MobileMoney: TheEconomicsofMPESA1

WILLIAMJACK 2 GEORGETOWNUNIVERSITY
AND

TAVNEETSURI 3 MITSLOAN October,2009

We gratefully acknowledge the support and collaboration of Pauline Vaughan and Susie Lonie, and other staff of SafaricomandVodafone.ThesurveywhoseresultsarereportedherewascommissionedbyFinancialSectorDeepening, a Nairobibased multidonor financial sector development programme, on behalf of the Central Bank of Kenya. The surveyitselfwasadministeredbytheSteadmanGroup,alocalsurveyfirm.ThanksareextendedtoStephenMwauraof the CBK, David Ferrand and Caroline Pulver of FSD, and to Carol Matiko and Moses Odhiambo of Steadman, and to seminarparticipantsatMITSloanandSafaricom. 2 wgj@georgetown.edu 3 tavneet@mit.edu
1

I.Introduction
Mobilephonetechnologyhasreducedcommunicationcostsinmanypartsofthedevelopingworldfrom prohibitivelevelstoamountsthatare,incomparison,virtuallytrivial.Nowherehasthistransformation beenasacuteasinsubSaharanAfrica,wherenetworksofbothfixedlinecommunicationandphysical transportation infrastructure are often inadequate, unreliable, and dilapidated. While mobile phone callingratesremainhighbyworldstandards,thetechnologyhasallowedmillionsofAfricanstoleapfrog thelandlineenrouteto21stcenturyconnectivity. Earlyoninthisrevolution,cellphoneusersfiguredoutthattheycouldeffectivelytransfermoneyacross wide distances. Phone companies have long allowed individuals to purchase airtime (i.e., prepaid cell phone credit that can be used for voice or SMS communication) and to send this credit to other users.Itwasasmallstepfortherecipientusertoonsellthereceivedairtimetoalocalbrokerinreturn forcash,orindeedforgoodsandservices,thuseffectingatransferofpurchasingpowerfromtheinitial sendertotherecipient. InMarch2007,theleadingcellphonecompanyinKenya,Safaricom,formalizedthisprocedurewiththe launch of MPESA, an SMSbased money transfer system that allows individuals to deposit, send, and withdrawfundsusingtheircellphone.MPESAhasgrownrapidly,currentlyreachingapproximately38 percentofKenyasadultpopulation,andiswidelyviewedasasuccessstorytobeemulatedacrossthe developingworld. Thispaperprovidesadescriptionoftheserviceandareviewofthepotentialeconomiceffectsprimarily atthehouseholdlevel,butalsointermsofmacroeconomicandmonetaryaggregates.Itthenprovides adetailedportrayalofpatternsofuseacrossurbanandruralpopulations,usingdatafromthefirstlarge householdsurveyfocusedonmoneytransferservicesinKenya. 4

II.Context
MobilephonesandmobilebankinginKenya
Theadoptionofmobilephoneshasoccurredatperhapsthefastestrateandtothedeepestlevelofany consumerlevel technology in history. Figure 1 illustrates the speed of adoption compared with a varietyofproductinnovations.Whilecumulativeforcesareofcourseimportant,makingitdifficultto compare directly across innovations, it is nonetheless informative to note that cell phones have been adoptedmorethanfivetimesasfastasfixedlinetelephoneservices,whichtook100yearstoreach80 percentofcountrypopulations.
Mobilepaymentsystemshavealsobeendevelopedinotherdevelopingcountries.InthePhilippinesGlobeTelecom operates GCASH, and in South Africa WIZZIT facilitates mobile phonebased transactions through the formal banking system(IvaturyandPickens,2006).SimilarlymobilebankingtechnologieshavedevelopedinSudanandGhana,andina numberofcountriesisLatinAmericaandtheMiddleEast(Mas,2009).Forrelatedoverviews,seealsoMasandRotman (2008) and Mas and Kumar (2008), as well as other publications of the Consultative Group to Assist the Poor, at www.cgap.org.
4

JackandSuri


Mobilephones CATscan Internetuse Personalcomputers Aviation Radio Steel(electrichearth) Telephones Steel(openhearth) Railways 0 20 40 60 80 Years 100 120 140

Figure1:Technologyadoptionforselectinnovations(numberyearstoreach80%coverage) 5
One of the reasons mobile phone technology has spread quickly is that it has followed other technologiesthatmayhaveeasedtheway.Figure2confirmsthissequencingpropertyislikelyatwork, at least in the US: many of the new technologies that were introduced before about 1950 (with the exceptionofradio)wererelativelyslowtodiffusethroughthepopulation,whereasthoseintroducedin the second half of the century saw generally steeper adoption rates. Nonetheless, the speed of adoptionofcellphones,especiallyinthedevelopingworld,remainsunprecedented.

Figure2:Technologyadoptionisgettingfaster 6

5 6

DatafromWorldBank. Source:NewYorkTimes,February10,2008.

