Behavioral Economics: decision making & opportunities ? It is all about Free lunches ! “Free Lunches” In Economics there is no free lunch Standard economic theory assumes that all agents (consumers and organizations alike) are rational agents working effectively to maximize their own welfare. Since all actors in the marketplace are assumed to be at a state of maximized returns, it follows that there are no “free lunches.” After all, the world is optimal at its current state “Free Lunches!” In Behavioral Economics there are are all kind of inefﬁciencies, and thus many free lunches The emerging ﬁeld of behavioral economics routinely demonstrates that individuals and organizations do not maximize their own welfare and instead follow suboptimal decision strategies and succumb to different decision traps. While this view of human rationality can be a bit depressing, it also implies that the marketplace is full of free lunches. Differences have implications for: Theory Individuals’ decisions Businesses’ strategies and offerings Policy (taxes, medical insurance, welfare, infrastructure & educational investments, role of EPA, the efﬁciency of free markets etc.) Psychological views on rationality Bounded Rationality: (Herbert Simon, 1957) A human mind has limited information processing and storage capabilities, humans must use simple rules of thumb and heuristics to help make decisions and solve problems. Systematic Cognitive Biases: (Kahneman and Tversky, 1974) Many heuristics, or simple rules, that people use to make judgments and decisions lead to systematic and predictable errors. Visual illusions as a metaphor for DM Prospect theory Amos Tversky Kahneman How a “simple” change in assumption can have far reaching implications Prospect theory Kahneman Daniel & Tversky Amos (1979) “Prospect Theory: An analysis of Decision Under Risk” Econometrica 47 263-291 Kahneman Amos Tversky Expected utility Subjective Value Value Prospect utility Subjective Value So ???? Value Prospect utility Diminishing returns Losses are steeper Pattern changes at the origin V Diminishing sensitivity V Diminishing sensitivity I Buying a calculator for $15 The sales person tells you that you can buy the same calculator for $8 by walking 15 min to the other store. Would you do it? Buying a suit for $1,165 The sales person tells you that you can buy the same suit for $1,158 by walking 15 min to the other store. Would you do it? Diminishing sensitivity II Psychophysics of money Spending more on cars? houses? Spending more on tomatoes? Implications…. Reference point V Investments Ia As the president of an airline company, you have invested $10 million of the company's money into a research project. The purpose was to build a plane that would not be detected by conventional radar,. When the project is 90 percent completed, another ﬁrm begins marketing a plane that cannot be detected by radar. Also, it is apparent that their plane is much faster and far more economical that the plane your company is building. The question is: should you invest the last 10 percent of the research funds to ﬁnish your radar-blank plane? NO - It makes no sense to continue spending money on the project. YES - As long as $10M is already invested, I might as well ﬁnish it. Investments Ib As the president of an airline company, you have invested $1,000 of the company's money into a research project. The purpose was to build a plane that would not be detected by conventional radar,. When the project is 0.01 percent completed, another ﬁrm begins marketing a plane that cannot be detected by radar. Also, it is apparent that their plane is much faster and far more economical that the plane your company is building. The question is: should you invest the last 99.99 percent of the research funds to ﬁnish your radar-blank plane? NO - It makes no sense to continue spending money on the project. YES - As long as $1,000 is already invested, I might as well ﬁnish it. Sunk cost The investment size should not relevant to the decision to continue or not. This is called the sunk cost effect Losses are more painful So we try to eliminate or delay them Thus, invest more because we are already “deep into it” Buying tickets for a BB game You have a ticket for a basketball game. The drive is 60 miles, it snows and the roads are bad. Is the likelihood that you will go to the game different if: A) you paid $10 for the ticket yesterday B) you paid $100 for the ticket yesterday C) you paid $100 for the ticket last year D) you got the ticket for free E) you got the ticket for a discount Diminishes over time (Arkes) Reference points The current state is considered the statusquo Related to defaults Actions are considered as deviations from the status-quo Deviations are losses and gains from that starting point Losses are steeper V Loss aversion I Would you take the following gamble: 50% to get $100 50% to lose $50 50% to get $1,000 50% to lose $500 Loss aversion II Which would you choose A sure gain of $240 25 % to win $1,000 and 75% to