journal of philippinedevelopment 4 ii u xlx n s ibriib · 2003. 4. 25. · journal...

Journal of Philippine Development 4___II_S Number35, VolumeXlX, No.2 SecondSemester1992 IBRIIB

IMF STABILIZATION PROGRAM

AND ECONOMIC GROWTH:

THE CASE OF THE PHILIPPINES

Joven Zamoras Balbosa"

INTRODUCTION

Relevance of the Study

In 1970, the countrywas in a mess, The standard of living wasdecliningrapidly and an insurgencyproblem in the countrysidehadgained strength and momentum. Moreover, the 1969 reelectioncampaign of Mr. Marcos, which was financed by massive moneycreation, precipitateda balance-of-paymentscrisis. In response tothe crisis, the government sought an InternationalMonetary Fund(IMF)-sponsored adjustmentprogram which called for a 43 percentdevaluation and a reduction in selected tariff rateS. Despite themeasures taken, the country remained in a state of political andeconomicchaos.This promptedthe regime to declareMartial Law in1972 to institutepolitical discipline with the promise of economicprosperity for all. Unfortunately, the promised prosperity neverhappened.

In 1983, the countrysuffered anotherbalance-of-paymentscrisis,this time more severe. Due to both political and economicuncertaintiesprecipitated by the assassinationof Senator BenignoAquino(a politicalrivalof Mr. Marcos),capitalflightincreased,foreigncredit availability decreased, and foreign exchange reserves fellsharply. Again, an IMF-sponsored adjustment program was put inplace. This time, the political and economic implications cost theregime its downfallin 1986.

And yet again in 1989, an impendingbalance-of-payments crisisbegan to unravel. The new governmentsubmittedin March 1989 a

'The authorisAssistantDirectorat theCongressionalPlanningand BudgetOffice,Houseof Representatives.The valuableassistancerendered by Ms, JuandaDimayuga in therevisionof thispaperis acknowledged,

!I

30 JOURNAL OF PHILIPPINE DEVELOPMENT

Memorandum of Economic Policy to the IMF stating its compliancewith IMF conditionalities in exchange for financial assistance.Already, the government had contended with six attempted coupd'etats and with a growing clamor from the people to reform aworsening economy. Was it then to suffer the same fate as itspredecessor? What, in the short-run period of a Fund-sponsoredprogram, needed to be learned in order to avoid or at least minimizethe social costs of adjustments?

A better knowledge of the immediate/short-run impact ofadjustment programs is an important ingredient in coming up withpolicies to ensure the political survival of the government. Theobjectives of this paper, therefore, are the following:

a) To analyze the short-run impact of adjustment programs in.order to generate policies which are consistent with a morestable adjustment process.

b) To test simple models showing the supply-side effects of IMFadjustment program policies.

Limitations of the Study

Due to data constraints, the study focuses only on the 1983balance-of-payments crisis and the subsequent IMF adjustmentprogram carried out in responseito the crisis. Consequently,annualdata from 1960 to 1986 are utilized to highlight the impact ofadjustmentpolicieson key macroeconomicvariables.

Organization of the Study

Section II begins with an outline of typical macroeconomic policiesthat are commonly utilized in Fund-sponsored-programs. Adiscussion of the policy implications on key macroeconomicvariables is then presented with emphasis on the arguments of themonetarist approach to the balance of payments and of thestructuralist models. A review of relevant literature ends the chapter.The theoretical framework and empirical results are presented insection II1.The supply-side effects of the IMF adjustment programare highlighted in this chapter. In section IV, we evaluate the policyinstruments used in the 1984 and 1989 adjustment programs for thePhilippines. And finally, lessons are drawn to guide us in theimplementation of the 1989 IMF-supported adjustment program.

BALBOSA:IMF STABILIZATIONPROGRAM 31

REVIEW OF LITERATURE

This section presents typical Fund adjustmentpolicies and theirimpacton the economy.The monetaristandstructuralistapproachesare discussed to show the contendinghypotheseson the effects ofFund policieson keymacroeconomicvariables.A review of relevantempiricalstudiesis alsopresented.

Fund-Type Stabilization Policies

AccOrdingto Tanzi (1987), the objectivesof a Fund-supportedstabilizationprogram includea balance of payments that is viableover the medium run, the promotion of growth under a stableeconomic environment, price stability, and the prevention ofexcessive growth in external debt. Now,-let us suppose a countrywere facing balance-of-payments difficulties,accelerating inflation,and a blatant mismanagement of its key sectors. If we were todeduce, the standardset of rem_ediesthat the IMF may offer,we maycome up with two macroeconomicpolicieswhichare almost alwaysrecommendedamongthe array of policychoices.These policiesare:

a) monetarycontraction,usuallydue tothe impositionofseparateceilingson creditfrom the bankingsystemto theprivate andgovernmentsectors;and

b) devaluation.

