Finance and Development: Institutional and Policy Alternatives to Financial Liberalization

Theory
Author(s): Philip Arestis, Machiko Nissanke and Howard Stein
Source: Eastern Economic Journal, Vol. 31, No. 2 (Spring, 2005), pp. 245-263
Published by: Palgrave Macmillan Journals
Stable URL: http://www.jstor.org/stable/40315489 .
Accessed: 09/01/2014 11:37

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.

Palgrave Macmillan Journals is collaborating with JSTOR to digitize, preserve and extend access to Eastern
Economic Journal.

http://www.jstor.org

This content downloaded from 168.176.5.118 on Thu, 9 Jan 2014 11:37:32 AM

All use subject to JSTOR Terms and Conditions
 FINANCE AND DEVELOPMENT:

INSTITUTIONAL AND POLICY ALTERNATIVES TO

FINANCIAL LIBERALIZATION THEORY

Philip Arestis
ofCambridge
University

MachikoNissanke
ofLondon
University

and

HowardStein
ofMichigan
University

INTRODUCTION

On August7, 2002,Brazilreceiveda $30 billiondollarpackagethatwas thelarg-

estloangrantedinInternational Monetary Fund(IMF) history andbrought totalIMF
lending to the countryto $63 billion since 1998. The bailout was simply the latest
chapter in a recent of
saga unprecedented financial instability and crisisaffecting
virtuallyeveryregionoftheglobaleconomymercilessly. Since 1997therehave also
beenmajorcrisesinArgentina, Ecuador,Thailand,Russia,Uruguay,Columbia,Indo-
nesia, Kenya,and Korea. This instability has been associatedwithrapidfinancial
liberalizationwithoutexception.Forexample,in Korea,thecrisisofNovember1997
followed thederegulation ofinterestrates,theopeningofthecapitalmarket,foreign
exchangeliberalization,thegranting ofnewbankinglicenses,and thedismantling of
government monitoringmechanisms thatwere part ofthe policy loan system.
Thepost-1997patternofliberalization leadingtocrisisis a continuation ofearlier
trendsthathave becomeubiquitous in Latin America and elsewhere. In 1989,Ven-
ezuela implemented financialliberalizationas partofa standardorthodox IMF and
WorldBankadjustment package and sectoral loan. Policiesrelated tofinance included
the removalofquotas forpriority lending, the liberalization of interestrates,the
openingup ofthebankingsectorto foreign ownership, and the privatizationofcom-
mercialbanks.By 1994,thebankingsystemwas in a full-fledged meltdown. Between
January 1994 and August 1995, 17 financial institutions failed,encompassing 60 per-
centofthetotalassets ofthefinancialsystemand 50 percentofthedepositsand an

Philip Arestis: CambridgeCentreforEconomicand Public Policy,Departmentof Land Economy,

ofCambridge,19 SilverStreet,CambridgeCB3 9EP, UK. E-mail:pa267@cam.ac.uk.
University

EasternEconomicJournal,Vol.31,No. 2, Spring2005
245

This content downloaded from 168.176.5.118 on Thu, 9 Jan 2014 11:37:32 AM

All use subject to JSTOR Terms and Conditions
246 EASTERNECONOMICJOURNAL

estimated20 percentoftheGDP tocleanup [Vera2002].In Mexico,liberalization and

privatizationintheearly1990shas provenalso tobe enormously costly.By 1999,the
costofthegovernment intervention reached$65 billionorroughly17 percentofthe
1998GDP [FinancialTimes1999]. Othercountries in differentregions,including the
CaribbeanandAfrica, havefollowed orthodoxcourses of financial with
liberalization,
verysimilarresults(see, forexample,Stein,Ajakaiye,and Lewis [2002]and Stein,
Cuesta,and McLennon[2002],fora numberofindividualcases in theseregions).
These examplesare notisolatedexceptionsbut morethe rule.Demirguc-Kunt
and Detragiache[1999] surveybankingcrisesin 53 countriescoveringthe period
between1980and 1995and findthat78 percentofall criseswerelinkedtoperiodsof
financialliberalization(see also WorldBank [2001,83] Table 2.1 fora comprehensive
listofthecostsinvolvedin theseand othercrises,amounting to 50 percentofGDP in
somecases). Giventheubiquityofthesecrises,whydo governments pursuefinancial
Answerstothisquestionincludetheinstitutionalization
liberalization? inrecentdecades
ofnormsof"acceptable"financialpolicies,theperceivedpotentialgainsofattracting
privatecapitalinflows, thenatureofglobalsystemsand theasymmetric powerrela-
tionsembeddedinglobalstructures thatdelimitnation-state options,and,finally,the
expectedgains arisingfromthe economiclogicembeddedin the theoryunderlying
financialliberalization.
Thispaperwillfocuson thelatterquestion,arguingthatfinancialliberalization
policyis builton shakytheoreticalpremises.We suggestthatthe recentfinancial
crisesthat have hit so manydevelopingcountriesand transitionaleconomiesare
inducednotmerelybytheinappropriate sequencing,pacing,ortimingofinternaland
externalfinancialliberalization, buttheyare theinevitableoutcomeofadoptingthe
policythatis based on a veryshallowunderstanding ofthe dynamicrelationships
betweenfinanceand economicdevelopment. In ourview,financialtransformation in
the imageofMcKinnon-Shaw has engenderedwidespreadbankingcrisesprecisely
becauseoftheweakfoundations ofthetheory.Ourviewis thatbad theorygivesrise
topoliciesthatgiveriseto crises.In contrast, theproponents ofthefinancialliberal-
izationthesissee financialcrisesas somehowirrelevant totheirtheoryand thepolicy
thatit has inspired.This is paradoxicalin thatproponents argue,whencrisestake
place,that more oftheir policyprescriptions should thecure,whenit is precisely
be
thosepoliciesthatcausedthecrisesinthefirstinstance.Polanyi[1957]describesthis
paradoxverywellwhenhe arguesthattheapologists, thedefenders offinancialmar-
ketliberalization,"arerepeatingin endlessvariationsthatbutforthepoliciesadvo-
catedbyits critics,liberalismwouldhave deliveredthegoods;thatnotthecompeti-
tivesystemand the self-regulating market,but interference withthatsystemand
interventions withthatmarketare responsible forourills"(p. 143).Hence,thepaper
aims at sketchingan alternativetheoreticalperspective byexamininginstitutional
for
requirements buildingand transforming financialsystemsforeconomicdevelop-
ment.The focusis howto enhancetheoperationaland developmental roleofbanks
and otherfinancialentitieswithinthebroaderfinancialsystem.The aim is totrans-
form themeso-level, whichmediatesmicroandmacrofinancial relationsinan economy.
We are notlookingspecifically at themicrolevel,butat factorsthataffect themicro
dimensions.

This content downloaded from 168.176.5.118 on Thu, 9 Jan 2014 11:37:32 AM

All use subject to JSTOR Terms and Conditions
 TO FINANCIALLIBERALIZATION
ALTERNATIVES THEORY 247

The paper is structuredas follows:We firstgive a briefpresentationofmain

thesesin financialliberalizationtheory(section2). Thisis followedbya critiquefrom
boththeoretical(sections 3 and 4) and empirical(section5) pointsofview.Section6
offersan alternativeperspectiveon financialtransformation moreconsistentwith
economicdevelopment (thatis, consistentwithbothan institutionalist theoryofeco-
nomicdevelopment and withtherealityoftheinstitutional structureofdeveloping
economies)thatdrawson a ratherdifferent setoftheoretical toolsand ideas. Finally,
section7 summarizesand concludes.

