Suggested answers for written Exam for the B.Sc.

in
Economics,
Winter 2010/2011

Macro B/Macro 3

Final Exam

January 13, 2011

(3-hour closed-book exam)

Academic Aim: The aim of the course is to describe and explain the macroeco-
nomic fluctuations in the short run, i.e. the business cycles around the long run
growth trend, as well as various issues related to this, and to teach the methodol-
ogy used in formulating and solving formal models explaining these phenomena.
Students are to learn the most important stylized facts about business cycles and
to acquire knowledge about theoretical dynamic models aimed at explaining these
facts. In connection with this, the aim is to make students familiar with the dis-
tinction between deterministic and stochastic models. Furthermore, students are
to gain an understanding of the distinction between the impulses initiating a busi-
ness cycle and the propagation mechanisms that give business cycles a systematic
character. Finally students are to learn how to use the models for analyzing the
effects of macroeconomic stabilization policy under various assumptions regarding
the exchange rate regime.To obtain a top mark in the course students should at
the end of the course be able to demonstrate full capability of using the techniques
of analysis taught in the course as well as a thorough understanding of the mech-
anisms in the business cycle models for open and closed economies, including the
ability to use relevant variants and extensions of the models in order to explain
the effects of various shocks and the effects of macroeconomic stabilization policies
under alternative monetary and exchange rate regimes.
Problem A

1
max U = u (C1 ) + u (C2 ) (A.1)
C1 ,C2 1+φ


1 L 1  L 
s.t. C1 + C2 = Y1 − T1 + Y − T2 + V1 (A.2)
1+r 1+r 2

1. Equation (A.1) is the utility function stating that the consumer gains utility
through consumption in period 1 and period 2. Since the consumer is im-
patient (s)he discounts period-2 consumption with the parameter φ which is
also called the rate of time preference. Equation (A.2) is the intertemporal
budget constraint stating that the present value of consumption must equal
the present value of financial and human wealth. Human wealth — or human
capital — is given as the present value of labour income earnings net of taxes.

2. The problem can be solved by substitution.

First, solve (A.2) with respect to C2 ;


L 1  L 
C2 = (1 + r) Y1 − T1 + Y − T2 + V1 − C1 (S.1)
1+r 2

Insert (S.1) into (A.1) and state the problem like


1 L 1  L 
max U = u (C1 )+ u (1 + r) Y1 − T1 + Y − T2 + V1 − C1
C1 1+φ 1+r 2
(S.2)
Find the first-order condition with respect to C1 ;
∂U 1+r ′
= 0 : u′ (C1 ) − u (C2 ) = 0 ⇐⇒
∂C1 1+φ
1+r ′
u′ (C1 ) = u (C2 ) (S.3)
1+φ
σ−1
σ
Since utility is given by u (C) = σ−1
C σ , we have

∂u (Ci ) −1
u′ (Ci ) = = Ci σ ,
∂Ci

2
 and (S.3) can be stated as
− σ1 1 + r − σ1
C1 = C ⇐⇒
1+φ 2

−σ
1+r
C1 = C2 ,
1+φ

and by using (S.1) we have

−σ

1+r L 1  L 
C1 = (1 + r) Y1 − T1 + Y − T2 + V1 − C1 ⇐⇒
1+φ 1+r 2

−σ 
1+r
1+ (1 + r) C1 =
1+φ



1 + r −σ L 1  L 
(1 + r) Y1 − T1 + Y − T2 + V1 ⇐⇒
1+φ 1+r 2
 −σ
1+r
(1 + r)

1+φ L 1  L 
C1 =
 −σ  Y1 − T1 + Y − T2 + V1 ⇐⇒
1+r 1+r 2
1 + 1+φ (1 + r)


1 L 1  L 
C1 =
 Y1 − T1 + Y − T2 + V1 ⇐⇒
1 1+r 2
1+r −σ
+1
( 1+φ ) (1+r)


1 L 1  L 
C1 =   Y1 − T1 + Y2 − T2 + V1
1
+ 1 1 + r
(1+r)−σ+1 (1+φ)σ


1 L 1  L 
C1 =   Y1 − T 1 + Y − T2 + V1 ,
(1 + r)σ−1 (1 + φ)−σ + 1 1+r 2

and then we have (A.3)


L 1  L  1
C1 = θ Y1 − T1 + Y − T2 + V1 , θ≡
1+r 2 1 + (1 + r) σ−1
(1 + φ)−σ
(A.3)

3. This follows directly from eq. (A.3). The derivative is negative, reflecting
that an increase in taxes will decrease consumption. However, since 0 <
1
θ ≡ 1+(1+r)σ−1 (1+φ)−σ
< 1 the effect from taxes to consumption will be less
than one-for-one (numerically). Thus, current consumption will decrease
by less than the increase in taxes. This is due to the consumer’s desire

3
 to smooth consumption (which is facilitated by the assumption of perfect
capital markets that is the basis for the intertemporal budget constraint in
eq. (A.2)) which implies that the consumer will choose to “pay” the increase
in current taxes by lowering both period-1 and period-2 consumption.