How are these policieslikelyto work?We limitour discussionto twoapproaches(the monetaristapproachto the'oalanceofp_ymentsandthe structuralistapproach) in order to shed lighton the workingsofthe above policieson the economy.

The Structuralist Approach and the Monetarist Approach:

On Monetary Contraction. The simple form of the monetaristapproach emphasizes the direct influence of excess demand andsupply of money on the demand for real resources and on the balance

of payments. Under the assumption of a stable demand for moneyfunction, an expansion of money stock in excess of what residents arewilling to hold creates a disequilibrium in the money market. Theexcess supply of money leads to greater demand for real resources.Under the assumption of full-employment output, the greater demand


for real resources can only lead to an upward pressure on prices;hence, continuous monetary expansion is inflationary. However, theopportunity to trade (i.e., to import) diffuses the inflationary pressuresbrought about by excess demand. Increased demand for importsbrought about by excess demand in the economy results in thedrawing down of foreign exchange. Under a fixed exchange ratesystem, this results in the depletion of international reserves and,consequently, balance-of-payments deficits. The depletion ofreserves leads to a contraction of money stock, and finally to arestoration of money market equilibrium. Note that a continuousexpansion of money stock through domestic credit expansion willresult in a worsening of the external imbalance for as long as thereare foreign exchange reserves to run down; otherwise, foreignborrowing will have to be resorted to. Thus, to counteract thisconstant pressure on the external account and the inflationary trendin the economy, restrictive monetary and fiscal policies must beimplemented. These policies are meant to influence the rate of growthof domestic demand and absorption. The basic objective is to restorethe balance between absorption and aggregate supply to achieve aviable balance-of-payments position within a reasonable period oftime.

The structuralists (see Taylor 1983, van Wijnbergen 1982, Lim1987, among others), on the other hand, point to the slow, inadequateor delayed decline of prices in countries that have followed themonetarist approach as proof that the approach, in its simplistic viewof the economy, has ignored important supply-side effects that mayhave prevented the desired monetarist cure from taking effect. Thestructuralists emphasize the transmission channel between monetaryinstruments and the supply side.of the economy via credit financingof working capital needs. Working capital is needed to finance thestocks of raw materials, semifinished goods, intermediate importedinputs, advance payments to workers, among others. Typically,working capital needs are financed by bank or curb-market credit.This in turn implies that the cost of credit (the real interest rate) is acomponent of input costs. Tight money drives up interest rates onloans to firms for working capital and investment and thus increasesinput cost. The normal business response would be to cut back onactivity and attempt to pass increased costs through to higher prices.Even if aggregate demand fell under monetary constraint, aggregatesupplymay fall even more, so that a stagflationary phenomenon (i.e.,recession with inflation) results.


The implicationis that monetary restraintmay be stagflationary inthe short run, increasing prices and causing output contraction ifinterest rate cost-push is strong enough. Thus, the first monetaristargument that "less money means lower prices" is contrary to thestructuralist view, at least in the short run. However, the secondmonetarist argument that "less money improves the foreign balance"may be cgmpatible with the structuralist hypothesis in the sense thatoutput contraction reduces the intermediate imported input demandof firms.

On Devaluation m The Real Effects. The monetary approachto the balance of payments sees devaluation as a useful instrumentin improving the external payments position of the economy.Underlying this argument is the assumption that the country is a smallopen economy. Consequently, the domestic price level is significantlyaffected by the country's exchange rate as well as by world prices(Rivera-Batiz 1985). A devaluation therefore raises the domesticprice level, and consequently lowers real money balances. Theexcess demand for money balances leads to a reduction inabsorption. Thus, an improvement in the external payments positionis expected. In the meantime, the expenditure-reducing effect maylead to a fall in real output via a fall in aggregate demand. Thestructuralists, on the other hand, view devaluation as primarilyincreasing the costs of intermediate imported inputs, therebyincreasing costs. This leads to a fall in aggregate supply via areduction in the intermediate imported input demand of firms. Thus,an improvement in trade balance is also expected, but it comes at ahigh social cost. That is, a devaluation may lead to contraction notonly on the demand side but on the supply side as well. Thus, anincrease in the price level and a deeper recession is predicted.

