Download Lecture 20 Keynesian Model and Policy Analysis

Lecture 20
Keynesian Model and Policy Analysis
Noah Williams
University of Wisconsin - Madison
Economics 702/312
Williams
Economics 702/312
Aggregate Demand and Supply Curves
Can define aggregate demand AD as demand for current
output dependent on price level. Derive by changing P
which shifts LM , trace out effect on Y for fixed IS.
In discussion so far have considered horizontal (short-run)
aggregate supply curve AS. With sticky prices (fixed in
short run), assume that producers meet whatever demand
at the current price.
Firms’ effective labor demand then determines output.
With P = P̄ fixed, AD determines Y . With current
K = K̄ , find labor demand from production
Y = F (K̄ , N ) → N = F −1 (Y ).
In long-run prices adjust to clear labor market, so long-run
aggregate supply curve is vertical: money is neutral in
long-run.
Williams
Economics 702/312
Figure 12.11
The Aggregate
Demand Curve
12-12
Copyright © 2005 Pearson Addison-Wesley. All rights reserved.
Williams
Economics 702/312
Figure 12.12
A Shift to the
Right in the IS
Curve Shifts the
AD Curve to the
Right
12-13
Copyright © 2005 Pearson Addison-Wesley. All rights reserved.
Williams
Economics 702/312
Figure 12.13
A Shift to the
Right in the LM
Curve Shifts the
AD Curve to the
Right
12-14
Copyright © 2005 Pearson Addison-Wesley. All rights reserved.
Williams
Economics 702/312
P
AD
SRAS
P*
Y*
Y
Short-run equilibrium in the Keynesian Model:
Sticky prices.
Williams
Economics 702/312
Labor, N
Output, Y
F(K*,N)
Output, Y
Labor, N
Production
Effective Labor Demand
Williams
Economics 702/312
P
LRAS
AD
SRAS
P*
Y*
Y
Long-run equilibrium in the Keynesian Model:
Sticky prices.
Williams
Economics 702/312
Nominal Wage Rigidities
Other sources of nominal rigidities lead to similar effects.
The book considers the case of nominal wage stickiness.
Wages may be fixed in the short-run due to nominal wage
contracts. Ex.: union contracts typically set nominal wages
for one year at a time.
This was the original argument of Keynes. Leads to a
different aggregate supply curve.
With flexible prices, sticky nominal wages, aggregate
supply curve is upward sloping.
Williams
Economics 702/312
Figure 12.3 Construction of the
Aggregate Supply Curve
12-4
Copyright © 2005 Pearson Addison-Wesley. All rights reserved.
Williams
Economics 702/312
Figure 12.14 The Keynesian Sticky
Wage Model
12-15
Copyright © 2005 Pearson Addison-Wesley. All rights reserved.
Williams
Economics 702/312
Figure 12.15 An Increase in the Money
Supply in the Sticky Wage Model
12-16
Copyright © 2005 Pearson Addison-Wesley. All rights reserved.
Williams
Economics 702/312
Real Wage Rigidities
There are also situations in which the real wage itself is
rigid. The prime example is the efficiency wage model due
to Shapiro and Stiglitz (1984).
Effort put forth by worker depends on wage e(w). Effort is
not directly observable by employers.
Workers who feel well treated will work harder and more
efficiently: “the carrot”.
Workers who are well paid won’t risk losing their jobs by
being caught shirking: “the stick”.
e(w) is S-shaped: increases in w, flattens out at high w.
Williams
Economics 702/312
Figure 16.17 Effort of the Worker as
a Function of His or Her Wage
16-19
Copyright © 2005 Pearson Addison-Wesley. All rights reserved.
Williams
Economics 702/312
Efficiency Wages
Firms take as given effort curve. Problem now:
max F (K , e(w)N ) − wN
N
First order condition: e(w)FN (K , e(w)N ) = w. Gives firm
labor demand for any real wage.
But how to choose wage to set? Want to minimize cost of
inducing effort. Choose efficiency wage to maximize
e(w)/w.
max e(w)/w ⇒ w = e(w)/e 0 (w)
w
Note the real wage is then rigid. Changes in labor supply
then only affect level of unemployment, not employment.
