Individual Choice Behavior:

Transcription

1 Individual Choice Behavior: This is a large, sprawling literature, in economics and psychology, much of which is devoted to testing the predictions of different theories of rational behavior, and to documenting deviations from those predictions. The experiments are mostly simple often just carefully crafted questions about hypothetical choices. They have given rise to some vigorous debates, on experimental methodology, on the interpretation of the observed choice behaviors, and on their implications for economics. One of the chief methodological debates concerns the use of hypothetical versus real choices. We ll talk about some series of experiments in which phenomena initially observed in hypothetical choices were reproduced in experimental environments in which subjects made real choices Many of these experiments address questions that arise out of the mathematical structure of expected utility theory and its near relations. So these experiments will also give us a chance to look at the interaction between theory and experiment. Often experiments start as tests of the predictions of a theory of rational behavior, and then move on to become investigations of unpredicted regularities. 1

2 What do we mean by rationality? There are lots of formulations, involving assumptions of different strength. These assumptions have shaped the experiments designed to test them. Do people have fixed preferences? If so (and if not) how do they make decisions? 2

3 Let s do a simple in class experiment. Two thirds of you have just received a windfall gift of a classy, stylish, desirable Stanford pen. It is yours to keep, or sell, or buy another one. Please examine it closely, taking note of its sleek lines and prestigious lettering. Think of yourself taking notes with such an instrument. In what follows, you will be referred to as owners. Those of you did not receive a pen. You will be referred to as non-owners. Will each owner now please pass his/her pen to a neighboring non-owner, so that the non-owners can also fully examine the pen. Because I randomly chose who would become an owner, there may exist some possible gains from trade. In order to assess this, I want to elicit from each owner who got the letter A the minimum price at which he/she would be willing to sell the pen. From each non-owner, I would like to elicit the maximum price she/he would be willing to pay to buy the pen. From owners with the letter B, I would like to elicit the maximum price she/he would be willing to pay to buy the pen. Of course, eliciting this price is a problem that itself presents some challenges to experimental design 3

4 Becker, DeGroot, and Marschak (1964) designed a procedure that would give utility maximizers the incentive to reveal their reservation price. Suppose owners do have stable preferences and care only about maximizing their utility. Suppose we only want to know the value of the pen to owners: Each owner has a true value, a true price, say p, at which she/he rather have the money than the pen, and for any lower amount of money the owner rather keep the pen than receive this amount of money. Procedure: Each owner of the object will be asked to name the amount of money for which he would be willing to sell it, say pa (and we would like this announced pa to be equal to p.) Next a random price pt will be determined. Suppose an owner has pa < pt. Then the owner was willing to sell her/his pen for pa so, she/he is for sure willing to sell for pt. So, every owner with pa pt receives pt and the experimenter takes back the pen. Note that the transaction takes place at pt and not the price announced by the owner (the potential seller of the pen) Every owner who states pa > pt wanted more than pt for the pen, so (claims to) values the pen more than pt and keeps the pen. 4

5 Why do owners have an incentive to announce the truth, that is have pa = p their true willingness to accept to trade price? Suppose an owner states pa > p, that is, an owner who values an item, say, for $20, but announces to only want to sell it for at least $25. If pt, the randomly generated price is higher or equal than $25, the owner could have stated the true willingness to accept. If pt<20, in no case would there have been a transaction. The owner again did not gain from misrepresentation. If 20 pt <25, e.g. pt=23, the owner would actually like to transact and sell the pen for 23, but, by stating pa =25 loses out on this transaction. Similarly the owner cannot gain from stating pa < p. Suppose now the owner announces 15 instead of the true price 20. If pt<15 or 20 pt it makes no difference the owner does not gain from misrepresentation. If 15 pt<20, e.g. pt = 17, the owner has to transact at 17, even though at that price the owner would rather keep the pen. So, the owner has incentives to announce her/his true willingness to accept! That is, to announce the true minimum price at which she/he wants to sell. (Note: probably all owners will be happy to sell for $ , so, we want the lowest price at which they want to sell ) 5

6 Now, suppose we want to elicit the willingness to pay that is for every non-owner the highest possible price at which she/he wants to buy the object. (again, probably many nonowners would be willing to receive a nice object for $0.01 so, we want the highest price at which they want the object, the price at which they are indifferent between the object and the paying the money.) Each non-owner has a true value, a true price, say p, at which she/he rather have the money than the pen, and for any lower amount of money the non-owner wants to buy the pen than keep this amount of money, and fort any higher amount, rather keep the money than take the pen at that price. Procedure that ensures incentive compatibility: Each non-owner of the object will be asked to name the amount of money for which she would be willing to buy it, say pa (and we would like this announced pa to be equal to p.) Next a random price pt will be determined. Suppose an non-owner has pa < pt. Then the non-owner was willing to buy the pen for pa but not pt, so, the non-owner keeps her/his money. Every non-owner who states pt pa wanted the pen for pa, so, wants it for pt and pays pt and receives a pen. Note that the transaction takes place at pt and not the price announced by the owner (the potential seller of the pen) 6

