MonetaryTheoryandElectronicMoney: ReflectionsontheKenyanExperience

WILLIAMJACK TAVNEETSURI
AND

ROBERTTOWNSEND 1 April2,2010

Abstract
Thispaperusesaclassofmodelsofmoneyandthepaymentssystemtoinformananalysisofmobile banking in the context of the rapid expansion of MPESA, a new technology in Kenya that allows paymentsviamobilephones(evenwithoutanyaccesstoabankaccount),andcurrentlyreachescloseto 38 percent of Kenyan adults. The separation of households and firms in space and time suggests, in theory, from various separate models, a number of implications. These include (i) the potential gain, under some circumstances, from allowing net emoney credit creation, (ii) the impact that the associatedenhancementofcreditmarketscanhaveonmonetarypolicyandontherealeconomy,(iii) the roles that emoney could play not only in credit but also in insurance, unrelated to its payment function,(iv)thepotentialroleforanactivistmonetarypolicyandemoneymanagement,(v)theroleof emoney as a circulating private debt and as a store of value though with potential coordination problems associated with achieving balanced security transformation, (vi) the potential welfare losses from insisting on continuous net clearing of cash and emoney and the difficulty, in any event, of achieving this in practice, and (vii) the management of shortages in the context of fixed rates of exchange of emoney for cash. We provide some summary statistics from data collected on MPESA Agents and users that are reminiscent of the environments of the models and that support some of these implications. Other implications of the models suggest reforms to enhance the systems efficiency.

William Jack (wgj@georgetown.edu) is at the Department of Economics, Georgetown University, Tavneet Suri (tavneet@mit.edu) is at MIT Sloan School of Management, and Robert Townsend (rtownsen@mit.edu) is at the Department of Economics, MIT. The authors would like to thank the editor, Ned Prescott, the referees, Huberto M. EnnisandKartikAthreya,andIgnacioMasandTomSargentfordetailedcommentsandfeedback.

1. Introduction
In 2007, the leading cell phone company in Kenya, Safaricom Ltd, launched MPESA, an SMSbased money transfer system that allows individuals to deposit, send, and withdraw funds from a virtual account on their cell phones, and is separate from the banking system. MPESA has grown rapidly, currentlyreachingmorethansevenmillionusers,approximately38percentofKenyasadultpopulation, anditiswidelyviewedasasuccessstorytobeemulatedacrossthedevelopingworld.Indeed,similar products have recently been launched in a growing number of countries across Africa, Asia, and Latin America,withtheviewofexpandingfinancialservicestopreviouslyunreachedpopulations. 2 MPESA is used not just for remittance purposes, but also to save, to purchase prepaid phone credit and other goods and services, to pay bills and to execute bank account transactions. However, consumersdonotneedbankaccountsinorderto useMPESA,which JackandSuri (2009)foundwas usedbymorethanhalfoftheunbankedintheirsample.ItisusedbyabroadcrosssectionofKenyan society, but has increasingly been adopted by those at the lower end of the income distribution, as evidencedbythesteadyreductionintheaveragetransactionsizesinceitsinception.AlargepartofM PESAssuccessisattributedtothebroadanddensenetworkofover16,000agentsacrossKenya,which providestheretailinterfacewithconsumers. Inthispaper,weexaminetheroleofmonetarytheoryinunderstandingthisnewgenerationofmobile bankingproducts,especiallythosethat,likeMPESA,donotsimplyprovideelectronicaccesstoexisting bank accounts. Deposits of money in a mobile phonebased account reflect holdings by the account ownerofacommoditywerefertoasemoney.Becauseemoneycanbeeasilytransferredfromone individualtoanother,aslongasitisexpectedtoretainitsvalue,itcanbeusedinequilibriumasameans ofexchange,aswellastotransferpurchasingpowerbetweenindividuals. Indeed,atthetimeofthelaunchofMPESA,theservicewasseenasameansofovercomingthehigh transactions costs associated with sending cash remittances that were faced by the 80 percent of the individualsintheeconomywithoutbankaccounts. 3 Butsincethen,emoneyhasbeenincreasinglyused both as a store of value and as a means of exchange, with users able to pay utility bills, make loan repayments, and even pay for taxi rides with emoney. The coexistence of essentially two forms of money,evenifcloselyrelatedandlinked,raisescertaintheoreticalmodelingissuesinitself.Butwhen one form of money is issued by a profitmaximizing entity and the other by the central bank, further issuesofcompetition,regulation,andcoordinationnaturallyemerge. Although mobile banking is in its infancy in the US, payroll cards have provided a similar payment function, albeit without the geographic reach of mobile phone communication. Funds are typically depositedbyanemployerintotheaccountofanemployee,whocaneitherwithdrawcashatanATMor use the card to make purchases at stores possessing debit card machines. As in the case of mobile
Forexample,MPESAissimilartotheservicecalledGCASHinthePhilippines. Theoriginalpilotprogram,supportedbytheUKgovernmentandthemobilephoneproviderVodafone,wasaimedat increasingtheefficiencyofmicrofinanceproductsbyallowingborrowerstomakerepaymentsmoreeasily.Howeverby thetimeofthefulllaunch,thefocushadshiftedtofacilitatingthesendingofremittancesmoregenerally.
3 2

banking,payrollcardusersdonotneedabankaccount,andtransactionsareexecutedusinganexisting communicationnetwork.Inaddition,payrollcardsintheUSaregenerallycheaperthancheckcashing services and money orders, just as MPESA in Kenya is cheaper than most alternatives. Foster et al (2010)describetheuseofvariouspaymentssystemsintheUStheyfindthat93.4%ofconsumersin theUShaveadoptedapaymentcard,butonly17.2%haveaprepaidcard. Thispaperpresentsafirstlookathowexistingmodelsofmonetarytheorycanbeusedtothinkabout the impact of mobile banking on the operations of the financial system, and the implications for monetary and regulatory policy decisions that face the central bank. We are not yet in a position to developafullyarticulatedmodelofmobilebanking,butwehopethisdiscussionwillbeafirststepin thisprocess.Inaddition,thispaperisnotanexhaustivediscussionofallmodelsofmoney,butmoreof asubsetofmodelsthathavedifferentimplicationsfortheroleofemoneyinaneconomy. Most theoretical models of money and credit include both a temporal dimension and some kind of generalized locational heterogeneity. Sequential trades over time require promises to be made (and kept), and records to be maintained. On the other hand, spatial separation can mean that it is not always possible for two parties to a trade to meet each other at the right time, so more complicated multilateralchainsofindividualsarerequiredtoeffectthedesirednettrades. Intheseenvironments,financialinstrumentssuchasfiatmoneyandprivatedebtcansometimeimprove the efficiency of resource allocations by facilitating intertemporal and interspatial trades. However equilibrium allocations may continue to be inefficient without the intervention of either a public institution(suchasacentralbank)orawellregulatedprivateagent(suchasaclearinghouse). Mobile banking has the potential to effectively reduce the distances that separate individuals, both literally and figuratively, thereby lessening the frictions that characterize models of incomplete intermediation,relaxingliquidityconstraints,andreducingtheneedformonetaryinterventions.Onthe other hand, new liquidity constraints could arise, binding for individuals who trade with the new financialinstrument,emoney. The paper proceeds as follows: Section 2, which draws heavily on Jack and Suri (2009), provides background on the recent evolution of mobile technology and mobile banking in Kenya and on the practical operational features of MPESA; Section 3 reviews a number of strands of the literature and discusses the specific lessons that we might draw regarding both the equilibrium impact of mobile banking, and its implications for policy; Section 4 presents some empirical facts from a survey on M PESA customers and Agents that provide some insights into the implications from the models and lessonsinSection3;Section5concludes.

2.BackgroundonMPESA
2.1MobileMoneyinKenya:anIntroduction
Mobilephonetechnologyhasreducedcommunicationcostsinmanypartsofthedevelopingworldfrom prohibitivelevelstoamountsthatare,incomparison,virtuallytrivial.Nowherehasthistransformation beenasacuteasinsubSaharanAfrica,wherenetworksofbothfixedlinecommunicationandphysical transportation infrastructure are often inadequate, unreliable, and dilapidated. While mobile phone callingratesremainhighbyworldstandards,thetechnologyhasallowedmillionsofAfricanstoleapfrog thelandlineenrouteto21stcenturyconnectivity.AsthenumberoflandlinesinKenyahadfallenfrom about 300,000 in 1999 to around 250,000 by 2008, mobile phone subscriptions had increased from virtuallyzerotonearly17millionoverthesametimeperiod(Figure1).Assuminganindividualhasat mostonecellphone, 4 47%ofthepopulation,orfully83%ofthepopulation15yearsandolder,have accesstomobilephonetechnology.InMarch2007,followingadonorfundedpilotproject,Safaricom launched a new mobile phonebased payment and money transfer service, known as MPESA. 5 The serviceallowsuserstodepositmoneyintoaccountslinkedtotheircellphones,tosendbalancesusing SMS technology to other users (including sellers of goods and services), and to redeem deposits for regular money. Charges, deducted from users accounts, are levied when emoney is sent, and when cashiswithdrawn. 6

Figure1:PhoneUseinKenya

Thisisnotquitetrue,assomeindividualsowntwo(ormore)phones,soastotakeadvantageofdifferenttariffpolicies ofthecompetingproviders. 5 Pesa is Kiswahili for money hence M[obile]Money. A second mobile banking service called ZAP has since been launched,operatedbyZain,thesecondlargestmobilephoneoperatorinKenya.ZAPsmarketshareremainsverysmall atthispointintime. 6 The marginal cost of depositing and sending money is very low. These fees cover the costs of maintaining and expandingtheagentnetworkandphysicalinfrastructure,marketing,andprofits.
4

Inparticular,SafaricomacceptsdepositsofcashfromcustomerswithaSafaricomcellphoneSIMcard and who have registered with Safaricom as MPESA users. Registration is simple, requiring an official formofidentification(typicallythenationalIDcardheldbyallKenyans,orapassport)butnoneofthe other validation documents that are typically necessary when a bank account is opened. Formally, in exchangeforcashdeposits,Safaricomissuesacommodityknownasemoney,measuredinthesame units as money (denominated in shillings), which is held in an account under the users name. This account is operated and managed by MPESA, and records the quantity of emoney owned by a customeratagiventime.Thereisnochargetoacustomerfordepositingfundsintotheiraccount,buta slidingtariffisleviedonwithdrawalsfromMPESAaccounts(forexample,thecostofwithdrawing$100 isabout$1). 7 AnMPESAuserwhosendsemoneyischargedaflatfeeofabout40UScentsifsending toanotherregistereduser,andaslidingfeeifsendingtoaphonenumberthatisnotregisteredforM PESA. 8 Figure 2 below illustrates the schedules of total net tariffs for sending money by MPESA, including the cost of withdrawing the funds incurred by the recipient, and compares these with two othermoneytransferservices,WesternUnionandPostapay(operatedbythePostOffice).TheMPESA tariffsshownincludeboththesendingandwithdrawalfees,andaredifferentiatedaccordingtoreceipt byregisteredandnonregistereduser.Feesarechargedtotheusersaccount,fromwhichemoneyis deducted. Additional cash fees are officially not permitted, but there is evidence that they are sometimeschargedonaninformalbasisbyagents.Wereturntothisissuebelow.

