際際滷

際際滷Share a Scribd company logo

Irrationality in Economics

Chris Waggoner
Chris Waggoner

Do people and groups maximize expected utility? A survey of alternative theories, for the non-economist.

1 of 22
Download to read offline
Implications Economic theory informs: anti-trust law contracts law Federal Reserve policy business decisions 鍖nancial markets (traders, arbitrageurs)
Traditional Theory Utility Expected Utility Maximization Calculate how to do this Rationality An approximation, a metaphor, or an ideology?
Cost-Bene鍖t Analysis max E dt 揃 ut (ct ) , given It At t
Some Departures from Homo Economicus Reciprocity Calculational Capacity Heuristics Absolute Value vs Relative Value Probability
Some Departures from Homo Economicus Reciprocity Calculational Capacity Heuristics Absolute Value vs Relative Value Probability
Probability Reasonable guess: if people who know exactly whats coming maximize utility, maybe people who can guess whats coming maximize expected utility. People have a very good inborn heuristic for averaging. They can do it in a blink. Plus the mathematics of probability are already worked out (Boltzmann, Kolmogorov), so economists can hypothesize about behavior without having to invent new mathematical tools.
Expected Utility Say an agent is choosing what to do in the face of uncertainty. Option X might result in one thing, or it might do something else. Same for Options Y and Z. Example: Messrs. X,Y, and Z are guys you could date. Of course there are a variety of possible outcomes associated with each of these choices, but lets just reduce it to: {break up after 1 month, break up after 1 year, break up after 10 years, die together}
Probability That turned out to be insuf鍖cient to describe peoples behavior. (After all, people do play the lottery and buy insurance, which on average is a losing proposition.) The next guess was that people care about variance as well as mean. (Chalkboard: pictures of distributions.)
Risk Attitudes Since this persons utility is proportional to wealth, they weigh the 鍖ftieth dollar the same as the millionth dollar. For this person, having only a few dollars will be extremely painful: they cant risk it!
Risk Attitudes
Uncertainty Aversion Say there are two classes you are considering enrolling in. You want to know how good the teacher is, so you go to a website that rates teachers. 1. One of the teachers, you cannot 鍖nd any information about. 2. The other teacher, the reviews are half positive, half negative. Most people would prefer the second teacher because they have more information, hence less uncertainty. However, both situations would be represented probabilistically as uniformly random, i.e. completely random. That means risk aversion does not suf鍖ce to describe preferences.
Ellsberg Paradox Suppose you have an urn containing 30 red balls and 60 other balls that are either blue or black. You don't know how many black or blue balls there are, just that the total number of black balls plus the total number of blue balls equals 60. You are now given a choice between two gambles: Gamble A You receive $100 if you draw a red ball Gamble B You receive $100 if you draw a black ball A B Also you are given the choice between these two gambles (different draw, same urn): Gamble C You receive $100 if you draw a red or blue ball Gamble D You receive $100 if you draw a black or blue ball C D According to expected utility theory, you should prefer Gamble A to Gamble B if, and only if, you believe that drawing a red ball is more likely than drawing a black ball. Similarly you should prefer Gamble C to Gamble D if, and only if, you believe that drawing a red or yellow ball is more likely If drawing a red ball is more likely than than drawing a black or blue ball. drawing a black ball, then drawing a red or blue ball is also more likely than drawing a black or blue ball.
Allais Paradox 1A 1B 2A 2B Both gambles give the same outcome 89% of the time, so in expected utility, these equal outcomes should not affect the desirability of the gamble. If the 89% common consequence is disregarded, both gambles offer the same choice; a 10% chance of getting $5 million and 1% chance of getting nothing as against an 11% chance of getting $1 million.
Some Departures from Homo Economicus Reciprocity Calculational Capacity Heuristics Absolute Value vs Relative Value Probability
Absolute Wealth A friend of mine works twelve hours a week at a $6/hour job. = $3600/year Thats not a lot, right? source: globalrichlist.com
from Myths of Rich & Poor: Why We're Better Off than We Think by W. Michael Cox and Richard Alm, New York: Basic Books, 248 pages, $25.00 A half gallon of milk cost the average worker 10 minutes of labor in 1970, 8.7 minutes in 1980, and only 7 minutes in the latest year for which data are available. A gallon of gasoline cost 11 minutes in 1950 but is now less than half that. But these are nothing compared to some price drops. A scratchy-sounding three-minute phone coast-to-coast phone call cost an incredible 90 hours of work back in 1910. Today, it's less than two minutes of work time. A hundred kilowatt hours of electricity in 1900 cost a shocking 107 hours of worker time in 1900, a bit over an hour by 1960, and less than 45 minutes today. quot;A typical American at the turn of the century spent $76 out of every $100 on food, clothing, and shelter. By the 1990s, this portion had fallen to $37 of every $100.
from Myths of Rich & Poor: Why We're Better Off than We Think by W. Michael Cox and Richard Alm, New York: Basic Books, 248 pages, $25.00 Economists point: See, were doing better. Psychologists point: So why dont we appreciate it?
So why do people making $90,000 a year still feel poor? (Im just an accountant.) Comparison to peers (rank) (mate choice) Hedonic treadmill Dont accurately predict what will make them happy, so they spend on things that dont actually make them better off. Happiness log(income)
Prospect Theory Reference point, not absolute Gains are valued, less than losses are hated Small probabilities overweighted; large probabilities underweighted
Implications GDP growth less important than stability, jobs Low, steady in鍖ation rather than stable prices Precautionary principle Downside risk rather than standard deviation in investments Libertarian paternalism (Thalers) Better theoretical basis for government regulation Buy more experience goods, fewer durables
Ef鍖cient Markets? WSJ.com - As Two Economists Debate Markets, The Tide Shifts Page 1 of 5 FORMAT FO R P'tneyBowes PRINTIN G tilt, spor' se red by October 18, 200 4 Stock Characters DOW JONES REPRINTS! m p This copy is for your personal , As Two Economist s non-commercial use only. To orde r Debate Markets, presentation-ready copies fo r distribution to your colleagues , clients or customers, use the Orde r The Tide Shift s Reprints tool at the bottom of an y article or visit: Belief in Efficient Valuatio n www.djreprints .com . Yields Ground to Role See a sample reprint in PDF Of Irrational Investors format . Order a reprint of this article now . Mr. Thaler Takes On Mr . Fama By JON E. HILSENRATH Staff Reporter of THE WALL STREET JOURNAL October 18, 2004 ; Page A I For forty years, economist Eugene Fama argued that financial markets were highly efficient i n reflecting the underlying value of stocks . His long-time intellectual nemesis, Richard Thaler, a member of the quot;behavioristquot; school of economic thought, contended that markets can veer off course when individuals make stupid decisions . In May, 116 eminent economists and business executives gathered at the University of Chicago Graduate School of Business for a conference in Mr . Fama's honor . There, Mr. Fama surprised some in the audience . A paper he presented, co-authored with a colleague, made the case tha t poorly informed investors could theoretically lead the market astray. Stock prices, the paper said , could become quot;somewhat irrational . quot; Coming from the 65-year-old Mr . Fama, the intellectual father of the theory known as the quot;efficient-market hypothesis,quot; it struck some as an unexpecte d concession. For years, efficient market theories were dominant, but her e was a suggestion that the behaviorists' ideas had become mainstream . e
Current Events Credit crunch illiquid markets Loss-averse traders Uncertainty-averse traders