FINANCIAL LIBERALIZATION THEORY: RETURNING TO THE ORIGINAL

TEXTS

Financialliberalizationtheory has itsoriginsintheworkofMcKinnon[1973]and

Shaw [1973].It was Patrick[1966],however, whopublishedtheseminalworkonthe
relationshipbetweenfinancialdevelopment and economicgrowth.He hypothesized
twopossiblerelationships, a "demand-following" approach,inwhichfinancialdevelop-
mentarisesas theeconomydevelops,and a "supply-leading" phenomenon, in which
thewidespreadexpansionoffinancialinstitutions leads toeconomicgrowth. Priorto
Patrick[1966]therehad been a greatdeal ofdebateon theissue,withcontributors
ranging fromBagehot[1873]andSchumpeter [1912],whosupported thesupply-leading
view, toRobinson [1952],who voicedstrongsupport for thedemand-following approach,
tomention onlythemainprotagonists. Thefinancial liberalization
schoolleanstowards
thesupply-leading relationship betweengrowthand development [McKinnon, 1973].
Theargument arisesoutofa highlysimplified worldwithoutfinancialintermediaries,
wherebythe purchaseofcapitalcan onlyarise fromself-finance; forwhenan indi-
vidualwhois limitedtoself-finance wishesto"purchasephysicalcapitalofa typethat
is differentfromhis own output...He may storeinventoriesofhis own outputfor
eventualsale whenthecapitalassetsare acquiredorhe maysteadilyaccumulatecash
balancesforthesamepurpose"McKinnon,1973,57].
In theirview,priorsavingsare seen to help the accumulationprocess.Conse-
quently,accordingto thisview,thekeyis to altertheincentivesbetweenconsump-
tionand saving.Followingclassicaleconomics, interestrates are seen as providing
thereturnforthischoice.1 Wheninterestratesarekeptartificially low,theresultwill
For
be shallowfinancing. example, Shaw [1973,8] argues:"Deepeningimpliesthat
interestratesmustreportmoreaccuratelytheopportunities thatexistforsubstitu-
tionofinvestment forcurrentconsumption and the disinclination ofconsumersto
wait. Real rates ofinterestare highwherefinanceis deepening."UnlikeKeynes
[1936],inwhichinterest ratesaffectthedemandforand supplyofmoney, Shaw [1973]
followstheclassicalmodel,inwhichtheequilibrium betweensavingsand investment
is determined by interestrates. A rise in real interestrates increasesthe flowof
savingsand reducestheexcessdemand.Ratesofreturnonholdingmoneyalso playa
rolein increasinginvestment levels[McKinnon, 1973].
In thisparadigm,thedesiretoholdmoneyis also positively affectedbytherateof
returnon capital,contrary to theportfolio approach(whichhas a negativerelation-
ship).Withhigherinvestment, therewillalso be a relatedimprovement inthequality

This content downloaded from 168.176.5.118 on Thu, 9 Jan 2014 11:37:32 AM

All use subject to JSTOR Terms and Conditions
248 EASTERNECONOMICJOURNAL

ofinvestment and a riseinthesavingslevelsallocatedthrough themarket.Rationing

ofcreditreducesthequalityofintermediation, however, withnegativeconsequences
to investment: "Rationingis expensiveto administer.It is vulnerableto corruption
andconspiracy individing betweenborrowers andofficers oftheintermediary monopoly
rentthatarisesfromthedifference betweenlow,regulatedloanrateand themarket-
clearingrate.Borrowers whosimplydo notrepayloans and keep theirplace in the
rationqueuebyextending maturitiescan frustrate it.Therationing processdiscrimi-
nates poorlyamonginvestment and
opportunities... the social cost of thismisalloca-
tionis suggestedbythehighincremental ratiosofinvestment tooutputthatlagging
economiesreport"[Shaw,1973,86].
It is further arguedthatcompetition throughprivateownershipcan shrinkthe
difference betweendepositandloanrates,encouraging "optimal" agreements between
banks,and amongborrowers andlenders,inturnincreasingtheefficiency ofinterme-
diation.Moreover, "fragmentation" indeveloping countries: "...inthesensethatfirms
and householdsface...different effective
prices for land, labor, capitaland produced
commodities. ..hasbeenlargelytheresultofgovernment policy"[McKinnon, 1973,5-7].
Reversing fragmentation bycreatinga singlecapitalmarketthrough theretractionof
state intervention is thenregardedas the sine qua non ofeconomicdevelopment:
"Arbitrary measuresto introducemoderntechnology via tariffs,or to increasethe
rateofcapitalaccumulation on
byrelying foreign aid or domestic forced saving,will
notnecessarily lead toeconomic development. Thusitis hypothesized thatunification
ofthecapitalmarket,whichsharplyincreasesratesofreturnto domesticsaversby
widening exploitable investmentopportunities,is essentialforeliminating otherforms
offragmentation" [McKinnon, 1973,9].2Byimplication, McKinnon [1973]and Shaw
[1973]supporttheliberalizationofthecapitalaccountin orderto providea unified
capital market forprivate decision makers to undertake utility-maximizing
intertemporal choice.3

A THEORETICAL COMMENT ON THE FINANCIAL LIBERALIZATION

HYPOTHESIS

Thereare severalfundamental problems withthefinancialliberalizationhypoth-

esis. To beginwith,theargumentis developedin an almostRobinsonCrusoeframe-
work,in whichall investment is self-financing.Thisis abstractedfromthecomplexi-
tiesofmoneyas a socialinstitution. In reality,moneyis bynaturesociallyembedded.
Theholdingofmoneyevenin thesimpleruralsettingdiscussedbyMcKinnon[1973],
is subjectto socialobligationsand constraints, and notsimplydrivenbyinvestment
needs,theproductivity ofcapital,and real returnon holdingmoney.Moreover, the
is It
presentation contradictory. justis not possible totalkabout self-financing no
as if
otherfinancialoptionsexist,and discussa returntoholdingmoneyas some"weighted
averageofnominalinterestratesofall formsofdeposits"[McKinnon, 1973,39],which
presupposestheexistenceofa sophisticated financialsystem.
McKinnon[1973]is, ofcourse,determined to showa positiverelationshipamong
higherinterestrates,financialdevelopment, and investment and growth.He is criti-
cal oftheneoclassicalportfolio approach,inwhichindividualschoosebetweenmoney

This content downloaded from 168.176.5.118 on Thu, 9 Jan 2014 11:37:32 AM

All use subject to JSTOR Terms and Conditions
 TO FINANCIALLIBERALIZATION
ALTERNATIVES THEORY 249

and assets,sinceit positsa negativerelationship betweenthe demandforreal bal-

ancesand thereturnonnonfinancial assets.He reallyis notescapingfroma portfolio
approach,however;he simplyredefines it as a choicebetweena noninterest-bearing
asset(own stocks of products) money(whichis interestbearing).Byintroducing
and a
laggedtimedimensionintothe choicebetweenmoneybalances and nonmonetary
assets,holdingtheportfolio as a storeofvaluebecomesthefocalpointinhis analysis.
The totalofthisportfolio is relatedto thedesiredrate offuturecapitalinvestment.
Thehighertheexpectedrateofreturn, thegreaterthedesiretoholdmoneybalances.
McKinnon[1973]wouldarguethat,inproperly operatingcapitalmarketswithout
fragmentation, all that is necessary is a single unifying marginal rate ofreturn.
Economy-wide choices then simplyoperate like individual choices.The problemwith
thisperception is thatcapitalmarketshave neveroperatedin thismanner.Markets
are alwaysfragmented and repletewithdifferent levelsofriskand uncertainty even
in the mostadvancedeconomies.Perceptionsofrates ofreturnvarynotonlywith
differenttypesoffinancialvehiclesbutalso byindividuals. Anindividualona particu-
lar projectperceivesinterestrates paid by governments verydifferently fromthe
anticipated futurereturn.Nothingwillautomatically unifythemintoa singleconcat-
enatingvisionofa futurepayout.
Thetheoryoffinancialliberalization also reliesontheassumptionofthecompeti-
tivemodel.The divergence betweenthefinancialworldand thecompetitive modelis
profound, however.Financeis repletewithasymmetries ofriskand information that
are less evidentin goodsmarkets.Stiglitz[1989,1994] pointsto a hostofmarket
imperfections embeddedinfinancialmarketsthatgobeyondthewell-known issuesof
moralhazard and adverseselection.Theyinclude:1) the largedivergence between
thesocialand privatecostsofbankfailures;2) thepublicgoodnatureofthesolvency
ofinstitutions that are likelyto be undersupplied;3) the externality effectofthe
presenceofa fewbad banks on the confidence of the sector;and 4) the divergence
betweentheprivate(thosewithrapidturnover) and social(projectsare likelytohave
longerturnover periodsandhigherrisk)benefits ofloans.Furthermore, and contrary
tothestandardassumptionofmarkets,financialmarketswillnotbe Paretoefficient
(wherethepricerepresents themarginalbenefit tothebuyerand themarginalcostto
thesupplier),sincetheborrower (buyer)willingto pay themostfora loan,maynot
the
provide highestprofit to the lender (seller).An important recentcontribution by
theimperfect information schoolgoes to theheartoffinancialliberalizationthesis,
thatis,theenhanceddegreeofcompetition itcreates[Hellman,Murdock, and Stiglitz,
2000].Morecompetition, according tothiscontribution, erodesfranchise value,which
reducesincentivesforprudentialbehavior,therebysubstantially increasingriskin
thesystem.The problemoffinancein developing countriesis muchdeeperand more
multifaceted, however, thansuggestedbytheimperfect information school,as argued
below.
Thefocalpointoffinancialliberalization is onretracting thesourcesofrepression
thathavedistorted thesignalingeffect ofinterestrates.Eitherthesignalsthemselves
are disruptedbythe"fragmentation" createdbygovernment policyorthesignalsare
misreaddue to thestateownershippatternoffinancialorganizations. On thelatter
point,it is an axiomaticbeliefthatthe publicownershipor controlalwaysleads to