4. Use the fact that when dT1 = dT2 we have ∂T2

∂T1
= −1 and based on eq. (A.3)
one finds





∂C1 1 ∂T2 1 1
= θ −1 + − = θ −1 + (−1) = −θ 1 +
∂T1 1+r ∂T1 1+r 1+r
1
Since 1 + 1+r > 1 this derivative is numerically greater than the derivative
in question 1, meaning that the permanent tax change will have a greater
impact on current consumption. This is no surprise and may once again be
explained by the consumer’s desire to smooth consumption: just as part of
the current tax increase will be ’paid’ through lower future consumption, a
part of the future tax increase will be paid by lower current consumption
making the overall effect on current consumption greater.
When r = φ it follows from the expression for θ that
1 1 1 1 1+φ
θ≡ σ−1 −σ = −1 = 1 = 1+φ+1
=
1 + (1 + φ) (1 + φ) 1 + (1 + φ) 1 + 1+θ 1+θ
2+φ
1+φ
Inserting θ = and r = φ into into the expression in eq. (A.5) one finds
2+φ


∂C1 1+φ 1 1+φ1+φ+1 1+φ2+φ
=− 1+ =− =− = −1
∂T1 2+φ 1+φ 2+φ 1+φ 2+φ1+φ
implying that a permanent tax increase will decrease consumption one-for-
one. The reason is that when φ = r the consumer wants to perfectly smooth
consumption, i.e. to choose the same level of consumption in both periods,
C1 = C2 .1 When this is the case, a permanent increase in taxes, which
1
This may be seen from the FOC for utility maximization which reads
1+r ′
u′ (C1 ) = u (C2 )
1+φ
and when ρ = φ one obtains u′ (C1 ) = u′ (C2 ) which implies C1 = C2 , since u′′ (C) < 0.
(u′′ (C) = 0). Intuitively the reason is that r measures the benefit from postponing consumption,
since the extra saving earns an interst. On the other hand, φ measures the disadvantage (’cost’)
of postponing consumption, since future utility carries less weight in the overall utility function.
When ρ = φ the benefit and cost of postponing consumption cancel out, and the consumer
chooses a constant level of consumption over time. I went through this at the lecture but it
probably doesn’t hurt to mention it again.

4
 decreases disposable income in both periods by the same amount, will lead
consumers to just decrease consumption in both periods by the increase in
taxes.

5. Calculations are not required in this question. It suffices to simply interpret

eq. (A.6) as requiring initial debt plus total present value of government con-
sumption expenditure (the LHS) to equal total present value of tax revenue
(the RHS).
Alternatively one may start by arguing that public debt at the beginning of
period 2 must be (this is really a definition and not a budget constraint)

D2 = (1 + r) (D1 + G1 − T1 )

In period 2 the budget constraint of government then requires that

T2 = D2 + G2

i.e. that taxes are sufficient to finance accumulated debt plus consumption
expenditure of the period. Inserting the first expression for D2 into the latter
and rearraning, eq. (A.6) obtains:

T2 = D2 + G2 ⇒
T2 = (1 + r) (D1 + G1 − T1 ) + G2 ⇔
1 1
T2 = D1 + G1 − T1 + G2 ⇔
1+r 1+r
1 1
D1 + G1 + G2 = T1 + T2
1+r 1+r
6. According to eq. (A.6) we find that (taking the derivative w.r.t. T1 )

1 ∂T2
0=1+ ⇒
1 + r ∂T1
∂T2
= − (1 + r)
∂T1
(One can also first isolate T2 in eq. (A.6) leading to

1
T2 = (1 + r) D1 + G1 + G2 − T1
1+r

5
 from which it follows that ∂T2
∂T1
= − (1 + r).) Using this, it now follows from
eq. (A.3) that




∂C1 1 ∂T2 1
= θ −1 + − = θ −1 + (− (− (1 + r))) = θ (−1 + 1) = 0
∂T1 1+r ∂T1 1+r

Thus, a change in taxes with public consumption unchanged will not affect
private consumption. The reason is that when public consumption is not
changed, consumers will realize that higher taxes now will imply lower taxes
in the future (when the governemnt IBS holds) and the present value of the
lower future taxes will be exactly equal to the higher current taxes. Thus
the total life time income of cunsumers will be unaffected in which case
consumers do not want to change the consumption pattern. To be specific,
in case T1 is increased by dT1 consumers will leave C1 unaffected and just
borrow dT1 in the current in order to pay the extra taxes. In period 2 the
consumer will then have to pay (1 + r) dT1 on the extra loan made in period
1 but since taxes in period 2 will be lowered by dT2 = (1 + r) T1 the saved
taxes in period 2 can exactly finance this payment.

