World Inflation and Monetary Accommodation in Eight Countries

69 Pages·2007·5.85 MB·English

ROBERT J. GORDON
Northwestern University
World Inlation and
Monetary Accommodation
in Eight Countries
The "monetarist"v iew of ... the accelerationo f inflations ince 1965 [is] that it
has been the ultimatec onsequenceo f an increasei n the rate of world monetary
expansion, an increase attributablep rimarilyt o the excessively expansionary
monetary policy pursued by the United States in recent years.-Harry G.
Johnson'
I find the alternativee xplanation. .. which regardst he basic cause as increased
trade-unionm ilitancy. .. morep lausible.-Nicholas Kaldor2
There is little doubt, then, that the wage-inflationi dea does not apply to the
inflationarye xperiencesw e have seen in the real world.... Only when the mac-
roeconomic implicationso f the wage-inflationh ypothesis are traced through
and confronted with the facts does it become apparentt hat it cannot be sus-
tained.-Arnold C. Harberger3
Note: This research has been supported by the National Science Foundation. I
am gratefult o my colleague John Bilson for aiding me in the acquisitiono f data from
the InternationalM onetary Fund, and to my research assistant James Glassman for
an absolutely outstanding job. Also helpful were the suggestions of Victor Argy,
Jacques Artus, Jacob Frenkel, Hans Genberg, John Helliwell, Paul Krugman,D avid
Laidler, Michael Parkin, Richard Sweeney, George Zis, and members of the Brook-
ings panel. Of special value were a number of discussions with ChristopherS ims.
1. Harry G. Johnson, "Inflation: A 'Monetarist' View," in Harry G. Johnson,
Further Essays in Monetary Economics (Allen & Unwin, 1972), p. 335.
2. Nicholas Kaldor, "Inflationa nd Recession in the World Economy," Economic
Journal,v ol. 86 (December 1976), p. 710.
3. Arnold C. Harberger," Inflation,"i n John Van Doren, ed., Symposium on the
Emerging World Economy (Great Ideas Today series) (Encyclopaedia Britannica
Educational Corporation, 1976), pp. 94-106.
409
410 Brookings Papers on Economic Activity, 2:1977
WHILE" SUPPLYS HOCKS"in oil and food are widely agreed to have ag-
gravatedi nflationi n many countriesi n 1973-75, no consensush as yet
emergedt o explain the accelerationo f inflationb etween the mid-1960s
and 1970-72 in the majori ndustrializedn ationso utsideo f North Amer-
ica. Instead,t wo majors chools of thought,t he "internationaml onetarist"
and the "wage push," have developed alternativee xplanationsh aving
radically differentp olicy implications. The first group views inflation
within a conventionalm acroeconomicf rameworka s an "international
monetaryp henomenon"a nd identifiesi ts fundamentacl ause as an excess
demandf or commoditiesg eneratedb y governmenta ctions.A ny attempt
to bringi n other factors, particularlyt hose of the wage-pushv ariety,i s
dismissedA. dherentso f the wage-pusho r "sociological"s chool of thought
disagree, pointing to the allegedly spontaneousw age explosions that
occurredi n a numbero f Europeanc ountriesi n 1968 and 1970 as evi-
dence of the specialn oneconomicc haractero f the recenti nflation.W age
claims are viewed as part of a continuingc onflict over income shares
among competings ocial groups, and the events of the 1968-70 period
reflectedla bor's" long-smoulderinrge sentmenta nd dissatisfaction."4
Aims of the Paper
The sourceso f inflationi n the maini ndustrializedn ationsi n the period
of fixed exchanger ates have not yet been adequatelye stablished.B oth
the international-monetariasnt d wage-pusha rgumentsa re unsatisfactory
and incomplete. The accelerationo f wages and pricesi n 1969-70 in the
majorn ationso utsideo f the UnitedS tatesf olloweda continuingd ecelera-
tion of their rates of monetaryg rowthb etween 1963 and 1969. For the
"world"-the United States plus seven other importanti ndustrialn a-
4. Kaldor, "Inflation and Recession," p. 710. In his more eclectic analysis of
wages, George L. Perry considers wage push, closely associated with a dissatisfaction
with shares, an important part of the wage explosions of the 1968-70 period, and
consequently one of the factors that brought the average rate of wage inflation in
the 1970s to a new, higher, plateau. See his "Determinantso f Wage Inflation around
the World," BPEA, 2:1975, pp. 403-35. Earlier, William D. Nordhaus had rejected
a related explanation labeled "frustrationt heories" in "The Worldwide Wage Ex-
plosion,"B PEA, 2:1972, pp. 431-64.