This content downloaded from 168.176.5.118 on Thu, 9 Jan 2014 11:37:32 AM

All use subject to JSTOR Terms and Conditions
250 EASTERNECONOMICJOURNAL

politicalorpatronageinfluences, whileprivateownerswillreactto thesignalswith

Paretoefficient decisionson the allocationofcredit.Stiglitz[1989,1994],and the
imperfect informationschool,recognize thatfinancialmarketslefttotheirowndevices
aregenerally of
incapable providing correctsignals.Stateintervention shouldbe aimed
at creatingmoderateformsoffinancialrepression to alterthe signalsthatlead to
moresociallyoptimaloutcomes. Theproblem is that,giventhepoorlydevelopednature
offinancialmarketsin developingcountries, evenadjustedsignalingmaynothave
thedesiredresultsand mayevenlead tounintendedconsequences.In countrieslike
Nigeria- whereregulationis poor,normsoftrustare notdeveloped,and a military
government handsoutbankinglicensestomilitary officials- subsidizing interestrates
wouldhave donelittleto reversethe financialchaos createdbyliberalizationafter
1986.Whileinterestratesare important, theyare onlyone dimensionoftheincen-
tivesanddisincentives thatinfluence thedecision-making processinfinancial systems.
The McKinnon-Shaw worldoffinanceis one wherefinancialintermediaries via
marketsfordepositing andlendingsimplysetinterestratestobalancethesupplyand
demandforsavingsofborrowers and depositors. In thisworld,thereare threeactors
in twoexchangeswiththedifference in thepriceofthetwoexchangessimplyreflect-
ingthecostofintermediation (whichwillbe keptlowwithsufficient competition). One
only needs the unfettered of
operation self-seeking atomistic individuals to arriveat
Paretooptimality. Thereis noneedforinstitutions. However, in the real world,inter-
estratesandincentivesare onlyonedimensionoffinance, whichis a complexinstitu-
tionembeddedin a broadersystemofnonfinancial institutions. As we willsee below,
many of the econometric studies have difficultypinpointing clear causalitybetween
financeand development. We wouldarguethisis preciselybecauseoftheintertwin-
ingand interaction betweenthedevelopment offinancialinstitutions andthefinance
ofdeveloping institutions.
Theseissueswillbe exploredbelow.
Financialtransformation in theimageofMcKinnon-Shaw has engendered wide-
spreadbanking crisespreciselybecause ofthe weak foundations ofthe theory.How-
ever,McKinnon[1993]has arguedthatit is not a problemwiththe theoryor the
policiesarisingfromthetheory, butone ofsequencing,particularly whenderegula-
tionis introducedbeforemacroeconomic stabilization is completed. Weturnouratten-
tionnextto thisissue.

SEQUENCING, MACROSTABILITY, AND POLITICAL ECONOMY

McKinnon[1993]attemptedto accountforinstitutional capabilitiesand weak-

nesses under"theoptimumorderofeconomicliberalisation." He arguesthat"How
fiscal,monetary, andforeignexchangepoliciesaresequencedis ofcriticalimportance.
Government cannot,and perhapsshouldnot,undertakeall liberalizing measuressi-
multaneously. Instead,thereis an order
'optimal' of economicliberalisation, which
mayvaryfordifferent liberalizingeconomiesdependingon theirinitialconditions"
[McKinnon' 1993,4] Thisoptimalorderwouldbeginwiththecontrol ofinflation, lead-
ingtothe deregulationofinterest and
rates,bankingprivatization commercialization,
foreign exchangerateunification, tradeliberalization,
and onlythenopeningup the
capitalaccount.Caprio,Atiyas,andHanson[1994]reviewfinancialreforms in a num-
berofprimarily developingcountrieswiththeexperienceofsix countriesstudiedat

This content downloaded from 168.176.5.118 on Thu, 9 Jan 2014 11:37:32 AM

All use subject to JSTOR Terms and Conditions
 ALTERNATIVES
TO FINANCIALLIBERALIZATION
THEORY 251

somedepthand length.Theyconcludethatmanagingthereform processratherthan

adoptinga laissez-faire stanceis important, andthatsequencingalongwiththeinitial
conditions infinanceandmacroeconomic stability are criticalelementsinimplement-
ingsuccessfully financial reforms. It is thusrecommended nowthatgradualfinancial
liberalizationis to be preferred. In thisgradualprocess,a "sequencingoffinancial
liberalization" is recommended, emphasizing theachievement ofstabilityinthebroader
macroeconomic environment and adequatebanksupervision withinwhichfinancial
reforms are to be undertaken[Choand Khatkhate,1989a; McKinnon,1988; Sachs,
1988;VillanuevaandMirakhor, 1990]. Employing credibilityarguments, Calvo[1988]
and Rodrik[1987]suggesta narrowfocusofreforms withfinancialliberalization left
as last.
The argumentembeddedin theorderoffinancialliberalization has increasingly
beenchallenged. A morerecentliterature has indicatedthatfinancialliberalization in
has
anysequence engendered the same Even
difficulties. where the "correct"sequencing
tookplace (forexample,Chile),wheretradeliberalizationhad takenplace before
financialliberalization, not muchsuccesscan be reported.It is also truein those
cases,likeUruguay,wherethe"reverse"sequencingtookplace- financialliberaliza-
tionbeforetradeliberalization - thatthe experiencewas verymuchthe same as in
Chile.The experience with financial liberalization, in bothdevelopedand developing
countries, in the 1980s and 1990s suggestsa markedincreasein thefrequency and
severity of financial crisesirrespective ofthe order of sequencing [Lindgreen, Garcia,
and Saal, 1996; Demirguc-Kunt and Detragiache,1998; Grabel,1995; Arestisand
Demetriades, 1997].
Moreover,the studyby Weller[2001]surveys26 countriesrepresenting every
region ofthe world to evaluate the relationship between financial liberalizationand
crisis.Amongthesampleare manyvariationsin thesequencingofpolicies.Drawing
ontheworkofMinsky[1984], a numberofhypotheses abouttherelationship between
financial liberalization andthebankingcrisesthattypically followaretested.Internal
and externalderegulation fostersfinancialfragility byencouraging flowsto specula-
tiveventures.Assetinflationcan raise the collateralofborrowers and increasethe
euphoria.Speculationbecomesself-fulfilling as greaterflowsintospeculationin turn
perpetuatethespeculativeboom.Short-term capitalinflowsafterliberalization can
raisetheexchangeratesleadingtodeterioration inthecurrentaccount,withimplica-
tionsto real growthin sectorslike industry.This is exacerbatedby the shiftaway
from investment finance tospeculation. Eventually, assetpricesbegintodeflate, default
riskrises,and maturity riskincreasesas short-term outflows increasein responseto
theworsening balancesheetofbanks.Ultimately, theeconomy is markedbya risein
interest rates, credit contraction,import and
priceinflation, depleteddomestic demand.
Weller's[2001]resultsconfirm thesehypotheses, especiallythatofthegrowthof
financialfragility afterfinancialliberalization.Indeed,morespeculativefinancing
greatlyenhancesthechancesofa bankingcrisisafterfinancialliberalization. Ofpar-
ticularinterest is howlongfinancial liberalization willcontinue toincreasethechances
offuturefinancialcrises.It has beenfrequently arguedthatfinancialliberalization
mightlead to short-term dislocationbut it will be beneficialin the longterm.The
resultsindicatethatthelikelihoodofa bankingcrisisdoes notdisappear,however,

