Pakistan Institute of Development Economics, Islamabad

Investment, Inflation and Economic Growth Nexus

Author(s): Nasir Iqbal and Saima Nawaz
Source: The Pakistan Development Review, Vol. 48, No. 4, Papers and Proceedings PARTS I and
II The Silver Jubilee Annual General Meeting and Conference of the Pakistan Society of
Development Economists Islamabad, March 16-18, 2010 (Winter 2009), pp. 863-874
Published by: Pakistan Institute of Development Economics, Islamabad
Stable URL: http://www.jstor.org/stable/41261352 .
Accessed: 23/06/2014 02:04

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

Pakistan Institute of Development Economics, Islamabad is collaborating with JSTOR to digitize, preserve and
extend access to The Pakistan Development Review.

http://www.jstor.org

This content downloaded from 195.78.108.147 on Mon, 23 Jun 2014 02:04:27 AM

All use subject to JSTOR Terms and Conditions
©ThePakistanDevelopment Review
48 : 4 PartII (Winter2009) pp. 863-874

Inflationand Economic
Investment,
GrowthNexus
Nasir Iqbal and Saima Nawaz

1. INTRODUCTION
High and sustainableeconomic growthand low inflationare the two main
objectives of policy-makersand the central bank. It is generallybelieved that
inflationhas negativeand significantimpacton economic growthin medium-and
long-run[Khan and Senhadji (2001)]. However, the existence and nature of
relationshipbetweeninflationand economicgrowthand thechannelsthroughwhich
it affectsreal economic activitieshas been the subjectof considerableinterestand
debatedue to inconclusiveresults.Recentliterature on thisissue has uncoveredsome
important findings.
Empiricalliteratureon inflationgrowthnexusis dividedintotwo mainstrands.
One strandof literaturehas found negative and significantrelationshipbetween
inflationand economic growth[Fisher (1993); Barro (1995); Bullard and Keating
(1995); Malla (1997); Bruno and Easterly(1998) and Faria and Carneiro(2001)]
whileotherhas confirmedpositiveand significant associationbetweeninflationand
economic growth[Lucas (1973); Mallik and Chowdhury(2001) and Gillman and
Nakov (2004)]. These strandsof literaturehighlightthe possibilityof non-linear
relationshipbetweeninflationand economicgrowth.Several recentempiricalstudies
have exploredthattherelationshipbetweeninflationand economicgrowthis in fact
nonlinear.They are tryingto supportthe hypothesisthatlow and stable inflation
promoteseconomicgrowthand vice versa.
Fischer(1993) exploredthis possibilityand notedthe existenceof nonlinear
relationshipbetweeninflationand economicgrowth.He foundthattherewas a positive
associationbetweeninflationand economic growthat low rate of inflation,and a
rose.Findingsof Fischer(1993) generatenewdebateamongthe
negativeone as inflation
economiststo determine thatpromoteseconomicgrowth.
preciselythelevel of inflation
Nasir Iqbal <nasir@pide.org.pk>is StaffEconomistat the PakistanInstituteof Development
Economics,Islamabad.Saima Nawaz <saimanawaz-2006@yahoo.com> is PhD Studentat thePakistanInstitute
ofDevelopment Economics,Islamabad.
Authors'Note: The authorsthankto Dr Musleh ud Din JointDirectorat the PakistanInstituteof
DevelopmentEconomics,Islamabadfortheirvaluablesuggestionsand MuhammadJavidStaffEconomistat
thePakistanInstitute The authorsalso wishto
of DevelopmentEconomics,Islamabadforhelp in estimation.
thankDr Mohsin S. Khan,Dr WaqarMasood Khanand Dr WasimShahidMalikfortheirusefulcomments on
an earlierversion.

This content downloaded from 195.78.108.147 on Mon, 23 Jun 2014 02:04:27 AM

All use subject to JSTOR Terms and Conditions
 Investment, and EconomieGrowthNexus
Inflation