JackandSuri

 ThespreadofmobilephonetechnologyhasbeenespeciallyrapidandbroadinAfricawherepenetration ratesstoodatsome32percentin2008,stillwellbelowtheglobalaverageof60percentatthattime, but much higher than the 7 percent coverage rate that prevailed just four years before. This pattern stands in contrast to the adoption of other technologies such as improved seed and fertilizer, which have been frustratingly weak. Since Solows (1956) seminal contribution to the theory of economic growth, and following later developments (e.g., Romer 1986 and Lucas, 1988), economists have understood that higher rates of adoption of modern technologies may accelerate the development process. In Kenya, the first mobile phone companies were publicly owned, and began operations in the mid 1990sonasmallscale.OvertimemobilephonesinKenyahaveeclipsedlandlinesastheprimarymeans oftelecommunication:whilethenumberoflandlineshadfallenfromabout300,000in1999toaround 250,000by2008,mobilephonesubscriptionshadincreasedfromvirtuallyzerotonearly17millionover the same time period (Figure 3). 7 Assuming an individual has at most one cell phone, 8 47% of the population,orfully83%ofthepopulation15yearsandolder,haveaccesstomobilephonetechnology.
18 16 Millionsofsubscribers 12 10 8 6 4 2 0 1998 2000 2002 2004 2006 2008 0% 2010 Percentofoursample 14 80% 60% 40% 20% 100%

Fixedlines Mobilelines MPESAusers Yearoffirstcellphoneuse(ourdata,righthandaxis)

Figure3:PhoneuseinKenya
Safaricom, which began operations in 1997, is currently the largest mobile phone operator in Kenya, controllingnearly80percentofthemarket,aheadofitstwonearestrivals(ZainandOrange).Recent
7

Figure3includesinformationontheshareofoursamplewhohadstartedusingacellphonebyyear.Theevolutionof thisfigurefollowscloselythatfromtheaggregatedataon cellphoneuse,providingpartialvalidation ofoursampling methodology. 8 Thisisnotquitetrue,assomeindividualsowntwo(ormore)phones,soastotakeadvantageofdifferenttariffpolicies ofthecompetingproviders.

JackandSuri

 andprospectiveentryintothesectorisexpectedtoputasqueezeonSafaricomsmarketshare,which somecommentators(includingitschiefexecutive)expecttofalltoaround65percentoverthenext3to 4years. 9 InMarch2007,followingadonorfundedpilotproject,Safaricomlaunchedanewmobilephonebased paymentandmoneytransferservice,knownasMPESA. 10 Theserviceallowsuserstodepositmoney into an account stored on their cell phones, to send balances using SMS technology to other users (includingsellersofgoodsandservices),andtoredeemdepositsforregularmoney.Charges,deducted fromusersaccounts,areleviedwhenefloat(thecurrencyinwhichMPESAbalancesaredenominated) issent,andwhencashiswithdrawn. MPESAhasspreadquickly,andhasbecomethemostsuccessfulmobilephonebasedfinancialservicein the developing world. 11 The average number of new registrations per day exceeded 5,000 in August 2007, and reached nearly 10,000 in December that year (see Figure 4). By August 2009, a stock of about7.7millionMPESAaccountshadbeenregistered.Ignoringmultipleaccountsandthoseheldby foreigners,thissuggeststhatabout38percentoftheadultpopulationhasgainedaccesstoMPESAin justover2years.
15,000

10,000 Numbernew usersperday

5,000

Figure4:AveragedailygrowthinMPESAregistrationsbymonth
MPESA is not a bank, a point that Safaricom has been at pains to stress, in part due to wariness regarding the potential regulatory response by the Central Bank of Kenya. On the other hand, the ubiquityofthecellphoneacrossbothurbanandruralpartsofthecountry,andthelackofpenetration
SeereportbyInternationalTelecommunicationUnion,http://www.itu.int/ITU D/ict/newslog/Safaricoms+Market+Share+To+Dip+From+80+To+65+As+Competition+Toughens+Kenya.aspx. 10 PesaisKiswahiliformoneyhenceM[obile]Money. 11 Similar services in Tanzania and South Africa, for example, have penetrated the market much less. See Mas and Morawczynski(2009).
9

JackandSuri

 ofregularbankingservices, 12 ledtohopesthatMPESAaccountscouldsubstituteforbankaccounts,and reachtheunbankedpopulation.Ourdata,presentedinmoredetailinthenextsection,suggestthisis partiallytrue,althoughMPESAhasbeenadoptedbyboththebankedandunbankedinroughlyequal proportions. 13 While the sustained growth in MPESA registrations is notable, the volume of financial transactions mediatedthroughMPESAshouldnotbeexaggerated. Table1reportsthatthevolumeoftransactions effectedbetweenbanksundertheRTGS(RealTimeGrossSettlement]methodisnearly700timesthe dailyvaluetransactedthroughMPESA.Ontheotherhand,theaveragemobiletransactionisabouta hundred times smaller than the average check transaction (Automated Clearing House, or ACH), and evenjusthalfthesizeoftheaverageAutomaticTellerMachine(ATM)transaction. 14 ThusMPESAisnot designedtoreplaceallpaymentmechanisms,buthasfoundandfilledanicheinthemarketinwhichit providessignificantlyenhancedfinancialservices.