This content downloaded from 168.176.5.118 on Thu, 9 Jan 2014 11:37:32 AM

All use subject to JSTOR Terms and Conditions
252 EASTERNECONOMICJOURNAL

butit in factincreasesovertime.The studybyArestisand Glickman[2002]reaches

similarconclusions inthecaseoftheSouthEastAsiancrisis,supporting thesehypotheses.
Otherstudieshavealso confirmed thelackofsignificance of"sequencing." Arestis,
Demetriades, and Fattouh [2003]survey theliteratureand offertheir own empirical
investigation, and findno evidencethatvaryingthesequenceoffinancialliberaliza-
tionalongMcKinnon's[1993]optimallinesleads toanydifferent results.In linewith
Weller's[2001]findings, thereis strongevidenceofincreasingfrequency and severity
offinancialcrisesinthewakeofliberalization. A WorldBankeconometric studyofthe
relationship betweenbankingcrisesand a seriesofexplanatory factors [Demirguc-
Kuntand Detragiache,1999],foundthemtobe highlycorrelated withfinancialliber-
alizationpoliciesevenwhencontrolling forfactorslikethesequencingrecommended
by McKinnon [1993]. Other problems with "sequencing"includethe questionofits
timing, thatis, howdo policymakersknowwhenit is timetomovefromone stageto
thenext?Sequencing,ofcourse,can easilycreateinertiain reform, ifindeedreform
is necessary.
The notionoftheexistenceofsomeoptimalsequencecan be questionedforother
reasons.Anyevaluationoffinancialliberalization shouldnotbe merelybased oneco-
nomiccriteria,but shouldalso containelementsfromthe realmofpolitics(Armijo,
1999).Thereis also theratherflawedbeliefthatorthodox financialliberalization will
lead toan improvement thatis growth enhancing. Underhill [1997], contrast, far
in is
less sanguineconcerning theprospectofeconomicgainsfromfollowing someoptimal
sequenceoffinancial transformation. Bynature,anychangeinexisting financialarrange-
mentscreatesnew winnersand loserswithimplications to the politicalrealm.He
characterizes thenewchangesin a financialsystemas constituting desegmentation
(unifying variousbranchesoffinanceintoa singlebranch),marketization (domestic
financialmarketliberalization) and transnationalization(integrating financialmar-
ketsacrossnationalboundaries).The actualoutcomeofchangevariesin accordance
withthe relativestrengthofthe constituent membersofthe financialcommunity
(government players,including centralbankersandregulators, nationalandmultina-
tionalbanks,pressuresfrominternational financialinstitutions,
strength ofprivate
involvement in privatization, etc.).The finaloutcomeis unpredictable withno guar-
anteeofanyeconomicimprovements and involvesconsiderable downsideriskfollow-
inganysequence.
ExamplesofUnderhill's[1997]observation abound.In Nigeria,underthemilitary
rule ofBabingida,applicationsforall new bankinglicenseswerereviewedby the
President'sofficeand FederalExecutiveCouncilcontrolled bythemilitary.Retired
militaryofficials withnobankingexperience wereinstrumental inobtaining banking
licenseswithoutthe properproceduresbeingfollowed.Centralbank officials who
wereinterviewed indicatedthenearimpossibility ofturningdownapplicationsinthe
climateofmilitary control.It is hardtoconceiveofanysequenceofreform working in
thisclimateor changingtheresultsofthenowwell-documented bankingcrisisthat
followed.In addition,politicsplayeda majorrolein the attemptto reorganizethe
financialsystemafterthebankingcrisis.In Nigeria,after1996,GeneralAbachaused
theopportunity topunishpoliticalopponents and challengetheindependence ofsome
businessgroups[Lewisand Stein,2002].

This content downloaded from 168.176.5.118 on Thu, 9 Jan 2014 11:37:32 AM

All use subject to JSTOR Terms and Conditions
 TO FINANCIALLIBERALIZATION
ALTERNATIVES THEORY 253

Finally,somehave suggestedthatMcKinnon[1993]was implicitly arguingfor

greaterprudenceandgradualisminfinancialreform. Based onourabovearguments,
however, thattheproblemis neithertimealonenorsequence,but
itis ourcontention
linkedto a fundamental misconceptualizationofthe underlying
theorybehindthe
financialliberalization oftheorder.
thesis,irrespective

EMPIRICAL STUDIES OF THE FINANCIAL LIBERALIZATION

HYPOTHESIS

The importance ofexaminingthebroaderconditions beforetransforming finan-

cial systemsgoesbeyondtheframework ofpoliticaleconomy totheunderlying struc-
turesofany developingeconomy. While there is broad agreement on the complica-
tionsofundertaking liberalization inthemidstofmacroinstability, whereeconomies
are structurallyweak,subjectto thevicissitudesofinternational commodity prices
and shiftingfinancialflowswithfewstabilizingreserves,instability is likelytobe the
ruleratherthantheexception. Evenwhenbankersare completely honestandregula-
torysystemsare in place,macroinstability or thelikelihoodofa futureoccurrence
will encouragebankersto holdreservesin government paper and to limitloans to
short-term duration. In places like Venezuela,Nigeria, and Russia,financialliberal-
izationledtoa decreaseinthedurationofloansandincreaseintheholdingofgovern-
mentpaper [Stein,Ajakaiye,and Lewis,2002]. In Tanzania, althoughmostofthe
bankingsystemis nowin foreign privatehandsand inflation has fallenbelow5 per-
cent, banks are sitting on 60 percent excess reserves with much ofit beingheld in
government paper[TheAfrican, 2002].
Giventhecentralroleofthefinancialliberalization hypothesis in liberalization
programs and the evolvingliterature, there has been a proliferation ofeconometric
testingofthetheory and attendant policycorrectives. In general,despitecontinuing
effortstodiscernthepostulatedpositiverelationship betweenfinancialliberalization
andgrowth bymainstream economists through cross-country studies,4 supportforthe
financialliberalization hypothesis is not verystrong and there is growingevidence
confirming Keynes'[1936]viewofthelinkageamong interest rates,investment, and
savings.
We beginwiththeliteraturethatteststhefinance-to-development causationor
supply-leadingrelationship. Habibullah [1999] surveys the literature for the financial-
led growthhypothesis and undertakes his own testing,finding little evidence to sup-
portthis hypothesis.5 Akinboade[1998] also uses Grangercausalitytestingin a
cointegrationframework on Botswanadata coveringtheperiod1972to 1995,tofind
clearbi-directionalcausality.Similarly, Sahoo,Geethanjali,andKamaiah [2001]note
in
thatinthecase ofIndia the 1970s highlevelsofsavingsdidnotlead tohigherlevels
ofgrowth.Later periodsofgrowthseemedto occurwithoutan appreciablerise in
savings.To examinethis moresystematically, theyapplycausalitytestingofreal
savings and realGDP data from 1950/51 to 1998/99. Theymanagetoestablisha strong
one-waylinkage from to
growth savings and therebyrefutetheproposition thatsav-
ingswas theengineofgrowthin thecase ofIndia.

This content downloaded from 168.176.5.118 on Thu, 9 Jan 2014 11:37:32 AM

All use subject to JSTOR Terms and Conditions
254 EASTERNECONOMICJOURNAL

The growth-to-savings causalityhas also been confirmed bysomeWorldBank's

studies.Countriesin Asia wereverysuccessfulin mobilizing savings,yetstrongevi-
denceindicatesthatinvestment and growthled to savingsratherthan savingsto
growth. It was notsavings thatled tothephenomenal investment and growth ratesof
recentdecades,buttheriseofincomethatincreasedsavings.The WorldBank [1993]
foundthecausationfromgrowth tosavingsinfivecountries (Indonesia,Japan,Korea,
Thailand, and Taiwan) and ambiguity in two (HongKong and Malaysia);inoneitwas
due tootherfactors(in Singapore,thestateprovident fundwas salient).
Surveysoftheliteratureindicatelittleor no evidenceofa positiverelationship
betweeninterestratesand savings[Dornbuschand Reynoso,1993]. Variousecono-
metrictestingfromAsia, LatinAmerica,and Africaalso confirms thelack ofcorre-
spondence between interest and savings, even amongstrongproponents oforthodoxy
likeFry[1988]as wellas manyothers(forexample,Giovannini[1985],Gupta[1987],
Cho and Khatkhate[1989b], GonzalezArrieta[1988],De Melo and Tybout[1986],
WarmanandThirlwall[1994],Oshikoya[1992],Taiwo[1992],andReichel[1991]).The
overwhelming evidencehas evenencouraged McKinnon[1993]toabandonthehigher-
interest-to-prior-savings argument in favorofa risein the"quality" ofinvestment after
liberalization.
Worseforthefinancialliberalization thesis,thereare studiesthathave actually
showna negativerelationship between interest ratesand savings.Matsheka[1998]
teststherelationship betweenthevariablesfortheperiod1976to 1995in thecase of
Botswana.A negativeand significant relationship betweenrealdepositinterestrates
and thelogofrealdomesticsavingsis foundtoprevail.The samestudyalso examines
theeffect ofdepositrateson privatesavinglevels(sincesavingsare overwhelmingly
dominatedbythegovernment in themineraleconomyofBotswana)and stillfindsit
negativeand significant. Matsheka[1998]takesitonestepfurther and disaggregates
theimpactofrealinterestratesintothenominalportionandtheinflation component.
The financialliberalization schoolpredictsthatthenominalinterestratewillhave a
positiveimpactonprivatesavingsand theinflation ratea negativeimpact.Contrary
to theirprediction, the resultsare the opposite.The nominalrate is negativeand
significant and the inflationrate positiveand significant. This is consistentwitha
Keynesian-type precautionary motiveofsavingsratherthana monetarist-type portfo-
lio shiftfromsavingsto assetsthatare inflation hedges.
Studiesutilizingdata on real interestratesand financialsavingshave produced
moremixedresults,eventhoughtheyare consistent withbothKeynesianand finan-
cial liberalization theory.Warmanand Thirlwall[1994]showa strongcorrelation be-
tweenrealinterestratesand M4,netofdemanddepositsintheMexicancontext. Seek
and Yasim [1993]indicatea correlation betweenM2/GDPand real depositinterest
ratesfora pooledsampleof21 countries withinAfrica(and also fora subgroupofnine
ofthe countriesthatincludesBotswana).This is usinga simpleregressionwithan
absurdlylow adjustedR2(.084), however.Moreover,Matsheka [1998]is unable to
confirm theseresultsforBotswana.Therealdepositrateis negativeandinsignificant.
Further, literaturetestingtherelationship betweenrealinterest ratesand invest-
mentseldomconfirms therelationship predictedbyMcKinnonand Shaw. Warman
and Thirlwall[1994]use Mexicandata and findan overallnegativerelationship. In a
sample of nine Africancountries, Seek and Yasim [1993]find a positiverelationship

