6-1

INFLATION AND UNEMPLOYMENT

Figure6-1 shows the unemployment rate since 1960. With a
that the economy was in bad shape at the end of quick glance one can see
2010. Contrast this with the low unem
ployment rate that a healthy U.S. economyenjoyed in 2006 and 2007. In this section we
discuss the Phillips curve, which gives the tradeoff
tion. Later in the chapter we give a more between unemployment and infla
tion between the aggregate supply rigorous derivation, demonstrating the transla
curve and the Phillips curve. (GDP
unemployment; potential GDP connects to the natural rate of connects to
level connects to the inflation rate.) On an
everyday
unemployment: the price
figures for unemployment on the Phillips curve basis it's much easier to work with
gate supply curve. than with GDP numbers on the aggre

THE PHILLIPS CURVE

In 1958A. W. Phillips, then a
professor at the London School of
comprehensive study of wage behavior in the United
The main finding is
Economics, published a
Kingdom the years 1861-1957.
for
summarized
lips curve is an inverse in Figure 6-2, reproduced from his article: The Phil
of inerease in money relationship between the rate of
wages.
rate of wage inflation. In The higher the rate of unemployment and the rate
and unemployment. other words, there is a unemployment, the lower the
tradeoff between wage inflation
'A. W. Phillips, "The Relation between
United Kingdom,
186]-1957" Economica,Unemployment and the Rate of
November 1958. Change of Money Wages inthe
 10

rates Curve fitted to 1861-1913 data

wage
year)
per
(percent
money 6

of
change

2
of
Rate

-2

2 3 4 6 7 9 10 11

Unemployment (percent)
FIGURE 6-2 THE ORIGINAL PHILLIPS CURVE FOR THE UNITED KINGDOM.
(Source: A. W. Pbillips, "Tbe Relation between Unemployment and the Rate of Change of
November 1958.)
Money Wages in the United Kingdom, 1861-1957" Economica,
decreases with the un
The Phillips curve shows that the rate of wage inflation
the
employment rate. Letting W, be the wage this period, and W,., the wage next period,
rate of wage inflation, g., is defined as
W,+) - W, (1)
W,
can write the simple
With u* representing the natural rate of unemployment," we
Phillips curve as
(2)
8 =-e(u - u*)
unemployment. Thisequation states
Wnere e measures the responsiveness of wages to
exceeds the natural rate, that is,
nat wages are falling when the unemployment rate
the natural rate. The difference
Wnen u > u*, and rising when unemployment is below unemployment gap.
Detween unemployment and the natural rate, u - u* is called the unemployment at the
Suppose the economy is in equilibriumwith prices stable and percent. Prices and
of, say, 10
talural rate. Now there is an increase in the money stock
between the nalural rate of unemployment, u*.
) TOu will see below that there is aclose conneciion "nonacceleratimg intlation rate of unemploy
ad potential output, y*, (2) Manyeconomists prefer the term Mankiw, "The NAIRU in
N. Gregory
(NAIRU)to the term tnatural rate." See Laurence M. Ball and
Novenber 2002. See also Chap. 7, footnote 13 in
cOry and Practice," Journal of Economic Perspectives,
this text.
 MACROECONOMICS
122
both have to rise by 10 percentfor the economyto get back to equilibrium. Bu
the Phillips curve shows that, for wages to rise by an extra 10 percent, the un-
wages
employmentrate will have to fall. That will cause the rate of wage increase to go up.

rising, prices too will rise, and eventually the economy will retum to
Wages will start point can be readily
full-employment level of output and unemployment. This inflation, Seen
the definition of the rate of wage in order to
using the
by rewriting equation (), today level:
relative to the past
look at the level of wages
W, = W[| - e(u - u*)) (2a)
above their previous level, unemployment must fall below the natural rate
For wages to rise
increase of wages or wage infl.
Although Phillips's own curve relates the rate of
above, the term Phillips urve" gradually
tion to unemployment, as in equation (2) Phillips curve ora curve relating the rate
came to be used to describe either the original
unemployment rate. Figure 6-3
of increase of prices-the rate of inflation-to the
the 1960s that ap
shows inflation and unemployment data for the United States during
pear entirely consistent with the Phillips curve.

THE POLICY TRADEOFF

The Phillips curve rapidly became a cornerstone of macroeconomic policy analysis.

It suggested that policymakers could choose different combinations of unemployment
and inflation rates. For instance, they could have lowunemployment as long as they put
up with high inflation-say, the situation in the late 1960s in Figure 6-3.Or they could
maintain low inflation by having high unemployment, as in the early 1960s.

6.0
5.5
5.0 Q1969
Inflation
rate
(percent)
4.5 1968
4.0
3.5 1966
3.0
1967
2.5 1965
2.0 1962
1.5
1964 1963 1961
1.0
0.5

3.5 4.0 4,5 5.0 5.5 6.0 6.5 7.0

Unemployment rate, civilian (percent)
FIGURE 63 RELATIONSHIP OF INFLATION AND
(Source: DRI/McGraw-Hill.,) UNEMPLOYMENT: UNITED STATES, 196 I-I
 CHAPTER 6"AGGREGATE SUPPIY AND THE PHILLIPS CURVE 123

Vou already know that the idea of apermanent uncmployment-inflation tradeoff

bccause you knov that the long-run aggrcgatc supply curvc is vertical.
must be wrong
The piece of the puzzle that is missing in the simple Phillips curve is the role of price
expectations. But the data in Figure 6-3 should lcavc you with twoimpressions that are
First. therc is ashort-run tradcoff bctwcen unemployment and infla-
clear and correct.
on 4Sccond, the Phillips curve (and thereforc the aggregatc supply curve) really is
quite flatinthe short run. Applying ocular cconomnctrics to Figure 6-3, you should sce
ibat lowering unemployment by afull perccntage point (which is a lot) incrcases the
:nlationrate in the shortrun by about half a point (a relatively modest amount). Note
too that at very low unemployment rates the inflation/unemployment tradeoff becomes
quite a bit stceper.