win 0 Which would you choose A sure loss of $240 25 % to lose $1,000 and 75% to lose 0 Buyers & sellers Lets trade [X]: Owners Write the minimum price that you will charge to sell your [X] Non-owners (buyer) Write down the maximum price that you are willing to pay for [X] Trades Given that the roles of sellers and buyers were selected randomly, how many exchanges would you predict? The Endowment effect The endowment effect Paying for things you own Basketball tickets you have vs. not Your old car Mugs & chocolate Kids ;) Loss aversion The common ratios of the effects of gains and losses across many studies is 2:1 Loss aversion is the cause for “timid decisions” Particularly when considering decisions one at a time Loss aversion: implications Insurance? Attachment? Other implications? Gains and losses Reference points is the current point Changes are considered relative to the reference point Losses loom larger than gains Loss aversion … Also statues-quo bias Value Lessons from prospect theory I Diminishing sensitivity Loss aversion Reference point Endowment Sunk cost Value Lessons from prospect theory II The standard theory: The carriers of utility are states of wealth, and people are risk-averse in wealth Prospect theory The carriers of utility are changes, gains and losses People are loss averse Accumulating empirical evidence for PT Other applications? Stock market? Equity premium puzzle I The equity premium puzzle Stocks have been outperforming bonds Large difference between rate of return on equities and treasury bills. Yet many people invest in bonds A dollar invested in stocks on 1/1/26, would yield $1800 in 1/1/98 A dollar invested in treasury bills 1/1/26, would yield $15 in 1/1/98 Equity premium puzzle II Variance in returns can cause people to experience loss aversion frequently As a consequence the “pain” of stocks will be higher than the “pain” of bonds 75 60 Bond X Stock Y 45 30 15 Time Overcoming the equity premium If (myopically loss averse) investors evaluated returns about once a year, puzzle would disappear. Personal advice: don’t look often at your portfolio and don’t look at individual stocks Risk taking by professionals Zur Shapira (2000) found that: Traders took riskier positions the following day if they ended the previous day with a loss rather than a proﬁt. Traders also showed a tendency to take ‘long shots’ in the last hour or two of trading at the end of the day if they were in loss at 3 o’clock. Investments by individuals Observation: The propensity to sell winners is twice as large as the propensity to sell losers (except in December) Why? People hold separate ‘mental accounts’ for different stocks For each stock they store the original buying price (which is irrelevant) As a consequence people hold on to “losers” too long The disposition effect: The propensity to hold onto losers has been called by Odean the disposition effect The buying price for each stock is a salient (yet irrelevant) reference price Leading to reluctance to realize the loss by selling “losers” Another aspect of prospect theory The probability weighting function Probability weighting function Associated weights 1.0 0.8 0.6 0.4 0.2 0.0 0 0.2 0.4 0.6 Probability 0.8 1 Probability weighting function 1.0 Associated weights Probability is not linear 0.8 0.6 0.4 0.2 0.0 0 0.2 0.4 0.6 Probability What are some applications of this different pattern? Probability weighting function Associated weights 1.0 0.8 0.6 0.4 0.2 0.0 0 0.2 0.4 0.6 Probability 0.8 1 0.8 1 Probability weighting function Lottery tickets Associated weights 1.0 0.8 0.6 0.4 0.2 0.0 0 0.2 0.4 0.6 Probability I have 100 tickets. One of which will win $1,000. How much would you pay for 1 ticket If you had 49 tickets, how much would you pay for 1 more ticket If you had 99 tickets, how much would you pay for 1 more ticket 0.8 1 The Allais paradox Which would you prefer? $0 G1a $1M $5M 100% G1b 1% 89% 10% G2b 89% 11% 0% G2a 90% 0% 10% The Allais paradox $0 G1a G1b G2b G2a $5M 1% $1M 100% 89% +1% -11% +10% 89% 90% 11% 0% 0% 10% +1% -11% +10% 10% Probability weighting function Market mechanisms? Associated weights 1.0 0.8 0.6 0.4 0.2 0.0 0 0.2 0.4 0.6 Probability What are some market mechanisms that use the shape of the probability weighting function? 0.8 1 “Winning is crucial to my retirement plan” Prospect theory summary 2 aspects: The psychological value function The probability weighting function Prospect theory summary Probability weighting function Subjective Value Value Associated weights 1.0 0.8 0.6 0.4 0.2 0.0 0 0.2 0.4 0.6 Probability 0.8 1 Prospect theory summary The psychological value function Diminishing returns Loss aversion Reference point The probability weighting function Probability counts mostly at the two extremes Behavioral economics: summary People behave in a way that is different from standard economics Sometimes we can model these behaviors and incorporate them into economics Prospect theory captures some of the differences in a simple form, and with many implications