ECON4260 Behavioral Economics. 4 th lecture

Transcription

1 ECON4260 Behavioral Economics 4 th lecture Mental accounting and the Equity premium puzzle You are going to participate in an experiment NO RESEARCH VALUE! Well studied in literature Not sufficient control But experience show: HUGE value TO YOU. Procedure Use session name: xkab15 There can be at most 20 players, so you are encouraged to form pairs. If there all players are taken when you try to log in, try to find a partner who has been able to log in. Use a fake name if you want to hide your identity to me Contrary to instructions: Trading period is 1 week (not 5 minutes) More information on 1

2 2

3 3

4 4

5 5

6 6

7 7

8 Critical summary this far Imperfect probability assessment Use Heuristics, but: Heuristics works very well most of the time Over-weighing low probabilities Important for some lotteries Loss aversion and endowment effect Not if commodity is produced to be sold No endowment-effect for experienced traders So. Does it matter? 8

9 Real rates of return Highly dependent on risk Stock return about 8% Long term bonds about 3.5% USA Treasury Bills (short term) 1% Difference too large to explain with risk aversion One argument: it is loss aversion NOTE: THERE ARE OHTHER THEORIES Habit formation Transaction costs Non-expected utility If it is loss aversion, So what? A discount rate of 8% or 1% for public projects matters a lot. Private and public project may have similar risk But what are the losses in public projects? Should prospect theory be a normative theory How should we then treat the editing? How often should we evaluate public projects? If we should be consistent with EU How do we account for the large deviation from EU in the asset market 9

10 Mental accounting Imagine that you are about to purchase a jacket for ($125)[$15] and a calculator for ($15)[$125]. The calculator salesman informs you that the calculator you wish to buy is on sale for ($10)[$120] at the other branch of the store, located 20 minutes drive away. Would you make the trip to the other store A: Numbers in ( ). Most will make the trip B: Numbers in [ ]. Few will make the trip Both cases save $5 at the cost of a 20 minutes trip. Why do people choose differently in A and B? Mental accounting To simplify decisions we isolate different decisions. Keep separate mental account The calculator purchase is seen isolated We do NOT focus on the global preference question Travelling 20 minutes Versus saving 5 dollars 10

11 Cab-drivers in New York Some days demand is low, with long delay between customers Low hourly wage Other days (rainy days or with problems on the subway) demand is high. High hourly wage Cab-drivers work long hours some days and short hours other days When do you think they work long hours? A: On rainy days B: On days with low demand? Prediction and results Standard theory: max h wh-c(h), [w=c (h)] predict: h (hours worked) is increasing in w (wage) Prediction: long hours on rainy days Empirical finding: Short hours on rainy days Alternative theory: Mental accounting Each day is a separate account Follow rules like: Earn at least $100 per day. When the wage is $10, it takes 10 hours When the wage is $20, it takes only 5 hours Note that 10 hours on the rainy day, and a day off after would yield the same income. 11

12 Transaction vv acquisition utility A hot day on beach. Friend is going to buy a beer from a [small, run-down grocery store] (a fancy resort hotel). What is your maximum willingness to pay? $ 2.65 (hotel) $ 1.50 (store) Transaction utility = Perceived value of deal = Amount paid and reference price. Acquisition utility = value of good relative to price Opening and closing accounts Purchase a stock at price P0. Sold at time t (mental account closed) If Pt<P0, we would close with a loss Utility function is locally convex (risk seeking) Accepting the loss is painful Thus: Keep losers If Pt>P0 Utility function is concave (risk aversion) We can close the account without losses Thus: Sell winners Observe a tendency to keep looser and sell winners Rationality predicts: Optimal to sell losers (tax deductible) 12

13 Successive lotteries Samuelson s colleague Turned down (-100,50%,200,50%) Would accept the same lottery played 100 times as long as he did not have to watch the bet being played out Two such lotteries = (-200,25%;100,50%,400,25%) Consider value function: v(x)=x for x 0 but v(x)=2,5x for x<0. Once: -2.5*100*50%+200*50% = - 25 Twice, watching: (-25) = -50 not watch: -2.5*200*25%+100*50% +400*25% = +25 Equity Premium Puzzle Mehra and Prescott Equity return: 8% (real) Treasury bills: 1 0.5% (real) Consumption is growing, 2% per year Marginal utility declining Risk aversion required, (rra=30) Lottery in future consumption ( , 50%, ,50%) ~ ( ) No one is that risk avers. 13

14 Alternative explanations Nonexpected utility (Kreps and Porteus) Separates declining marginal utility and risk aversion But why do anyone buy treasury bills? Habit Formation (Constantinides) Requires very high (implausible?) degree of habit formation. Explaining the equity premium puzzle How is a stock kept 36 months valued As 36 bets (watching the bets played out) As almost 1000 daily bets (watching) As 12 quarterly bets (watching) As 3 yearly bets (watching) As one bet (or no watching) To explain the eq. prem. paradox, we must assume that it is seen as 3 yearly bets. Benartzi and Thaler argues that this is the most natural. E.g. tax reports are due yearly. 14

15 Why does it matter when mental accounts are closed? Consider an asset held for two years First year yield a nice gain Second year yield a bad -500 loss Evaluated as one account Total gain is good Account closed every year: First year a benefit Second year loss value 2.5(-500)=-1250 Net value -250 Perceived as a bad choice What if this is the correct theory? What interest rate should we use For public project Evaluate environmental policy (climate change) Is Prospect theory a good theory of how we ought to act? What are the losses of climate abatement? Or is expected utility a better Normative theory? Are investors overly reluctant Should government stimulate risk taking? 15

16 Summary: Behavioral decision theory Imperfect probability assessment Over-weighing low probabilities Loss aversion and endowment effect Does it matter? Some think it matter for interest rates Maybe the impact otherwise is small (limited to cabdrives and other odd cases)? But decisions are the basic unit of economic theory, if we get them wrong, more may be at stake. 16