Econ 421, Fall 2023 – Example Midterm

NAME:
STUDENT ID:

Please answer all questions. You may use a calculator if you wish, though it is not required.
If you need extra space, scratch paper is available; attach any extra sheets to your exam
when you turn it in. Remember to put your name and student ID number on your test. You
will have 1 hour and 20 minutes to complete the exam.

Note: There are 100 total points with 10 free points, so the min. score is 10.

Do NOT open this test until instructed to do so.

1
Short Questions: True or False [30 pts]

Decide whether each of the following claims is true, false, or uncertain. Explain your answers.

1. [5 pts] The income and substitution effects of a wage change on labor supply work in
opposite directions.
Solution: True. When the wage falls, the substitution effect pushes the worker to
enjoy more leisure and consume less, but the income effect pushes the worker to decrease
both consumption and leisure. When the wage rises, the substitution effect pushes the
worker to consume more and enjoy less leisure, but the income effect pushes the worker
to enjoy more of both. The income and substitution effects of a wage change on leisure
(and therefore labor supply) thus work in opposite directions.

2. [5 pts] Using an instrumental variables strategy, Brown (1980) finds evidence that
workers who work under worse conditions are compensated with higher wages.
Solution: False. Brown (1980) uses a fixed-effects regression to control for unobserved,
time-invariant worker characteristics that could induce “ability bias” in the estimation
of the hedonic wage function. He finds mixed evidence that is largely inconsistent with
the presence of compensating differentials for unattractive work environments.

3. [5 pts] Immigration reduces native employment when the labor demand curve is per-
fectly elastic.
Solution: False. When labor demand is perfectly elastic, firms are willing to hire as
many workers as are are willing to work at the prevailing market wage. Therefore, an
increase in labor supply due to immigration has no effect on the wage, and no native
workers will be displaced. In the Johnson (1980) model, the effect of an increase in
immigrant employment Ei on native employment En is given by:

dEn ϵn (1 − µ)
β=− = ,
dEi ϵn (1 − µ) − ηd

2
 where ϵn is the elasticity of native labor supply, µ is the immigrant share of the labor
market, and ηd is the elasticity of labor demand. Perfectly elastic labor demand means
ηd → −∞, which implies β → 0.

4. [5 pts] In theory, the Earned Income Tax Credit may cause some workers to exit the
labor force.
Solution: False/Uncertain. If we consider only labor supply responses to an EITC,
this statement is false. Anyone who would have worked without the EITC would
still work with an EITC, because the EITC 1) provides no benefit to those who do
not work, and 2) increases the effective wage workers earn at the endowment point.
Therefore, if the market wage was above the worker’s reservation wage without the
subsidy, it will also be above the worker’s reservation wage with the subsidy (and so
no one will leave the labor force). If we consider employer responses in equilibrium
as in Rothstein (2010), then employers will cut wages in response to the EITC policy,
and some non-EITC-eligible workers may leave the labor force.

5. [5 pts] According to the intertemporal labor supply model (with many periods), if the
wage in period t increases temporarily, the impact on leisure in period t will be bigger
than on consumption.
Solution: True. According to the intertemporal model, M ULt = wt λ, and M UCt = λ
at the worker’s optimal choices of leisure and consumption in all periods t. If a wage
increase in period t is expected, the marginal utility associated with an increase in the
worker’s lifetime wealth λ will not be affected and so her optimal choice of Ct will not
change. However, the worker will choose to work more in period t, because M ULt will
rise, implying that Lt falls. If the wage increase is unexpected, the results are largely
the same, except that one-time increase in wages will cause λ to fall slightly, but this
change will be much smaller than the change in the wage (since the income effect of
the wage change must be spread out over all remaining periods).

3
6. [5 pts] Angrist and Krueger (1991) present definitive evidence in favor of the human
capital theory of schooling relative to the signaling model.
Solution: False. In order to falsify the signaling model relative to the human capital
model, it is necessary to compare outcomes of (otherwise identical) workers who have
the same schooling inputs but received different signals about their schooling. Angrist
and Krueger compare the outcomes of students who received different levels of school-
ing, but also (likely) different signals about their total schooling, so it is not possible to
definitely say that the return to schooling they estimate is only due to human capital.

4
Long Question 1 [30 pts]

Nick has the utility function:

where C is consumption and L is leisure.

A. [6 pts] Find Nick’s marginal utilities of leisure and consumption.

 1/4
3 L
M UC =
4 C
 3/4
1 C
M UL =
4 L

B. [6 pts] Find Nick’s marginal rate of substitution between leisure and consumption. In
words, what is the interpretation of the MRS?
Solution: The MRS is

 3/4  1/4
M UL 1 C 4 C
M RSLC = = ×
M UC 4 L 3 L
C
= .
3L

In words, the MRS is the amount of consumption that Nick would be willing to give
up to get another unit of leisure.