RobertJ . Gordon 411
tions5-the 1969-70 accelerationo f wage rates preceded by almost a
year the 1971 explosiono f money and internationarl eserves,i n apparent
conflict with the international-monetarisetx planation.W age-pushp ro-
ponentsh ave not made their case either, because autonomousw age in-
creasesd o not automaticallyp ass into the pricel evel.
This papera cceptsf rom the outset the propositiont hat world inflation
must in the long run be a world monetaryp henomenon.C onfirmationo f
a connectionb etweent he rates of growtho f world money and prices, as
in severalr ecents tudies,i s vieweda s only the firsts tep in the development
of a fuller understandingo f the inflationp rocess, because the sourceso f
changes in world money are left unexplained.6A correlationb etween
worldp ricesa nd money does not rule out wage push as a sourceo f world
monetaryg rowth.I nstead of viewing internationalm onetarya nd wage-
push factorsa s competitivee xplanationso f inflation,a more comprehen-
sive theoretical frameworki s presented which incorporatest hese two
ingredientsa s possible explanationso f the behavior of the monetary
authorities,t ogether with fiscal deficits, supply shocks, and a counter-
cyclicalm onetaryr eactionf unction.
Since a time-seriesc orrelationb etweeni nflationa nd monetaryg rowth
is consistentw ith several sources of monetaryg rowth, a more discrimi-
nating empirical methodology is required. An impirical study of the
United Statesa nd seven other majori ndustrialn ationsa ttemptst o deter-
mine whethert he data are consistentw ith an effect of any or all of the
possible sourceso f monetarya ccommodation.
Among the majorq uestionst o be answereda re the following: If U.S.
monetaryg rowthi s to be blamedf or the accelerationo f inflationi n other
countries,b y what channelsw as this impulset ransmitted?D id patternso f
causationd iffera mong countries?W ere some countriesb etter able than
otherst o pursuec ountercyclicarl athert hanp rocyclicalm onetaryp olicies?
Did differencesa crossc ountriesi n the degreeo f accommodationto supply
shocks in 1974 correspondw ith the degree of accommodationt o wage
5. The "Other Seven" countries are Canada, France, West Germany, Italy,
Japan, Sweden, and the United Kingdom.
6. See, for instance, David I. Meiselman, "WorldwideI nflation: A Monetarist
View," in Meiselman and Arthur B. Laffer, eds., The Phenomenon of Worldwide
Inflation (American EnterpriseI nstitute, 1975), pp. 69-112. See also Hans Genberg
and Alexander K. Swoboda, "Causes and Origins of the Current Worldwide Infla-
tion," discussion paper (Geneva: Graduate Institute of International Studies,
November 1975; processed).
412 Brookings Papers on Economic Activity, 2:1977
push or reservei nflowsi n the earlierp eriod?D o episodeso f wage push
identifiedb y othersa ppeart o have been genuinelye xogenouso r were they
precededb y episodes of monetarya ccommodation?C an the breakdown
of the BrettonW oods systemb e treatedu ltimatelya s anotherc asualtyo f
the VietnamW ar,o r did domestice ventsi n otherc ountriesp lay a role?
Answerst o these questionsa re useful not only for an understandingo f
history, but for future policymaking:I s control of the money supply
eithern ecessaryo r sufficientf or a nationt o controli ts inflationr ate?I s it
likely that monetarya uthoritiesw ill be able in the futuret o insulatet he
inflationr ate from the tendenciest o wage push, or are incomes policies
requireda s a cure?
Finally,t he paperr aisesa methodologicalq uestiono f interestt o many
economistsc aught in the middle without any particulara llegiancet o an
international-monetariosrt wage-pushv iew of the world: Can economet-
ric techniquesu ncovers ystematict endenciest owardp rocyclicala ccom-
modationo r countercyclicaal ctivismo n the part of centralb anks, or do
shiftingt argetsa nd prioritiesd efy statisticalg eneralizationsa nd call in-
stead for a mored escriptivea nd anecdotala pproacht o monetaryh istory?
