stock market purchases, mean monthly stock performance. Sum of the absolute values of purchases and sales per month and divide bu the end-of-month stock portfolio position.
  • Overconfidence affects the expectations of future stock price performance = when selling a security the effect of overconfidence is mixed with reference point dependent decisions behaviour = different analysis of buy transactions. So the effect of overconfidence is stronger when only buying transactions are considered.
• Measures of overconfidence:
  
  • Miscalibration: upper and lower bounds of 90%
  • Stock market forecasts: overconfident investors underestimate the volatility of stock returns.
    
    • Better than average effect.

Question 34

Conclusion on Glaser and Weber (2007), Overconfidence and Trading Volume?

Answer 34

• Overconfidence as measure by calibration questions is not significantly related to trading volume. This result is inconsistent with theory = overconfidence models are almost always by miscalibration.
• But: investors who think that they are above average do trade more, the better-than-average effect explains trading volume = perceived competence.

Question 35

What is Kahneman and Tversky (1974), Judgements Under Uncertainty: Heuristics and Biases about?

Answer 35

• This article discusses how people assess the probability of unvertain events of quantities. People rely on a limited number of heuristic pricniples which reduce the complex task of assessing probabilities to simpler judgemental operations.
  
  • These heuristics are in general quite helpful, but can lead to severe systematic errors.
• Representativeness:
  
  • Probabilities are evaluated by the degree to which A is representative of B = highly representative.
  • Eg. Steve is very shy and withdrawn. People estimate that he is more likely to be a librarian than other possible occupations. People order the occupations by probability and by similarity in the same way.
  • Insensitivity to prior probability of outcomes = neglect of base rate. There are more farmers than librarians but that is neglected.
• Availability
• Adjustment and anchoring

Question 36

What is the article “Testing for salience effects in choices under risk” by Nielsen, Sebald and Sorensen about?

Answer 36

• They construct and run an experiment to isolate and test the most basic choice effect predicted by Salience Theory (Bordalo).
• They vary an apparent payoff ratio to influence salience, treatments have economically equivalent consequences. Most other theories predict zero effect. The experimental findings are strongly consistent with a continuous version of Salience Theory.
• They also introduce novel structural estimates regarding the salience effect. Increasing the relative payoff contrast by one percent is equivalent to an increased odds ratio by 0,4%.

Question 37

What is salience theory about?

Answer 37

• Our cognitive resources are limited, so it can be drawn to salient features of our environment. Bordalo, Gennaioli and Schleifer have developed salience theory to explore the economic consequences of the way salience captures attention.

Question 38

What are the psychophysical concepts of perception?

Answer 38

• Weber-Fechner law:
  
  1. Weber’s principle: it states that dissimilarit between 2 stimuli magnitutes is determined by the ratio of the large magnitute to the low:
     
     • Eg. difference between 11 and 10 is perceived as the same as 22 and 20.
  2. Fechnerian sensitivity: there is diminishing sensitivity to a given difference in stimuli magnitude
     
     • Human perception is more sensititive to a difference in 11 to 10, than 16 to 15.

Question 39

What was the experiment in “Testing for salience effects in choices under risk” by Nielsen, Sebald and Sorensen about?

Answer 39

• Individuals were asked to make simple choices of allocating wealth to a risky investment. The choice problems were designed in a way that they have economically equivalent consequences.
• Majority of theories of choice under risk predict zero treatment effect.
  
  • But: vary an apparent payoff ratio to potentially influence the salience of the risky choice.
  • So salience theory predicts a choice effect.
• The experiment allows for an identification of the diminishing sensitivity effect predicted by salience theory.
• Two treatments provide identical state-dependent wealth consequences to any amount bet on the risky lottery.
• Salience theory predicts that subjects will bet more on the risky lottery in treatment B (Because the relative payoffs to the risky bet differ: bad consequence is more salient in treatment A (1.1/0,4<0,4/1,1) than in B (1,6/0,9>0,9/0,6)
• More salient consequences get assigned larger decision weights in salience theory. In treatment B, the good consequence is more salient and weighted, causing risky betting to seem attractive.