C. [6 pts] There are 12 hours available for work or leisure, and Nick has non-labor income
of $32. His wage is $4. Write an equation for his budget constraint. Draw a graph of
the constraint, labeling key points.
Solution: The budget constraint is:

C + wL ≤ wT + V.

5
 Plugging in, we have:
C + 4L ≤ 80.

The graph of Nick’s budget constraint is:

D. [6 pts] Find Nick’s optimal choices of leisure and consumption. Draw his optimal
choices in a graph.
Solution: Nick will set the MRS equal to his wage rate:

C
=4
3L
=⇒ C = 12L

6
 Plugging this into the budget constraint yields:

12L + 4L = 80

=⇒ 16L = 80

=⇒ L = 5, C = 60.

The graph is:

E. [6 pts] Nick’s wage falls from $4 to $2. For this wage change, is the substitution effect
or the income effect stronger?
Solution: When Nick’s wage falls, the substitution effect will pushes him to enjoy
more leisure than before, but the income effect will push him to enjoy less leisure than
before. With a wage of 2, Nick’s choice satisfies the new optimality condition:

C
=2
3L
=⇒ C = 6L.

7
Plugging this in to his budget constraint gives:

6L + 2L = 80

=⇒ L = 10.

This means Nick takes more leisure when the wage is lower, which immediately tells
us that the substitution effect is stronger.

8
Long Question 2 [30 pts]

A firm has the production function:

f (E, K) = E α K 1−α

A. [6 pts] Find the firm’s marginal rate of technical substitution and its elasticity of
substitution between labor and capital.
Solution: The MRTS is:

M UE αE α−1 K 1−α α K
M RT S = = α −α
=
M UK (1 − α)E K 1−αE

The elasticity of substitution can be found using the equation:

w
M RT S =
r
α K w
=⇒ =
1−αE r
=⇒ log(K/E) = log(w/r) − log(α/(1 − α)).

Then:

∂ log(K/E)
σ= = 1.
∂ log(w/r)

B. [6 pts] Suppose α = 1/2, and K is fixed in the short run at K0 = 1. Find the short-run
demand curve for labor if the firm acts as a price taker.
Solution: Plugging in α = 1/2, and K0 = 1, the firm’s profit maximization problem
is:
√
max p E − wE
E

9
 Taking the first order condition, we can solve for the firm’s short-run labor demand:

p
V M PE = √ = w
2 E
√ p
=⇒ E =
2w
p2
=⇒ E =
4w2

p
Since V M PE = √
2 E
is always declining, V M PE ≤ V APE for all values of E. This
p2
means the firm’s short-run demand for labor is to E(w) = 4w2
for any w ≥ 0.

C. [6 pts] Now suppose the firm is a monopsonist. The short run labor supply curve is
E = w2 , and the product price is p = 1. What is the firm’s marginal cost of hiring?
Find the firm’s optimal choice of labor and the wage.
Solution: The monopsonist faces a marginal cost of hiring of:

M CE = w(E) + w′ (E)E

From the supply curve we have:

√
w(E) = E
1
=⇒ w′ (E) = √ ,
2 E

and so

√ 1
M CE = E+ √ E
2 E
3√
= E.
2

10
 The monopsonist sets V M PE equal to M CE :

1 3√
√ = E
2 E 2
1
=⇒ E = .
3

Plugging this in to the (inverse) labor supply curve, the wage is then:

√ 1
w= E=√ .
3

D. [6 pts] The government imposes a payroll tax of t = 1 per hour on the firm. Will the
firm’s employment go up, or down? (No need to explicitly solve for the new employment
level.) Judging by your results, how does the effect of a payroll tax in a monopsonistic
labor market compare to its effect in a perfectly competitive market?
Solution: The firm’s new marginal cost of hiring is just the old marginal cost plus
the payroll tax:
M CE = w(E) + w′ (E)E + t.

This makes hiring more expensive at every level of E. Because V M PE is declining, the
firm will further cut back on hiring. The payroll tax will therefore reduce employment
and take-home wages. This is the same direction as the corresponding effects in a com-
petitive market. Note the difference between this effect and a minimum wage, which
reduces employment in a competitive market but increases it under monopsony. The
payroll tax example shows that not all policies have different effects under monopsony
and competition.

E. [6 pts] Consider the long run. Assume the firm produces q = 10, the wage is 4 , and
the rental rate is r = 1. Find the substitution effect of a decrease in wage from 4 to 1.

11
Solution: The firm’s initial choice satisfies the conditions:

√ √ K
| E {z
K = 10} and
E
=4 .
isoquant for q=4 | {z }
M RT S=w/r

The MRTS condition can be written as K = 4E. Plugging this in to the isoquant
gives:

√ √
E 4E = 10

=⇒ 2E = 10

=⇒ E ∗ = 5.

When the wage falls to 1, we find the substitution effect by keeping the firm on the
same isoquant, but setting the MRTS equal to the new price ratio:

√ √
E K = 10
K
=1
E
=⇒ K = E

=⇒ E sub = 10

The substitution effect is therefore:

E sub − E ∗ = 10 − 5 = 5.

12