In thiscontext,variousempiricalstudiesare conducted.Sarei (1996) foundthatbefore

1970s inflation rateswere modestin mostcountriesand empiricalstudiesduringthis
period show the evidenceof a positiverelationship betweeninflationand economic
growth while after 1970s inflation rates started to be highand a negativerelationship
betweenthesevariables,beyondthattimeperiod,was observed.Bruno and Easterly
(1998) examinethedeterminants of economicgrowthusingannualCPI inflation of 26
countrieswhichexperiencedinflation crisesduringtheperiodbetween1961 and 1992.
Theirempiricalanalysispredictsthatinflation rateof 40 percentand over is considered
as thethreshold levelof inflation.
Khan and Senhadji (2001) explored the inflationand growthrelationship
separatelyfor industrialand developingcountriesand re-examinedthe issue of the
existenceof "Threshold"effects. Theirresultsshowthatthethreshold rateof inflationis
1-3 percentforindustrial countriesand 7-11 percentfordevelopingcountries.Lee and
Wong (2005) estimatedthe thresholdlevels of inflationforTaiwan and Japanusing
quarterly data set fromthe periodbetween1965-2002 forTaiwan and 1970-2001 for
Japan.Theirestimation of thethreshold modelssuggeststhatan inflation ratebeyond7.3
percent is detrimental forthe economic growth of Taiwan. On theother hand,theyfound
twothreshold levelsforJapan,whichare 2.5 percentand 9.7 percent.Theysuggestthat
inflationratebelow theestimatedlevelof 9.7 percentis favourableto economicgrowth
and beyondthisthreshold value it is harmful fortheeconomicgrowthin Japan.Munir
and Mansur(2009) investigatethe non linearrelationship betweeninflationrate and
economicgrowth duringtheperiodof 1970-2005forMalaysia.Theyfoundthatthreshold
level of inflationis 3.9 percentand supportthe view thatthe relationship between
inflationrateand economicgrowthis nonlinear. Inflationrateabove thethreshold level
significantly retards growth rate of GDP and below the thresholdlevel, it promote
economicgrowthsignificantly. Sergii(2009) investigate thegrowth-inflation interaction
forCIS countriesforthe periodof 2001-2008 and foundthatwheninflationlevel is
higherthan8 percenteconomicgrowth is sloweddown,otherwise, itis promoted.
How theninflation affectgrowthand moreparticularly, whatdo givesriseto the
so-calledthreshold effectin therelationship? Whatis thechannelthrough inflationcan
affectgrowthin nonlinear settings? Recentliterature highlightsthatinvestment mightbe
consideras an important channelthroughwhichthe impactof inflation is transmitted
nonlinearly in economicgrowth.Investment, inflationand economicgrowthnon linear
nexuscan be explainedby usingfinancialmarketdevelopment. Nonlinearity between
inflationand financialdevelopment is well documentedin literature [Boyd and Smith
(1998); Huybensand Smith(1998, 1999); Boyd,et ai (2001); Khan,et ai (2001)]. A
predictable increasein therateof inflation can slow downfinancialmarket development.
Inflation,a tax on real balance,reducesreal returns to savingswhichin turncauses an
informational frictionafflictingthe financialsystem.These financialmarketfrictions
resultsin creditrationingand thuslimittheavailabilityof investment and finallythis
reductionin investment adverselyimpactseconomicgrowth.In endogenousgrowth
literature,economicgrowthis dependent on rateof return and inflation decreasesrateof
return[Nelson (1976); Fama and Schwert(1977); Gultekin(1983) and Boyd, et ai
(1996)], whichleads to reduction in capitalaccumulation and hencedecreasethegrowth
rate.Inflation createsuncertainty in thefinancialmarket and increasestheriskassociated

