Slide04 (Heijdra)

Firm Investment
Dynamic IS-LM
Punchlines
Foundations of Modern Macroeconomics

Second Edition
Chapter 4: Anticipation effects and economic policy
B.J. Heijdra
Department of Economics, Econometrics & Finance

University of Groningen
1 September 2009
Foundations of Modern Macroeconomics - Second Edition Chapter 4 1 / 60

Firm Investment
Dynamic IS-LM
Punchlines
Outline
1 Capital accumulation decision by firms

Adjustment cost theory
Microeconomic effects of an investment subsidy
Macroeconomic effects of an investment subsidy
2 Dynamic IS-LM model and the term structure of interest rates

An ecclectic model
The perverse effects of an anticipated/permanent fiscal boost
3 Punchlines

Firm Investment
Dynamic IS-LM
Punchlines
Aims of this lecture
Complete our discussion of the forward looking theory of

investment (commenced in Chapter 2).
Study the effects of investment stimulation by the government.
Study a dynamic IS-LM theory with an endogenous term
structure of interest rates.

Firm Investment Theory
Dynamic IS-LM Investment subsidy (micro)
Punchlines Investment subsidy (macro)
Dynamic investment theory
Redo the basic model in continuous time.

Real profit:
π(t) ≡ F (N (t), K(t)) − w(t)N (t) − pI (t) [1 − sI (t)] Φ(I(t))
F is a CRTS production function.

Φ is the adjustment cost function.
pI is relative price of investment goods; w is real wage.
sI is the investment subsidy.
N , K, and I are, respectively, employment, capital stock, and
investment.

Quadratic adjustment cost function:
Φ(I(t)) = I(t) + b [I(t)]2
b > 0 so that:
Φ(0) = 0, ΦI = 1 + 2bI > 0, and ΦII = 2b > 0
Capital accumulation:
K̇(t) = I(t) − δK(t)
δ > 0 is the depreciation rate.

Value of the firm:

Z ∞
V (0) ≡ π(t)e−rt dt
0
Z ∞h i
= F (N (t), K(t)) − w(t)N (t) − [1 − sI (t)] Φ(I(t)) e−rt dt
0
Firm must choose paths for labour demand, investment, and

the capital stock such that the value of the firm is maximized,
given the constraints imposed by (a) the capital accumulation
identity and (b) the initial capital stock (K(0)).

How to solve this problem?
Set up so-called current-value Hamiltonian expression (similar

to Lagrangian).
HC (t) ≡ F (N (t), K(t)) − w(t)N (t) − [1 − sI (t)] Φ(I(t))

+ q(t) [I(t) − δK(t)]
q(t) is a co-state variable (similar to a Lagrange multiplier).

N (t) and I(t) are control variables.
K(t) is the state variable.


First-order (necessary) condition for employment:
∂HC (t)
= FN (N (t), K(t)) − w(t) = 0
∂N (t)
We get the usual result: w = FN .

First-order (necessary) condition for investment:
∂HC (t)
= q(t) − (1 − sI (t))ΦI (I(t)) = 0 ⇒
∂I(t)
q(t) = (1 − sI (t))ΦI (I(t)) (S1)
|{z} | {z }
(a) (b)
(a) Shadow price of installed capital (marginal benefit of

investment).
(b) Net marginal cost of investing.

For the quadratic adjustment cost function, (S1) becomes very
simple:
q(t)
ΦI (I(t)) = 1 + 2bI(t) = ⇒
1 − sI (t)

1 q(t)
I(t) = · −1
2b 1 − sI (t)
The first-order (necessary) condition for the capital stock:

∂HC (t)
q̇(t) − rq(t) = − =⇒
∂K(t)
h i
q̇(t) − rq(t) = − FK (N (t), K(t)) − δq(t) =⇒
q̇(t) = (r + δ)q(t) − FK (N (t), K(t)) (S2)

An intuitive way to write (S2) is in the form of an arbitrage

equation:
q̇(t) + FK (N (t), K(t))
= r| {z
+ δ}
q(t)
| {z } (b)
(a)
(a) The return to installed capital consists of a capital gain (q̇)

plus the marginal product of capital (FK ). By dividing the
return by q we obtain a rate of return.
(b) The opportunity cost of invested funds consists of the rate of
interest on other assets (r) plus the rate of depreciation (δ)
(capital evaporates).