Table1:Dailyfinancialtransactions,Oct2007Sept2008 15
RTGS ACH ATM Mobile Valueperday(billionKSh) 66.3 8.5 1.0 0.1 Transactionsperday(thousands) 1.0 39.2 180.2 107.2 Valuepertransaction(millionKSh) 64.67 0.216 0.006 0.003

HowdoesMPESAwork
AlthoughMPESAdoesnotpayinterestondeposits,anddoesnotmakeloans,itcanusefullybethought ofasabankthatprovidestransactionservicesandthathasoperated,untilrecently,inparallelwiththe formalbankingsystem. SafaricomacceptsdepositsofcashfromcustomerswithaSafaricomcellphoneSIMcardandwhohave registeredasMPESAusers.Registrationissimple,requiringanofficialformofidentification(typically the national ID card held by all Kenyans, or a passport) but no other validation documents that are typicallynecessarywhenabankaccountisopened.Formally,inexchangeforcashdeposits,Safaricom issues a commodity known as efloat, measured in the same units as money, which is held in an account under the users name. This account is operated and managed by MPESA, and records the quantityofefloatownedbyacustomeratagiventime.Thereisnochargefordepositingfunds,buta slidingtariffisleviedonwithdrawals(forexample,thecostofwithdrawing$100isabout$1).16 Figure5
In2006itwasestimatedthat18.9percentofadultsusedabankaccountorinsuranceproduct,andby2009thishad increasedto22.6percent.(FinaccessI.) 13 Inthetimesinceoursurveywasfirstadministered,therehasbeensignificantgrowthinthenumberofindividuals,and households,withabankaccount,duetotheexpansionofsuchinstitutionsasEquityBankandFamilyBank.Inaddition, anumberofbankshaveveryrecentlyallowedconsumerstolinkthereMPESAandbankaccounts.Howthesechanges haveaffectedtherelationshipbetweenMPESAregistrationandaccesstobankingservicesremainstobeseen. 14 ThesedatarefertoaperiodbeforeMPESAcouldbeusedatATMs. 15 Source: Central Bank of Kenya, presentation at conference on Banking & Payment Technologies East Africa, 1719 February2009,Nairobi. 16 Thecompletetariffscheduleisavailableat http://www.safaricom.co.ke/fileadmin/template/main/downloads/Mpesa_forms/14th%20Tariff%20Poster%20new.pdf.
12

JackandSuri

 illustratesthescheduleoftotalnettariffsforsendingmoneybyMPESA,WesternUnionandPostapay (operated by the Post Office). The MPESA tariffs include withdrawal fees, and are differentiated accordingtoreceiptbyregisteredandnonregistereduser.
1,400 1,200 1,000 Tariff 800 600 400 200 0 0 10,000 20,000 30,000 40,000

Amountdepositedandsent Postapay MPESA:Regtoreg MPESA:Regtononreg WesternUnion

Figure5:TotalnettariffratesfordepositingandsendingmoneybyPostapayandbyMPESA toaregistereduserandtoanonregistereduser
Efloat can be transferred from one customers MPESA account to another using SMS technology, or sold back to Safaricom in exchange for money. Originally, transfers of efloat sent from one user to another were expected to primarily reflect unrequited remittances, but nowadays, while remittances are still an important use of MPESA, efloat transfers are often used to pay directly for goods and services,fromelectricitybillstotaxicabfares.Thesenderofefloatischargedaflatfeeofabout40US cents,buttherecipientonlypayswhens/hewithdrawsthefunds.

Table2:Safaricomcelltowerdistributionbyprovince
Province Nairobi Rift Valley Coast East Central Nyanza Western North-East Total Towers 584 375 247 214 206 162 90 45 1923 Population per tower 4,872 22,448 12,046 24,871 19,048 30,771 46,122 29,467 17,653 Area per tower (sq mi) 0.5 179.0 130.7 288.5 24.7 38.5 35.9 1,088.8 117.0

Feesarechargedtotheusersaccount,fromwhichefloatisdeducted.Additionalcashfeesareofficiallynotpermitted, butthereisevidencethattheyaresometimeschargedonaninformalbasisbyagents.

JackandSuri

 Transfers are, of course, subject to availability of network coverage, which has expanded consistently overthepastdecade.Therearenownearly2,000Safaricomtowersacrossthecountry(inadditionto towers operated by competing providers), concentrated in areas of high population density. Table2 givesabreakdownbyprovince,andthemostrecentnetworkcoveragemapisshowninFigure6.

Figure6:Safaricomnetworkcoverage,September2009 17
To facilitate purchases and sales of efloat, MPESA maintains and operates an extensive network of over12,000agentsacrossKenya.AscanbeseeninFigure7,thegrowthofthisnetworklaggedbehind that of the customer base for the first year of MPESAs operation during which time the number of users per agent increased fivefold, from a low of 200 to a high of 1,000. But since mid2008, agent growthhasacceleratedandthenumberofusersperagenthasfallenbacktoabout600. RegisteredMPESAuserscanmakedepositsandwithdrawalsofcash(i.e.,makepurchasesandsalesof efloat) with the agents, who receive a commission on a sliding scale for both deposits and withdrawals. 18 MPESAagentsholdefloatbalancesontheirowncellphones,purchasedeitherfrom Safaricom 19 orfromcustomers,andmaintaincashontheirpremises.Agentsthereforefaceanontrivial inventory management problem, having to predict the time profile of net efloat needs, while maintainingthesecurityoftheiroperations.
17
18

Source:http://www.safaricom.co.ke/index.php?id=388

The commission amounts are nonlinear (and concave) in the size of the transaction. Some reports suggest that in response to this, agents encourage customers to split their transactions into multiple pieces, thereby increasing the overallcommission. 19 MPESArequiresthateachagenthasabankaccount,sothatfundscanbetransferredeasilybetweenthem.