This content downloaded from 195.78.108.147 on Mon, 23 Jun 2014 02:04:27 AM

All use subject to JSTOR Terms and Conditions
 IqbalandNawaz 865

withtheinvestment whichtranslated intoreductionin economicactivities[Hellerstein

(1997)]. Inflationcan discourage investors by reducingtheirconfidencein investments
thattakea long timeto maturein stockmarket. Thereare fewstudiesthatexploredthe
investment inflationrelationshipand hence its impacton growth.Accordingto Barro
(1995) reductionis economicgrowthis occurreddue to reductionin thepropensity to
investment thatis outcomeof inflation. He further shows thatan increasein average
inflationby 10 percentagepointsper yearcause reductionin theratioof investment to
GDP by 0.4-0.6 percentagepointsand thisreductionin investment reducesthereal per
capitaGDP by 0.2-0.3 percentage pointsper year.Therefore, inflation reducesthelevel
of investment and hencereductionin investment adverselyaffectseconomicgrowth.Li
(2006) estimatesrelationship betweeninflation and investment for27 developedand 90
developing countriesover the period 1961-2004 and found that relationship is nonlinear
forbothdevelopedanddevelopingcountries.
Pakistan'seconomyhas lostsignificant momentum in lastfewyears.Deteriorated
economicgrowthcoupledwithhighinflation and low investment is majorproblemsof
Pakistan'seconomy.The growthrateof realGDP has gonedownto 1.2 percentin 2008-
09 from9 percentin 2005-05. Investment is a key determinant of growthand its
fluctuation reflectsthe intensificationof economicactivities.The totalinvestment has
declinedfrom22.5 percentof GDP in 2006-07 to 19.7 percentof GDP in 2008-09 and
privatesectorinvestment was decelerating persistentlysince2004-05and itsratioto GDP
has declinedfrom15.7 percentin 2004-05to 13.2 percentin 2008-09.The inflation rate,
measuredas ConsumerPrice Index (CPI), has climbedto 22.3 percentduring2008-09
overthecorresponding increaseof 10.3 percent[Pakistan(2010)].
These statisticsreveals that it is important to investigatethe nexus among
inflation,investment and economic growth.Few studies,in Pakistan,envisage the
existenceof non linearrelationship betweeninflation and economicgrowth[Mallikand
Chowdhury (2001); Mubarik (2005) and Hussain (2005)]. Mubarik(2005) estimatesthe
threshold levelof inflationforPakistanusingan annualdatasetfromtheperiodbetween
1973 and 2000. He foundthatan inflation ratebeyond9-percentis detrimental forthe
economicgrowthwhile inflationratebelow thislevel is favourableforthe economic
growth.Hussain (2005) empiricallyestimatesthe thresholdlevel of inflationusing
standard econometric techniqueused forestimations of threshold effectfortheperiodof
1973-2005 in Pakistan.He foundno thresholdlevel of inflationforPakistan.These
resultsare in sharpcontrastto thefindings of Mubarik(2005) wherethreshold level for
Pakistanis at 9 percent.He suggeststhattargeting inflation exceedinga rangeof 4-6
percentwillbe a deterrent to economicgrowthand thisrangeof inflation is tolerablefor
Pakistan.
Ourstudyis different in threerespectsfromotherstudiesthathavebeenconducted
forPakistan.First,we focus on a morerecentand long timeseries (1961 to 2008).
Secondly,thesestudiesfocuson theexistenceof onlyone threshold levelbetweenthese
two variablesby ignoringthe possibilityof second thresholdin the relationship of
inflationand growth. Thirdly, thesestudies have not examined therole of investment as a
channelthrough whichinflation affectseconomicgrowth.
The paper has twofoldobjectives.Firstly,the impactof the inflationrate on
economicgrowthwiththepossibility of two threshold levels forPakistanusingannual

This content downloaded from 195.78.108.147 on Mon, 23 Jun 2014 02:04:27 AM

All use subject to JSTOR Terms and Conditions
 Investment, and EconomieGrowthNexus
Inflation

data from1961 to 2008 is examinedand secondly,nonlinearrelationshipbetween

inflation has been investigated.
and investment Followingquestionsare analysedin this
context:
(1) Does a secondthresholdlevelexistin theinflation-growth
relationship?
(2) What is the betweeninflation
relationship and investment?
Does theeffectof
inflationon investment
show a similarpatternto thatinflation
on economic
growth?
The remainder of paperis organisedas follows:Model specification is discussed
in Section 2. Data and descriptivestatisticsare explainedin Section 3. Resultsare
presentedinSection4 whileconclusionandpolicyrecommendation are in lastsection.

2. THE MODEL SPECIFICATION

The relationship
betweeninflationand economicgrowthcan be derivedusingthe
standard
growth equation[Barro(1991) and Sala-i-Martin
(1997)]:
¿logK = Xß+ e ... (1)

Where Y is real output,X is a set of explanatory variables,ß is slope coefficients

attachedwithexplanatoryvariablesand e is theerrorterm.Thisbasic growth equationis
extendedto capturesthelinkbetweeninflation and economicgrowthby usingfollowing
equation:
JlogK = a0+a1/n/ + Xß+ e (2)
Whered log Y is growthrateof real GDP, //2/ is growthrateof CPI and X is matrixof
otherexplanatory variables,ß matrixof slopecoefficients ande is theerrorterm.
Neoclassicalgrowthmodeluses investment and populationgrowthin thegrowth
analysis.Increase in investment togetherwitha decrease in populationgrowthrate
promoteseconomicgrowth.International tradetheoryproposesto includeopennessof
the economyin the growthregressionwhichis positivelyrelatedto growth.Money
supplyis important indicator forfinancialdevelopment. Developmentin financialsector
is positivelylinkedwitheconomicgrowth.Finally,our empiricalanalysisuses the
following explanatoryvariables:investment, populationgrowth, M2 and opennessof the
economy.Choice of variablesis consistentwiththechoice made by otherresearchers
[Khanand Senhadji(2001); Drukker, et ai (2005); Mubarik(2005); Hussain(2005); Li
(2006) and Sergii(2009)].
So, ourfinalregression modelis as follow:

rflogy= a0+a1(//i/) + ß1(P)+ß2(WV) + ß3(F) + ß4(O) + e (3)

Whered log Y is growthrateof real GDP, /л/is growthrateof CPI and P is population
growthrate,INV is investment to GDP ratio,F is M2 to GDP ratio,О is openness
+
((Export Import)/GDP) and e is theerrorterm.
Theoreticalas well as empiricaldebate predictsthat thresholdeffectsare
associatedwitha rateof inflation exceedingsome"criticalvalue"or belowsome"critical
value".ThresholdModel was developedbyKhanand Senhadji(2001) fortheanalysisof

