Economics explored through Freakonomics What is assessable Incentives

Economics explored through Freakonomics What is assessable? • Incentives matter – Importance of incentives – Conflicting incentives – owners versus labour • Correlation versus causation • Supply, demand equilibrium price • Effect of technological change on the market for goods and services – Shifts in supply

Incentives • Fundamental economic principle: people respond to incentives • An incentive is a benefit, reward, or cost that motivates an economic action. • Types of incentives: – Financial – monetary gain – Moral – it is the right thing to do – Natural or social – curiosity (discover the consequences of an action), shame – Coercive – failure to act in a particular way will result in force being used against oneself

Incentives in Freakonomics 4 case studies: • Day care centres • School teachers in Chicago • Sumo wrestling in Japan • Paul Feldman’s bagel business

Day-Care Centre 1/16 • Issue: parents late to pick child up from day care centre. • Suggested solution: fine parents $3 if 10 minutes or more late. • What happened? – Late incidents per week increased from 8 to 20!!!

Day-Care Centre What went wrong? • $3 is relatively small – indicates that late pickups not a problem. • The program substituted an economic incentive ($3 penalty) for a moral incentive (guilt parents were supposed to feel when they came late). • Parents could buy off their guilt for $3! 5 / 19

School Teachers • Standardised testing, such as NAPLAN, has become mandatory in schools. • In the USA there is a law called “No Child Left Behind” - it includes “high-stakes testing” 8 / 22 – Awards schools that make progress on these tests (with funding; teacher may be praised, promoted, given bonuses) – Punishes schools that chronically lag behind the rest of the nation (with probation and withholding funding; teachers may be fired) – NB: this punishment scheme does not operate in Australia.

School Teachers • So what incentives exist so far? – Financial for both schools and teachers (CA introduced bonuses of $25, 000!) – Coercive (probation / dismissed) – Natural (teacher cheating is rarely detected)

How did the teachers cheat? Write scores on the board Extra time for students Teach to the test Instruct students to fill in blank questions at the end • Erase wrong answers and fill in correct answers • • 9 / 23

How can you catch a cheater? Think like one! • Identify what incentive they have to cheat – Young / inexperienced – Not concerned about students’ growth into the next year – Lowest scoring classrooms • Identify what incentive they have to not cheat – Concerned about students’ learning growth – Experienced – don’t need to cheat for students to get good scores

Catching the cheaters 11 / 25 • Freakonomics’ authors designed an algorithm to identify the cheaters. • Searching for unusual answer patterns – Blocks of identical answers in a class, especially among harder questions – All the tricky questions were answered correctly – Students test scores spiked in one year and dropped the next (but if they spiked and continued to improve the next year, it is likely a good teacher not a cheater)

What does a cheating teacher look like? • Cheating Teacher: – Students have dramatic spikes in test scores – Students get more than 6 answers correct in a row • Good Teacher: – Easier questions answered correctly by majority of students – Student learning gains carry over into the next year

How do we increase the incentive to 18 / 32 not cheat? • Arnie Duncan, then CEO of Chicago Public Schools, organised a re-test: – Control group = best teachers – >50% = classrooms with teachers suspected of teaching – CPS officials administered the test, not teachers • Result? – No cheating in control classrooms (scores same or increased, in the 3 week test gap) – Suspected cheaters (far worse, by avg of more than 1 full grade level!)

How do we increase the incentive to 18 / 32 not cheat? • Arnie penalised the cheaters • Set a precedent for being tough on cheaters by firing 12 cheating teachers. • Final outcome: the following year, cheating by teachers fell more than 30%.

Key Takeaways (1) Economics is, at its root, the study of incentives: how people get what they want or need, especially when other people want or need the same thing. (2) A slight tweak of incentives can produce drastic and often unforeseen results.

1. Which of the following statements best captures the essence of economic analysis? • a) No pain, no gain. • b) Everything has a price. • c) Incentives matter. • d) Scarcity is only relative.

2. Which of the following is not an example of an economic incentive? • a) A factory is required to pay a fine for each unit of pollution it produces in excess of the legal limit. • b) A tax is imposed on each bottle of French wine imported into the country. • c) A jewellery salesman is paid on commission. • d) Posting the pictures of convicted prostitutes on the Internet.

3. From an economic perspective, the incentive to cheat will be greatest when: • a) the total benefits are as large as possible. • b) the expected benefits exceed the expected costs by the greatest amount. • c) the marginal expected costs of cheating are zero. • d) the cheater thinks he is unlikely to get caught.

4. From a teacher’s perspective, the knowledge that teacher cheating is rarely looked for and hardly ever punished has the effect of: • a) reducing the expected cost of cheating. • b) increasing the economic benefits of cheating. • c) increasing the moral benefits of cheating. • d) reducing the net benefits of cheating.