No downward pressure on wages with excess labor supply,
since if firms reduce wages effort will decline.
Williams
Economics 702/312
Figure 16.20 Determination of the
Efficiency Wage
16-22
Copyright © 2005 Pearson Addison-Wesley. All rights reserved.
Williams
Economics 702/312
Implications of Real Wage Rigidity
Efficiency wage leads to vertical output supply curve Y s
(FE): w determined by efficiency considerations, then N
determined, giving Y s via production. Does not depend on
N s , hence no dependence on r.
Effect on aggregate depends on whether prices are sticky or
not. Typically assume sticky prices as well: horizontal
SRAS.
The model is qualitatively like the sticky price model but
now long-run equilibrium is independent of labor supply.
Williams
Economics 702/312
Figure 16.21 Unemployment in the
Efficiency Wage Model
16-23
Copyright © 2005 Pearson Addison-Wesley. All rights reserved.
Williams
Economics 702/312
Figure 16.22 The Output Supply
Curve in the Efficiency Wage Model
16-24
Copyright © 2005 Pearson Addison-Wesley. All rights reserved.
Williams
Economics 702/312
r
FE
LM
IS
r*
Y*
Y
Long-run equilibrium in the Keynesian Model:
Efficiency wages
Williams
Economics 702/312
Effects of Fiscal Policy
Increase in government spending: Shifts the IS curve out.
In short run, Y and r interest rate increase.
The multiplier mg = ∆Y
∆G gives the increase in output due
to government spending.
Much original Keynesian theory thought mg > 1.
Government spending leads to higher private output.
With efficiency wages, FE line unaffected. Labor supply
only determines unemployment.
As prices adjust, LM curve shifts up. New general
equilibrium has higher r, same Y .
Williams
Economics 702/312
r
IS
FE
IS’
LM(P)
G increase
r’
r*
Y*
Y’
Y
Short-run Effect of Increase in Govt Spending
Williams
Economics 702/312
LM(P’)
r
IS
FE
IS’
LM(P)
G increase
P increase
r’
r*
Y*
Y’
Y
Long-run Effect of Increase in Govt Spending
Williams
Economics 702/312
Effects of Monetary Policy
Increase in money supply: not neutral in short run. Shifts
LM curve out, increasing Y decreasing r.
Shifts AD curve out. Effect on prices depends on whether
they’re sticky or not.
With sticky prices no effect on price. With sticky wages,
price level increases, shifting LM curve part of way back in.
In long run, prices or nominal wages adjust, shifting LM
curve and SRAS curve. Y decreases and r increases to
original levels.
Short-run non-neutralities lead to role of monetary policy
in smoothing responses to shocks: stabilization policy.
Williams
Economics 702/312
LM(P)
FE
r
LM’(P)
IS
M increase
r*
r’
Y*
Y’
Y
Short-run Effect of Increase in Money Supply
Williams
Economics 702/312
r
LM(P)=LM(P’)
LM’(P)
FE
IS
M increase
P increase
r*
r’
Y*
Y’
Y
Long-run Effect of Increase in Money Supply:
Money is neutral in the long-run.
Williams
Economics 702/312
Stabilization Policy
Classical theory gave no role for government to offset
shocks. At best, government policy was neutral.
In Keynesian model, since markets may be out of
equilibrium in the short run, role for government to smooth
fluctuations.
Example: Reduction in investment demand shifts the IS
curve down. Keynes’s “animal spirits”. Causes recession in
short run.
If government does nothing, price level will eventually
decline, shifting LM . Output decline will endure for a
while.
Govt. could increase M , shifting LM curve down to general
equilibrium. Same as doing nothing but happens faster.
Govt. could increase G, shifting IS curve back up.
Taking policy action has potential benefit of shortening
recession. Leads to higher price level than if no action.
