59435 December 2010 | Edion No. 3 Kenya at the Tipping Point? with a special focus on the ICT Revoluon and Mobile Money December 2010 | Edion No. 3 Kenya at the Tipping Point? with a special focus on the ICT Revoluon and Mobile Money Poverty Reducon and Economic Management Unit Africa Region TABLE OF CONTENTS FOREWORD i ACKNOWLEDGEMENT ii ABBREVIATIONS AND ACRONYMS iii EXECUTIVE SUMMARY iv THE STATE OF KENYA'S ECONOMY 1 1. The 2010 Performance 2 1.1 Macroeconomic Management 4 1.2 Fiscal Performance 5 1.3 Balance of Payments 7 1.4 Financial Sector Developments 7 1.5 Snapshot on EAC Integraon 8 2. The Outlook for 2011 and 2012 9 2.1 Growth Expectaons 9 2.2 Key Risks to the 2011-12 Forecast 11 THE ICT REVOLUTION AND MOBILE MONEY 12 3. Kenya's ICT Revoluon 13 3.1 The Transformaon of the Last Decade 13 3.2 Explaining Kenya's ICT Revoluon 14 3.3 Will Kenya be the Home of Africa's "Silicon Valley"? 16 4. Mobile Money 17 4.1 What is Mobile Money? 17 4.2 Exponenal Growth 17 4.3 Users 19 4.4 Uses 21 4.5 Benefits 23 4.6 The Future of Mobile Money 25 BIBLIOGRAPHY 25 LIST OF FIGURES Figure 1: A strong recovery in 2010, growth at almost 5% iv Figure 2: Kenya's ICT revoluon v Figure 3: Deepening financial access in Kenya v Figure 4: In 2010, there has been broad-based recovery 2 Figure 5: Stable growth in 2010 ...Kenya is catching up with its neighbors 3 Figure 6: Declining inflaon as Nairobi Stock Exchange rebounds strongly 4 Figure 7: Credit to public sector has been increasing 4 Figure 8: Exchange rate fluctuaons are mainly reflecng shis among major currencies 5 and not of the Kenyan shilling Figure 9: The fiscal smulus was countercyclical and not inflaonary 6 Figure 10: Posive balance of payment with strong capital and financial flows 6 Figure 11: Interest rates movement 8 Figure 12: Banks have kept the interest rates spread at 10% due to structural rigidies 8 Figure 13: Kenya currently contributes 40% of EAC GDP and has the most diversified export sector 9 Figure 14: Kenya's recovery connues 10 Figure 15: Kenya could be entering another decade of high growth, but it will remain 10 below its neighbors Figure 16: Since 2000, the telecom sector outperformed all sectors in the economy 13 Figure 17: Africa's telecom revoluon 14 Figure 18: Subscribers increase exponenally while prices plummet 14 Figure 19: Private sector investment in mobile telephones peaked in 2006 while 15 employement connued to rise Figure 20: Mobile money: expected to exceed 20% of GDP at the end of 2010 17 Figure 21: End 2010, Kenya will have 15 million mobile money customers 19 Figure 22: M-Pesa stores outnumber bank branches by a factor of 20 19 Figure 23: By August 2010, M-Pesa had enlisted 12.6 million customers and nearly 20,000 agents countrywide 19 Figure 24: Mobile moneey usage is highest among the younger, but 45% of those above 50 years 20 also use it Figure 25: The rich use mobile money dispropoonately, but poor Kenyans are catching up 20 Figure 26: Usage rates are higher in Central and West of Kenya, and men are slightly more 20 likely to use mobile money Figure 27: Mobile money usage trails mobile phone ownership in Coast and Eastern provices 21 Figure 28: Sending and receiving transfers are the dominant uses of mobile money 21 Figure 29: Within 3 years, mobile money has replaced tradional means of reming funds 21 Figure 30: Nairobi accounts for the largest share of mobile money flows, but the fastest 22 growth is in remote regions - Figure 31: Richer quinles deposit substanally more than they withdraw 22 BOXES Box 1: What is a Tipping Point? iv Box 2: Enabling Reforms and Regulaon Permied Growth 16 Box 3: How does Mobile Money Work? 18 TABLE Table 1: Key Macroeconomic Indicators 11 - ANNEXES Annex 1: An Update on the KEU Edion1 (December 2009) 27 Annex 2: An Update on the KEU Edion 2 (June 2010) 28 Annex 3: Global Commodies Prices Trends 29 Annex 4: Key Assumpons to the Growth Forecast 30 Annex 5: Key Factors determining the Compeveness in the ICT Sector 31 ­ Comparison of Kenya and India Annex 6: Other Sector Innovaons on the Mobile Plaorm 32 Annex 7: Data Sources and Projecons for Mobile Phone and Mobile Money Stascs 33 Annex 8: The Genesis of Mobile Money in Kenya 34 Annex 9: Background Stascs 35 FOREWORD T his is the third edion of the Kenya Economic Update. This report, like its predecessors, provides an update of recent economic trends as well as a special focus on a selected topical issue. The special topic of this report features Kenya's ICT revoluon and the implicaons that mobile money has for the Kenyan economy. The report aims to support all those who want to improve the economic management of Kenya. It is intended to help inform and smulate debate on topical policy issues, and so to make a contribuon to unleashing Kenya's growth potenal. This edion of the Economic Update has three main messages. First, Kenya's economy is growing faster than we predicted in the last update. It could also be at a pping point, in the sense that it may have reached a crical threshold beyond which it may enter a period of sustained growth. We have increased our 2010 growth forecast by almost a full percentage point, to 4.9 percent. Second, economic growth in 2011-12 could range between 5.3 and 6.0 percent, approaching levels of growth last seen in 2004-07. The report analyzes the factors that will need to remain posive for this to happen as well as the risks that could threaten the return to high growth. Third, the report's special topic analyzes the implicaons of Kenya's ICT revoluon and the introducon of mobile money for future economic growth. The report notes that mobile money is transforming the financial sector, creang op- portunies for more effecve transfers to the poor, and influencing a range of producve acvies. More importantly, we observe that the implementaon of essenal reforms and uptake of ICT throughout the economy could provide the impetus required for high and sustained growth. Over the last decade, Kenya's ICT sector has grown phenomenally, aracng global aenon, especially aer the introducon of mobile money. Today, Kenya has the largest mobile money plaorm in the world. An esmated 15 million mobile phone users are expected to be using mobile money by end 2010, the equivalent of three out of every four adult Kenyans. Kenya has posioned itself to become a global ICT hub, aracng investors who want to extend the ICT revoluon domescally as well as look for applica- ons in other developing countries. Johannes Zu World Bank Country Director for Kenya December 2010 | Edion No. 3 i ACKNOWLEDGEMENT T his edion was prepared by a team led by Jane Kiringai and Wolfgang Fengler, together with Gabriel Demombynes, Millicent Gitau, Fred Owegi and Roger Sullivan. The core team also included Ed Al-Hus- sainy, Allen Dennis, Peter Warutere, Sachin Gathani, Yira Mascaró, Rosemary Oeno, Aly Khan Satchu, Dimitri Stoelinga and Catherine Gachukia. The team also appreciates the inputs from Andrew Roberts, Andrew Karanja, and Kaoru Kimura. The team is very grateful for the connued close collaboraon of the Economic Roundtable members, in- cluding the Ministry of Finance, Central Bank of Kenya, the Kenya Naonal Bureau of Stascs, the Kenya Instute of Public Policy and Research Analysis, and the Internaonal Monetary Fund. This report reflects the inputs of the Roundtable meeng held on November 5, 2010, which also included the parcipaon of the Office of the Prime Minister, the Communicaons Commission of Kenya, and the Ministry of Informa- on and Communicaons. The Economic Roundtable is chaired by the Governor of the Central Bank, Prof. Njuguna Ndung'u. The team wishes to acknowledge the important contribuons made by the Ministry of Informaon and Communicaons, the Communicaons Commission of Kenya, the ICT Board, Safaricom, and Zain-Kenya, parcularly in helping us understand ICT's dynamic story in Kenya. Special thanks go to Michael Joseph (outgoing CEO of Safaricom) and his team for their invaluable assistance in providing background informa- on for the report's mobile money secon. In addion, the report benefited greatly from the peer reviews of Dr. Mbui Wagacha (Independent Con- sultant) and Isabel Neto (Senior ICT Policy Specialist, World Bank), as well as feedback from Ragnar Gud- mundsson (Resident Representave, IMF). The team remains thankful for the support and advice received from the World Bank managers: Johannes Zu (Country