This content downloaded from 168.176.5.118 on Thu, 9 Jan 2014 11:37:32 AM

All use subject to JSTOR Terms and Conditions
 ALTERNATIVES
TO FINANCIALLIBERALIZATION
THEORY 255

betweenreal depositratesand investment ratesin a simpleregressionwithridicu-

louslylow adjustedR2(0.039). Oncetheeffect is brokendownintonominalinterest
ratesand inflation and othervariablesare introduced, however,therelationship no
longer holds. Nominal interest rates have a and
negative significant relationship with
investment. In thecase ofBotswana,Matsheka[1998]showsa negativebutinsignifi-
cantrelationship betweenrealdepositratesand thelogofgrossdomesticinvestment.
the
Bycontrast, availability ofcreditand a laggedaccelerator relationshiparepositive
and significant. As theauthorpointsout,credithas actuallybecomeless availableto
theprivatesectorin thewake offinancialliberalization, a phenomenon also seen in
Nigeria and in Jamaica. In the case of Jamaica, for real to
example, lending manufac-
83
turingplummeted percent between 1989, when financialliberalizationbegan,and
1999 [Stein,Cuesta,andMcLennon2002].Similarly, inNigeria,therewas a negative
andverysignificant relationship betweenthenumberofbanksandreallendingtothe
privatesectorand thenumberofbanksand theleveloffinancialsavingsrelativeto
GDP. Financialdisintermediation and a viciouscirclearosefromliberalization, not
thevirtuouscirclepredicted byMcKinnon-Shaw [Lewisand Stein,2002].
Othersurveyshave indicatednegativeconsequencesoffinancialliberalization.
Ndung*u [1997]surveysnineEnglish-speaking African countriesintroducing orthodox
financialliberalization and findsdeclininginvestment; fewexamplesofa risein sav-
ings;reducedefficiency ofintermediation, as measuredbytherisingspreadbetween
depositand lendingrates;and fallingGDP growthrates.Otherauthorshave docu-
mentedrisinginterestratespreadsinplaceslikeVenezuela[Vera,2002]and a mixof
African countries[Nissanke,2002].ForJamaica,Stein,Cuesta,andMcLennon[2002]
testtherelationship betweenthegrowth offinancial institutions
andthespreadbetween
deposit and lending rates,and establish a positive and significant
relationshipbetween
thenumberoffinancialinstitutions andthespread,completely contraryto the predic-
tionofMcKinnon-Shaw. The chaoscausedbyfinancialliberalization was leadingto
growing inefficiencyofintermediation.
It followsfromthe aboveanalysisand empiricalresultsthatit is paramountto
developalternativesto the McKinnon-Shaw thesis,givenits weak theoreticalbase
andpoorempiricalperformance. Forthis,we proposetoadoptan institutional-centric
viewoffinance anddevelopment as a waytowards alternativefinancial formulation.
policy

FINANCIAL SYSTEM DEVELOPMENT: TOWARD AN INSTITUTIONAL

PERSPECTIVE

Our analysisso farclearlyindicatesthat the focuson priceformation largely

rendersfinancialliberalization italso lendsitselftomisunder-
theorya-institutional;
standinghowinstitutions in developingcountries
work,whenscantattention is paid
tothem.Yet,thecriticalimportanceof"institutional
endowments" foreconomic growth
beenemphasized
has increasingly bymanyrecenteconometric - see,forexample,
studies
Rodrik[1999]and Acemogluet al. [2002]fortheinstitutions-growth-macroeconomic
performances,and Chinnand Ito [2002]forthe institutions-financial development
link.Even authorslargelyin favorofthegoal ofmonetary restrainthave recognized
theimportanceofdesigning newpoliciesinthecontextofthe"institutional
endowments"
ofcountries[Ball, 1999].The performance ofnew institutional arrangements, like

This content downloaded from 168.176.5.118 on Thu, 9 Jan 2014 11:37:32 AM

All use subject to JSTOR Terms and Conditions
256 EASTERNECONOMICJOURNAL

independent centralbanks,dependsonthecountry's otherformalarrangements, such

as fiscalpolicy,and manyinformal institutional arrangements, likedomesticinterest
grouporganizations; international relations;andthehistory, norms,andideology ofa
country. What is useful from Ball's [1999]analysis is the recognition that financial
transformation is fundamentally an institutional phenomenon, whichinteractswith
theexistinginstitutional endowment. Whatis less usefulis thedichotomous distinc-
tionbetweenformaland informal institutions. Theformal-informal distinction arises
fromtheworkofNorth[1990]and is aimedat explainingthehiddenconstraints on
formallevelsoftransformation. In the case ofBall [1999]thesehiddenconstraints
affecttheextentoftherigidity neededin newmonetary institutions in creatingrules
formonetary constraint. The greateris the informal commitment to monetary con-
the
straint, greater is the in
flexibility usingmonetary in
policy reacting to unfore-
seen shocks.
Thereare problemswithBall's [1999]framework. First,thedistinction between
formaland informal is ratherarbitrary. As Sindzingreand Stein [2002]note,differ-
encesbetweenthe formaland informalconstitutea continuumofactivitiesrather
thana dualityofpolaropposites.Second,theformal/informal distinction is usedinthe
sense ofmaximizing an objectivefunction subjectto constraints. The relationship is
unidirectional in thesensethattheinformal acts to limittheformalin reachingthe
goal or objectivefunction. In fact,the aim oftransformation is notone ofdesigning
formalrulestobe consistent withinformal institutional constraints, buttotransform
theexistinginstitutional endowment forspecificpurposes.Thisis complicated in the
framework duetoa thirdproblem, the
namely conflating ofdimensions ofinstitutions
withtheframework ofinstitutions themselves. NormsintheVebleniansenseof"hab-
itsofthoughtcommontothegenerality ofmenand women"can be institutions them-
selves.Thus,ideologyand historyplay a centralrolein creatingthe framework of
institutional while
transformation, organizations, or even interest groups, are an impor-
tantdimension ofinstitutions in thesenseofconcatenating peoplein a structure with
commonrulesand purposes.Lumpingall thesetogetheras informal institutions is
ratherconceptually problematic.
A greatdeal ofbackingoftheimportance ofinstitutions emanatesfromtheimper-
fectinformation school(forexample,Stiglitz,[1985,1994]).Imperfect information in
thefinancial markets dictates theexistence ofinformation-gathering institutions. Institu-
tionsmatter, therefore, especiallythoseofthefinancialintermediation variety.Under
such circumstances, the structureoffinance,debtversusequity,becomesofpara-
mountimportance. Threeschoolsofthoughtcan be identified onthisscore:thebank-
based view,whichemphasizesthepositiveroleofbanksin development and growth
[Gerschenkron, 1962;Stiglitz,1985;Singh,1997];themarket-based view,whichhigh-
lightsthe advantagesofwell-functioning markets[Beckand Levine,2002]; and the
financialservicesview,accordingto whichneitherbanksnormarketsmatter - it is
financialservicesthemselvesthatare byfarmoreimportant thantheformoftheir
delivery. They are different components of thefinancial system;theydo notcompete,
and as such amelioratedifferent costs, transaction and information, in the system
[Boydand Smith,1998].These viewsare concernedwithbuildinginstitutions that
supportthedevelopment ofmarkets.The workingsoftheseinstitutions becomethe
focusoftheanalysis,especiallytheregulatory aspectsoftheinstitutional framework.6

This content downloaded from 168.176.5.118 on Thu, 9 Jan 2014 11:37:32 AM

All use subject to JSTOR Terms and Conditions
 TO FINANCIALLIBERALIZATION
ALTERNATIVES THEORY 257