It is worth noting that it would be misleading to suggest that Fundprograms rely exclusively on one or two policy instruments directedsolely at restraining domestic demand. Khan and Knight (1985)emphasized that a Fund program is a much more complex policypackage than a mechanical application of the simple monetaristapproach to the balance of payments. Indeed, Fund programs haveincorporated many alternative policies including those proposed bycritics, particularly those relating to the supply side. For our purpose,however, it is convenient to simplify and categorize Fund-typepolicies into demand-side policies and supply-side policies. In linewith this dichotomy, we argue in section IV of this paper that, in the


case of the Philippines, the demand-side policies are the mainemphasis of Fund-supportedadjustmentI_rograms.Thus, demand-side policiescan be synonymouswith Fund-supportedadjustment •policies,at least inthe shortrun.

Empirical Survey

Ourconcerninthispaperis tohighlightthe impactof monetaryandexchangerate policiesonthe levelof outputandprices.What follows,therefore,is a review of relevantempiricalstudiesfromwhich we candraw importantlessons.

Invan Wijnbergen(1982), a macroeconomicmodelofSouth Koreaincorporatingstylizedfacts aboutits financialstructureis presented.The role of the commercial banking system as an intermediarybetween privatewealth holdersand firms, the role of foreigncapitalinflows,the importanceof the informalmoneymarkets in the Koreanfinancialsystem, and the use of credit obtainedfrom those sourcesand abroad to financefixed and workingcapital, are all highlighted.The followingresultswere obtainedfrom the study:

a) Both export prices and domestic prices show a significantsensitivityto thecost offinancingworkingcapital.Thiscredit-supplysidelinl_givesa stagflationarybiasto restrictivecreditpolicies.

b) The roleofthecurb marketis quitepersuasive.Time depositsare shownto be a closersubstitutetocurb market loansthantoM1 (definedascurrencyincirculationanddemanddeposits),indicatingthata decreaseintime depositrateswouldmoderatestagflationaryimpacteffects by increasingthe availabilityoffundstothebusinesssectoras peopleshiftfromtime depositstocurbmarket loans.This wouldlowercurbmarketratesand,via the credit-workingcapital link,have a favorable effect oninflation.

c) It was also found that the phenomenal increase in timedepositsafter the interestrate reformin 1965 was caused bya switch from lendingat thecurb market to time depositsandnot by additionalsavingsas is usuallyclaimed. This impliessignificant substitutability between curb market lending andtime deposits.


Likewise,Lim(1987)arguedthatmonetaryandcreditcontraction,espousedby monetaristsas a keypolicyforcontrollinginflation,willincreasethecostof financingworkingcapitalneeds,thusnegativelyaffectingtheproductivesectorof theeconomy.Regressionanalysiswas utilizedto findouttheeffectsof contractionarypolicyon realoutputandprices.The studyusesa one-sectormacromodelWithannual data from the Philippinesfrom 1958 to 1980. The keyvariablesusedto explainrealoutputandpricesweretherealwagerate, expectedpriceinflation,i'ealexchangerate,domesticpriceofimportedinputs,levelofcapitalstockandinvestment,realbankloansavailable,anddomesticandforeigninterestrates.The resultsseemto indicatesome initial increase in price inflationimmediatelyfollowingthemonetaristprescription.This is thenfollowed_bydeeprecession.Then and0nlythendoespriceinflationstartto decline.Thus, these resultsgive empiricalsupportto the claim that theworkingcapitalcost-push,supply-sideeffectisstrongenough,whichmeansthat austeritymeasuresmay requirea heavy and costlyrecessionbeforepriceinflationcanbebroughtdown.

A formalassessmentof theFundstabilizationprogramwasdoneby Gylfason(1987).The authorreviewedthe relationshipbetweencredit policyand growthperformanceof 32 countriesunder thestabilizationprogramsupportedbythe IMF from1977to 1979.Theaim of the studywasto determinewhethereconomicgrowthwasadverselyaffectedinthe shortrunby the creditpoliciesadoptedundertheadjustmentprogramreviewed,andifso,towhatextent.Ananalytical framework was also developed to show themacroeconomicimplicationof thesupplyandcosteffectsof creditpolicy.

A simple before-and-aftercomparisonapproachwas utilized,complementedwitha nonparametrlcU-testproposedby MannandWhitneyto measurethe statisticalsignificanceof the differencesbetweenthechangeintheperformanceof thecountriesunderstudyfromtheyearbeforetheprogram,duringtheprogramyear,andtheyear after the program.Tests were also done to compare thesignificanceof differencesbetweenthe behaviorof the majormacroeconomicaggregatesincountriesthatenteredintoIMF stand-byarrangementsandthebehaviorof thesameaggregatesinanothergroupofcountriesthathadpersistentcurrentaccountorbalance-of-paymentsproblemsduring1975-1977butdidnotenterintostand-byarrangements.

