Economics 849 Midterm
Answer Key

Prof. Bryan Caplan

Fall, 2003

__Part 1:__ True,
False, and Explain

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

State whether each of the following six propositions is true
or false. In 2-3 sentences (and
clearly-labeled ** diagrams**,
when helpful), explain why.

1. "Suppose that getting to *The Armchair Economist*

__T, F, and Explain:__
**Landsburg concludes that price
discrimination of this sort is inefficient. **

FALSE. Landsburg concludes that this price
discrimination will be efficient *if*
individuals maximize Kaldor-Hicks efficiency.
In that case, the traveler will buy the ticket, and transfer $800 to the
airline by giving up $700 in surplus for themselves. If they follow their own self-interest,
however, the outcome of the airline's pricing policy will have a deadweight
cost of $100.

2. Suppose that there are 25,000,000 voters other than
yourself. There is a 99% chance that the
electorate is *not *evenly
divided. In that case, p=.51. But there is a 1% chance that the electorate *is* evenly divided (p=.5).

__T, F, and Explain:__
**The overall probability of
casting the decisive vote is, as usual, less than one-in-a-million.**

FALSE. The probability of casting a decisive vote is
given by:

_{}

This simplifies to:

_{}

3. Suppose the government currently taxes two goods, labor and movies. The supply of movies is much more elastic than the supply for labor.

__T, F, and Explain:__
**Under the Ramsey Rule for optimal
taxation, if both elasticities fall (i.e., become more inelastic), then
the taxes on both goods will rise.**

FALSE. The Ramsey Rule tells us the optimal way to
raise a GIVEN quantity of revenue; it prescribes *relative* tax rates, not absolute tax rates. If both elasticities fall by the same
percent, then, tax rates will not change at all.

I gave partial credit for answers that pointed out that the Ramsey rule is about demand elasticities, not supply elasticities. That is not true in more complex versions of the rule, but I did create that impression in the notes.

4. 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' *preference* rankings are as follows:

Voter #1: _{}

Voter #2: _{}

Voter #3: _{}

__T, F, and Explain:__
**A election will NOT reveal social intransitivity in this electorate,
and spending level c will win.**

TRUE. Graphing the preferences shows that these
preferences are single-peaked: Voter #1's peak at *a*, #2's at *d*, and #3's at
*c*.
We can therefore apply the Median Voter Theorem, which reveals that *c* is the median.

5. Delli Carpini and Keeter's measure of education ranges from 1 (no high school) to 6 (graduate school). Their measure of income ranges from 1 (<$10,000/year) to 5 ($50,000+).

__T, F, and Explain:__
**Their regression results (Table
4.1) for knowledge of political Substance show that the male-female knowledge
gap is comparable to more than 2 steps of education, but less than 5 steps of
income.**

FALSE. From Table 4.1, we can see that the gender
gap is -1.76, the one-step education gap is .72, and the one-step income gap is
.20. The gender gap is greater than 2
steps of education (1.44), but it is also greater than 5 steps of income (1.00).

A couple of students used the
standardized regression coefficients rather than the unstandardized. (Standardized coefficients are unstandardized
coefficients divided by the standard deviation of the independent variable in
question). This is incorrect, but since
I did not explain this point, I still gave credit for answers based on these
numbers.

6. In Cooter's model
of bicameralism, assume that $L<$U, but $L_{0}>$U_{0}.

__T, F, and Explain:__
**Given these assumptions, the
Pareto set and the bargain set are actually the same.**

TRUE. The Pareto set is the set of policies such
that making one party better off makes the other worse off. That remains the set of points from $L to
$U. The bargain set, however, is the *subset* of all points __in the Pareto__
set that both parties prefer to $0. Since
both $L<$U<$U_{0}<$L_{0}, it is clear that both houses
prefer $U to $0, and therefore prefer any point in the range $L-$U to $0. Intuitively, parties would never bargain for
more than $U, because they both prefer $U to any amount greater than $U.

__Part 2:__ Short
Essays

**(20 points each)**

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

1. We discussed four different views of "tax equity": the standard cardinal utility view, Rawls, Landsburg, and Rothbard. Analyze the equity effects of federalism from any TWO of these perspectives. Which view is less sympathetic toward federalism, and why?

As a general rule, the more
redistribution a thinker favors, the less favorably he will view federalism,
because federalism means tax competition.
This tax competition allows citizens to escape redistribution by
relocating.

Rawls favors more
redistribution than cardinal utility recommends, so he is likely to be most
hostile toward federalism.

Since cardinal utility argues
for redistribution from rich to poor, however, it too is fairly hostile to
federalism (at least on tax equity grounds), though not as hostile as Rawls.

Rothbard sees taxation as
theft, so he is likely to be most positive toward federalism.

There are two strands in
Landsburg: If he really wants the "Perfect Tax," he would naturally
oppose federalism for thwarting redistribution.
But based on his "principled objection" to redistribution, he
would see federalism's ability to limit redistribution as equity-enhancing.

2. Suppose that
before the recent __two__
different ways that voters could have compensated for their inability to verify
the President's claims.

One method would be the
probability multiplier if Bush is later found to be lying. He could be punished by denying him
reelection, impeachment, or worse.

Another method would be
blanket skepticism: Assume no WMD until Bush proves otherwise. This would give Bush an incentive to provide
information as soon as it became available.

One student made the novel
suggestion of a Hansonian betting market.

3. "If the assumptions underlying the Miracle of Aggregation are correct, then all else equal we should expect citizen welfare to be greater in more populous states." Is this correct? Why or why not?

Yes,
it is correct. The Miracle of
Aggregation is based on the Law of Large Numbers. It says that ignorant voters *tend to* cancel each other out *as N goes to infinity!* Therefore as the number of voters N gets
bigger, the Miracle is increasingly likely to hold. On the other hand, if there were only 3
voters and 2 voted randomly, the worse candidate would still win 25% of the
time.