Economics 849 Midterm

Prof. Bryan Caplan

Fall, 2006

Part 1: True, False, and Explain

(10 points each - 2 for the right answer, and 8 for the explanation)

1.  T, F, and Explain:  The "marriage tax" – the extra taxes married couples have to pay compared to unmarried couples – is inconsistent with the Ramsey rule.

FALSE.  "Being married" is also a good that could be taxed under the Ramsey rule, and it seems to be fairly inelastic.  Current policy shows that a lot of people will remain married even if there is a substantial financial cost.  Of course, it is possible – though it seems unlikely - that "being single" is even more inelastically demanded than being married, in which case a Ramsey rule would recommend a tax on being single.  (I gave full credit to students who answered TRUE on the grounds that demand for being married is more elastic than demand for being single).

2.  Suppose there are three voters (#1, #2, and #3) and four spending options (a, b, c, and d).  a is the highest spending level, b the second highest, c the third highest, and d the lowest.  Voters' willingness to pay for each policy is given by the following table:

 a b c d Voter #1 \$1000 \$800 \$400 \$200 Voter #2 \$200 \$400 \$600 \$800 Voter #3 \$800 \$1100 \$1200 \$0

T, F, and Explain:  A election will NOT reveal social intransitivity in this electorate, and the efficient spending level, c, will win.

FALSE.  These preferences are single-peaked, and the median value, c, will win in an election.  However, c is not the efficient outcome.  Adding up the columns shows that the social value of b is \$2300, versus \$2200 for c.

3. T, F, and Explain:  In the Median Voter Model with imperfect information, extremists can move policy in their preferred direction by refusing to vote for moderates.

TRUE.  As long as the extremists drop out conditional on what they perceive to be excessive moderation, they are giving "their" politician an incentive to move somewhat in their direction.  There is however a cost – the less moderate their candidate is, the greater the probability that the other side wins.  With perfect information, the side with more flexible extremists wins for sure; with imperfect information, however, less flexible extremists basically pressure their candidate to choose a more extreme platform with a smaller probability of victory.

4.  Suppose that there is a 3-party parliamentary system that uses proportional representation.  Party A offers better policies than either Party B or Party C.  As a result, informed voters always vote for Party A.

T, F, and Explain:  If the Miracle of Aggregation holds, Party A will run the government as long as at least one-third of voters are informed – even if Parties B and C are willing to form a coalition government.

TRUE.  With three parties, Party A gets 100% of the informed vote (1/3), plus 1/3 of the uninformed vote (1/3*2/3=2/9).  The sum total is 5/9, enough to beat even a coalition of Parties B and C.

(A few students interpreted the question to mean "Party A will NOT run the government if the fraction of informed voters is LESS than one-third," and therefore answered FALSE.  Students who gave this answer and correctly calculated that Party A wins as long as more than 25% of voters are informed got full credit).

5.  T, F, and Explain:  The externalities rationale for redistribution explains why countries redistribute more to their own citizens than foreigners, and more to the elderly than the poor.

FALSE.  The externalities rationale can explain why countries focus on the domestic poor rather than foreigners.  After all, almost all crime and begging in a country is committed by people who are IN that country.  However, the externalities rationale cannot explain why we spend so much money on the old, who are highly unlikely to turn to crime or begging in any case.

(A few students argued that foreign aid might reduce the externalities of terrorism, etc.  I thought this argument was quite implausible – the number of domestic murders over the last ten years is roughly one hundred times as large as the number killed on 9/11/2001.  But I did give partial credit for this answer).

6.  Caplan doubts whether non-profit competition works very well.

T, F, and Explain:  It turns out, however, that federalism can improve the quality of policy that the average citizen lives under even if sub-national governments select their policies randomly.

TRUE.  If governments choose policies randomly, at least there will be some variation in policy.  Variation gives citizens an incentive to move to places where (a) the overall quality of governance is high, and (b) the mix of taxes and services fits well with their particular tastes.  This is true even if sub-national governments make no effort to copy each other.  In the extreme, if one sub-national government happened to have terrible policies, everyone would leave, so no one would actually have to live under these policies, and the effect on citizens would be negligible.

Part 2: Short Essays

(20 points each)

In 6-8 sentences, answer all of the following questions.

1.  In the U.S., most government revenue comes from taxation of labor, the supply of which is highly inelastic.  Discuss TWO simple revenue-neutral tax reforms that would nevertheless be more efficient than the status quo.  Explain your reasoning.

Example 1: Cutting the income tax and increasing the tax on activities like gasoline consumption with negative externalities (such as pollution and congestion).  Even if the DW cost of labor taxation were zero, the DW cost of externalities taxes can be NEGATIVE.

Example 2: Cutting the income tax and increasing the tax on land.  Since labor supply is not perfectly inelastic, but land supply virtually is, and since labor is currently taxed at a lower rate than land in most places, the Ramsey rule recommends a shift from labor taxes to land taxes.

2.  Analyze one example from Bastiat's essay "What Is Seen and Not Seen."  What implicit assumptions does Bastiat make about elasticities?  Is he correct?

Bastiat discusses the alleged social benefits of using taxes to create government jobs.  Assuming the government worker does not produce anything, there is a pure waste:

But when James Goodfellow hands over a hundred sous to a government official to receive no service for it or even to be subjected to inconveniences, it is as if he were to give his money to a thief. It serves no purpose to say that the official will spend these hundred sous for the great profit of our national industry; the more the thief can do with them, the more James Goodfellow could have done with them if he had not met on his way either the extralegal or the legal parasite.

Two assumptions seem implicit in Bastiat's analysis: (1) Overall labor supply is perfectly inelastic, but (2) Labor supply to farming versus government jobs is perfectly elastic.  If overall labor supply were not perfectly inelastic, there would be an additional DW cost on top of the wasteful government spending.  At the same time, unless labor supply for farming versus government jobs were perfectly elastic, the DW cost would be less than it seems.  The bureaucrat who earns 100 sous, for example, might only be able to produce 50 sous as a farmer, so the DW loss to society of his sinecure is only (100-50)=50 sous.

3.  Building on the Tiebout model and Delli Carpini and Keeter's evidence on the distribution of voter information, how would you expect policy in localities with highly educated citizens to differ from policy in localities with poorly educated citizens?  Are there reasons to doubt whether this kind of sorting – and the attendant policy differences - would be stable?  Why or why not?

Since Delli Carpini and Keeter find that educated voters know a lot more than less educated voters, it is reasonable to conclude that – especially in small polities – educated voters will lead to better policy.  Educated voters will have a better idea about who to blame for what, and will be less likely to "accidentally" vote for candidates whose policies they don't like.

At first glance, it seems like sorting-by-education would be unstable.  Less educated voters would at least see the economic consequences of good polices, so they would move to highly-educated districts, which would in turn reduce the median level of voter education and the quality of policy.  But upon reflection, this instability is probably illusory.  High-quality policy would raise property values, which in turn would price less-educated voters out of the market.  As the sign at the ritzy Springfield Heights Promenade (The Simpsons ep. 320) says, "Our prices discriminate because we can't."

(Several students mentioned zoning restrictions as an alternate way to prevent migration from the less-educated.  That might work, too, but is not necessary).