1. The point are given in the main text on pp. 486-489, but can be summarized
in the following bullets:

A. Consumer may not be sophisticated enough to understand the public

IBC. In this case the consumers do not understand the consequences of
current tax reductions and therefore they do not expect them to induce
future tax increases.
B. The planning horizon of cunsumers and government may not be iden-
tical. The state will continue to exist beyond the finite lifetime of the
individual consumers. Many current taxpayers may therefore rationally
expect that some of the future tax payments on serviceing existing pub-
lic debt will be levied on future generations. An objection against this
argument, is that a bequest motive would reinforce Ricardian Equiva-
lence. If parents internalize the welfare of their children who care about
the welfare of their children, the current generation will effectively be-
have as if it has an infinite time horizon.

6
C. Consumers are different and taxes are redistributive in nature. There-
fore at the individual basis a given tax reduction might not be expected
to be counteracted exactly by a future tax increase on the same indi-
viduals.
D. Real taxes are distortionary — not lump sum. Real-life taxes are rarely
lump sum. If a debt-financed cut in current income taxes generates
expectations of higher future tax rates, consumers may want to increase
their current labour supply and reduce future labour supply. In that
case, the government might be able to stimulate activity in the economy.
E. Some consumers may be credit constrained. If a tax increase today leads
to lower taxes in the future, consumption smoothing gives an incentive
to borrow money today and pay them back in the future. However, if
consumers are credit restricted, they cannot borrow money and have
to face a lower level of consumption today than an optimally planned
consumption path states. Likewise, the credit-constrained consumer
would increase current consumption, if current taxes were reduced.

7
Problem B
1. Equation (B.1) states the goods market equilibrium value of real output.
(It should be accepted if students write that eq. (B.1) is the goods market
equilibrium condition, although it is not.) The model is log-linearized around
the long run equlibrium in which output is given by natural output, ȳ. In
equlibrium real output is equal to demand, a part of which is public con-
sumption, gt . This explains the positive effect from gt to ḡ. The real interest
rate, rt , affects real output negatively due to the former’s negative effect
on goods market demand through private investment and possibly through
private consumption.
Equation (B.2) defines the (ex ante) real interest rate.
Equation (B.3) states the monetary policy rule according to which the nom-
inal interest rate is an increasing function of the inflation gap defined as the
difference between actual inflation, π t , and the target level of inflation, π ∗ .
Combining eq. (B.3) with eq,. (B.2) it is seen that an increase in inflation
will also increase the real interest rate which will in turn decrease goods
market demand and hence real output. As will be seen below this will af-
fect inflation negatively thorugh the supply side of the economy. Thus the
policy in eq. (B.3) aims at stabilizing inflation around the target value π ∗ .
Eq. (B.3) may be thought of as a special case of the Taylor rule where the
central bank does not react to changes in the output gap, yt − ȳ.
Eq. (B.4) is the fiscal policy rule, which aims at stabilising real output
around ȳ by decreasing (increasing) public consumption when real output is
above (below) ȳ thereby decreasing (increasing) goods market demand.
Eq. (B.5) is the short run aggregate supply curve, SRAS, which states that
inflation is an increasing function of real output and expected inflation. The
SRAS may be explained in a number of ways one of which is: given nomi-
nal wages higher real output increases employment which will decrease the
marginal product of labour thereby increasing marginal costs of firms who
react by increasing prices and this leads to higher inflation. The expected
inflation affects positively because a labour union sets nominal wages based
on the expected inflation (price level) and a higher expected inflation will
cause the nominal wages to increase thereby leading to higher marginal costs
of firms and ultimately to higher inflation.

8
 Eq. (B.6) is the assumption of static expectations according to which the
expectation about the value of a variable (here: inflation) is given as the
actual value of the variable in the previous period.