36 JOURNALOF PHILIPPINEDEVELOPMENT

The results indicate that credit expansionwas reduced markedlyandthat theoverallbalance of paymentsimprovedsubstantially.Theinflatipn rate, although increasing on average, was generally keptbelow the,rates, prevailing in other non-oil-producing developingcountriesl"rllese results, it appears, were achieved at the cost of arelatively modest reduction in the average growth rate of outputduringand immediatelyafter the programperiod. It shouldbe kept inmind, however, that most of the programs under review wereaccompaniedbydevaluation,microeconomicsupply-sidemeasures,or increased foreign borrowing,all of which tend to counteract thedirect negativeeffectof contractionarycreditpolicy.Nevertheless, itappears that the experiencewith these stand-byarrangementsdoesnot give occasion for grave concern about the short-termcontractionaryor even stagflationaryconsequences of adjustmentprogramssupportedby ,theFund.

Gylfason (1987) also pointedout the weaknesses of the before-and-after approachincapturingthe"true" effectof Fundpolicies.Theissuesunderlyingthisapproachare adequatelydiscussedinthe 1986paper of Goldsteinand Montiel. ,

Goldstein and Montiel (1986) discuss several methodologicalproblems associated.withestimatingthe effectsof fund programs inFund-supportedcountries.Two common approaches in evaluatingFund programs are discussed.Thepurpose of the paper is to pointout problems that can cause "true" program effects to differ from"estimated" Program effects when either the before-and-afterapproach or the control-groupapproach is used. Both approachesuse multicountrydata. The before-and-afterapproachcalculates fortheselected countriesthe average macroeconomicoutcomesbeforeand after a Fund-supportedprogram is implemented.The approachthen tests for statisticalsignificanceof the difference inthe value ofthe macroeconomicvariables before and after the program. On theother hand, the control-group approach, by utilizing nonprogramcountriesas the basis for comparison,accountsfor changes in theinternationaleconomicenvironmentthatcouldalter macroeconomicoutcomesindependentlyof theprogram.The authoremphasized fourfactors that may influence key macroeconomicvariables (e.g., thecurrent account, rate of inflation,growthrate of real output,amongothers),namely:

a) changes in macroeconomic policy instruments (e.g., the rateof domestic credit expansion, government tax revenues andexpenditures, the exchange rate, etc.);


b) changes in world economicconditions(suchas changesinworld oil prices or changesin real economicactivityinindustrialcountries);

c) thetotaleffectofa Fundprogramifthecountryalreadyhadaprograminplaceduringtheperiod;and

d) a hostof unobservableshocksthatarespecificto a country.

Amajorflawof thebefore-and-afterapproachisitsrelianceontheassumptionof "otherthingsbeingequal"thatis highlyimplausible,There existsan empiricalproblemin isolatingtheeffectsof Fundpolicies on macroeconomicvariables. Independentpolicies ofgovernmentsandexternalshocksencounteredbythecountryaretwo majorreasonswhy thebefore-and-aftercomparisonof meaneconomicoutcomesforprogramcountriesis unlikelytoyielda goodestimateof true programeffects.As such,ascribingall of theobservedchangesinoutcomesto theprogramalonewillinvariablyoverstateorunderstatethetrueindependenteffectsof theprogram.

It isalsoworthnotingthattheauthorspointedoutthefundamentalissue of Fund-type policies vis-a-vis Fund involvement. Ininvestigatingthe effects of Fund:typepolicies,it is no longernecessary to differentiatebetween program and nonprogramcountries.Instead, the relevantcomparisonwould be betweenmacroeconomicoutcomesunderFund-typepoliciesandthoseundersome otherset of policies.The purposeof this type of analysis,therefore,is to findoutwhetherFund-typepoliciesare effectiveorineffectiveinthecountryconcerned.

Inthisstudy,anattempttoupdatetheresultsofLim(1987)ismadeutilizingannualdataof thePhilippinesfrom1960to 1986.Finally,thebefore-and-afterapproachis utilizedtocapturetheshort-runeffectsof Fundpoliciesonthesectoraloutputoftheeconomyovertheperiod1981to 1986.Althoughthisstudyisalsosubjecttothepitfallsof theapproach;as pointedout by Morrisand Goldstein(1986), weneverthelessarguethatit isadequateenoughto capturetheimpactof the 1984 Fund-sponsoredprogramon sectoral output.Theempiricalspecificationsand resultsare the subjectof the nextchapter.