Director) and Kathie Krumm (Sector Manager). December 2010 | Edion No. 3 ii ABBREVIATIONS AND ACRONYMS ATM Automated Teller Machines BPO Business Process Outsourcing CBK Central Bank of Kenya CCK Communicaon Commission of Kenya DRC Democrac Republic of Congo EAC East African Community FDCF Financial Deepening Challenge Fund FDI Foreign Direct Investment GDP Gross Domesc Product HA Hectare ICT Informaon and Communicaon Technology ISP Internet Soluon Providers IT Informaon Technology KEU Kenya Economic Update KIHBS Kenya Integrated Household Budget Survey KNBS Kenya Naonal Bureau of Stascs KPA Kenya Ports Authority KPTC Kenya Posts and Telecommunicaons Corporaon KRA Kenya Revenue Authority KSHS Kenya Shillings KYC Know-Your-Customer MT Metric Ton MOT Ministry of Transport NCPB Naonal Cereals & Produce Board NBFI Non Bank Financial Instuons NSE Nairobi Stock Exchange OECD Organizaon for Economic Co-operaon and Development PBCS Port-Based Community System PPP Public-Private Partnerships SME Small & Medium Enterprises SSA Sub-Saharan Africa US United States WRS Warehouse Receipng System WB World Bank December 2010 | Edion No. 3 iii EXECUTIVE SUMMARY K enya may be at a "pping point", the theme of the third Kenya Economic Update which has a spe- cial focus on the transformave impact of Informaon and Communicaons Technology (ICT) and mobile money. Over the last decade, ICT has outperformed all others sectors growing at an average of 20 percent per year. The benefits of ICT are starng to be felt in other sectors, and have contributed to the condions for Kenya to reach this pping point. Kenya has entered the new decade with renewed and stronger than expected growth. The passing of the new constuon, connued strong macroeco- nomic policies, and a favorable regional environment have created a new posive economic momen- tum. Kenya may again be posioned to experience high growth. Over the last three decades Kenya has experienced only two short episodes when economic growth exceeded five percent and was sustained for at least three consecuve years: 1986-88 and 2004-2007. Is Kenya again at the verge of experienc- ing another growth spurt? Will it last longer and go deeper than the previous two episodes? The State of Kenya's Economy (SSA) region as a whole. Per capita growth will also be closer to the per capita growth for SSA, though I n 2010, Kenya has seen the return to higher sll slightly lower. By 2011, African growth rates growth. Services, the driver of previous years' are expected to again exceed 5 percent. There are growth, have moderated while agriculture and in- also indicaons that closer East African integraon dustry are rebounding aer two weak years (see and aempts to improve Kenya's infrastructure are Annex 1 for a summary of the first Kenya Economic contribung to Kenya's growth momentum, though Update, December 2009, which discussed Kenya's accelerated growth and a take-off of in exports will economic performance in 2008-09). Growth in the require reliable energy supplies and improved trans- Kenya economy has accelerated since the first quar- port infrastructure. The port of Mombasa remains ter of 2010 leading us to project a rate of growth for a potenal boleneck and constrains business op- 2010 of 4.9 percent, almost a full percentage point eraons in the larger East Africa Community (EAC) higher than the 4.0 percent predicon in the second region (see Annex 2 for progress on recommended Kenya Economic Update (June 2010) and the same reforms at the port of Mombasa as discussed in the growth rate is predicted for the Sub-Saharan Africa second Kenya Economic Update). Box 1: What is a Tipping Point? Kenya is also expected to sustain balanced growth through all the quarters in the year 2010 (see fig- The concept of pping point was popularized by the writer Malcolm Gladwell in his book, The Tipping Point: ure 1). For the first me since 2007, Kenya is ex- How Lile Things Can Make a Big Difference, published in Figure 1: A strong recovery in 2010, growth at almost 5% 2000. As Gladwell notes, the concept of pping Point comes from the world of epidemiology. It's the name given to that moment in an epidemic when a virus reaches a crical mass, when it starts to spread exponen- ally aer a slow build up. Gladwell applied the concept to different situaons to show how certain factors can come together with dramac outcomes. We are using the concept in the case of the Kenya economy to analyze whether condions may be ripe for a pping point in terms of economic growth. We are examining, in parcu- lar, the rapidly expanding ICT sector in Kenya and how its applicaon in other sectors might be one of the key factors contribung to high levels of growth in the Kenyan economy. Certainly ICT related innovaons in the finan- cial sector have seen a revoluon with remarkable outcomes in access to financial services even for the poor. Source: World Bank Global Economic Prospectus December 2010 | Edion No. 3 iv pected to grow above 4 percent in each quarter of 72.0 billion in 2009. Kenya will connue to benefit 2010, peaking in the third quarter at an esmated from EAC integraon, which already accounts for a 5.5 percent. Economic growth could exceed 5 per- quarter of its export trade. cent in 2010, if the fourth quarter performs beer than expected and achieves a growth rate of 4.5 The government's monetary policy and the fiscal percent or higher. This stability in economic growthsmulus, aer some implementaon delays, are throughout the year is unlike previous years when now contribung to Kenya's strong growth per- growth was highly volale and negavely impacted formance in 2010. Public investment has increased by a number of shocks. to its highest levels in a decade. At the same me, Kenya's government borrowing has increased Five structural factors give hope for a more sus- sharply and debt levels have increased over the last tained path of growth in the next decade. First, two years from 40 percent of GDP to an esmated Kenya is home to a growing market of 40 million 47 percent by end 2010. This is a much smaller in- people and is becoming more closely integrated crease compared to most countries in the world but with the EAC, which offers a market of more than reason enough to revert back to ghter fiscal poli- 130 million, many of which share a common herit- cies in 2011. age and language. Furthermore, Kenya's populaon is increasingly urbanized (39 percent) and educated For 2011, the World Bank forecasts growth at 5.3 which is associated with posive economic develop- percent. Driven by strong public investments and ment. Second, Kenya has a new constuon which improved business confidence the economic indi- will address the governance concerns and improve cators point to a full recovery and possibly takeoff Kenya's business environment. Third, Kenya's pri- in the medium term. Under the high case scenario vate sector remains among the most dynamic in the economy could achieve a growth of 6.0 percent Africa, demonstrang resilience during crisis and in 2011 and maintain that level in 2012. However, producing global innovaons, especially in ICT, the there remain a number of important risks. Weath- special topic of this report. Fourth, Kenya's eco- er-related shocks, such as moderate drought condi- nomic policies have been improving and can now ons in early 2011--which some have forecasted-- build on a strong track record in managing external could impede growth. On the polical front, the shocks. Inflaon and interest rates have declined, run-up to the 2012 naonal elecons, if not man- debt remains at manageable levels, and the bal- aged properly, could affect investor confidence and ance of payments returned to surplus contribung dampen the growth outlook. to higher reserves. Fih, investments in infrastruc- ture, parcularly roads and energy, are begining to Kenya's ICT Revoluon and pay off and will reduce the cost of doing business. Mobile Money Kenya is also benefing from the posive growth momentum in Africa, especially East Africa. The connent has recovered from the 2008-09 global I n the last decade, Kenya has undergone a trans- formaon in ICT which has also had an impor- tant impact on Kenya's social and economic struc- crises beer than most parts of the world, and more tures. In 1999 less than 1 in 1000 Kenyan adults had rapidly than it has from past global crises. With an mobile phone service. By mid-2010, there