Figure2:TotalNetTariffRatesforDepositingandSendingMoneybythePostOfficeandby MPESAtoaRegisteredUserandtoaNonregisteredUser

Thecompletetariffscheduleisavailableat http://www.safaricom.co.ke/fileadmin/template/main/downloads/Mpesa_forms/14th%20Tariff%20Poster%20new.pdf 8 Nonregisteredindividualscanreceivemoneysentbyaregistereduseraslongastheyhaveacellphone.Therecipient receivesatextmessagewithacodethatcanbetakentoanMPESAAgentwhoprovidesthecashlessanyfees.Thefee scheduleisdesignedsoastoencouragerecipientstoregister.Notethatanonregisteredusercannotonsendemoney toathirdindividualfromhis/herphone.

7

EmoneycanbetransferredfromonecustomersMPESAaccounttoanotherusing SMStechnology,or soldbacktoSafaricominexchangeformoney.Originally,transfersofemoneysentfromoneuserto another were expected to primarily reflect unrequited remittances, but nowadays, while remittances are still an important use of MPESA, emoney transfers are often used to pay directly for goods and services,fromschoolfeestothewagesofdomesticstaff. 9

2.2TheGrowthofMobileMoney
MPESAhasspreadquickly,andhasbecomethemostsuccessfulmobilephonebasedfinancialservice. 10 TheaveragenumberofpeopleopeningupanMPESAaccount(i.e.,newregistrations)perdayexceeded 5,000inAugust2007,andreachednearly10,000inDecemberthatyear(seeFigure3).ByAugust2009, 7.7 million MPESA accounts had been registered. Ignoring multiple accounts and those held by foreigners,thismeans38%oftheadultpopulationhadgainedaccesstoMPESAinjustover2years. Since the launch of MPESA, wary of regulation by the Central Bank of Kenya, Safaricom has been at painstostressthatMPESAisnotabank.Ontheotherhand,theubiquityofthecellphoneacrossboth urban and rural parts of the country, and the lack of penetration of regular banking services, 11 led to hopes that MPESA accounts could substitute for bank accounts, and reach the unbanked population. DatareportedinJackandSuri(2009)suggestthisispartiallytrue,althoughMPESAhasbeenadopted byboththebankedandunbankedinroughlyequalproportions. 12 Inaddition,morerecently,MPESA usershavebeenabletowithdrawfundsfromtheirMPESAaccountsatATMsoperatedbyoneofthe commercial banks (Equity Bank) and some banks have begun to use MPESA as their mobile banking platform. However, deposits cannot (yet) be made at ATMs, and the network of ATMs and bank branches,whilegrowing,remainlimited:whileinthelongruntheycouldreplaceagents,bothcapital costs, and costs of security, operation, and maintenance, suggest agents will continue to play an importantroleforsometime. 13

Transactionsarenotalwayslimitedtoinnocenttrades.Forexample,therearereportsofpeopleusingMPESAtopay bribestotrafficpolice.Evenworse,rumorshavecirculatedinNairobithatkidnappersarerequestingransomtobepaid byMPESA,althoughtheserumorshavenotbeenconfirmed. 10 Similar services in Tanzania and South Africa, for example, have penetrated the market much less. See Mas and Morawczynski(2009). 11 In2006itwasestimatedthat18.9percentofadultsusedabankaccountorinsuranceproduct,andby2009thishad increasedto22.6percent(FinaccessI). 12 These data are from a survey fielded in late 2008. Since then, there has been some growth in the number of individuals,andhouseholds,withabankaccount,duetotheexpansionofsuchinstitutionsasEquityandFamilyBank. 13 In2003therewere230ATMsinKenya(seeCentralBankofKenya,2003,at http://www.centralbank.go.ke/downloads/nps/nps%20old/psk.pdf).Recentdatasuggesttherearearound1,200.
9

Figure3:AverageDailyGrowthinMPESARegistrationsbyMonth
ThesizeofMPESAtransactionshasfallenovertimeasithasreachedmoreofthepopulationandhas been used more extensively, as show in Figure 4. In the two years following its introduction, the averagetransactionsizefellabout30%,havingstartedatKShs3,300(about$50).Mostofthisdecline hasprobablybeenduetotheexpansionoftakeupamongthepoorerindividualsandhouseholds.

Figure4:AverageTransactionSize(KenyaShillings):MovingDownmarket 14
Table1belowshowsthevarioustypesoftransactionsforwhichMPESAwasused,whichincludednot just sending and receiving money, but also storing or saving money, purchasing airtime (the prepaid creditusedforvoiceandtextcommunications),andpayingbills. 15
14

ThisdatacomesdirectlyfromSafaricom.

Table1:WhatDoIndividualsUseMPESAFor?
Receivemoney Sendmoney Store/savemoneyforeverydayuse Buyairtimeformyself Buyairtimeforsomeoneelse Store/savemoneyforemergencies Store/savemoneyforunusuallylargepurchases Paybills Receivemoneyforabill/elsepaybills

FractionofSample (BasedonMultipleResponses) 28.40% 25.08% 14.39% 13.58% 8.30% 6.69% 0.27% 1.35% 0.77%

While the sustained growth in MPESA registrations is notable, the volume of financial transactions mediatedthroughMPESAshouldnotbeexaggerated.Table2reportsthatthevolumeoftransactions effectedbetweenbanksundertheRTGS(RealTimeGrossSettlement)methodisnearly700timesthe daily value transacted through MPESA. Related, the average mobile transaction is about a hundred timessmallerthantheaveragechecktransaction(AutomatedClearingHouse,orACH),andjusthalfthe sizeoftheaverageAutomaticTellerMachine(ATM)transaction. 16 MPESAisnotdesignedtoreplaceall paymentmechanisms,buthaseffectivelyfilledanicheinthemarket.

Table2:DailyFinancialTransactions,Oct2007Sept2008 17
RTGS ACH ATM Mobile Valueperday(billionKShs) 66.3 8.5 1.0 0.1 Transactionsperday(thousands) 1.0 39.2 180.2 107.2 Valuepertransaction(millionKShs) 64.67 0.216 0.006 0.003

2.3TheAgentNetwork
Tofacilitatepurchasesandsalesofemoney,andinlightoflowratesofbankaccountcoverageamonga widelydispersedpopulation,MPESAmaintainsandoperatesanextensivenetworkofmorethan16,000 agentsacrossKenya.Theseagentsarelikesmallbankbranches,oftenmannedbyasingleperson.As canbeseeninFigure5,thegrowthofthisnetworklaggedbehindthatofthecustomerbaseforthefirst yearofMPESAsoperationduringwhichtimethenumberofusersperagentincreasedfivefold,froma low of 200 to a high of 1,000. But since mid2008, agent growth has accelerated and the number of usersperagenthasfallenbacktoabout600.
EachentryistheshareofregisteredMPESAusersinoursamplewhoreportedthecorrespondingfunctiontobethe mostcommonlyused.Thebillpaymentservicehadonlyjuststartedatthetimeofthesurveyandhassincebecome ratherpopular. 16 ThesedatarefertoaperiodbeforeMPESAcouldbeusedatATMs. 17 Source: Central Bank of Kenya, presentation at conference on Banking & Payment Technologies East Africa, 1719 February2009,Nairobi.
15

Figure5:ExpansionoftheAgentNetwork 18
RegisteredMPESAuserscanmakedepositsandwithdrawalsofcash(i.e.,makepurchasesandsalesof emoney) with the agents, who receive a commission on a sliding scale for both deposits and withdrawals. 19 Clearly,withdrawalsofcashcanonlybeeffectediftheagenthassufficientfunds.But symmetrically,cashdepositscanonlybemadeiftheagenthassufficientemoneybalancesonhis/her phone.Agentsfaceanontrivialinventorymanagementproblem,havingtopredictthetimeprofileof net emoney needs. Figure 6 shows a representation of the flow of money and emoney among individuals,Safaricom,commercialbanksandthecentralbank,andillustratesthecoreworkingsofM PESA.Theroleofwhatwecallthecoordinator,whichinpracticeisaheadoffice,anaggregator,ora superagent,isdescribedinmoredetailbelow. ThenetworkofcommercialbankbranchesacrossKenya,whilegrowing,remainsmuchsmaller.Asof November 2009, the Central Bank of Kenya 20 reports 44 commercial banks had 849 branches across Kenya(aboutonebranchforevery40,000Kenyans),with50percentofbranchesconcentratedinthe largest(bysizeofbranchnetwork)fourbanks.Asof2008intheUS,therewere7,086institutionswith 82,547branchesthatcameunderFDICprotection,yieldingadensityofbankbranchesabout10times thatinKenya(whosepopulationisabout10%oftheUS).

Source:Safaricom. The commission amounts are nonlinear (and concave) in the size of the transaction. Some reports suggest that in responsetothis,agentsmayencouragecustomerstosplittheirtransactionsintomultiplepieces,therebyincreasingthe overallcommission. 20 http://www.centralbank.go.ke/financialsystem/banks/Register.aspx
19 18

Figure6:FlowsofFiatMoneyandEmoney
In practice, MPESA Agents are organized into groups. Originally, MPESA required that agent groups operatedinatleastthreedifferentlocations,sothattheprobabilityofcashoremoneyshortfallscould beminimized.Thisdiversificationwithinthegroupwouldonlybeeffective,ofcourse,iftheinventories ofmoneyandemoneywereefficientlyreallocatedacrossagentsinthegroupaccordingly.Thereare nowthreeagentmodelsinoperation,inwhichthereisacentralbodythatmanagesandcoordinatesthe operationsofagroupofsubsidiaryagents.Thesemodelsaredifferentiatedwithregardtotheformal statusofthecoordinatingbodyandtheownershipstructureofthegroup,andwhetherthecentralbody conductsdirecttransactionswithindividualusers,asshowninFigure7.