The effects of investment stimulation measures

Policy question: What happens if the government subsidizes investment spending by
firms?
Summary of the model developed so far:
K̇ = I(q , sI ) − δK
+ +
q̇ = (r + δ)q − FK (N , K )
+ −
w = FN (N , K )
− +
We have dropped time index where no confusion is possible.

Signs of partial derivatives below variables.

The effects of investment stimulation measures
There are three ways to interpret the model.

(Microeconomic) At firm level: w is constant (constant
capital-labour ratio).
(Macroeconomic) At the economy-wide level: w endogenous
(we need to close the model by looking at the labour market).
(a) Exogenous labour supply (labour scarcity).
(b) Endogenous labour supply (labour supply effects).

Capital-investment dynamics at the level of a firm
If w is constant then so is the marginal product of labour

(since w = FN ).
Since F features CRTS (homogeneous of degree one) it
follows that FN (N, K) is homogeneous of degree zero, i.e. we
can write FN (N, K) = FN (1, K/N ).
Hence, the labour demand equation can be written as
w = FN (1, K/N ).
Since w is constant so is the optimal capital-labour ratio for
the firm (K/N ).
But then the marginal product of capital, FK , is also constant
(since FK (N, K) = FK (1, K/N )).

The micro model
So our model at firm level simplifies to:
K̇ = I(q , sI ) − δK
+ +
q̇ = (r + δ)q − FK
where FK is a constant.
We can derive the phase diagram of this model in Figure 4.1.

Figure 4.1: Investment with constant real wages
q
.
! BN K=0
A
AO ! ! ! AN
B
.
q* = FK/(r+*) ! ! q=0
! BO
K* K

Features of the phase diagram

Start with the K̇ = 0 line: combinations of q and K for which
net investment is zero (I(q, sI ) = δK).
Slope of this line is obtained in the usual fashion:

∂q δ
= >0
∂K K̇=0 Iq
⇒ The line is upward sloping.

For points off the K̇ = 0 line we have:
∂ K̇
= −δ < 0
∂K
⇒ For points to the right (left) of the K̇ = 0 line gross
investment is less than (more than) replacement investment
and net investment is negative (positive). This is indicated
with horizontal arrows in Figure 4.1.

Now look at the q̇ = 0 line: combinations of q and K for
which there are no capital gains or losses (q = FK /(r + δ)).
Slope of this line:
∂q
=0
∂K q̇=0
⇒ The line is horizontal.

For points off the q̇ = 0 line we have:
∂ q̇
=r+δ >0
∂q
⇒ For points above (below) the q̇ = 0 line the shadow price of
capital is higher (lower) than its long-run equilibrium value (of
FK /(r + δ)) so that part of the rate of return on installed
capital is explained by capital gains (losses). Hence, q̇ > 0
(< 0) for point above (below) the q̇ = 0 line. See the vertical
arrows in Figure 4.1.
By combining all the information derive so far we obtain

Figure 4.2. Let us derive (heuristically) the properties of the
model.
There is a unique steady state where the q̇ = 0 line intersects
the K̇ = 0 line (at point E0 ).
By combining the “arrow” information we get the dynamic
forces operating in the four regions (see the hands of the
clock).
We can try out some arbitrary trajectories in the various
regions. None of them seem to go to the equilibrium at E0 !
But that is not quite right! The q̇ = 0 line itself is a stable
trajectory (leading back to E0 ).
We call the unique stable trajectory the saddle path. In this
particular model the saddle path is equal to the q̇ = 0 line (in
the other models this will no longer hold).

Figure 4.2: Derivation of the saddle path

BO
q .
K=0
BN
!
C E0 B
.
q* ! < ! = ! q=0
A CN
! AN
AO
K* K

Experiment 1: Unanticipated and permanent increase in sI

The first policy experiment studies an unanticipated and
permanent increase in the investment subsidy.
“Unanticipated” because announcement date (tA ) and
implementation date (tI ) are the same (agents cannot prepare
for the policy measure and are taken by surprise).
“Permanent” because policy maker announces that the policy
measure is permanent and the agents believe it.
The increase in the investment subsidy lowers the cost of
investing to firms and shifts the K̇ = 0 line to the right in
Figure 4.3. In formal terms:

∂q Is
=− <0
∂sI K̇=0 Iq
The new long-run equilibrium is at E1 .

