The Short-Run Trade-off Between

35 Inflation and Unemployment

In this chapter, look for the answers to
these questions:
▪ How are inflation and unemployment related in the
short run? In the long run?
▪ What factors alter this relationship?
▪ What is the short-run cost of reducing inflation?
▪ Why were U.S. inflation and unemployment both
so low in the 1990s?

1
 Introduction
▪ In the long run, inflation & unemployment are
unrelated:
• The inflation rate depends mainly on growth in
the money supply.
• Unemployment (the “natural rate”) depends on
the minimum wage, the market power of unions,
efficiency wages, and the process of job search.
▪ In the short run,
society faces a trade-off between
inflation and unemployment.

2
 The Phillips Curve
▪ Phillips curve: shows the short-run trade-off
between inflation and unemployment
▪ 1958: A.W. Phillips showed that
nominal wage growth was negatively
correlated with unemployment in the U.K.
▪ 1960: Paul Samuelson & Robert Solow found
a negative correlation between U.S. inflation
& unemployment, named it “the Phillips Curve.”

3
 Deriving the Phillips Curve
▪ Suppose P = 100 this year.
▪ The following graphs show two possible
outcomes for next year:
A. Agg demand low,
small increase in P (i.e., low inflation),
low output, high unemployment.
B. Agg demand high,
big increase in P (i.e., high inflation),
high output, low unemployment.

4
 Deriving the Phillips Curve
A. Low agg demand, low inflation, high u-rate
P inflation

SRAS
B B
5%
105
A
103 3% A
AD2
PC
AD1

Y1 Y2 Y 4% 6% u-rate

B. High agg demand, high inflation, low u-rate

5
 The Phillips Curve: A Policy Menu?
▪ Since fiscal and mon policy affect agg demand,
the PC appeared to offer policymakers a menu
of choices:
• low unemployment with high inflation
• low inflation with high unemployment
• anything in between
▪ 1960s: U.S. data supported the Phillips curve.
Many believed the PC was stable and reliable.

6
Evidence for the Phillips Curve?
During the 1960s,
U.S. policymakers
opted for reducing
unemployment
at the expense of
higher inflation

7
 The Vertical Long-Run Phillips Curve

▪ 1968: Milton Friedman and Edmund Phelps

argued that the tradeoff was temporary.
▪ Natural-rate hypothesis: the claim that
unemployment eventually returns to its normal or
“natural” rate, regardless of the inflation rate
▪ Based on the classical dichotomy and the
vertical LRAS curve.

8
 The Vertical Long-Run Phillips Curve
In the long run, faster money growth only causes
faster inflation.
P inflation
LRAS LRPC

high
P2 infla-
tion
P1 AD2 low
infla-
AD1 tion
Y u-rate
natural rate natural rate of
of output unemployment
9
 Reconciling Theory and Evidence
▪ Evidence (from ’60s):
PC slopes downward.
▪ Theory (Friedman and Phelps’ work):
PC is vertical in the long run.
▪ To bridge the gap between theory and evidence,
Friedman and Phelps introduced a new variable:
expected inflation – a measure of how much
people expect the price level to change.

10
 The Phillips Curve Equation
Natural
Unemp. Actual Expected
= rate of – a –
rate inflation inflation
unemp.

Short run
Fed can reduce u-rate below the natural u-rate
by making inflation greater than expected.
Long run
Expectations catch up to reality,
u-rate goes back to natural u-rate whether inflation
is high or low.

11
 How Expected Inflation Shifts the PC
Initially, expected &
actual inflation = 3%,
inflation
unemployment = LRPC
natural rate (6%).
Fed makes inflation
2% higher than expected, B C
5%
u-rate falls to 4%.
3% A
In the long run,
expected inflation PC2
increases to 5%, PC1
PC shifts upward,
4% 6% u-rate
unemployment returns to
its natural rate.
12
ACTIVE LEARNING 1:
Exercise
Natural rate of unemployment = 5%
Expected inflation = 2%
Coefficient a in PC equation = 0.5
A. Plot the long-run Phillips curve.
B. Find the u-rate for each of these values of actual
inflation: 0%, 6%. Sketch the short-run PC.
C. Suppose expected inflation rises to 4%.
Repeat part B.
D. Instead, suppose the natural rate falls to 4%.
Draw the new long-run Phillips curve,
then repeat part B.
13
ACTIVE LEARNING 1:
LRPCD
Answers
PCB LRPCA
7
An increase
in expected 6
inflation
5
inflation rate
shifts PC to
the right. 4
PCD
3
PCC
A fall in the 2
natural rate
1
shifts both
curves 0
to the left. 0 1 2 3 4 5 6 7 8
unemployment rate
14
The Breakdown of the Phillips Curve

Early 1970s:
unemployment increased,
despite higher inflation.
Friedman & Phelps’
explanation:
expectations were
catching up with
reality.