2. In the long run actual and expected inflation must coincide, π t = π et . It then
follows from the SRAS in eq. (B.8) that (in the absence of shocks, i.e. when
st = zt = 0) yt = ȳ. From the AD curve in eq. (B.7) it then follows that
π t = π ∗ . Thus the long run equlibrium is given by the intersect between the
vertical LRAS and and AD curves in Figure B.1 where yt = ȳ and π t = π ∗ .
Figure B.1 also shows an economy initially being in a recession at point E0 ,
the intersect between AD and SRAS0 , with inflation and real output being
π 0 and y0 < ȳ, respectively. Due to the assumption of static expectations
we then have that in period 1 πe1 = π 0 . Since it follows from the SRAS that
y = ȳ ⇒ πt = π et the SRAS will shift down in period 1 to SRAS1 which
goes thorugh (ȳ, π 0 ). The equilibrium will then be given by point E1 where
inflation has decreased and real output increased compared with period 0.
The economic explanation is the following: between period 0 and period 1
there is a decrease in expected inflation (from π e0 to π e1 = π 0 ). This will
cause the labour union to reduce the growth rate of nominal wages which
will in turn make the marginal costs of the firm increase at a slower pace.
Consequently firms will increase their prices at a slower pace, i.e. inflation
will be reduced. The reduction in inflation will then lead monetary authori-
ties to lower the nominal interest rate according to the monetary policy rule
and this results in a lower real interest rate. This increases goods market
demand which finally leads to increased real output.
In period 2 we then have π e2 = π 1 < π e1 and consequently this process of
downward-shifting SRAS curves will continue until the economy has reace-
hed the long run equilibrium which is seen to be stable.

9
 Figure B.1

3. According to eq. (B.10) expectations are now so-called adaptive expecta-

tions. The consequence is shown graphically in Figure B.2. In period 0 the
economy is in a recession in E0 where actual inflation is π 0 and expected
inflation is π e0 . With static expectations we have in period 1 that π e1 = π 0
and therefore the SRAS will shift down to SRAS(π e1 = π 0 ) i.e. by the full
vertical distance π 0 −π e0 , moving the economy from E0 to E1s . With adaptive
expectations we have that

π e0 > π e1 = φπ e0 + (1 − φ) π 0 > π 0 (7)

Consequently the SRAS in period 1 shifts down only to SRAS(π e1 = φπ e0 + (1 − φ) π 0 ),

i.e. by a smaller vertical distance thereby moving the economy only to E1a .
This still moves the economy towards the long run equlibrium, i.e. the
economy is stable, but since the vertical shift in the SRAS is smaller, the
convergence will be slower.
The basic intuition is the following: when the economy is not in a long run
equilibrium we necessarily have π t = π et , i.e. there is an expectation error.
Over time the revision of expectations will cause the economy to converge
towards the long run equilibrium. This happens both with static and with

10
 adaptive expectations. However, with adaptive expectations, the revision of
expectations is slower, since expected inflation is based partly on past ex-
pectations which are necessarilly incorrect outside long run equlibrium.

Figure B.2

4. We assume that the economy is in a long run equlibrium at E0 in period 0

when in period 1 there is a supply shock. The effect of the supply shock is
shown in Figure B.3 where the SRAS shifts up by the vertical distance st in
period 1 to SRAS1 . This moves the economy from the long run equlibrium
at point E0 to point E1 causing inflation to rise and real output to decrease.
The economic explanation is the following: A negative supply shock directly
leads to higher inflation through, e.g. lower productivity, higher prices of raw
materials etc. which causes firms’ costs to increase thereby leading them to
increase prices. Through the monetary policy rule the increased inflation wil
increase the nominal and then the real interest rate reducing goods market
demand and finally real output. It could be noted that the lower real output
will increase the marginal product of labour thereby reducing the marginal
cost increase of firms thus dampening the increase in inflation.
In period 2 the negative supply shock has disappeared but due to the fact
that the higher inflation of period 1 compared with period 0 has now in-

11
creased expected inflation, the SRAS only shifts down to SRAS2 which goes
through (ȳ, π e2 = π 1 ). This moves the economy from E1 to E2 whereby in-
flation decreases and real output increases. The reason for the decreased
inflation is the disapperance of the negative supply shock but the decrease in
inflation is not equal to the supply shock becuase expected inflation has in-
creased. The decrease in inflation leads to an increase in real output through
monetary policy as described above. From E2 the economy then moves back
to long run equlibrium as described in question 2.

