Download NBER WO~G PAPER SERIES MACROECONOMIC POLICY ~ THE PRESENCE OF STRUCTURAL

NBER WO~G
PAPER SERIES
MACROECONOMIC POLICY ~ THE
PRESENCE OF STRUCTURAL
MALADJUSTMENT
Robert J. Gordon
Working Paper 5739
NATIONAL BUREAU OF ECONOMIC RESEARCH
1050 Massachusetts Avenue
Cambridge, MA 02138
September 1996
This research has been supported by the National Science Foundation. I am grateful to Martin
Cohan and Tomonori Ishikawa for help with the data and graphs. This paper is part of NBER’s
research program in Economic Fluctuations and Growth. Any opinions expressed are those of the
author and not those of the National Bureau of Economic Research.
0 1996 by Robert J. Gordon. All rights reserved. Short sections of text, not to exceed two
paragraphs, may be quoted without explicit permission provided that full credit, including 0 notice,
is given to the source.
NBER Working Paper 5739
September 1996
MACROECONOMIC POLICY IN THE
PRESENCE OF STRUCTURAL
MALADWSTMENT
ABSTRACT
This paper analyzes two-way interactions between structural reform and macro policy. If
structural reforms increase the flexibility of labor markets, they are likely to improve the short-run
inflation-unemployment tradeoff, providing an incentive for policymakers to expand aggregate
demand. In turn, the promise by policymakers that they will encourage a decline in unemployment
in response to good news on inflation can be used to strike a political deal with political interests
opposed to the introduction or extension of structural reform. Expansionary monetary policy also
provides relief on the fiscal front, directly by bringing the actual budget deficit closer to the
structural budget deficit, and indirectly, by encouraging structural reform, potentially reducing the
structural budget deficit itself.
In 1992-93 several European countries dropped out of the ERM to pursue more expansionary
monetary policies. The difference in the performance of these countties and those countries which
maintained a peg between their currencies and the Deutsche mark provides an important test case
of the consequences of expansionary monetary policy. The depreciating nations by 1995 enjoyed
a substantial relative acceleration of nominal GDP and, surprisingly, an even greater deceleration
of inflation, so that their growth rate of real GDP accelerated more than their growth rate of nominal
GDP in relation to the pegging countries. The continued deceleration of inflation in the depreciating
countries provides evidence that their natural unemployment rate has declined and that expansionary
monetary policy has interacted beneficially with structural reform.
Robert J. Gordon
Department of Economics
Northwestern University
Evanston, IL 60208
and NBER
[email protected]. nwu.edu
1. Introduction
Perhaps
the most important
single economic
linked to the topic of this paper, macroeconomic
Is it possible for Europe
monetary
subsidies of inefficient
are the connections
these connections
currently
policy and structural
to pursue the macroeconomic
union despite the existence of structural
fiscal debt and deficits,
performance
firms, and overregulated
union, monetary
programs,
state
and labor markets?
to maintain
data on major aspects
What
reform?
Do
demand
the growth
required
a fixed exchange
reform ?
It then turns to a theoretical
and supply analysis.
on the
of macroeconomic
between macro policy and structural
section
maladjustment
In this view, the task of macro
rate of nominal aggregate
demand,
while structural
can be viewed as an adverse supply shock, and successful structural
reform as a beneficial supply shock.
between the effects of structural
oil price shocks.
product
tightness required
begins by examining
in terms of aggregate
maladjustment
and welfare
strife with interest groups resisting structural
the interactions
policy is to control
such as excessive
to resolve the conflicts among fiscal stringency
in Europe and the United States.
that interprets
policies needed to achieve
maladjustments,
state pension
maladjustment.
suggest any new avenues for policy in countries like France that are
rate, and the political
The paper
underfunded
today can be
and feedbacks between macro policy and structural
struggling
route to monetary
issue in Europe
We shall analyze the similarities and differences
maladjustment
and adverse supply shocks, such as
Macroeconomic
The supply shock model forces us to distinguish
change of structural
to a negative
maladjustment.
change
in the
Structural
degree
Policy, Page 2
between the level and rate of
reform may be viewed as analogous
of structural
maladju~ment.
On
closer
examination
some types of structural
productivity
growth as to raise it. The analysis of demand and supply shocks forces
us to distinguish
as well between
reform are just as likely to reduce the rate of
two leading models
of aggregate
supply — the
and the hysteresis model — and a hybrid that combines them.
natural rate hypothesis
How does the outcome
of alternative
policy interventions
depend
on knowing
the
right model in advance?
The next section of the paper examines feedback from reform to macro policy:
countries with more flexible markets may have a more favorable short-term
unemployment
hysteresis
achieved
tradeoff.
hypothesis,
This is particularly
which
important
from the perspective
implies that a reduction
in unemployment
at the cost of only a finite increase in the inflation
which depends on the slope of the short-run
from macro policy to reform:
tradeoff
inflation-
curve.
of the
may be
rate, the amount
There is also feedback
a policy that leans in the direction
of expansion
make it possible to create a political deal with interest groups that resist reform.
expansionary
policy also reduces the impact of transition
There is still the need to determine
may
An
costs from reform.
Most of our discussion of macro policy revolves around monetary
exchange rate policy.
of
policy and
where fiscal policy fits in.
Macroeconomic
Is there still a role for fiscal policy in managing
aggregate
the monetary-fiscal
policy
mix states that
monetary
demand;
fiscal policy mainly influences
long-run
capital accumulation
the real interest
demand?
mainly
Policy, Page 3
The theory of
controls
aggregate
rate and hence the rate of
and growth.
Should monetary policy be conducted
high ratios of public debt to GDP?
differently in nations that have relatively
When fiscal deficits exceed the level consistent
with a stable debt-GDP ratio, the need to run a budget surplus becomes an imperative
and requires a stimulative
exchange
monetary
policy to compensate,
rate stability may not be consistent
with the implication
with fiscal convergence.
The empirical section of the paper examines data on the experience
European
experiment
countries
since the 1992 breakdown
of the consequences
that
of ERM, which
of expansionary
aggregate
of selected
provides a controlled
demand
policy.
We
examine the behavior of nominal GDP growth, as well as the “split” of nominal GDP
growth between inflation and real GDP growth, in those countries
major effective exchange
rate depreciations
those that did not depreciate.
We conclude
in 1992-93
as compared
by developing
push toward
monetary
union.
with some of
the implications
empirical results for future macro policy within Europe and exploring
of a continued
that experienced
of the
the desirability
Macroeconomic
2. The Primary Puzzles in Macroeconomic
2.1
Inflation, Unemployment,
We begin by comparing
Behavior
and Labor’s Share
basic macroeconomic
indicators
United States, in order to identify the puzzles to be explained.
well-known
United
divergence
European
roughly
Union
(labelled
2 percent
unemployment
in the time series of unemployment
In 1995 the unemployment
States.
for
same
countries
for Europe and the
Figure 1 displays the
rates in Europe and the
rate for the current
members
1) was 11.0 percent,
in the
early
of the
compared
1960s,
The
to
1995
rate in the United States was 5.7 percent, exactly the same as in 1963.
The upsurge of European
primarily
“Europe” in Figure
the
Policy, Page 4
between
unemployment
relative to U. S. unemployment
1975 and 1985, suggesting
should begin with major structural
occurred
that the search for an explanation
changes that occurred
within Europe during that
decade.
The major theories that describe the interrelation
and inflation
rates are the natural
rate hypothesis
between the unemployment
and the hysteresis
Inflation
rates for the Unit~~d States and the same set of European
displayed
in Figure 2. Here we see that the average inflation
has been remarkably
European
behavior
hypothesis.
countries
are
of “Europe”
similar to that of the United States since the late 1960s, with the
inflation rate exceeding that for the United States by one or two percentage
points per year in most years. Given the similarity of the inflation performances,
the
Macroeconomic
natural
rate hypothesis
unemployment
but increased
is consistent
with the data only if the natural
was roughly stationary
unemployment
This leaves the causes of the increase
and U. S. unemployment
the contrast
flexibility in the U. S.
shared in common
explanation
(1980),
for the divergence
between
(1979),
and
real wage
rigidity
in Europe
Following the early- 1970s slowdown
by all industrialized
nations,
real wage rigidity in Europe prevented
wage ~/P)
definition
By definition,
natural
The seminal
(1985),
and real wage
of productivity
growth
in the U, S.
in tandem with productivity
growth,
such a deceleration.
by examining
data on labor’s
labor’s income share (S) is equal to the real
divided by output per hour (Q/H).