7 Now, we have owners and non-owners. Each owner of the object will be asked to name the amount of money for which he would be willing to sell it. Each nonowner will be asked to name the amount of money for which he would be willing to buy it. Next a random price will be determined. (between $0.05 and $5.95.) Each buyer seller pair will then transact (the buyer will buy the object from the seller at the random price) if and only if the random price is higher than the seller s demand and lower than the buyer s offer. Note that the transaction takes place at the random price, not at the stated willingness to pay or willingness to accept. In case there are not enough owners, or non-owners who want to transact, I will randomly determine who gets to transact. It is a dominant strategy for a utility maximizer to state his true reservation price, since if he overstates or understates it he will miss some desirable selling opportunities or be forced to enter into some undesirable transactions. 7

8 (Note that not only is using the BDM elicitation procedure a part of the design, so is explaining it some theories of rationality could be taken to imply that the explanation is unnecessary.) And of course, if subjects aren t utility maximizers, they may not even have unique reservation prices, so the BDM technique may have unpredictable effects. But because many modern economic theories are based on the assumption that agents are expected utility maximizers, so that it is frequently the case that the predictions of the theory can only be known if the utility functions of the subjects are elicited in an incentive compatible way, this technique has found wide application even in experiments whose primary purpose is not to estimate utility functions. Are there any questions? Question 1: Would everyone please write down a price, and their name, and whether they are owners A (price to sell the pen), owners B (price to buy another pen) or non-owners (price to buy a pen) 8

9 Owner A (WTA) Non-Owner (WTP) Owners B (WTP) 9

10 A version of this experiment with owners and non-owners only was reported in Kahneman, Daniel, Jack L. Knetsch, and Richard H. Thaler 1990, Experimental Tests of the Endowment Effect and the Coase Theorem, JPE, 98, 6, What can account for the price differences between owners and non-owners? Can we learn anything from the stated prices of owners to buy a second pen? 10

11 Extra Homework: Watch: Dan Gilbert TED talk on YouTube: The Experiment we did in class with owners A and nnowners: half the people receive a pen, and state their willingness to accept (WTA), that is, the lowest price at which they want to sell the pen. The other half received nothing, and stated their willingness to pay (WTP), that is, how much they were at most willing to pay for the pen. We found: WTA>WTP Possible problem: People who received the pen: They are richer, and rich people may be more willing to pay for luxury items than poorer people. Let s assume instead of a pen: use BMW cabriolet. Well, maybe then we would not be surprised to find WTA>WTP when we consider a subject pool that is wealth constrained. How can we design an experiment that gets around this explanation (apart from using the design in class) 11

12 The solution of Kahneman, Knetsch and Thaler 1990: Half the participants receive a coffee mug (their item of choice) and state the willingness to sell (average price $7.12). The other half received nothing, but, for each price from a range of 0 to $9.25 were asked whether at this price they would prefer the mug, or the money. In this case, participants have to choose between money or mug, as opposed to choose for what price they want to buy the mug. (so, they were choosers, not buyers as we had in class.) Another possibility is to consider exchanges between two goods, as opposed to trading goods for money. KK&T 1990: 1. class: Receives mug at the beginning of the class and is asked at the end whether they want to exchange it for a bar of Swiss chocolate. 2. class: other way round: receive chocolate and can after the class exchange it for a mug. 3. class: Choose between mug and chocolate bar. Proportion of students selecting the mug: Mug class Choose class Chocolate class 89% 56% 10% 12

13 The endowment effect in the field: Recently, experimenters (most notably John List) have started to do field experiments. John List (Does Market Experience eliminate market anomalies, QJE 2003) went to sports cards show at Orlando and collector pins market constructed by Walt Disney world in Orlando. He receives endowment effects with traders at these shows. Only when he uses dealers with a lot market experience does the endowment effect disappear. 13

14 Incidentally, this experiment (with real choices, controlled incentives, etc.) arose from the observation that, in hypothetical questionnaires, that there was often a big gap between reported WTPs and WTAs e.g. in contingent valuation studies, with questions like these: How much would you pay to eliminate some risk that presently gives you a.001 chance of sudden death over the next five years? How much would you have to be paid to accept an additional.001 chance of sudden death over the next five years? Putting prices on unfamiliar things turns out to be a difficult task, and the results are sensitive to how you re asked. (Dan Ariely has some nice results on this 14

15 Do you have stable Preferences (beliefs?) Everyone please write down: Last 3 digits of social security number. Guess: How many countries are members of the UN? The person who guesses correctly receives $5. 15

16 SSN Number of countries in UN

17 COHERENT ARBITRARINESS : STABLE DEMAND CURVES WITHOUT STABLE PREFERENCES Ariely, Loewenstein and Prelec, QJE A nice demonstration that people may not have a clear idea how much they value and how to price - unfamiliar things. Experiment 1: 55 MBA students: see 6 products (without the market price: on average $70). First: Subjects were asked if they were willing to buy each product for a dollar figure = last 2 digits of their Social security number. Then: subjects stated their Willingness to pay. A random device was used to decide which answer is decisive and for the WTP a BDM procedure was used. 17