6-2
STAGFLATION, EXPECTED INFLATION,AND THE INFLATION
EXPECTATIONS-AUGMENTED PHILLIPS CURVE
The sinple Phillips curve relationship fell apart after the 1960s, both in Britain and in
the United States.Figure 6-4 shows the behavior of inflation and unemployment in the
United States over the period since 1960. The data for the 1970s and 1980s do not fit the
simple Phillips curve story.
Something is missing from the simple Phillips curve. That something is expected.
or anticipated, inflation. When workers and firms bargain over wages, they are con
cerned with the real value of the wage, so both sides are more or less willing to adjust
the level of the nominal wage for any inflation expected over the contract period. Unem
ployment depends not on the level of inflation but, rather, on the excess of inflation over
what was expected.
A little introspection illustrates the issue. Suppose that on the first of the year your
employer announces a 3 percent across-the-board raise for you and your coworkers.
While not massive, 3percent is a nice increase, and youand your colleagues might be
reasonably pleased. Now suppose we tell you that inflation has been running 10 percent
a year and is expected to continue at this rate. You will understand that if the cost of liy
ing rises 10 percent while your nominal wage rises only 3percent, your standard ot
living is actually going to fall, by about 7(= 10 - 3) percent. In other words, you care
about wage increases in excess of expected inflation.
We can rewriteequation (2), the original wage-inflation Phillips curve,to show
that it is the excess of wage inflation over expected inflation that matters:
T)= -e(u - u*) (3)
(8
where is the level of expected price inllation.

"N. Gregory Mankiw, "The Inexorable und Mysterious Iradeotf between lntlation and Unemployment,"
Economic Journal 1 , May 2001.
Inother words, applying eyeball to dala.
 MACROECONOMICS
124

80
14

79
12
74
81
10 75

7%

(69 73 90 77 82

70 76
91
89
060s 88 08>71 85
68 (00
87 84
67 1295 9392 12 11 83
(66 98 04 037 02-94 86 10
65 o61
64 62
09

-2
3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0
Unemployment rate, civilian (percent)
FIGURE 6.4 RELATIONSHIP OF INFLATION AND UNEMPLOYMENT: UNITED STATES,
1961-2012.
(Source: Bureau of LaborStatistics.)

Maintaining the assumption of a constant real wage, actual inflation, t, will equal
wage inflation. Thus, the equation for the modern version of the Phillips curve, the
(inflation-)expectations-augmented Phillips Curve, is
TT = T - e(u - u*) (4)

Note two critical properties of the modern Phillips curve:

Expected inflation is passed one for one into actual inflation.
Unemployment is at the natural rate when actual inflation equals expected intlation.
We have now an additional factordetermining the height of the short-run Phillips
curve (and the corresponding short-run aggregate supply curve). Instead of intersecting
the nalural rate of unemployment at zero, the modern Phillips curve intersects the natu
ral rale at the level of expected inflation. Figure 6-5 shows stylized Phillips curves for
the early 1980s (when inflation had been running 6 to 8 pereent) and the early oughts
(when inflation had been running at about 2 percent).
Firns and workers adjust their expectations of iflation in light of the recent
history of inflation.o Theshort-run Phillips curves in Figure 6-5 reflect the low levelof
inflation that was expected in the early oughts and the much higher level that was
expected in the early 1980s. The curves have two properties you should note. First, Uie

rather than to recen

 CHAPTER 6"AGGREGATE SUPPIY AND THE PHILLIPS CURVE 125

Inftation
rate
(percent)
7 T carlys0s 7%

Early 1980s Phillips curve

2 -Tcarly 00% 2%

Early 2000s Phillips curve

u*
Unemployment rate
FIGURE 65 INFLATION EXPECTATIONS AND THE SHORT-RUN PHILLIPS CURVE.

curves have the same short-run tradeoff between unemployment and inflation; that is to
say. the slopes are equal. Second, in the early oughts full employment was compatible
with roughly 2 percent annual inflation; in the early 1980s full employment was com
patible with roughly 7 percent inflation.
The height of the short-run Phillips curve, the level of expected inflation, , moves
up and down over time in response to the changing expectations of firms and workers.
The role of expected inflation in moving the Phillips curve adds another automatic
adjustment mechanism to the aggregate supply side of the economy. When high aggre
curve,
gate demand moves the economy up and to the left along the short-run Phillips
inflation results. If the inflation persists, people come to expect intlation in the future
(r rises) and the short-run Phillips curve moves up.

STAGFLATION
Stagflation is a term coined to mean highh unemployment ("stagnation") and high
inflation. For example, in 1982 unemployment was over 9 percent and intlation
to see how
aPproximately 6 percent. Point S in Figure 6-5is a stagtlation point. lt is casyincludes sig
Stagilation occurs.? Once the economy is on ashort-run Phillips crve that
below expected
nilicant expected inflation, a recession will push acual intlation down Figure 6-5), but
InTlation (e.g., a movement Lo the right on the 1980s Phillips curve in

don'tunderstand stagfation. This was prob

ror some reason, journalists delight in reporting thaleconomists
expectations was fully appreciated. The
dbly true in the 1960s and early 1970s, before he role of inflation
700S were a long time ago. As you see, stagflation is no longer a puzzle.