Transcript Chapter 9: Inflation, Activity, and Nominal Money Growth

CHAPTER
9
Inflation, Activity,
and Nominal Money
Growth
Prepared by:
Fernando Quijano and Yvonn
Quijano
And modified by Gabriel
Martinez
The Volcker Disinflation
 In October 1979, the Fed, under Paul
Volcker, decided to reduce nominal money
growth and decrease inflation, then close to
14% per year.
 Five years later, after a deep recession,
inflation was down to 4% per year.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The Volcker Disinflation
 How did the Fed reduce inflation?
 It did it by changing the relationship between
inflation and unemployment
 It caused a recession to prove it was serious
about inflation. This changed expectations of
inflation.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
9-1
Output, Unemployment,
and Inflation
 This chapter builds on three relations:
1. Okun’s Law, which relates the change in
unemployment to output growth.
2. The Phillips curve, which relates the changes
in inflation to unemployment.
3. The aggregate demand relation, which relates
output growth to both nominal money growth
and inflation.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Output Growth, Unemployment, Inflation,
and Nominal Money Growth
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Output Growth, Unemployment, Inflation,
and Nominal Money Growth
 We are used to thinking in terms of AS-AD
 AS, which shows the effect of output on prices,
is split in this chapter into two parts:
 Output affects unemployment through Okun’s Law.
 A higher growth rate of output reduces unemployment.
 Previously, we assumed Y = N = L(1-u), but life is richer
and more complicated.
 Unemployment affects inflation through the Phillips
Curve.
 A lower unemployment rate causes inflation to rise.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Output Growth, Unemployment, Inflation,
and Nominal Money Growth
 We are still using AD.
 Higher prices lower output demanded.
 Suppose the Central Bank increases the
nominal money supply at a constant, positive
rate = 5%.
 Inflation = 3%, so the rate of growth of real
money supply is 2% = 5% – 3%.
 Suppose inflation rises unexpectedly to 4%.
 Then the real money supply will rise more
slowly, at 1% per year.
 Higher inflation reduces the rate of growth of real
money, which reduces the output growth rate.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Okun’s Law:
From
Output Growth to
Unemployment
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Okun’s Law: From
Output Growth to Unemployment
 If output grows, unemployment should
fall, right?
 Assume
Y=N
Y
= L
– U.
Yt – Yt-1 = (Lt – Lt-1) – (Ut – Ut-1)
 Assume the labor force doesn’t grow (Lt –
Lt-1=0). Then
Yt – Yt-1 = – (Ut – Ut-1)
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Okun’s Law: From
Output Growth to Unemployment
 Yt – Yt-1 = – (Ut – Ut-1)
 This also implies that the unemployment rate
(u) is negatively related to the output growth
rate (g):
Yt  Yt 1
Yt 1
ut  ut 1   g yt
 g yt
Ut
 ut
Lt
 We’ve made a lot of assumptions: no inputs
besides labor, no diminishing returns
 Particularly, we assumed no changes in labor
productivity, constant labor force, etc.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Okun’s Law: From
Output Growth to Unemployment
ut  ut 1   g yt
 The change in the unemployment rate could
be equal to the negative of the growth rate of
output.
 For example, if output growth is 4%, then the
unemployment rate should decline by 4%.
 Now, let’s be more realistic.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Okun’s Law: From
Output Growth to Unemployment
 The actual relation between output
growth and the change in the
unemployment rate is known as Okun’s
law.
 This relation allows for more realistic
production functions, labor market behavior,
etc.
 Particularly, it allows for changes in labor
productivity, and a growing labor force, etc,
so the economy can be expected to be
growing constantly.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Okun’s Law: From
Output Growth to Unemployment
Changes in the
Unemployment Rate
Versus Output
Growth in the United
States, 1970-2000
High output growth is
associated with a
reduction in the
unemployment rate;
low output growth is
associated with an
increase in the
unemployment rate.
 Using thirty years of data, the line that best fits
the data is given by: u  u   0.4( g  3%)
t
© 2003 Prentice Hall Business Publishing
t 1
Macroeconomics, 3/e
yt
Olivier Blanchard
Okun’s Law Across Countries
The coefficient β in Okun’s law gives the effect
on the unemployment rate of deviations of output
growth from normal. A value of β of 0.4 tells us
that output growth 1% above the normal growth
rate for 1 year decreases the unemployment rate
by 0.4%.
Table 1
Country
Okun’s Law Coefficients Across Countries and Time
1960-1980 β
1981-2003 β
United States
United Kingdom
Germany
Japan
© 2003 Prentice Hall Business Publishing
0.39
0.15
0.20
0.02
Macroeconomics, 3/e
0.39
0.54
0.32
0.12
Olivier Blanchard
Okun’s Law: From
Output Growth to Unemployment
ut  ut 1   0.4( g yt  3%)
 According to the equation above,
If g yt  3%, then ut  ut 1   0.4( )  0
If g yt  3%, then ut  ut 1   0.4( )  0
If g yt  3%, then ut  ut 1   0.4(0)  0
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Okun’s Law: From
Output Growth to Unemployment
 To maintain the unemployment rate
constant, output growth must be 3% per
year. This growth rate of output is called the
normal growth rate.
gy
 Output growth 1% above normal leads only
to a b%<1 reduction in unemployment.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Okun’s Law: From
Output Growth to Unemployment
ut - ut- 1 = - β( g yt - g y )
 Output growth above normal leads to a
decrease in the unemployment rate.
 If output grows below normal, the
unemployment rate. Increases.
 This is Okun’s law:
g yt  g y  ut  ut 1
g yt  g y  ut  ut 1
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Okun’s Law: From
Output Growth to Unemployment
ut  ut 1  0.4( g yt  3%) Assume u = 6%
t-1
gyt
ut
4%
5%
6%
gyt
ut
2%
1%
0%
gyt
ut
3%
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The Phillips’s Curve:
From
Unemployment to Inflation
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The Phillips Curve: From
Unemployment to Inflation
πt = π t - α(ut - un )
e
 Inflation depends on expected inflation and on
the deviation of unemployment from the natural
rate of unemployment. Suppose et is well
approximated by t-1. Then:
πt - πt- 1 = - α( ut - un )
 The Phillips curve implies that
ut  un   t   t 1
ut  un   t   t 1
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The Aggregate Demand Relation
From
Nominal Money Growth and
Inflation
To
Output Growth
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The Aggregate Demand Relation:
From Nominal Money Growth and Inflation to Output Growth
 If the IS curve is
Y
 And the LM curve is
1
c0  b0  b2i  G0 
1  c1 (1  t )  b1
d1
M
i Y
d2
d2 P
 Then the AD curve is


