UNDERSTANDING
BUSINESS
Robert
E. Lucas, Jr.
University
CYCLES*
of Chicago I.
Why repeated Keynes’ the
is it
General
main
in capitalist
about
Theory,
outstanding
this challenge business
that,
fluctuations
trend,
the resolution
challenges
cycle theorists,
there
the
stand
the
problem
Keynes, aggregate
theory
of
the
consequence
of the Keynesian
away from this question of this
was a rapid increase economic
theories
at a point
underinter-
which
has
Lausanne
Revolution
was the redirection
onto the apparently in time, taking
Revolution,
due
in the level of precision were
with
general
by the
economics.2
consequence
the interwar
remains
quantities,
expressed
of output
to meet
among
Theory;’
primarily
of theoretical
effort
and attempts
into
theory,
Cycle
we here
perfectly
of the determination secondary
theory’
been
A primary
phenomena
Trade
of all economic
most
as one of
as to what it would mean to
contradiction,
of
modern
Prior to
was regarded
Moreover,
equilibrium
dependence School
of research
of cyclical
of economic
By ‘equilibrium
undergo
as a leading example:
they are in ‘apparent crucial
variables
the same character?
research,
theory.
was wide agreement
[T] he incorporation which
cycle
To cite Hayek,
the system
aggregate
of this question
to economic
were called business
solve this problem.
economies,
all of essentially
formulated.
more
simpler
history
to Tinbergen
and explicitness
As a result,
question
as given.3 than
A to
with which
Keynesian
macro-
*
Paper prepared for the Kiel Conference on Growth without Inflation, June 22-23,1976;revised, August 1976. I would like to thank Gary Becker, Jacob Frenkel, Don Patinkin, Thomas Sargent, and Jose Scheinkman for their comments and suggestions. 1
Hayek (1933), p. 3311.
‘Hayek (1933), p. 42~ 3 Thtsredirection was conscious and explicit on Keynes’ part. See, for example, the first sentence of his chapter on the trade cycle. “Since we claim to have shown in the preceding chapters what determines the volume of employment at any time, it follows, if we are right, that our theory must be capable of explaining the phenomena of the trade cycle ” (1936), p. 313.
economics has benefited from several decades of methodological improvement whereas, from this technical point of view, the efforts of the business cycle theorists
appear hopelessly
outdated.
Yet from another that
point of view, they seem quite modern.
macroeconomics
commonplace,
is in need
and
need and about economists
though
there
what it would
would
problem
have
as roughly
of a microeconomic is much
no
difficulty
equivalent
a coherent
and useful aggregate
from
further
statement attempts
macroeconomics there
one’s
is a more
immediate
a theoretically
sound
contribution
once
which
sources
the
toward and not
Keynesian
discussion
of the could from
effect
hoped
could
The process
and legislative to explain
immediate,
an undesirable
scientific
or very state,
had been
the hope
or their influence envisaged
that,
mitigated
was the painfully
on the other
side, there
improvement.
cycles
accompanied
short-term, however
of the
the public
cycles”
with
institutional
business
current
and which
business
reform;
for
interest
in the nature
of instability,
be removed
but
search
of the work of pre-
of
to offer
to “explain
or “permanent”
effort
aim in itself,
the contemporary
redirection
sources
changes.
of long-term
abandonment policy
theorists,
to which
was a sharp change
The effort
institutional
hope
economy
cycle
structures
as a resumption
the
economists
institutional
these
slow one of public
that
of
will result from the acceptance
is a worthwhile
economics
Revolution
to accept.
understood,
the
theory
for interpreting
Accompanying
at identifying
by appropriate was
statement
or not this is so, I wish
by the business
ancestors
aggregative
to policy largely
directed
Hayek’s
Whether
the jerry-built
reason
by the Keynesian
has come
economic
intellectual
theorists.
occasioned
of this
it, it is likely that many modern
has led us.
Honoring
Keynesian
own.
has become
the nature
be so, or that the most rapid progress
as advanced
to refine
about
accepting
to their
in this essay to argue that it should of the problem
confusion
mean to satisfy
The observation
foundation
The a belief
movement
arrived
of the
at, to a better
state. The to come
belief
closer
economics
is so
are
only
symmetric.
If
manipulation diverts entirely
the
from
argument
much
erosion
the
cycle
discussion
such postponement
understandable
is attainable, to
to make life more
improve
business
on which
that
attempt to
objective
it is the only legitimate
widespread
trying
attention
be effective;
this latter
a willful
“destructive,” who
that
to achieving
and that
the attempt
task of research
in aggregate
the
contrary
difficult
lot
of mankind.
theorists
were
of aggregative is, moreover, in the belief
8
Yet
correct,
economics
of stabilization
policies accompanied
is viewed
as
for one’s colleagues the
situation
the
is now which
is
short-term
focused might
by the steady
on the part of noneconomists
only
actually and that
aggrrgative
economics
has anything
In the next section, the events these and
we call business
cycles,
facts, to the progress finally
apparent.
to
the
Hayek and other
limits
of the
phenomena
and then turn to the Keynesian
made along the line Keynes
severe
The remainder
for cyclical
useful to say.
I will review some of the main qualitative
to
this
theory,
cycle theorists
have
of to
and Tinbergen
which
essay will consider
by an economic
business
progress
features response
have
initiated,
now
the prospects
become
of accounting
in the narrow
sense in which
that term.
LI&
II. Let me begin to sharpen
the discussion
by reviewing
the main qualitative
features of economic time series which we call “the business cycle.” movements about trend in gross national product in any country described
by a stochastically
These
movements
which
is to say, they
sometimes
do not
arise in the natural
The
principal
broadly
defined
exhibit
high
conformity;
amplitude
prices
of
aggregates
(iv)
time
of producer
goods
are
and profits
show
series
procyclical; measures
long-term
which
are observed
(i) Output
movements
terminology,
language,
they
durables
they
have
exhibits
high tnuch
of nondurables.
(iii) Production
resources
lower
than
average
and
much
greater
high
series. (v) Prices generally
and velocity
which
(In Mitchell’s
and consumer
natural
or amplitude,
wave motions
time series.
fol1owing.4
move together.
of very low order. period
regularities
aggregative
are the
in modern
Business
than other rates
Those
than does the production
agricultural
conformity. interest
sectors
(ii) Production
greater
sciences. these
equation
of either
the deterministic
among different among
across
coherence.)
difference
uniformity
do not resemble
are in the co-movements
amplitude
disturbed exhibit
Technically, can be well
rates
have
conformity are procyclical. slightly
so.
and
(vi) Short-term (vii)
Monetary
are procyclical.
4
The features of economic time series listed here are, curiously, both “well known” and expensive to document in any careful and comprehensive way. A useful, substantively oriented introduction is given by Mitchell (1951), who summarizes mainly interwar, U.S. experience. The basic technical reference for these methods is Bums and Mitchell (1946). U.S. monetary experience is best displayed in Friedman and Schwartz (1963). An invaluable source for earlier British series is Gayer, Rostow, and Schwartz (1953), esp. Vol. II. The phenomena documented in these sowces are, of cowse, much more widely observed. Most can be inferred, though with some difficulty, from the estimated structure of modem econometric models. An important recent contribution is Sargent and Sims (1976), which summarizes postwar U.S. quarterly series in several suggestiveways, leading to a qualitative picture very close to that provided by Mitchell, but within an explicit stochastic framework, so that their results are replicatable and criticizable at a level at which Mitchell s are not.