EMPIRICALANALYSIS

Inthischapterwe presenta simpleempiricalmodelof priceandincomedeterminationwhereinboththemonetaristapproachandthestructuralisthypothesescan be tested. The before-and.after


approach is also presented to test the effects of the 1984 Fund-program on sectoral output. For conciseness of presentation we willalso give the corresponding empirical results rather than collect themin a separate section.

The Output Function

Following the model of Lim (1987), we assume a one-sectoreconomy. The .productionfunction is neoclassical and output isassumed to have three inputs: a) labor, b) capital stock, and c)importedraw materialsand imported intermediateinputs.To capturethe determinants of the real output of the economy, the followingsemireduced form equationis formulated:

Y = F (w, RER, MS, ITB, INFe, K)where,Y = real outputw = real wage rateRER = real exchangerateMS = money supplyITB = Jnterestrate on 90-day Treasury billsINFe = expectedinflationrateK = level of capitalstock.

On the supplyside,we postulatethat realoutput (Y) is determined-by the level of capitalstock(K), as well as by the real costsof inputs.The realwage rate (w) accountsfor thecosts of labor,while the realexchangerate (RER) accountsforthe costsof intermediateimportedinputs.The moneysupply(MS), interestonTreasury bills(ITB), andexpected inflation (INFe) capture the cost of working capital.Following simple rules of adaptive expectations, INFe can berepresentedby lag values of the inflationrate.

The interestonTreasurybillsservedas a proxyvariablefor lendingrates due to data constraints.We assume that lenders see perfectsubstitutabilitybetweenTreasury billsandIoanablefunds forworkingcapital needs, Thus, the higher the ITB, the higher the cost offinancingworkingcapital.The moneysupply (MS) isalsoexpected tohave a significanteffectonoutputvia theworkingcapital hypothesis.Contractionin the money supply decreases the supply of Ioanablefunds, thus increasingthe cost of financingworkingcapital. Anotherfactor affecting the lending rate is expected inflation. A higher

BALBOSA: IMF STABILIZATION PROGRAM 39

expected inflationrate reducesthe real lendingrates.•Thus, theinvestormustredirectIoanablefundsto otherassetswhosevaluesdonotfallwithpriceinflation.Theresultingscarcityof Ioanablefundsexertsan upwardpushonthecostofworkingcapitalviaa riseinthelendingrates.

Ontheaggregatedemandside,a reductioninMSresultsina fallinaggregatedemand;thus,a fall inoutputmay follow.Anincreaseinlendingrates (given'by ITB) resultsin a decrease in desiredinvestmentandthereforeleadsto a fall inaggregatedemand.ForINFe,weexpectthata higherexpectedinflationrateinducespresentconsumptionoverfutureconsumption,therebyincreasingaggregatedemandforgoodsandservices.

•Followingthemonetaryapproach,wenotethatthemechanismbywhichtheexchangeratecanaffectoutputisthroughthedemandandsupply of real money balancesand absorption.We note thatabsorptionis influenced,among other factors, by real moneybalances,whichareconsequentlyinfluencedby thedomesticpricelevel.In itssimpleform,theapproachyieldsthepropositionthatthedomesticpricelevelisdeterminedbythenominalexchangerateaswell as worldprices.Thus, a devaluation,holdingall otherthingsconstant,willreduceabsorptionviareductioninrealmoneybalances.Moreover,asthetradedgoodspricesriserelativetonontradedgoodsprices,expenditureswitchingmay occur,causingan expansionaryeffecton thenontradedgoodsector.

Table1summarizestheresultoftherealoutputmodel.Table1,aswellas Table2, usesbothOLSandInstrumentalVariable(i.e.,TSLSprocedure) estimatesto accountfor the fact that ITB is anendogenousvariablewhichisusedasanindependentvariableintheequation.The functionalrelationshipisinlogform.Thus,coefficientsrepresentelasticitymeasures.

The resultspointtoRER, ITB,andKas importantdeterminantsofchangesinrealoutput,withRER beingsignificantwithOLSbutnotwithTSLS. We findRERandITBsignificantlyandnegativelyrelatedto realoutputat the5 percentlevelofsignificance.It is interestingtonote that moneysupply(MS) has notshownmuch influenceindeterminingreal output (even consideringlagged variables).However,tightmonetarypolicyisbestreflectedbythehighnegativecorrelationbetweenITBandY, suchthata 1percentincreaseinITBmayleadto about0.14 percent(in OLS) or0.22 percent(in TSLS)reductioninoutput.A 1 percentrealdepreciation,ontheotherhand,may leadto an outputcontractionof 0.15 percent(in OLS) or 0.08


iiii -ill I Emil i

Table 1

REGRESSION RESULTS FOR REAL OUTPUT

Dependent Variable: In YExplanatory OLS TSLS a

Variables Coef. t-stat. Coef. t-stat.