International-MonetariAstp proach
The international-monetarisatp proachb egins from the proposition
that underf ixed exchanger atest he world inflationr ate is determinedp ri-
marily by previous changes in the rate of growth of the world money
supply. This follows from two basic propositionsw hich previouslyh ad
been emphasizedb y U.S. monetaristsa ddressingis sues of a closed econ-
omy: the stabilityo f the demand-for-monefyu nction,a nd the lack of any
long-run tradeoff between inflation and unemployment.B oth of these
elements have been tested by international-monetarisetc onomistsw ho
have estimated structural equations describing the behavior of the
demand-for-moneyfu nction and the expectationalP hillipsc urve for the
"world"( usually the Group of Ten).7 Further,r educed-formt ests have
7. See M. R. Gray, R. Ward, and G. Zis, "The World Demand for Money Func-
tion: Some PreliminaryR esults,"a nd Nigel Duck and others, "The Determinationo f
the Rate of Change of Wages and Prices in the Fixed Exchange Rate World Econ-
omy, 1956-71," both in Michael Parkin and George Zis, eds., Inflation in the World
Economy (Manchester University Press and University of Toronto Press, 1976), pp.
151-78 and 113-43, respectively.
RobertJ . Gordon 413
found that changesi n the world rate of monetaryg rowthh ave preceded
changesi n the world inflationr ate.8
The behavioro f individualc ountriesi s characterizedn ot only by stable
national demand-for-moneyf unctions and vertical long-run Phillips
curves, but by two additionalf eatures. First, capital is mobile among
nations,s o that internationarl eservest end to flow in or out as necessary
to set a nation's money supply equal to its demand for money, which
dependsp rimarilyo n its price level, real output, and interestr ate.9S ec-
ond, commoditya rbitragem aintainst he tradable-goodsp ortion of the
domesticp ricel evel fairlyc lose to the worldp ricel evel of tradableg oods,
while labor mobility communicatesc hangesi n prices of tradableg oods
to the nontradabless ector.10T hus any event that raises the foreignp rice
level tends to push up both the domestic price level and the domestic
money supply, irrespectiveo f the reaction of the domestic monetary
authorities.
The international-monetariaspt proacht racest he accelerationo f infla-
tion outsidet he United Statesi n the late 1960s back to U.S. fiscal deficits
incurredt o pay for Vietname xpendituresw, hich induceda n acceleration
in the growthr ate of the U.S. money supplya nd pricel evel. Inflationt hen
spreadf rom the United Statest o otherc ountriesb y four main channelso f
transmission."F irst, the "directp rice influence"w orkingt hroughc om-
modity arbitrager aised the prices of tradableg oods everywhere.T his
then pushedu p the marginalv alue producto f labor and of other factors
of productiona nd hence domesticc osts, raisingt he priceso f nontradable
8. See Genberga nd Swoboda, "Causesa nd Origins."
9. See Robert A. Mundell, International Economics (Macmillan, 1968), chap.
18; also Harry G. Johnson, "The Monetary Approach to Balance-of-Payments
Theory,"i n Johnson, FurtherE ssays in MonetaryE conomics, pp. 229-49.
10. Rudiger Dornbusch, "Devaluation, Money, and Nontraded Goods," Ameri-
can Economic Review, vol. 63 (December 1973), pp. 871-80. The Dornbusch model
is extended to the case of imperfectly flexible prices in Robert J. Gordon, "Interrela-
tions Between Domestic and International Theories of Inflation," in Robert Z.
Aliber, ed., The Political Economy of Monetary Reform (Macmillan, 1977), pp.
126-54.
11. An excellent general discussion of the channels through which inflation is
transmittedc an be found in Walter S. Salant, "InternationalT ransmission of Infla-
tion," in LawrenceB . Krause and Walter S. Salant, eds., WorldwideI nflation: Theory
and Recent Experience (Brookings Institution, 1977), pp. 167-227. See also Alex-
ander K. Swoboda, "Monetary Approaches to Worldwide Inflation," in the same
volume, especially pp. 3 3-44.