Figure7. (a) The coordinating body is the head office which owns agents, and can transact directly with customers. (b) The coordinating body is referred to as an aggregator and has arms length contractual relationshipswithagents.(c)Thecoordinatingbodyisabankbranchandiscalledasuperagent,butneither ownstheagentsnortransactsdirectlywithcustomers.
10

In the first model, one member of the agent group is designated as the headoffice, which deals directlywithSafaricom,whilesubsidiaryagentsthatareownedbytheheadoffice,managecashande moneybalancesthroughtransactionswiththeheadoffice. 21 Boththeheadofficeandtheagentscan transact directly with MPESA users. The second model is the Aggregator model, with the aggregator actingasaheadoffice,dealingdirectlywithSafaricomandmanagingthecashandemoneybalancesof agents. However, the agents can be independently owned entities, with which the aggregator has a contractualrelationship.Afinalandmuchmorerecentmodel 22 allowsabankbranch,referredtoasa superagent, to make cash and emoney transactions with agents on an ad hoc basis.However the bankdoesnottradeemoneywithMPESAcustomers.Thesuperagent modelisoneexampleofthe integration of MPESA services into the banking system. Other developments in this vein include the ability to transfer funds, often via ATMs, between a users MPESA account and accounts at certain commercial banks with which MPESA has forged partnerships. But even as MPESA has facilitated transactions fortheapproximately72percentofuserhouseholdsin JackandSurissamplewithbank accounts,itremainspopularwiththeunbanked,ofwhommorethanhalf,(54percent)usedMPESA. 23 ThecashcollectedbyMPESAAgentsinexchangeforsalesofemoneyiseitherkeptonthepremisesor deposited in the agents (or head offices) bank account. When they wish to replenish their emoney balances,agentstransfermoneyviathebankingsystemtooneoftwobankaccountsheldbySafaricom. Safaricomisrequiredtolimitthequantityofemoneyitissuestotheamountofmoneyitreceivesfrom agentsthatis,emoneyis100%backedbydepositsincommercialbanks.However,thesedepositsare subjectonlytotheregular6percentCentralBankreserverequirement.

3.ModelsofMoneyandMeansofPaymentwithSpatialSeparation
MPESAs rapid expansion means that a large share of the Kenyan population now conducts at least some of their financial transactions by phone. In this section we discuss the implications of this new kind of payment system for the management of the financial system as a whole, and of central bank regulatoryandmonetarypoliciesinparticular.Toaddressthesequestions,inthissectionwedescribein some detail a number of models of money, the payment system, and clearing and settlement. The purposeistofocusonthefeaturesofthemodelsthatcanprovideinsightsintotheoperationaldesignof mobile banking and inform policy choices facing regulators and monetary authorities. Therefore we followthesummaryofeachmodelwithadiscussionofitsimplicationsformobilebanking.Weproceed incrementally,beginning withsimple butsurprisinglyrichmodelsofmoney,thenprogressivelyreview morecomplexmodelswhichwebelievereflectparticularfeaturesoftheKenyanfinancialenvironment.

MPESA requires that each coordinating body has a bank account, so that funds can be transferred easily between them.InordertoopenanMPESAbusiness,thecoordinatingbodymusthaveaminimumbalanceinabankaccount, whichisusedtopurchaseinitialholdingsofemoney. 22 ThismodelstartedafterthefirstroundoftheJackandSuri(2009)survey. 23 About50percentofhouseholdshadatleastonememberwithabankaccount.Ofbankedhouseholdsinthesurvey, about60%usedMPESA,comparedwiththe54percentofunbankedhouseholdsreportedabove.
21

11

3.1Townsend(1983)ModelofFinancialDeepeningandGrowth
Thismodelfocusesdirectlyonimprovementsinthetechnologyofcommunication,andlinksthedegree offinancialinterconnectednessofagentswiththelevelofeconomicdevelopment,inacrosssectionand also over time. The idea is that as connectedness increases, with electronic payments connecting otherwise spatially separated agents, there is an increase in the specialization of labor, an increase in theconsumptionofmarketproducedgoods,andashifttowardsemoneyrelativetofiatmoney.Thisis thestoryofhowfinancialdeepeningandgrowthareintertwined,andhowMPESAcouldhelpKenyato increaseGDPovertimeatthesametimeasitincreasesmonetizedexchange. Eachhouseholdoftypeicanproduce (bysupplyinglabor)onlygoodi,andeachhasautilityfunction overitsownconsumptionofgoodiandagooditcannotproduce,i+1,wellasleisure.Whenhouseholds areinautarky,withoutphysicalorelectroniccontact,notradeispossible,soeachhouseholdconsumes allitsproductionofgoodi,only.Inthissituation,thereisnoneedforameansofpayment.Incontrast, with some travel, as in the Cass and Yaari (1966) or Lucas (1980) versions of the Wicksell triangle applied therefore many times, household i can only trade either with household i+1 (whose good i values), or with household i1 (who values good i). But because of the structure of preferences (e.g., because household i+1 does not value good i, and instead wants goods i+1 and i+2), narrow bilateral exchangebetweeniandi+1isinnoonesselfinterest.Thisisthekeylack ofdoublecoincidenceof wants.Decentralizedtradewouldgiverisetoautarky,ifitwerenotforvaluedfiatmoney. 24 The timinglocation is shown in Figure 8 below where, in any given period, household i has two members,ashopperandaseller,whocanonlymovehorizontallytotradewithhouseholdsi+1andi1, respectively. Between time periods, members of household i, i even, shift down one line, and households i, i odd, stay put. Thus, debt issued by a household of type i, i even, can only be passed alongtohouseholdverticallyabovetheissuerandsohasnovalue.Onlyfiatmoneyisusedanditcan havevalue.Specifically,onememberofeachhouseholditravelstothemarketwithi+1andpurchases someofgoodi+1atpricepi+1withfiatmoneyacquiredpreviously,andthesecondtravelstothemarket withi1andsellssomegoodiformoneyatpricepi.Notethatittakesoneperiodforgoodsproduced and sold to come back via money holding in the interim as goods purchased. With constant prices across time and space and with a positive intertemporal discount rate, this makes it less beneficial to supplylabor.Thisisacrucialaspectofthisandotherrelatedmodelsbelow. i1
mkt(i,i1)

mkt(i,i+1)

i+1

Figure8:TradingSchemeforPairedHouseholds

24

Thesemodelsruleoutprivatedebtandfuturecontractsinfiatmoneybyassumingtherearenopairingssuchthatdebt canberedeemedbytheissuer.Seebelow.

12

InaWalrasian,centralizedexchangeregimewithelectronicdebitsandcredits,householdscannowhold intraperioddebtforwithinperiodpurchasesand,atthesametime,sellandreceiveelectroniccredits. Attheendoftheperiod,accountsarecleared.Intuitively,whenonememberofhouseholditravelsto market(i,i+1)tobuygoodi+1fromhouseholdi+1,itisasifthatmemberwereusingacreditcard(or phone)linkedelectronicallytoacentralaccount,whichwillnotbepaidtilltheendoftheperiod.The secondmemberofhouseholdiwho travelstomarket(i,i1)andsellsgoodiispaidwith acredit card from household i1. At the end of the period, these electronic debits and credits are cleared and accounts must balance (we return to interperiod debt in the Lacker model below). Note that goods producedandsoldcanbetransformedinthiswaytogoodspurchasedwithinthesameperiod,sothere is no inefficiency associated with holding idle money balances. In fact, in the equilibrium of this electronic accounting system, fiat money plays no role and its price is zero. The prices of goods themselves are in some (arbitrary) unit of account. Related, though households remain separated in space,itisasiftheyweretransactingwithoneanotherinacentralizedmarketthatignoredthespatial segmentation, as far as prices and values are concerned. However, that this system works only if households are allowed to overdraft their electronic accounts and there enough commitment or punishmenttomakesuretheyhonordebtsaccruedwithintheperiod. Insummary,ifwethenassumethatsubstitutioneffectsdominateincomeeffectsandfocusonprices, thecostofconsumptionofthenonproducedgoodintermsoflaborisinfiniteinautarky,andhighin thefiatmoneyregimerelativetothecentralizedWalrasianelectronicclearingeregime.Movingfrom autarky to the decentralized money regime and then to the centralized Walrasian regime, the model predicts that labor supply increases, output of the produced commodity rises, consumption of non produced good rises, consumption of produced good drops, trade volume increases and welfare increases.Ifaneconomyhadamixofdecentralizedandcentralizedregimes,aswithsomefractionof lines (see Figure 8) using fiat money and other lines using Walrasian credit, and these fractions variedacrosscountries,thenpercapitanationalincomerisesasfinancialinterconnectednessincreases, fiatmoneydecreases,percapitaprivatedebtincreases,buttheratiooffiatmoneytoincomedecreases andtheratioofcredittoincomeincreases.Thispatterntendstobewhatweseeincrosssectionaldata. Similarcomparisonsarevalidforaneconomythatisbecomingmorefinanciallyintegratedovertime,as is Kenya, where forward looking households in the fiat currency part of the economy treat financial integration into the Walrasian esystem as an exogenous random event which happens with positive probability (essentially changing the discount rate). Note, however, that thus far, in this particular model,nohouseholdneedstousemultiplemeansofpayment.

Implicationsformobilebanking
WhataretheimplicationsofthiskindofmodeloffinancialdeepeningforasystemlikeMPESA?Itis clearthatMPESAwillchangethefinancialconnectednessoftheindividualsintheeconomy,whichin the model above will cause higher economic development. The main takeaway from this model is thereforethatMPESAcanbeviewedasatechnologicalinnovationwhichlowerstradingcosts,orbetter put, allows financial transfers, credits and debits, across agents who are still separated in space. This improves welfare, at least in the model economy without government and no vested interests in the currentintermediationsystem(andwithoutotherheterogeneity).Fiatmoneyandelectronicpayments 13

cancoexistifsomehouseholdshaveaccesstoMPESAandsomedonot.However,inthemodel,but perhaps not in the MPESA system, the household buying goods in effect creates a net increase in e money within the period. If emoney were essentially only a debit card, then an initial deposit of currencywouldhavetounderliethedebittransaction,undercuttingthiskeyadvantage.Puttheother wayaround,thetheoryarguesthatwemightseefeaturesofnetcreditcreationinthefunctioningofthe actualMPESAsystem,thoughperhapsatanaggregatedoragentlevelandnotnecessarilyatthelevel ofindividualhouseholds.However,forthisfeaturetoexisttheremustbea(harsh)meansofpreventing renegingordefaultsothataccountsactuallyclearattheend.Eventhatrequiresforesightoftheoverall equilibrium,e.g.heretheshopperknowsthepricesatwhichthesellerisreceivingcredits.Again,we comebacktothismismatchandinterperiodcarryoversintheothermodelsbelow.