C 1.50 1.56 0.82 0.68

1n w _-0.01 -0.07 0.01 0.071n RER -0.15 -2.25 =_ -0.08 -0.86In MS *0.10 -1.00 -0.14 -1.20in ITB -0.14 -2.74 _ -0.22 2.86 °_INFe -0.13 -1.48 -0.06 -0.521n K 0.83 4.13 ° 0.93 3.98 °

R-squared 0.9896 0.9870Adj. R-squared 0_9834 0.9792F-value 159.08 _ 126.28 _DW 1.40 1.46n 17 17

Sample Range 1970-86 1970-86

Noles; a The RHS endogenous variable is In ITB,Instrument list: C In w In RER In MS INFe In K In G In y6

• Denotes significance at the 1 _.,ercent level,• al Denotes significance at the 5 percent level,,i•• Denotes significance at the 10 percent level.

percent (in TSLS). We found multicollinearity problems in the model.This may cause a downward bias on our t-scores but it will not biasour estimators.

Comparing with Lim's findings, his OLS and TSLS functions showthat monetary-related variables do affect the supply functionsignificantly. Real bank loans have positive effects on outputparticularly in the post-1971 period. Interest rates were found to havesignificant negative influences on the supply function. In this regard,Lim's results are consistent with this study.

The Price Inflation Function

The semireduced form equation derived from Lim (1987) yields thefollowing function for price inflation:


• Ill I I •

Table 2 ,REGRESSION RESULTS FOR PRICE INFLATION

Dependent Var able: Ap

Explanato[y OLS TSLS aVariables Coal. t-stat. Coef. t-stat.

C 0.22 4.50 e 0.22 4.38 e^w -0,07 -0.36 -0.08 -0.43

^RER -0,41 .3.01 e° -0.43 -2,98 ee^MS 0,27 1.42 0.32 1.47dlTB 0.02 6.14 e 0.03 4.13 °dlNFe -0.04 -0.40 -0.07 -0.56^K -1,49 -2,64 °e -1.60 -2.60 ee

R-squared 0.8529 0.8452Adj. R-squared 0,7548 0.7420F-value 8,70 e 8.19 eDW 2.14 2.12n 16 16

Sample Range 1971-86 1971-86

Notes: a The RHS endogenous variable Is dl'rB.

Ihstrument llst'. CAW ARER _ dlNFe ^KIn GIn ya

I Denotes $ignlfloartce at the 1 parent level,

,lie Oenotes significance el the $ peroent level,

goo Denotes ei_lNfk_rloe at the 10 percent level.

^P = F ( ^w, ^RER,^MS, dlTB, dlNFe,^K )where,^P = the inflationrate^ = the growthrateofvariablesd = thefirstdifference.

Arisein^RER (adepreciation)isexpectedtohavea positiveeffecton the inflationratewhetherfromthemonetarists'orstructuralists'pointof view.On theotherhand,monetarycontractionwill haveanambiguouseffectonthepricelevel.Monetarycontractionisexpectedtohavea positiveeffect'ontheinflationrateviathedemandsideanda negativeeffectvia the supplyside.The workingsof the moneysupplyontheinflationrateviathelendingratesarebestcapturedby

42 jOURNAL OF PHILIPPINE DEVELOPMENT

the instrumental variable method using dlTB as the right-hand sideendogenous variable.

Table 2 summarizes the result of the Price Inflation model. Overall,the RER, ITB, and K show a significant influence on price inflation.The result for ITB strongly confirms th_belief of the structuralists. Wealso note that MS has an insignificant effect on prices although itscoefficient is much bigger (i.e., it has a more important effect) thaniTB. Moreover, the sign of the MS coefficient confirms thedeflationary effect of money supply on the price lever. The result onRER is curiously contrary to what we expected.

Concerning the capital stock level, the result in this model as wellas in the real output model affirms the crucial role of capital formationin economic growth. No autocorrelation problem was detected in bothmodels. Lim's findings on price inflation reveals that, in the short run,the working capital cost-push effect seems to dominate. Lendingrates were found to have a significantly positive effect, while the signfor the growth rate of real bank loans was negative. His findings seemto point to a "stronger supply-side, working capital cost-pushinflationary push than to the demand-side, monetarist deflationarypull." Lim's result on the lending rates seems to corroborate ourfinding on the effect of ITB on the inflation rate. On the other hand, thesign of our MS coefficient (which is positive) reveals the dominanceof the monetarist's deflationary pull. However, its t-score fails toestablish its significance.

