Mit14 04f20 Exam1

Midterm 1
14.04, Fall 2020

Prof: Robert Townsend
TA: Laura Zhang and Michael Wong
Please answer all four questions. There are 80 points in total. You have 80mins to
complete the exam.
1 Short Answer (30 pts)

Answer the following questions in no more than one paragraph each. True and
False answers need to have explanations to receive full credit.
a) (6 pts) In a utility maximization problem with two goods X1 , X2 , suppose

we have found a maximum that is a corner solution where X1 = 0. What
do we know about the relationship between M RS = M U1 /M U2 relative
to the price ratio p1 /p2 ?
Solution: At the corner solution X1 = 0, then either you have a tangency
at zero or you would prefer to have an even lower X1 without the non-
negativity constraint. Therefore M U1 /M U2 ≤ p1 /p2 .
b) (6 pts) True or False: You can consume two goods X1 , X2 and have a fixed
budget I. Prices for the two goods are p1 , p2 . If p2 increases, it is always
the case that consumption of X1 will increase since it is now relatively
cheaper.
Solution: Not always true. The substitution effect would lead you to
want to consume more X1 , but if X1 is a normal good, there is an offsetting
income effect from higher prices that would lead you to want to consume
less X1 . Which one dominates depend on the utility function. The utility
function is not specified here, so X1 or X2 could be inferior goods, or
there may not be any substitution effects (e.g. in the case of perfect
complements). One specific example of why the statement may fail is
needed for full credit.
c) (6 pts) Define “Pareto-optimal allocation” and “feasible allocation.”
Solution: A feasible allocation satisfies the technology and endowment
constraints. An allocation is Pareto-optimal iff it is feasible and there is
no other feasible allocation that is weakly preferred by all and strictly
preferred by some.
d) (6 pts) There is a feasible allocation x and some other feasible allocation x0

such that for some i, xi0 i xi but for some other j, xj j xj0 . Does that
tell us anything about whether x is a Pareto-optimal allocation or not?
Solution: This does not tell us about whether x is Pareto-optimal or not.
If x0j j xj for all j and xj0 j xj for some j, then we would know that
1
x is not Pareto-optimal. However, there is not enough information in the
given statement to say whether x is Pareto-optimal or not.
e) (6 pts) Describe the experiment Jensen and Miller performed with poor
households in China. Summarize the evidence for and against Giffen be-
havior.
Solution: They experimentally subsidize the price of several staple food
items, and find Giffen good behavior with rice and weakly with wheat
purchases. Lowering the price for these staple goods reduced household
demand for them. These results are stronger in Hunan because households
have fewer substitution possibilities and are much poorer, as predicted by
theory.
2 Production Functions (10 pts)

In this section, we will ask about returns to scale for production functions.
a) (4 pts) How is homogeneity of degree one related to returns to scale?
Solution: Homogeneity is the same as constant returns to scale.
Now using your answer to part (a) or any other techniques, show whether the
following production functions are decreasing, constant, or increasing returns to
scale.
b) (3 pts) F (X, Y ) = X 1/3 Y 2/3
Solution: This is constant returns to scale
c) (3 pts) F (X, Y ) = 2X 1/2 + 2Y 1/2
Solution: This is decreasing returns to scale
3 Expected Utility, Risk, and Insurance (20 pts)

An agent has a CRRA utility utility function
c1−θ
u(c) =
1−θ
An agent has poor health, and there are two possible states s they could be in
(1) s = H for healthy and (2) s = S for sick, each with probability 1/2. Suppose
that consumption when healthy is c = 2 and consumption when sick is c = 1.
Recall from lecture that expected utility would be
X
E[u(c)] = π(s)u(c; s)
s∈{H,S}
Here π(s) = 1/2 for s = H, S. The agent can pay for insurance which will lead
consumption to be 2 − p in both states, eliminating all risk. We can think of p
as the cost of insurance.
2
a) (10 pts) What is the expected utility of the agent without insurance and
with insurance?
Solution: Expected utility without insurance is