3.2ManuelliandSargent(2009)TurnpikeModelwithCurrencyandDebt
AcloselyrelatedmodelofManuelliandSargent(2009)rationalizesthecoexistenceoffiatmoneyand private credit. As in Townsends turnpike models, agents meet in pairs, and while they have long enoughrelationshipstoundertakesomeefficiencyenhancingintertemporaltradesviatheextensionof private credit, they do not stay together for long enough to effect fully Pareto efficient allocations. More specifically, time is divided into periods (think of these as years), each composed of foursub intervals(e.g.,seasons).Individualsmeetforjusthalfayearonly,i.e.twoconsecutivesubintervals, andthenmoveon,sometotheeast,sometothewest(seeFigure9).Inthefirstsubintervalofahalf year one person in a given pair has a positive endowment of the single perishable consumption good and the other has none, and in the second subinterval these roles are reversed, giving rise to short term(twosubinterval)privatecreditarrangements.However,thepositiveendowmentsineachsub intervalcanbeeitherhighorlow,forexample,a>0,b>0,anda/b>1,whileaggregateoutputineach halfyear(a+b)isconstant,andeachindividualsannualaggregateendowmentisconstant,alsoequal to (a + b). Because agents remain together for only two subintervals (one half year), they cannot implementtradesacrosshalfyearsthatis,theycannotissuelongtermdebt.Fiatmoneyplaysarole in facilitating the trades that such debt would effect. Manuelli and Sargent generalize this to include laborsupply,sothatoutputisendogenous.
EastHeadedAgents

(a,0) (0,b) 1

(0,b) (a,0) 2

(a,0) (0,b) 3

(0,b) (a,0) 4

West HeadedAgents

Figure9:TurnpikeSetupforManuelliandSargent(2009)
One interpretation of Manuelli and Sargents model is as a generalization of Townsends original 14

turnpikeinwhichendowmentsfluctuatedwithaperiodicityoftwoandmeetingslastedonlyoneperiod. Insteadofmeetingfortwoperiods,wecaninterpretManuelliandSargentasamodelwherehouseholds continue their travels after one period but remain linked electronically for two periods (though we ignoretherequisitecostlyshippingofgoodsinthesecondperiodsomeofthemodelsbelowaremore complicated so as to eliminate this flaw in our attempted interpretation). As the time and spatial limitationsofcommunicationfall,e.g.,withtheexpansionofthenetworkofMPESAagents,accounts andtheuseofcellphones,debtsofincreasinglylongmaturitycan,inprinciple,beissuedandrepaid.

Implicationsformobilebankingandmonetarypolicy
Totheextentthatmobilebankingfacilitatestheoperationoftheprivate(ofteninformal)creditmarket, amodelthataccommodatessuchproducts,withnontrivialimplicationsforpolicy,canbeinformative. To start, as in the Townsend models, the laissez faire, noninterventionist monetary equilibrium (withoutdebt)isnotParetooptimal.Essentially,thewedgethatwediscussedintheearlierCassYaari model, where money is earned through production and held without interest for one period, can be eliminated with intervention, by paying interest on cash balances. This equates intertemporal substitutioninconsumptiontothenaturalrateoftimediscountandensuresthatnohouseholdhitsa bindingcorner,runningoutofcash. But, the impact of monetary policy interventions in the form of changes to base money, depends on whether private credit is allowed, or under the interpretation here, whether emoney which allows borrowing and lending is in the system. An increase in the growth rate of the money supply has ambiguouseffectsontheaveragelevelofoutputbutincreasesthevolatilityofoutputwhenthereare norestrictionsonprivateborrowingandlending.However,ineconomieswhereindividualsdonothave access to private loan markets, say because they move on without cell phones, the results are quite different:anincreaseintherateofmoneygrowthdecreasesmeanoutputandhasnoeffectonvolatility ofoutput(whichremainszero).Likewise,iftheeconomywereliberalized,orotherwiseexperiencesa surpriseinnovationthatallowsprivateborrowingandlending,thenpricesincreaseandoutputbecomes more volatile. Financial innovation is welfare improving, but intimately connected with the impact variablesthatCentralBankstypicallymonitororattempttocontrol. As Manuelli and Sargent (2009) emphasize, the potential destabilizing effects of actual financial liberalizationsarehighlightedinboththeacademicandpolicyliteratures.Moregenerally,theeffectsof monetary policy depend on the way private credit markets are operating, even if in the process of borrowingandlendingthereisnonetcreationofemoney.Thus,whenformulatingmonetarypolicy, theCentralBankwillneedtotakeintoaccounttheeffectivechangeinfinancialregimeswhichMPESA hasbroughtwithit.Indeed,intheaboveclassofmodels,optimalmonetarypolicyintermsofcontrol over fiat base money is still relatively straightforward but not without interest. Specifically, the allocationachievedunderoptimalpolicydiffersfromthatassociatedwiththecorrespondingeconomy withnolocationalrestrictionsandcentralizedtradespermittedattimezero.Whilebothallocationsare Pareto optimal, they are not the same, implying that efficient monetary policy has redistributive consequences. Further, optimal government issued currency continues to play an essential role even wheninterestisoptimallypaidonholdingsofsuchcurrency.And,theinterestoncurrencypolicydoes 15

notworkinawaythatcanbereplicatedbyfreebankinginaWalrasianworld.Related,movingfroma suboptimal policy to one with interest on currency may redistribute income and not be Pareto improving.Inthismodel,unlikethefirst,emoneydoesnotdriveoutfiatmoney,northeneedforan optimalmonetarypolicy.Thisisreminiscentofaclassofrelatedmodelsofmonetarymanagementin whichimplementationofpolicydependsontheabilityofagentstotradeinassetmarkets 25 .Financial marketsegmentationreliesoncostswhichmaybearguablydecreasing.

3.2Townsends(1987,1989)ModelsofActivistMonetaryPolicyandMoneyas aCommunicationDevice
To generalize Manuelli and Sargent, credit arrangements can be used to implement trade among individualswhoremainintheirhomelocationanddealwitheachotherrepeatedlyovertime.Thisis also similar to the Walrasian accounting system of the first model above, except that here again the tradeisintertemporal,withborrowingandlendingovertime,sothatanyindividualsbalancedoesnot havetonettozeroattheendofeachperiod.Inthenextsetofmodelscreditisidentifiableasdirect communication and promises. Fiat money then coexists with credit and serves as a communication devicefordealingwithstrangers,acrosslocations. 26 But,themodelsherefeatureDiamondandDybvig (1983) style preference shocks, with patient and urgent households generating the desired intertemporal trade. Moreover, the models here deliver welfare gains from an activist monetary authority responding to shocks, managing liquidity needs. More generally, the quantity and kinds of money in the system are determined optimally in an effort to compensate for missing credit and insurancemarkets.Inthisway,onecanbuildontheplatformofetransferstocreateahighlyeffective recordkeepingsystem,inwhichelectronicaccountsallowforarichvarietyoffinancialinstruments. Townsends(1989)modelenvisionsascenariowherethereareNislandseachwithNinhabitants(the caseofN=3isshowninFigure10belowbut,moregenerally,Nisalargenumber). 27 Preferenceshocks thatarecorrelatedamongasegmentofeachislandsresidentsoccurinthefirstperiod.Thatis,some fraction of the residents are patient, in principle willing to lend, and the residual fraction are urgent, wantingtoborrow.However,ashare(1)ofthepopulationofeachislandmoves,spreadingoutacross alltheotherislandsinsuchawaythatnomoverencountersanyonefromhishomeislandathisnew destination.Thiscreatesaproblemifrecordkeepingislimitedtolocations,thatis,ifthereisnocross island communication or accounting system, so that only nonmovers can borrow and lend: promises
For example, see Fuerst (1992), Grossman and Weiss (1983), Lucas (1990), PerezVerdia (2000),Romer (1987) and Rotemberg(1984). 26 See also Kocherlakota and Wallace (1998), Ireland (1994), Calvacanti and Wallace (1999), Kocherlakota (2005) and Wallace(2005)andthereviewinWallace(2000)inwhichoutsidemoneyandinsidemoneyissuedbybankswithknown tradinghistoriescoexist. 27 In this model, the agents all precommit to arbitrary tax and transfer schemes over time and to all institutions and resource allocation rules. In the language of the models, they commit to an economywide credit arrangement that specifiesconsumptionandtransferstoagentsconditionalonaggregatestatesandonindividualspecificlocationshifters (that are public) and individual announcements of preference shocks (private). Apart from these plans, there is no governmentandnodistinctionbetweenprivateandpublic.
25

16

involving movers (either between themselves but going to different locations, or between them and nonmovers),ontheotherhand,arenotcredibleastheycannotbeconsummatedatalaterdate. Asmoversareeffectivelyexcludedfromthecreditmarket,asocialplannercouldattempttoimplement efficientintertemporalconsumptionprofilesbyaskingmoversateachdatetoreporttheirpreferences, allocatingconsumptionaccordingly.Butiftheinformationreportedcannotbecrediblytransmittedto other islands, without a record keeping device, then the only incentive compatible mechanism is one thatgivesallmoversthesamelevelofconsumptioninbothperiods,independentoftheirpreferences. Portablefiatmoneyallocatedtomovers,andmonotonicallyrelatedtotheirfirstperiodannouncements, canfacilitatethetransmissionofinformationacrosstimeandspacetothestrangerstheymeetattheir destinations. In this interpretation, fiat money is a portable token. By allowing side trades between individuals,monotonicitycanbestrengthenedtolinearity,deliveringapriceoffiatmoneyortokensfor goods. Of course, the initial nominal price level remains arbitrary, as that is simply a matter of the denominationoftheunitofaccount.