1
M
Y
 G0 
c0  b0  b2
1  c1 (1  t )  b1  b2 d1 / d 2  
d2 P

 Aggregate Expenditure depends on all sorts of
parameters (the c’s, the b’s, the d’s, t, and G),
positively on M and negatively on P.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The Aggregate Demand Relation:
From Nominal Money Growth and Inflation to Output Growth
 More simply, we can say that
 Mt

AD Relation Yt  Y 
, Gt , Tt 
 Pt

 Even more simply, suppose that changes in
output are caused only changes in the real
money stock, then:
Yt = γ
© 2003 Prentice Hall Business Publishing
Mt
Pt
Macroeconomics, 3/e
Olivier Blanchard
The Aggregate Demand Relation:
From Nominal Money Growth and Inflation to Output Growth
Yt = γ






Mt
Pt
Let’s put this in terms of growth rates:
gyt = (Yt-Yt-1)/Yt-1
gmt = (Mt-Mt-1)/Mt-1
 = (Pt-Pt-1)/Pt-1
And since g is a parameter (gt-gt-1)/gt-1=0.
From this we can derive
© 2003 Prentice Hall Business Publishing
g yt  gmt   t
Macroeconomics, 3/e
Olivier Blanchard
The Aggregate Demand Relation:
From Nominal Money Growth and Inflation to Output Growth
 How do we go from Y = γ
t
Mt
Pt
to g yt  gmt   t ?
 The easiest way is to use logarithms and calculus:
Log Y = log (gM/P)
Log Y = log g + log M - log P
Taking a total derivative d logY d logg d log M d log P