Using lower-case
letters for logs, this
implies that the growth rate of the real wage is equal to the growth
of productivity
over
of European
Bruno-Sachs
real wage flexibility
It is easy to assess the validity of this hypothesis
share in national income.
behavior.
Sachs
allowed the growth of real wages to decelerate
while
in the European
rates was a contrast in labor-market
of Branson-Rotemberg
emphasized
rate in Europe
rate unexplained.
During the early 1980s a popular
work
rate of
in the United States over the last three decades
by roughly as much as the actual unemployment
the same period.
Policy, Page 5
rate
plus the growth rate of labor’s share:
Aw-Ap
=
(Aq-ti)
+ As
(1)
Macroeconomic
Thus any tendency
of the growth rate of real wages (Azu - Ap) to exceed the growth
rate of productivity
(~).
(Aq - Ab) would be reflected
In consequence,
share in Europe
1975-85,
Policy, Page 6
in positive growth in labor’s share
the real-wage rigidity hypothesis leads us to expect that labor’s
would have increased
relative to that in the United States during
the period of the rising relative European
unemployment
rate.
As the cost
of labor increased relative to its marginal product, profits would have been squeezed,
the demand
for labor
would
have decreased,
and unemployment
would
have
increased.
The real-wage
substantial
decades,
rigidity hypothesis
U, S. literature
(albeit small) growth
growth,
According
to equation
to have been consistent
per hour.
a
In the
for the failure of
(l), this common
substantially.
with a 1.0 percent
labor’s share should have declined at 1.0 percent
instance from 70 percent
annual
For
rate of
per year, for
in 1973 to 53 percent in 1993.
Both the real-wage rigidity hypothesis
perception
rate of output
implies that the U. S. wage share must have declined
zero real wage growth
productivity
has as its counterpart
features of U. S. labor markets account
real wages to keep pace with productivity.
perception
Europe
on the failure of real wages to grow over the past two
despite a positive
American view, structural
regarding
for Europe as well as the common U. S.
of stagnant real \/ages appear to be the reverse of the truth, as shown by
the display of wage shares ir~ Figure 3. These wage share series, constructed
by the
Macroeconomic
OECD, include in wage income an imputation
self-employed.
(dominated
percent
Policy, Page 7
for the labor income earned by the
Far from declining rapidly, the wage share series for North America
by the U. S. ) has remained
in 1973 to 66.5 percent
roughly
in 1993.
constant,
falling only from
The wage share series for Europe
68
did
increase relative to North America during 1974-81, which coincides with the period
when the European
the European
unemployment
rate began its relentless descent.
But since 1981
wage share has declined much more, from a 1981 value of 69.3 percent
to 63.4 percent
in 1993.
Thus any effect on unemployment
of the high European
wage share during the
late 1970s should have been more than reversed by the declining wage share during
the
1980s
and
unemployment
convincing
The
rate casts doubt
is parallel
slowdown
productivity
have
slowdown
Competing
absence
now
high European
countries
been
rigid-real-wage
unemployment.
the worldwide
in the
natural
hypothesis
as a
The difficulty in this
post-1973
productivity
to the oil price shocks of the 1970s, since
completely
reversed
in real
terms,
while
the
has not been reversed in most countries.
with the view that the increased natural rate of unemployment
in origin is the hysteresis hypothesis,
is “state dependent,”
of such a reversal
on the original
to that in linking
in the industrialized
oil shocks
structural
1990s.
cause of persistently
explanation
the
early
automatically
which postulates
was
that the natural rate
following in the path of the-actual unemployment
Macroeconomic
rate like the tail of a dog.1
actual unemployment
unemployment
temporary
turn,
The hysteresis hypothesis
in Europe
by a combination
unemployment
hysteresis
rate,
the
natural
rate
by restrictive
followed
Because the actual unemployment
process, the inflation
To reverse the process, some combination
natural
of
rate, especially adverse supply shocks (e.g., higher oil prices and a
rate during the transition
expansionary
can explain the evolution
of events that raised the actual
increase in labor’s share) accompanied
through
Policy, Page 8
demand
in the
demand
path
of the
In
actual
rate was above the natural
rate declined,
as shown in Figure 2.
of beneficial supply shocks supported
policy must bring the actual unemployment
rate, i.e., the unemployment
policy.
gap must become negative,
by
rate below the
in order to “drag
down” the natural rate.
As we shall see in Fig~re 4, none of the major European
having
a negative
expansionary
unemplc yment
The consequence
is close to
of such hypothetical
policies would be to raise the inflation rate until the economy stabilizes
at a new equilibrium
with a lower actual and natural unemployment
rapid rate of inflation.
is to determine
gap.
countries
A key issue in evaluating
the merits of expansionary
the tradeoff between inflation and unemployment
the new equilibrium.
The divergent experiences
of the ERM in 1992-93
highlight
(1995a),
especially
that dropped
of the tradeoff.
his own my
policy
in the transition
of those countries
the recent behavior
1. See the recent book edit(;d by Cro~
rate, and a more
(Cro~
1995 b).
to
out
Macroeconomic
2.2 Unemployment
Gaps, Structural Deficits, and the Stabilizing
Studies are available for many different
the natural
rate of unemployment,
rates of unemployment.
countries
Deficit Ratio
which provide estimates
and of the gap between
1995 unemployment
Policy, Page 9
of
the actual and natural
gaps for the G-7 countries,
plus Spain
and Sweden, are shown in Figure 4. The source is Giorno et. al. (1995), which uses
a method
developed
unemployment
by Elmeskov
is estimated
and MacFarlan
by solving an equation
wages is related simply to the unemployment
rate is equal to the actual unemployment
accelerating
nor decelerating.
(1993).
The natural
in which the rate of change of
gap; thus the natural
other extreme
percent.
the actual
The unemployment
gaps in Figure
act~al real GDP in excess of potential
France and Spain, with estimated
unemployment
The average gap for Europe is about 2 percent,
unemployment
unemployment
rate of 11 percent
(Figure
4 are arrayed
gap and was
real GDP, and at the
gaps in excess of 3
and subtracting
this from
1) implies that the natural
rate for Europe in 1995 was about 9 percent,
estimate of 6 percent
unemployment
rate in any year in which wages are neither
between the United States, which in 1995 had a negative unemployment
assessed to be producing
rate of
much higher than the
for the United States.*
2. In Gordon (1995a) I ha~~e recently estimated the U. S. natural rate of unemployment
as a timevarying parameter and have found it co have decreased gradually from about 6.4 percent in 1981 to about 5.8
percent in 1995.
Macroeconomic
Using a version
of Okun’s
between actual and potential
law to translate
unemployment
at potential
gaps into gaps
Giorno et. al. (1995) also compute
output,
budget deficits, i.e., the budget deficit that would be incurred
operating
Policy, Page 10
output instead of at actual output.
estimates for the same set of countries
structural
if the economy
was
Figure 5 displays the 1995
covered by Figure 4 and shows that most of
the fiscal problems
of the high-deficit
recession-induced.
None of the European countries have actual deficits that are more
than double their structural
deficits.
European
countries
deficits would be correspondingly
This provides a link between the hysteresis hypothesis
There is the possibility that monetary
actual and natural
rates of unemployment,
without
difficult
politically
depreciation,
monetary
the acceptance
rather than
Of course, if the natural rate of unemployment
were lower in these countries, the structural
Europe.
are structural
budget-cutting.
of additional
lower.
and the fiscal dilemma facing
expansion
could pull down both the
and thus reduce the structural
But this would require
inflation,
and
deficits
exchange
rate
the abandonment
of
union.
The consequence
of government
of large fiscal deficits is, of course, an increase in the ratio
debt to GDP.
ranges from 52 percent
Currently
for the major European
for France to 125 percent
countries
this ratio
for Italy, as shown in Figure 6.
A standard relation states that stability of the debt-GDP ratio requires that the deficitGDP ratio be equal to the growth
rate of nominal
GDP times the debt-GDP
ratio.