9
There
is, as far as I know,
restricting
them
regularities
common
absolutely conclude series,
that,
a unified
governing
market
to the qualitative
economies,
omitted
rather
countries
the
the
U.S.,
trade
For
1 think,
beg the
origins of cyclical ment,
market
economies
experiencing, attempts
exhibit
the
general
laws
character-
statistics
the above
bccausc,
high
from
for a large economy
enough
crucial
a phenomenon
purely
suggests
that
which
requires
of a
conformity export
question
to be
movements explanations
of the ultimate
to pass over without
com-
series in the twenty-five
years
descriptive Nevertheless,
there
is nothing
level,
it is impossible
so long a period inherent
and account
in the pre-World
for regular
in any way to a presumption
cyclical
to
of relative
in the workings
living with the level of instability
or to which we were subject
of capitalist
this
the possibility
or institutional
to be sure,
and
in amplitude
War Il. At this
to document
connected
trade
country,
good luck from good policy. strongly
in
among
economists,
movements. but too striking
stability
of co-movements
but to focus on open-economy
reduction
World
is to
grounded
in part
difficult
is the general
following
feature
cycles,
Also omitted,
distinguish
there
for it suggests
of foreign
do not
more
Though
by to be
is led by the facts
inclined
than in political
a smaller
would do much to “explain”
appear
or periods.
behavior
statistics
interesting.
would,
behavior
cycles,
of phenomena-to-be-explained,
cyclically
observations
they
economies. it, one
and challenging,
of business
these
periods:
are all alike. To theoretically
to particular
I have
to anticipate
be attractive
market
catalogue
reason
explanation
istics specific
like
to all decentralized
cycles
should
to qualify
or time
with respect
business
no need
countries
no theoretical
conclusion of
to particular
of
we are now
War II years. That is,
movements
that such movements
need not be
are an inevitable
economies. Ill.
The implications the time
series features
1937,
and
which
developed
early
work’
Modigliani, from
conform
of Keynesian reviewed 1944) this well
macroeconomic
above.
fit well qualitatively;
theory
and from
quantitatively.
5
models
Early versions
the
Tinbergen’s
These
confomr
(for example,
models
well to by Hicks,
econometric
models
largely
independent
located
the primary
For example, see Tinbergen (1939). This work was not explicitly Keynesian; indeed, it was conceived as an empirical complement to Haberler’s review and synthesis of theoretical work on business cycles (1936). Keynes, on his part, was actively hostile toward Tinbergen’s work. See Moggridf: ,‘1~3)~~~. , , 285-320. In referring to those who built in part on Tinbergen’s work as “Keynesian contributing to the continuation of an historical injustice.
10
disturbances to the then
in investment
highly
induce
bances
volatile general
interest
was a good
movements
rates
empirical
In
this
accounting theoretical
Keynes,
from
The empirical of the
movements
This feature
The
seem very important and historically
Tinbergen
success
generated
Goldberger answer,
Klein
model (1946)
to the
this criterion) business
cycles.
a &
in the most through
economics. of
meaning theories
emphasis
to
grounds: did not
could distinguish
surprise had
in an original
in their simulation
The Adelmans armed
between
a collection
programmed
of economic
to follow
series generated
posed,
with the methods the Klein-
by an actual economy.
of the Adelmans
(and,
in no way directed
their
one suspects, efforts
of
to meeting
was =.8
This achievement
behaves
(1959)
an observer
by a computer
who
the
factors
was measured
of the U. S. economy. of whether
evident
in
from
empirical
monetary
and Adelman
and the analogous
and Goldberger,
that
on
role
directly
all gave great was
of these developments
artificially
equations
money
(still
wage and price
no important
did not result
on discovered
apt way by Adelman
way, the question
procyclical
and Modigliani on
and prices sector
7
empirically.
of Burns and Mitchell series
Hicks,
distur-
with procyclically
on, a wage-price
play
series
Since these
of rigid wages
in money certainly
de-emphasis
Klein-Goldberger
in a precise
Later
U. S. model)
high-amplitude
they were consistent
The assumption
curve) was added to fit observed
models;
econometricians
in these
and employment.
” IS shifts,”
description,
forces.
via lags (in Tinbergen’s
Movements
first approximation.
for cycles.
monetary
linked
in output
and velocity.
later called a Phillips movements. 6
level
series.
were, in Hicks’ terms,
moving
The
behavior,
profit
signaled
One exhibits literal
a new standard understanding
sense: a fully articulated
time so as to imitate
The Keynesian
for what it means to understand
of business
closely
macroeconomic
explicitness
and
empirical
of the term
“theory”
cycles
artificial
by constructing economy
the time series behavior models
accuracy;
could not really be viewed as “theories”
were the first to attain
by doing
to such an extent
which of actual
so, they
altered
that the older business
this the cycle
at all.
These models are not, however, “equilibrium theories” in Hayek’s sense. Indeed, Keynes chose to begin the General Theory with the declaration (for ‘Klein and Goldberpr
(1955).
‘Tinbergen (1939). pp. 183-185. Tinbergen, as did most subsequent significance of interest rates to test the importance of money. 8
macroeconometricians,
used the
It ISnot correct that a search for “good tits” would have led to a model satisfying the Adelmans’ criteria; think of fitting polynomials in time to “explain” each series over the sample period.
11
Chapter
I1 is no more
than
this) that
an equilibrium
that unemployment
was not explainable
and that the failure
of wages to move as predicted
be treated
as due to forces
Keynes verifiable
wrote
serious
whether
between
sharper
fluctuating
than
at sharply
irregular
this crucial
behind
this
“cyclical market
rates through
problem.
nominal
should
an otherwise
classical
This decision
1 have described,
Now in possession economy,
means
to
actual
evaluate
various
proposed
as though
they had been experimel:tally
been attempted
conditional requires type hope
studied.
a property that
neither
countercyclical with
changes
structed choose
of a model
to imitate
has almost
nothing
to answer
had certain
invariance
being
course,
(1959) forecasts,
have differed
aCCUI’dte
measures.
emerged
It seemed
from this procedure
even if such policies
Invariance
had never
codifying
of the model
under
policy
can be assured
in advance,
nor technology
vary systematically
In contrast, how
agents’ with
to a new situation. the decision sample period,
will be of no use in predicting
stable
without
agents
explaining
the consequences
12
variations model
of the
is not, of
tastes
to
with variations
rules will in general
model
Any disequilibrium
rules which
behavior
but it seems reasonable
decision
An equilibrium agents
would
ways? This ability
in an economic
which
by
to make accurate
how
in specified
tastes
policies.
over some previous
in the way tested
of the form:
been different
of parameters
in the environment.
to respond
actual behavior
to do with its ability
questions
policies
of the structure
so as to predict
by simply
quantitatively
by
to have an inexpensive
policy
which
theory,
exploited
in any actual economy.