21−θ

1 1
E[u(c)] = +
2 1−θ 1−θ
Expected utility with insurance is
(2 − p)1−θ
E[u(c)] =
1−θ
b) (10 pts) Solve for the highest amount the agent would be willing to pay for
insurance p∗ .
Solution: The highest the agent would be willing to pay is to set the
two expected utilities equal.
21−θ (2 − p)1−θ

1 1
+ =
2 1−θ 1−θ 1−θ
With some rearrangement and simplication, the solution is
1 1
p∗ = 2 − ( + 2−θ ) 1−θ
2
4 Dynamic Consumption and Saving (20 pts)

Suppose that a village plants crops in t = 0 and the crops will produce output
in t = 1, 2, 3. Output is decreasing over time where R1 = 4, R2 = 1, R3 = 0.
The village only consumes in periods t = 1, 2, 3, so the world ends in t = 3.
The village can store crop output across time periods without any depre-
ciation or uncertainty. There is no discounting√ of future periods. The
utility function from consumption is u(Ct ) = Ct . If the village stores St crops
in period t − 1, then it can consume up to St crops plus any output Rt in period
t, but minus any storage for the next period St+1 .
a) (6 pts) Set up the village’s maximization problem to find the optimal con-
sumption and storage plan. Be clear about the objective function and
constraints.
Solution:
3 p
X
max Ct
Ct ,St
t=1
s.t. C1 = R1 − S2
C2 = R2 + S2 − S3
C3 = R3 + S3
3
b) (4 pts) Notice that the utility function is strictly concave and there is no dis-
counting over time. Do you have any conjectures about what the optimal
consumption plan looks like?
Solution: Since utility is strictly concave, we do not want extreme amounts
of consumption in each period. Since there is no discounting, we value con-
sumption in each time period equally, and therefore utility is maximized
by setting consumption equal across time.
c) (6 pts) Write the FOCs for the optimal consumption and storage amounts
Ct for t = 1, 2, 3 and St for t = 2, 3
Solution: We can rewrite the maximization problem by subbing in R1 , R2 ,
plugging Ct into the objective so we have a function in S2 , S3
p p p
max 4 − S2 + 1 + S 2 − S3 + S3
leading to the FOCs
1 1
− (4 − S2 )−1/2 + (1 + S2 − S3 )−1/2 = 0
2 2
1 1
− (1 + S2 − S3 )−1/2 + (S3 )−1/2 = 0
2 2
d) (4 pts) Solve for optimal consumption and storage Ct , St

Solution: Rearranging the FOCs, we find that S2 = 7/3, and S3 = 5/3.
We can plug these back into the consumption constraints to get C1 = C2 =
C3 = 5/3. Clearly, we can see that it is optimal to smooth consumption
over time since the utility function is concave and there is no discounting.
4
MIT OpenCourseWare
https://ocw.mit.edu/
14.04: Intermediate Microeconomic Theory

Fall 2020
For information about citing these materials or our Terms of Use, visit: https://ocw.mit.edu/terms.

Mit14 04f20 Exam1

Uploaded by

Copyright:

Available Formats

Mit14 04f20 Exam1

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Mit14 04f20 Exam1

Uploaded by

Copyright:

Available Formats

Midterm 1

14.04, Fall 2020

1 Short Answer (30 pts)

a) (6 pts) In a utility maximization problem with two goods X1 , X2 , suppose

d) (6 pts) There is a feasible allocation x and some other feasible allocation x0

2 Production Functions (10 pts)

3 Expected Utility, Risk, and Insurance (20 pts)

Solution: Expected utility without insurance is

4 Dynamic Consumption and Saving (20 pts)

d) (4 pts) Solve for optimal consumption and storage Ct , St

14.04: Intermediate Microeconomic Theory

You might also like