414 Brookings Papers on Economic Activity, 2:1977
goods.'2S econd,t he trades urpluseso f other nationsi nducedb y the fact
that U.S. income growthw as higher than their own, as well as by de-
terioratingU .S. price competitivenessb, oosted foreignl evels of produc-
tion and income throught he conventional" Keynesiand emand-pressure
mechanism."T hird,t he "BrettonW oods monetizationc hannel"a llowed
U.S. balance-of-paymentds eficitst o be paid for by the creationo f U.S.
dollar liabilitiesw hich expandedt he monetaryb ase of many nationsa nd
furtherf ueled their own domestici nflationr ates. Fourth, an acceleration
of U.S. inflationc ould have raised domestice xpectationso f inflationd i-
rectly, leading to higher wage and price increases.T he channels were
connected, since the direct effects of higher prices for tradableg oods,
higherr eal output,a nd highere xpectedp ricesf or nontradableg oods, all
raised a nation'st ransactiond emandf or money and, with freely mobile
capital,a ttractedt he internationarl eservesn eeded to bringt he domestic
moneys upplyi nto equalityw ithh igherm oneyd emand.
In the long runu nderf ixed exchanger ates,t hen,p olicymakersin small
countries should regard inflation as part of the external environment
rathert han under their own control. Domestic "wage push" by unions
cannot contributet o domestic inflation;w orker groups that achieve a
highern ominalw age when the price level is determineda broadc an raise
theiro wn incomeso nly at the expenseo f unemploymenat nd lowerp rofits,
particularlyi n the tradable-goodss ector. Incomes policies designed to
controlw age push may be able to influencet he unemploymenrt ate or the
distributiono f income,b ut not the inflationr ate.
But the internationaml onetarist'sd enial of a role for an autonomous
wage push in the worldi nflationp rocessi s not totallyc onvincing,b ecause
the symmetryb etween tradable-goodsp rices and domestic wages is ig-
nored. Both are determinantso f domesticp rices and the transactiond e-
mand for money. An autonomousw age push could raise the demand
for money and, with perfectc apitalm obility,s uck in the reservesn eeded
to provide the base for a higher domesticm oney supply. If commodity
arbitragei s perfect, the profitso f firmsi n the tradable-goodss ector will
be squeezed,b ut there is some evidencet hat commoditya rbitragei s not
12. This is the primary channel of international transmission in the "Scandi-
navian" model of inflation. See Gbsta Edgren, Karl-Olof Faxen, and Clas-Erik
Odhner, "Wages, Growth and the Distribution of Income," Swedish Journal of
Economics, vol. 71 (September 1969), pp. 133-60. Also Odd Aukrust, "Inflationi n
the Open Economy: A Norwegian Model," in Krause and Salant, eds., Worldwide
Inflation,p p. 107-53.
RobertJ . Gordon 415
perfect, at least in the short run.'3 The increase in domestic monetary
growth would contribute to an acceleration in world monetary growth
and in the world rate of inflation, particularly if an autonomous wage
push were to occur in several countries simultaneously.
The Wage-Push, or "Sociological," Explanation
In its most extreme form, proponents of the wage-push view argue
that the inflation rate depends entirely on the aggressiveness of labor
unions in pressing wage demands. Peter Wiles has claimed, for instance,
that "we have moved from wage claims based on the actual situation in
the trade ... to claims picked out of the air...."14 The underlying source
of wage push is viewed variously as a conflict over the fairness of the
income distribution and wage structure; as the result of the rise of the
tactics of the New Left and the decline of authority; and as a consequence
of a communications revolution that increased awareness of foreign wage
claims. The independent influence of money and aggregate demand is
often viewed as negligible.'5
While money is sometimes rejected as a cause of wage behavior,
changes in money are viewed as a consequence of wage push in many dis-
cussions. Sooner or later the central bank will have to raise the money
supply in order to accommodate the higher transaction demand for money
created by higher wages. Richard Cooper has made this point succinctly:
The wage level in the modern economy is indeterminantb ecause in the
final analysist he monetarya uthoritiesm ust-for political reasons-provide a
money supply adequatet o ratify any given level of money wages, no matter
how it was reached,i n ordert o avoid excessiveu nemployment.'6
13. See Irving B. Kravis and Robert E. Lipsey, "Export Prices and the Trans-
mission of Inflation," American Economic Review, vol. 67 (February 1977), pp.