Yet the ability the Adelmans
appeared
tested,
of his day freed
by equilibrium
of detailed,
recommendations
saw
of rigid
into a model
and fruitfully
economic
to treat policy
it, Keynes
postulate
theorist
imposed
economists
legitimate
labor
had avoided
time series.
was rapidly
replicas
to supply
col!!d be transformed
from the discipline
macroeconometricians. of the
choose
cycle theorists
it had not resolved
for observed
this freedom
“apparent
equilibrium”
Why, in the face of moderately
on the part of the most prestigious
of economists
there were
is the
with the unexplained
model
were
look at a market
“economic
households
time? Most business
which did a fair job of accounting
and,as
behavior.
this problem
of unemployment
for nowhere and
was to
to illuminate.
or not. Nevertheless,
choice,
choices
theory
theory
one could somehow
and those who addressed
sidestepping
a generation
nature
phenomena”
wages and prices,
that by simply prices,
as though
in labor
nominal
of economic
it is in equilibrium
reasons
contradiction”
the power
was unattainable:
of individual
by the classical
the “involuntary”
observation,
directly
empirical
theory
as though
by direct
and verify
beyond
theory
as a consequence
in
change
is, by definition,
con-
and technology
will
model,
have found
constructed
it useful to use
why these rules were used, of nontrivial policy changes.
The quantitative
importance
by examination
of specific
settled
elsewhere’
that it is of fatal importance
economic
models,
these models. graphic
primarily
Rather
illustration:
expansionary
4 percent,
has averaged
were
widely
the
decrease
In this
the policy
deeper
curves followed
very
homogeneity specific
the initial response
sustained
inflation,
were
equilibrium
an
equilibrium
case
apparently variance, monetary flexible
of
on the
and supply
and without that,
(or
“natural”)
sustained
inflation
growth rather
than
of real output
were adopted? fixed
has
Address. aggregative
models. On the equilibrium: the
Thus, without to forecast
using in any
one can, in the case of
rate prior to the inflation the
same
is a relatively
not -too simple, as it is still highly one would need a more moments,
Phillips which
rate
will
be an
is underway.
questions,
and other
of a specific
the ability
then
The issue is
Presidential
to an inflation, rate,
review.
models
functions.
is no
inflation.
to econometric
the standard of economic
if the unemployment
that
unfortunately,
not be lost.
his
in had
behavior
sustained
detailed
basis
claiming
of the economy
once the inflation
The policy
reason
from
in
than characteristic
of demand
model,
clear
sector”
it was based on apeneral
zero-degree detail
made
(1968)
inflation,
in macroeconomic
proceed
“wage-price
Phelps
was, most
of this episode
revision
did not
a better
result
in question
forecasts
involved
equilibrium
of a few new variables
as Friedman
argument
with
contrary,
the addition
it),
that
of rates
than unemploy-
lo These
by sustained
would
the lesson
this is the only
from
Friedman’s model,
any
than
(though
observation
that
inflation
closely
and
is now too clear to require
that
lower
themselves
(1968)
represented
experiment
out, and its outcome It is important
much
of the
change
implied
unemployment
point
not
Friedman
in unemployment
instance,
carried
basis
units
models
to a sustained
of U. S. history.
economists
in
“stagflation.”
leading
any long period
macro-
in detail, let me cite the most
a full percentage
Earlier,
on the
under
of modern
of expectations
wo~11d lead also to sustained
by many
forecasting.
purely
sustained
during
to be
I have argued
all sectors
U. S. econometric
policies
a matter
models.
treatment
during the recent
or about
endorsed
econometric
in virtually
these arguments
fiscal
per annum
ment
argued,
and
is, of course,
in specific
of the faulty
as 1970, the major
4 percent
invariant
than review
monetary
of less than
because
our experience
As recently about
of this problem relationships
exchange
Under
simple
controversial). explicit change
a balanced
model. budget
‘Lucas (1976). “Hirsch (1972), de Menil and Enzler(1972).
13
(though
For other kinds of How
if a policy
rates? One can generate
one
would
the
of 4 percent
fiscal rule? numerical
Under answers
to questions of this sort from current macroeconomic n~odels, but there is no reason for anyone to take these numbers seriously. On the other hand, neither can quantitative answers be obtained by purely theoretical reasoning. To obtain them, one needs an explicit, equilibrium account of the business cycle.
IV. I have summarized, in section II, the main features of the cyclical behavior in quantities and prices. In section III, I have argued the practical necessity of accounting for these facts in equilibrium (that is, non-Keynesian) terms. That is, one would like a theory which accounts for the observed movements in quantities (employment, to observed movements
consumption, in prices.
In the next section, 1 individiral decision making to .explain, in particular, why the importance. Given this general
investment)
as an optimizing
response
will describe the general point of view toward be taken in the remainder of the paper, and will recurrent character of business cycles is of central view, I shall consider in sections VI and VII the
way in which relative price movements induce fluctuations in employment and investment. Sections VIII, IX, and X examine the conditions under which these safie quantity responses may be triggered by movements in general. or nominal, prices. Not surprisingly, the sour’:e of general price movements is located, in section
XI, in monetary
changes.
V. The view of the prototypical individual decision problem taken by modern capital theory is a useful point of departure for considering behavior over the cycle, though it is in some respects highly misleading. An agent begins a period with stocks of various kinds of capital accumulated in the past. He faces time: paths of prices at which he can trade in the present and future. Based on his preferences over time paths of labor supplied and goods consumed, he formulates a plan. Under certainty, he is viewed as simply executing a single plan without revision; with uncertainty, he must draw up a contingency plan, saying how he will react to unforeseeable events. Even to begin to think about decision problems of this general form, one needs to imagine a fairly precise view of the future in the mind of this agent. Where does he get this view, and how can an observer infer what it is? This aspect of the problem has received rather offhand treatment in traditional capital theory, and no treatment at al1 in traditional macroeconomics. Since it is absolutely crucial for understanding business cycles. we must pursue it here in some detail.
14
At a purely a subjective which
joint
impinge
probability
distribution
“uncertainty”)
(such
agent.
the general
hypothesis
makers
in many
applications,
way
has,
of inferring
hypothesis
of the decision
any distinction
distinction
between
between
“risk”
and
is, at this level, meaningless.
Unfortunately, decision
philosophical
on the structure
(1921)
variables
The link between
is a most complex
In particular,
as Knight’s
formulate
random
opportunities.
and “reality”
as seen by an individual
agent must
over all unknown
and future market
view of the future
of randomness
some
that a rational
but the way it is solved has little effect
problem types
level, we know
on his present
this subjective question,
formal
what
an agent’s
that economic little
subjective
is of no help in understanding
can be (and
today,
view of relevant
is) understood
probabilities.
of psychoanalysis,
as “rational,”
one hopes)
are Bayesian
content:
without
view of the future
his behavior.