Whileinformation gatheringis an important aspectofinstitutional design,it is

onlyonedimensionoffinancialinstitutions. We proposegoingbeyondtheideas that
underliethoseofthe imperfect information school,and also beyondthe misleading
distinctionbetweenformalandinformal institutions,toan institutional-centric theory
ofthetransformation ofa developmentally orientedfinancialsystem.It is important
to distinguish institutional forcesfrominstitutional purposesand institutional out-
comesin termsofthegenesis,evolution, and maturationofinstitutions. Forconcep-
tual clarity,thereneeds to be a carefulunderstanding ofthe relationshipbetween
institutionalcontexts andpotentialinstitutional transformation paths.Financialsys-
temscan be disaggregated intofiveinstitutionally relatedcomponents thatare inter-
activein producing particular outcomes. Each operates in a particularinstitutional
context.Thefivecomponents arenorms, incentives,regulations, andorganizations.7
capacities,
In thecontextoffinancialsystems,normsare habitsofthoughtthatarise from
socialesteemandsanctionsderivedfrom establishedpatternsofbanking.Theyinvolve
rulesofthumb,thedevelopment oftrustand professional habitsthatencouragepro-
bity,and theproperconductthatis thebackboneofbanking.Whiletheseare central
to the development and operationofanybankingsystem,bankingfordevelopment
mustalso incorporate normsthat encouragethe extensionoftimehorizonsas an
integralpart ofintermediation.
Incentivesfocuson therewardsand penaltiesthatarisefromdifferent modesof
behavior.The institutional-centric viewofincentivesis dissimilartomarginalcalcu-
latingutilitymaximizersembeddedin neoclassicaleconomics.First,incentivesare
notsimplydeliveredvia marketsbutcan arisewithina varietyofdifferent organiza-
tionalconstructs. Second,humansare foremost socialbeingswhoare motivatedby
rewardsand penaltiesthatgowellbeyondincomeormaterialfactors.In thecontext
ofbanking,financialvariableslikeinterestratesare onlyonedimensionofa variety
offactorsthatshape bankingdecisionsand behavior.Promotions, the loss ofsocial
esteem,threatsofostracism,socialresponsibility, legal repercussions, professional-
ism, and pride,etc., are all central to generating the incentives for expandingand
operating in
bankingsystems developing countries.
Regulationsconstitute thelegalboundariesthathelpsettherulesofoperationin
financialsystems.The regulatory dimensionsare wellknownand includeprudential
guidelines on the provisions forand categorization ofassetrisk;accounting standards;
auditingschedules;deposit insurance stipulations;capitalrequirements; licensingrules
and procedures;regulationon interestrate determination and interbankmarkets;
scopeofoperationsin termsofthetypesoffinancialdevicessold;and property right
issues,including rules to access collateral when loan payments are in default, etc.
Whatis particularly important is a careful specification ofthe spheres ofinteraction
amongthecomponents oftheeconomy, includingownership amongthedifferent seg-
mentsofthefinancialsystemand theirlinkagetoindustrial,agricultural, and other
servicesectors.One ofthereasonswhyempiricaltestingbetweenfinanceand growth
is so indeterminate is preciselybecause cases ofsuccessfulbankingin developing
countrieshave arisenwhenthereis a dynamicinterface amonginvestment, produc-
tion,and banking.
The issue oflegallysettingoutincentivesto loan,monitor, and superviseactivi-
tiesthathave higherriskbutare moredevelopmental becomesan important partof

This content downloaded from 168.176.5.118 on Thu, 9 Jan 2014 11:37:32 AM

All use subject to JSTOR Terms and Conditions
258 EASTERNECONOMICJOURNAL

thejuridicaldesignoffinancialsystems.An equallyimportant issue is the mecha-

nismstoinstitutionalize thelegalsystemin thesenseofencouraging theinternaliza-
tionoftherulesofoperation.Rewardsand punitivemeasuresmaybe necessaryto
enforce regulationsin theinitialphases.The ultimateaim ofthedesignofanyregu-
latorysystemis tohave monitoring becometheprimefunction and have intervening
enforcement becomethe exception, nottherule.
Capacitiesare relatedtotheunderlying capabilitiesoftheconstitutive members
(individuals and other subunits) of to
organizations operate in an effectivemanner to
achievethegoalsofan organization withintheconfines ofits norms and rules.These
capabilitiesmustbe developedin a consonantmanneron boththe regulatory and
bankingsides. One ofthegreattragediesofliberalization has been the asymmetric
expansionofbankingentitiescomparedto theauditingand otherregulatory capaci-
ties ofsupervisingagencies.Whilenewlegal organizations expandedin places like
Nigeria,thecapabilitiesoftheindividualswithinthesenewstructures wereextremely
weak,providing theopportunity formisuserelativetotheirstatedpurpose.
Organizationsare legallyrecognizedstructuresthat combinegroupsofpeople
withdefinedcommon rulesand purposes.Theyincludebothstateregulatory agencies
andfinancialintermediaries. As intermediaries, countries shouldfocusoncreatingan
assortmentofownershipand bankingtypesto deal withthe multitiered financial
needs ofa developingeconomy(merchant, development, commercial, microfinance,
local,state,international, and cooperativeownership, etc.). In all thesestructures
thestatewillneedto assumeriskbothon thedepositside and loan side (giventhat
themostdevelopmental projectwilloftenhave thehigherrisks).Withoutthesocial-
izationofprivaterisk,it is difficult to see howprivateinvestment and accumulation
will occurin developingcountries.There are manyoptionsforensuringthat the
criteriaforsubsidizationor access to fundsare beingmetand are consistentwith
developmental needs(Korean-style policyloans,Japanesemainbanksystem,busi-
ness-government councils,planningagencies,partialstateownership ofbanks,devel-
opmentalbanks,etc.).To avoidinstability, capitalaccountsneedtobe carefully con-
trolled,includingtheaccessofbankingsystemsto international loans.
Unfortunately, littleofthisis currently happening.In thewakeofthewidespread
failureoffinancialliberalization(includingprivatization to domesticowners)in the
1990s,developingand transitional economieshavebeenturningincreasingly to sell-
ing offfinancialinstitutions to foreignbanks usinga singletypeoforganizational
construct, the commercialbank. The movehas been particularly strongin transi-
tionaleconomieslikeHungaryand theCzechRepublic.The CzechRepublichas paid
an estimated21 percentoftheGDP tocleanup thefinancialsystemafteran exercise
in orthodox liberalization.In response,95 percentofthebankingsectorhas beensold
off.Thefocusofthesebanksis onservicing richerclientsand multinational investors.
Few fundsare beingmadeavailableto domesticinvestors.Small and mediumenter-
prises(SMEs) employnearly60 percentofthecountry's workforce andgenerateabout
40 percentoftheGDP. However,itis estimatedthatonly2 percentofSMEs wereable
toobtaina loan in 2000.Manyare holdingtheirassets in government paperorloans
totheinterbank market{FinancialTimes,2001;November21,2002).Forsomecoun-
tries,likeTanzania,theyhave simplymovedfromstateownership toforeign owner-
ship. While this has avoided the enormously costlyexercisein financialliberalization

This content downloaded from 168.176.5.118 on Thu, 9 Jan 2014 11:37:32 AM

All use subject to JSTOR Terms and Conditions
 ALTERNATIVES
TO FINANCIALLIBERALIZATION
THEORY 259

experiencedby theirneighborsin East Africa,the new foreignbanks are holding

massiveexcessreservesin theformofgovernment paperwithonlyfewloans in the
handsofa handfulofwealthycustomers.Moreover,insteadofaccessingglobalfinance,
banks
foreign in many countries
are national
exporting savingsto saferhavens.

SUMMARY AND CONCLUSIONS

We have arguedin thispaperthatthefinancialliberalization thesisis weak on

boththeoreticaland empiricalgrounds.An alternativeis desperatelyneeded.We
have sketchedthe essentialsof such an alternative,but moreresearchis clearly
required.At the coreofthe projectis the institutionalizationoffinance.For new
financialsystemstobe institutionalized,
theymustbecomelegitimateentitiesin the
sensethattheyare embeddedin thecircuitsofsocialand economicproduction. Ulti-
for
mately, banking norms to be developmental,they need to be absorbed intothe
consciousnessofthegeneralpopulation, whichis morelikelyto happenwhenstruc-
turesarediverse,
participatory, A beliefthatfinancial
andaccessible. systems indevelop-
ingcountriescan be builtbyadjustingpricesignalsand retracting stateintervention
willcontinuetolead tothechaoswe havewitnessedin manyplacesforfartoomany
years.