JackandSuri


14000 12000 Numberofagents 10000 8000 600 6000 4000 2000 0 Apr07 Jun07 Aug07 Oct07 Dec07 Feb08 Apr08 Jun08 Aug08 Oct08 Dec08 Feb09 Apr09 Jun09 Aug09 400 200 0 1200 1000 800 Usersperagent

Numberagents(LHaxis)

Usersperagent(RHaxis)

Figure7:Expansionoftheagentnetwork 20
Inpractice,agentsareorganizedintogroups.Originally,MPESArequiredthatagentgroupsoperatedin atleastthreedifferentphysicallocations,sothattheprobabilityofimbalancesarisingwithinthegroup could be minimized. There are currently three agent group models in operation. In the first, one member of the agent group (the headoffice) deals directly with MPESA, while subsidiary agents, which are owned by the head office, manage cash and efloatbalances through transactions with the headoffice.BoththeheadofficeandtheagentscantransactdirectlywithMPESAusers. ThesecondmodelunderwhichagentsareorganizedintogroupsistheAggregatormodel.Thismodelis similar to the first, with the aggregator acting as a head office, dealing directly with Safaricom and managingthecashandefloatbalancesofagents.However,theagentscanbeindependentlyowned entities,withwhichtheaggregatorhasacontractualrelationship. A final and more recent model allows a bank branch, referred to as a superagent, to perform the functionsoftheaggregatorofthesecondmodel.Thebranchmanagescashandefloatbalancesofa group of nonbank MPESA agents, but unlike the regular and aggregator models, the bank does not tradeefloatdirectlywithMPESAusers. ThesuperagentmodelisoneexampleoftheintegrationofMPESAservicesintothebankingsystem. Otherdevelopmentsinthisveinhaveseenuserswithaccountsatcertaincommercialbanks(about72% of user households in our data have at least one bank account see Table 8 below), being able to transferfundsbetweenthoseaccountsandtheirMPESAaccounts,oftenviaATMs.

20

Source:Safaricom.

JackandSuri

 ThecashcollectedbyMPESAinexchangeforefloatisdepositedinbankaccountsheldbySafaricom. Originally, all funds were held in just one account at the Commercial Bank of Africa, but recently Safaricomhasopenedanaccountatasecondcommercialbanktodiversifyingitsrisk.Theseaccounts areregularcurrentaccounts,withnorestrictionsonSafaricomsaccesstofunds.Inturn,thebanksface nospecialreserverequirementswithregardtoMPESAdeposits,whicharetreatedasanyothercurrent accountdepositintermsofregulatorypolicyoftheCentralBank.Thereisnoexplicitrequirement,for example,forSafaricomtogivenoticeofitsintentiontowithdrawlargequantitiesofcashatagiven pointintime.AsMPESAcontinuestoexpand,andasthesebalancesgrow,theauthoritiesmaydecide to revisit this arrangement. An alternative approach, adopted in the Philippines, is to institute a 100 percentreserverequirementvisvismobilebankingdepositbalancesheldinaccountsatcommercial banks. The success of MPESA has rested in part on the trust that customers have in one of Kenyas mostwellrespectedprivatecompanies,theparent.Butiffaithinthebankingsystemerodes,arunon MPESA could be sparked, thereby jeopardizing the position of the banks in which it holds deposited funds. Becausetheyareheldinregularcurrentaccountsatcommercialbanks,MPESAdepositsinthebanking systemareinsuredundertheDepositProtectionFund. 21 Howeverthisdepositinsurance,designedfor individual bank account holders, provides insurance on deposits up to a maximum of KSh 100,000, or about$1,300.ThusMPESAdepositsarevirtuallycompletelyuninsuredagainstbankfailure. Finally,asMPESAdepositsenterthebankingsystem,theyonlyreducecashincirculationtotheextent that banks comply with or exceed official reserve requirements. But as efloat becomes more widely acceptable as an easily transferable store of value, it will adopt the features of money. The practical implicationofthisisthatMPESAcouldincreasetheeffectivemoneysupply,withpossibleimpactson inflation and /or output. Of theoretical interest is the possibility that two monies could coexist in equilibrium.Wewilladdresstheseissuesinmoredetailinfuturework.

III.Potentialeconomicimpactsonhouseholds
MPESA facilitates the safe storage and transfer of money. As such, it has a number of potential economiceffects.First,itsimplyfacilitatestrade,makingiteasierforpeopletopayfor,andtoreceive payment for, goods and services. Electricity bills can be paid with a push ofa few buttons instead of traveling to an often distant office with a fistful of cash and waiting in a long queue; consumers can quicklypurchasecellphonecredit(airtime)withoutmoving;andtaxidriverscanoperatemoresafely, withoutcarryinglargeamountsofcash,whentheyarepaidelectronically. Second,byprovidingasafestoragemechanism,MPESAcouldincreasenethouseholdsavings. 22 Third, because it facilitates interpersonal transactions, it could improve the allocation of savings across householdsandbusinessesbydeepeningthepersontopersoncreditmarket.Thiscouldincreasethe averagereturntocapital,therebyproducingafeedbacktothelevelofsaving.
21 22

Seehttp://www.centralbank.go.ke/dpfb/background.aspx Bynet,wemeannetoflossesduetotheft,etc.