Macroeconomic
Table 1 compares
the actual and structural
in Figure 6 with a computed
ratio is the current
growth
rate of nominal
ratio from Figure 6, multiplied
rate of inflation,3
Table 1, for the U. S., in which nominal
deficit-GDP
and steady output
ratio is 2.3 percent
GDP growth
growth
by the “warranted
of 5.0 percent
at the potential
is consistent
rate, the required
(.46 times .05 equals .023), a bit above the 1995
on line 5 for the U. S. is -0.5, indicating
than the stability value, implying a slight shrinkage
while the figure displayed
The stability
As shown in the first column of
actual deficit and exactly equal to the 1995 structural
displayed
ratio.
included
GDP, which in turn is set equal to the rate of potential
output growth plus the current
with steady inflation
deficits for the same countries
“stability value” of the deficit-GDP
debt-GDP
Policy, Page 11
deficit.
Hence
the figure
that the actual deficit is smaller
in 1995 of the debt/GDP
c~n line 7 is 0.0, indicating
that the structural
ratio,
deficit is
exactly equal to the stability value.
Table 1 includes simil:~r calculations
for the other European
countries
covered
by Figures 4-6. Line 5 shows that all but Germany and Italy have actual deficits well
in excess of the stability value, implying continued
The surprising
growth
in the debt-GDP
ratio.
inclusion of Italy in the stability group results from a combination
a huge debt/GDP
ratio and relatively
rapid inflation
of
(and hence high warranted
3. For countries with positive unemployment
gaps in Figure 4, the natural rate hypothesis predicts
that inflation is demlerating.
Thus the warranted nominal GDP growth rate allows the rate of real GDP
growrh to accelerate ~ti -u
with the deceleration of inflation until the unemployment
gap is eliminated.
Macroeconomic
nominal GDP growth).
The list of countries
with structural
stability value on line 7 is the same as the list of countries
5, although
adopted
2.3
of course the debt/GDP
expansionary
Policy, Page 12
deficits in excess of the
with positive gaps on line
ratio would grow more slowly if these countries
policies to eliminate their output and unemployment
gaps.
Exchange Rates, Demand Growth, and Inflation
The macroeconomic
has been dominated
by the Maastricht
over the significance
effective exchange
policy discussion within Europe over the past few years
conditions
of the breakdown
1987 to 1992, and then the breakdown
divergence
nominal
influence
GDP growth
rates influenced
between
countries
nominal
countries
Figure 7 plots the
from 1981 to 1995.
from 1981 to 1987, the ERM period from
period starting in 1992.
we will return to the recent period and ask how the post-1992
of exchange
four largest
of the ERM in 1992.
rates of the four largest European
The history has three stages: convergence
Subsequently
for monetary union, and the debate
real GDP growth
is provided
GDP growth,
nominal
in Figures
GDP growth
and inflation.
8 and 9.
Many
and the split of
A preview for the
different
factors
and we can focus on several major episodes.
The
major events that occurred after 1987 were the British boom of 1987-89, the German
reunification
boom
of 1990-91,
and the divergence
growth from the French rate after 1992.
of British
nominal
demand
Italy presents a puzzle, with the appearance
Macroeconomic
of convergence
divergence
of nominal
GDP growth
closer to the French
Policy, Page 13
rate rather
than
after the 1992 Italian exchange rate depreciation.
Further
puzzles are evident
in the behavior
of inflation
French and German inflation rates were relatively close together
for the period of the German reunification
and its aftermath,
rates in Figure 9,
after 1987, except
1991-93.
The puzzle
is that British and Italian inflation rates were much closer to the French and German
inflation
rates after
implications
for demand management
3. Stwctuml
We shall return
1992 than before.
Maladjustment
and Aggregate Demand-Supply
to several simple analytical
distinction
between
hysteresis
hypothesis,
demand
and its
policy, in section V of the paper.
This section links the two main topics of the paper,
maladjustment,
to this puzzle,
macro policy and structural
tools and theories.
and supply shocks, the natural
and the response
Analysis
of demand-management
These include the
rate hypothesis,
the
policy to supply
shocks.
3.1
Demand Shocks and the Natural Rate Hypothesis
We begin with the familiar expectational
which plots the inflation
hypothesis
rate against the unemployment
is valid, the long-run
natural unemployment
Phillips curve diagram in Figure 10,
rate.
If the natural
rate
Phillips curve (LP) is a vertical line rising above the
rate (U*). The initial short-run
Phillips curve (SPJ is drawn
Macroeconomic
on the assumption
actual inflation
Policy, Page 14
that the expected inflation rate is p’. which in turn is equal to the
rate (p~, the point in the vertical dimension
intersects
the vertical LP line. The SP cume shifts whenever
expected
inflation
at which the SP curve
there is a change in the
The SP curve can also shift upward
rate (p’).
in the case of an
adverse supply shock, and down in the case of a beneficial supply shock.
shortly link the concept
of structural
maladjustment
We will
to that of supply shocks.
The influence of aggregate demand is shown by the DG [for Detnarui Growth)
schedule.
of nominal
The position
of the DG schedule is determined
GDP growth
over potential
output
by the excess of the rate
growth,
or “excess nominal
growth” (x), as well as the previous period’s unemployment
to the inflation
rate, then by definition
output growth,
and the unemployment
rate,
then
actual
unemployment
output
growth
actual output
gro~h
rate is fixed.
exceeds
rate.
GDP
When x is equal
is equal to potential
When x exceeds the inflation
potential
output
growth,
and
the
rate declines.4
Consider the effect of a permanent
acceleration
of excess nominal GDP growth
from XOto xl. The DG line shifts upward as shown, and the economy initially moves
to point El. With adaptive expectations
and an adjustment
coefficient
of unity, the
4. When this analysis is done on a diagram with the output gap on the horizontal axis then the DC
=hedule is a negative 45 degree line. If the Okun’s law coefficient linking the unemployment
gap to the
output gap were unity, then the DC schedule in Figure 10 would be a positive 45 degree line. If the Okun’s
law coefficient is 2 (an output gap of 2 corresponds to an unemployment
gap of 1), then the slope of the DC
line is 2, indicating that a shortfall of inflation below x of one percentage
point corresponds
to an
unemployment
gap of minus 0.5.
Macroeconomic
expected
inflation
subsequent
rate is equal to last period’s actual inflation
rate.
Policy, Page 15
Hence in the
period the SP line shifts upward to cross the LP line at a point (marked
C) directly to the right of point El. In the subsequent
to intersect the long-run
period the DG schedule shifts
rate of inflation directly above the initial equilibrium
(at the point marked A).5 If the rate of excess nominal GDP growth
at xl permanently,
point
is maintained
the economy will go through the loop shown by the dashed spiral
line.
This
hypothesis.
permanent
diagram
A permanent
acceleration
which the inflation
Maintaining
summarizes
of excess nominal
of the same amount
by the
as shown
deceleration
natural
GDP growth
rate
causes a
after a transition
rate (UJ below the natural
period in
equilibrium
unemployment
value.
rate (U*)
of excess nominal GDP growth and results in a steady
of the inflation rate.
and a temporary
implied
rate oscillates above and below its long-run
an unemployment
curve is linear,
permanent
acceleration
of inflation
requires a steady acceleration
acceleration
the basic results
The process is symmetric
in Figure
10.
A deceleration
if the short-run
of inflation
Phillips
requires
a
of excess nominal GDP growth, results in oscillating inflation,
period of unemployment
above the natural rate.
5. Imagine a DGZ line drawn through point A. This indicates that the unemployment
rate would be
unchanged if the inflation rate at that point (pJ equal led the value of excess nominal GDP growth (xl) at the
same point.
Policy, Page 16
Macroeconomic
3.2
Supply Shocks and t~
Interpretation
of Structural Maladjustment
Figure 11 uses the same model to examine
the effects of an adverse supply
shock.
The shock is viewed as shifting the SP schedule upwards without changing its
slope.
If the shock is an agricultural
increase, and the initial SPOschedule will shift upward to SP’. But
will temporarily
after the cause of the agricultural
relative price of farm products
beneficial
crop failure, the relative price of farm products
problem is over and normal conditions
will decline.
The economy will temporarily
shocks of 1974-75
the transition
and 1979-81
period
back to normal
raised the relative
prices.