The Before-and-After Approach

We used the before-and-after approach to investigate whethersectoral outputwas adverselyaffected in the shortrun by the creditpolicies adopted under an adjustment program. Due to datalimitations,we utilized the real output of the manufacturing sectorover the years 1981 to 1986. FollowingMontes (1987), we assumedthe implementationof the Fund programto have begun in earnest in1984. The task was to compare real sectoral output before theprogram to real sectoral output after 1984. Moreover, wehypothesizedthatthegreater the need forworkingcapitalof a sector,the more it is affected by contractionary monetary policies. Weassumed thatworkingcapitalisneeded to financethe laborcostsandintermediateimported inputcosts of a firm. Thus, the more import-dependent and labor-usinga firm is, the more it is affected by thecosts of obtainingworkingcapital.


By usingthe Wilcoxon-Mann-Whitneytest (or theW-test),we candeterminewhethersectoraloutputsignificantlydecreasedafter 1984.We utilized the null hypothesis (Ho) that real output remainedconstant from 1981 to 1986against the alternativehypothesis(Ha)that itsignificantlydeclinedafter 1984.Table3 summarizesthe resultof theW-teston sectoraloutput.

I II I II II I II IIlI

Table 3SUMMARY OF W-TEST RESULTS ON TREND

OF REAL SECTORAL OUTPUT, 1981-1986

Sector Decision

Food Manufactures Reject HoBeverage • Accept HoTobacco Accept Ho

• Textiles Reject HoFootwear Reject HoWood and Wood Products Reject HoFurniture and Fixtures Reject HoPaper Products Accept HoPrinting Reject HoLeather Products Reject HoRubber Reject HoChemical Products Reject HoPetroleum Products Reject HoNon-metallic Mineral Products

(cement and glass products) Reject HoBasic Metal

(including iron and steel products) Accept HoMetal Products

(fabricated metals) Reject HoMachinery

(non-electrical machinery,.equipments and parts) Reject Ho

Electrical Machinery Accept HoTransport Equipment Reject Ho

TOTAL OUTPUT Reject Ho

Source of raw date: 1988 Philippine Statistical Yearbook.

Note: Ho: Real output remained constant from 1981 to f 986,

Na: Real output significantly declined after 1984,

-...-,


Overall, 13 (or 72 percent) out of 18 industries listed in Table 3show a significant decline in real output after 1984. However, cautionmust be exercised in interpreting the results because there are otherfactors which may determine the trends of sectoral output over time.The W-test simply described the industry behavior during the IMFStabilization Program,

LESSONS AND POLICY IMPLICATIONS

The 1984 and 1989 IMF Program

Under the December 1984 agreement (the 18th Stand-byAgreement with the IMF), the Philippinegovernmenthad taken thefollowingmeasuresbefore financingwas extended:

a) a reduction in high-powered money through ceilings ondomestic borrowing;

b) the floating of the exchange rate, which .resulted in thedepreciationof the currency;

c) a reformof the foreignexchangesystem by adoptinga moreliberaltrade policy;

d) an increaseininterestratesonCentralBankbillsandTreasurybills to approximately40 percent at the time of.the float,togetherwith increases in several other lending and depositrates;

e) the implementation of significant,tax revejnue generatingmeasures; and

f) increases inadministeredprices to reflectmarket conditions.

On the Memorandum on Economic Policy submitted by thegovernment in March 1989, the following policy measures andadjustmentswere outlined:

a) a reductioninthe growthrateof moneyandcreditbyabout 12percent overtheperiod 1989to 1992 fromitsactual 1987 levelof 15 percent annually;

b) the removal of impedimentsto privatesector activitythrougha dismantlingof monopolies and special privileges in somesectors;

c) trade policyreform emphasizinga competitiveand outward-lookingstrategy;


d) public sector reform through the privatization of governmentparastatals, a reduction in the government deficit through taxreforms, and decreases in investment spending; and

e) a reform package to generate domestic saving largely throughthe government sector.

Indeed, a cursory review of the 1984 and 1989 adjustmentprograms indicates that there was not much difference between thetwo. Monetary policy through the control of domestic credit andexchange rate programs were the main policies put in place. It shouldalso be noted that the 1989 program called for a strict reduction in thegovernment's role in the economy (i.e., government activities thatoverlap with the private sector activities were eliminated or at leastminimized). Not surprisingly, these policy measures were the oneswe referred to as typical Fund-type stabilization policies.