JackandSuri

10

 Fourth,bymakingtransfersacrosslargedistancestriviallycheap,MPESAimprovestheinvestmentin, and allocation of, human capital as well as physical capital. Households may be more likely to send memberstohighpayingjobsindistantlocations(e.g.,thecapital),eitheronapermanentortemporary basis,andtoinvestinskillsthatarelikelytoearnareturninsuchplacesbutnotnecessarilyathome. Fifth, MPESA could affect the ability of individuals to share risk. Informal risksharing networks have been found to be an important, although not fully effective, means by which individuals spread risk, makingstatecontingenttransfersamonggroupmembers.Byexpandingthegeographicreachofthese networks, MPESA may allow more efficient risk sharing, although the riskreducing benefits might be mitigatedduetoissuesofobservabilityandmoralhazardwhenpartiesareseparatedbylargedistances. Sixth,afurtherriskrelatedeffectarisesifMPESAfacilitatestimelytransferofsmallamountsofmoney. Insteadofwaitingforconditionstoworsentolevelsthatcauselongtermdamage,MPESAmightenable supportnetworkstokeepnegativeshocksmanageable.Forexample,ahouseholdheadwithaccessto MPESAwhosuffersamildhealthshockmightreceiveasmallamountofmoneyviaMPESAthatallows himtokeephischildreninschool.Ifthismoneywasdelayed,orthesenderwaiteduntiltherecipient reallyneededit,thechildrenmighthavequitschool,theeffectsofwhichmaybehardtoreverse. Seventh,ifMPESAallowshouseholdstospreadrisk,theymaybeledtomakemoreefficientinvestment decisions,relaxingthetradeoffbetweenriskandreturnthattheywouldotherwiseface. MPESA could conceivably alter bargaining power and weaken incentives within households or other networks. Economically weaker family members might expect larger and more regular remittances frombetteroffcitydwellingrelatives,whothemselvesmightfindithardtojustifynotsendingmoney home.Thiscouldweakenincentivesforruralhouseholdmemberstoworkorinnovate,offsettingsome oftheefficiencyenhancingbenefitsofimprovedgeographiclaborallocationandrisksharing. Conversely, MPESA could have the effect of empowering certain household members who have traditionallyhadlessbargainingpower,inparticularwomen.Especiallyamongpoorersegmentsofthe population, remittances and transfers received (and sent) via MPESA are less visible than those transmitted by other means, such as delivery by a friend or relative. Granted this information advantage,recipientscouldbeinapositiontokeepmoreofthefundstheyreceive.Evidencesuggesting thespendingpatternsofwomenandmendiffer(see,e.g.,ChattopadhyayandDuflo,2004)thenimplies thattheadventofMPESAcouldhaverealeffectsontheallocationofhouseholdspending.Theseare issueswehopetoexploremorefullyinfuturework.

IV.Surveyanddata
Surveymethodology
InSeptember2008weundertookasurveyof3,000randomlyselectedhouseholdsacrossKenya.Atthe time,bothcellphonetowerandMPESAagentcoveragewereverylimitedintheremotenorthernand easternpartsofthecountry,sotheseareaswereexcludedfromthesampleframe.Thenonexcluded areacoveredbythesampleframeincluded92percentofKenyaspopulation,and98percentofMPESA JackandSuri 11

 agentsasofApril2008.Werandomlyselected118locations(thesecondsmallestadministrativeunit), in which there were 300 enumeration areas routinely visited by the Kenyan National Bureau of Statistics.Tenhouseholdsineachenumerationareawererandomlychosentotakepartinthesurvey theGPSrecordedlocationsofthesehouseholdsareshowninFigure8.Inordertoincreaseourchances ofinterviewinghouseholdsinwhichsomeoneusedMPESA,weoversampledlocationsonthebasisof thenumberofMPESAagentspresent.Allfigurespresentedbelowhavebeenreweightedaccordingly.

Figure8:Interviewedhouseholds(+Householdsvisited;lighterareashavehigherpovertyrates)
Duringtheinterviewswecollectedinformationonbasichouseholdcompositionanddemographicdata, dataonhouseholdwealthandassets,consumption,positiveandnegativeshocks,andremittances.We alsoaskedforinformationontheuseoffinancialservices,savings,etc.,andcollecteddetaileddataon cellphoneuseandknowledgeingeneral,anduseofMPESAinparticular.

Summarystatistics
Table 3 reports summary statistics of the households we interviewed, weighted so as to be
representativeofthe92percentoftheKenyanpopulationlivingintheareasfromwhichoursamplewas drawn.Thefirstpanelreportsthatabout44percentofhouseholdshadatleastonememberwhohad used MPESA at least once. By our definition, a user does not have to have a registered MPESA account,ass/hecouldusesomeoneelsesphonetomakeatransaction. The second panel reports householdlevel income and wealth indicators users report annual expenditures which are on average 67 percent higher than those of nonusers, and they report asset holdingsthatareonaverage21percenthigher. Figure9graphstheempiricalwealthdistributionsfor bothusersandnonusers.Thesedistributionssharealargelycommonsupport,althoughthereisclear evidencethatusersaretypicallywealthierandbetteroff. JackandSuri 12

Table3:Householdcharacteristics
Numberofhouseholds Shareoftotal Incomeandwealth AnnualHhldExpenditure(KSh) Assets(KSh) Wealthindex Othercharacteristics Shareofhouseholdswithatleastonecellphone Shareofhouseholdswithatleastonebankaccount Shareoftheunbankedpopulationineachcategory Shareofthebankedpopulationineachcategory Shareofruralpopulationineachcategory Shareofurbanpopulationineachcategory
Notes:Allfiguresreweightedaccordingly.Standarddeviationsin().