This content downloaded from 195.78.108.147 on Mon, 23 Jun 2014 02:04:27 AM

All use subject to JSTOR Terms and Conditions
 Iqbal and Nawaz 867

thresholdlevel of inflation
forindustrialised and developingcountries.Mubarik(2005)
andHussain(2005) use thesame modelfortheestimation of threshold
levelof inflation
in Pakistan.In thismodelonlyone threshold level was captured.We extendthismodel
withthepossibility of twothresholdlevel in inflation growthnexus.By introducing two
thresholdlevelof inflation;
followingfinalregression modelis designed:
d logY = ce,+ a2 (/л/)* /(//i/
< n{ ) + cc3(Inf) * 1(щ < Inf< n2

Wheredependant variableand thecontrolvariablearedefinedas thesameas in Equation

3 whileЩ andп2 aretwothreshold levelof inflation.
I(Inf< Tij),1(щ < Inf< n2)and ¡(Inf
< n2)areindicators functions whichtakethevalueofone ifthetermbetweenparentheses
is trueand are zero otherwise.This modelspecifiestheeffectsof inflationwiththree
coefficients:a2, a3, and a4. a2 denotestheeffectof inflation below thefirstthreshold
level7ii,a3 denotestheeffectof inflation on economicgrowth betweenщ andn2,and ct4
denotestheeffectof inflation on economicgrowthexceedingthesecondthreshold level

Identificationof thresholdis based on the methodology definedby Khan and

Senhadji(2001). Regressionequation is estimatedfordifferentvaluesof thresholdwhich
is chosenin an ascendingorder(i.e., 1, 2 and so on), the optimalvalue thresholdis
obtainedbyfinding thevaluethatmaximisestheR2fvom therespective regressions.This
also impliesthattheoptimalthreshold level is thatwhichminimises theresidualsumof
squares(RSS). This procedurehas becomewidelyacceptedin theliterature on thistopic.
Search of optimalthresholdfor widerrangeof thresholdis verytedious.Moreover,
Hansen(2000) proposedto searchoptimalvalue onlyin theregionwheredo expectthe
thresholdshouldbe.
Theoreticalliterature
indicatesthatinvestment mightbe thechannelthrough which
inflation
hitseconomicgrowth.Followinglinearmodelspecification is used to measure
therelationship
betweeninvestment andinflation:

INV=80+5]Inf + 82INVt_l+E (5)

WhereINV is thegrossfixedcapitalaccumulation as shareofGDP andfirstlag of INV is
includedto controltheeconomicconditions in thelastperiod.Withthepossibility
of two
in investment
thresholds inflation
nexus,followingmodelis designed:

7(71,<Inf<n2
',-i) + e (6)
Selectionof thresholdlevel is based on the similarprocedureexplainedfor
inflation
andeconomicgrowth

3. DATA AND DESCRIPTIVE STATISTICS

The dataaretakenfromEconomicSurveyofPakistan(VariousEditions)and Fifty
Year Economyof Pakistan (SBP). Data are rangingfrom1961 to 2008. Following
variablesare used in analysis.Growthof real GDP is measuredas annualpercentage
growthrateof GDP at constantpricesbased on 2000 prices.Inflationis measuredas

This content downloaded from 195.78.108.147 on Mon, 23 Jun 2014 02:04:27 AM

All use subject to JSTOR Terms and Conditions
358 investment, and EconomicGrowthNexus
Inflation

annualpercentchangeof averageconsumerpriceindex.Data forinflation are averages

fortheyearand indexis based on 2000=100. Growthrateof populationis measuredas
annualpopulationgrowthrate. Investment is measuredas grosscapital formation as
percentof GDP. Openness of the economy is measured as shareof exportplus importin
GDP.
of sampledatashowsthattheaveragevalueof growthrateof
Descriptivestatistics
outputis 5.5 and
percent, growth rateof populationhas theaveragevalueof 2.7 percent,
investment has averagevalue 18.1 percentand opennessof theeconomyhas average
value35.4 percent.Inflationhas mean7.8 percentwhileM2 as shareof GDP has average
valueof 34.7 (Table 1).

Statistics
Descriptive
Variables N Mean SD Min Max
GDP GrowthRate 48 5.48
548 ТШ 1.20 9.80
Investmentas % ofGDP 48 18.08 2.15 12.93 22.95
Inflation 48 7.81 5.29 -0.52 26.66
Growth RateofPopulation 48 2.67 0.33 1.78 3.19
M2as%ofGDP 48 34.76 4.89 24.28 46.69
Openness 48 35.39 3.19 28.85 42.62

of theseriesis confirmed
The stationarity by applyingAugmented Dickey-Fuller
(ADF) test.Table 2 gives theresultof ADF forall series.Real GDP growthrateand
opennessare stationary at level whileInvestment
to GDP ratio,Inflation, Population
GrowthrateandM2 as percentof GDP are nonstationary at leveland becomestationary
at first
difference.