To practice
agents
empirical
behavior
given a sufficiently
economics,
of understanding
is, this
Even psychotic
abnormal
we need some way (short
which
decision
problem
agents
are solving. John Muth (1961) proposed subjective
probabilities
or with “true”
“true”
with
probabilities,
probabilities
situations
in which
relevant:
which
in situations
defined
recurrent
situations
event,
economic
recurrent
character similar
changes
have fairly they
stable utilize
free of systematic 11
the
why
business and
why
it will be reasonable
arrangements information
In
terminology.
in terms of economic
m Muth’s cycle
sense.
for collecting
In cases
to treat
emphasized
hope
as repeated agents
they
the
as reacting
are rational future
were
instances to
that they
and processing=ation,
in forecasting
and easily correctable
theorists
we must
as “risk,” or to assume their expectations this
likely be a fairly well
on the part of agents will
cycles can be viewed
in a stable
and way,
biases.
Kmght(1921).
tbffe;;;
Knight’s
are rational
cycle,
as business
events,
concern
are
will be of no value.
explain of
in
behavior
in
frequencies
It will m
of interest
“risk”
and
will not be
will it be applicable
may be explainable
expectations reasoning
considerations
of essentially
of
of subjective
this hypothesis
“uncertainty.”
of rational
so that behavior
right in doing so. Insofar
that
1 1 called
agents
to be forecast,
if any, observable
the probabilities
situations
In such situations, These
cyclical
Knight
coincidence Neither
which,
by identifying
of the events
Evidently,
behavior.
guess
of risk, the hypothesis
of uncertainty, the
psychotic
in which
have usable content, theory.
expectations.
one cannot
situations
useful
frequencies
calling the assumed
rational
of value in understanding
to resolve this problem
observed
I am hterpreting the risk-uncertainty distinction as referring not to a classification of types of individual decision problems but to the. relationship betwe~decision maker and
15
VI. In moving will bc helpful
from
these
general
to consider
considerations
as an exan~ple
a single worker-producer,
confronted
each period
a good which he then makes to order, he
The good he receives with
for grunted.
in exchange
the historical
This money,
from
day to day. Some purchases
from
work,
general
or average
fluctuate
level
but simply
on a wide variety
of goods,
he makes on his way home,
of prices
let us postulate
Now
rcsqond?
does
not
an increase
to the average
The answer
change,
take it different
in an hour’s break
though
this
price
work no harder,
by economic
individual
prices
unfortunate
price
change
mcans.
theory
If he believes
since
will be immediate)
be the case if each period’s The aliswer
producer
is willing
is highly
substitutable
the
labor
change
extreme. price
for labor
he will work
longer
to transitory
price
the
indicates
nearby
by Rapping
than
about
leisure
periods.
and myself
(1970);
(as would
from
a fixed
the rate at which the If “leisure”
on high price
actual labor supply response,
is an excellent
evidence Ghrz
or negative.
tomorrow.
“long-run”
in one period Systematic
drawing
to knowing
today
we do know
that
a
to a permanent
is transitory
were an independent labor
over time, movements
signals
that he will
are zero
the price change
in this case amounts
to substitute
response
elasticities
about
was obtained
the price
“long-run”
supply
close early on low price days. Less is known
in other,
producer
must be: who knows’? At this
a little less hard. That is, we know that “long run”
terminology,
What if, at the opposite distribution)?
selling price,
in his selling price, we know from much evidence
and probably
(very
in today’s
How will this hypothetical
which would enable one to imagine what the producer
movement
change
of 10 percent
of past prices.
given
I have said nothing
permanent
leisure
I shall not be
is “money”;
from day to day.
as compared
thinks
how
or several days later. I ~SSLIII~~ for now that he holds no other 1 assume also that this agent lives in a cycle-free world, in which the
securiti:s.
point,
selling price, determines
for this arrangcmcnt,
is spent
price for
per hour. That is,
then goes home to relax.
for the effort
reasons
in turn,
theory, it agent. 13L Imagine
with a given market
his current
many hours to work that day, sells his produce, concerned
specific
at a fixed rate of output
to his place of work, observes
COIR~
to more
a “representative”
at the
and Becker
days and responses but what
substitute aggregate (1975)
for level
reached
12 Many of the arguments in this and subsequent sections lwe been developed more explicitly elsewhere. The closest sin e parallel treatment is in Lucas (1975). See also Pbelps, e_tal. (1970), Barre (1976), Sargent and Wallace (1 d 73, Sargent (1976). In what follows, I will not document particular arguments, nor will I attempt to apportion credit (or blame) for ideas discussed.
16
the same conclusion induce
workers
Saturday,
two
conclusion,
at a disaggregative
to shift weeks
and this
its probabilistic of this evidence,
holidays in March
level. The small premiums
and
vacations
(take
rather
than
in August)
evidence
“cas~~al”
simplicity: holidays one would predict
is somewhat
Monday
more
required
to
off instead
of
point
to the
impressive
same
because
of
are known to be transitory. On the basis highly elastic response to transitory price
a
.
changes
Before
dealing
for business price
cycle
with complications theory.
fluctuations
exactly
what
intelligible prices
with
large
we observe
effects,
our aggregative
generally;
the example
not
in
responds
output
and rests
unintelligible
on
note its promise
us
who
The description
observations refers
let
a producer
fluctuations
over the cycle,
substitution
us go slowly:
to this example,
I have described
on
economically
“disequilibria.”
refer to co-movements
to relative
to small
employment: Yet let
of output
price movements
and
in a stationary
environment. Before
facing
the example seems
just
mart’
producers
this
difficult
considered.
realistic
to
by quantity to be
There
seems
as signals of current
me harmless
and
interchangeably. equilibrium
let
from
a descriptive
of
accurate
determination relevant
A second
variation
when
producers
reason
on
it often
conveyed
to
and the like.
to focus
exclusively
on
price
increase
and sales increase
however,
rigorous
set prices
is extremely
difficult,
than consider
a worker-
cycle behavior
these
being rundowns,
functions,
analysis
of
exist.
is easy to carry out. Rather separate
variations
of view,
At this verbal level, it seems to
terms
surprisingly,
to business
one could
information
demand.
to use the
some
point
inventory
substantive
and future
Somewhat
us consider
demand
new orders,
compelling
no
and no examples entrepreneur,
thinK
changes:
prices
issue,
First,
introduce
firms,
and consider
labor and product markets separately. In the present context, this would introduce a distinction between wages and prices, and raise the issue of riskallocating permit
arrangements
the study
None
of
these
peripheral the
tendencies.
employers
different is without
cycle theory.
neither
This suggests
Accordingly,
130ne
but
a central
“wages”
questions
for business
cycle,
ments
between
of possibly
do
that any attempt
I will proceed
and “prices”
interest, Observed
they ‘exhibit
role in an explanation as though
and
l3
workers.
information
It would
but
all are,
in my
opinion,
real wages are not constant
consistent
pro-
or
to assign systematic
of business
also
sets for firms and workers. over
countercyclical real wage move-
cycles is doomed
to failure.
the real wage were fixed, using the terms
interchangeably.
such arrangement is the practice of “laying off’ workers. See Azariadis (1975).