The OPEC oil
price of oil for a substantial
In this case the upward shift of the SP curve to SPYis followed
to the initial
complete,
(SPJ when
position
the transition
ignoring the impact of shifting inflation
to the higher
by a return
relative
excess nominal GDP growth remains unchanged
the DG schedule
economy
would
unemployment.
go
remains
initially
to
fixed.
point
in which
with
higher
supply shock the
inflation
and
higher
Even if in tile subsequent period the supply shock goes away, the SP
schedule will stay above its initial position
previously
is
at XO,and so for the initial transition
Then with an adverse
L,
price
expectations.
A central case useful for the analysis of supply shocks is a situation
period
enjoy a
supply shock due to the relative price decline, and the SP curve will shift
down to SF’ during
period.
return, the
in Figure 10, reflecting
if expectations
the response
are formed as we assumed
of expected
inflation
to the higher
Macroeconomic
actual inflation
caused by the supply shock.
Gradually
Policy, Page 17
higher unemployment
will
push down the inflation rate, and the economy will slide back to the initial position
by the route shown by the dashed line.
But policymakers
supply shock.
will be tempted
If they accelerate
policy of “accommodating”
unemployment
to fight the unemployment
nominal demand growth sufficiently,
but at the cost of a permanent
increase in inflation.
higher rate of excess nominal GDP growth.
EOto a point like N provides a realistic description
made a transition
i.e., pursue a
the supply shock, the economy will avoid an increase in
will arrive at a point like N, with both actual and expected
permanently
caused by the
The economy
inflation
equal to the
The shift from a point like
of the process by which the U. S.
from 5 percent inflation in the early 1970s to 10 percent inflation
in the late 1970s.
The opposite extreme is a policy of “extinguishing”
GDP growth is dropped
M.
Because inflation
Eventually,
sufficiently
the supply shock. Nominal
to push the economy rightward
does not accelerate,
when the initial transition
there is no adjustment
is completed
to a point like
of expectations.
(i.e., the relative
price of oil
settles down to its new higher level), the SP curve will shift back to its initial position.
However,
maintained,
subsequent
excess nominal GDP growth has been diminished,
there
will be a permanent
transition.
reduction
In short the policymakers
and if this reduction
in the rate of inflation
is
in the
face a tradeoff in reacting to supply
Macroeconomic
shocks,
with a neutral
temporary
increase
implying
a permanent
accommodating
experience
L) or extinguishing
in unemployment,
increase
and extinguishing
European
countries
until the major monetary
Numerous
responses.
(point
The contrast
policy
between
(point
did experience
restriction
Franz-Gordon
a
~
the effects of
did not
rate in the late 1970s, whereas the U. S. and
an acceleration
of inflation
that
of the early 1980s began.
of supply shocks and demand
(1992) show that a change in labor’s share,
such as that implied by the real-wage rigidity hypothesis
inflation-unemployment
implying
policies helps to explain why Germany
events can be fit into the framework
In particular,
M) policy
but an accommodating
of inflation.
an increase in its inflation
several other
persisted
(point
Policy, Page 18
discussed in Part II, shifts the
relation in exactly the same way as a change in the relative
price of oil. The SP schedule shifts upward during the period when labor’s share is
increasing.
Whether
of accommodation,
Thus
far
unemployment,
or not it returns to its original position
if any, and the adjustment
the
analysis
does
not
depends on the degree
of expectations
allow
for
simply because the natural rate hypothesis
of inflation.
a permanent
is maintained
not allowed for an increase in the natural rate of unemployment.
in two ways. First, the nature of the structural
functioning
there
of the labor market
could be a hysteresis
sufficiently
mechanism
maladjustment
in
and we have
This could occur
could be to impair the
to increase the natural
in place which
increase
automatically
rate.
Second,
translates
a
Macroeconomic
period
of high actual
unemployment
into
an increase
in the
Policy, Page 19
natural
rate
of
unemployment.
Krugman
(1994) asks whether
with a relatively
high reservation
rise in the natural
unemployment
generous European
wage (or legal minimum
wage) could explain the
Such policies could explain high
rate in Europe.
but not rising unemployment,
unemployment
welfare policies, combined
as was observed
between
1970s and late 1980s, unless the benefits had increased significantly.
welfare states were already generous
or if there
decreased
demand
unemployment
is a minimum
for low-productivity
in many European
unemployment
countries,
workers
low-wage
This hypothesis
wages
then the
into
is consistent
of young people has increased
higher
with
significantly
while in the United States the same demand twist has
of the demand twiw, Krugman
countries,
workers
If reservation
will be translated
emerged in the form of an il lcrease in the inequality
explanations
workers.
wage that does not change,
rate than reduced relative wages.
the facts that long-term
with a twist in the
a decrease in the demand for low-productivity
and an increase in the demand for high productivity
do not change,
But European
back in the early 1970s when unemployment
was relatively low. Instead, he argues that the facts are consistent
demand for labor, combining
the early
of wage rates,
rejects international
which he argues would change the sectoral
In searching for
competition
from
mix of high skill
relative to low skill employment.
been a generalized
Macroeconomic
Policy, Page 20
Instead, he favors the explanation
that there has
skill-biased shift in technology
that has affected all industries.
Krugman places his emphasis on skill-biased technology
evolution
that
of European
no one factor
European
and American labor markets.
appears
unemployment
than a single cause.
of the American and European
contrast
emerges
declining)
the increase
uses the differing performance
to assess the plausibility
unemployment
natural
in
causes, rather
rate in Europe
rate in the United States.
labor market diagram in which, following
equilibrium
occurs at the intersection
of
in
He
Layard
of a labor demand
sloped “wage setting curve” which displays the wage that
from the bargaining
process
factors could have caused the European
U. S. wage-setting
fully to explain
as a criterion
causes of a rising natural
the
Bean (1994) argues
that there must be multiple
economies
centers his analysis on a standard
curve with a positively
In contrast,
(1995), like Krugman,
to a stable (and recently
et. al. (199 1), short-term
enough
and concludes
Gordon
a long list of potential
significant
shifts by comparing
at alternative
wage-setting
levels of labor input.
curve to shift upwards
What
or the
curve to shift downwards?
1. An increase in the tax wedge.
Since firms pay pre-tax wages but workers
receive after-tax wages, any increase in payroll or income taxes can shift up the wagesetting schedule.
The tax wedge in the Europe is both higher and increased
more
than in the United States between the late 1960s and late 1980s (Bean, 1994, p. 586).
Macroeconomic
2. The rigid real wage hypothesis
the European
labor share between
seems consistent
Policy, Page 21
with the observed bulge in
1974 and 1982 (Figure 3 above), which coincides
with the period of most rapid increase in the natural rate of unemployment.
3. A leading candidate
for causing divergent behavior across the Atlantic is the
marked decline in U. S. trade union membership,
percent
from 26.2 percent in 1977 to 15.8
in 1993 (union members as a fraction of wage and salary workers).
4. The real minimum wage has fallen sharply in the United States while rising
in some European
countries,
particularly
France.
5. Both legal and illegal immigration
added substantially
pressure
6.
market
schedule.
r~:gulation in Europe,
particularly
hours, reduce the demand fot unskilled labor as contrasted
emerge with an unregulated
product
to Krugman’s
and provide a convincing
set of macroeconomic
by the phrase “structural
maladjustment.”
the evolution
maladjustment
of productivi~]
German
to the demand that would
phenomena
skill-biased
demand twist
that can be summarized
We will return below to the interactions
and reform with demand management
growth.
shop-closing
market.
These six factors are complementary
of structural
into the U. S. has
to the supply of unskilled labor and plausibly added to downward
on the U. S. wage-setting
Product
of unskilled workers
policy and with
Macroeconomic
3.3
The Hysteresis
Hypthesis
The second approach
unemployment
to explaining
is the hysteresis
hysteresis hypothesis
management
revives
the original
demand-induced
rate.