TestforNon-stationarity
of Variables
Level FirstDifference
Variables No Trend WithTrend Result No Trend WithTrend
GDP GrowthRate -5.73 -6.01 Stationary - -
Investmentas % ofGDP -2.12 -2.39 Non-stationary -6.98 -6.99 Sta
Inflation -2.15 -2.86 Non-stationary -4.60 -3.89 Sta
GrowthRateof Population -0.25 -2.44 Non-stationary -3.77 -4.28 Sta
M2as%ofGDP -0.73 -1.44 Non-stationary -5.92 -3.75 Sta
Openness -3.62 -3.76 Stationary - -
Note:5 percentcriticalvalueis -2.87 forthecase of no-trend,
and-3.42 whena trendis included.A1C is used
forlag selection.

4. MODEL ESTIMATION

4.1. Inflationand EconomicGrowthNexus

The simplelinearmodelof economicgrowthand inflation as definedin Equation
3 has beenestimated.The basic purposeof simplelinearregression
is to revealtheshape
of thegrowthfunction relatingtheinflationwitheconomicgrowth. Resultindicatesthat

This content downloaded from 195.78.108.147 on Mon, 23 Jun 2014 02:04:27 AM

All use subject to JSTOR Terms and Conditions
 Iqbal and Nawaz 869

has significant
inflation negativeimpacton economicgrowthat secondlag.1One percent
increasesin inflation
causes 0.2 percentreductionin growthrateof GDP. Investmenthas
positiveand significantimpacton economicgrowthwhilepopulationgrowthalso has
positiveand significantimpacton economic growth(Table 3). The coefficientof
investment / GDP ratiois 0.65 whichindicatesthata 1 percentagepointincreasein
investment willcause a 0.65 percentage pintincreasein growth.OthervariablelikeM2
to GDP ratioand opennesswas also used in theregressionequationand finallyboth
variablesweredropdue to theirinsignificantrelationshipwitheconomicgrowth.

LinearEstimation
Results(DependantVariableis GDP GrowthRate)
Variable Coefficient t-Statistic
Constant -14.35769 -2.368223
Inflation -0.198856 -3.274821
Investment 0.651056 3.573079
Population 0.036202 2.696865
R-Squared= 0.3 1; DW = 2. 12; Jarque-Bera
= 0.12; RamseyRESET Test( 1, 41) = 1. 11 [0.30].

Nonlinearmodelhas beenestimated usingEquation4. Forestimation of щ andn2,

we applythemethodologygiven in Section 2. we
First, estimate theEquation3 withone
threshold
level.Withthepossibility
of one threshold level,we reformulate
Equation3 as
follow:

uflogr= a1+a2(//t/)*/(/n/<^) + a3(//i/)*/(//2/>7r1)

+ ßl(/))+ ß2(/) + e (7)
We applya rangeof threshold levelrangingfrom1 to 8 and choosethevaluethat
minimises theerrorsumof squareas mention byHansen(2000). Finally,resultindicates
thatthevalueofщ is 6 and inflation below6 percenthaspositivebutinsignificant impact
on economic growth(AppendixTable 1-A and Table 1-B). Then, we carryout a
significanttest of no thresholdagainst one thresholdщ. The null hypothesisis
Ho = a2 = oc3against the alternativeof#0 =oc2 * oc3. The resultindicatesthat null
hypothesis is rejectedat 5 percentlevel of significance whichconfirms theexistenceof
one threshold levelin inflationdata.
The existenceof secondthreshold in therelationship
betweengrowthand inflation
is testedby usingEquation4. By usingsame processwe findthesecondthreshold level
whichis 11 (Table 4). Then,we carryouta significant testof one threshold
againsttwo
thresholds.The nullhypothesis is existenceof onlyone threshold the
against alternative
ofexistenceof twothresholds. The resultsupports theexistenceof twothresholdsagainst
one at 5 percentlevelof significance.
Our findingsshow thatforthe low inflation below the fistthresholdlevel, the
coefficient of inflation(0.18) is positivewhichimpliesthat1 percentageincreasein
inflationwill cause a 0.18 percentage pointincreasein economicgrowth.However,this
positiveimpact is not This
significant. indicatesthat,in Pakistan,low inflation upto6 is
'Mubarik(2005) and Hussain(2005) also foundthatinflation
effect
economicgrowth
at secondlag.

