Macro Tut 4
Macro Tut 4
Question 4.3
Ben J. Heijdra
Outline
1 Some techniques
2 Question 4.3
(1) Draw a diagram with y(t) on the vertical axis and z(t) on the
horizontal axis.
(2) Draw the isoclines.
The y-isocline depicts combinations of y(t) and z(t) for which
ẏ(t) = 0.
The slope can be determined by using the Implicit Function
Theorem. For example:
∂y(t) Fz (y(t), z(t), w0 , x0 )
=− .
∂z(t) ẏ(t)=0 Fy (y(t), z(t), w0 , x0 ))
(3) Mark the steady state, which is such that ẏ(t) = ż(t) = 0 (at the
intersection of the isoclines). Denote the steady-state values by y ∗
and z ∗ .
(4) Draw the arrows of motion.
Calculate either ∂ ẏ(t)/∂y(t) or ∂ ẏ(t)/∂z(t) (whichever is easier) and
analyze its sign. This determines whether y(t) is increasing or
decreasing over time for points off the y-isocline.
Calculate either ∂ ż(t)/∂y(t) or ∂ ż(t)/∂z(t) (whichever is easier) and
analyze its sign. This determines whether z(t) is increasing or
decreasing over time for points off the z-isocline.
Intermediate Macroeconomics EBB842B05, 2022–2023 Tutorial 4: Macroeconomic dynamics, II 4 / 31
Some techniques
Question 4.3
Stable Unstable
y(t) y(t)
. .
z(t) = 0 z(t) = 0
E0 E0
y* ! y* !
A A
y0 ! y0 !
. .
y(t) = 0 y(t) = 0
z0 z* z(t) z0 z* z(t)
y(t)
.
B! z(t) = 0
!
A BN
y(0) !
C E0
!
y* !
!
CN
SP
.
y(t) = 0
z0 z* z(t)
Analysis of a shock
[How to: Analyze a saddle-point stable model in continuous time]
Analysis of a shock
ṗ = φ[−σRL + g − ȳ]
ṘL [λ + γσ]RL − γg + (m − p) − α
=
RL λ
Intermediate Macroeconomics EBB842B05, 2022–2023 Tutorial 4: Macroeconomic dynamics, II 11 / 31
Some techniques
Question 4.3
Part (a)
∗ g − ȳ λ λ + γσ
RL = , p∗ = g− ȳ + m − α
σ σ σ
(5) Arrows of motion:
∂ ṗ ∂ ṘL RL
= −φσ < 0, =− <0
∂RL ∂p λ
(6) The matrix of partial derivatives is given by:
!
0 −φσ
∆= RL∗
λ+γσ ∗
− λ λ
RL
such that:
φσRL ∗
|∆| = −<0
λ
Hence the system is saddle-point stable.
RL ṘL = 0
∗
RL ṗ = 0
p∗ p
The steady state values and transitional paths for y and RS can be
deduced.
y = −σRL + g (M1)
γ[−σRL + g] + α − (m − p)
RS = (M2)
λ
ṗ = φ(y − ȳ) (M3)
ṘL
RS = RL − (M4)
RL
RL ṘL = 0
E0
(ṗ = 0)0
E1
(ṗ = 0)1
∗
RL,1 p∗1
tA = tI t tA = tI t
RS y
∗
RS,0 ȳ1
∗
RS,1 ȳ0
tA = tI t tA = tI t
Part (c)
(c) Now redo part b for the case in which the shock is announced (i.e.
becomes known to the public) before it actually takes place. The
time at which the shock actually occurs (the “implementation time”)
is tI , so we assume that tA < tI .
RL ṘL = 0
E0
(ṗ = 0)0
E1
(ṗ = 0)1
RL,0
∗
p∗0
RL,1
∗
p∗1
tA tI t tA tI t
RS y
RS,0
∗
ȳ1
RS,1
∗
ȳ0
tA tI t tA tI t
Part (d)
E1
(ṗ = 0)1
(ṗ = 0)0
E0
∗
RL,0 p∗0
tA = tI t tA = tI t
RS y
∗
RS,1
ȳ
∗
RS,0
tA = tI t tA = tI t
Part (e)
E1
(ṗ = 0)1
(ṗ = 0)0
E0
∗
RL,0 p∗0
tA tI t tA tI t
RS y
∗
RS,1
ȳ
∗
RS,0
tA tI t tA tI t
Part (f)
(ṗ = 0)1
(ṗ = 0)0
E0
∗
RL,0 p∗0
tA tI = tE t tA tI = tE t
RS y
∗
RS,1
ȳ
∗
RS,0
tA tI = tE t tA tI = tE t
(g) Assume that the real (rather than the nominal) long-term interest
rate, rL ≡ RL − ṗ, features in equation (Q4.1). Assume furthermore
that 0 < σφ < 1. Study the (impact, transitional, and long-run)
effects of an unanticipated and permanent technology shock (an
increase in ȳ). Illustrate your answers in an impulse-response
diagram.
y = −σ[RL − ṗ] + g
−σ[RL + φȳ] + g
y=
1 − σφ
The rest of the question is similar to what was done in part (b)