The adjustment occurs gradually along the saddle path from
E0 to E1 (see the arrows).
Figure 4.3(a): An unanticipated permanent increase in the

investment subsidy
q .
(K = 0)0
.
(K = 0)1
E0 E1 .
q* ! < < ! q=0
K0 K1 K
Figure 4.3: An unanticipated and permanent increase in the

investment subsidy
K1
K
K0
I
I
q .
sI
(K = 0)0
.
(K = 0)1
sI
E0 E1 .
*
q ! < < ! q=0
K0 K1 K
t A = tI time
(a) Phase diagram (b) Impulse-response diagrams
Experiment 2: Unanticipated and permanent increase in r

The second exercise with the model concerns an unanticipated
and permanent increase in the interest rate.
The increase in the interest rate reduces the long-run
equilibrium level for q because the future marginal products of
capital are discounted more heavily. Hence, the q̇ = 0 line
shifts down in Figure 4.4.
The new long-run equilibrium is at E1 .
Adjustment path is an immediate jump in q from E0 to A and
impact (because K is predetermined and can only move
gradually). This is a “financial correction” in the light of new
information (concerning the interest rate).
Economy must jump to the new saddle path because that is
(by definition) the only trajectory leading to the new
equilibrium.
During transition the economy moves gradually along the
saddle path from point A to point E1 .
Figure 4.4(a): An unanticipated permanent increase in the

rate of interest
.
q K=0
E0 .
q *
0 ! (q = 0)0
E1 A
q1* .
! = = ! (q = 0)1
K1 K0 K
Figure 4.4: An unanticipated permanent increase in the rate

of interest
K0
K
E1
K1
I
I
A
.
r1
q K=0 A
E0 .
q*0 ! (q = 0)0
r0
E1 A
q1* .
! = = ! (q = 0)1
K1 K0 K t A = tI time
Experiment 3: Anticipated and permanent increase in r
The third exercise with the model concerns an anticipated and

permanent increase in the interest rate. Agents hear (at
announcement time tA ) that the rate of interest will increase
permanently at some later date (implementation date tI ).
“Anticipated” because announcement date (tA ) and
implementation date (tI ) are not the same (agents can prepare
partially for the shock; the news arrives at time tA ).
Case can be solved technically, but intuitive solution principle
is useful.

Intuitive solution principle:

Discrete jump in q only allowed when news arrives (which is at
time tA ).
K is predetermined at impact (accumulated in the past).
When shock occurs (at time tI ) the economy must be on the
stable trajectory to the new equilibrium (the saddle path).
Between tA and tI the economy must be on a trajectory which
reaches the saddle path at exactly the right time (at tI ). Since
the shock has not occurred yet, dynamics of the old
equilibrium (E0 ) determine the laws of motion.
In Figure 4.5 we deduce the equilibrium adjustment path
from E0 to A to B to E1 .

Figure 4.5(a): An anticipated permanent increase in the rate

of interest
q .
K=0
E0 .
q0* ! (q = 0)0
q(tA) !A
E1 B .
q1* ! = ! (q = 0)1
K1 K0 K
Figure 4.5: An anticipated permanent increase in the rate of

interest
A
K0
B K E1
K1
q,I
A
B q,I
q .
K=0 B
E0 r1
q0* !
.
(q = 0)0 A
q(tA) !A
r0
E1 B .
q1* ! = ! (q = 0)1
K1 K0 K
tA tI time
Intuition for why q falls over time. Integrating the arbitrage

equation q̇ + FK = (r + δ)q from t to ∞ yields the expression
for q(t) at some time t.
Z ∞ Z τ
q(t) ≡ FK (τ ) exp − [r(s) + δ] ds dτ
t t
Hence, q(t) represents the discounted present value of

marginal capital productivities.
If something happens to the interest rate in the future q(t)
reacts immediately.
As time gets closer to implementation of the shock, few years
of low discounting remain so that q(t) falls over time.

Capital-investment dynamics in the aggregate economy (N

fixed)
As a second case we interpret our investment model at the

level of the aggregate economy. Instead of assuming a
constant real wage (which is hard to justify in this case]) we
assume that the supply of labour is exogenous (N = 1).
The model that we wish to analyze is:
K̇ = I(q , sI ) − δK
+ +
q̇ = (r + δ)q − FK (1, K )
−
where we have substituted N = 1 in the expression for the

marginal product of capital (labour market clearing).