dY
dg
dM
dP
dY dg dM dP



Y
g
M
P
Yt  Yt 1 g t  g t 1 M t  M t 1 Pt  Pt 1



Y
g
M
P
g yt  g mt   t
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The Aggregate Demand Relation:
From Nominal Money Growth and Inflation to Output Growth
 In terms of the growth rates of output,
money, and the price level:
g yt  gmt   t
 According to the aggregate demand relation:
gmt   t  g yt  0
gmt   t  g yt  0
 Given inflation, expansionary monetary policy
leads to high output growth.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The Aggregate Demand Relation:
From Nominal Money Growth and Inflation to Output Growth
g yt  gmt   t
gmt
gyt
7%
Assume t = 3%
5%
3%
1%
Assume gmt = 6%
t
gyt
6%
© 2003 Prentice Hall Business Publishing
4%
Macroeconomics, 3/e
2%
Olivier Blanchard
The Three Relations
•
Okun’s
Law
•
Phillips
Curve
•
AD
Relation
ut - ut- 1 = - β( g yt - g y )
πt - πt- 1 = - α( ut - un )
© 2003 Prentice Hall Business Publishing
g yt  gmt   t
Macroeconomics, 3/e
Olivier Blanchard
Output Growth, Unemployment, Inflation,
and Nominal Money Growth
g yt  gmt   t
ut - ut- 1 = - β( g yt - g y )
πt - πt- 1 = - α( ut - un )
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
9-2
The Medium Run
The Medium Run
 In chapter 6 we defined the medium run
as the time when Pe=Pt-1=Pt.
 This meant that there was no reason for Pe
to change.
 No changes in Pe meant that the WS would
stay put, yielding a “steady-state”, mediumrun level of unemployment, the natural rate
of unemployment.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The Medium Run
 The medium run: Pe=Pt-1=Pt.
 In growth rates rather than levels, t = e.
 Then, by the Phillips curve, u = un.
 So inflation is constant.
 Because un is constant, so is
unemployment.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The Medium Run
 Inflation is constant.
 Because nominal money growth is a policy
variable, it changes exogenously and it is more
natural to imagine that its constant. g  g
m
m
 Then, by the aggregate demand curve,
g yt  gmt   t
output growth must be constant.
© 2003 Prentice Hall Business Publishing
g yt
Macroeconomics, 3/e
Olivier Blanchard
The Medium Run
 Is this constant g yt equal to
gy
which is normal output growth?
 It has to be. If it weren’t, u would have to
be changing, which is inconsistent with
u = un.
u - u = - β( g - g )
t
© 2003 Prentice Hall Business Publishing
t- 1
yt
y
Macroeconomics, 3/e
Olivier Blanchard
The Medium Run
 t = t-1.
 t = e.
 ut = ut-1.
 ut = u n .
 g yt  g y (t 1)

g yt  g y
 For any level of gm.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The Medium Run
 Alternatively,
 Start by assuming that
ut  ut 1
 This makes sense as a definition of the
medium run because we want the MR to be a
time of rest, stability, constancy.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The Medium Run
ut  ut 1
 So Okun’s Law implies that output grows
at its normal rate.
g yt  g y
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The Medium Run
ut  ut 1
g yt  g y
 Assume nominal money growth is
gm  gm
 We know output growth is equal to its normal rate
 Then the aggregate demand relation (g yt
gy
 gmt   t )
 implies that inflation is constant:
  gm  g y
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The Medium Run
  gm  g y
 According to the equation above, in the medium
run, inflation equals the difference between
nominal money growth and normal output
growth.
 Call gm  g y
growth.
© 2003 Prentice Hall Business Publishing
adjusted nominal money
Macroeconomics, 3/e
Olivier Blanchard
The Medium Run
 If inflation is constant, then
t = t-1,
if this is true, the Phillips curve implies
that
ut = un.
Therefore, in the medium run, the
unemployment rate must equal the
natural rate of unemployment.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The Medium Run
 Changes in nominal money growth have no
effect on output or unemployment in the
medium run, because in the medium run
ut = un and g yt  g ,y and neither un nor normal
output growth depend on the money supply.
 So changes in nominal money growth must be
reflected one for one in changes in the rate of
inflation.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The Medium Run
Inflation and
Unemployment
in the Medium Run
In the medium
run,
unemployment is
equal to the
natural rate of
unemployment, at
any level of
inflation.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The Medium Run
Inflation and
Unemployment
in the Medium Run
In the medium
run, inflation is
equal to
adjusted
nominal money
growth.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The Medium Run
Inflation and
Unemployment
in the Medium Run
In the medium
run, a decrease in
adjusted nominal
money growth
reduces inflation
at the same level
of unemployment.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The Medium Run
 Suppose
gm  7%, un  6%, and g y  2%
 In the medium run, gyt =