This content downloaded from 195.78.108.147 on Mon, 23 Jun 2014 02:04:27 AM

All use subject to JSTOR Terms and Conditions
g70 Investment, and EconomicGrowthNexus
Inflation

notharmful forthecountry.2In themiddlerangeof inflation i.e. betweentwothreshold

level (inflationbetween6 and 12), thecoefficient of inflation (-0.32) is negativeand
significantat one percent level. Results show thatan increase in one percentagepoint
inflationper year is associatedwith a reduction of the growth of real GDP by 0.32
rate
percentage point.Wheninflation rateis exceedingthe11 percentage pointi.e. above the
secondthreshold level,thecoefficient (-0.06) is stillnegativeand significant.
of inflation
However,thisnegativeeffectis smallerthanthatwheninflation is in therangeof 6 to 11.
A one percentage increasein inflation, wheninflation rateis morethanelevenpercentage
point,a reduction of0.06 percentage pointis occurin realGDP growth rate.

withThresholds
Estimation (DependantVariableis GDP GrowthRate)
Effect
Variable Coefficient t-Statistic
Constant -10.16507 -1.596767
<6
Inflation 0.183643 0.517296
<=1 1
andInflation
Inflation>=6 -0.322854 -2.61 1079
> 11
Inflation -0.056985 -3.827330
Investment 3.449236 2.593879
Population 0.512724 2.622883
R-Squared = 0.37; DW = 2.02; Jarque-Bera= 0.09; Ramsey RESET Test (1, 39) = 0.34 [0.56].

The existenceof two thresholdlevels impliesthatinflation can be dividedinto

threeparts.As inflation risesfromzero to six percentagepoint, effecton economic
the
growthis negligibleor even positive.As inflationcrosses low threshold
the level,it has
and negativeimpacton theGDP up to a certainlevel.Wheninflation
significant crosses
secondthreshold level,themarginaladverseimpactof inflation on growthdiminishes.
The smallernegativecoefficient thattheinflation
illustrate growthrelationshipflattens
whentheeconomyhas highinflation. we
Intuitively, can say that
once inflation
exceeds a
thresholdlevel,all of thedamageto thefinancialsystemhas alreadybeendone,and then
perfectforesight dynamicscomes into being.When theseoccur,further increasesin
inflation
haveno additionaldetrimental on economicgrowth.3
effects

4.2. Inflationand InvestmentNexus

Theoreticalliteraturehas suggestedthatinvestment mightbe thechannelthatlink
inflationto economicgrowth.The linearmodel is estimatedby usingEquation5 to
uncovertherelationship betweeninflation and investment. Resultsindicatethatinflation
and negativeimpacton investment/GDP
has significant ratio.The coefficientof inflation
(-0.08) showsthata 1 percentage pointincreasein inflationwillcause a 0.08 percentage
pointreduction in investment.The firstlag of investmentis usedto controltheeconomic
conditionsin the last period which has significant and positiveimpacton current
investment (Table 5). This linearanalysis confirmsthe inflation-investment/GDP nexus
likeinflationand GDP growth.A dummyvariablerangingfrom1973 to 1981 is used to
makedatastableandnormal.
2Mubarik(2005) foundthatin Pakistan,inflation up to 9 percentis notharmful
whileHussain (2005)
between4 percentto 6 percentis feasible.
foundthatinflation
3Li(2006) also founda similarresultfordevelopingcountries.

This content downloaded from 195.78.108.147 on Mon, 23 Jun 2014 02:04:27 AM

All use subject to JSTOR Terms and Conditions
 ìqbal and Nawaz 87 1

LinearEstimationResults(DependantVariableis Investment/GDP
Ratio)
Variable Coefficient t-Statistic
Constant 7.682031 3.179267
Inflation -0.084268 -1.940828
Lag ofInvestment 0.589304 4.515194
Dummyfrom1973 to 1981 -0.945999 -1.699239
= 0.08; RamseyRESET Test( 1, 40) = 1. 13 [0.32].
R-Squared= 0.55; DW = 1.80; Jarque-Bera

Nonlinearmodelof investment and inflation is estimatedusingEquation6. By

applyingsameprocessas givenforinflation andgrowth, a singlethreshold at 7 percentis
estimatedbecause we cannot reject the null hypothesisof one thresholdagainst2
thresholds.Table 6 presentstheestimation resultsof theinflation-investment relationship
withthresholdeffects.The coefficient of inflation(0.05) is positivebut insignificant
when inflationis below the thresholdlevel. However,as inflationrates exceed the
thresholdlevel, the effectof inflationon the level of investment is negativeand
The
significant. coefficient of inflation
(-0.07) shows that a 1 percentagepointincrease
will cause a 0.07 percentage
in inflation point reduction in investmentas inflationrose
fromthreshold level.These evidencessuggestthatduringa periodof highinflation, the
levelof investment be adverselyaffectedbyinflation.