15
 Another PC Shifter: Supply Shocks
▪ Supply shock:
an event that directly alters firms’ costs and
prices, shifting the AS and PC curves
▪ Example: large increase in oil prices

16
How an Adverse Supply Shock Shifts the PC
SRAS shifts left, prices rise, output & employment fall.
P inflation
SRAS2

SRAS1
B B
P2

P1 A A
PC2

AD PC1
Y2 Y1 Y u-rate

Inflation & u-rate both increase as the PC shifts upward.

17
 The 1970s Oil Price Shocks

Oil price per barrel The Fed chose to

1/1973 $ 3.56 accommodate the
first shock in 1973
1/1974 10.11
with faster money growth.
1/1979 14.85
Result:
1/1980 32.50
Higher expected inflation,
1/1981 38.00 which further shifted PC.
1979:
Oil prices surged again,
worsening the Fed’s tradeoff.

18
The 1970s Oil Price Shocks

Supply shocks & rising expected

inflation worsened the PC tradeoff.

19
 The Cost of Reducing Inflation
▪ Disinflation: a reduction in the inflation rate
▪ To reduce inflation,
Fed must slow the rate of money growth,
which reduces agg demand.
▪ Short run: output falls and unemployment rises.
▪ Long run: output & unemployment return to
their natural rates.

20
 Disinflationary Monetary Policy
Contractionary monetary
policy moves economy
inflation
from A to B. LRPC
Over time,
expected inflation falls, A
PC shifts downward.
B
In the long run, C
point C: PC1
the natural rate PC2
of unemployment,
and lower inflation. u-rate
natural rate of
unemployment
21
 The Cost of Reducing Inflation
▪ Disinflation requires enduring a period of
high unemployment and low output.
▪ Sacrifice ratio: the number of percentage points
of annual output lost in the process of reducing
inflation by 1 percentage point
▪ Typical estimate of the sacrifice ratio: 5
• Reducing inflation rate 1% requires a sacrifice
of 5% of a year’s output.
▪ This cost can be spread over time. Example:
To reduce inflation by 6%, can either
• sacrifice 30% of GDP for one year
• sacrifice 10% of GDP for three years
22
Rational Expectations, Costless Disinflation?
▪ Rational expectations: a theory according to
which people optimally use all the information
they have, including info about govt policies,
when forecasting the future
▪ Early proponents:
Robert Lucas, Thomas Sargent, Robert Barro
▪ Implied that disinflation could be much less
costly…

23
Rational Expectations, Costless Disinflation?
▪ Suppose the Fed convinces everyone it is
committed to reducing inflation.
▪ Then, expected inflation falls,
the short-run PC shifts downward.
▪ Result:
Disinflations can cause less unemployment
than the traditional sacrifice ratio predicts.

24
 The Volcker Disinflation
Fed Chairman Paul Volcker
• appointed in late 1979 under high inflation &
unemployment
• changed Fed policy to disinflation
1981-1984:
• Fiscal policy was expansionary,
so Fed policy needed to be very contractionary
to reduce inflation.
• Success: Inflation fell from 10% to 4%,
but at the cost of high unemployment…

25
The Volcker Disinflation
Disinflation turned out to be very costly:

u-rate near
10% in
1982-83

26
The Greenspan Era: 1987-2006

Inflation and unemployment

were low during most of
Alan Greenspan’s years
as Fed Chairman.

27
 1990s: The End of the Phillips Curve?
▪ During the 1990s, inflation fell to about 1%,
unemployment fell to about 4%.
Many felt PC theory was no longer relevant.
▪ Many economists believed the Phillips curve
was still relevant; it was merely shifting down:
• Expected inflation fell due to the policies of
Volcker and Greenspan.
• Three favorable supply shocks occurred.

28
 Favorable Supply Shocks in the ’90s
▪ Declining commodity prices
(including oil)
▪ Labor-market changes
(reduced the natural rate of unemployment)
▪ Technological advance
(the information technology boom of 1995-2000)

29