were 21 esmated growth of almost 5 percent in 2010, SSA million acve mobile phone numbers, equivalent to will experience the second strongest recovery in the one per adult. Not all of the poor have access yet world ­ aer the surging economies in Asia. East Af- to a mobile number, as there are mulple subscrip- rica is expected to grow at 6 percent and, aer two ons by richer individuals and many teenagers also years of low growth, Kenya is now starng to close have phones, but the "access gap" is closing rapidly. the gap with its neighbors which connue to grow While subscripons increased exponenally, calling strongly, albeit from a lower level of income. East rates and the cost of devices have dropped sharply, African integraon has created a single market with making Kenya's communicaon costs among the an esmated Gross Domesc Product (GDP) of US$ lowest worldwide. December 2010 | Edion No. 3 v ICT has been the main driver of Kenya's economic have dramacally opened up financial access for growth over the last decade. Since 2000, the sector the poorer segments of the populaon. About five has outperformed all other segments of the econo- years ago, only a small percent of the adult popu- my, growing on average by 20 percent annually and laon had tradional bank accounts. Equity Bank propelling the combined transport and communi- pioneered a new form of banking, making accounts caons sector into the economy's second largest. accessible and affordable to the middle class, in- Since 2000, Kenya's economy grew at an average of creasing the access many Kenyans had to formal 3.7 percent. Without ICT, growth would have been banking services. a lackluster 2.8 percent--similar to the populaon growth rate--and income per capita would have Mobile money has included the "middle of the stagnated. ICT has had a transformave impact on pyramid" into the financial system (see figure 3). the financial sector and has contributed to impor- Mobile money is sll used much more frequently by tant indirect economic effects in other sectors, such richer Kenyans. However, since 2009 it has rapidly as health care and public informaon (see Annex penetrated rural areas as well as middle and lower 6). income groups, closing the access gap between in- come strata and urban/rural residents. Among the many ways cell phones have been used in Kenya, the most innovave is mobile money. Mo- Figure 3: Deepening financial access in Kenya bile money is a global innovaon with the potenal to become an addional engine of Kenya's growth and an important tool for poverty reducon. Start- Banks ing with the M-PESA system launched by Safaricom Adults access to Banks in 2007and later joined by Zain's Zap and Yu's Yu- finance Mobile Cash in 2009, and Orange Money in 2010mobile phones only money has rapidly become a fixture in the lives of Mobile money only Kenyans, extending a sophiscated form of financial Adults access to finance access to a wide populaon. The World Bank es- No access to Mobile phones only finance or phones mates that by end 2010, 15 million Kenyans (3/4 of No access to the adult populaon) will use mobile money (see finance or phones figure 2), transferring an esmated US$ 7 billion an- 36.6 million 20.9 million 39.4 million 23.1 million nually (20 percent of GDP) by phone. Figure 2: Kenya's ICT revoluon Source: World Bank staff esmates Note: "Banks" also include other financial instuons 25 The economic benefits of mobile money are sub- 20 stanal. This new mode of money transfer has con- siderably reduced transacon costs in undertaking 15 Populaon age 15+ social and economic acvies. Just-in-me pay- Millions Mobile phone subscripons Mobile money customers ments in areas of tradional agriculture or in more 10 Internet subscribers remote parts of the country are expected to lead to 5 greater producvity in those areas. The inclusion of mobile money flows into a semi-formal financial 0 system also helps with macroeconomic policy mak- 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 ing because it increases informaon about liquidity in the financial system. In addion, there have also Source: Internaonal Telecommunicaon Union, KNBS, World Bank projecons been a number of important social benefits, includ- ing higher personal security as it reduced the need Mobile money and innovaons in retail banking, to carry large amounts of cash and improved finan- as well as a new business philosophy to also tar- cial security for women giving them an independ- get the lower segments of the wealth pyramid, ent place to store and manage funds. December 2010 | Edion No. 3 vi Investments in ICT have been an engine of eco- · Private entrepreneurship thrives with good reg- nomic growth up to the present, but alone will not ulaon. The explosive growth in ICT would not sustain future development unless they enable have occurred if the government had retained its other sectors of the economy to grow. In 2010, monopoly and shut-out the private sector. Priva- there are a number of indicaons that ICT is start- zaon and good regulaon has proved far-sight- ed and beneficial to the economy. ing to reshape the "old economy", especially in the financial sector. This sector has been among the · Private ownership increases jobs in growth indus- strongest performers in 2010 and benefied from a tries, not reduces them. By privazing Telecoms number of innovaons. This includes Equity Bank's and providing a good regulatory environment the government opened the window for thriving and targeng of the "middle of the pyramid" of income compeve private sector operators, leading to earners to open savings accounts and M-KESHO, a new employment opportunies. joint venture allowing mobile phone users to earn interest on their mobile phone-based savings ac- · Mobile money has the potenal to become a game changer for the economy. There is evi- counts. dence that mobile money transfers to rural areas have a producve impact by enabling just-in-me Several lessons emerge from the sustained growth payments for farm labor or to procure agricultural in ICT over the last decade and its emergence as a inputs. The poor also benefit from mobile money major sector of the Kenyan economy: as it permits transfer payments in a low-cost envi- · The potenal of ICT to reshape social and eco- ronment and in a way that safeguards their cash nomic development cannot be underesmated. and reduces insecurity. Innovave uses of the ICT infrastructure, such as the introducon of mobile money, demonstrate ways ICT can create opportunies for growth and social inclusion. December 2010 | Edion No. 3 vii The State of Kenya's Economy T he first part of the Kenya Economic Update examines Kenya's recent economic performance. Sec- on one notes that 2010 will be a good year for Kenya's economy, East Africa's largest economy, with an expected growth rate of 4.9 percent. The report states that the improvement in Kenya's growth can be aributed to three main factors. First, agriculture and manufacturing have rebounded due to good rains and high prices for Kenya's export crops. Second, investments have increased, especially through the implementaon of the fiscal smulus program. Third, business senment has improved, parcularly since the adopon of the new constuon, beer regulaon and the first signs of improve- ments in infrastructure. Secon two provides growth forecasts for 2011-2012. The report esmates that economic growth in 2011 could be 5.3 percent and might even reach 6 percent, approaching lev- els of growth last seen in 2004-07. For 2102, the report esmates that growth could connue in the 6 percent range, as long as weather condions remain favorable and the polical situaon connues to be stable leading up to the elecons at the end of 2012. Figure 4: In 2010, there has been broad-based recovery 1. The 2010 Performance K 10.0 enya is experiencing a strong recovery in 2010. The World Bank is upgrading its growth forecast 8.0 the second me and projects growth at 4.9 percent 6.0 in 2010. The recovery in 2010 has been robust and 4.0 broad based (see figure 4); agriculture (+5 percent) and industry (+7.6 percent) rebounded strongly af- 2.0 ter two weak years. Services will grow at a moder- 0.0 ate 4.0 percent. -2.0 -4.0 Kenya has overcome the quadruple shock of 2008 and 2009 (post-elecon violence, drought, and the -6.0 global food and financial crises) and achieved bal- 2007 2008 2009 2010 