Nonusers 1,685 0.56 197,344 (318,896) 856,663 (3,151,751) 0.520 (1.599) 0.53 0.36 0.44 0.39 0.71 0.48

Users 1,315

0.44 329,348 255,211 (430,102) (377,428) 1,033,965 934,387 (3,530,126) (3,323,535) 0.667 0.000 (1.577) (1.695) 0.91 0.72 0.56 0.61 0.29 0.52 0.70 0.52 1.00 1.00 1.00 1.00

Allhhlds 3,000 1.00

Thebottompanelof Table3showsthat70percentofhouseholdsreporthavingacellphone, 23 and that userhouseholds are much more likely to own one (91 percent) than nonuser households (53 percent).Inoursample,52percentofallhouseholdshadatleastonebankaccount, 24 andtheshareof userhouseholdswithabankaccountwastwicethatofnonusers.Atthetimeofthesurvey,MPESA hadreached56percentofhouseholdswithoutabankaccount,and61percentofbankedhouseholds.
100% 80% 60% 40% 20% 0% 0 500,000 1,000,000 Wealth,KSh 1,500,000 2,000,000

Nonusers

Users

Figure9:Empiricalwealthdistributionsofusersandnonusers

23 24

77percentofhouseholdsreportedhavingusedone. Individuallevel data from other sources (e.g., Finaccess?) suggest that at the time of our survey, the banked population was at most 20 percent. However, that figure reflects the share of individuals with an account, and we believetheaccessofahouseholdtobankingservicesisperhapsmoresuggestiveoffinancialintegration.

JackandSuri

13

 Wealsocollectedindividualleveldemographicandotherdataonallhouseholdmembersinoursample. Table4 reports summary statistics by user status. The average age of users and nonusers were the same,althoughusersweremorelikelytobementhanwomen,andweremorelikelytobeliteratethan the population average. MPESA users have typically completed a higher level of education: for example,45percentofusershavecompletedsecondaryschool,and9percenthaveauniversitydegree, whilethecorrespondingfiguresfornonusersare37percentand4percent,respectively.

Table4:Individualcharacteristics
Count Demographics Age(years) Sex(sharemale) Sharewhocanread Sharewhocanwrite Educationalattainment(share) None Primary Secondary University Other

Non users 5,985 35.0 (15.1) 0.48 (0.50) 0.89 (0.32) 0.88 (0.33) 0.22 (0.41) 0.28 (0.45) 0.37 (0.48) 0.04 (0.20) 0.09 (0.28)

Users 1,091 36.6 (13.1) 0.57 (0.50) 0.96 (0.18) 0.96 (0.19) 0.09 (0.29) 0.19 (0.39) 0.45 (0.50) 0.09 (0.29) 0.18 (0.38)

Total 6,985 35.2 (14.8) 0.49 (0.50) 0.90 (0.30) 0.89 (0.31) 0.20 (0.40) 0.27 (0.44) 0.39 (0.49) 0.05 (0.22) 0.10 (0.30)

Remittances
The primary function of MPESA, at least as it was conceived, is to reduce the costs of making remittances from one individual to another, especially across large distances. We collected detailed dataonallkindsofremittances,bothmonetaryandinkind,andsentbyallmeans.Table5reportsthe sharesofhouseholdsinoursamplewhosentorreceivedremittances,byrural/urbanlocation,andby MPESAuse.

JackandSuri

14

Table5:Whomakesremittancesbothmoneyandgoods
Total Bygeographiclocation Rural Urban ByMPESAuse Nonuser User

Send 53% 38% 61% 38% 72% Receive 44% 42% 45% 28% 63%

On average, more households send remittances than receive, although many no doubt do both. The share receiving transfers is similar across rural and urban households (42% and 45%), but sending is performedbyalargershareofurbanhouseholdsthanrural.MPESAusersaremuchmorelikelytosend orreceiveremittancesthannonusers.Ourdata(notreportedabove)indicatethatofhouseholdswho makeorreceiveremittances,about38%arenetsenders,30%arenetrecipients,and32%areneither. IndicatorsoffrequencyandsizeofremittancessentandreceivedarereportedinTable6.Onaverage, householdssendandreceiveremittancesonceeverythreetofourmonths.Monthlyremittancessent aresmaller,amountingtoapproximately4.5%ofmonthlyexpenditure,whilethosereceivedareabout 5.6%.AboutathirdofremittancesaresentandreceivedviaMPESA,andtheytendtobesmallerthan theaverageremittance,amountingtoabout1.5%ofmonthlyhouseholdexpenditureonaverage. 25

Table6:Remittancessentandreceived
Numberpermonth Valuepermonth(%consumption) Valuepertransaction(KSh) Numberpermonth Valuepermonth(%consumption) Valuepertransaction(KSh) All Total 0.17 2.3% 3,112 0.13 2.8% 3,738 Users Total MPESA Other Sending(N=1,741) 0.33 0.17 0.17 4.6% 2.3% 2.3% 3,447 3,112 4,016 Receiving(N=1,327) 0.23 0.13 0.10 5.9% 2.8% 3.2% 5,854 3,738 10,291* NonUsers Total 0.27 4.4% 3,269 0.22 5.0% 3,685

*Receivedhastwolargevaluesofmore700,000KSh(aboutUSD1,000)forrepaymentsofdebts.