ut =

t =
 Adjusted money growth =
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
From the Short Run to
the Medium Run
From the Short Run to the Medium
Run
 Above we defined Okun’s Law as
ut - ut- 1 = - β( g yt - g y )
 “Unemployment falls if output grows above the normal
growth rate of output.”
 But other authors define it as
ut  un  b ( g yt  g y )
 “Cyclical unemployment arises if if output grows below
the normal growth rate of output.”
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
From the Short Run to the Medium
Run
 If we use this definition of Okun’s Law
ut  un  b ( g yt  g y )
 And we remember that the Phillips curve is
e
πt = π t - α(ut - un )
 Then we can write an “Inflation Adjustment”
curve.
 It will say that inflation rises when output is
above normal.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
From the Short Run to the Medium
Run
 An “Inflation Adjustment” curve.
 It says that inflation rises when output is above
normal.
ut  un  b ( g yt  g y )
Okun’s Law
πt = π t - α(ut - un )
e
Phillips’ Curve
 t   e  b ( g yt  g y )
Inflation-Adjustment curve
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
An “Inflation Adjustment” curve
 t    b ( g yt  g y )
e
t
IA
Higher output
growth reduces
unemployment and
increases inflation.
b

If gyt = gy,
e
t=e
 t   e  bg y  b g yt 
e  bgy
g yt
gy
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
From the Short Run to the Medium
Run
 The Aggregate Demand curve.
 (For a given rate of nominal money growth),
 The AD curve says that if inflation rises, the real
money supply grows more slowly.
 This raises interest rates, lowers the growth rate
of spending, and lowers the growth rate of
actual real output.
g yt  gmt   t
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The Aggregate Demand curve
g yt  gmt   t
t
gmt
Higher inflation
reduces the real
money supply and
reduces real output
growth.
AD
g yt
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
From the Short Run to the Medium
Run
IA
= OL + PC
 t   e  b ( g yt  g y )
t
AD
g yt  gmt   t
IA
At point A, output
growth is below the
natural rate of
output growth, and
inflation is below
expected inflation.
e
A
AD
g yt
gy
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
From the Short Run to the Medium
Run
IA
AD
g yt  gmt   t
 t   e  b ( g yt  g y )
t
In the short-run, t<e.
IA
IA
e
In the medium-run,
wage-setters will lower
their expectations of
inflation, which shifts
the IA curve down.
A
This happens until
gyt = gy.
AD
g yt
gy
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
From the Short Run to the Medium
Run
t
Imagine the Central
Bank shifted the AD
curve up repeatedly by
raising gmt in order to
get short-run reductions
in unemployment.
IA
In the medium-run, the
IA curve will shift up
over and over, keeping
gyt = gy.
gyt = gy means that
AD
u=un,
gy
© 2003 Prentice Hall Business Publishing
g yt
Macroeconomics, 3/e
At any level
of gmtBlanchard
and t.
Olivier
From the Short Run to the Medium
Run
In the medium run, t=e
The short-run relation between
output growth and inflation
disappears and the IA
equation    e  b ( g  g )
t
yt
y
t
becomes
g yt  g y
While AD determines inflation.
AD
g yt
gy
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
From the Short Run to the Medium
Run
 In the short run,
 Aggregate demand influences output: output
grows faster if inflation is below nominal money
supply growth.
 Inflation adjusts upward if output grows above
the normal rate.
 In the medium run,
 Expectations of inflation adjust so g yt  g y and
u=un.
 AD only determines inflation.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The Medium Run
Inflation and
Unemployment
in the Medium Run
In the medium
run, a decrease in
adjusted nominal
money growth
reduces inflation
at the same level
of unemployment.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The Medium Run
 Suppose gm  7%, un  6%, and g y  2%
 The CB changes g m to 3%
 In the medium run, gyt =