Inflation
converges
It can be shown
tradeoff
alter the unemployment
between
recession boosts unemployment,
rate.
gap as
inflation
Instead,
and unemployment.
it
A
which in turn boosts the natural
rate.
natural
rate
In reverse, the hypothesis
at the cost of a finite, not ever-accelerating,
But, unlike the original stable Phillips tradeoff,
increase
the tradeoff
at any given time depends on all of past history.
of high unemployment
push the unemployment
the
policy always is faced with the choice of achieving
schedule available to policymakers
The experience
that
that demand-
rate.
stabilizes at a new lower level when the new higher
of unemployment
rate of
by the change of the unemployment
to the new higher actual unemployment
in the inflation
formally
of the natural rate hypothesis
permanently
implies that demand-management
a reduction
equation
the basic implication
Phillips
rise in the natural
to replacing the level of the unemployment
is equivalent
policy cannot
a permanent
hypothesis.
the driving variable in the inflation
rate.G It contradicts
Policy, Page 22
implies that European
policymakers
cannot
rate as low as they could were they taking the same policy
actions as fifteen or twenty-five
years ago.
6. Our diseuxion here refers to a “linear” version of the hysteresis hypothesis in which the equilibrium
unemployment
rate “like an elephant” remembers all pastshocks. Cross (1995b, p. 190) distinguishes this from
the nonlinear version in which the memory of past shocks is selective.
Macroeconomic
If hysteresis
insider-outsider
is present
model of wage determination
to convert a favorable
rather
in fact, this calls for a theoretical
demand
or supply shock into wage increases
model goes in the same direction.
In addition to total unemployment
in addition
of real wages or of labor’s income share from target levels.
stemming
from
hysteresis-like
deviations
The
for themselves
The target real-wage
nominal wage increases are influenced
labor’s share responds
explanation.’
shows how employed insiders are able
than into new jobs for the unemployed.
curve approach,
Policy, Page 23
bargaining
in the Phillips
by the deviation
If the target level of
to its actual level, then any pressure on wages
of the actual
share
from
the target
share
gradually
disappears.
3.4 Implications of Structural Maladjustment
How can the structural
in the diagrammatic
possibility
and hysteresis hypotheses be interpreted
previously introduced?
In Figure 12 we consider
a
in the 1970s and early 1980s, the conjunction
adverse supply shock caused by higher oil prices and/or an increase
income
unemployment
workers,
framework
that might have occurred
of a temporary
in labor’s
maladjustment
and Hysteresis
share,
with
a permanent
increase
in . the
natural
rate
of
caused by Krugman’s labor demand twist toward more highly skilled
together
with
some
7. A wide variety of theoretical
combination
and empirical
of my previous
list of supply-side
papers on hysteresis is found in Cross (1988).
Macroeconomic
impediments,
tax wedge.
Policy, Page 24
including an increase in the real minimum wage and an increase in the
In Figure 12 the temporary
SP curve, and the permanent
shock is indicated by the upward shift in the
increase in the natural rate by the rightward
shift in the
LP line.
What choices are open to policymakers?
relative to potential
output
growth
(x) remains
moves to point L, just as in Figure 11.
removed
the economy
initially
Once the source of the supply shock is
returned
12 there no longer is a positive unemployment
than excess nominal
natural
unchanged,
(e.g., the relative price of oil or labor’s share stabilizes at a new level), in
Figure 11 inflation and unemployment
potential
If the growth rate of nominal GDP
output,
rate.
GDP growth,
gap at point L. Inflation
and so output
and the unemployment
m~ladjustment
must grow more slowly than
the disinflationary
dashed line in Figure 12, initially experiencing
then a partial recovery to the equilibrium
a further
loop shown
One apprcach
the validity of the condition
effect of the unemployment
by the
rise in unemployment
and
level UI *.
approach
and the hysteresis approach
explain the observed increase in Europe’s natural rate of unemployment.
be distinguished?
is higher
rate must rise relative to the new higher
The econom]~ goes through
Both the structural
to their original levels. But in Figure
is to estimate wage and price equations
can
Can they
to determine
for “pure hysteresis,” namely the absence of a “level”
gap.
In one such attempt,
Franz and Gordon
(1992)
Macroeconomic
Policy, Page 25
found that hysteresis was partial rather than full in both Ge~many and the United
States, since both level and rate of change
countries.
Because the German and U. S. coefficients
of unemployment
potential
effects are highly significant
in the two countries
are so similar, yet the evolution
is so different,
there
through this route for explaining that evolution.
appears
properties
of a permanent
depends
The structural
gap), it provides
since the equilibrium
natural rate model
only on the level of the unemployment
maladjustment
approach
share had been reversed,
aspects of the European
gap.
as depicted in Figure 12 has the appeal
Oil shocks, an increase in labor’s share, and a skill-biased
all began to occur in the 1970s.
translated
in unemployment,
of the economy are the same as with a straightforward
in which inflation
of realism.
increase
to be little
If hysteresis is partial (i.e.,
inflation depends on both the level and change of the unemployment
no explanation
in both
demand twist,
By the time the oil shocks and the increase in labor’s
the natural rate of unemployment
welfare system that prevented
had been increased
by
the demand twist from being
into greater inequality of wages, as it was in the United States and to some
extent in the United Kingdom.
Macroeconomic
4. Intemctions
Responses
4.1
between Macro Policy and Structura! Maladjus~ent
of Demand Management
We can now discuss interactions
management
policy.
macroeconomic
management
to Successful Structural Reform
between structural reform and macro demand-
If the sources of structural
policy reform
policy respond?s
maladjustment
line Ieftward.
are identified
begins to reverse their effects, how should
Our previous framework
in Figure 13, we assume that successful structural
~
Policy, Page 26
We have relabeled
the points
demand
is easiest to interpret
reform instantaneously
on the diagram,
and
if, as
shifts the
so that
the
economy’s initial high level of the natural rate is at UO* and moves leftward suddenly
to
U1*.Because
the unemployment
gap is now zero at point .E1 there is a new SPI
schedule that has shifted to the left at the initial inflation rate and expected inflation
rate.
However,
the econom:r does not move instantly from point EOto El.
it moves to point A, which is at the intersection
of the new SP1 line with the initial
DGO line, which holds fixed the initial rate of excess nominal GDP growth
point A inflation
has declined
output growth.
in which inflation remains below XOallow a temporary
which in turn allows the actL,al unemployment
8. A detailed
evaluation
(x~. At
below the initial value of x, and this allows actual
output growth to rise above the rate of potential
point El. An alternative
Instead,
acceleration
Successive periods
of output growth,
rate to decline to the new equilibrium
poiicy could achieve the same output path by temporarily
anti chronology
of structural
reform efforts is provided
in OECD
(1994 b).
Macroeconomic
Policy, Page 27
raising excess nominal GDP growth by enough to maintain a constant
inflation
rate,
thus allowing the economy to move straight left from EO to El rather than following
the curved dashed path drawn through
How can such a policy procedure
Reserve Board appears
point A.
be carried out in practice?
to follow a policy procedure
targeting the unemployment
that could be interpreted
rate consistent with steady nonaccelerating
adjusting interest rates to keep the actual unemployment
the natural
immediate
unemployment
instrument,
unemployment
rate (U*).
forecast of the actual unemployment
value of U*. Symmetrically,
estimated
as
inflation, i.e.,
rate as close as possible to
Since the Fed believes that the effects of its
the Federal Funds rate, take roughly one year to influence the
rate, it leans toward
its best forecast
The U. S. Federal
raising the Federal
Funds rate when its best
rate one year from now is below its estimated
it leans toward
of the actual unemployment
reducing
the Federal Funds rate when
rate one year from now is above its
value of U*.
Although
U* may not move around as much as the actual unemployment
but it is not immutably
carved in stone as a single precise number.
turns out to be lower than the value forecast
private forecasters,
rate,
When inflation
by the Fed’s staff or a consensus
of
as in much of 1995, the Fed concludes that U* has declined from
the value previously
assumed by forecasters.
This makes it more likely that the Fed
will reduce interest
rates and less likely that it will raise them.
Accordingly,
a low
realized value of inflation
to lower long-term
Macroeconomic
Policy, Page 28
or a high realized value of unemployment
can both lead
interest rates as speculators
be to reduce short-term
guess that the Fed’s next move will
interest rates.
The Fed appears to act as if it is operating
and it does not appear to adjust its operation
movements
of the exchange
attempt by an individual
could
of this policy procedure
rate of the dollar.