Question 40

What is Malmendier & Tate (2005), CEO Overconfidence and Corporate Investment?

Answer 40

• They argue that managerial overconfidence can account for corporate investment distortions. Overconfident managers overestimate the returns to their investment projects and view external funds as unduly costly.
• Therefore, they overinvest when they have abundant internal funds, but curtail investment when they need external financing.
• They test the overconfidence hypothesis, using panel data on personal portfolio and corporate investment decisions of Forbes 500 CEOs. We classify CEOs as overconfident if they persistently fail to reduce their personal exposure to company-specific risk. We find that investment of overconfident CEO’s is signficiantly responsive to cash flow, particularly in equity-dependent firms.

Question 41

What are the traditional explanations for investment distortions in Malmendier & Tate (2005), CEO Overconfidence and Corporate Investment

Answer 41

1. Misalignment of managerial and shareholders interest: agency theory:
   
   • Private benefits for the manager, but limited in the capital markets so when there is a lot of cash flow, the manager can create more investment distortions.
2. Asymmetric information between corporate insiders and the capital markets.
   
   • Restrict external financing to avoid diluting the (undervalued) shares of their company. Cash flow thus increases investment, but reduces the distortion.

Both cause investment to be sensitive to the amount of cash in the firm.

Question 42

What is Malmendier and Tates explanation for the investment distortion?

Answer 42

• Tension between the beliefs of the CEO and the market about the value of the firm. Overconfident CEOs systematically overestimate the return to their investment projects.
  
  • If they have sufficient internal funds, they overinvest relative to the first-best.
  • If they do not have sufficient funds, they are reluctant to issue new equity because they perceive their stock as being undervalued by the market. So they curb their investment.
• Overconfidence based on the better-than-averae effect: illision of control, high degree of commitment to good outcomes (eg. personal wealth) and abstract reference points (ability to pick profitable investment project is difficult), make it hard to compare performance across individuals.

Question 43

How do Malmendier & Tate (2005), CEO Overconfidence and Corporate Investment measure overconfidence?

Answer 43

• Exploit the overexposure of typical CEOs oto the idiosyncratic risk of their firms = stock options. Because of under-diversification, risk averse CEOs should exercise their options early given a sufficiently high stock price.
  
  • Benchmark for the minimum percentage in the money. If he persistenly exercise later, he is overconfident because he thinks he can keep the company price rising, thus profiting.
  • End of the option’s duraction: if a CEO is optimistic enough, he holds all the options until expiration = overconfident
  • CEOs that habitually increase their holdings of company stock are also overconfident.
• Used on panel data set on the options and stock holdings of CEO’s of 477 large US companies.
  
  • They do not earn significantly over the S&P 500 = they do not have inside information which would make it a good decision to exercise option late.
  • They show that investment-cash flow sensitivity is significantly higher for “late exercisers” or “stock purchasers” than for peers. So overconfident investors invest more when they have cash on their hands.

Question 44

What is the conclusion on Malmendier & Tate (2005), CEO Overconfidence and Corporate Investment?

Answer 44

• Main goal of the paper is to establish the relation between managerial overconfidence and corporate investment decisions.
• 3 mains steps:
  
  1. Simple model of the corporate investment decisions: prediction that the sensitivity of investment to cash flow is strongest in the presence of overconfidence.
  2. 3 measures of overconfidence, if one is yes = overconfident:
     
     1. Does the CEO hold his options beyond a theoretically calibrated benchmark for exercise?
     2. Does the CEO hold his options even until the last year before expiration
     3. Does the CEO habitually buy stock of his company during the first 5 sample years?
  3. Regress investment on cash flow, the overconfidence measure and the interaction between cash flow and overconfidence. Strong postiive relationship between the sensitivity of investment to cash flow and executive overconfidence. Coefficients of the interactions term are highly significant for all out measures. Also find that overconfidence matters more in firms that are equity dependent.
• Conclusion: important implications for contracting practices and organizaitonal design.
  
  • Standard incentives such as stock and option compensation are unlikely to mitigate the detrimental effects of managerial overconfidence.
  • As a result: board of directors might need alternative disciplinary measures (debt overhang).
  • Need for independent and vigilant directors.
    *