5. In which of the following scenarios would cheating in a sumo match be most likely to occur? • a) Wrestler A has a 6 -8 record and wrestler B has an 8 -6 record. • b) Wrestler A has a 7 -7 record and wrestler B has a 10 -4 record. • c) Wrestler A has an 8 -6 record and wrestler B has a 7 -7 record. • d) Wrestler A has a 7 -7 record and wrestler B has a 7 -7 record.

6. Referring to the previous question, cheating most likely occurs in the scenario you selected because: • a) the gains to the winner outweigh the losses to the loser by the greatest amount. • b) the marginal gains and losses are equal. • c) the combined total benefits are greatest. • d) the combined total losses are lowest.

7. What incentive would a sumo wrestler with a winning record in a tournament have to help an opponent who is “on the bubble”? • a) The good feeling he gets from helping his fellow man, i. e. , a moral incentive. • b) The knowledge that the opponent would do the same thing for him in the next tournament, which translates into one less win he has worry about the next time around. • c) By losing an occasional match, the wrestler with the winning record can lull future opponents into believing that he is not that good. • d) A sumo wrestler with a winning record in a tournament has no incentive to help an opponent who is “on the bubble. ”

8. Using economic principles, how can we explain why a particular individual would be more likely to steal from someone they don’t know than they would be to steal from someone they know and like? • a) The cost the thief incurs stealing from the friend is greater than the cost he incurs stealing from the stranger. • b) The marginal cost the thief incurs stealing from the friend is less than the marginal cost he incurs stealing from the stranger. • c) The benefit the thief incurs stealing from the friend is greater than the benefit he incurs stealing from the stranger. • d) There is no difference. Someone who steals is just likely to steal from one person as another, regardless of whether they know and like the person.

9. In the course of analyzing the data from his bagel business, Paul Feldman discovered that executives at one business were more likely than the lower level administrators to cheat. This suggests that: • a) the argument that as income increases so does willingness to pay is wrong. • b) executives are more likely to be criminals, all else constant. • c) willingness to pay is determined by more than simply one’s income level. • d) executives get promoted to their current position precisely because they are willing to cheat to get what they want.

Briefly describe what we mean by the word “incentive” as it is used in economic analysis. Give examples to illustrate your answer. • An incentive is a benefit, reward, or cost that motivates an action. • Examples: – Parking fines – Employee bonus based on productivity

How does a moral incentive differ from an economic incentive? • A moral incentive appeals to one’s conscience while an economic incentive appeals to one’s financial or economic well being.

Referring to the example in the book regarding the day-care center and the problem of parents picking up their children late, why did charging a fee to parents for picking up children late cause parental tardiness to increase? • Charging a fee for tardiness had the effect of replacing a moral incentive, i. e. , it’s not polite to be late, with a small economic incentive, i. e. , a fine. • In effect, the fee (fine) placed a price on being late. • Because the fee was relatively low, and obviously lower than the “cost” incurred when the parent was considered impolite, more parents were willing to pay it.

According to Freakonomics, economics is the study of incentives. Explain this statement. • Economists study, among other things, the choices people, e. g. , consumers and firms, make. They make these choices based on the incentives they face. • Therefore, to understand incentives is to better understand an individual’s choices.

Does a string of correct answers on a test, duplicated by several students in the same class, prove the teacher is cheating? Why or why not? • No. A string of the same answers may be evidence of a cheating teacher, but not proof. There are other possible explanations for this. • For example, it could simply be coincidence. Alternatively, it could be that all of the questions involved are easy and the students are all exceptionally bright.

Why did some Chicago public school teachers cheat to increase their students’ test scores? • Teachers are rewarded or punished depending on how their students perform on standardized tests. • This reward/punishment structure in turn created incentives for teachers to cheat. • Some teachers responded to those incentives by cheating. Others did not.

Briefly, how did Levitt arrive at the conclusion that some sumo wrestlers were cheating? • Given an individual wrestler’s records, he predicted whether the wrestler would win the next match. • He did this for wrestlers who were “on the bubble. ” • He concluded that cheating was likely when a wrestler won with significantly greater frequency than his record would indicate.

Essay # 1 • Explain, using economic principles, why Levitt expected cheating to be more likely when a wrestler on the bubble, i. e. , with a 7 -7 record, was competing against a wrestler with an 8 -6 record than when the same wrestler was competing with someone with a 10 -4 record.

Essay # 2 • Referring to Paul Feldman’s bagel business, use economic principles to explain why cheating would be more likely in an office with a few hundred employees than it would be in an office with a few dozen employees.

Use your time wisely! • Your exam includes MCQ, short answer and essay questions. • The questions provided to you are similar, but not necessarily identical to those that you will be tested on – use them as practice! • You WILL need to have an understanding of the case studies in Freakonomics: day care centre, cheating teachers, sumo wrestlerss, the bagel business.