155-63. Also see Rudiger Dornbusch and Paul Krugman, "Flexible Exchange Rates
in the Short Run,"B PEA, 3:1976, especially pp. 559-68.
14. Peter Wiles, "Cost Inflation and the State of Economic Theory," Economic
Journal,v ol. 83 (June 1973), p. 378.
15. A clear statement emphasizing the fairness issue is contained in John Hicks,
The Crisis in KeynesianE conomics (Basic Books, 1974), chap. 3.
16. Richard N. Cooper, statement in "Commentaries"o n paper by Thomas D.
Willet, "The EurocurrencyM arket, Exchange-RateS ystems, and National Financial
Policies," in Carl H. Stem, John H. Makin, and Dennis E. Logue, eds., Eurocur-
rencies and the International Monetary System (American Enterprise Institute,
1976), p. 252.
416 Brookings Papers on Economic Activity, 2:1977
Some British commentatorsa dmit the monetaryc onnection.'7B ut this
acknowledgmenits by no meansu niversal.'8
Severali nvestigatorsh ave found that wage equationsf or majorE uro-
pean countriesr equiret he use of dummyv ariablest o explainp articular
episodes of sudden accelerationi n the rate of wage change.'"W hile a
significantp ositive coefficiento n a dummyv ariable indicates only that
somethingi s happeningt hat cannot be otherwisee xplained,t he timing
of the wage accelerationsc orrespondt o widely recognizedi ncidentso f
aggressivel abor behavior,p articularlyt he Frenchg enerals trikeo f May
1968, andt he Italian" hota utumn"o f 1969.
The wage-pushe xplanationh as been universallyc ondemnedb y pro-
ponents of the international-monetarivstie w, but empiricalc ritiquesb y
internationaml onetaristsh ave involved tests of unnecessarilyr estrictive
versions of the sociological approach.F or instance, any influence on
wages of excess commodityo r labor demando r proxies for inflationary
expectationsi s cited as negative evidence, implicitlyr uling out a more
eclectic frameworki n which both the aggressivenesso f workers and
conventionale conomic variablesm ight be influential.F urther,t ests by
internationalm onetaristsh ave measuredw age-pushe ffects by including
variablesr epresentingti mel ost in strikesr athert han the dummyv ariables
cited earlier.T o the extentt hatw orkersa chievet heirw age aimsb y threats
of strikes that are not actually carriedo ut, wage push could exist but
neverthelessb e uncorrelatedw ith strike variables.20U nfortunately,t his
17. Wiles, "Cost Inflation,"p . 385; Stephen Marris, "Panel Discussion: World
Inflation,"i n Claassen and Salin, eds., StabilizationP olicies in InterdependentE con-
omies, p. 303.
18. Aubrey Jones, for instance, argues that "a tightening of the supply of
money is not, therefore, .. . a solution to the problem of rising prices," in The New
Inflation:T he Politics oj Prices and Incomes (London: Andre Deutsch, and Penguin,
1973),p. 39.
19. See in particular George L. Perry, "Determinantso f Wage Inflation,"e spe-
cially table 4, p. 424, in which significant coefficients are found for a dummy for
1968 in France and for 1970 in Italy, West Germany, Sweden, and the United King-
dom (a 1968 dummy for Japan is only marginally significant). A similar approach
was followed by Erich Spitaller, who found significantd ummy coefficientsi n various
periods for France, West Germany, and the United Kingdom; see "Semi-Annual
Wage Equations for the ManufacturingS ectors in Six Major Industrial Countries,"
Review of WorldE conomics, no. 2 (1976), pp. 300-37.
20. The most comprehensive tests of strike variables are presented in David
Laidler, "Inflation-Alternative Explanations and Policies: Tests on Data Drawn
from Six Countries," in Karl Brunner and Allan H. Meltzer, Institutions, Policies
and Economic Performance (Amsterdam: North-Holland, 1976), pp. 251-306.
RobertJ . Gordon 417
problemm akes it difficultt o link puzzlinge pisodes identifiedb y signifi-
cant coefficientso n dummyv ariablest o any quantifiablep roxy for labor
militancy.