Figure10:OnEachIsland,aShareofEachPopulationLeavestoOtherIslandsinPeriod2
However, if additional periods are added to the model, e.g. another round of movers, future movers mustalsobeallocatedfiatmoneyinordertoengageinintertemporaltrade.Thepurchasingpowerof eachunitofmoneyallocatedtosecondroundmoversmust,forefficiencyreasons,bethesameasthat offeredtofirstroundmovers,butthequantitywillbeincreasinginthenumberofmoversandtheshare ofthemwhoarepatient. 28 Asthepreferencesofnewgenerationsofconsumersarerevealed,planned consumptionlevelssupportedbyallocationsoffiatmoneyinearlyperiodsmayberevised.Sincethe purchasing power as previously explained is constant across early movers in a given period, the associated adjustment to consumption levels is effected through changes in the price level. That is, inflationeatsupthepurchasingpoweroffirstroundmoversifitisjudgedthattheyshouldbegetting less given what new information tells the monetary authority about the way they and second round moversshouldbetreated.Notethatthisactivistpolicyisquitedifferentfrom,say,aFriedmanrule,as described in the earlier class of models above, in which a constant rate of deflation can remove the distorting wedge. This is not enough here. Optimal policy is state contingent (Manuelli and Sargent (2009)anticipatesuchresultsintheconcludingsectionoftheirpaper).
28

Individualswhoarepatientconsumemorelater,andthereforeneedmoreunitsofmoneytoconfirmthistofuture strangers.

17

Notehoweverthatfiatmoneyastokensconveyonlytheinformationthatahouseholdhadbeenpatient inthepast,notthatthehouseholdhasbeenafirstorsecondroundmover.Ifevenmoreinformation, suchasthedatesandthenatureofpasttransactionswereencodedinthesystem,thenthedistinction betweenlocalandinterislandaccountscoulddisappear.Thatis,onecanimagineonekindoffiatmoney e.g.,redtokensforfirstroundmovers,andanotherkindgreentokensforsecondroundmovers, and electronic accounts for those who stay home. Indeed, accounts which distinguish all these space time transactions could be accomplished with the electronic record keeping that mobile technologies andmarketsallowatleastinprinciple.Indeed,withallofthat,wecouldinprincipalgofurtherandhere again completely mimic the outcome of a perfect Walrasian accounting system, in which changing locationspersehasnoconsequences.Thefractionofagentsleavinganislandisexactlythesameasa fractionarrivingand,financiallyspeaking,therewouldbenostrangers. Townsend (1987) generalizes this idea of multiple monies (or differentiated eaccounts) in a similar framework with four agents, spatial separation and private information on preference shocks. In particular, suppose there are two islands with two individuals each, as illustrated in Figure 11 below. Agentsaandbliveontheleftislandinperiod1,andagentsaandbliveontherightisland.Inperiod 2, b and b switch places, while a and a (who are subject to shocks) remain on their home islands. Agentsbandbareriskneutralandinprinciplecanprovideinsurancetoagentsaandawhoarerisk averse.Withonegood,preferenceshocksdeterminenotonlythedegreeofriskaversionbutrelative patience. With two goods, there can be preference shocks for each good over time, e.g. patient for goodoneandurgentforgoodtwo,andanoverallintertemporalshockdeterminingutilityinperiodone versusperiodtwo.

Figure11:Islandswithstayers(a)andmovers(b)
Townsendthenexaminesthepropertiesoftradefacilitatedbyalternativecommunicationdevicesinthis environment,bothforthecasesofasinglegoodaswellasformultiplegoods.First,oralcommunication can take place only between agents in the same location, and so cannot be used to convey credible information across time to strangers (if agents cannot carry tokens, commodities or messages). The equilibrium is thus Pareto inefficient. On the other hand, tokens (money) that are appropriately distributedinperiod1canbeusedtoverifyinformationinperiod2,helpingwithincentivestoreveal informationcorrectlyandactingagainasatechnologyforstoringthatinformation.Thepreviousmodel providesintuitionforthecaseofonegood.However,withtwogoods,onetypeoftokenmaynotbe enough.Intuitively,onewantstoconveythefullhistoryofshocksforeachgoodinthefirstperiod,yet ensure incentive compatibility in the second when agents can turn out to be very desirous of consumptionoverall.Forexample,onetypeoftoken,saythegreen,ishandedoutinperiodonegivena certainrealizationofpreferenceshocks,whiletheredtype ishandedoutgivenanotherrealizationof theseshocksagaininperiodone(alternatively,thesearedifferentcreditsindifferentcellaccounts). 18

Theninthesecondperiod,theagentisrequiredtoshownotjustthecorrectnumberoftokens,butalso thecorrectcoloredtoken(orhavetherequisitebalancesinaspecificcellaccount).Indeed,muchcan bedoneevenwithncommoditiesandmcombinationsofshocksusingcombinationsofredandgreen tokens (two types of emoney) as an encryption system. The point more generally is that multiple monies are used to convey the history of trade, borrowing and lending, and insurance, not simply a meansofpaymentortransfersystem.

Implicationsformobilebanking
The bottom line of these models of money as a communication device is that the better the communicationofpastshocksortransactions,themoreefficientcanbetheallocationofconsumption, but (with initial heterogeneity) this may be wealth redistributing. The model features tokens or fiat money, but again portable cell devices linked to some of the account history of earlier transactions wouldprovidesimilarfeatures.But,toachieveanefficientallocation,therecanariseasinthesemodels theneedforactiveliquiditymanagement.WecanseethatinascenariowhereMPESAemergesasthe entitybehindalargefractionoftransactions,emoneycouldsubstituteforfiatmoneyortokens.This would not necessarily replace the need for an activist monetary policy, but would alter that policy, where the level of tokens created on net by the financial system ideally responds to mobility and the stateofdemand,sowouldelectroniccreditsifallowedoptimallytofunctionthatway.Hereadistinction betweenprivatecreditandpublicmoneybecomesblurredasweconsiderquestionsofoptimalmarket design.Thesocialgoodisservedbyhavingmutuallyagreeduponandcollectivelyenforcedrules. Another lesson from these models is that electronic records of past transactions allow new financial instruments,inthiscasebetterborrow/lendingandinsuranceoverspaceandtime.Tyingfiatmoneyto emoneyandthinkingofbothasfacilitatingpayments,only,mayleadonetomissotherwisebeneficial arrangements that have to do with insurance against spatial and intertemporal idiosyncratic and aggregateshocks.Indeed,underthecurrentMPESAsystem,thepricesatwhichmoneytradesfore moneyaresupposedtobefixedovertimeandacrossspace;emoneyandcashtradeforeachotherone for one, (as described above however there are nonlinearities in the transactions costs by amount traded)yetthesefeescanbeseenasallowinginprincipleatradingpricebetweencashandemoney thatisdifferentfromone.Whetherornotonewantstoallowmoneypricesandtherateofexchangeof moneyforemoneytomovewiththestateoflocaldemandandinventoryoftheactorsagainbegsthe questionofwhatemoneyissupposedtobeameansofpaymentonly,ifitfacilitatesanexpansionof the monetary base, or is rather a partial substitute for missing, more centralized economywide insuranceandcreditmarkets. 3.3TownsendandWallace(1982)CirculatingPrivateDebtanda

CoordinationProblem
Thereisyetanotherwaytothinkofmoney,namelyasanobjectwhich,evenifprivatelyissued,appears frequentlyinexchange,i.e.withahighvelocity.Wecanunderstandthisbysimplyextendingthemodel environment in the previous section to four periods with households b and b continuing to switch 19

locations from one period to another, back and forth, with households a and a remaining in a single location.TownsendandWallace(1982)replacepreferenceshockswithtimevaryingendowmentsofa singlegood,butwithdifferentprofilesforthedifferentagents,toinducethedesireforintertemporal trade.Theyalsoassumetherearemanyagentsofeachtypeinanygivenlocation,tojustifypricetaking behavior.Inoneoftheequilibria,thefirstperiodhouseholdbmakesadepositofgoods(butcouldbe money)to,thatis,lendsto,agenta,asifagentawereabankissuinglongtermdebt(oratleastdebt payable on demand). However, household b does not hold this debt but rather moves in the second period to a different location, inhabited by agent a. At this new location, neither party is physically connectedtobanka.Then,subsequent,inthethirdperiod,agentawillpassthedebttobwhointurn redeemsitinthelastperiod,sincebmeetsupwiththeoriginalissuer,agenta.Notethatlongterm debtisalsothedebtthatcirculates,thathasahighvelocity.Inthatsensecirculatingdebthassomething todowithmaturitytransformation.Shorttermdebt,e.g.twoperiodbetweenagentsaandb(oraand b),notonlyextinguishessooner,itdoesnotcirculate. With private debt transferable electronically, one has the same equilibria but with the more realistic interpretationofagentaasanMPESAAgentissuingdebt(inthiscaseinexchangeforgoods,notfiat money, but see below). That is, household b uses the emoney account to trade with subsequent householdsand,again,theemoneyisnettedoutbacktozerowhenathirdpartycomestoagentato redeem it. In this model, agent a can also play this role as banker, or MPESA Agent, instead of a. However,withoutcoordination,anotherproblememerges.Theamountofemoneyissuedbyagentsa andahastobecoordinatedsoastobeconsistentwiththeoverallequilibrium.IfMPESAagentsaand aarenotcommunicatingacrossspace,thenitishardtoimaginehowthiswouldhappen.Townsend and Wallace refer to various historical episodes such as the crash of markets in bills of exchange as evidencethatthemodelwithcoordinationproblemsispickingupproblemsthatmayoccurinpractice.