ut =

t =
 Adjusted money growth =
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The Effects of Money Growth
The Effects of Money Growth
 Okun’s law relates the change in the
unemployment rate to the deviation of output
growth from normal:
t
t  1
yt
y
u u
  b( g  g )
 The Phillips curve relates the change in
inflation to the deviation of the unemployment
rate from the natural rate:
t  t  1    (ut  un)
 The aggregate demand relation relates output
growth to the difference between nominal
money growth and inflation.
gyt  gmt  t
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The Effects of Money Growth
Output Growth, Unemployment,
Inflation, and Nominal Money
Growth
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The Medium Run
Assume that the central bank maintains a constant
growth rate of nominal money, call it gm. In this case,
the values of output growth, unemployment, and
inflation in the medium run:
 Output must grow at its normal rate of growth, g y
 If we define adjusted nominal money growth as equal to
nominal money growth minus normal output growth, then
inflation equals adjusted nominal money growth.
 t  g mt  g yt
 The unemployment rate must equal to the natural rate of
unemployment.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The Short Run
Now suppose that the central bank decides
to decrease nominal money growth. What will
happen in the short run?
 Given the initial rate of inflation, lower nominal
money growth leads to lower real nominal money
growth , and thus to a decrease in output growth.
 Now, look at Okun’s law, output growth below
normal leads to an increase in unemployment.
 Now, look at the Phillips curve relation.
Unemployment above the natural rate leads to a
decrease in inflation.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The Short Run
In words: In the short run, monetary tightening
leads to a slowdown in growth and a temporary
increase in unemployment. In the medium run,
output growth returns to normal, and the
unemployment rate returns to the natural rate.
Table 9-1
The Effects of Monetary Tightening
Year 0
1 Real money growth %
Year 1
Year 2
Year 3
(gm-π)
3.0
0.5
5.5
3.0
2 Output growth %
(gy)
3.0
0.5
5.5
3.0
3 Unemployment rate %
(u)
6.0
7.0
6.0
6.0
4 Inflation gate %
(π)
5.0
4.0
4.0
4.0
5 Nominal money growth %
(gm)
8.0
4.5
9.5
7.0
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
9-3
Disinflation
Disinflation
 To achieve lower inflation, the rate of nominal money
growth must be reduced.
 This implies a (possibly long) transition between one
“medium-run” equilibrium and another “medium-run”
equilibrium.
 This transition happens in the short run, so the
downward sloping (Original) Phillips Curve becomes
relevant again.
 Disinflation moves the economy (in the short run)
along the short-run (Original) Phillips Curve.
 In the medium run, Unemployment above natural
causes a shift down of the Phillips curve, until
unemployment = un in the medium run.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The Medium Run
Inflation and
Unemployment
in the Medium Run
In the medium
run, inflation is
equal to
adjusted
nominal money
growth.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Disinflation
Inflation and
Unemployment
in the Medium Run
In the short run,
inflation is decreased
by increasing
unemployment.
In the medium run,
higher unemployment
and lower inflation
cause a fall in
expected inflation
and a shift down of
the SR Phillips
Curve.
© 2003 Prentice Hall Business Publishing
Decrease in
expectations of
inflation.
Macroeconomics, 3/e
Olivier Blanchard
Disinflation
 To achieve lower inflation, the rate of nominal money
growth must be reduced. Here is what happens in the
Short Run:
 In the aggregate demand relation,
g yt  gmt   t
 gm  ( gm   )  g y 
 Then, from Okun’s law,
ut - ut- 1 = - β( g yt - g y )
 gy  u 
πt - πt- 1 = - α( ut - un )
 Finally, according to the Phillips curve relation:
 u 
•
Notice that now u > un.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Disinflation
 But over time, in the Medium Run,
 According to the Phillips curve relation: π - π = - α( u - u )
t
t- 1
t
n
u  un   
 As  falls, it falls far enough below gm.
In the aggregate demand relation,
g yt  gmt   t
  gm  g y  0
 Eventually gy rises enough that gy > gy.
ut - ut- 1 = - β( g yt - g y )
Then, from Okun’s law,
gy  gy  u 
 After a decrease in nominal money growth, unemployment
first increases, but eventually it starts decreasing.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Working Out the Path
of Nominal Money Growth
Table 9-1 Engineering Disinflation
0
1
2
3