Table7reportsthedestinationandoriginofhouseholdremittances.Remittancesappeartogofrom
youngertooldergenerations,as47%ofthosesentaretoparents,while12%ofremittancesreceived arefromthem.MPESAuseiscorrelatedwithasmallerpercentageoftransferswithparents:nonusers
NotethatthesefiguresrefertotheaverageofMPESAremittances,nottheaverageofallremittancessentbyMPESA users.
25

JackandSuri

15

 send half of their remittances to their parents, and receive 16% from them, while users send 41% to their parents and receive 5% of remittance receipts from them. Users have correspondingly larger sharesoftheirremittanceportfolioslinkedtootherrelativesandfriends.

Table7:Destinationandoriginofremittances
Spouse Parent Child Otherrelative Friend Other Total

Destination Nonusers Users 9% 8% 50% 41% 10% 8% 16% 24% 6% 13% 8% 6% 100% 100%

Total 9% 47% 10% 19% 8% 7% 100%

Origin Nonusers Users 8% 18% 16% 5% 23% 15% 27% 32% 19% 24% 7% 6% 100% 100%

Total 12% 12% 20% 29% 21% 7% 100%

Saving
Becauseindividualsdonotneedtowithdraworsendbalancesimmediately,theyareabletoaccumulate savingsontheirMPESAaccountsovertime.ThusMPESAhasbecomeasavingsinstrument,aswellas ameanstosendmoney. 26 Table8reportssharesofallhouseholds,andbyuserstatus,usingvarious savingsinstruments.Manypeople77%ofhouseholdssavemoneyathomeunderthemattress, but slightly more nonusers do than users. Users are more likely to own stocks (possibly Safaricom shares, sold in an IPO shortly before our survey, though we donot have data to supportthis), and to haveabankaccount.ButfullythreequartersofhouseholdswithanMPESAuserreportusingittosave.

Table8:Savingsinstrumentsusedbyhouseholds
MPESA Bankaccount Mattress SACCO Merrygoround Householdmember Familymember Friend Advancepurchase Stocks Other

Nonusers 0.00 0.36 0.81 0.14 0.38 0.13 0.04 0.03 0.04 0.06 0.57

Users 0.75 0.72 0.72 0.24 0.41 0.16 0.05 0.04 0.04 0.19 0.68

Allhhlds 0.33 0.52 0.77 0.19 0.39 0.14 0.04 0.04 0.04 0.12 0.62

26

SometimesmoneyisstoredinanMPESAaccountsimplytosaveapersonfromcarryingtoomuchcash,especiallyfor exampleonlongandpotentiallydangerousbustrips.

JackandSuri

16

 MPESAusersvaluethesavingfunctionitprovides.Whenaskedtoranksavingsinstrumentstheyusein orderofimportance,21%sayMPESA isthemostimportant,and90%sayit isoneofthethree most important.AnMPESAaccountappearstoprovideasafermeansofsavingthanotherinstruments,and onethatiscomparabletoabankaccount.10.5%ofhouseholdsreportsavingsbeinglostorstolen.But only1.4%reportlosingsavingsheldinabankand1.6%saytheylostsavingsfromMPESA.

CustomerexperiencewithMPESA
The perceived safety of MPESA and its convenience are major reasons that early adopters of the technologychosetouseit.Table9reportshouseholdsprimaryreasonsforusingornotusingMPESA. Among users, 26% report that safety was their main motivation for adopting it, nearly twice as many (45%) say ease of operation was the main reason. About 12% say they use MPESA for emergencies. Fornonusersthereasonthatwasmentionedmostoftenastheprimarycauseofnonadoptionwaslack ofadequateaccesstothenetworkofagents.Intheyearsincethesurveywasfielded,thenumberof agentshasrisenfromabout2,500tomorethan12,000,sothisconstraintislesslikelytobindnow.

Table9:Reasonsforuse,andnonuse,ofMPESA
Reason Safety Ease Cost Noaccess Confidentiality Emergency NoReason Noneed Other

Nonusers 0.03 0.01 0.08 0.21 0.02 0.00 0.28 0.18 0.19

Users 0.26 0.45 0.07 0.00 0.02 0.12 0.08 0.00 0.01

Nonetheless, problems have been experienced by users, either due to their own errors or associated withthechallengesfacedbyagentsinmanagingtheircashandefloatbalances.Forexample4.3%of users report having sent money to the wrong person at least once, although twothirds of these retrievedthemoney,abouthalfofthemwithinaday.

Table10:DelaystowithdrawingmoneyfromMPESA
Reason DeletedSMS Agenthadnomoney Agentnotavailable Agentsystemdown Safaricomnetworkdown NoID Other

Shareofdelays 0.00 0.69 0.00 0.01 0.11 0.07 0.08

Delayuntilwithdrawalpossible Hourorless 0.19 Halfaday 0.29 Aday 0.35 Afewdays 0.13 Aweek 0.03 Severalmonths 0.01 Never 0.00 17

JackandSuri

 20%ofusersreportatleastoncenotbeingabletowithdrawmoneyfromanagentwhentheywanted. Table10reportsthatofthese,69%wereduetotheagenthavingnocash,and11%duetotheSafaricom networkbeingdown.Ontheotherhand,83%ofdelayedwithdrawalswereresolvedwithinaday. Weaskedusersabouttheirexperienceswiththeagentwhowasmostconvenientlylocatedtothem,as reportedinTable11.Fortheseagents,alowershareofrespondents,15%,reportednotbeingableto withdrawfunds.Just6%ofusersreporteddelaysinbeingabletodepositfundsinMPESA,associated with inventory management issues faced by agents. Unlike a bank branch, an MPESA agents cannot simplytakecashandrecordthedepositinanaccount, 27 butmustexchangecashforefloat.