NOTES

1. In neoclassicalterms,the first-order conditionforintertemporal utilitymaximization fromcon-

sumption is suchthattheratiobetweenmarginalutilitiesin anytwoperiodsmustbe equal to the
expecteddiscountrate.In thismodel,it is assumedthatfinancialliberalization notonlyraisesreal
interestrates,but also allows individualsnew access to borrowing to smoothconsumption over
timewithina lifecycleframework. In thecredit-constrained world,themarginalutilityofpresent
consumption exceedsthemarginalutilityoffutureconsumption. The newaccessto creditincreases
consumption initiallysincethe consumption oftheyoungrises.The fallin savingsthatresultsis
shortlivedas individualsadjust theirconsumption overtime(consumption will fall as theyget
older).Whatis mostimportant is an increasein sensitivitytovariableslikeinterestrates.A risein
the interestrate decreasesthe incentiveto borrowand lowersthe utilityofconsumption, raising
the inducement to save and loweringthe excessdemandforsavings.See, forexample,Gersovitz
[1988],Bayoumi[1993],and Mavrotasand Kelly [2001].
2. McKinnon[1973] uses a Fisher/Hirshleifer (see, forexample,Hirshleifer[1970]) approachto
capitaltheory, in whichtheutilityofan entrepreneur in intertemporal decisionmakingis related
to threeissues: his endowmentor potentialself-deployed capital,his investment opportunities,
and his marketopportunities forexternallendingor borrowing. In a fragmented markettypicalof
developing countries, the threecomponents ofdecisionmakingare badlycorrelated. For example,
thosewithinternalfundsmighthave fewprofitable opportunities.The wayforward is through the
reductionofthe dispersionofrates ofreturnto a "singleallocativemechanism"that can "accu-
ratelyreflecttheprevailing scarcityofcapital"[McKinnon, 1973,11-12].BydrawingonHirshleifer's
[1970]viewofcapital,McKinnon[1973]is takinga ratherextremeposition,in whichtheinterest
rateis notonlyequal to thereturnto capital- theopportunity costofusinginternalfunds - butit
therateofintertemporal
also reflects preference (forexample, the rate of the
discounting future).
3. The financialliberalizationschoolhas heavilyinfluencedWorldBank thinkingthroughout the
1980sand 1990s.This is evidentthroughout theirownpublications: see, forexample,WorldBank
[1983,58-59;1989,171; 1994,114-115].
4. For mostrecentcross-country exercisesforthis purpose,see Reinhartand Tokatlidis[2002],
Galindo,Micco,and Ordonez[2002],and Bekaert,Harvey,and Lundbrad[2002].
5. Habibullah's[1999]statisticalworkuses Grangercausalitytestsand quarterlydata spanningthe
period 1981-94forseven Asian countries.This periodwas one that exhibiteda good deal of

This content downloaded from 168.176.5.118 on Thu, 9 Jan 2014 11:37:32 AM

All use subject to JSTOR Terms and Conditions
260 EASTERN ECONOMIC JOURNAL

in thesecountries,
financialliberalization including and a varietyofnew
interestrateliberalization
financialinstruments.In fiveof the seven cases the causal relationshipwas fromgrowthto
financeor bi-directionality.
6. This definitionofinstitutions as "rules,enforcement mechanisms,and organizations... that sup-
portmarkettransactions"[WorldBank, 2002, 6] drawson twocontributions. The firstis North
[1990], in which,as mentionedabove, thereis the formal/informal aspect, complemented by
"humanlydevisedconstraints," such as "codesof conduct,""formalrules,"and "laws,"that are
aimed"tocreateorderand reduceuncertainty in exchange"[North,1991,97]. The secondis the
workofNabli and Nugent[1989],whoare concerned withhowinstitutions changewithrespectto
theirorganizational nature.Theirfocusis on the extentto whichinstitutions and organizations
coincide.Our approach,by contrast,looksat fiveinstitutionally relatedbut conceptuallydistinct
components thatinteractto producevalid outcome.
7. See Nissankeand Stein [2003]fora moredetailedbreakdownofthesecomponents.

REFERENCES

Acemoglu D., Johnson, S., Robinson, J., and Thaicharoen, Y. InstitutionalCauses, Macroeco-
nomicSystems:Volatility,Crises and Growth.NBER WorkingPaper 9124. September2002.
The African. June2002.
Akinboade, O. A. FinancialDevelopmentand EconomicGrowthin Botswana:A Test of Causality.
Savingsand Development 22 (3) 1998,331-48.
Arestis,P. and Demetriades, P. FinancialDevelopmentand EconomicGrowth:Assessingthe Evi-
dence.EconomicJournal,May 1997, 783-99.
Arestis,P., Demetriades, P., and Fattouh, B. FinancialPoliciesand theAggregateProductivityof
the Capital Stock:EvidencefromDeveloped and DevelopingCountries."Eastern Economic
Journal,Spring2003, 219-44.
Arestis,P. and Glickman,M. FinancialCrisisin SouthEast Asia: DispellingIllusionthe Minskyan
Way. CambridgeJournalofEconomics,March2002, 237-60.
Armijo, L. MixedBlessing:Expectationsabout ForeignCapital Flows and Democracyin Emerging
Markets.In Financial Globalizationand Democracyin EmergingMarkets,editedbyL. Armijo.
Basingstoke:Macmillan,1999.
Bagehot, W. LombardStreetHomewood,111.:RichardD. Irwin,1873.
Ball, R. The InstitutionalFoundationsof MonetaryCommitment: A ComparativeAnalysis.World
Development, October1999, 1821-42.
Bayoumi, T. FinancialSaving and HouseholdSaving.EconomicJournal,November1993, 1432-43.
Beck, T. and Levine, R. IndustryGrowthand Capital Allocation:Does Havinga Market-or Bank-
Based SystemMatter?NBER WorkingPaper No. 8982,2002.
Bekaert G., Harvey, C. R., and Lundbrad, C. Does FinancialLiberalizationSpur Growth?Paper
presentedat WorldBank Conference May 2002.
on FinancialGlobalization,
Boyd, J. H., and Smith,B. D. The EvolutionofDebt and EquityMarketsin EconomicDevelopment.
EconomicTheory,November1998, 519-60.
Calvo, G. Servicingthe PublicDebt: the Role ofExpectations.AmericanEconomicReview,September
1988,647-61.
Caprio, G. Jr., Atiyas, I., and Hanson, J. A. (eds.) Financial Reform:Theoryand Experience,
Cambridge:CambridgeUniversity Press, 1994.
Chinn D. M. and Ito, H. Capital AccountLiberalization,Institutionsand FinancialDevelopment:
Cross CountryEvidence."Departmentof Economics,Universityof California-SantaCruz,
2002.
Cho, Y-J. and Khatkhate, D. Lessons of Financial Liberalisationin Asia: A ComparativeStudy.
WorldBank DiscussionPapers,50, Washington, D.C.: WorldBank, 1989a.
20 May
. FinancialLiberalization:Issues and Evidence.Economicand PoliticalWeekly,
1989b.
De Melo, J. and Tybout, J. The EffectsofFinancialLiberalizationon Savingsand Investment in
Uruguay.EconomicDevelopment and CulturalChange,April1986,561-87.
Demirguc-Kunt,A. and Detragiache, E. The Determinants ofBankingCrises in Developingand
DevelopedCountries.IMF StaffPapers,45, 1998,81-109.

This content downloaded from 168.176.5.118 on Thu, 9 Jan 2014 11:37:32 AM

All use subject to JSTOR Terms and Conditions
 ALTERNATIVES TO FINANCIAL LIBERALIZATION THEORY 261

. FinancialLiberalizationand FinancialFragility.In Proceedings ofthe1998 WorldBank