Agriculture(25.5) Industry(18.8) Service(55.7) ance growth in all sectors (see figure 4). Favorable weather condions have led to the recovery of agri- Source: Kenya Naonal Bureau Stascs and World Bank esmates (sector share in GDP in parenthesis) culture and also contributed to more reliable ener- gy which has an immediate posive impact on the ICT sector has been driving growth over the last manufacturing sector. In addion, the economic decade. Recently the financial services sector has smulus program, which only came into full effect adopted the innovaons introduced by ICT, result- in 2010, is now also contribung to the economic ing in increased compeon and efficiency gains rebound. Kenya's growth in 2010 will equal the SSA among the leading firms in the sector. We project average and be similar to Ghana's, while exceeding the financial sector to be among the fastest grow- that of South Africa (see figure 5). However, Kenya's ing, expanding by about 10 percent in 2010. growth will be lower than that of several neighbor- ing countries. The agriculture sector is rebounding and is expect- ed to grow at 5 percent. This is an important de- The "new economy" is also starng to spill-over velopment aer two consecuve years of decline, into the "old economy" as witnessed by the per- when the sector contracted by a combined 6.7 per- formance of financial services, Kenya's best per- cent. Favorable weather condions and specific pol- forming sector in 2010. As we reported in the June icy intervenons under the economic smulus pro- 2010 edion of the Kenya Economic Update, the gram helped to turn around the sector. The most December 2010 | Edion No. 3 2 The State of Kenya's Economy Figure 5: Stable growth in 2010 ...Kenya is catching up with its neighbours 8 8 Economic Growth 2010 7 7 6 6 5.3 4.9 G DP grow th % 5 4 4 Percent 2.6 3 2 1.6 2 1 0 0 G hana Ethopia S outh Africa R w anda Kenya S S A average Tanzania Uganda 2007 2008 2009 2010 2011 -2 GDP Kenya GDP SSA Percapita Kenya Percapita SSA Source: World Bank staff esmates successful intervenon has been the rehabilitaon structure and regional economic integraon. With of old irrigaon schemes, which increased the area a populaon of more than 130 million people and a under irrigaon from 120,000 ha to 400,000 ha. For combined GDP of US$ 72 billion, the EAC common instance, it is esmated that maize output increased market is also expected to provide a new impetus to 2.7 metric tons (mt) by August 2010 compared to for growth for the sector. Recent trends show that the previous average of 2.5 mt. the sector likely to benefit most from the larger market include light manufactures especially food The performance of Kenya's main agriculture ex- processing. ports was strongest for tea which recovered rap- idly from 2009 weather condions. Coffee exports Growth in the services were mixed and horculture contracted (see Annex sector will moderate 3 for global commodity prices). In parcular: to about 4 percent in · Tea is experiencing a very strong year. A combi- Investments in ICT 2010. The service sector naon of volume and price increases will see the have paid off, and has been driving growth sector perform even beer than in 2008, which the remarkable over the last decade, had previously been the best year for the sector. specifically the explo- innovaons in the sive growth in ICT, which Global tea prices increased by another 1.5 per- sector have en- cent in 2010 topping the record prices of 2009. along with wholesale sured its connued and retail trade contrib- · Coffee is sll recovering. Although coffee is ben- growth. uted significantly to ex- efing from an increase in global prices, output pansion of the sector. contracted as coffee producon was slow to re- cover from the prolonged drought in early 2009. Domesc investments will drive growth in 2010. · Horculture exports contracted for the third Consumpon was the key driver of growth between consecuve year. The sector connues to be af- 2003 and 2007 before the mulple shocks in 2008. fected by a muted recovery in Europe, especially However, the expansionary fiscal policy, with heavy the fruits and vegetables. In addion, the volcan- investments in infrastructure, will provide the in- ic ash crisis in April 2010 disrupted access to the tended smulus and investment will drive growth in key source markets in Europe. 2010 expanding in excess of 10 percent. Consump- on too will rebound at 4 percent growth, but stay Industry is projected to grow at 7.6 percent in below the pre-crisis average of 4.8 percent. Exports 2010, suggesng that the resumpon to near nor- and imports will grow at a similar pace (about 8 mal levels of ulity supplies has reduced the cost percent) to maintain the structural current account of doing business. In addion, the industrial sec- deficit in the range of 4-5 percent of GDP, with a tor is starng to benefit from investments in infra- negave contribuon to growth. December 2010 | Edion No. 3 3 The State of Kenya's Economy 1.1 Macroeconomic Management The government maintains strong credenals in World Bank esmates show that starng in 2010, macroeconomic management, and the impact of economic growth will reach potenal output sug- the ambious smulus program has contributed gesng that it's me to exit the fiscal smulus in to the strong recovery in 2010. Macroeconomic 2011. Aer the quadruple shocks, economic growth fundamentals remain broadly stable. In 2010, Ken- was below potenal and Kenya met all the condi- ya's average inflaon rate declined to below 4 per- ons for a countercyclical smulus. The govern- cent (see figure 6). This is below the Central Bank ment had reduced debt to sustainable levels creat- of Kenya's (CBK) policy target of 5 percent, and the ing space for a fiscal expansion. Interest rates were lowest average rate since 2002. In addion, the Nai- stable, the financial sector was on sound foong, robi Stock Exchange connued the rebound begun and the condions were also right for a monetary in 2009, outperforming the Dow Jones in 2010. smulus. Consequently, the government has been able to finance a large budget deficit from the do- Monetary policy remained broadly neutral and mesc market. Credit to government grew by 50 high liquidity in the market dampened the up- percent, with public sector borrowing an equivalent ward pressure on interest rates which have started of 4.1 percent of GDP in the first half of 2010, with declining. CBK reduced the Central Bank Rate by a commensurate increase in the stock of domesc 100 basis points during the year, from 7.0 percent in January to 6.0 percent in July 2010. The 91-day debt. Treasury bill rate also boomed out in July. Figure 6: Declining inflaon as Nairobi Stock Exchange rebounds strongly 30.0 25.0 16500 6000 Dow Jones overall inf 15500 5500 20.0 14500 Food inf. 5000 13500 15.0 12500 4500 NSE Dow Jone s NSE 10.0 11500 4000 10500 5.0 3500 9500 3000 0.0 8500 7500 2500 Jan Jan A p r il A p r il A p r il Oct Oct Oct July July July Jan A p r il A p r il A p r il A p r il Oc t. Oc t. Oc t. Oc t. July July July July Jan Jan Jan Jan 2008 2009 2010 Source: KNBS, NSE, Dow Jones and World Bank staff esmates Figure 7: Credit to public sector has been increasing 10.0 70 Growth New credit % GDP 60 Public 8.0 Private 50 Total 6.0 40 4.0 30 pe r c e nt 2.0 20 10 0.0 Household 0 -2.0 Private Mar c h Mar c h Mar c h Jan Jan Jan May May May July July July Nov Nov Se pt Se pt Se p t Public -10 Total new credit -4.0 -20 2008 2009 2010 2007 2008 2009 2010 June Source: Central Bank of Kenya and World Bank staff esmates December 2010 | Edion No. 3 4 The State of Kenya's Economy Government bonds issued in the local currency ciaon of the dollar in the global market. The US market and earmarked for infrastructure financing dollar appreciated by about 20 percent against the have aracted funding from investors, and could euro between November 2009 and June 2010. The lead to crowding out credit to the private sector. shilling exchange rate mimicked these movements However, excess liquidity in the market suggests and depreciated by 14 percent against the dollar, that this is not yet hap- but appreciated against the euro. However, the real pening. Credit to the exchange rate, which is a good indicator of Kenya's private sector grew by compeveness, remained broadly stable (see fig- 17 percent in the first The real exchange ure 8). half of the year (equiva- lent to 1.5 percent of rate, which