More Related Content

Irrationality in Economics

  • 1. Implications Economic theory informs: anti-trust law contracts law Federal Reserve policy business decisions 鍖nancial markets (traders, arbitrageurs)
  • 2. Traditional Theory Utility Expected Utility Maximization Calculate how to do this Rationality An approximation, a metaphor, or an ideology?
  • 3. Cost-Bene鍖t Analysis max E dt 揃 ut (ct ) , given It At t
  • 4. Some Departures from Homo Economicus Reciprocity Calculational Capacity Heuristics Absolute Value vs Relative Value Probability
  • 5. Some Departures from Homo Economicus Reciprocity Calculational Capacity Heuristics Absolute Value vs Relative Value Probability
  • 6. Probability Reasonable guess: if people who know exactly whats coming maximize utility, maybe people who can guess whats coming maximize expected utility. People have a very good inborn heuristic for averaging. They can do it in a blink. Plus the mathematics of probability are already worked out (Boltzmann, Kolmogorov), so economists can hypothesize about behavior without having to invent new mathematical tools.
  • 7. Expected Utility Say an agent is choosing what to do in the face of uncertainty. Option X might result in one thing, or it might do something else. Same for Options Y and Z. Example: Messrs. X,Y, and Z are guys you could date. Of course there are a variety of possible outcomes associated with each of these choices, but lets just reduce it to: {break up after 1 month, break up after 1 year, break up after 10 years, die together}
  • 8. Probability That turned out to be insuf鍖cient to describe peoples behavior. (After all, people do play the lottery and buy insurance, which on average is a losing proposition.) The next guess was that people care about variance as well as mean. (Chalkboard: pictures of distributions.)
  • 9. Risk Attitudes Since this persons utility is proportional to wealth, they weigh the 鍖ftieth dollar the same as the millionth dollar. For this person, having only a few dollars will be extremely painful: they cant risk it!
  • 11. Uncertainty Aversion Say there are two classes you are considering enrolling in. You want to know how good the teacher is, so you go to a website that rates teachers. 1. One of the teachers, you cannot 鍖nd any information about. 2. The other teacher, the reviews are half positive, half negative. Most people would prefer the second teacher because they have more information, hence less uncertainty. However, both situations would be represented probabilistically as uniformly random, i.e. completely random. That means risk aversion does not suf鍖ce to describe preferences.
  • 12. Ellsberg Paradox Suppose you have an urn containing 30 red balls and 60 other balls that are either blue or black. You don't know how many black or blue balls there are, just that the total number of black balls plus the total number of blue balls equals 60. You are now given a choice between two gambles: Gamble A You receive $100 if you draw a red ball Gamble B You receive $100 if you draw a black ball A B Also you are given the choice between these two gambles (different draw, same urn): Gamble C You receive $100 if you draw a red or blue ball Gamble D You receive $100 if you draw a black or blue ball C D According to expected utility theory, you should prefer Gamble A to Gamble B if, and only if, you believe that drawing a red ball is more likely than drawing a black ball. Similarly you should prefer Gamble C to Gamble D if, and only if, you believe that drawing a red or yellow ball is more likely If drawing a red ball is more likely than than drawing a black or blue ball. drawing a black ball, then drawing a red or blue ball is also more likely than drawing a black or blue ball.
  • 13. Allais Paradox 1A 1B 2A 2B Both gambles give the same outcome 89% of the time, so in expected utility, these equal outcomes should not affect the desirability of the gamble. If the 89% common consequence is disregarded, both gambles offer the same choice; a 10% chance of getting $5 million and 1% chance of getting nothing as against an 11% chance of getting $1 million.