Summary, Lessons, and Policy Implications

We learned that a primary determinant of output and the domesticpricelevel is the lendingrate.The resultssuggestthat tightmonetarypolicymay indeedinducepricesto fall, corroboratingthemonetarists'belief. However, the effectof thispolicyon the lending rates shouldnot be ignored. One has to consider the profound effects of thevolatilityof lending rates on the economy via the supply-sideeffectsof financing working capital. Our nonparametric test (the W-test)indicates a significant decline in the real output of selectedmanufacturing industries which can be traced to tight monetarypolicy. Moreover, policiesthat raise the real effectiveexchange ratehave a contractionaryeffect on output,but their inflationary effectcould not be firmly established by the results. And finally, weconfirmedthe importanceof investmentandthe level of capitalstockon the growth of real output.Such expansionin outputmitigatestheinflationarypressuresinthe economy.

The task for the 1990s, therefore, is to come up with policiesthattake into consideration the stagflationary effect of contractionarypolicies,the role of the real exchange rate on outputand prices, andthe importanceof investmentand capitalstockinthe economy.Withthese in mind, caution should be exercised in implementing futureadjustmentprograms.Targets to lower monetaryaggregates basedon the monetary approach framework alone may underestimateornot capture the "true" effects on output and prices. Moreover, credit ....


which is shown to have an importantrole in the real sector, mustbeadequately providedto preventstagflationarypressures. However,thisshouldbe clonewithinthe contextof developinga more.dynamiceconomythat can fully explore itscomparativeadvantage. Finally, areductionin the governmentdeficit is called for in order to min!mizethe crowding-outeffectexperienced bythe privatesector duringthe1970s and 1980s. This can be donewithinthe context of:

_a) a general reduction in government expenditures throughdetermined and programmed reduction in the size of thegovernment bureaucracy and the establishment of a systemofprioritizing government projects (especially in infrastructure)which strongly adheres to economic efficiency criteria; arid

b) an improvement in the tax _¢.,o_ectionmachinery throughstructural and administrative reforms in the tax system.


BIBLIOGRAPHY

Alburo, F.A. "Philippine Trade in Manufactures, Structural Changeand Adjustment." UPSE Discussion Paper 8509. Quezon City:University of the Philippines School of Economics, 1985.

Bautista, R.M. "-Ihe 1981-85 Tariff Changes and Effective Protectionof Manufacturing Industries." UPSE Discussion Paper 8213.Quezon City: University of the Philippines School of Economics,1982.

Goldstein, M. and Montiel, P, "Evaluating Fund StabilizationPrograms with Multicountry Data: Some Methodological Pitfalls."IMF Staff Paper No. 33, pp. 304-344, June 1986.

Guzman, V.S. "The 1983 Balance-of-Payments Crisis of thePhilippines: A Painful Lesson." Masteral thesis, Center forDevelopment Economics, Williams College, 1985.

Gylfason, T. "Credit Policy and Economic Activity in Developing•Countries with IMF Stabilization Program." Princeton Studies inInternational Finance, No. 60, August 1987.

Khan, M.S. and Knight, M.D. "Fund-Supported Adjustment Programsand Economic Growth." IMF Occasional Paper No. 41.Washington, D.C.: International Monetary Fund, November 1985.

Lira, J. 'q-he New Structuralist Critique of the Monetarist Theory ofInflation: The Case of the Philippines." Journal of DevelopmentEconomics 25 (1987): 45-61.

Montes, M. F. "Macroeconomic Adjustment in the Philippines, 1983-85." PIDS Working Paper Series No. 87-01. Makati: PhilippineInstitute for Development Studies, 1987.

Rivera-Batiz, F.L. International Finance and Open EconomyMacroeconomics. New York: McMillan Publishing Co., 1985.

Tanzi, V. "Fiscal Policy, .Growth and Design of StabilizationPrograms." In External Debt, Savings and Growth in LatinAmerica, edited by A. Martirena-Mantel. Washington, D.C.:International Monetary Fund, 1987.

Taylor, L. Structuralist Macroeconomics. New York: Basic Books,inc., 1983.

van Wijnbergen, S. "Stagflationary Effects of Monetary StabilizationPolicies: A Quantitative Analysis of South Korea." Journal ofDevelopment Economics 10 (1982): 133-169.

Wonnacott, T.H. Introductory Statistics for Business and Economics.2nd ed. New York: John Wiley and Sons, 1977.


Central Bank of the Philippines. "Memorandum on Economic Policyof the Government of the Philippines," October 11, 1984.

"Memorandum on Economic Policy During thePeriod 1989-92," March 7, 1989.

International Monetary Fund (IMF). International Financial StatisticsYearbook. Washington, D.C.: IMF, 1989.

National Statistical Coordination Board (NSCB). Philippine StatisticalYearbook. Makati: NSCB, 1988.

journal of philippinedevelopment 4 ii u xlx n s ibriib · 2003. 4. 25. · journal...

Documents