is a 1.2 Fiscal Performance GDP). Credit to house- good indicator of The government's fiscal deficit reached 7 percent holds took the highest Kenya's compe- in fiscal year 2009/2010, higher than projected in share of credit to private veness, remained the last Kenya Economic Update and explained by sector signaling a re- broadly stable. an acceleraon of the implementaon of the fiscal covery in consumpon smulus. The smulus, which has been extended growth which has been into 2011, will increase government spending as the key driver of growth in Kenya. Aer contracng a share of GDP to 33.1 percent and generate rev- in 2009, credit to households has recovered and ex- enues equivalent to 24.9 percent of GDP in fiscal panded by 30 percent in the first half of 2010, be- year 2010/11. The deficit inclusive of grants, at 6.8 coming again one of the main sub-sectors geng percent of GDP, will be financed through domesc loans (see figure 7). and external borrowing which will increase the to- tal debt stock to 47 percent of GDP by the end of Kenya has been affected by the fluctuaons in glo- 2010. bal currencies, but overall the Kenya Shilling has been broadly stable if exchange rates are com- Revenue is targeted to record a growth of 20 per- pared to a basket of the major internaonal cur- cent in 2010/2011 fiscal year. In the first quarter rencies. In first half of 2010, the debt crisis in the (July-September 2010) revenue collected amount- euro area was transmied to the Kenyan economy ed to Kshs 140.4 billion, represenng a growth of through a weakening of the shilling against the dol- 13.2 percent compared to the same period in the lar. This crisis in the euro area, which started in previous year. However, Value Added Tax revenue Greece, saw a weakening of the euro and an appre- registered only a 2.5 percent growth indicang an Figure 8: Exchange rate fluctuaons are mainly reflecng shis among major currencies and not of the Kenyan Shilling 120 115 Effecve Exchange rate (Jan 2003=100) Exchange rate 110 110 105 100 100 95 90 Kshs/ 90 Euro 85 Nominal 80 Real 80 75 70 70 65 US$ July Jan Jan Jan Mar Mar Mar Mar Se p t Se p t Se p t Se p t May May Jan May May July July July Nov Nov Nov 60 Jan Jan Jan Jan Oc t Oct Oct Oc t A pr A pr A pr A pr July July July July 2007 2008 2009 2010 Source: Central Bank of Kenya December 2010 | Edion No. 3 5 The State of Kenya's Economy Figure 9: The fiscal smulus was countercyclical and not inflaonary 8 Deficit financing Actual versus potental GDP growth 8.0 6 7.0 F inancing 4 6.0 2 % of GDP 5.0 0 4.0 -2 Deficit 3.0 -4 2.0 -6 GDP growth 1.0 Potenal GDP -8 2008 2009 2010 2011 2012 2013 - Deficit Foreign Domesc Privazaon 2007 2008 2009 2010 2011 Source: Ministry of Finance and World Bank staff esmates Figure 10: Posive balance of payment with strong capital and financial inflows 30.0 Balance of Payments % of GDP 5 months 25.0 of Import 4 cover 20.0 15.0 3 US$ billions 10.0 Official 5.0 2 reserves 0.0 Overall -5.0 1 balance -10.0 0 -15.0 -20.0 -1 -25.0 A pr A pr A pr A pr A ug A ug A ug A ug De c De c De c De c 2007 2008 2009 2010 July Capital Service 2007 2008 2009 2010 Goods overall Source: Central Bank of Kenya and World Bank staff esmates expected slowing in the growth in consumpon. Although the smulus can be credited with in- In fiscal year 2009/10, the government's revenues creased economic acvity in 2010, there are con- stayed 2 percent below target despite a remarkable cerns about the increase in government debt, par- increase of 10 percent compared to the year be- cularly when conngent liabilies like pensions fore. These lower than expected revenues, in con- are included. In the medium term, fiscal consolida- juncon with accelerated expenditures in the first on will be essenal for the government to reduce half of 2010, led to a relavely high budget deficit and maintain the debt to GDP rao at its targeted of 7 percent. In 2009/10 the fiscal deficit was met range of 45 percent. through increased domesc borrowing at 4.8 per- cent of GDP compared to a 3.0 percent target, with Kenya has adopted a pragmac approach to debt the balance coming from external financing (see fig- management. Kenya is one of the few African coun- ure 9). tries that has a debt management strategy with December 2010 | Edion No. 3 6 The State of Kenya's Economy clear medium-term and long-term debt targets. an average of US$ 735 in remiances from abroad, Debt levels have grown in Kenya but not as much as at an average of US$ 105 per transacon and 7 in other countries. Furthermore, debt restructuring transacons per year. While this resource flow rep- has been on going to reduce the cost of the debt resents a significant share of GDP, the effecve role and increase the maturity profile. that remiances can play in dealing with economic shocks, in providing general access to financial re- 1.3 Balance of Payments sources and indirectly helping to reduce poverty is somemes overlooked. Most Kenyans who receive The overall balance of remiances rely on this money to cover at least payments posion has some of their daily expenses such as food, clothes, been posive in 2010 housing, ulies, and medicine. A quarter of re- and the outlook remains Internaonal spondents say that all of their remiance money is posive. The strong per- remiances used for these daily necessies. formance of the service account will dampen the flowing into Kenya 1.4 Financial Sector Developments pressure on the external for twelve months account. However the to June 2010 stood Kenyan banks are well-capitalized and soundness structural current ac- at US$ 1.9 billion. indicators of the banking system remain above count deficit will remain the statutory minimum. The total capital to total in the range of 5 to 6 per- risk-weighted assets rao stood at 20 percent, sll cent of GDP as imports of goods will outpace the well above the statutory minimum of 12 percent. growth of exports. The current account deficit was Kenyan banks are also on track in implemenng the financed by strong inflows in the capital and finan- CBK's requirement to have a minimum core capital cial account with a posive overall balance (see fig- of Kshs 1 billion by 2012. As of September 2010, ure 10). Consequently, the Central Bank has rebuilt 27 banks had reported core capital in excess of the and increased its reserves, now equivalent to 3.9 Kshs 1 billion, well in advance of the deadline. Also, months of import cover. the rao of non-performing loans (net) to gross loans declined further by 200 basis points, from 9.4 Internaonal remiances to Kenya exceed Foreign to 7.4 percent during the first half of 2010. Liquidity Direct Investment (FDI) and aid combined. Aer levels remained in excess of 40 percent, well above a strong performance in 2008, FDI petered out in the CBK statutory requirement of 20 percent. 2009-10, but remiances connued to be a stable and reliable source of foreign exchange. Remianc- Kenya's commercial banks operate within a tradi- es recorded remarkable growth before the financial onal banking model where interest on loans with crisis increasing to about 6 percent of GDP, more short maturies is their principle source of earn- than double the level of SSA which is at 2.4 percent ings, generang about three quarters of the gross of GDP. From a miscellaneous trade accounng income. For Kenyan banks, commercial loans and item, remiances are now a widely recognized flow consumer advances are the principle assets largely of foreign financing with important implicaons for funded from retail savings deposits (85 percent). development. This model negavely affects the growth of medium and long-term lending which is required for projects Recent esmates indicate that internaonal re- with long gestaon periods, especially those in the miances flowing into Kenya for twelve months real sector. to June 2010 stood at US$ 1.9 billion (Kshs 154 billion).