  • 14. Some Departures from Homo Economicus Reciprocity Calculational Capacity Heuristics Absolute Value vs Relative Value Probability
  • 15. Absolute Wealth A friend of mine works twelve hours a week at a $6/hour job. = $3600/year Thats not a lot, right? source: globalrichlist.com
  • 16. from Myths of Rich & Poor: Why We're Better Off than We Think by W. Michael Cox and Richard Alm, New York: Basic Books, 248 pages, $25.00 A half gallon of milk cost the average worker 10 minutes of labor in 1970, 8.7 minutes in 1980, and only 7 minutes in the latest year for which data are available. A gallon of gasoline cost 11 minutes in 1950 but is now less than half that. But these are nothing compared to some price drops. A scratchy-sounding three-minute phone coast-to-coast phone call cost an incredible 90 hours of work back in 1910. Today, it's less than two minutes of work time. A hundred kilowatt hours of electricity in 1900 cost a shocking 107 hours of worker time in 1900, a bit over an hour by 1960, and less than 45 minutes today. quot;A typical American at the turn of the century spent $76 out of every $100 on food, clothing, and shelter. By the 1990s, this portion had fallen to $37 of every $100.
  • 17. from Myths of Rich & Poor: Why We're Better Off than We Think by W. Michael Cox and Richard Alm, New York: Basic Books, 248 pages, $25.00 Economists point: See, were doing better. Psychologists point: So why dont we appreciate it?
  • 18. So why do people making $90,000 a year still feel poor? (Im just an accountant.) Comparison to peers (rank) (mate choice) Hedonic treadmill Dont accurately predict what will make them happy, so they spend on things that dont actually make them better off. Happiness log(income)
  • 19. Prospect Theory Reference point, not absolute Gains are valued, less than losses are hated Small probabilities overweighted; large probabilities underweighted
  • 20. Implications GDP growth less important than stability, jobs Low, steady in鍖ation rather than stable prices Precautionary principle Downside risk rather than standard deviation in investments Libertarian paternalism (Thalers) Better theoretical basis for government regulation Buy more experience goods, fewer durables
  • 21. Ef鍖cient Markets? WSJ.com - As Two Economists Debate Markets, The Tide Shifts Page 1 of 5 FORMAT FO R P'tneyBowes PRINTIN G tilt, spor' se red by October 18, 200 4 Stock Characters DOW JONES REPRINTS! m p This copy is for your personal , As Two Economist s non-commercial use only. To orde r Debate Markets, presentation-ready copies fo r distribution to your colleagues , clients or customers, use the Orde r The Tide Shift s Reprints tool at the bottom of an y article or visit: Belief in Efficient Valuatio n www.djreprints .com . Yields Ground to Role See a sample reprint in PDF Of Irrational Investors format . Order a reprint of this article now . Mr. Thaler Takes On Mr . Fama By JON E. HILSENRATH Staff Reporter of THE WALL STREET JOURNAL October 18, 2004 ; Page A I For forty years, economist Eugene Fama argued that financial markets were highly efficient i n reflecting the underlying value of stocks . His long-time intellectual nemesis, Richard Thaler, a member of the quot;behavioristquot; school of economic thought, contended that markets can veer off course when individuals make stupid decisions . In May, 116 eminent economists and business executives gathered at the University of Chicago Graduate School of Business for a conference in Mr . Fama's honor . There, Mr. Fama surprised some in the audience . A paper he presented, co-authored with a colleague, made the case tha t poorly informed investors could theoretically lead the market astray. Stock prices, the paper said , could become quot;somewhat irrational . quot; Coming from the 65-year-old Mr . Fama, the intellectual father of the theory known as the quot;efficient-market hypothesis,quot; it struck some as an unexpecte d concession. For years, efficient market theories were dominant, but her e was a suggestion that the behaviorists' ideas had become mainstream . e
  • 22. Current Events Credit crunch illiquid markets Loss-averse traders Uncertainty-averse traders