17
Additional LWS
variations
can be obtained
of the worker-producer’s
attempted,
for example,
in job search. some
other
of substitution is little when
unemployed
than of the
elasticity
movements,
“leisure”
unemployment
not
the reasons
matter,
a testimony
there
that
that
measured
the important
with
why
that the unwillingness
is more
Nevertheless,
at all. Economically, of employment
i&
in the importance
that search is less costly
for that
or,
have
substitues
one
illustrations.
in job search,
various
as time engaged
that one’s belief
employed,
any activity
I suspect
Indeed,
doesn’t
when
unemployment
plausible
time is spent
among
Many writers
away from work
shows
by some
measures
and price
as enjoying
if one substitutes
that much
is the, magnitude
measured
and experience
is bolstered
evidence
unemployment wage
to interpret
Certainly, activity,
by distinguishing
time when he is not working.
respect
issue
to transitory
elasticity
is what
to speak of workers
to the force of Keynes’
it is.
in recession
insistence
that
is “involuntary”
want
than a response to observed phenomena. One that people !& depressions! Of course, the hypothesis
to suggest
of a cleared
labor
observation
that people
market
carries
with it no
go hungry
suggestion,
SLIC~I
in cleared
food markets
any more than the suggests
that people
enjoy hunger.
More
complex
variations
VII. on this example
kinds is introduced. Let us do this, retaining over time in the general level of prices. Three
possibilities
can be stored the account surely labor
of interest
as finished
supply
relative
sketched storage
also dampens
in the elasticity
to price, and an increase
in the real sales-price
As a second
suppose
production future
to acquire
periods.
the same
As a’ third,
effect.
may be considered
Since
earlier,
provided
the producer
which
it is clear that neither
the new capital
can be applied
will
On the industry The net result
can use a part of his current
will raise
his output-per-hour
transitory
of these options his original
to production,
will have vanished.
18
in school
do not differ
of purely
with
is
with respect
elasticity.
the producer
two possibilities
was satisfied
The producer
movements.
s~lppose he can take a course
these
made it appear profitable
above.
of employment-production
as one. In the example
discussed time
a machine
production
seems to work against
is precluded.
price
likely to be a reduction
possibility,
of various of stability
for sale later when price is high, smoothing
to the case where
this behavior
capital
that current
This possibility
co-movements
in low price periods
level, however,
arise. First, suppose
goods inventory.
of price-output
produce
arise when
still the assumption
in all
which will have
economically,
they
price movements,
will ever be exercised stock
--
of capital.
By the
the price movement
which
Current capital
relative
price
accumulation
permanent.
In
insensitive moving
movements
when,
this
case,
to price
at the however,
movements.
systematically a situation,
engineers variables
(the transitory
observed
directly;
knowledge change, his
these
decisions
components, depends leisure
our
increase
in
unobserved. factors:
and
(price)
in what changing
fundamental
which
cannot
be
together
with
his
sources
of price Based on
calculation,
out
producer
he takes
to be an average
Under
supplied,
expansion
in productive
symmetric;
the responses
optimal
the way he interprets concerning
assumptions
his response
a
of the
to face stochastic
of transitory
response
to
and permanent price
the information intertemporal consistent
to an unforeseen
a decline
capital
is taken
as a mix
His
his preferences
evidence, labor
turns
hypothetical
is describable
and consumption.
and available
of price)
movements,
be
elements.
to engage
in more
probability
decision
it must
in the two components.
conditional the
obliged
of the two unobserved
the movements
be
to the two extremes.
both
changes,
price
as
will
and permanent
movements
on
and employment
a single variable
components
observed
implied
which
on two
from
importance
infers
appropriate
arise
these
surprisingly,
variability,
these
will find himself he observes
and permanent
To recapitulate, price
are a mix of transitory
movements
this
employment
price movements,
from
to
Not
noted,
of relative
of the relative
solution
as I have
effect
are regarded
in the direction
he imperfectly
decision.
maximal they
investment
the producer
time;
their
extreme,
to observe
call “signal processing”:
through
have
Thus,
the case that such movements In such
will
opposite
in finished
movements contained
with rational price increase
goods
of all kinds. This are the opposite. 14
of
behavior is a sizable
inventory,
accumulation
to price decreases
in
substitution
and
an
behavior
is
VIII. It economy goods
to
comprised
and subject
go behind underlie
is time
price them.
think
of
of similar
situating agents,
though
to different
individual
movements
to the
These
changes
this
representative of course
price movements.
changes
are occurring
producing
time
in an different
To do this, one must
in technology all the
producer
and taste and, ‘indeed,
which their
14 What is happening to consumption expenditures as these employment and investment responses take place? In his critique of equilibrium business cycle models, Grossman (1973) argues that consumption must necessarily move in them direction from labor supplied. Since this is not what is in fact observed over the cycle, it would indeed by a serious paradox if a negative correlation were a consequence,of utility theory. One M derive it,for special cases (see Lucas, 1972, Fig. 1) but this implication is certamly ma general fact for optimizing households; it does & for example, follow from Rapping’s and my (1970) theory or from that of Ghez and Becker (1975,ch. 4).
19
importance ments
to individual
which
agents
constitute to
the
general,
lead
relative,
reducing
costs of producing
business
not
production
of other reduced
economy, small
SLICII
in the
mere
presence,
to total
per
investments.
ous!y.
Such shocks
of a
and away from the of one good
in a complex
modern
in any given period,
There
will be much
affect
each
“averaging
ments
fit the description
reason
seriously,
It is surely
there
many
price movements
instances
of the economy
in section
(other
of
simultane-
and they will in-
They will not, however,
sketched
for a large
their work weeks
in the way I have described,
in the aggregate.
cycles
in individual
possible
have been
sectors
why one
we call business
of conditions
tight.
all, or many,
will not cancel
fluctuations
movements
important
movements
to feel an urge to increase
More
which
to countercyclical
the production
shifts
the most
is not entirely
duce output
procyclical
output.
in
technology,
of the purchase
se, of unpredictability
spontaneously
to supply
which
SLICII
of the general
Yet this argument of agents
shocks
possible
move-
should,
A new
benefits,
Moreover,
of
movements
movements.
in favor
others.
of this sort is, 1 think,
seek an explanation
and expand
shifts
on
relative
Cancellation
number
these
minor
across markets.
cannot markets.
Taste
will be a large number
effects
Yet
price
into the good which
goods.
in importance
out” of
cycle.
general
expenditures
there
by far the relatively
an old good or making
new one, will draw resources involve
dominates
II: all supply
lead to moveshifts will lead
things being equal) in contrast
to the
we observe.