¹ A recent World Bank-CBK survey indicates Structural rigidies in the credit market have his- that 14 percent of Kenyan adults regularly receive torically made it difficult to squeeze the interest ¹The World Bank esmates the volume of remiances at approximately US$ 1.9 billon for the 12 months ending June 2010, while CBK esmates this at US$ 0.6 billon. The differences in the esmates can be reconciled by accounng for: (i) remiances captured in the errors and omissions item of the Balance of Payments, (ii) remiances coming in through informal channels and retained in foreign currency, (iii) mis-classificaon of remiance transfers as business transacons and investments by banks and money transfer operators, and (iv) informal remiance transfers through the under-invoicing of import receipts (the "hawala" system). Adjusng for these issues will impact both capital and current account items of the BoP. December 2010 | Edion No. 3 7 The State of Kenya's Economy Figure 11: Interest rates movement 15.0 Real interest rates 11.0 10.0 10.0 CBR 9.0 5.0 8.0 7.0 pe r c e nt 0.0 p e rc e n t 6.0 91 days -5.0 5.0 TB -10.0 4.0 3.0 -15.0 Lending rate 2.0 Deposit rate -20.0 1.0 J an J an J ul y J ul y J ul y J ul y Jan Jan Jan Jan Oc t Oct Oc t Apri l Apri l Apri l Apri l Se p t Se p t Se pt May J an J an May May May Se p t 2007 2008 2009 2010 Source: Central Bank of Kenya and World Bank staff esmates Figure 12: Banks have kept the interest rates spread at 10% rate spread. The spread² has remained relavely due to structural rigidies steady hovering around 10-12 percent for the last 12 few years (see figure 11). As can be seen in figure 12, which decomposes the spread's components 10 3.19 over 2006-09, overheads, provisions and profit mar- 3.01 8 3.24 gins remain key drivers that have proven inelasc to 3.85 1.54 change. Overheads have come down from historic pe r c e nt 6 1.51 1.39 highs in the early 2000s, but have increased since 1.65 1.2 1.62 2006 as banks have aggressively expanded their 4 1.64 0.85 branch networks and invested heavily in Informa- 2 4.22 3.89 on Technology (IT). Provisioning for bad debts also 3.1 3.16 increased in response to spikes in food and energy 0 prices and the economic slow down that followed 2006 Overheads Provisions 2007 Reserve Requirement 2008 Taxes 2009 Profit Margin the global financial crisis in 2009. Source: Central Bank of Kenya and World Bank staff esmates Spreads are likely to remain around 10 percent in from investments in monitoring and evaluang the short to medium term, but ongoing reforms credit risk, as well as improvements in the external and new legislaon will improve access to finan- environment brought about by the introducon of cial services. The introducon of agency banking credit reference bureaus, land registries, and more in 2009, along with the licensing of deposit taking aggressive enforcement of creditor rights by the ju- micro finance instuons, are expected to increase dicial system, will also reduce the need for provi- access to financial services and improve efficiency sions and bring down spreads in the medium-term. in the sector. In addion, branchless banking regu- laons enacted in 2010 will allow banks to expand 1.5 Snapshot on EAC Integraon outreach and reduce costs without invesng in brick In July 2010, the EAC adopted a common market and mortar infrastructure. Non-performing loans protocol which creates a free market of more than have connued declining and banks are expected 130 million people with a combined GDP of US$ to relax their provisioning buffers as the Kenyan 72 billion. Within the EAC, Kenya has the strong- economy connues to grow, and banks expand and est economy contribung 40 percent to EAC's to- diversify their credit porolios. Efficiency gains tal GDP (see figure 13). Overall, the EAC region has ²Defined as ex-post lending minus deposit rates (calculated from banks' balance sheets and income statements). December 2010 | Edion No. 3 8 The State of Kenya's Economy been a dynamic economic area over the last 5 years · Uganda and Rwanda enjoy a strong fiscal posi- averaging 5 percent growth. However, at US$ 550, on, with fiscal deficits below 2 percent of GDP. the average incomes of EAC cizens are only half of But as observed in earlier secons of this update, SSA's which is esmated at US$ 1,115 (2009). While Kenya's current fiscal posion can be aributed there has been a process of convergence in incomes to its current fiscal smulus program and remains due to high and sustained growth in Rwanda, Ugan- broadly sustainable. With regard to external vul- da and Tanzania, a number of differences remain: nerability, Burundi has the highest levels of exter- nal debt and a high current account deficit--signs · The EAC has broadly three income groups. Kenya of external vulnerability--while the rest of the has East Africa's highest standard of living with a EAC enjoy broadly favorable debt dynamics. per capita income of US$ 757. Tanzania, Ugan- da, and Rwanda have very similar income levels of some US$ 500 per capita, about two thirds of 2. The Outlook for 2011 and 2012 Kenya's income levels. Burundi has by far the low- est levels of income. With a per capita income of 2.1 Growth Expectaons US$ 159 per capita the average Burundian only earns 20 percent of the average Kenyan (see fig- ure 13). F or the first me since 2007, Kenya is expected to grow above 5 percent in 2011. The higher growth will create another hope that Kenya may · Kenya is more diversified than the partner coun- again be at a point of its economic development tries, and manufacturing and services account where high sustained growth is possible. In the last for more than two thirds of GDP, slightly above three decades, Kenya experienced only two epi- the average for SSA. However, the enre EAC re- sodes of such relavely high growth for three years gion performs poorly in export capacity. Kenya or more: from 1986-1988 the economy expanded has the highest index of exports per capita at US$ in excess of 5 percent and during 2004-2007 Kenya 200, but is sll significantly below the SSA aver- achieved an average growth rate of 5.8 percent. age of about US$ 350. As we noted in the sec- Both periods were characterized by polical stabil- ond edion of Kenya Economic Update, Kenya's ity, a beer investment climate, and a posive glo- export engine is weak, but it is even weaker for bal economic environment. the EAC region (see figure 13). Figure 13: Kenya currently contributes 40 percent of EAC's GDP and has the most diversified export sector 45.0 757 Exports are very low in the entire EAC Region 40.0 Share of EAC GDP 400 GDP per capita 35.0 SSA 503 30.0 481 506 300 Export 25.0 per capita 200 20.0 $ Kenya 15.0 Tanzania 100 159 Uganda 10.0 Rwanda Burundi 5.0 0 45 50 55 60 65 0.0 Manuf acturing and Serv ices % GDP Kenya Tanzania Uganda Rwanda Burundi Size of the Bubble GDP per capita Source: World Development Indicators and World Bank staff esmates 2009 December 2010 | Edion No. 3 9 The State of Kenya's Economy The outlook for 2011 and 2012 is promising: for the to address the infrastructure gap are expected to year 2011 the World Bank forecasts that growth emerge from private sources. The recent set up of will be 5.3 percent and could even reach 6.0 per- the Public Private Partnership secretariat (effecve cent (key assumpons underlying the forecast are since February 2010), housed at the Ministry of Fi- presented in Annex 4). Aer a peaceful and broadly nance, is expected to enhance Kenya's ability to fi- successful referendum on the new constuon, the nance investment in infrastructure. economic indicators point to a full recovery and possibly takeoff in the medium term. Successful Growth in private consumpon is expected to re- and mely implementaon of the constuonal re- main stable at about 4 percent in 2011 and 2012. forms will send a posive signal to the private sec- As the government phases out the fiscal smulus, tor and bolster business confidence. Under the high government consumpon will be scaled back and case scenario the economy could achieve a growth is projected to grow at 3.4 percent in 2011. Within of 6.0 percent in 2011, maintaining the momentum the EAC, Kenya's overall growth will be less than its into 2012 (see figure 14). neighbors, but closely matching SSA's average (see figure 15). In the medium term growth will be driven by in- vestment which is projected to grow at 13 percent The impact of infrastructure investment is expect- in 2011 and 12 percent in 2012. Public Investment ed to pay off, resulng in a robust performance for will connue to grow in industry, notably manufacturing. The performance line with planned capi- of the sector is likely to be negavely impacted if tal spending. However, ulity prices increase because of dry condions at recent spikes in public Aer a peaceful the end of 2010 and in the first part of 2011. How- investment, notably in and broadly ever, demand