The macro model
The key complication is that FK is no longer constant but

diminishing in K (the more capital is added the scarcer is
labour).
As a result the q̇ = 0 line is downward sloping:

∂q FKK
= <0
∂K q̇=0 r+δ
For points above (below) the q̇ = 0 line there are capital gains
(losses):
∂ q̇
=r+δ >0
∂q
Using the same tricks as before we can deduce that the saddle
path is now downward sloping–see Figure 4.6.

Figure 4.6: Investment with full employment in the labour

market
q
.
B! K=0
! CN !
BN
E0 !C
!
A!
DN
! AN ! SP
!D
.
q=0
K
Experiment 1: Anticipated and permanent abolition of sI
First policy shock to be studied concerns an anticipated and

permanent abolition of the investment subsidy (as occurred in
the Netherlands in the 1980s).
The K̇ = 0 line shifts to the left and the long-run equilibrium
shifts from E0 to E1 in Figure 4.7.
Following our “intuitive solution method” we deduce that the
adjustment path is from E0 to A to B to E1 .
We reach the intuitively appealing conclusion that investment
rises at impact (enjoy the subsidy while it exists).

Figure 4.7(a): An anticipated abolition of the investment

subsidy
.
(K = 0)1
q
E1 .
(K = 0)0
! B
!
A
! SP1
E0 !
.
q=0
K1 K0 KN K
Figure 4.7: An anticipated abolition of the investment

subsidy
KN B
K A
K0
K
K1
I B
A
I
I
B
.
q
(K = 0)1
q
q A
E1 .
(K = 0)0
! B
A
!
!
SP1
sI
E0 ! B
sI
.
q=0
K1 K0 KN K tA tI time

Experiment 2: Unanticipated and temporary increase in sI

Second policy shock to be studied concerns an unanticipated
and temporary increase of the investment subsidy (as is
sometimes used to boost the economy).
By “temporary” we mean that the policy maker announces at
time tA = tI that the policy shock will be undone at some
future date tE .
The K̇ = 0 line shifts to the left (if the shock were permanent,
the long-run equilibrium would shift from E0 to E1 in Figure
4.8).
While the higher subsidy is in place (between tA and tE ) the
equilibrium E1 dictates the laws of motion.
Following out “intuitive solution method” we deduce that the
adjustment path is from E0 to A to B to E1 .
Intuitive conclusion: investment rises at impact (“make hay
while the sun shines”) (temporary shock has higher impact
effect on investment than permanent shock).

Figure 4.8(a): A temporary increase in the investment

subsidy
.
(K = 0)0
q
.
E0 (K = 0)1
!
B
!
A!
SP0
!
!
AN !
E1 SP1
.
q=0
K0 K1 K
Figure 4.8: A temporary increase in the investment subsidy
B
K1
K A
K0
K
A B
I
I AN !
B I
q
.
(K = 0)0
q
E0
.
(K = 0)1
A q
!
B
!
A!
!
SP0
sI B
AN
!
! sI sI
E1 SP1
.
q=0
K0 K1 K tA = tI tE time

Capital-investment dynamics in the aggregate economy (N

variable)
We continue to interpret our investment model at the level of
the aggregate economy, but . . .
Instead of assuming a constant employment level, we assume
that the supply of labour is endogenous and depends on the
after-tax wage rate.
The model that we wish to analyze is:
K̇ = I(q , sI ) − δK
+ +
q̇ = (r + δ)q − FK (N , K )
+ −
w = FN (N , K )
− +
g(N ) = w(1 − tL )
+

The macro model
In Figures 4.9 and 4.10 we study the effects on investment

and the capital stock of a decrease in the labour income tax
rate, tL .
For a given capital stock, the decrease in the tax rate
stimulates labour supply (because the substitution effect
dominates the income effect by assumption) so that
employment increases (see Figure 4.10).
Since capital and labour are cooperative factors of production
the marginal product of capital rises and the q̇ = 0 line shifts
to the right in Figure 4.9.
The adjustment path is from E0 to A to E1 in both figures.
Hence, measures which impact directly on the labour market
also have an induced effect on investment and capital
accumulation!