It would therefore
rate
toward Monetary
fluctuations
Union.
sufficiently
A policymaker
large
rate in response
to a lower natural
(achieved
nlicroeconomic
reform
rates
by a successful
is turning
to stimulate
unemployment
inflation
rate.
maintaining
appear that an
to compromise
policy)
unemployment
would
notice
the
in order
economy
to push
down
rate
first that
and would react by reducing
dolnestic
the
to reduce
out to be below forecast,
the
interest
actual
The c(:ntral aspect of this policy is taking the good news on
to be a signal that stimulative
unemployment
in response to
in Europe attempting
the actual unemployment
inflation
environment,
central bank within Europe to emulate the Fed’s procedure
lead to exchange
movement
in a closed economy
rate, rather
policies should be adopted
than just “accommodating”
the existing unemployment
to reduce the
the lower inflation
rate by
rate.
4.2 Structural Reform and l>roductlvity Growth
The analysis of Figurl: 13 did not take into account
reform on the growth rate of potential
any effect of structural
output, which would reduce x (the excess of
Macroeconomic
nominal GDP growth over potential
were to remain unchanged.
potential
Policy, Page 29
output growth) if actual nominal
Uncertainty
regarding
the effect of structural
output growth stems from the conflict between two forces.
gains should raise productivity
grotih.
less expensive
to hire, such as the reduction
for employers
GDP growth
reform on
Pure efficiency
But reforms in labor markets that make labor
or elimination
of the
minimum wage, may raise the demand for labor at a given level of output and reduce
productivity.
reforms,
The same effect would
especially
a loosening
would trade consumer
be expected
of German
convenience
of particular
regulations
product
on shop-closing
market
hours that
for an increase in the labor requirements
needed
to achieve a given total of real retail sales.
The interactions
between changes in the degree of structural
both in an adverse and beneficial direction,
evolution
of European
wage-setting
unemployment
shocks (e.g., an increase
initially raises unemployment
are explored
is portrayed
maladjustment,
in Gordon
(1995 b). The
as resulting from a set of adverse
in the real effective minimum
northwest
up a labor
demand curve, also boosts the marginal and average products ‘of labor.
This is then
followed
by a profit squeeze and a period
productivity
of disinvestment,
gain but further increases the unemploymel]t
by achieving
minimum
and, by shifting the economy
wage) which
a beneficial
wage-setting
the
rate. Reversing this process
shock (e.g., a reduction
wage) would initially reduce unemployment
which eliminates
in the real effective
and both the marginal
and
Macroeconomic
average products
of labor,
increase of investment,
This would then be followed
by a profits boom and an
which would offset the initial productivity
reduce the unemployment
Policy, Page 30
loss and further
rate.
Feedback from Macro Policy to Structural Reform and Vice Versa
The analysis of Figure 13 emphasizes
the opportunity
in the natural rate of unemployment
for macro policy to
respond
to a reduction
reform.
A policy response that stabilizes inflation and responds to a reduction
natural rate as an opportunity
to accelerate
achieved by structural
nominal demand growth
in the
differs from a
passive policy in which excess nominal GDP growth (x) is stabilized.
This policy choice may influence
reform.
When
there
reducing the minimum
is strong
the likelihood
and magnitude
opposition
to such reform
political
wage and reducing government
central bankers and finance ministers to respond
authorities
measures
as
subsidies, the willingness
of
aggressively to successful reforms
may make those reforms more likely. There is potential
the demand-management
of structural
for a political deal in which
promise to lean toward lower interest rates and
faster nominal demand growth, in trade for the willingness of political interest groups
to support those reform measures that appear to have the greatest chance of reducing
the natural
unemployment
reduce the transition
rate.
Demand
expansion,
by creating
costs of reform policies which create temporary
through the required restructuring
of a particular
jobs, may also
unemployment
industry that, for instance, has been
Macroeconomic
overstaffed
as a result of previous
Policy, Page 31
and subsidies
state ownership
(the “olive-belt”
airlines come to mind).
Another
expansion
related feedback is from monetary
which reduces the unemployment
deficit closer to the structural
which demand expansion
reducing
the
themselves.
natural
policy to fiscal policy.
gap brings the actual government
deficits displayed
in Figure 5.
goes hand in hand with structural
rate
of unemployment
and
hence
Reducing the deficit then raises the potential
There
is feedback
budget
A successful deal in
reform may succeed in
the
structural
deficits
for using the fiscal dividend
to reduce tax rates and hence the tax wedge, at least in those countries
on an explosive track for the debt-GDP
Demand
which are not
ratio.
in the reverse
direction
as well.
Countries
with more
flexible labor markets may have a more favorable inflation-unemployment
tradeoff,
i.e., a flatter SP schedule in Figures 10-13.
This is particularly
from the
perspective
which
of the
unemployment
hysteresis
hypothesis,
implies
Current
States, with its flexible labor market,
estimates
unemployment
that
a reduction
in
may be achieved at the cost of only a finite increase in the inflation
rate, the amount of which depends on the slope of the short-run
United
important
show
that
a one-year
rate by one full percentage
tradeoff curve. The
has a very flat short-run
sustained
reduction
point below the natural
of
tradeoff.
the
actual
unemployment
Macroeconomic
rate raises the inflation
(Gordon,
rate by only 0.35 of a percentage
point
Policy, Page 32
after one year
1995a, Table 1).
5. Policy Lessons fmm tie 1992-93 ERM Breakdown
An important
controlled
experiment
was carried
countries,
United Kingdom,
broke away from the ERM and achieved substantial
mark,
including
European
country
against the countries
France,
analysts
the
Italy, Portugal,
in 1992-93
several important
of their currencies
particularly
out in Europe
which remained
Low Countries,
have previously
believed
is futile, since any transitory
output
depreciation
depreciation
to evaporate.
be eroded
by an
gains from the exchange
In its most extreme form, t:his view holds that countries
p:.st a transition
period
of a year or two, over their real
rates.
A pessimistic
provided
Some
by any individual
gains soon would
of inflation that will soon cause the competitive
exchange
depreciations
and Switzerland.
acceleration
do not have any control,
and the
aligned with the Deutsche
Austria,
that
Spain, Sweden,
when
interpr~:tation
by some authors.
and Spain.
He points
of the outcome
I )eGrauwe
divergence
is
(1995, p. 9) focusses on the two cases of Italy
to tne temporary
initially the sharp depreciations
of the 1992-93
nature
of the competitive
gains.
While
“did not affect inflation very much in these countries
(mainly because of the rece:;sion), since 1994 inflation
differentials
with Germany
Macroeconomic
have started to increase significantly.”
to guide their inflation
He concludes that in order for Italy and Spain
rates to the differential
a new policy of “painful disinflation”
prospect
Policy, Page 33
would
prescribed
by the Maastricht
have to begin.
Treaty,
Since he views the
of success of such policies as low, he thinks that ‘the door to monetary
union will be shut for a long time for these countries.”
DeGrauwe’s
treatment
on two depreciating
in those countries
countries.
than Gemny
inflation
boom.
that depreciated
with Germany
that country with all countries
(not including
Switzerland.
of inflation
high in 1992-93
failing to
as a result of
would look at
them with a set of countries
in 1994-95
a misleading
makes the inflation
indicator
other
differential
of the inflation
of any
differential
of
other than Germany.
Germany)
in Table 2. The five appreciating
are Austria, Belgium, France, Netherlands,
On average these countries experienced
of 10.2 percentage
the acceleration
The omission of Germany is crucial, because
The results of the analysis are summarized
countries
He focusses only
A more balanced appraisal
and compare
that did not depreciate.
reasons.
rates only with Germany,
rate was temporarily
the time path of its disinflation
other country
their inflation
of the reunification
a set of countries
for numerous
He greatly exaggerates
by comparing
note that the German
the aftermath
is misleading,
a nominal-effective
appreciation
points, from exchange rate indexes of 99.5 in 1992:Q2
in 1995 :Q2. In contrast the depreciating
countries — Italy, Portugal,
and
to 109.7
Spain, Sweden,
Macroeconomic
and the United Kingdom — experienced
an average depreciation
an average index value of 99.1 in 1992:Q2
Policy, Page 34
of 22.2 points, from
to 76.9 in 1995 :Q2. All of the average
values in Table 2 are weighted across the five countries in each group using SummersHeston
1985 PPP weights.
Thus among the appreciating
Among the depreciating
France is 59 percent.
countries,
countries,
the weight of
the respective
weights of
Italy, Spain, and the U. K. are 36,2, 17.0, and 37,2 percent.