18 Even though people seem to have no clue how to price certain items, they seem to know the relative ordering within the categories of wine and computer accessories. Experiment 3: Annoying sounds of 100, 300 and 600 seconds. At the beginning of the experiment, subjects are asked about the last 3 digit number of their social security number. Subjects were then asked whether, hypothetically, they would listen again to the sound they just experienced (for 300 seconds) if they were paid the money amount they had generated from their social security number. In the main part of the experiment, subjects had three opportunities to listen to sounds in exchange for payment. The three different durations were again ordered in either an increasing set (100 seconds, 300 seconds, 600 seconds) or a decreasing set (600 seconds, 300 seconds, 100 seconds). In each trial, after they indicated their WTA, subjects were shown both their own price and the random price drawn from the distribution (which was the distribution used in Experiment 2 but multiplied by 10). If the price set by the subject was higher than the computer s price, subjects continued directly to the next trial. If the price set by the subjects was lower than the computer s price, subjects received the sound and the money associated with it (the amount set by the randomly drawn number), and then continued to the next trial. This process repeated itself 3 times, once for each of the three durations. 18

19 Results: 19

20 Problem with a trade-off between money and annoying stimuli only? Do people know how to trade-off two averse stimuli? Use annoying sounds from before, and ounces of unpleasant liquid (composed of equal parts of Gatorade and Vinegar). After being exposed to both stimuli, subjects were shown 3 containers of unpleasant liquid (1 oz., 2 oz., and 4 oz.) and were asked to please answer the following hypothetical question: Would you prefer the middle size drink or X minutes of the sound Where X=1.5 minutes or X=3 minutes (low and high anchor). Now, participants in the experiment have to choose between the liquid and the unpleasant sound. Depending on the anchor (1 or 3 minutes) people have different trade offs between these two unpleasant things The experimenters ask: For each drink size, what is the highest number of seconds of unpleasant noise, s.t. you prefer the unpleasant noise 20

21 Mean Maximum duration at which subjects prefer Tone to Drink. 21

22 Not only is it difficult for people to know what preferences they have, their preferences may sometimes be unstable because we have troubles remembering, or correctly evaluating an experience. (Kahneman, Daniel, Peter Wakker and Rakesh Sarin, 1997, Back to Bentham? Explorations of Experienced Utility, Quarterly Journal of Economics. And Kahneman et al Psychological Science 1993). When remembering an experience, some people argue that we follow a peak-end rule. That is, the peak and the final unpleasantness of an event, determine (largely) how we evaluate that event. In the Short trial the subject kept one hand in water at 14 C for 60 seconds, after which he was allowed to remove the hand from the water and to dry it with a warm towel. In the Long trial the immersion lasted a total of 90 seconds. Water temperature was kept at 14 for the first 60 seconds, at which point (unbeknownst to the subject) the experimenter caused the temperature of the water to rise gradually from 14 to 15 over the next 30 seconds. Different hands were used for the Short and for the Long trials. Half the subjects experienced the Short trial before the Long one; the sequence was reversed for the other subjects. The trials were separated by seven minutes during which the subject performed an unrelated task, which was resumed after the second trial. 22

23 Seven minutes later, the subject was called in for a third trial, informed that one of the two previous procedures would be repeated exactly, given a choice of whether the first or the second trial should be repeated, and asked to answer several questions about the first two trials. On a discomfort meter from 0 (no pain at all) to 14 (intolerable pain) the 32 participants continuously indicate the pain they experience. 21 subjects record a decline in pain in the warmer water. The long trial had a lower average of peak and end pain than the short trial for those subjects. As expected, a large majority of them (17 of 21) preferred to repeat the Long trial. The other eleven subjects did not indicate a significant decrease of pain as the temperature of the water was raised. For these subjects the Peak-End average was very similar on the Short and on the Long trial. Only five of the eleven subjects preferred the Long trial. The difference between the preferences of the two groups was statistically significant, as was the overall preference for the Long trial (69 percent for the 32 subjects). They did a similar thing with colonoscopies, leaving the colonoscope in place for a minute (extending discomfort!) yielded a significant improvement in the global evaluations of the procedure There are some nice empirical finance paper on biases and their effects in saving and choice of funds by Thaler et al 23

24 Summary: Preferences may not always be fixed and immovable, they may depend on whether we own an object or not. (This implies we may have a hard time, as non-owners, appreciating how much we like something once we own it). While preferences may seem moveable, they are nonetheless coherent, in that more valuable items are valued more. Sometimes remembered experiences may not fully account for the second-by-second experience we have, leading us to choose more actually painful experiences, which we only remember as being less painful. Technical note: Becker-DeGroot-Marschack procedure to know how much individuals value an item, that they either own or do not own. 24