Figure 4.9(a): A fall in the tax on labour income:

investment and employment effects
q .
K=0
A
qN ! E1
q1 !
SP
q0 ! E0
. .
(q = 0)0 (q = 0)1
K0 K1 K
Figure 4.9: A fall in the tax on labour income: investment

and employment effects
A N
K
K1
K A
K0
A
q .
q,I
A
K=0
q,I
qN ! E1
q1 !
SP
tL
q0 ! E0 tL
. .
A
(q = 0)0 (q = 0)1
K0 K1 K tA = t I time

Figure 4.10: The short-run and long-run labour market

effects
w
S
N0
S
N1
E0 E1
! !
A
!
D
N1
D
N0
N0 NN N1 N
Firm Investment
Theory
Dynamic IS-LM
Anticipated fiscal policy
Punchlines
A forward-looking IS-LM model

One of the things some economists do not like about the
IS-LM model is the lack of (forward looking) dynamics
(Blinder-Solow is an example of backward looking dynamics).
It is not difficult, however, to add interesting dynamic effects
to the IS-LM model. We study the model of the New
Keynesian economist Olivier Blanchard.
The model is described by the following equations:
Aggregate demand for goods depends on Tobin’s q (a > 0), on
aggregate production in the economy (Y , 0 < β < 1), and on
government consumption.
Y D = aq + βY + G
Production is changed only gradually (σ > 0):

Ẏ = σ Y D − Y

Firm Investment
Theory
Dynamic IS-LM
Punchlines
A forward-looking IS-LM model

Model features (continued):
Money market equilibrium (RS is the interest rate on short
term securities):
M/P = kY − lRS , k > 0, l > 0
Term structure of interest rates (RL is the yield on

perpetuities):
RS = RL − (1/RL )ṘL (S3)
Arbitrage equation between shares and short bonds:

q̇ + π
= RS (S4)
q
Profits depend positively on aggregate output:

π = α0 + α1 Y
Firm Investment
Theory
Dynamic IS-LM
Punchlines
The asset structure
Especially (S3) and (S4) need some further comment. They

incorporate a more complicated asset structure than the
standard IS-LM model.
There are three financial assets: shares, short bonds,
perpetuities.
All assets are perfect substitutes in the portfolios of investors.
Yields on three assets must be same.
Yield on short bonds is RS .
Yield on shares is (q̇ + π)/q.
Yield on perpetuities is (1 + ṖB )/PB , where PB = 1/RL is
the price of a perpetuity paying 1 euro each period.

Firm Investment
Theory
Dynamic IS-LM
Punchlines
The model
Model summary:
Ẏ = σ [aq − bY + G] , b ≡ 1 − β, 0 < b < 1

RS = (k/l)Y − (1/l)(M/P )
q̇ + α0 + α1 Y
RS =
q
RS = RL − (1/RL )ṘL
Main differences with standard IS-LM model:

Tobin’s q theory of investment.
Expectations play vital role (PFH).
Term structure of interest rates.
Model illustrated graphically with the aid of Figure 4.11.
Firm Investment
Theory
Dynamic IS-LM
Punchlines
Figure 4.11: Dynamic IS-LM model and the term structure

of interest rates
.
. q Y=0
q . Y=0
q=0
"1l/k .
q=0
SP
q0 !
E0 E0
q0 !
SP
"1l/k
!
M/k "0/"1 Y0 Y
M/k G/a Y0 Y
(a) Bad News Case (b) Good News Case

Firm Investment
Theory
Dynamic IS-LM
Punchlines

Ẏ = 0 line is upward sloping. For points above (below) the
line investment is too high (low) and output gradually rises
(falls). See horizontal arrows in Figure 4.11.
Slope of the q̇ = 0 line is ambiguous:

∂q α1 − qk/l
=
∂Y q̇=0 RS
In the steady state q = (α0 + α1 Y )/RS and a rise in Y raises

both the numerator and the denominator. If the LM curve is
relatively steep (so that k/l is high) then the interest rate
effect dominates and the q̇ = 0 line slopes down. This is called
the “bad news case” by Blanchard.
There is a unique saddle-point stable equilibrium at E0 .
Slope of the saddle path is case-dependent.
Firm Investment
Theory
Dynamic IS-LM
Punchlines
Anticipated and permanent boost in public consumption
Focus on “bad news case”.

At time tA agents learn that G will be increased permanently
at some future time tI (tI > tA ).
Dynamic adjustment path for q and Y deduced with intuitive
solution principle; see Figure 4.12.
Paths for remaining variables can be deduced from model
structure.
Initially a perverse effect on output!