Both groups of countries
deceleration
of inflation,
But there the similarity
depreciating
points.
countries
enjoyed an acceleration
of nominal GDP growth,
and thus an even greater acceleration
of real GDP growth.
stops,
GDP growth
The acceleration
of nominal
exceeded that in the appreciating
countries
a
in the
by 1.3 percentage
Yet none of this was absorbed by inflation; inflation actually decelerated more
in the depreciating countries than the appreciating countries.. And as a result the
acceleration
of real GDP growth in the depreciating
depreciating
countries
It is implausible
by 1.7 percentage
countries
exceeded
that in the
points.
that this favorable
outcome
for the depreciating
countries
could continue forever, but the IMF forecasts for 1996 (from the same source as used
for Table 2) show no acceleration
of five depreciating
discovered
countries.
a “macroeconomic
route to macroeconomic
of inflation to be predicted
for the same average
Thus it appears that the depreciating
free lunch,” an improvement
expansion
without
inflation.
countries
in competitiveness
The implication
have
and a
is that there
Macroeconomic
is substantial
room for individual
nations in Europe to reduce their unemployment
gaps and their actual budget deficits through
inevitable
depreciation
depreciation
Policy, Page 35
in the nominal
expansionary
exchange
monetary
rate will translate
in the real exchange rate and an improvement
policy.
The
into a durable
in competitiveness.
With inflation in Germany at only 2 percent, some of which doubtless reflects
the well-known
upward
lead the remaining
nations
nominal GDP growth,
“split” of additional
bias in price indexes, there is ample room for Germany
tied to its exchange
The experience
nominal
of the depreciating
GDP growth
highly favorable under present conditions.
of inflation
rate into a further
between
growth
and inflation
is
way, the benign behavior
in Europe suggests that the natural rate of unemployment
decline, just as the Federal Reserve believes has occurred
of
nations suggests that the
output
Stated another
expansion
to
has begun to
within the United States,
Macroeconomic
Policy, Page 36
6. Conclusion
The management
of demand
policy interacts
with structural
maladjustment,
both when maladjustment
is getting worse, as in the 1970s and early 1980s, and
when successful structural
reforms reduce the influence
of structural
and allow a decrease in the natural
(i.e., constant-inflation)
In our interpretation
of structural
a combination
maladjustment
rate of unemployment.
factors, rather than a single “silver
bullet”, explains the substantial increase in Europe’s natural unemployment
decade 1975-85
and the divergence
declining natural
during the
of that natural rate from the stationary
rate in the United States.
In the 1970s two-supply
shocks, in the
form of higher real oil prices and an increase in labor’s share in national
pushed
European
unemployment.
economies
Reactions
in some countries
in the
direction
of higher
inflation
of central banks varied, with monetary
— but not others — leading to a substantial
or even
income,
and
higher
accommodation
divergence
of inflation
rates in Europe by the early 1980s.
These
two supply
both
of which
were
in the mid- 1980s,
combined
with
influence
reversed
structural
maladjustments
shocks,
to boost the European
natural
low single digits in the early 1970s to nearly 10 percent
prices and labor’s income share declined subsequently,
reflects
the role of other
longer-lasting
temporary
more
and had their
deeply
unemployment
by 1985.
entrenched
rate from
While real oil
the natural rate did not. This
maladjustments,
especially
inflexible
real
Macroeconomic
wages for the unskilled who were harmed by skill-biased
inflexibility
in turn reflects three dimensions
technical
Policy, Page 37
change, and this
along which Europe (on average if not
for every country) differs from the United States – high and rising tax wedges, more
powerful
unions, and a higher and (in some countries)
rising real minimum
This account does not place much emphasis on hysteresis.
Empirical
wage.
research
does not support the “pure hysteresis” view that requires the absence of level effects
of the unemployment
gap in dynamic wage and price adjustment
most plausible channel by which hysteresis operated
other sources of maladjustment
capacity utilization
natural
was through
to low capital accumulation,
rate,
Stated
differently,
Europe
The
feedback from the
with the result that the
rate in Europe has been roughly stationary
unemployment
equations.
despite the rise in the
no longer
has sufficient
capital to employ fully its existing labor force.
The paper stresses two-way interactions
policy.
With flexible labor ]narkets, the short-run
is likely to be favorable, providing
demand.
between structural
reform and macro
inflation-unemployment
an incentive for policymakers
The benign effect of the British depreciation
to expand aggregate
in 1992 on the inflation
provides one example of the potential payoff of more flexible labor markets.
the promise by policymakers
that they will encourage
tradeoff
rate
In turn,
a decline in unemployment
in
response to good news on inflation can be used to strike a political deal with political
interests opposed to the introduction
or extension of structural reform.
Expansionary
Macroeconomic
monetary
policy
unemployment
also provides
relief on the fiscal front,
gap and bringing
In 1992-93
countries
the
monetary
and those countries
Deutsche
expansionary
mark
policies.
monetary
dropped
The difference
which maintained
provides
policy,
an important
did not depreciate.
But, surprisingly,
reducing
the
test
the
budget deficit itself.
out of the ERM to pursue
.
in the performance of these
a peg between their currencies
Not surprisingly,
acceleration
reform, potentially
the structural
countries
enjoyed a substantial
inflation,
structural
rate and therefore
several European
more expansionary
by reducing
the actual budget deficit closer to the structural
budget deficit, but also, by encouraging
natural unemployment
both
Policy, Page 38
case of the
consequences
the depreciating
and
of
nations by 1995
of nominal GDP growth relative to the nations that
they enjoyed an even greater deceleration
so that their growth rate of real GDP accelerated
of
more than their growth
rate of nominal GDP.
This augurs well for a favorable outcome
may be that the natural
unemployment
of expansionary
rate has begun to decline
Europe, and that this explains why inflation has continued
which have depreciated
countries
their exchange
creates an environment
The set of issues addressed
litany
of contractionary
demand
demand policy.
to decelerate
rates significantly.
favorable to continued
significantly
It
in
in countries
Job creation
success in structural
in those
reform.
in this paper seems a long way from the depressing
decisions
currently
being
made
in numerous
Macroeconomic
European
countries
in the name of the Maastricht
explosive path for the debt-GDP
An important
if accompanied
a mix of tighter
abandoning
textbook
discussion
while maintaining
does not adopt
by a shift toward easier monetary
fiscal policy and easier monetary
two ways. First, the Bundesbank
Bundesbank
are on an
of the
mix, that a shift toward tighter fiscal policy need not reduce output
or raise unemployment
to expand
Some countries
ratio and need to bring their fiscal house in order.
lesson can be drawn from the traditional
fiscal-monetary
shift toward
criteria.
Policy, Page 39
A
policy can occur in
could adopt easier policies, allowing other countries
a fixed exchange
easier policies,
rate with Germany.
other
countries
Second, if the
have the option
the path to monetary union and adopting expansionary
of infhition.
of
monetary policies
that would bring down botl~ the actual and natural rates of unemployment
minimal acceleration
policy.
with a
Macroeconomic
Policy, Page 40
REFERENCES
BEAN, CHARLES R. (1994).
fionomic
“European
Unemployment:
A
Survey,”
Journal
of
Literature, 32 (June), 573-619.
BRANSON,WILLIAM H. ANDROTEMBERG,JULIOJ. (1980).
“International
European Economic kviezu, 13 (May), pp. 309-32.
with Wage rigidity,”
Economics
BRUNO, MICHAEL, AND SACHS, JEFFREY D. (1985).
Stagflation (Cambridge
CROSS, ROD (ED.) (1988).
Hypothesis.
(1995a).
Oxford
MA:
Harvard
Unemployment,
and New York:
(Cambridge
(1995 b). “Is the Natural
of Worldwide
University Press).
Hysteresis, and the Natural tite
Basil Blackwell.
The Natural Rate of Unemployment:
of the Hypothesis.
Adjustment
UK: Cambridge
Rate Hypothesis
Reflections on 25 Years
University Press).
Consistent
with Hysteresis?”
in Cross, ed. (1995a), pp. 181-200.
DEG~UWE, PAUL(1995).
“The Economics of Convergence
towards Monetary
Union
in Europe,n CEPR discussion paper 1213, July.
ELMESKOV, JORGEN, AND MACFARLAN, MAITLAND (1993).