Table11:Reportedexperienceswithagents
Fractionunabletowithdrawmoneyfromagent Fractionunabletodepositmoneywithagent FractionaskedbyagenttoshowID Fractionwhotrustagent

Fraction 0.15 0.06 0.76 0.65

Overall however, customers appear to value MPESA services highly, especially when compared with othermoneytransferservices.WhenaskedtocompareMPESAwithothersuchservicesintermsofa number of attributes, the responses were overwhelmingly positive, with large majorities responding thatitwasfaster(98%),easiertouse(99%),moreconvenient(96%),safer(98%),andcheaper(96%). Similarly, when asked how happy they were with the service, measured on a scale from 1 (extremely unhappy) to 10 (extremely happy), more than half reported a rank of 10, and nearly 90% reported valuesof8orabove(seeColumnI,Table12).AskedwhatimpacttheywouldexperienceifMPESAwas tobeshutdown,thelargemajorityreportedthatitwouldbelargeandnegative(ColumnII,Table12).

Table12:MeasuresofsatisfactionwithMPESA
I.HappinesswithMPESA Extremelyunhappy1 0.006 2 0.003 3 0.009 4 0.001 5 0.005 6 0.022 7 0.069 8 0.123 9 0.229 Extremelyhappy10 0.534
AlthoughtheagentisrequiredbySafaricomtorecordtransactionsinalogbook,thisisnotsufficientasitdoesnot leadtoachangeinthecustomerselectronicbalance.
27

II.ImpactofclosingdownofMPESA Largenegative 0.84 Smallnegative 0.12 None 0.02 Smallpositive 0.02

JackandSuri

18

V.Conclusions
As the developed world begins to rebuild the recently collapsed global financial system, the financial architectureinpartsofthedevelopingworldisbeingrapidlytransformed.Asthecostsofmobilephone technology have fallen, and as the technology has been adapted to support financial services, mobile banking innovations have begun to spread across and within poor countries. The low cost, and the widespreadunmetdemandforfinancialservices,ascapturedbylowratesofbankaccess,meansthat mobilebankinghasthepotentialtoreachremotecornersofthesocioeconomic,aswellasgeographic, spectrum. That potential appears to be being realized in Kenya, through MPESA, a mobile banking system operated by Safaricom. We estimate that MPESA has reached nearly 40 percent of the adult population after a little more than 2 years of operation. Part of this success is due to a rapidly expandingnetworkofMPESAagents,whonownumberover12,000. MPESA is an innovation that clearly dominates its moneytransfer predecessors on virtually all dimensions. Users say it is faster, cheaper, more reliable, and safer, and a very large majority report thattheywouldsuffersignificantnegativeconsequencesifitweretobeshutdown. TheseexpressedpreferencessuggestthatMPESAisvaluedmorebyindividualsthanitcosts.Onthe otherhand,theprecisesourceofthesebenefitsi.e.,thespecificeconomicimpactsofMPESAare not easy to calculate. We have identified a number of potential economic effects of MPESA at the householdlevelforexamplefromimpactsonsavingandinvestment,toriskspreadingandinsurance. Atthemacroeconomiclevel,therecouldbeimportantimpactsonthemoneysupplyandinflation,with implicationsfortheextentofCentralBankregulationandtheconductofmonetarypolicy.Wehopeto exploretheseissuesempiricallyinfuturework.

References
Chattopadhyay, Raghabendra and Esther Duflo (2004): Women as policy makers: evidence from a randomizedpolicyexperimentinIndia,Econometrica,72(5),140943. Ivatury,GautamandMarkPickens(2006):MobilePhoneBankingandLowIncomeCustomers:Evidence fromSouthAfrica,ConsultativeGrouptoAssistthePoor,WashingtonDC. Lucas, Robert (1988): On the mechanics of economic development, Journal of Monetary Economics, 22,342. Mas,Ignacio(2009):Theeconomicsofbranchlessbanking,Innovations,4(2),5776,MITPress. MasandMorawczynski(2009):DesigningMobileMoneyServices:LessonsfromMPESA,Innovations, 4(2),7792,MITPress. Mas, Ignacio and Sarah Rotman (2008): Going Cashless at the Point of Sale: Hits and Misses in DevelopedCountries,ConsultativeGrouptoAssistthePoor,FocusNoteNo.51,WashingtonDC. JackandSuri 19

 Mas,IgnacioandKabirKumar(2008):BankingonMobiles:Why,How,forWhom?,ConsultativeGroup toAssistthePoor,FocusNoteNo.48,WashingtonDC. Romer,Paul(1986):IncreasingReturnsandLongRunGrowth,JournalofPoliticalEconomy,October, 94:5,100237. Solow,Robert(1956):"AContributiontotheTheoryofEconomicGrowth,"QuarterlyJournalof Economics.

JackandSuri

20