Conference on Development Economics,editedby B. Pleskovicand J. E. Stiglitz.Washington,
D.C.: WorldBank, 1999.
Dombusch, R. and Reynoso, A. FinancialFactorsin EconomicDevelopment. In PolicyMakingin
theOpenEconomy: Concepts and Case StudiesinEconomicPerformance, editedbyR. Dornbusch.
Oxford:OxfordUniversity Press, 1993.
Financial Times. 12 March1999,
. 12 December2001.
. 21 November2002.
Fry,M. Money,Interestand Bankingin EconomicDevelopment, Baltimore:JohnsHopkinsUniversity
Press, 1988.
Galindo A., Micco, A., and Ordonez, G. FinancialLiberalizationand Growth:EmpiricalEvidence.
Paper presentedat WorldBank Conference on FinancialGlobalization, May 2002.
Gerschenkron,A. EconomicBackwardnessin HistoricalPerspective. A Book ofEssays, Cambridge,
Mass.: HarvardUniversity Press,1962.
Gersovitz, M. Savings and Development.In Handbook of DevelopmentEconomics,edited by H.
Cheneryand S. Srinivasan.Amsterdam: North-Holland, 1988.
Giovannini,A. Savingsand Real InterestRates in LDCs. JournalofDevelopment Economics,18 (2-3)
1985,197-217.
Gonzalez Arrieta, G. M. InterestRates, Savings and Growthin LDCs: An Assessmentof Recent
EmpiricalResearch.WorldDevelopment, May 1988,589-605.
Grabel, I. Speculation-LedEconomicDevelopment:A Post-KeynesianInterpretation of Financial
LiberalizationPrograms.International ReviewofAppliedEconomics,9 (2) 1995, 127-49.
Gupta, K. L. AggregateSavings,FinancialIntermediation and InterestRates. ReviewofEconomics
and Statistics,May 1987,303-11.
Habibullah, M. S. Financial Developmentand EconomicGrowthin Asian Countries:Testingthe
Financial-LedGrowthHypothesis.Savingsand Development, 23 (3) 1999,279-90.
Hellman, T. F., Murdock, K. C, and Stiglitz,J. E. Liberalization, MoralHazard in Banking,and
PrudentialRegulation:Are Capital Requirements Enough?AmericanEconomicReview,March
2000, 147-165.
Hirshleifer,J. Investment, Interestand Capital. EnglewoodCliffs,N.J.:Prentice-Hall, 1970.
Keynes, J. M. The General Theory ofEmployment, and
Interest, Money. New York: Harvest/Harcourt
Brace Johanovich, 1936.
Lewis, P. and Stein, H. The PoliticalEconomyofFinancialLiberalization in Nigeria.In Deregulation
and theBankingCrisisin Nigeria:A Comparative Study,editedby H. Stein,O. Ajakaiye,and
P. Lewis,21-52.Basingstoke:Palgrave,2002.
Lindgreen, C-J.,Garcia, G., and Saal, I. S. Bank Soundnessand Macroeconomic Policy.Washing-
ton,D.C.: International MonetaryFund, 1996.
Matsheka, T. C. InterestRates and the Saving-Investment Processin Botswana.AfricanReviewof
Money,Financeand Banking,(1-2) 1998,5-23.
Mavrotas, G. and Kelly, R. Savings Mobilizationand Financial SectorDevelopment:The Nexus.
Savingsand Development, 25 (1) 2001, 33-64.
McKinnon, R. Moneyand Capital in EconomicDevelopment. Washington, D.C.: BrookingsInstitu-
tion,1973.
. FinancialLiberalisationin Retrospect:InterestRate Policiesin LDCs. In The State of
Development Economics,editedby G. Ranis and T. P. Schultz.Oxford:Basil Blackwell,1988.
. The OrderofFinancial Liberalization.Baltimore:JohnsHopkinsPress, 1993.
Minsky,H. P. Can "It"HappenAgain?Essayson Instability and Finance.Armonk, N.Y.: M.E. Sharpe,
1984.
Nabli, M. K. and Nugent,J. B., eds. TheNew Institutional Economicsand Development Theoryand
Applicationsto Tunisia. Amsterdam: NorthHolland,1989.
Ndung'u, N. The Impactof Financial SectorLiberalizationon Savings,Investment,Growth,and
FinancialDevelopmentin Anglophone Africa.AfricanDevelopment Review,June 1997,20-51.
Nissanke, M. AfricanExperiencewithFinancialSectorReforms: WhatHas Been AchievedSo Far? In
Deregulationand theBankingCrisisin Nigeria:A Comparative Study,editedby H. Stein,O.
Ajakaiye,and P. Lewis,129-67.Basingstoke:Palgrave,2002.

This content downloaded from 168.176.5.118 on Thu, 9 Jan 2014 11:37:32 AM

All use subject to JSTOR Terms and Conditions
262 EASTERN ECONOMIC JOURNAL

Nissanke, M. and Stein, H. FinancialGlobalizationand EconomicDevelopment: Towardan Institu-

tionalFoundation.EasternEconomicJournal,Spring2003, 287-308.
North, D. C. Institutions, InstitutionalChangeand EconomicPerformance. Cambridge:Cambridge
University Press, 1990.
. Institutions.JournalofEconomicPerspectives, Winter1991, 97-112.
Oshikoya, T. W. InterestRate Liberalization, Savings, Investment, and Growth:The Case ofKenya.
Savingsand Development 16, 1992,305-21.
Patrick, H. FinancialDevelopmentand EconomicGrowthin DevelopingCountries.EconomicDevel-
opmentand CulturalChange,14 (2) 1966, 174-89.
Polanyi, K. The Great Transformation: The Political and Economicoriginsof Our Time. Boston:
Beacon Press,Inc., 1957.
Reichel, R. The Macroeconomic ImpactofNegativeReal InterestRates in Nigeria:SomeEconometric
Evidence.Savings and Development 15, 1991,273-83.
Reinhart, C. M. and Tokatlidis, L. Beforeand AfterFinancialLiberalization.Paper presentedat
WorldBank Conference on FinancialGlobalization, May 2002.
Robinson, J. The Generalizationofthe GeneralTheory.In The Rate ofInterestand OtherEssays,
editedbyJ. Robinson.London:MacMillan,1952.
Rodrik, D. Trade and Capital AccountLiberalisationin a KeynesianEconomy.JournalofInterna-
tionalEconomics,August1987, 113-29.
. InstitutionsforHigh-Quality Growth:WhatTheyAre and How to AcquireThem.Paper
presentedat the IMF Conference on Second-Generation Reform,8-9 November1999.
Sachs, J. Conditionally, Debt Reliefand the DevelopingCountries'Debt Crisis.In DevelopingCoun-
tryDebt and EconomicPerformance, editedby J. Sachs. Chicago:University ofChicagoPress,
1988.
Sahoo, P., Geethanjali, N., and Kamaiah, B. Savings and EconomicGrowthin India: The Long-
Run Nexus.Savingsand Development 25 (1) 2001, 66-80.
Schumpeter, J. A. The TheoryofEconomicDevelopment. Leipzig:Dunkerand Humblot,1912.
Seek, D. and Yasim, H. E. N. FinancialLiberalization in Africa.WorldDevelopment. November1993,
1867-90.
Shaw, E. Financial Deepeningin EconomicDevelopment. New York:OxfordUniversity Press, 1973.
Sindzingre,A. and Stein, H. Institutions, Development, and GlobalIntegration: A TheoreticalCon-
tribution. CNRS, July2002.
Singh, A. Financial Liberalisation,StockMarketsand EconomicDevelopment.EconomicJournal,
May 1997,771-82.
Stein, H., Ajakaiye, O., and Lewis, P., eds. Deregulationand theBankingCrisis in Nigeria:A
ComparativeStudy.Basingstoke:Palgrave,2002.
Stein,H., Cuesta, M. and McLennon,R. FinancialLiberalization and theJamaicanFinancialDebacle.
Departmentof Economics,RooseveltUniversity, 2002.
Stiglitz,J. E. CreditMarketsand theControlofCapital.JournalofMoney,Credit,and Banking,May
1985,133-52.
. FinancialMarketsand Development.OxfordReviewofEconomicPolicy,Winter1989,
55-68.
. The Role ofthe State in Financial Markets.In WorldBank Proceedingsof the World
Bank Annual Conference on DevelopmentEconomics,1993. Washington,D.C.: WorldBank,
1994.
Taiwo, I. O. A Flow-of-Funds Approachto SavingsMobilizationUsing NigerianData. Savings and
Development, 16 (2) 1992, 168-82.
Underhill, G., ed. The New WorldOrderin International Finance,Basingstoke:Macmillan,1997.
Vera, L. A Chronicleofa LatinAmericanCountryFinancialCrash:The Case ofVenezuela.In Deregu-
lationand theBankingCrisisin Nigeria:A Comparative Study,editedbyH. Stein,O. Ajakaiye,
and P. Lewis,168-92.Basingstoke:Palgrave,2002.
Villanueva, D. and Mirakhor, A. InterestRate Policies,Stabilisationand Bank Supervisionin
DevelopingCountries:StrategiesforFinancialReform. InternationalMonetary Fund Working
Papers,WP/90/8. Washington, D.C.: International MonetaryFund, 1990.
Warman,F. and Thirlwall,A. P. InterestRates,Savings,Investment and Growthin Mexico1960-90:
TestoftheFinancialLiberalization Hypothesis. JournalofDevelopment Studies,April1994,629-649.

This content downloaded from 168.176.5.118 on Thu, 9 Jan 2014 11:37:32 AM

All use subject to JSTOR Terms and Conditions
 ALTERNATIVES TO FINANCIAL LIBERALIZATION THEORY 263

Weller, C. Financial Crises and Financial Liberalization:ExceptionalCircumstancesor Structural

Weaknesses?Journalof Development Studies,October2001, 98-126.
World Bank. WorldDevelopment Report,1983. WashingtonD.C., 1983.
. Sub-Saharan Africa,From Crisis to Sustainable Growth:A Long TermPerspective
Study.Washington D.C., 1989.
. The East Asian Miracle:EconomicGrowthand Public Policy.Washington, D.C., 1993.
. Adjustment in Africa,Reforms, Resultsand theRoad Ahead. WashingtonD.C., 1994.
. FinanceforGrowth:PolicyChoicesin a VolatileWorld.A WorldBank PolicyResearch
Report.Washington D.C., 2001.
. WorldDevelopment Report,2002, BuildingInstitutionsforMarkets.Washington, D.C.,
2002.

This content downloaded from 168.176.5.118 on Thu, 9 Jan 2014 11:37:32 AM

All use subject to JSTOR Terms and Conditions