Implicationsformobilebanking
Electronic debits can be transferred across agents in spatially separated locations and have a high velocity. This seems to capture a big part of the Kenyan MPESA reality. This comes, however, with potential coordination issues which need to be thought through. In the current model with two locations,fouragenttypes,andfourperiods,oneachievesthefirstbestwiththerightcombinationof circulating private debt and othershort term noncirculating debt. In that equilibrium, prices/interest ratesaremovingaroundovertimeandspaceandallmarketsingoodsandfinancialinstrumentsclear. Again, fixing the price at which one object trades against another would seem to create additional problems.Butevenpriceswereflexible,itappearsthatagentsneedtocoordinateontheoverallcredit issue. Lack of initial coordination could show up as overissue or underissue of the correct financial instrument,orofthecombinationofinstrumentsthatissupposedtogivethecorrectmaturitystructure, showing up in turn later on as sharp movements in prices. This could even lead to doubts about the commitment or ability of agents to achieve the requisite transfers of purchasing power necessary for liquidity in intermediate periods, or to ensure redemption of debt at maturity. Manuelli and Sargent ponderwhetherfiatmoneycanhelpsolvethistypeofcoordinationproblem. 20

3.4LackersModelofClearingandSettlementandInterAgentMarkets
Lacker(1997)focusesonclearingandsettlement,forhimviaacentralbank,andtheimpactofcertain centralbankpoliciessuchasreserverequirementsandinterestpaidonreserves.Buildingontheearlier modelsofTownsend(1983,1989),hedevelopsamodelinwhichthereisalargenumberNofislands,on each of which live N individuals. Each island produces a single perishable good which must be consumed on the island. This geography is illustrated for the case N = 3 in panel (a) of Figure 12, in whichtheislandsarelabeledA,B,andC,andindividualsare1,2,and3.Eachperiod,allbutoneofthe individuals who live on a given island travel to all the other islands at random, one to each, with one stayingbehind.Inpanel(b)ofFigure12,itisindividual3whoremainshome.Allindividualsconsume thegoodthatisproducedontheislandtheyvisit(sotheonewhoremainsconsumesthegoodproduced onhisisland).AsinTownsend(1989),beforetheyleavehome,travelersentrusttheirendowmentof goodstotheindividualwhostaysbehind(calledthemerchantbanker),whoisresponsibleforhandingit outtoarrivalsfromotherislands.Therecordoftheamountentrustedtothestayathomeagentisan individuals deposit and the merchant banker is thought of as operating a bank holding his islands deposits. AsillustratedinFigure12,eachislandreceivesafullydiversifiedgroupofvisitorseachperiod,onefrom eachotherisland.Ifpreferencesandendowmentsweresuitablyfixed,e.g.,ifeachindividualconsumed 1/Nunitsofthegoodoftheislands/hevisited,consumptionandincomewouldbalanceonapersonby personbasis.However,Lackerassumesasinsomeofthemodelsabovethateachperiod,islandsarehit by DiamondDybvig idiosyncratic i.i.d. preference shocks that affect the urgency of consumption. All individuals from a given island get the same shock. Since each island is visited by an individual from every other island and since shocks are independent across islands, there is no aggregate uncertainty aboutthedemandforeachislandsgood(asNgoesto).However,anislandthatsuffersarunoflarge urgentshocksovertimeconsumesmoreovertimethanonethatsuffersarunofsmallshocks. Because goods do not move between islands, there is no possibility to directly exchange one for another. Instead, an individual purchases consumption from the merchant on the island he visits by providingabillorcheckdrawnonhisdepositheldbyhisownmerchantwhostayedathome.(Thiscould beanelectronicchargetotheeaccountbutagainnotonepaidinstantaneously.)Inturneachmerchant collectsbillsorecreditsfromallotherislands,oneforeachvisitor.Inanygivenperiod,someislands willconsumemorethantheyproduce(i.e.,issuemorebillsthantheycollectorbeleftwithnegativee balances), while for others the opposite will be true. Intertemporal trade between islands across periods,i.e.interbankborrowingandlendingofebalances,isthusefficient. Inthefinalstageoftheperiod,allthemerchantbankerstravelwiththeirbillstoacentrallocationand submit them (to each other) for payment (panel (c) of Figure 12). With cell technologies, physical meetings would not be necessary. Payment is effected through an accounting mechanism, with each islandsaccountbeingcreditedanddebitedaccordingtothebillsoremoneypresentedtoandbyit.The residualthatdoesnotcleariscarriedover,insurplusordeficit.

21

Figure12:Lacker'smodelofsettlement Implicationsformobilebanking
Lackerreferstothecentralinstitutionthatkeepstheaccountsofeachislandasthecentralbank,and these accounts are thought of as reserve accounts. However this could equally be a private clearing house run by Safaricom or some another independent entity, as the model focuses on the account keeping and clearing functions of the institution, not the issuing of money per se. Positive account balances with the institution are the liabilities of that institution, while negative balances represent overdrafts.Inthemodelbillsarecleared(i.e.,acceptedbytheclearinghouse/centralbank)attheend oftheperiod,andsettled(i.e.,depositstransferredbytheinstitutionfromoneislandsreserveaccount toanothers)attheendoftheperiodforacrossperiodborrowingandlending. Beyond Lackers model, limits on withinperiod bank overdrafts with the clearing house/central bank caninducesomebanks(oneswithapositivepreferenceshock)toborrowfromothers.Likewise,limits onovernightoverdraftscaninduceresidentsofislandsthathavehadastringofpositive,urgentshocks toconstraintheirconsumptionbelowtheefficientlevel,astheyareunabletoborrowenough.Lackers model is a useful motivation for thinking about another aspect of the MPESA system, in particular overallclearingandtherelatedinventorymanagementproblemfacedbyagents.Crucialwouldbethe kindofcontractualconditionsSafaricommightwanttospecify,giventherealityoftheactualeconomy inwhichthedistinctionbetweenwithinperiodandacrossperiodclearingandborrowing/lendingishard tomaintain. WeidentifyeachMPESAAgentwithamerchantbankerinLackersmodel,althoughindividualsarenot bound to agents as residents are to islands. An MPESA Agents trading account at Safaricom 22

corresponds directly to the reserve account held by each bank with the central bank. To parallel the model,eachperiodindividualsdeposittheirendowments(ofcash)withanagent,whichrequiresthat the agent hold sufficient emoney. An agent would take out an overdraft loan from Safaricom if he wererequiredtoissueemoneytoacustomerbeforehavingpresentedtheequivalentamountofcash to Safaricom. As transferring a bank note or cash is slower than transferring emoney, it seems likely thattherecouldbedemandforsuchoverdrafts. Emoneyissentbetweenindividuals(i.e.,checksareexchanged),andrecipientspresenttheiremoney (i.e.,checks)toagents.ThishappensattheendoftheperiodinLackersmodel.Ifagentshaveenough cash to purchase the emoney from customers, their trading accounts are credited with the relevant amounts. In reality, as individuals visit the agent over the day, his net demand for emoney will fluctuate, and he might require shortterm overdrafts from Safaricom or acquire cash in some other way. In the absence of such a facility, he will need to trade off the costs of holding zerointerest reserves (emoney balances on his trading account) against the costs of reduced trade (and commissions).Alternatively,agentscouldlendemoneybetweenthemselves,creatingtheequivalentof an interbank market as envisioned in Lackers model. Again, this might be organized by another institution(likeaclearinghouse)thatitselfpurchasedemoneyfromSafaricomandlentitouttoagents atsomeinterestrate.Likewise,theheadofficesorsuperagentscouldperformthisrole,thoughneither appearstochargeinterest.Eachheadofficeorsuperagentwouldfaceasimilarinventorymanagement problemofcourse,havingtoholdenoughemoney,and/orcashtolendoutduringtheday/period.

3.5FreemanandGreensModelsofLiquidityOptimalBaseMoney Management
Freemans (1992) model and Greens (1996) reformulation, related to Townsend (1989) as exposited above,focusongettingmoneyandcirculatingdebtinthesamesettingsimultaneouslyduetoimperfect meetingsbetweencreditorsanddebtors.This,again,hasimplicationsforliquidityandmonetarypolicy. In Greens OLG model, there are two types of individuals (Creditors and Debtors) who live for two periods each. A creditor is someone who in equilibrium will be willing to defer consumption, while a debtorwillwishtoborrow.Wefollowtraditionandrefertoyoungandoldagents,simplytoimplythe first and second periods of the twoperiod, dynamic transactions profiles of the households. When young, Creditors and Debtors are endowed with perishable goods x and y, respectively. In the first period, old creditors are endowed with fiat money, and old debtors have nothing. Creditors and Debtorsdifferalsointheirpreferences:creditorswishtoconsumewhentheyareyoungandold,while debtors wish to consume only while young. Both types prefer to consume a mix of goods x and y insteadofjusttheirown. SupposeyoungDebtorsmeetyoungCreditorsfirst,andonlythengoontomeetoldCreditors.Young DebtorspurchasexinreturnfordebtdthattheyissuetoyoungCreditors,asillustratedinthefirstpanel

23

of Figure 13. 29 Subsequently, young Debtors sell their own good y to old Creditors in exchange for money.Atthebeginningofthenextperiod(thesecondpanelofFigure13),nowold(previouslyyoung) Debtors settle their debts using money with nowold (previously young) Creditors. Once the debt is settled, the process repeats, with the nowold Creditors holding money, and the new young cohort endowedwithgoods. 30

Figure13:TradewithMoneyandCredit
Nontrivial monetary dynamics can arise when Creditors and Debtors do not necessarily meet at the righttime.Withvariouswavesofmovers,oldagentseitherarrivelateorleaveearly:inparticular,a fraction(1)ofDebtorsarrivelate,andafractionofCreditorsleaveearly.Thisnaturallycomplicates thedebtsettlementprocess,andcanleadtoinefficiencies.Inparticular,whileefficiencyrequiresthat allCreditorsconsumethesamequantityofgoodswhenold(purchasedfromyoungDebtors),thosewho leave early may in equilibrium consume less than this amount while those who leave later consume more.ThusearlyleavingCreditorscanendupfacingliquidityshortageswhichconstraintrade. ThissituationisillustratedinFigure14,inwhichperiodt+1isdividedintotwosubperiods(t+1,inthe secondpaneldown,andt+1,inthethirdpanel).Agentsnotinthemarketattherelevantmomentin timeareshownasboxeswithbrokenlines.ItisassumedthattheCreditorsarefullydiversifiedatt+1, holding debt issued by each and every old Debtor. At the beginning of t+1, all the old Creditors are present,asareafraction ofoldDebtors.TheDebtorssettletheirdebts,eachCreditorreceivesashare <1oftheamountowedtohim.Atdatet+1thelatearrivingoldDebtorsareabletosettletheirdebts infullwitholdCreditorswhoremain.EarlyleavingoldCreditorswillconsumelessthantheirefficient levelofconsumption,andlateleavingoneswillconsumemore.

29 30

Thedateswithineachboxrefertotheperiodofbirth. Inthefigure,debtsettlementandtradebetweentheyoungandoldoccurinthesameperiod,t+1.

