1 Problem Set 1 Solutions
1 Problem Set 1 Solutions
Revision
Mark the following statements as True (T) or False (F). No explanations required in this
part.
0) so that x falls. These statements imply in combination that x does not deviate
from x ∗ but gravitates back to x ∗ . Here:
i
∂k̇ h
= sαkα−1 − (n + g + δ) = ...
∂k ∗
k=k ∗
k=k
= (α − 1)(n + g + δ)
hence, if n + g + δ > 0 is assumed to hold,
∂k̇
<0 ⇔ α<1
∂k k=k∗
Verbally: For k < k∗ , k̇ > 0 as investment outweighs depreciation and dilution
(dilution: as capital stock is defined in efficiency units, additional capital in effi-
ciency units is only added if additional effective workers AL are also equipped
with additional capital K.). Hence, k converges to k∗ for k < k∗ . Vice versa,
k > k∗ implies k̇ < 0 as depreciation and dilution outweigh investments.
Graphically:
Oliver Landmann Advanced Macroeconomics I
Markus Epp PS1: Solow model
Y
c) As y = AL = kα ,
α
∗ s 1− α
y =
n+g+δ
follows immediately. Moreover, c = (1 − s)y∗ implies
∗
α
s 1− α
c ∗ = (1 − s )
n+g+δ
d) k GR : capital stock (in eff. units) that maximizes steady state consumption (in eff.
units). Two ways to solve the problem:
1. Find the golden rule savings rate sGR for which steady state consumption is
at its maximum. Compute the corresponding capital stock.
2. Find the golden rule capital stock k GR directly by noting that consumption
is the difference between output y and investments (n + g + δ)k.
If we are not directly interested in computing the golden rule savings rate sGR
we solve therefore
k GR := arg max {kα − (n + g + δ)k }
k >0
All capital stocks k > k GR imply that a lower saving rate improves consumption
c in the short and the long run. Hence, all saving rates s > α cannot be dynami-
cally efficient as any generation can be made better of by lowering s. Thus, [0, α]
is the range of dynamically efficient saving rates.
e) Factor prices are derived from the representative firm’s problem:
max Π := Y − wL − RK
K,L
The FOCs define implicitly aggregate demand for K and L dependent of the
prices w and R:
∂Π
= α( AL)1−α K α−1 − R = 0
∂K
∂Π
= (1 − α) A1−α L−α K α − w = 0.
∂L
Because of R = r + δ:
r = αkα−1 − δ
w = (1 − α) Akα
r∗ > n + g
α −1
⇔ αk∗ > n + g + δ
α −1
s 1− α
⇔ α > n+g+δ
n+g+δ
n+g+δ
⇔ α > n+g+δ
s
⇔ α>s
All statements are tautological (i.e. equivalent) and so are r ∗ > n + g and s < α.
Hence, we could also have started from s ≤ α to derive the even more general
result r ∗ ≥ n + g.
Oliver Landmann Advanced Macroeconomics I
Markus Epp PS1: Solow model
c) A one time jump in the number of workers leads to a temporary departure from
the old (=new) steady state. After a discrete jump in the level of output, the
growth rate of output is temporarily above the pre-shock growth rate during the
transition to the new (=old) steady state. This is due to the fact, that gk goes up
while n and g remain constant.
d) A one-time increase in the labor supply amounts to an immediate fall in the
capital-labor ratio of the economy. As a consequence, per-capita income and
per-capita consumption are reduced on impact. But since the steady-state level
of per-capita income is not affected at an unchanged saving rate, per-capita con-
sumption gradually recovers and returns to its initial steady-state path.
Revision
Mark the following statements as True (T) or False (F). No explanations required in this
part.
2. In the Solow model, convergence takes place when α, n, g and δ are similar even
when K differs initially.
α
1− α
3. As y = n+ sg+δ falls in g, a higher rate of g implies less output per effective
worker. Note however, that output in aggregate and per capita terms still grows
faster in this case.
5. It’s the other way round: break-even investment exceeds actual investment if k > k∗ .