arising from the larger EAC common infrastructure projects successful market and lower infrastructure costs resulng from under the government's referendum on the recent investments will cushion the sector from the smulus program, will impact of higher ulity costs. new constuon, moderate as government reverts to more stringent the economic The EAC common market is also expected to boost fiscal policies in order to indicators point to trade within the region, though non tariff barriers maintain debt at sustain- a full recovery and to trade will constrain the rapid growth that would able levels. As highlight- possibly takeoff in otherwise be achieved. The prices of primary com- ed in Vision 2030, the vast the medium term. modies will hold above the 2010 levels, boosng majority of investments Kenya's overall trade performance, though the slug- Figure 15: Kenya could be entering another decade of high Figure 14: Kenya's recovery connues growth, but it will remain below its neighbors 8 9.0 Selected Comparators 7 8.0 7.0 2011 6 6.0 2012 5 P e r c e nt 5.0 4 4.0 3 Base case 3.0 2 Low case 2.0 High case 1 1.0 0.0 0 2007 2008 2009 2010 2011 Ethiopia Tanzania Uganda Rwanda Kenya SSA South average Africa Source: World Bank staff esmates Source: World Bank staff esmates December 2010 | Edion No. 3 10 The State of Kenya's Economy gish recovery of the North American and European through an increase in the cost of ulies, threat- economies will dampen a rapid increase in exports ening the economic forecast. Unless countervailing to those markets. Overall, exports and imports are measures are taken early enough, the effects of a expected to expand at the same pace, and the cur- drought could slow down consumpon growth and rent account deficit will remain in the range of 5 to crowd out fiscal space for crucial investments as 6 percent of GDP (see government responds to drought-related emergen- table 1 and Annex 9 for cies. more details). Investors will be The run up to the elecons at the end of 2012 2.2 Key Risks to the 2011- could dampen investment flows and growth. In looking for signs of 12 Forecast the past, naonal elecons have been associated polical stability, with lower growth and oen with major negave As in previous years, the which if it material- shocks, such as in 2008 following the post-elecon agriculture sector re- mains vulnerable if the izes, and external violence. However the successful campaign and markets remain largely peaceful vote for a new constuon in Au- drought cycle returns. favorable, will gust 2010 offers hope that the coming elecon cycle The recent weather fore- will follow the same path. Investors will be looking cast for a dry spell at the minimize risks to for signs of polical stability, which if it materializes, end of 2010/early 2011 Kenya's economic and external markets remain favorable, will mini- could spill over from outlook. mize risks to Kenya's economic outlook. agriculture to industry Table 1: Key Macroeconomic Indicators Variable 2007 2008 2009 2010 2011 Actual Esmate Projecons GDP 7.0 1.6 2.6 4.9 5.3 GDP per Capita 4.3 -1.1 -0.2 2.3 2.6 Private Consumpon 7.3 -0.4 3.8 5.1 3.9 Government Consumpon 7.2 3.7 5.5 4.1 3.4 Gross Fixed Investment 13.3 8.9 0.6 13.8 12.6 Exports, GNFS 6.0 3.6 -7.0 12.0 8.9 Imports, GNFS 12.7 5.3 -0.2 14.5 8.6 Current account Balance (% of GDP) -3.8 -6.6 -5.5 -5.8 -6.0 Populaon 2.7 2.7 2.8 2.7 2.7 Potenal GDP 4.3 5.5 5.4 5.4 5.4 Source: World Bank staff esmates December 2010 | Edion No. 3 11 The ICT Revoluon and Mobile Money T his special focus reviews Kenya's transformaon in the ICT sector and analyses the impact of mobile money transfers, an area where Kenya has become a world market leader. The ICT sector in Kenya has undergone a revoluon which is transforming the lives of Kenyans. Home-grown innovaons in the sector have been a major contributor to economic growth. Since 2000, mobile prices have plummeted and usage rates have increased exponenally. Today, 9 out of 10 Kenyans have access to a mobile phone, and the introducon of mobile money has brought financial access to millions of Kenyans. 3. Kenya's ICT Revoluon bust growth in ICT growth overall GDP growth would have averaged a lackluster 2.8 percent--close to the 3.1 The Transformaon of the Last Decade populaon growth rateand incomes per capita I n the first decade of the new millennium, Kenya would have stagnated. has undergone a remarkable ICT revoluon. At the close of the 1990s, less than 3 percent of Ken- Kenya has experienced a triple technology trans- yan households owned a telephone, and fewer than formaon: mobile phones, mobile money, and in- 1 in 1000 Kenyan adults had mobile phone serv- ternet. The first mobile telephone services in 1993 ice. At the end of 2010 there are nearly 90 mobile began on an analog system and the service was ex- phone subscripons for every 100 Kenyan adults. pensive. Only 20,000 subscribers were connected Among the many uses to which cell phones have over the next seven years. Following deregulaon been put in Kenya, the most innovave is mobile and paral privazaon of the telecommunicaons money. Starng with the M-PESA system launched sector, the number of mobile phone subscribers by Safaricom in 2007and later joined Figure 16: Since 2000, the telecom sector outperformed by Zain's Zap system and Yu's yuCash in all sectors in the economy 2009, and Orange Money in 2010mo- bile money has rapidly become a fix- Annual average growth 2000-2009 by sector ture in the lives of Kenyans, extending Post and telecommunicaons Hotels and restaurants a sophiscated form of financial access Construcon to a wide populaon. Transport and storage Wholesale and retail trade, repairs The ICT sector's growth has outper- Water supply formed every other sector, expanding Health and social work by 23 percent annually during the last Farming of animals decade (see figure 16). The sector is Manufacture of food, beverages and tobacco Mining and quarrying now six mes larger than it was at the Real estate, renng and business services beginning of the decade. This remark- All other manufacturing able growth can be aributed in part to Other community, social and personal services innovaons, such as the introducon Educaon of mobile money discussed in the next Electricity supply GDP average growth in secon. Private households with employed persons the period is 3.7% Growing of crops and horculture Financial intermediaon Investment in ICT during the last dec- Forestry and logging ade was a major contributor to Ken- Fishing ya's growth. In the period 2000-2009, Agricultural and animal husbandry services Kenya experienced moderate to strong -5.0% 0.0% 5.0% 10.0% 15.0% 20.0% 25.0% growth of 3.7 percent. However, our simulaons show that without the ro- Source: Kenya Naonal Bureau of Stascs and World Bank staff esmates December 2010 | Edion No. 3 13 Special Focus ­ The ICT Revoluon and Mobile Money exploded from 114,000 government enacted the Kenya Communicaons in 2000 to 5.26 million Act, implemented by the Communicaons Commis- in 2005, and is projected sion of Kenya (CCK) which introduced compeon to reach 22 million at the The average call in the cellular mobile industry and allowed new end of 2010. Since mobile tariff declined players to compete with Telkom Kenya. Currently, money was introduced in from Kshs 16.8 per there are four licensed operators: Safaricom, Celtel/ early 2007, the number minute in 2002 to Zain, Yu and Telkom Kenya. of mobile money custom- about Kshs 3 per ers has now reached 15 Compeon, high call volumes, and infrastructure million. Access to the in- minute in 2010. investments have been the basis for the sharp re- ternet, via personal com- ducon of average call tariffs. The average call tar- puter or mobile phone, now exceeds 8 million and iff declined from Kshs 16.8 per minute in 2002 to has now reached a pping point for further accel- about Kshs 3 per minute in 2010 (from 20 US cents eraon. to 4 US cents -- see figure 18). Furthermore, these call tariffs have now been extended to several des- Kenya is part of a telecom revoluon which is naons in North America and Asia, enabled by the spreading across all of Africa. Many other countries recent successful installaon of three fiber opc ca- on the connent have also experienced dramac ex- bles. Even as the reducons in call tariffs are wel- pansion in mobile phone access. Along with South comed, close aenon to other crucial segments Africa, Ghana, and Nigeria, Kenya's mobile penetra- of this sector should be upheld, and parcularly on on rate is among the highest (see figures 2 and 17). the data market. Providers could compensate for The mobile phenomenon in Kenya is disnguished lower call tariffs by seng higher prices and by re- by the unparalleled rise of mobile money. ducing investments in the data market, hence con- Figure 17: Africa's telecom revoluon straining market development. The Kenyan cost of accessing the internet, for example, has remained 160.0 Mobile phone subscripons per 100 adults (Age 15+) prohibive. 