. 93
( 126 .)


D1 stimulus to aggregate demand normally pulls up both real
output and prices.
Figure 1, which is familiar from earlier chapters, reviews
S this conclusion. Initially, the economy is at point A, where
aggregate demand curve D0D0 intersects aggregate supply
curve SS. Then something happens to increase spending,
Price Level

and the aggregate demand curve shifts horizontally to D1D1.
The new equilibrium is at point B, where both prices and
A output are higher than they were at A. Thus, the economy
experiences both inflation and increased output. The slope
of the aggregate supply curve measures the amount of infla-
tion that accompanies any specified rise in output and there-
fore calibrates the trade-off between inflation and economic
But we also have learned in this book (especially in
Real GDP Chapter 10) that inflation does not always originate from the
demand side. Anything that retards the growth of aggregate
supply”for example, an increase in the price of foreign oil”
Demand-side inflation is can shift the economy™s aggregate supply curve inward. This sort of inflation is illustrated
a rise in the price level
in Figure 2, where the aggregate supply curve shifts inward from S0S0 to S1S1, and the econ-
caused by rapid growth of
omy™s equilibrium consequently moves from point A to point B. Prices rise as output falls.
aggregate demand.
We have stagflation.
Thus, although inflation can emanate from either the demand side or the supply side
Supply-side inflation
of the economy, a crucial difference arises between the two sources. Demand-side
is a rise in the price level
inflation is normally accompanied by rapid growth of real GDP (as in Figure 1), whereas
caused by slow growth
supply-side inflation is normally accompanied by stagnant or even falling GDP (as in
(or decline) of aggregate
Figure 2). This distinction has major practical importance, as we will see in this chapter.

Copyright 2009 Cengage Learning, Inc. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part.
Licensed to:

Chapter 16 319
The Trade-Off between Inflation and Unemployment

Let us begin by supposing that most economic fluctua- S0
tions are driven by gyrations in aggregate demand. In
that case, we have just seen that GDP growth and in-
flation should rise and fall together. Is this what the data

Price Level
show? B
We will see shortly, but first let us translate the prediction
into a corresponding statement about the relationship be-
tween inflation and unemployment. Faster growth of real out-
put naturally means faster growth in the number of jobs and,
hence, lower unemployment. Conversely, slower growth of
real output means slower growth in the number of jobs and,
hence, higher unemployment. So we conclude that if business D0
fluctuations emanate from the demand side, unemployment
Real GDP
and inflation should move in opposite directions. Unemploy-
ment should fall when inflation rises high and rise when
F I GU R E 2
inflation falls.
Inflation from the
Figure 3 illustrates the idea. The unemployment rate in the United States in 2007
Supply Side
averaged 4.6 percent (which we approximate by 5 percent in the figure), and the
Consumer Price Index was 2.8 percent higher than in 2006 (which we approximate by
2 percent). Point B in Figure 3 records these two numbers. Had aggregate demand
F I GU R E 3
grown faster, inflation would have been higher and unemployment would have been
Origins of the Phillips
lower. To create a concrete example, let us suppose that
unemployment would have been 4 percent and inflation
would have been 3 percent”as shown by point A in Fig-
ure 3. By contrast, had aggregate demand grown more
slowly than it actually did, unemployment would have A
been higher and inflation lower. In Figure 3, we suppose
that unemployment would have been 6 percent and in-
Inflation Rate

flation would have been just 1 percent (point C). This B
figure displays the principal empirical implication of
our theoretical model:
If fluctuations in economic activity are caused primarily C
by variations in the rate at which the aggregate demand
curve shifts outward from year to year, then the data
should show an inverse relationship between unem-
ployment and inflation.
4% 5% 6%
Now we are ready to look at real data. Do we actually
Unemployment Rate
observe such an inverse relationship between inflation
and unemployment? About 50 years ago, the economist
A. W. Phillips plotted data on unemployment and the
rate of change of money wages (not prices) for several extended periods of British his-
tory on a series of scatter diagrams, one of which is reproduced on the next page as Fig- A Phillips curve is a graph
ure 4. He then sketched a curve that seemed to fit the data well. This type of curve, depicting the rate of unem-
which we now call a Phillips curve, shows that wage inflation normally is high when ployment on the horizontal
unemployment is low and is low when unemployment is high. So far, so good. These axis and either the rate of
data illustrate the short-run trade-off between inflation and unemployment, one of our inflation or the rate of
change of money wages on
Ideas for Beyond the Final Exam.
Phillips curves are more commonly constructed for price inflation; Figure 5 on the next the vertical axis. Phillips
curves are normally
page shows a Phillips-type diagram for the post“World War II United States. This curve downward-sloping, indicat-
also appears to fit the data well. As viewed through the eyes of our theory, these facts sug- ing that higher inflation
gest that economic fluctuations in Great Britain between 1861 and 1913 and in the United rates are associated with
States between 1954 and 1969 probably arose primarily from changes in the growth rate of lower unemployment rates.

Copyright 2009 Cengage Learning, Inc. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part.
Licensed to:
Part 3
320 Fiscal and Monetary Policy

F I GU R E 4
The Original Phillips Curve

Change of Money Wages in the United Kingdom, 1861“1957,” Economica, New
SOURCE: A. W. Phillips, “The Relation between Unemployment and the Rate of

Rate of Change of Money Wage Rates

in Percent per Year





Series, 25 (November 1958).

1 2 3 4 5 6 7 8 9 10 11

Unemployment Rate in Percent

F I GU R E 5
A Phillips Curve for the United States, 1954“1969


6 1968
Inflation Rate

3 1954
1967 1959
2 1958

0 1 2 3 4 5 6 7 8 9 10

Unemployment Rate in Percent

aggregate demand. The simple model of demand-side inflation really does seem to
describe what happened.
During the 1960s and early 1970s, many economists thought of the Phillips curve as a
“menu” of choices available to policy makers. In this view, policy makers could opt for
low unemployment and high inflation”as in 1969”or for high unemployment and low
inflation”as in 1961. The Phillips curve was thought to measure the quantitative trade-
off between inflation and unemployment. And for a number of years it seemed to work.
Then something happened. The economy in the 1970s and early 1980s behaved far
worse than the historical Phillips curve had led economists to expect. In particular, given
the unemployment rates in each of those years, inflation was astonishingly high by past
standards. This fact is shown clearly by Figure 6, which simply adds to Figure 5 the data
points for 1970 to 1984. So something went badly wrong with the old view of the Phillips
curve as a menu for policy choices. But what? There are two major answers to this ques-
tion, and a full explanation contains elements of each.

Copyright 2009 Cengage Learning, Inc. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part.
Licensed to:

Chapter 16 321
The Trade-Off between Inflation and Unemployment

We begin with the simpler answer, which is that much
of the inflation in the years from 1972 to 1982 did not 12%
emanate from the demand side at all. Instead, the
1970s and early 1980s were full of adverse “supply
shocks””events such as crop failures in 1972“1973 10
1974 1980
and oil price increases in 1973“1974 and again in
1979“1980. These events pushed the economy™s aggre- 1979 1981
gate supply curve inward to the left, as was shown in 8 1975
1973 1978
Figure 2 (page 319). What kind of “Phillips curve” will


. 93
( 126 .)