4
5
6
7
8
Inflation (%)
14
12
10
8
6
4
4
4
4
Nominal money
growth (%)
17
10
13
11
9
7
12
7
7
Output growth
(%)
3
2
3
3
3
3
8
3
3
Unemployment
rate (%)
6
8
8
8
8
8
6
6
6
The Central Bank wants to cut inflation from 14% to
4%. To do this, it cuts nominal money growth
radically, which reduces output growth and
increases unemployment. Inflation gradually falls.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
How Much Unemployment?
and for How Long?
πt - πt- 1 = - α( ut - un )
 In the Phillips curve relation above, disinflation—
a decrease in inflation—can be obtained only at
the cost of higher unemployment.
( πt - πt- 1 ) < 0 Þ ( ut - un ) > 0 Þ ut > un
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
How Much Unemployment?
and for How Long?
 Do we have any idea of the amount of
unemployment we must inflict on an
economy to reduce the inflation rate?
 Are there measures of this sacrifice?
 What are the determinants?
 How should we design disinflation
programs?
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
How Much Unemployment?
and for How Long?
πt - πt- 1 = - α( ut - un )
 For example, let’s assume that  =1/2
 Then reducing inflation by 10 percentage points
requires increasing unemployment above the
natural rate by 20 percentage points over a
number of years:
 Notice that while the left-hand side is a year-per-year
change, the right-hand side is just the difference
between two variables, one of which doesn’t change
with time.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
How Much Unemployment?
and for How Long?
πt - πt- 1 = - α( ut - un )
 If  =1/2, reducing inflation by 10 percentage
points requires 20 percentage points of excess
unemployment over a number of years:
 Suppose we want to achieve the disinflation
over 5 years: then we need 5 years of
unemployment at 4 percentage points above
the natural rate.
 Achieving the disinflation over 10 years means
10 years of unemployment at 2 percentage
points above the natural rate.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
How Much Unemployment?
and for How Long?
πt - πt- 1 = - α( ut - un )
 If  =0.8, reducing inflation by 10 percentage
points requires 12.5 percentage points of excess
unemployment over a number of years:
 Achieving the disinflation over 2 years means
2 years of unemployment at 6.5 percentage
points above the natural rate.
 Achieving the disinflation over 25 years means
25 years of unemployment at 0.5 percentage
points above the natural rate.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
How Much Unemployment?
and for How Long?
 A point-year of excess unemployment is
a difference between the actual and the
natural unemployment rate of one
percentage point for one year.
 So if  =1/2, reducing inflation by 10 percentage
points requires 20 points-years of excess
unemployment.
 If  =0.8, reducing inflation by 10 percentage
points requires 12.5 points-years of excess
unemployment :
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
How Much Unemployment?
and for How Long?
πt - πt- 1 = - α( ut - un )
 The sacrifice ratio (=1/) is the number of pointyears of excess unemployment needed to achieve
a decrease in inflation of 1%.
Sacrifice Ratio 
Point- yearsof excess unemployme
nt
1 perc.pointdecreasein inflation
 For example, if the sacrifice ratio is 1.32, then
a 10% disinflation requires 13.2 point-years of
excess unemployment.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
How Much Unemployment?
and for How Long?
 If inflation is a bad thing and the number
of point-years of excess unemployment
is unchangeable (because  is fixed),
why not “get it over with” in one year?
 At the very temporary cost of high
unemployment.
 This policy would have the great benefit of
full credibility: there’s no need to wonder if
the disinflation program will continue.
 This works if the announcement of the
policy immediately changes e.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
How Much Unemployment?
and for How Long?
 But if e = t-1, then seeing is believing,
and you need at least two years.
 Moreover, the output loss could be huge,
by Okun’s Law.
 Many of the effects of the recession would
be permanent: discouraged workers,
bankruptcies, political instability, etc.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
How Much Unemployment?
and for How Long?
 Suppose the Central Bank wishes to reduce
inflation by 9%.
 If  = 1.15, what is the number of point-years of excess
unemployment?
 Given the goal of reducing inflation by 9%, can the
Central Bank affect the number of point-years of excess
unemployment calculated above?