24

Figure14:LateArrivingDebtorsandEarlyLeavingCreditors
AnalternativescenarioillustratedinFigure15allowsoldCreditorstoexchangedebt.Periodt+1isnow dividedintot+1,t+1,andt+1.Atthebeginningoft+1(thesecondpanel)earlyarrivingDebtorsrepay theirdebtstoallCreditors.Asbefore,allCreditorscontinuetoholdoutstandingdebtissuedbyDebtors whohavenotyetarrived.Next,attimet+1(thethirdpanel)earlyleavingCreditorsselltheirremaining holdingsofdebttolonglivedCreditorsinexchangeformoney.If<,witharelativescarcityofDebtors andanabundanceofearlyleavingCreditors,theCreditorsspendalloftheirmoneyondebt,whichhasa price less than one. Earlyleaving Creditors then purchase goods from young Debtors, and quit the market.Inthefinalpanel,attimet+1,latearrivingDebtorssettletheirdebtwiththoseCreditorswho remain, including the debt those Creditors bought from earlyleaving Creditors. They then purchase goods from young Debtors. The amount of consumption they enjoy is (1)/(1) times the efficient level.Thus,ifnotenoughDebtorsshowupintime,evenifCreditorstradetheirdebttheallocationis inefficient. As Freeman observed, a central bank can remedy the inefficiency by issuing money to some or all Creditors, and then withdrawing it from circulation later, say by taxing young Creditors as they enter thesecondperiod.Theimportantissueisthatnewmoneyisissuedtocreditors,andthatitbeissued beforetheearlyleavingCreditorsdepart.

25

Figure15:TradeinMoneyandCredit,WithaSecondaryDebtMarket Implicationsformobilebanking
In Green's version of the model, Debtors consume nothing in the second period of their lives, but Creditorsdo.TheonlyreasonforoldDebtorstocometothemarketistopayofftheirdebts.Soife money allows them, or some of them, to do this without coming to market, then the share that are "late"issmaller.Intheextremecase,therewouldbenolatearrivingdebtors,andnoliquidityproblems for creditors. But if some old debtors still didn't pay off their debts in time (maybe because they couldn't find an agent with emoney) then it would be possible that earlyleaving Creditors wouldn't haveenoughmoneytofinancetheefficientlevelofpurchasesfromyoungdebtors. Thesemodelscaninformourthinkingaboutmobilebankinginacoupleofways.First,focusingonthe reductionintransactionscostsassociatedwithtransferringemoney,mobilebankingmightreducethe proportions (1) of Debtors who arrive late and the proportion of Creditors who leave early. Debtors who previously had to physically meet their creditors in order to settle their debts can now settlethemwithemoney,sodonotneedtobepresent.Ontheotherhand,evenifnotalldebtsare repaidatthebeginningoftheperiod(i.e.,ifthereremainsomelatearrivingDebtors),theexistenceof

26

emoney could relax the liquidity constraint faced by earlyleaving Creditors, and make central bank interventionlessnecessary. However it is overlysimplistic to assume that mobile banking allows individuals to send money costlessly:itallowsthemtosendemoneycostlessly(oratleastatlowcost),buttheymustacquirethe emoneyfirst.Amorecompletemodelwouldthusincludeindividualsholdingoptimalmixesofmoney andemoney,andwoulddescribetheproductionprocesswherebyeachisconvertedintotheother.In practice,thisconversioniseffectedbyMPESAAgents,whosimplyperformtheroleofatechnological blackbox,butablackboxthatissometimesoutofservice. AlthoughthisfeatureisnotpartoftheFreemanandGreensetup,ifmoneyandemoneyarebothused in equilibrium, then a late arriving Debtor might correspond to an individual who is otherwise on time and has sufficient financial resources (money and/or emoney) to repay his debts, but who is frustratedinnotbeingabletofindanMPESAAgentwithsufficientemoneyormoney. 31 Similarly,an earlyleavingCreditorinthisenvironmentcouldbeonewhohasinfactbeenrepaid,sayinemoney,but who must use money to purchase the consumption good. 32 If he cannot find an MPESA Agent with sufficientcash,thenshecouldbeliquidityconstrainedasabove.First,ifheisstuckwithemoneybut musttradewithmoney,shewillsufferalossequaltohisexcessemoneyholdings.Ontheotherhand, even if she cannot find an MPESA Agent to trade with, she might trade her emoney with another Creditorforcash,justasearlyleavingCreditorsselltheirdebttolateleavingCreditorsforcashinthe thirdpanelofFigure15.Butsuchtradesmusttakeplacebetweenlocationallyproximateagents,andif thereislocallyanexcesssupplyofemoney,theallocationofconsumptionmightremaininefficient.

4.LinkingTheorywithData:ResultsfromHouseholdandAgentSurveys
In this section we present data that speaks to some of the issues raised by the models of money summarizedabove,especiallyasregardsshortagesofemoneyandcashandwhetherthereareindeed credits in the system because of the operational logistics of agents as described in Section 2 above. These data, some of which are reported in Jack and Suri (2009), derive from a survey of 3,000 households and 250 MPESA Agents in Kenya in late 2008. 33 We focus on issues related to MPESA Agents,asreportedbyconsumersandtheAgentsthemselves,againasmotivatedbythemodels.We alsolookatsomeofthecommontransactionsthatconsumersmakeusingMPESA. First,10%ofallconsumersreportedfacingatleastoneproblemwiththeAgentstheyhadvisited.Of thosewhoreportedproblems,Table3showsthebreakdownoftheproblemstheyhad.Byfar,themost
WhethertheDebtorneedstofindanMPESAAgentwithmoneyoremoneywilldependonwhatformoffinancial wealththeDebtorhasonhand,andhowtheCreditorwishestobepaid.Thisinturnwilldependonthespecificfeatures ofthetwomonies. 32 Again,whethertheCreditorneedsmoneyoremoneydependsonhowthesellerwantstobepaid. 33 PartofthesedataformthebasisofaconfidentialreportissuedbyFSDtotheCentralBankofKenya.Inaddition,Jack andSuri(2010)lookatsomeofthemicroeconomicrisksharingimpactsofMPESA.Otherpapershavealsolookedat themoremicrolevelimpactsofemoneythoughoncurrencydemand,forexampleseeFujikiandTanaka(2010).
31

27

commonproblemsareAgentslackofcashandemoney.Thefirstfourrowsinthetablebelow,infact, suggestthatinsomecasesAgentshavebeenabletoincreasethepriceofemoney,byvaryingthefees theychargeconsumers.Thisisanimportantimplicationofthediscussionabovefixingpricesforcash and emoney will require an accompanying policy stance. However, the penultimate two rows in the tablebelowconfirmthatthisstrategyisemployednowherenearenoughtoclearthemarket.

Table3:TheProblemsConsumersHaveHadwithAgents
Agentgavelessmoney/emoneythanIwasowed Agentchargedmetodeposit Agentoverchargedme Agentunderchargedme Agentwasabsent Agentrefusedtoperformedthetransaction Agentwasunknowledgeable Agentwasrude Agenthadnoemoney/notenoughemoney Agenthadnocash/notenoughcash Other MostUsedAgent 2.63% 1.11% 1.17% 0.52% 0.74% 0.80% 1.04% 3.66% 43.60% 34.78% 9.95% ClosestAgent 2.80% 1.68% 1.84% 0.78% 0.91% 0.61% 1.57% 6.02% 22.81% 51.33% 9.65%

Inadditioninthesurvey,consumerswerespecificallyaskediftheywereeitherunabletodepositmoney orunabletowithdrawmoneyattheMPESAAgentclosesttothemorattheAgenttheyusedthemost. Table 4 34 shows that approximately 6 percent of MPESA users are unable to deposit money with an Agent,i.e.theAgent doesnot haveanyemoneytogive theconsumerinreturn.And,as manyas15 percent of consumers were unable to withdraw money from the closest Agent, i.e. the Agent had no cashtogivetheconsumerinexchangeforemoney.

Table4:UnabletoDepositCash(NoEmoney)orUnabletoWithdrawCash(NoCash)
HaveyoueverbeenunabletodepositmoneyatthisAgent? Yes No Total

MostUsedAgent ClosestAgent 6.63% 93.37% 100.00% 6.22% 93.78% 100.00%

HaveyoueverbeenunabletowithdrawmoneyfromthisAgent? Yes No Total

6.63% 93.37% 100%

15.33% 84.67% 100%

34

NotethatthequestionsaskedforTables3and4arequitedifferent.Table3asksaboutthemainconsumerreported problemswithAgents,whileTable4asksabouttheincidenceoftwospecificproblems.

28

In the survey of Agents themselves, respondents were asked how often they run out of emoney and howoftentheyrunoutofcashtheseresultsarereportedinTables5and6below.Onaverage,about 29 percent of Agents run out of emoney once a month or more frequently and indeed a nontrivial fraction(14percent)runoutaboutonceaweek.Similarly,about26percentofAgentsrunoutofcash once a month or more frequently than that and about 10 percent run out once a week (and, in fact, about8percentrunoutonadailybasis).Clearlythereareliquidityissuesbothintermsofcashaswell asintermsofemoney,asanticipatedinsomeofthemodelsreviewedabove.

Table5:HowOftendoAgentsRunOutofEmoney?
Morethanonceaday Onceaday Onceaweek Onceamonth Onceeverythreemonths Onceeverysixmonths Lessoftenthanthat Never Fraction 3.2% 6.4% 14% 5.6% 1.2% 0.4% 12% 57.2%

Table6:HowOftendoAgentsRunOutofCash?
Morethanonceaday Onceaday Onceaweek Onceamonth Onceeverythreemonths Onceeverysixmonths Lessoftenthanthat Never Safaricom initially required all MPESA Agents to prepurchase emoney before they can trade it for money to the public. If an Agent runs out of emoney, he is required to purchase more, either from Safaricomorfromthepublicwhentheyredeemcash,beforebeingabletotakecashdepositsfromthe public. This suggests that there are no credits or debits involved between agents and Safaricom, and therefore no role for a formal settlement system. Indeed, even as the agent model has evolved, this feature has been maintained, at least with respect to the relationship between the coordinating bodiesofFigure7andSafaricom.Ontheotherhand,cashandemoneytransactionsbetweenagents and their headoffices or aggregators need not remain in continuous balance, and the parties can operateinanetcreditordebitpositionvisviseachother.Theseimbalancesareoflittleconcernfor theHeadOfficemodel(panel(a)ofFigure7),totheextentthattheagentsareownedandcontrolled 29 Fraction 3.2% 8.4% 10% 4.8% 1.2% 0.4% 22.4% 49.6%

closelyenoughthatinternalfinancialarrangementofthegroupdonotaffectitsviability.However,the more arms length relationship between agents and an aggregator (panel (b)) suggests that chronic imbalanceswithsuchagroupcouldproveproblematic. We note that while the potential exists for persistent financial imbalances within a group under the aggregatormodel,inprincipleMPESAusersontheonehand,andSafaricomontheother,donotface anyriskassociatedwiththebankruptcyofanyparticularagentoragentgroup,asdepositsofcashare matchedatthelevelofthecoordinatingbodybytransfersofemoney,andviceversa.Ifanindividual userfindsthatallagentswithinareasonabledistancegooutofbusiness,hewilllikelyfacealiquidity constraint,unlessheisabletousehisemoneytodirectlypurchasegoodsandservices. ThesurveyaskedAgentshowtheypayforemoneywhentheyrequestitfromtheirheadoffice.Inwell over half the cases, Agents receive transfers from their head offices without any immediate correspondingpayments(seeTable7below).