It is, then, possible to situate our hypothetical producer in a general equilibrium setting, in which his price and output fluctuate, yet aggregate levels do not. His responses
to these relative
responses
price
to general
prices movements
movements
which
will mimic the aggregate
constitute
the business
have then a coherent model, but not one which as yet accounts phenomena to be explained. This model can, without difficulty, permit
general,
supply-induced
blance to the modern Before most
of
present
leaving
the
in such a sdtting. a national
to the concert finance
By the time like these,
of stable
troubles
and
Will consumers joke?
education,
one has acquired
Will a doren hobby?
or tide
it is worth economic
years
stressing
agents
20
design,
in piano
over next month’s
necessary
to resolve
be or lead
wages strike?
questions
one is committed.
and magnitude,
in the year ahead
of training
that
would
Will this week’s overtime
the family
the information
to risks of this nature worked
bear no resem-
take to a novel automobile
it is too late; one way or the other, actually
but these
aggregates,
challenges
stage, or just a pleasurable a child’s
Compared the hours
fluctuations,
We
cycle.
this world
risk which
will it become help
business
output
cycle.
for the general be modified to
the question
will be 1.03 times
of whether
what one plans
for now,
or .97, seems
economic
theory,
imposed
a minor
on an otherwise
the transfer
one, and seems so because
we are accustomed stable
of abstractions
involve fatal distortions
to think of business
environment.
useful
it is. In aggregative
cycles as a kind of risk
Such habits
for some purposes
of thought
into contexts
reflect
where
they
of reality. IX.
Let us now drop point
of view
in the nature given
of the in his
or a transitory
a situation
which,
producer’s
part.
price
individual
as will relative
expansions,
price
the responses
at the aggregate
be solved.
the
change
Before,
a
relative
price
that al prices
are changing,
lead to no real response
on the
and transitory
relative
at the time, neither
General
movements
price increases,
can
exactly
in the same direction
as such ordinarily
traders
not know
as they
in
published those
prices
and carefully,
at all. Of the many behavior
is no reason
of most
sources
for traders
By the same reasoning,
Within
price
price increase
than will be con-
output,
seem
and investment
importance
to specialize
agents, their
not
how
An optimizing
many trader
problem
less so, and most to him, the business of no special
could world,
every day, nor would useful.
be
of the aggregative
implausible:
to his decision
of less importance
is, for most
movements
of a multi-commodity
all prices
particularly
price
the context
may
of risk of importance
generally
general movements
real decisions
indices
those
general
In the reality
want to observe price
even during
in real terms in prices,
that
assumption
t& price of goods?
in demand
earlier,
over markets.
over the actual cycle.
argument
this
described
to cancel
(this is what a general
will be expanding
are occurring.
used,
no one would
will process
declines
will be co-movements
to this
changes
will not tend
increases
level, just as is observed
models
find
and technology
will observe
more
It is essential
agents’
will induce
will observe
The net effect
diagnosing
be distinguished.
price increases
producers
but more
and therefore
aggregate
would
From
a slight
a permanent
out with certainty
to taste
to general
tracting.
there
be sorted
increases,
means),
frequently
mean
one. Now, it can also mean
cannot
To be sure, some
traders
which must
could
diagnosed,
only
and investment.
Unlike
however,
in average prices. involves
Yet, for the same reason that permanent
responses
perceived
this
problem
price”
and general movements
employment these
“own
if correctly
movements
relative
of stability
producer,
of the signal processing
movement
change
the assumption
prices
not
cycle and
importance,
own information
most
systems
and for
correctly. one can see that sustained
in the way that transitory
21
inflation
price movements
will not affect do. Nothing
is easier than to spot and correct involve
no changes
information. for
There
what
inflation
in agents’
may, of course,
it is; about as though
Changes effects
on
as often,
placed
bias in forecasts. Such corrections systems or in the costs of processing
be some lag in diagnosing agents
will incorrectly
sustained
inflation
perceive
a transitory
will have more
fundamental
it were sustained.
in the
agents’
“weights”
systematic
information
degree
of price
information
variability
processing
on price information
idea is that one trusts
“noisy”
behavior,
because
in forecasting
future
they
affect
prices.
the
The general
price signals less.
.r
The aggregate more
complex
Investment reasons
or average
response
as one considers
decisions
will be distorted
as will employment,
to general
investment
price
movements
becomes
as well as employment
responses.
by general price movements,
and in the same direction
for the same
as the responses
induced
by relative price movements. Further investment
complications
affects
seen to extend
follow,
future
capacity,
in time,
perhaps
however,
from the observation
and hence
future
even
to amplify,
detail,
imagine
prices.
that current
This effect
the initial effects
can be
of general
price movements. To spell
this out in more
would,
if correctly
Sooner
or later, then, this adjustment
than not perceive
perceived a relative
As a result,
employment
information
diffuses
mistaken.
and investment
through
postponing
unsystematic
however,
both
or short-term
shocks
of
increase.
however,
to prices
which
generally.
more traders in their favor.
Through
time,
as price
will see they have been
capacity
retards
initial
shock.
the
occurs
in prices
permanent,
these traders
the added
recognition
some event
Initially,
possibly
the economy,
the
that
an increase
will occur.
price movement,
In the meantime,
generally,
by all. induce
can lead to much
price increases In
this
longer
way,
swings in
prices. In addition, of capacity.
have to become There
is no
describable
general since
reason
to expect
scenario,
optimal
automatically
of general
less than normal
crucially price
is a downturn
recognition
as a “crash,”
This depends
there
When
inflation
built in to this expansion does occur,
for a time while capacity this
readjustment
investment
readjusts
to come
will
downward.
rapidly,
or to be
or “bust.”
like the earlier
on the confusion
movements.
This
investment
policy
description
of the employment
on the part of agents
is especially
clear
in the
between case
has a great deal of “smoothing”
22
response, relative
and
of investment, built
into it:
since investment
is a long-term
to be relatively This
of
purchases
contribution For
negative,
investment
and often
to what stems ment.
someone
who
amplitude response
shocks
to skepticism cycle.
rates
of return
in hundreds
a weak added
is often
the situation the capacity
is, at the level at which amplitude
faced by agents. variable,
A quick, current
often
response
the key to a successful is clear to everyone
to meet
invest-
is too late;
the high demand. response
decisions
movements
in durable
insist on the minor
are highly
of percent.
“signal”
until
as to the
can moderate
movements
level, to be a high-amplitude
to still higher
How
risk to the general risk situation
waits
at the aggregate
only to what seem
Here again, one must
projects,
else has already
appears,
grounds, business
lead to the high-amplitude
measured
to others
The agent
in the
are observed?
of economy-wide
individual
on serious
effects
in prices which
it will respond
price shifts.
has led,
accelerator
movements
goods
commitment,
relative
observation
importance cyclical
permanent
are made,
in returns
What
pattern
to low-
a high-amplitude
to individual
invest-
as repeated
fluctu-
ments.15
XI. I began section ations
a certain
typical
a competitive by agents rationalizing sections.
output,
pattern
employment to price
the observed
any we know prices
direction
secular
and output movements
movements
do extremely
in aggregative
economics,
of effect
kinds
appropriate
as rational
with
Since in
are chosen to begin by
or optimal
responses
in the preceding
five
of price movements.
of money
or the quantity
of various
it seemed
associated
other variables.
