Economics 370 Midterm #2 Answers

Prof. Bryan Caplan

Fall, 1998

Mean: 61.5

Approximate Grades: 60-100 A; 45-59 B; 35-44 C.

Part 1: True, False, and Explain

(10 points each - 3 for the right answer, and 7 for the explanation)

State whether each of the following six propositions is true or false. In 2-3 sentences, explain why.

1. Suppose that abolishing the EPA and OSHA would permanently raise the economic growth rate of the U.S. from 1% to 3%.

T, F, and Explain: 50 years after implementing this policy, the total output of the U.S. economy will be 100% greater than it otherwise would have been.

FALSE. As can be seen from the standard formula for compound interest, growing 2% faster for 50 years leaves you more than (2%*50)=100% wealthier. (An exact computation was unnecessary, but if your initial income was $1000, then 50 years of 1% growth leaves you with $1645 vs. $4384 at 3%. It should be clear that $4384 is more than double $1645).


2. Suppose that in the market for movies, there is no duplication effect.

T, F, and Explain: Banning discounted tickets for children and seniors will push the number of movies further away from the optimal level.

TRUE. With no duplication effect, the number of different movies is already too low because it is impossible to perfectly price discriminate. Banning existing price discrimination (like discounted tickets) makes a bad situation worse: movie variety falls even further below its optimal level.



3. New York Cityís mayor implements new rules for New York City buildings inspectors; the main provision of these rules imposes much harsher sentences upon corrupt inspectors who take bribes from landlords to "look the other way" at violations of the building codes.

T, F, and Explain: This new rule will improve the quality of NYC housing, and reduce the time and effort it takes someone to find a new apartment.

The answer I had in mind was:

FALSE. While the rule will raise the quality of rent-controlled housing, it will also make the housing shortage worse. When price controls exist, the market can still clear if housing quality falls enough; forcing the quality level up therefore makes the housing shortage worse, increasing the time and effort it takes a person to find a new place.

However, several people thought that "new" housing was newly built housing not subject to rent control. They therefore answered as follows, for which you could get full credit:

TRUE. Tightening quality standards worsens the shortage and induces more building owners to abandon their buildings. The result is higher quality housing, and an increase in the availability of newly-built apartments.


4. Ireland recently legalized divorce.

T, F, and Explain: Search theory predicts that Irish couples will now on average be younger on their wedding day.

TRUE. You search for a spouse until the expected marginal benefit equals the marginal cost. If you can divorce a spouse who turns out to be a bad match, the MB of search falls, so you do less searching. This enables people to marry sooner since they donít have to be quite as sure about the quality of the person they marry.



5. An auto insurance company only sells policies with large deductibles, but does not put a pre-existing condition clause into the insurance contract.

T, F, and Explain: In the auto insurance industry, moral hazard is probably a serious problem, but adverse selection is not.

TRUE. The use of large deductibles indicates that fully insured people take a lot more risks; the absence of a pre-existing condition clause indicates that adverse selection is not a serious problem in the industry because no steps have been taken to control it. If adverse selection were a serious problem in the industry, but one firm took no precautions to prevent it, it would find that virtually every person that no one else will insure would buy their policy.

(Most people said false, arguing that deductibles had eliminated the moral hazard problem, and the absence of pre-existing condition clauses left the adverse selection problem unsolved. At least the second claim requires that the auto insurance firm be really stupid Ė adverse selection is a problem, but they do nothing to prevent it? And the first claim ("moral hazard isnít a problem; deductibles solve it") is a rather strange interpretation: If you surrounded your home with an electrified fence, would you say "Crime isnít a problem around here. My electrified fence saw to that"? I would think you would say that because there is a serious crime problem, you put up the electrified fence.)


6. "When [my friend Ami Glazer] found a secondhand car he wanted, he asked the dealer if, for an additional payment, he would provide a one-year warranty. When the dealer refused, Ami went to another dealer. At last he found one willing to sell a suitable car with a warrantee. ĎAll right,í Ami said, ĎIíll take the car. I donít want the warrantee.í" (David Friedman, Hidden Order)

T, F, and Explain: If all customers thought like Ami did, used car dealers would have to raise the price of their warrantees to signal quality.

FALSE. If all customers thought like Ami did, warrantees on their own wouldnít signal anything: if Ė like Ami - no one actually buys the warrantees, then it would be costless to offer them! To signal quality, you would need to offer cheap warrantees that even someone like Ami would consider to be a good deal.

You only received partial credit if you answered true, on the grounds that if everyone asked for a warrantee, sellers of crummy cars would have to charge a lot for warrantees to stay in business. This missed the deeper point of the question: Ami did more than just ask for a warrantee; he asked for a warrantee he didnít intend to buy!










Part 2: Short Answer

(20 points each)

In 4-6 sentences, answer both of the following questions.

  1. Landsburg ("People Wanted") concludes that if anything the number of people on earth is too small. Analyze his arguments using the conceptual tools of optimal variety. In particular, which of his arguments suggest that the "non-appropriability effect" is important? Which of his arguments, if any, suggest that the "duplication effect" is important?
  2. The problem of the non-appropriability effect is that a producer canít capture the full surplus that the existence of a product creates (to do that, you need perfect price discrimination). Now with a person, the problem is that the parents can barely capture ANY of the surplus! For starters, your parents canít charge you for giving birth to YOU. It is true that parents get a lot of joy from their kids, but their canít charge you for YOUR joy of being alive at all. Similarly, your parents canít charge your friends, your future spouse, or people who enjoy the same products as you (if you are below the minimum efficient scale, more demand makes ACs lower) for the benefits of your existence. This is non-appropriability on a huge scale. It is much harder to think of a duplication effect of people. Thus: big non-appropriability effect, no clear duplication effect; a different approach replicates Landsburgís conclusion that the number of people is suboptimal.


  3. During an exam, you find yourself spending a lot of time on one question. When does search theory say you should skip the question? When does search theory say you should keep working even though time is running out? Empirically, do you think that studentsí behavior usually fits the predictions of search theory? Why or why not?

Search theory says you should skip the question if the expected marginal benefit of another minute of time is less than the marginal cost of the time. You should keep working of the expected marginal benefit exceeds the marginal cost of the time. Benefits and costs could be measured in terms of exam points or Ė as usual Ė in dollars (although remember that since test time is rationed, you probably value a minute of test time more than a normal minute; and you value a minute of time more if your test is taking longer than expected). Empirically, search theory usually works, but occasionally Iíve noticed students just get frustrated or flustered and leave a bunch of easy questions blank.