1.5
1.3
1.15
1
0.9
Sacrifice ratio
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Working Out the Path
of Nominal Money Growth
 An important question for policy makers
is what is the optimal path of money
growth to achieve a disinflation.
 This is worked out in this way:
 The path of inflation shows the values of
inflation before achieving a desired 4%.
 The path of unemployment shows the
unemployment required to achieve the
decrease in inflation.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Working Out the Path
of Nominal Money Growth
 The path of output shows the output
growth required to achieve the required
path of unemployment.
 The path of nominal money growth
shows the growth required to achieve the
required path of output.
gm  g y  ut  ut 1   t   t 1
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Working Out the Path
of Nominal Money Growth
Table 9-1 Engineering Disinflation
0
1
2
3
4
5
6
7
8
Inflation (%)
14
12
10
8
6
4
4
4
4
Nominal money
growth (%)
17
10
13
11
9
7
12
7
7
Output growth
(%)
3
2
3
3
3
3
8
3
3
Unemployment
rate (%)
6
8
8
8
8
8
6
6
6
This table shows the path of nominal money growth
needed to achieve 10% disinflation over five years,
which we assume requires 10 point-years of excess
unemployment. That is, u > un by 2 points every
year for 5 years.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Working Out the Path
of Nominal Money Growth
Table 9-1 Engineering Disinflation
0
1
2
3
4
5
6
7
8
Desired path of
Inflation (%)
14
12
10
8
6
4
4
4
4
Unemployment
rate (%)
6
Output growth
(%)
3
Nominal money
growth (%)
17
g y  gm  
ut  ut 1   0.4( g yt  3%)
© 2003 Prentice Hall Business Publishing
   un  ut  =1
Macroeconomics, 3/e
Olivier Blanchard
Working Out the Path
of Nominal Money Growth
Table 9-1 Engineering Disinflation
0
1
2
3
4
5
6
7
8
Inflation (%)
14
12
10
8
6
4
4
4
4
Nominal money
growth (%)
17
10
13
11
9
7
12
7
7
Output growth
(%)
3
2
3
3
3
3
8
3
3
Unemployment
rate (%)
1
6
8
8
8
8
8
6
6
6
If (Okun’s Law) ut  ut 1   0.4( g yt  3%) , then gy must fall by 5 points the
first year (2 / (-0.4)) = -5 below normal growth, from 3 to -2 percent.
Raising unemployment by 2 points means lowering inflation by 2 points.
Because g  g   ,
y
m
-5 = gm – (– 2), then gm = – 7.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Working Out the Path
of Nominal Money Growth
Table 9-1 Engineering Disinflation
0
1
2
3
4
5
6
7
8
Inflation (%)
14
12
10
8
6
4
4
4
4
Nominal money
growth (%)
17
10
13
11
9
7
12
7
7
Output growth
(%)
3
2
3
3
3
3
8
3
3
Unemployment
rate (%)
6
8
8
8
8
8
6
6
6
The second year, ut=ut-1. By Okun’s Law, ut  ut 1   0.4( g yt  3%)
2
gy must go back to normal growth, 3%. Because still u – un =
2 points, inflation falls by by 2 points, to 10%. From the AD
relation g y  gm   , 5 = gm – (-2), then gm = 3.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Working Out the Path
of Nominal Money Growth
Notice that although money
growth rises in year 7,
inflation does not :
20
the reason is that
unemployment goes back to
its natural level.
15
10
5
0
1
2
3
4
5
6
7
8
9
-5
Inflation (%)
Nominal money growth (%)
Output growth (%)
Unemployment rate(%)
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Working Out the Path
of Nominal Money Growth
A Disinflation Path
Five years of
unemployment above
the natural rate of
unemployment lead to
a permanent decrease
in inflation.
This figure shows a path of unemployment and
inflation similar to the disinflation path in Table 9-1.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Working Out the Path
of Nominal Money Growth
Credible Disinflation
causes the Phillips
curve to shift down.
But remember what
determines the
position of the
Phillips curve:
expectations of
inflation.
If expectations
change, this works.
If they don’t it
doesn’t.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Expectations, Credibility,
9-4
and Nominal Contracts
 This section examines how changes in
expectation formation might affect the
unemployment cost of disinflation.
 Two separate groups of macroeconomists
challenge the traditional notion that policy
can change the timing, but not the number
of point-years of excess unemployment.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Expectations and Credibility:
The Lucas Critique
 The Lucas critique states that it is
unrealistic to assume that wage setters
would not consider changes in policy when
forming their expectations.
 If wage setters could be convinced that inflation
was indeed going to be lower than in the past,
they would decrease their expectations of
inflation, which would in turn reduce actual
inflation, without the need for a change in the
unemployment rate.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Expectations and Credibility:
The Lucas Critique
 Thomas Sargent, who worked with Robert
Lucas, argued that any in order to achieve
disinflation, any increase in unemployment