The Demandf or and Supplyo f MonetaryA ccommodation
Internationalm onetary effects and wage push are only two of the
possible sources of inflation. A useful frameworkf or analysis of the
sources of inflationi s to distinguishf actors that create pressureo n the
centralb ank to "accommodate"-that is, to react by raisingt he money
supply-and factorst hat help to explainw hy the centralb ankr eactsa s it
does to these pressures.A "demandf or monetaly accommodation"is
createdb y domesticd emands hifts, domesticc ost push, and demanda nd
supply shocks from abroad.T he "supplyo f monetarya ccommodation"
by the centralb ank dependso n the weightsi n its countercyclicarl eaction
function, its own degree of independencef rom the government,a nd the
extento f the government'sa ttemptst o influencei ts chancesf or reelection
by manipulatingth e economy.21
THE DEMAND FOR MONETARY ACCOMMODATION
When the real money supplyi s held constant,a n increasei n domestic
expenditurews ill tend to alteri nterestr ates,w hetheri ts sourcei s a change
in consumer attitudes,b usiness expectations,g overnments pending, or
tax rates. An increasei n spendings hifts the "IS curve"o f intermediate
macroeconomicth eoryt o the rightu p a fixed" LMc urve,"a nd the central
banki s forcedt o raiset he moneys upplyi f it desirest o offsetp arto r all of
the increasei n interestr atest hatw ould otherwiseo ccur.
Central banks may be under pressure not only from autonomous
shiftsi n demand,b ut also from autonomousi ncreasesi n wages and prices
negotiatedi n the privates ector, and from externals upplys hocks.A t first
glancet he motivationf or wage push may seem elusive,a t least in a closed
Laidler has pointed out to me that strike variables were first used by British cost-
push proponents,n ot by internationalm onetaristsw ho were seeking to test the cost-
push hypothesis.
21. A more formal and extended presentationo f this framework is developed in
Robert J. Gordon, "The Demand for and Supply of Inflation,"J ournal of Law and
Economics, vol. 18 (December 1975), pp. 807-36.
418 Brookings Papers on Economic Activity, 2:1977
economy, since prices are likely to be markedu p over any autonomous
wage increasesa chievedb y workers,l eaving real wages unchanged.B ut
workersw ho "push"m ay achievea n increasei n theirr eal incomer elative
to holderso f assetsw ithr eturnsf ixedi n nominalt erms,w orkersw ho have
less marketp ower, and profitso f firmsi n the tradable-goodss ector.S up-
ply shocks-for example, crop failures or the formationo f commodity
cartels-are like any other type of cost push in creatingp roblemsf or the
monetarya uthoritiesb, ecausea failuret o accommodatew ill createu nem-
ployment, while accommodationm ay unleash an inflationarys piral if
workersa ttemptt o maintaint heiro riginalr ealw agel evels intact.
Internationacl apitalm obilitym ay emboldenw orkersa nd othersc on-
sideringa n autonomousp ush, since theiro wn actionsc an directlyi nduce
a rise in the money supply by pulling in internationalr eserves.I n this
case the push is automaticallya ccommodatedO. nly when capitali s less
than perfectlym obile can the central bank attempt to sterilize part of
the reservei nflowb y open-markest ales of its domestica ssets,a restrictive
rediscountp olicy, or an increase in the reserve requirementso f com-
mercialb anks.
THE SUPPLY OF MONETARY ACCOMMODATION
AND THE MONETARY REACTION FUNCTION
The mere existenceo f pressureo n the centralb ank does not implyt hat
it will act either to accommodateo r to resist. A central bank following
Friedman'sr ule of a constantg rowthr ate for the money supply would
ignore such pressurese ntirely.M ore likely, a central bank will attempt
to vary the growtho f the money supply,o r some other monetaryi nstru-
ment over which it believes it has more direct control, with the aim of
maximizinga social-welfaref unction. Higher unemploymenta nd infla-
tion, andl owerf oreign-exchangree serves,a re all evils thatm ayb e resisted
by countervailingsh iftsi n monetaryp olicy.
Two sets of conflicts constraint he reaction of policymakers.F irst,
even an idealistica ttemptt o maximizes ociety'sw elfarec ollides with the
incompatibilityo f short-run improvementsi n unemploymenta nd in
inflation,o r in unemploymenat nd in the foreignb alance.W agep ush and
supply shocks allow no easy solutions, for a centralb ank must choose
whethert o resistt heir stimulust o inflationb y contractingo r to resistu n-
employmenbt y expanding.