Figure B.3

From eq. (B.7) it follows that the AD curve may be written as

1
πt = π∗ − (yt − ȳ − zt ) (8)
α
This shows the slope of the AD curve to be
1 1 1 + α1 k
− = − α2 h = −
α 1+α k
α2 h
1

We can thus conclude that a higher value of k (relative to h) makes the AD

curve steeper while a higher value of h (relative to k) makes the AD curve
flatter. Figure B.4 shows the effect of a negative supply shock when the AD
curve is either flat or steep. It is seen that the steeper the AD curve the

12
smaller will be the effect on real output and the larger will be the effect on
inflation.
The graphical analysis is not required according to the formulation of the
question but the following explanation is: According to the SRAS curve in eq.
(B.6) a negative supply shock will have to either affect inflation positively,
affect real output negatively or a combination of both. The less inflation is
affected, the larger will be the effect on real output et vice versa.
The higher the value of k, i.e. the more focused fiscal policy is on stabilizing
real output around its long run level, the higher must thus be the effect on
inflation. And likewise: the higher the value of h, i.e. the monetary policy
reacts to inflation gaps, the more will real output be affected.
Hence the conclusion is that in case of a supply shock policy makers have to
choose between stabilizing real output or inflation.
This is not the case when shocks are due to demand. A graphical analy-
sis is somewhat complicated because both the slope and the vertical shift
given zt = 0 will be affected by h and k. However, the economic intuition
is the following: A demand shock will directly affect goods market demand
and thereby equilibrium real output. Real output in turn affects inflation
through the supply side of the economy. Consequently, by stabilizing real
output around its long run level, inflation will also be stabilized (and by
stabilizing inflation, real output will be stabilized through monetary policy).

13
 Figure B.4

5. The change of policy produces a new AD curve. Inserting eq. (B.3) into eq.
(B.2), eq. (B.11) eq. into (B.12) and the resulting expressions into eq. (B.1)
we find

yt = ȳ + α1 (ḡ − k (yt − (ȳ + ω)) − ḡ) − α2 r̄ + π et+1 + h (π t − π ∗ ) − π et+1 −
= ȳ − α1 k (yt − ȳ − ω) − α2 h (π t − π ∗ ) ⇔
(1 + α1 k) (yt − ȳ) = α1 kω − α2 h (πt − π ∗ ) ⇔

α1 k
AD’ : yt = ȳ + ω − α (πt − π ∗ ) (9)
1 + α1 k
It is seen that the change of policy shifts the AD to the right as shown in
Figure B.5. This moves the economy from the long run equlibrium at E0
to E1 where both real output and inflation have increased. The economic
reason is the following: With the change of policy now aiming at stabilizing
real output around y ∗ > ȳ the initial response is to increase government pur-
chases gt . This increases goods market demand and hence the equilibrium
real output. This will increase employment thereby decreasing the marginal
product of labour leading to increased marginal costs in firms. The result
will be an increase in firms’ prices leading to higher inflation.

14
 Figure B.5

Due to the increased inflation in period 1 expected inflation will increase in

period 2 shifting the SRAS shifts up to SRAS2 . This moves the economy
to E2 increasing inflation and decreasing real output. The economic mech-
anisms are exactly as described by the adjustment to long run equlibrium
in qurstion 2, except that the intitial situation is now one in which yt > ȳ.
The proces will continue until the economy has reached its new long run
equlibrium at ELR where π t = π et = π LR and consequently yt = ȳ. The value
of π LR may be found from the expression for AD’ in eq. (9) by letting yt = ȳ:
α1 k
yt = ȳ = ȳ + ω − α (π t − π ∗ ) ⇒
1 + α1 k
α1 k α2 h
0 = ω− (π t − π ∗ ) ⇔
1 + α1 k 1 + α1 k
α 1 k
πt − π∗ = ω⇒
α2 h
α1 k
π LR = π ∗ + ω (A.14)
α2 h
π LR is seen to depend positively on k and α1 . The reason is that the higher
the value of k the stronger will be the fiscal policy reaction to an increase
target for real output. The stronger the fiscal policy reaction in combina-
tion with the magnitude by which fiscal policy affects real output, which is

15
captured by the size of α1 , the stronger is the initial effect on real output.
Since the increase in real output is what ultimately drives up inflation, this
explains the effect on π LR. . At the same time π LR is affected negatively
by α2 and h. The reason is that inflation keeps rising until the increase in
inflation has increased the nominal and hence the real interest rate thereby
reducing real output to its long run level. Since a higher value of h implies
that an increase in inflation will cause a larger increase in the real interest
rate, and since the negative effect from the real interest rate onto real output
is stronger, the larger the value of α2 , the required increase in π in order to
reduce y to ȳ will be smaller the larger the values of η and α2 .

16