Estimation Effect(DependantVariableis Investment/GDP

withThresholds Ratio)
Variable Coefficient t-Statistic
Constant 7.878550 3.277259
Inflation<7 0.047665 0.608740
Inflation>=7 -0.067759 -1.949206
Lag of Investment 0.579052 4.459827
Dummyfrom1973 to 1981 -0.991533 -1.795522
R-Squared= 0.57; DW = 1.81;Jarque-Bera
= 0.06; RamseyRESET Test(1, 40) = 0.1 1 [0.74].

5. CONCLUSION AND POLICY OPTIONS

The objectiveof the presentstudyhas been twofold.Firstly,the impactof the
inflationrateon economicgrowthwiththepossibility of twothreshold levelforPakistan
using annual data from 1961 to 2008 has been examined and secondly,nonlinear
relationshipbetween inflation and investmentis also explored.
Inflation
and growthmodelssupporttheexistenceof a nonlinear with
relationship
twothresholds. Existenceof a doublethreshold dividestheinflationintothreecategories
i.e. low inflation,moderateinflationand high inflation.Inflation,below the first
threshold (6 percent),affectseconomicgrowth positively butinsignificantly;at moderate
ratesof inflation,
betweenthetwo threshold levels(between6 percentand 11 percent),
theeffectof inflationis negativeand significant; and at highratesof inflation,above the
second threshold(above 11 percent),the marginalimpactof additionalinflationon
economicgrowth diminishes butitis stillnegativeandsignificant.

This content downloaded from 195.78.108.147 on Mon, 23 Jun 2014 02:04:27 AM

All use subject to JSTOR Terms and Conditions
872 Investment, and EconomicGrowthNexus
Inflation

The second objectiveof the studyis to explorethe mechanismthroughwhich

inflationaffectslong-runeconomicgrowthin nonlinear Investment
settings. is one of the
possible channels through which inflation affects economic growth. The analysis
indicatesthenonlinear betweenthesetwovariableswithonlyone threshold
relationship
below thethreshold
at 7 percent.Rate of inflation level has a positivebut insignificant
impacton investment, whileabove thethreshold it has strongnegativeand significant
impacton theinvestment.
Thesefindings providesomeimportant policyimplications. The analysisshowsthat
itis desirableto keeptheinflation
belowthan6 percent and thereforecentralbankshould
concentrate on thosepolicieswhichkeeptheinflation ratebelowthefirst thresholdbecause
itmaybe helpfulfortheachievement ofrobusteconomicgrowth. Higherinflation thanthe
threshold wouldhaveadverseconsequencesforgrowth. Monetary policymustbe designed
Low inflation
to stabilisethepricesand curbinflation. is also helpfulforminimising the
uncertaintiesin thefinancialmarket in
which turnboost investment in thecountry. Better
coordination betweenmonetary and fiscalpolicesis requiredto achievebothobjectives,
i.e.,highandsustaineconomicgrowth andlow inflation.

Appendices

AppendixTable 1-A
Estimation
ofOne Threshold
Level(DependantVariableis GDP GrowthRate)
Variable Coefficient t-Statistic
Constant -13.94929 -2.110006
Inflation
<6 0.240981 0.924573
Inflation
>=6 -0.203435 -3.087819
Investment 0.639304 3.232172
Population 0.035727 2.572852
R-Squared= 0.32; DW = 2.14; Jarque-Bera
= 0.11; RamseyRESET Test(1, 40) = 1.19 [0.28].

AppendixTable 1-B
Estimation
ofOne Threshold
Level(DependantVariableis GDP GrowthRate)
Variable Coefficient t-Statistic
Constant -10.90925 -1.748677
Inflation
<=6 -0.561443 -2.596820
>6
Inflation -0.249473 -3.786838
Investment 0.515003 2.651303
Population 0.035711 2.725498
R-Squared= 0.36; DW = 2.01; Jarque-Bera
= 0.08; RamseyRESET Test(1 , 40) = 1.08 [0.30].

REFERENCES
Barro,R. (1991) EconomicGrowthin a Cross-section
of Countries.Quarterly
Journalof
Economics106:2, 407-443.
Barro,R. J. (1995) Inflationand Economic Growth.NationalBureau of Economic
Research.(WorkingPaper,No. 5326).