would have to be only small.
 The essential ingredient of successful
disinflation, he argued, was credibility of
monetary policy—the belief that the central
bank was truly committed to reducing
inflation. The central bank should aim for
fast disinflation.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Expectations and Credibility:
The Lucas Critique
 Recall that, although we assumed the Phillips curve
can be approximated by
πt - πt- 1 = - α( ut - un )
it is really
πt = π t - α(ut - un )
e
Our calculations assumed that agents didn’t form
expectations based on policy, just on history.
 What if policy were fully credible, so that if the CB
announces a future  =4%, e becomes 4%, wagesetters set their nominal wage increase at 4%, and
price-setters set price increases at 4%. Inflation
becomes 4% instantaneously, and u=un.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Expectations and Credibility:
The Lucas Critique
 Which program is more credible, taking
into account political pressures,
elections, etc.?
 A disinflation that happens in one year
(say, the first year out of a 4-year
presidential period), and then you’re
done with it?
 Or a disinflation that is announced to
start today and end in 20 years?
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Working Out the Path
of Nominal Money Growth
If the Central Bank
is so credible that its
ultimate forecasts of
inflation are
believed, inflation
expectations fall
right away to their
final level.
1
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Normal Rigidities and Contracts
 A contrary view was taken by Stanley
Fischer and John Taylor. They emphasized
the presence of nominal rigidities, or the
fact that many wages and prices are not
readjusted when there is a change in policy.
 If wages are set before the change in policy,
inflation would already be built into existing
wage agreements.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Normal Rigidities and Contracts
 Taylor argued that the staggering of wage decisions
imposed strong limits on how fast disinflation could
proceed.
 The way to decrease the unemployment cost of
disinflation is to give wage setters time to take the
change in policy into account.
 “Inflation won’t change much over the next year or two (to
avoid costs in output). But in two years, inflation will begin to
fall drastically.”
 If the government makes this announcement, how would you
negotiate your wages?
 Slow but credible disinflation might have a lower cost.
The central bank should go for slow disinflation.
 If inflation ain’t changin’, why should we believe it will?
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
Normal Rigidities and Contracts
Disinflation Without
Unemployment
in the Taylor Model
With staggering of
wage decisions,
disinflation must be
phased in slowly to
avoid an increase in
unemployment.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The
U.S.
Disinflation,
9-5
1979-1985
 What did happen in the early eighties?
 The U.S. disinflation of the early 1980s was
associated with a substantial increase in
unemployment.
 The Phillips curve relation proved more
robust than many economists anticipated.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The U.S. Disinflation,
1979-1985
Table 9-2 Inflation and Unemployment, 1979-1985
Percent
1979
1980
1981
1982
1983
1984
1985
GDP growth
2.5
0.5
1.8
2.2
3.9
6.2
3.2
Unemployment rate
5.8
7.1
7.6
9.7
9.6
7.5
7.2
CPI inflation
13.3
12.5
8.9
3.8
3.8
3.9
3.8
Cumulative
unemployment
1.0
2.6
6.3
9.9
11.4
12.6
Cumulative disinflation
0.8
4.4
9.5
9.5
9.4
9.5
Sacrifice ratio
1.25
0.59
0.66
1.04
1.21
1.32
Cumulative unemployment is the sum of point-years of excess unemployment
from 1980 on, assuming a natural rate of unemployment of 6%. Cumulative
disinflation is the difference between inflation in a given year and inflation in
1979. The sacrifice ratio is the ratio of cumulative unemployment to cumulative
disinflation.
© 2003 Prentice
Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The U.S. Disinflation, 1979-1985
The Federal Funds Rate
and Inflation, 1979-1984
A sharp increase in the
interest rate from
September 1979 to April
1980 was followed by a
sharp decline in mid 1980,
and then a second and
sustained increase from
January 1981 on, lasting
for most of 1981 and
1982.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard
The U.S. Disinflation,
1979-1985
 Laurence Ball, who examined 65 disinflation
episodes concluded that:
 Disinflations typically lead to a period of higher
unemployment.
 This contradicts a radical version of Lucas/Sargent.
 Faster disinflations are associated with smaller
sacrifice ratios.
 This supports a moderate version of Lucas/Sargent.
 Sacrifice ratios are smaller in countries that
have shorter wage contracts.
 This supports Fischer/Taylor.
© 2003 Prentice Hall Business Publishing
Macroeconomics, 3/e
Olivier Blanchard