Firm Investment
Theory
Dynamic IS-LM
Punchlines
Figure 4.12(a): Anticipated fiscal policy

.
(Y = 0)0
q
E0
q0 ! .
(Y = 0)1
qN A
!
!
B E1
q1 !
SP
.
q=0
Y0 Y1 Y
Firm Investment
Theory
Dynamic IS-LM
Punchlines
Figure 4.12: Anticipated fiscal policy
RL
RL RS
RS
YD
Y
Y
YD Y
YD
.
(Y = 0)0
q q
E0
.
q
q0 !
(Y = 0)1
qN A
!
!
q1
B
! E1 G G
SP
.
q=0
Y0 Y1 Y tA tI time
Firm Investment
Theory
Dynamic IS-LM
Punchlines
Test your understanding
**** Self Test ****
Make sure you know how to (a) deduce the dynamic

effects for the remaining variables, like Y , Y D , RS , and
RL , from the structure of the model; (b) study other
policy shocks; (c) study shocks in the good news case.
****

Firm Investment
Dynamic IS-LM
Punchlines
Punchlines
Key concept saddle-point stability.

Timing crucially important.
When is the news received by the agents?
When does the shock actually happen?
Is the shock (believed to be) permanent?
Intuitive solution principle can often yield the solution.
Policies can often have perverse effects (initially) due to
forward looking behaviour of agents.

Slide04 (Heijdra)

Uploaded by

Copyright:

Available Formats

Slide04 (Heijdra)

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

Slide04 (Heijdra)

Uploaded by

Copyright:

Available Formats

Firm Investment

Foundations of Modern Macroeconomics

Department of Economics, Econometrics & Finance

Foundations of Modern Macroeconomics - Second Edition Chapter 4 1 / 60

1 Capital accumulation decision by firms

2 Dynamic IS-LM model and the term structure of interest rates

Foundations of Modern Macroeconomics - Second Edition Chapter 4 2 / 60

Aims of this lecture

Complete our discussion of the forward looking theory of

Foundations of Modern Macroeconomics - Second Edition Chapter 4 3 / 60

Dynamic investment theory

Redo the basic model in continuous time.

π(t) ≡ F (N (t), K(t)) − w(t)N (t) − pI (t) [1 − sI (t)] Φ(I(t))

F is a CRTS production function.

Foundations of Modern Macroeconomics - Second Edition Chapter 4 5 / 60

Dynamic investment theory

Quadratic adjustment cost function:

Φ(I(t)) = I(t) + b [I(t)]2

K̇(t) = I(t) − δK(t)

δ > 0 is the depreciation rate.

Foundations of Modern Macroeconomics - Second Edition Chapter 4 6 / 60

Dynamic investment theory

Value of the firm:

Firm must choose paths for labour demand, investment, and

Foundations of Modern Macroeconomics - Second Edition Chapter 4 7 / 60

How to solve this problem?

Set up so-called current-value Hamiltonian expression (similar

HC (t) ≡ F (N (t), K(t)) − w(t)N (t) − [1 − sI (t)] Φ(I(t))

q(t) is a co-state variable (similar to a Lagrange multiplier).

Foundations of Modern Macroeconomics - Second Edition Chapter 4 8 / 60

How to solve this problem?

We get the usual result: w = FN .

(a) Shadow price of installed capital (marginal benefit of

How to solve this problem?

The first-order (necessary) condition for the capital stock:

Foundations of Modern Macroeconomics - Second Edition Chapter 4 10 / 60

How to solve this problem?

An intuitive way to write (S2) is in the form of an arbitrage

(a) The return to installed capital consists of a capital gain (q̇)

Foundations of Modern Macroeconomics - Second Edition Chapter 4 11 / 60

The effects of investment stimulation measures

Summary of the model developed so far:

We have dropped time index where no confusion is possible.

Foundations of Modern Macroeconomics - Second Edition Chapter 4 12 / 60

The effects of investment stimulation measures

There are three ways to interpret the model.

Foundations of Modern Macroeconomics - Second Edition Chapter 4 13 / 60

Capital-investment dynamics at the level of a firm

If w is constant then so is the marginal product of labour

Foundations of Modern Macroeconomics - Second Edition Chapter 4 15 / 60

The micro model

So our model at firm level simplifies to:

Foundations of Modern Macroeconomics - Second Edition Chapter 4 16 / 60

Figure 4.1: Investment with constant real wages

Foundations of Modern Macroeconomics - Second Edition Chapter 4 17 / 60

Features of the phase diagram

⇒ The line is upward sloping.

Features of the phase diagram

⇒ The line is horizontal.