This content downloaded from 195.78.108.147 on Mon, 23 Jun 2014 02:04:27 AM

All use subject to JSTOR Terms and Conditions
 IqbalandNawaz 873

Boyd,J.H. and B. D. Smith(1998) CapitalMarketImperfections in a MonetaryGrowth

Model. EconomicTheoiy 11,241-273.
Boyd, J. H., R. E. Levine, and B. D. Smith(1996) Inflationand FinancialMarket
Performance. FederalReserveBankofCleveland.(WorkingPaper9617).
J.
Boyd, H., R. Levine,and B. D. Smith(2001) The Impactof Inflationon Financial
SectorPerformance. JournalofMonetary EconomicsAl, 221-248.
Bruno,M. and W. Easterly(1998) InflationCrises and Long-runGrowth.Journalof
Monetary Economics41:1, 3-26.
Bullard, J. and J. Keating(1995) The Long-runRelationshipbetweenInflationand
OutputinPost-warEconomies.JournalofMonetary Economics36, 477-496.
Drukker, D., P. Gomis-Porqueras,and P. Hernandez-
Verme(2005) ThresholdEffectsin
theRelationship betweenInflationand Growth:A New Panel-DataApproach.(IMF
StaffPapers).
Fama, E., and G. Schwert(1977) Asset Returnsand Inflation.Journalof Financial
Economics5, 115-146.
Faria,J.andF. Carneiro(2001) Does HighInflation Affect Growthin theLong and Short
Run?JournalofAppliedEconomics1.
Fisher,S. (1993) The Role of Macroeconomic Factorsin Growth.Journalof Monetary
Economics32, 485-512.
Gillman,M. and A. Nakov (2004) GrangerCausalityof theInflation-Growth Mirrorin
AccessionCountries.EconomicsofTransition 12:4, 653-681.
Gultekin, N. (1983) StockMarketReturnsandInflation: EvidencefromOtherCountries.
JournalofFinance 38, 49-65.
Hansen,В. (2000) SampleSplitting andThresholdEstimation. Econometrica68:3, 575-
603.
Hellerstein,R. (1997) The ImpactofInflation.
RegionalReview7:1.
Hussain,M. (2005) Inflationand Growth:Estimationof ThresholdPointforPakistan.
PakistanBusinessReview7:3.
Huybens,E. and B. Smith(1998) FinancialMarketFrictions,MonetaryPolicy, and
CapitalAccumulationin a Small Open Economy.Journalof EconomicTheory81,
353-400.
Huybens,E. and B. Smith (1999) Inflation,Financial Markets,and Long-runReal
Activity.JournalofMonetary Economics43,283-315.
Khan,M. S., A. S. Senhadji,andB. D. Smith(2001) InflationandFinancialDepth. (IMF
Working Paper,WP/01/44).
Khan,M. S. and S. A. Senhadji(2001) ThresholdEffectsin theRelationshipbetween
Inflation andGrowth.IMF StaffPapers 48:1, 1-21.
Lee, С. С and S. Y. Wong (2005) Inflationary ThresholdEffectsin theRelationship
betweenFinancialDevelopmentand EconomicGrowth:EvidencefromTaiwan and
Japan.JournalofEconomicDevelopment30, 49-69.
Li, M. (2006) Inflationand Economic Growth:ThresholdEffectsand Transmission
Mechanisms.University of Alberta.(WorkingPapers).
Lucas, R. (1973) Some InternationalEvidenceon Output-InflationTradeoffs.American
EconomicReview63, 326-334.

This content downloaded from 195.78.108.147 on Mon, 23 Jun 2014 02:04:27 AM

All use subject to JSTOR Terms and Conditions
874 Investment, and EconomicGrowthNexus
Inflation

Malla, S. (1997) Inflationand EconomicGrowth:Evidencefroma GrowthEquation.

Department ofEconomics,University ofHawai'I at Monoa. (Mimeographed.).
Mallik, G. and A. Chowdhury(2001) Inflationand EconomicGrowth:Evidence from
SouthAsianCountries.AsianPacificDevelopment Journal8:1, 123-135.
Mubarik,Y. A. (2005) Inflationand Growth:An Estimateof the ThresholdLevel of
InflationinPakistan.StateBankofPakistanResearchBulletin1:1, 35-44.
Munir,Q. and K. Mansur(2009) Non-linearity betweenInflation Rate and GDP Growth
inMalaysia. EconomicsBulletin29:3, 1555-1569.
Nelson,С (1976) Inflation and Ratesof Returnon CommonStocks.Journalof Finance
31,471-483.
Pakistan,Governmentof (2010) Pakistan Economic Survey(2009-10). Islamabad:
EconomicAdvisor'sWing,Ministry ofFinance.
X.
Sala-i-Martin, (1997) I JustRan Two Million Regressions.AmericanEconomic
Review87, 178-183.
Sarei,M. (1996) NonlinearEffectsof Inflation on EconomicGrowth.IMF StaffPapers
43, 199-215.
Sergii, P. (2009) Inflationand Economic Growth:The Non-LinearRelationship.
EvidencefromCIS Countries. MA ThesisinEconomics,KyivSchoolofEconomics.

This content downloaded from 195.78.108.147 on Mon, 23 Jun 2014 02:04:27 AM

All use subject to JSTOR Terms and Conditions