Williams
Economics 702/312
r
IS’
FE
IS
LM(P)
Shock
r*
r’
Y’
Y*
Y
Short-run Effect of Shock to IS Curve
Williams
Economics 702/312
r
IS’
FE
IS
LM(P)
LM(P’)
Shock
P fall or M increase
r*
r’
r*’
Y’
Y*
Y
Response to Shock: Nothing or Monetary Policy
Williams
Economics 702/312
r
IS’
FE
IS
LM(P)
Shock
G increase
r*
r’
Y’
Y*
Y
Response to Shock: Fiscal Policy
Williams
Economics 702/312
P
LRAS
AD’
AD
Shock
SRAS
P*
Y’
Y*
Y
Effect of the IS Shock: AD-AS
Williams
Economics 702/312
P
LRAS
AD’
AD
Shock
P adjusts
SRAS
P*
SRAS’
P’
Y’
Y*
Y
Effect of the IS Shock: Nothing
Williams
Economics 702/312
P
LRAS
AD’
AD
Shock
M or G increase
SRAS
P*
Y’
Y*
Y
Effect of the IS Shock: Monetary or Fiscal Policy
Williams
Economics 702/312
Limitations on Stabilization & Liquidity Traps
Generally have monetary policy smooth cyclical
fluctuations. Fiscal policy takes longer to implement,
money is more direct.
Limitation on monetary policy: nominal interest rates are
bounded at zero. Can’t set nominal interest rates negative:
everyone would want to borrow.
Means LM curve very flat near zero interest rates. So
increases in money supply have little effect at low rates: a
liquidity trap.
Very relevant for many economies: Japan, US, Europe.
From 1960 to 1990, Japan’s economy grew over 6% per
year. But the Japanese economy slumped in the 1990s,
with growth near zero. Stock and land prices fell from
excessive levels, hurting banks. Banks’ financial distress
caused lending to fall, reducing investment.
Williams
Economics 702/312
Growth in Interest Rates in Japan
Japan: Annualized Growth of Real GDP, 1993−2004
4
2
0
−2
−4
1994
1996
1998
2000
2002
2004
Japan: Money Market Inerest Rate, 1993−2004
3.5
3
2.5
2
1.5
1
0.5
0
1994
1996
1998
Williams
2000
2002
Economics 702/312
2004
r=R
IS
IS’
FE
LM(P)
Shock
r*
0
Y’
Y*
A Liquidity Trap: Increase in M Powerless
Williams
Economics 702/312
Policy in a Liquidity Trap
Interest rates near zero and remained there. Low inflation
or even deflation, so real rates also near zero.
Possible to run expansionary fiscal policy to shift IS curve
to restore equilibrium.
Was tried but unsuccessful – some say it wasn’t enough. In
combination with expansionary
Problem with monetary policy is zero (expected) inflation.
If could engineer an inflation, then possible to have
equilibrium with negative real rates but positive nominal
(Krugman).
Also no reason to force households to consume.
Not clear how to commit to it – most central banks like
Bank of Japan known inflation fighter. Possible to do so by
depreciating currency (Svensson).
Williams
Economics 702/312
r=R
IS
IS’
FE
LM(P)
Shock
G increase
r*
0
Y’
Y*
A Liquidity Trap: Fiscal Policy
Williams
Economics 702/312
R=r+i
r
IS
IS’
FE
LM
LM’
Shock
r*
i
0
r’<0
0
Y’
Y*
A Liquidity Trap: Increase Inflation
Williams
Economics 702/312
US Policy Recently
Fed has recently raised interest rates after having kept
them at near zero for extended period of time.
Fed has also backed away from explicit short-term inflation
target, seeming to be willing to allow inflation to rise in
short run. Has stressed that 2% inflation target is for the
“medium term”.
Fed has become more explicit recently, announcing forecast
paths for future interest rates. (Dot plots)
All of these steps geared toward guiding expectations, and
committing to allowing inflation to rise. Inflation has
remained low, but success of policy would require low rates
even when inflation picks up. Given dissent on Board, in
press, profession, not clear this would happen.
Others have argued that Fed should raise inflation target
as long as economy remains below trend growth. Potential
problem on how to reduce inflation once it takes hold.
Williams
Economics 702/312