Persistence,”
OECD Economic Studies (no. 21, Winter),
“Unemployment
pp. 57-88.
Macroeconomic
FRANA WOLFGANG, ANDGORDON, ROBERTJ. (1993).
Germany
and
America:
Differences
and
Policy, Page 41
“Wage and Price Dynamics in
Common
Themes,”
European
Economic Review, 37 (May), pp. 719-54.
GIORNO, CLAUDE,RICHARDSON,PETE, ROSEVEARE,DEBOM,
pAUL (1995).
“Potential
Output,
Output
Gaps,
ANDVANDENNOORD,
and
Structural
Budget
Balances,” OECD Economic Studies (no. 24), pp. 167-209.
GORDON, ROBERTJ, (1995a).
paper presented
November
“Estimating the NAIRU as a Time-varying
to Panel of Economic Advisers, Congressional
Budget Office,
16.
(1995 b). “Is There a Tradeoff between Unemployment
Growth?”
Parameter,”
NBER woiking
paper
5081,
April.
and Productivity
Forthcoming
in Dennis
J.
Snower and Guillermc) de 1a Dehesa, eds., Uwmployment Policy: How Should
Governments
Respond to Unemployment?
Cambridge
UK:
Cambridge
University Press for CEPR.
KRUGMAN, PAUL (1994).
“Past and Prospective
Federal Reserve Bank oftinsas
Causes of High Unemployment,”
City Economic &view (fourth quarter), pp. 21-
43.
LAYARD, RICHARD,
NICmLL,
STEPHEN, AND JACKMAN, RICHARD
Unemployment: Macroeconomic Pe+ormance and the bbor
Oxford University
Press).
(1991).
Wrket (Oxford:
Macroeconomic
OECD
(1994a).
95.
Policy, Page 42
The OECD Jobs Study: Unemployment in the OECD Area, 19sO-
(Paris).
(1994b).
Assessing Structural Reform:
SACHS, JEFFREYD. (1979).
Comparative
269-319.
“Wages, Profits,
kssons for the Future.
and Macroeconomic
(Paris).
Adjustment:
A
Study,” Brookings Papers on Economic Activity, 10 (no. 2), pp.
Table 1
Gaps between Actual and Structural Deficit/GDP Ratios
and Deficit/GDP Ratio Required to Stabilize Debt/GDP Ratio,
1995, Selected Countries
11.
s.
-. -.
-Fmnce
------
11.K.
-. .-
Camlsnv
. . . .... ..
-Snnin
---
Sweden
-------
Ttslv
..-. J
Debt-GDPRatio
46,0
51.5
52.5
58.8
64.8
81,4
2. Warranted Nominal
GDP Growth
5.0
4.1
4.2
5.0
7.9
4.9
7,1
3. Stability Value of
Deficit/GDP
Ratio
2.3
2.1
2.2
2.9
5.1
4.0
8.9
4. Actual
1.8
5.0
4.7
2.4
6.1
10.2
7.8
-0.5
2.9
2.5
-0.5
1.0
6.2
2,3
3.5
3.1
1.8
4.0
8.2
7.5
0.0
1.4
0.9
-1.1
1.1
4.2
-1.4
1.
Deficit
5. Actual Deficit
minus Stability Value
6. Structural
Deficit
7. Structural Deficit
minus Stability Value
Soume by line:
1.
2.
3.
4,6
124.9
-1.1
Figure 6.
1995 Potential Output growth from Giomo er. al. (1995), Table 2, plus 1995 rate of
change of GDP deflator, from IMF World Economic Outfook, October 1995, Table
A-9.
Line 1 times line 2, calculated as perunt.
Figure 5.
Table 2
Change in Effective Exchange Rates, and Growth Rates
of Nominal GDP, Real GDP, and GDP Deflator, 1992-95,
Five Appreciating Countries and Five Depreciating Countries
Appreciating
Countries
Effective Nominal
Exchange Rate
1992:Q2
1995:Q2
1995-1992
Percent Change
Nominal GDP
1992
1995
1995-1992
in
Percent Change
Real GDP
1992
1995
1995-1992
in
Percent Change in
GDP Deflator
1992
1995
1995-1992
Soume Notes:
Depreciating
Countries
Difference,
Depr - Apr
99,5
109.7
10.2
99.1
76.9
-22.2
-0.4
-32.8
-32.4
4.2
4.6
0.4
5.0
6.7
1.7
0.8
2.1
1.3
1.7
2.7
1,0
0.2
2.9
2.7
-1.5
0.2
1.7
2.5
1.9
-0.6
4.8
3.8
-1.0
2.3
1.9
-0.4
Appreciating countries are Austria, Belgium, France, Netherlands, Switzerland.
Depreciating muntries are Italy, Portugal, Spain, Sweden, and the United Kingdom.
All data are aggregated using 1985 Summers-Heston GDP weights.
Nominal and real GDP and GDP deflator growth rates are from IMF Wodd
October 1995, Tables A2 and A9.
Economic Outlook
I
.
,
.
.
,
,
/
,
.
.
.
1
\
. .
. .
‘.
‘.
.
I
1
I
n
I
t
.
0
,
,
,
{
1
In
\\
1-
0
5
w
■
--l
\
I
.
\
\
\
.
.
/
1
-I
I
I
\
\
t
I
I
1
I
J
I
I
\
I
I
I
0
m
m
.
“-
1
I
1
1
I
I
I
N
ai
1
q
0
m
q
‘1
1
I
1
I
I
I
I
I
I
0
\ / ..
)
““””””’
I
\~
/
I
f
\
.:””””
.,
,...’
I
I
,#
:“
\
)
...‘“
4
/
‘...
\’ “:
\:
w
a
■
■
F
\
o
0
o
CD
““””[)”,
...
4
.-
“.
.’
“.
~’
,’
/:”
/,”1
!
c .“”
(
.:~...
~:”
) ..,.
\ “::
n
:
“>
\:
.
\
..
..
..
..
..
..
,.”
. ..
.,””
,..
!/
..
..
\
.
:/
/’
I
1,
“.
.,
..
..-”
..
..””
..
“:-
---
..
---
..
/
I
>
,%
1
\’
.“
I
-.
-. ---
\
“>
~z- “\
.“
.“
“
...
.. ..
--
‘.
#\”
I
(
.>
/
‘,
I
I
.
7
\“’,
.’:/~
.
..’”/
.. \,’ ‘ ‘
d
..../\
b
\
3
“=
!
,
<..’
../)”
..”>
.... ‘<
.
t
..’
I
“. *
‘.
#’
t.
:.
I
1“”..,
\ ,:...’”” \
.’
\
I
.
I
*
I
.)
\;
.\
.. .
)
. ...\
“,
“’”\
“~.,
/
“k
.. <~
,.. ..
.’
‘$
‘.<’
....
# >
,
#
#
/
. #.
..:, ‘/ #
,
9-
.’
●
““““.\R
/
/“
I
/
.... ‘\ #
I
,
......
I
‘1
...
..
..-”
..””
.. ,,.
.,.
..1
..””
--
..
..
k
..”
.
‘..
..
...
.,
. . . ..””
.
0
N
/“
In
o
m
o
Inflation
Rate
Long-Run
Phillips Curve
LP
I
t
DG1(X1)
P2 -. ...-. ............
DGO(XO)
R
Po --------
--------
--------
-.
1
1
1
SP1(P?)
sPo(p;)
1
1
1
1
1
1
1
1
1
u,
u*
Figure 10
Unemployment
Rate
Inflation
Rate
X
L
PO
----
I
1
---------.---- 4---”
\
\
u’
Figure 11
M
SP- (adverse
SPO (no shock)
SPU (beneficial shock)
Unemployment
Rate
LPO
LPI
N
\
/
L
-.
-.
!
PO
--------
. -------
--------
.. -. -..
...
-P’
/
--------El
--------
-.
M
\
\
SP’
(adverse shock)
\
SPO (no shock)
/
u;
Figure 12
u;
Unemployment
Rate
Inflation
Rate
LPI
LPO
DG1(xO)
\
Po
-------
DGO(XO)
/
X
--------.
\
. --- . . . . . . . . . . . .
El
/
x
/
EO
‘b-. .--”*
A
\
SPO(F?:)
/
\
u;
u:
Figure 13
sP1(py)
Unemployment
Rate