Table7:HowisEmoneyPaidForWhenanAgentRequestsIt?
Itisadirectpurchaseofemoneyfromtheheadoffice(involvesacashtransfer) Receiveadirecttransferfromtheheadofficewithnoconcurrentpayment Receiveadirecttransferfromtheheadofficewithnopaymentrequired Other Refusedtoanswer Total

Fraction 36.2% 31.2% 18.1% 12.6% 2% 100%

Similarly,AgentswereaskedhowtheygetcashforMPESAtransactionswhentheyrunout.Asreported inTable8,inmorethanhalfofcasesAgentsdonotimmediatelyexchangeemoneyforcashreceived.

Table8:HowDoAgentsGetCashWhenTheyRunOutofIt?
Redeemfromheadofficeinexchangeforemoney Redeemdirecttransferofcashfromheadofficewithnoemoneyexchange Useownsavings Other* Dontknow Total Fraction 17.6% 20.4% 42.8% 18% 1.2% 100%

*Ofwhich27%isfromsaleofcreditcard,27%iswaitfordeposits,21%isborrowfrommanagementand11%isfrom theotherbusiness

ThestatisticsinTables7and8suggestthatcreditarrangements,explicitorotherwise,betweenAgents andtheirheadofficesoraggregatorsappeartobewidespread.Wedonotknowthematurityofthese credits,butgiventhelargenumberofAgentsreportingthem,itispossiblethatatleastsomeofthem 30

arelongerthansimpleovernightpositions.Indeeditseemsinevitablethattheremaybeanontrivial amountofcreditinthesetransactionsasthesupplychainofemoneyinvolvesexchangesinspatially and temporally separated markets, leading naturally to one way transfers of either cash or emoney. However,giventhemechanicsofMPESA,thesecreditscannotbeissuedbetweenAgentsandindividual users, or between the coordinating body and Safaricom. Nonetheless, the evidence above clearly illustratesthatsuchnetcredits/debitsmayexistbetweenAgentsandtheircoordinatingbodies.

5.ConclusionandFurtherModeling
The most successful version of mobile banking in Kenya (and perhaps the world), MPESA, is quite literally everywhere. In many cases, the scenarios envisioned in existing monetary theory models appeartomatchtherealityofMPESA,andassuchthesemodelspromisetoinformdecisionstakenby both Safaricom in managing MPESA, and the Central Bank in managing the Kenyan economy. The empirical evidence presented, from surveys of both MPESA users and Agents, further serves to illustratetheimportanceoftheselessons. For example, just as the Central Bank may intervene to relax liquidity constraints, it is arguable that Safaricom should actively manage eliquidity, by issuing emoney that is at some times in some locations,unbackedbymoneydeposits,assumingthatsuchactivismwouldbecostlessandallowedby the Central Bank. In fact, the data suggest that some MPESA Agents are engaging in such eliquidity managementalready,forexamplewhentheyreceiveemoneytransfersfromtheirheadofficeswithout a corresponding transfer back of cash. This has implications, of course, for the measurement and meaningofmonetaryanddebtaggregates.Improvedsystems,however,requirethatthecompanyhave betterinformationonnetdemandsforemoneyacrossAgentsthantheAgentsthemselveshave,orat leastbebetterabletoactonthisinformation,withoutthespace/timecoordinationproblemsthatthe modelssuggest.Notsurprisingly,SafaricomhasbeenchangingtheirAgentmodelovertimetobetter deal with cash and emoney liquidity issues. Time series and geographically disaggregated data on fluctuationsindemandwouldbeusefulforfurtherevaluatingtheseissuesandmakingimprovementsto theirsystem. On the modeling side, understanding the operations of the MPESA Agent network seems key to the developmentofanoverall,comprehensivemodelofemoney.Forexample,modelingthedecisionsand constraintsofAgentswouldpotentiallyallowustoendogenizethetimingpatternsassumedinsomeof theexistingmodels.Frictionsthatimpedeefficientandimmediatereallocationsofmoneyandemoney balancesacrossAgentswouldtherebyreplacethesetimingassumptionsasthefundamentalsourceof liquidityconstraints.Similarly,realisticheterogeneityacrossconsumersforexample,intermsofphone ownership, access to mobile coverage, safety of the local environment, frequency of market access, accesstoMPESAagentscouldbemodeledmoreexplicitly

31

Bibliography
Bewley, Truman, (1980), The Optimum Quantity of Money, in Models of Monetary Economics, ed. JohnKarekenandNeilWallace,FederalReserveBankofMinneapolis. Bewley,Truman,(1983),ADifficultywiththeOptimumQuantityofMoney,Econometrica,51:1485 1504. Calvacanti,RicardoandNeilWallace,(1999),AModelofPrivateBankNoteIssue,ReviewofEconomic Dynamics,2(1),104136. Cass, David and Yaari, Menahem, (1966), "A ReExamination of the Pure Consumption Loans Model," JournalofPoliticalEconomy,August1966,74,35367. CentralBankofKenya,(2009),PresentationatConferenceonBankingandPaymentTechnologies,East Africa,1719February2009. Diamond,DouglasandDybvig,(1983),BankRuns,DepositInsuranceandLiquidity,JournalofPolitical Economy,91(3):401419. FinancialSectorDeepening,FinaccessReports,Various. Foster, Kevin, Erik Meijer, Scott Schuh, and Michael A. Zabek, 2010 The 2008 Survey of Consumer PaymentChoice,FederalReserveBankofBoston,PublicPolicyWorkingPaper,January2010. Freeman, Scott, (1992), The Payments System, Liquidity, and Rediscounting, American Economic Review,86(5):11261138 ___,(1996),ClearinghouseBanksandBanknoteOverIssue,JournalofMonetaryEconomics,38:101 115. Fuerst,T.S.,(1992),Liquidity,LoanableFunds,andRealActivity,JournalofMonetaryEconomics,29: 324. Fujiki, Hiroshi, and Migiwa Tanaka, (2010), Currency Demand, New Technology and the Adoption of ElectronicMoney:EvidenceUsingIndividualHouseholdData,WorkingPaper. Green, Edward, (1999), Money and Debt in the Structure of Payments, Federal Reserve Bank of MinneapolisQuarterlyReview,23(2):1329. Grossman, Sanford J., and Laurence Weiss, (1983), A Transactions Based Model of the Monetary TransmissionMechanism,AmericanEconomicReview,73(5):871880. Ireland, Peter, (1994), Money and the Gain From Enduring Relationships in the Turnpike Model, FederalReserveBankofRichmondWorkingPaperNo9407. Jack,William,andTavneetSuri,(2010),TheRiskSharingBenefitsofMobileMoney,WorkingPaper. Jack,William,andTavneetSuri,(2009),MobileMoney:TheEconomicsofMPESA,WorkingPaper. Kocherlakota,Narayana,(2005),DiscussionofFromPrivateBankingtoCentralBanking:Ingredientsof aWelfareAnalysis,InternationalEconomicReview,46(2),633636. Kocherlakota, Narayana, and Neil Wallace (1998), Incomplete Record Keeping and Optimal Payment Arrangements,JournalofEconomicTheory,81(2),272289. Lacker,Jeffrey,(1997),Clearing,SettlementandMonetaryPolicy,JournalofMonetaryEconomics,40: 347381. Lucas,Robert,(1980),EquilibriuminaPureCurrencyEconomy,EconomicInquiry,18(2):203220. Lucas,Robert,(1990),LiquidityandInterestRates,JournalofEconomicTheory,50. 32

Mas, Ignacio and Morawczynski, (2009), Designing Mobile Money Services: Lessons from MPESA, Innovations,4(2),7792,MITPress. Manuelli,Rodolfo,andThomasSargent,(2009),AlternativeMonetaryPoliciesinaTurnpikeEconomy: VintageArticle,WorkingPaper,NewYorkUniversity,June2009. PerezVerdia,Carlos,(2000),TransactionCostsandLiquidity,PhDDissertation,UniversityofChicago, 2000. Romer,David,(1987),TheMonetaryTransmissionMechanisminaGeneralEquilibriumVersionofthe BaumolTobinModel,JournalofMonetaryEconomics,20:105122. Rotemberg,JulioJ.,(1984),AMonetaryEquilibriumModelwithTransactionsCosts,JournalofPolitical Economy,92(11):4053. Townsend,Robert,(1980),ModelsofMoneywithSpatiallySeparatedAgents,inModelsofMonetary Economics,ed.JohnKarekenandNeilWallace,FederalReserveBankofMinneapolis. ___,(1983),FinancialStructureandEconomicActivity,AmericanEconomicReview,73(5):895911. ___,(1987),EconomicOrganizationwithLimitedCommunication,AmericanEconomicReview,77(5): 954971. ___,(1989),CurrencyandCreditinaPrivateInformationEconomy,JournalofPoliticalEconomy,97(6): 13231344. Townsend, Robert and Neil Wallace, (1982), A Model of Circulating Private Debt, Staff Report 83, FederalReserveBankofMinneapolis. Wallace, Neil, (2005), From Private Banking to Central Banking: Ingredients of a Welfare Analysis, International Economic Review, 46, 619631.Wicksell, Knut, (1935) Lectures on Political Economy, London:MacMillan,1935. Wallace, Neil, (2000), Knowledge of Individual Histories and Optimal Payment Arrangements, QuarterlyReview,FederalReserveBankofMinnesota,24(3):1121.

33