This has been accomplished
1 turn next to the sources
For explaining
cycles
of output,
in prices and
movements,
quantity
price movements.
in the quantity
of business
and the composition
of co-movements
economy,
in response
to observed
either
II with a definition
in employment.
in prices
and is not sensitive
of money.
in this relationship;
generally,
secular
movements
well. This fact is as well established I6 There
to how one measures
is no serious
no one argues
as
doubt
as to the
that the anticipation
of
15“Austrian” or “monetary-over-investment” business cycle theory (see Haberler, 1936, or Hayek, 1933 ) was based on this same idea of mistaken investment decisions triggered by spurious price signals. However, the price which this theory emphasized was the rate of interest, rather than product prices as stressed here. Given the cyclical amplitude of interest rates, the investment-interest elasticity needed to account for the observed amplitude in investment istoo high to be consistent with other evidence. 16Friedman and Schwartz (1963).
23
sixteenth-century to finance
inflation
rcfcrs
to averages
strong
to be ignored:
real variables that.
sent Columbus
it. This evidence
over much longer periods, we have accounted
“long
run,”
general
in the
quantity
of money.
enough
to be quantitatively
shock as the force triggering The direct indicate
that
however,
more
between
general
price
This connection identifying
general
one
could
function far from would
between
to a monetary money,
episodes
money-induced.17
and other
to prices,
cycle theory.
describe
short-term
of lagged
movements
output, appear
to
In general,
variables
is agreed
to be
evidence
linking
is encouraging
from
money
To see why, recall the theoretical
and
economic
activity
as skctchcd
that the signal processing
from observations by agents.
of link
above.
problem
of a few individual
of
prices
Now s~lppose it were true that
general
price
in some
published
movements
the signal processing
to
the point
by a simple.
monetary
problem
aggregate.
to be solved
fixed Then,
by agents
trivial, they could simply observe current monetary aggregates, -) the predicted current and future price movements thev imply,and cor-
rect their behavior
for these
tight relationship
belticcn
relationship These
answer
units changes money
at all between remarks
long and variable depending
diffcrcnt
price
would
suggest
that one should by
aggregates
“Again, see Friedman
explain
some
and changes e
describe
expansion
monetary but closely
and Schwartz (1963).
24
effects
work with
into the system,
with
way is selected.
This
“state” aggregate,
related
and
can occur in a variety
on which
the monetary
periods,
in real variables.
It seems likely that the
is “injected”
depending
unobservable
over short periods
monetary
little is known.
that a monetary
implications
The result would be a very
over even very short
movements
on the way the money
response
determined
observed
these
do not, of course,
lies in the observation
perfectly.
and prices,
lags. On this question
of ways,
being
changes are large
bc
calculate
no
are
on the hypothesis
difficult,
point
extreme
in the short-term
to be solved perfectly
being
correlations
and these
price movements
was too difficult
in money
to “long and variable lags.”
movements
rested
movements
we know
from
cycle.
recoveries
money
business
price movements; arise primarily
to read. Certain
and in particular
view of monetary
among
All these arguments
this weakness
activity,
are too
interesting.
terms,
Paradoxically, economic
connections
cyclical
and
the gold
cycles, since it
of co-movements
Moreover.
on short-term
depressions
in Friedman’s
to general
price movements
difficult
the link between
subject,
but the indirect
the real business
evidence
is much
to business
for the pattern
over the cycle as responses
in the
and prices
to the New World to locate
has no direct connection
of the economy loosely
secularly.
related
as to
XII Let
me
sketched fluctuating
tastes
and studied agents
recapitulate
the co-movements
behaved
We then
in a monetary
individual
comovements summarized wonders
price
movements.
among
aggregate
in section
why it seems
which
theoretical
pass
the
seems
ordering
which posed
rather
to undo
them
in
way. To date, Adelmans
for economic
policy.
By seeking rather
policy are induced increased
variability
and
variability
would
There
explicit
Simons,
then
purpose,
more
of Henry
fluctuations
increase remain
is no prima
18Pmceedin further of Sargent’s $1976).
if
unsystematic
source
generate
of “noise”
a pattern
of
embarrassingly
simple:
which
terms,
to the hope
that this can be done model
has
at the same time, could
My own guess would
be that
business
cycle theory the theory
much of the above is simply’an
attempt
the implicit
the policy
model and
stability
other
of business scope
by the theory.
promises
the smoothest
25
of activist cycles,
one
serving
Insofar
as
no social
to reduce
aggregate,
real
however,
that come
real
monetary
variability
out on this limb, it is likely that such a “successful”
to
of governmental
instability,
is no doubt,
facie case that this residual
critics
on the
monetary
There
underlying
account
be rationalized
by gratuitous
even under
enough
years off.18
limitations
monetary
no equilibrium
and which,
an equilibrium
welfare.
and
time
equilibrium
Friedman,
might
at. I think
economic
easy to guess at even when
Milton severe
one
to arrive at it. Yet
of observed
policy of a successful
state. Indeed,
and make
countercyclical
sizable,
a Revolution
(1959).
are, I think,
here
in advance
emerge
information
the observations
however,
these standards
by the
itself is in a preliminary
aggregative
with
to match
in the behavior
to lend confidence
meets
The implications
accepts
would
what has, in fact, been arrived
is nothing
exist
of the sort outlined
proposals
which
is to
theory
relative prices,
their incomplete
appears
success in this sense is five, but not twenty-five
understand
changing
economy
result
cycle
an economy
an additional
series which
to be necessary
and rigorous
test
this
adding
not to overstate
examples
been developed
on
account
precludes
in an explicit
business
and prices
and utilized
The
the
II. this
be careful
of
by imagining
continually
in quantities
aggregate,
it is fairly clear that there series
implying
superimposed
retrospect,
one must
features
We began
in their own interest
movements
In
main
sections.
and technology,
effectively. to
the
in the preceding
and fiscal policies.
would
be better
dealt
model will be a close descendant
with
by centralized,
governmental
policies
than
by individual,
decentralized
responses.19 In view question
of this
that in a democratic must
lack of novelty
to ask: why
be able
do we need society
in the realm
the theory?
it is not enough
to believe
unemployment
why rate fluctuates
between,
both
situations
are attainable,
and it is clear
three
than at ten. It is also clear that government
which
to explain
of these
natural,
then.
covering
situations than
which
advocating
their
even
when
now,
appeal
adoption? the
is understandable The
subject
of scientifically
should
instability.
should
As long as the business
economic
theory.
both
determine
how business
positions cycles
that
some
then
economics.
and
is most social
forty
evident,
science
years
its
offers
by flexible
arc in possession
them to determine, if many
of these
cycle remains tenable.
and of a
at any time,
who are not profcssion-
beliefs.
This in itself settles in stabilization
to offset
“in apparent There
to become
are inherently
be eliminated
part or actively
appear
of dis-
and
economies
the role of government
its own disruptive
situation,
state of affairs.
enabling
to the latter
at
be more
as that
of Keynesian
economists
be. It is doubtful
the that
are happier
economics
market
that
is
have much to do with
desirable
has had can only
and
knowledge
today,
in which It follows
time. What could
the hope
that
which
as to whether
be to reduce
more
a fair
I think,
to be right; one’
people
of this promise
share
ideas:
responses:
hold,
in the dispute
the
public
related
tested
responses
ally committed little
most policies
of the existing
fluctuations
governmental
what these
to all who literate
two
to violent
forceful body
with
to
emptiness
rationalization
economically
comfortable
that
This was the promise
scientific
oneself
We live in a society
at any particular
will lead
it seems
answer,
say, 3 and 10 percent.
to view the task of aggregative
policies
more than elegant
one is right.
prevails
of policy,
The general
seems
private
contradiction”
policy sector to
to bc no way to
are to be dealt with short of understanding
what
they are and how they occur.
19 That is to say, active countercyclical policy would require the same kind of cost-benefit defense used in evaluating other types of government policies. See Pbelps (1972), and also Prescott’s review (1975).
26
REFERENCES Adelman,
I.. and Adelman,
Goldberger
Model.”
F. L., “The Dynamic
Econometrica,
Properties
of the Klein-
27, No. 4, (October
1959),
596-
625. Azariadis,
C.,
Journal
“Implicit
Contracts
of Political
and
Economy,
Underemployment
83, No.
Equilibria,”
6, (December
1975),
1183.
1302. Barro,
R. J., “Rational
Journal Burns,
of
Friedman,
7.
8.
1, (March
1968),
Friedman,
M., and
Schwartz,
National
Bureau of Economic
A. D.. Rostow, of
the
Press.
1953.
Presidential Economic
A. J. A Monetary Princeton Research,
New York:
Address Review,
to the 58, No.
Economv.
New
National
of the United Press
A. J. The Growth
1790-1850.
G. S. The Allocation York:
History
University
for
the
1963.
W. W., and Schwartz,
British
Cvcle.
Cycles.
l-32.
1946.
American
Princeton:
G. R.. and Becker, Life
Business
Policy,”
1976),
1-17.
1867-l 960.
ation
of Monetary
1, (January
Policy,”
Association,
States.
the
Research,
M., “The Role of Monetary Economic
Ghez,
2, No.
W. C. Measuring
Bureau of Economic
American
Gayer,
and the Role
Economics,
A. F., and Mitchell,
National
6.
Expectations
Monetary
Oxford:
and Fluctu-
The
Clarendon
of Time and Goods
Bureau
of Economic
Over
Research,
1975. 9.
Grossman,
H. I.,
Journal
“Aggregate
of Political
Demand,
Economy,
Job
Search,
and
Employment,”
81, No. 6, (November/December
1973)
1353-1369. 10.
Haberler,
Il.
Hayek,
G. Prosperity F.
Jonathan
A. von. Cape,
and Depression. Monetary
Theory
1933.
27
Geneva: and
the
League Trade
of Nations, Cvcle.
1936.
London:
12.
Hicks,
J. R.. “Mr. Keynes
Econometrica. 13.
Hirsch,
A.
Reserve Keynes,
System
Klein.
of
1929-52.
Board
Theory
OBE
Econometric
of
Council,
Model.”
Conference.
Governors
(ed.
of the
0.
Federal
1972.
of Employment,
Interest
and Money.
1936. A. S. An Econometric
Amsterdam:
F. H. Risk,
the
Determination
and Social Science
L. A., and Goldberger,
Knight,
with
Price
D. C.:
J. M. The General
States, 16.
Simulations
Macmillan,
Interpretation.”
147-159.
Washington,
London: 15.
“Price
Econometrics
Eckstein),
14.
5, (1937),
A.,
in The
and the ‘Classics’: A Suggested
North
L’ncertainty
Holland,
and Profit.
Model of the United
1955. Boston:
Houghton
Mifflin,
1921. 17.
Lucas,
R. E, Jr.,
,Economic
“Expectations
Theoy,
18.
, “An Equilibrium of Political
19.
Economy,
Curve and Labor
Carnegie-Rochester North Lucas,
Holland,
R. E., Jr..
Intlation,” tion 21.
de Menil,
22.
Conference
and
Rapping,
(eds.
of the Business 1975).
Journal
of
Evaluations:
Cycle,”
Journal
11 13-l 144. A Critique,”
(eds. K. Brunner
Series on Public
G., and Enzler,
ference
, (ed. 0. Eckstein),
in &
and A. H. Meltzer),
Policy,
1, Amsterdam:
Wages,
1970.
and Wages in the FR-MIT
Econo-
of
New
Price
During Business
28
Determination
D. C.: Board of Governors
and Social Science
195 1.
York:
and
and Infla-
Norton,
Washington,
Resarch,
Employment,
of Employment
--et al.),
Econometrics
System
W. C. What Happens
Bureau of Economic
“Real
Foundations
J. J., “Price
in The
Reserve
L. A.,
E. S. Phelps.
Model,”
Mitchell,
Policy
of Money,”
103-l 23.
1976, 19-46.
metric
the Federal
Model
Markets,
in Microeconomic
Theory.
1972).
83, No. 6, (December
, “Econometric Phillips
20.
and the Neutrality
4, No. 2, (April
Council,
Conof
1972.
Cycles. New York: National
23.
Modigliani,
F.,
Money,” 24.
“Liquidity
Econometrica,
Moggridge,
Muth,
J.,
Phelps,
the
12, No. 1, (January
Theory
of Interest
“Rational
1944). 45-88.
of
1973.
Expectations 29,No.
“Money
ES.,
Journal
and
Vol.
Macmillan,
Econometrica, 26.
and
D. (ed.) T>
XIV. London: 25.
Preference
and the Theory
3, (July
Wage Dynamics
Political
of Price Movements,”
1961), 315-335.
Economy,
and
76,
Labor
No.
4,
Market
Equilibrium,”
II, (July/August
1968)
687-7 11. 27.
. Inflation Benefit
28.
Phelps,
Policy
Approach
and
to Monetary
Unemployment Planning.
Theory:
London:
E. S., -et al. Microeconomic Foundations Theory. New York: Norton, 1970.
The
Macmillan,
Cost1972.
of Employment
and
Inflation 29.
Prescott,
E. C.,
Economy,
30.
Sargent,
“Efficiency
of the
83, ?:o. 6, (December
T.
J.,
States,”
“A Classical
Journal
Natural 1975),
Rate,”
Macroeconometric
of Political
Journal
of Political
1229-l 236.
Economy,
Model
for
the
84, No. 2, (April
United
1976)
207-
Without
Pre-
237. 31.
Sargent,
T. J.. and
tending
to Have
of Michigan 32.
Sargent,
Too
working
Much
Instrument,
of Political Tinbergen, Geneva:
Economy,
J. Business
Cycle
A Priori
Economic
Modeling Theory,”
University
paper, March 1976.
T. J., and Wallace,
Monetary
33.
Sims, C. A., “Business
N.,
” ‘Rational’
and the Optimal 83, No. 2, (April Cycles
League of Nations,
29
Supply
the Optimal Rule,”
Journal
1975), 241-254.
in the United 1939.
Expectations,
Money
States
of America.
1919-32.