140.0 South Africa Figure 18: Subscribers increase exponenally while 120.0 prices plumment Ghana 100.0 Mobile phone users and average call tariffs Kenya 25 25 80.0 Nigeria Tanzania 20 20 60.0 Uganda kshs/min 40.0 15 15 Million 20.0 Ethiopia 10 10 0.0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 5 5 Source: World Bank staff esmates - 0 3.2 Explaining Kenya's ICT Revoluon 2000/01 2002/03 2004/05 mobile phone subscribers 2006/07 2008/09 ave tariff khs/min The regulatory environment permied compe- Source: World Bank staff analysis of data from CCK 2010 on and played a catalyc role in the phenomenal growth of ICT. The rapid growth in Kenya's ICT sector followed major reforms in the telecommunicaons Openness to private-public policy dialogue creat- sector in the late 1990s, aer years of government ed incenves for private sector investments. The resistance. The turning point was in 1998 when the ICT sector is primarily private sector driven³ even ³The investors in these cables include internaonal firms and the Internaonal Finance Corporaon (IFC)--the World Bank's private sector lending arm. The Bank has also invested US $114.4 million in Kenya's Transparency and Communicaons Infrastructure Project, which is supporng establishment of digital villages and other acvies through the Kenya ICT Board. Kenya is connected to three undersea cables, SEACOM, TEAMS and EASSy, which landed in Mombasa in 2009/10. The cables are shared between Kenya and other countries in Eastern and Southern Africa region. December 2010 | Edion No. 3 14 Special Focus ­ The ICT Revoluon and Mobile Money though the government has played a major role More recently, the agency banking regulaons through investment in crical infrastructure (see have revoluonized the financial services land- figure 19). Available data shows that over US$ 3.2 scape in Kenya. Branchless banking regulaons fa- billion was invested in mobile services between cilitated the extension of (limited) banking services 2001/02 and 2009/10 while another US$ 3 billion through agent networks, addressing concerns raised was invested in fixed telephone services. Invest- by banks on the need for a level playing field. The ment in the three undersea fiber opc cables alone agent banking model turns non-bank outlets into is esmated at US$ 700 million. Another US$ 60 financial services providers. Building on networks million was esmated to have been invested in in- developed by the telecommunicaons industry, an ternet and data services between 2006 and 2010, addional 5,900 agents have been approved, push- and US$ 500,000 in Business Process Outsourcing ing forward financial inclusion froners to provide (BPO) in 2007. financial products such as savings and insurance to the unbanked. Balancing innovaon with prudence in regulaon has been key in this process. In the case of mobile As mobile money operators provide financial prod- money, regulaon followed innovaon, and not ucts such as savings and insurance to the unbanked, vice versa. The CBK, the CCK, and other relevant further enhancements to the regulaon of mobile regulatory agencies allowed innovaon to proceed money can be made to accelerate financial inclu- even before they developed an appropriate regula- sion and deepening in Kenya. The enhancements tory framework for the operaon of mobile money. would include: (i) consumer protecon regulaons Mobile money and associated financial services and financial literacy tailored to mobile money use; have been led by telecommunicaons firms (unlike (ii) ered Know-Your-Customer (KYC) regulaons most financial innovaons which are bank-led), and that permit immediate account opening with mini- have benefited from the robust financial regulatory mum barriers for poor people, with a progressive environment (see box 2). ghtening of KYC as their usage of financial services Figure 19: Private sector investment in mobile telephones peaked in 2006 while employment connued to rise Investment in ICT Cummulave New Employment 50 100,000 90,000 40 80,000 70,000 Khs 30 billion 60,000 50,000 20 40,000 30,000 10 20,000 - 10,000 2001 2003 2005 2007 2009 - Investment in other ICT Investment in mobile phone Source: CCK, KNBS and World Bank staff esmates Source: Alexandre, Claire, Ignacio Mas, and Dan Radcliffe. 2010. Regulang New Banking Models that can Bring Financial Services to All. Bill & Melinda Gates Foundaon Note. December 2010 | Edion No. 3 15 Special Focus ­ The ICT Revoluon and Mobile Money Box 2: Enabling Reforms and Regulaon Permied Growth Deregulaon facilitated compeon. Telecommunicaons sector reforms followed long, protracted years of govern- ment's reluctance to dismantle the monopoly of the Kenya Posts and Telecommunicaons Corporaon (KPTC). The Kenya Communicaons Act (1998) unbundled KPTC into three enes: Telkom Kenya, the Postal Corporaon of Kenya, and a market regulator - the CCK. The Act, which was amended in 2009, regulates the informaon and communicaons sector, and also enables CCK to enforce specific regulaons to encourage fair compeon. There are at least 20 regu- laons relang to issues such as universal access, tariffs, interconnecvity, fair compeon, broadcasng, consumer protecon, dispute resoluon and others. Consequently, the telecom market has evolved into a compeve sector notably: · 1999 ­ Telkom Kenya established its subsidiary, Safaricom (1st mobile operator) · 2000 ­ Celtel Kenya (formerly KenCell) entered the market (2nd mobile operator) · 2004 ­ Econet Wireless was awarded the 3rd mobile operator license (Yu) · 2004 ­ over 70 Internet Soluon Providers (ISPs) licensed in Kenya (Telkom Kenya monopoly came to an end) · 2007 ­ award of 3G license to Safaricom · 2009 ­ landing of TEAMS cable in Mombasa In the case of mobile money, regulaon followed innovaon. CBK inially regulated and supervised mobile payments service plaorms from a strict payments systems perspecve, allowing innovaon outside the purview of strict banking supervision rules. This approach permied the growth of innovaon with minimum barriers to entry. All four mobile phone providers now have a mobile money product: · 2007 - M-Pesa was launched by Safaricom · 2009 - Zap money was launched by Zain Kenya, · 2009 - YuCash was launched by Yu · 2010 - Orange Money launched by Telkom Kenya Source: World Bank staff grows; and (iii) creang regulatory space for a class represenng 1.6 percent of the populaon. By of non-bank e-money issuers authorized to raise 2007, the number increased to 2.9 million (7.5 per- deposits and process payments, but not to interme- cent of the populaon), and at the end of 2010 the diate funds. number of users exceeds 8 million, a penetraon rate greater than 20 percent. A major innovaon The goal of an "opmal regulatory regime" should by the mobile phone companies is the introducon be to have rules ght enough to protect users and of data enabled smart phones, which give Kenyans discourage fraud, but loose and open enough to access to internet through their mobile phones. encourage innovaon and the development of new services. Therefore, banking and telecommunica- The ICT infrastructure has also created opportu- on regulators will need to cra a strategy to act nies for BPO. The first call center was licensed in in concert to influence compeon in the mobile 2004 and by 2007 the number had increased to 18. money and retail payments space in Kenya, without The government has floated a public-private part- undermining growth. This will be parcularly im- nership to develop an ICT city with a 5,000 BPO seat portant at a regional level as well, as the EAC moves park projected to cost US$ 1 billion. Private inves- towards common markets in telecommunicaon tors are also building smaller parks in several parts services, mobile money, and banking. of the country. Other policies, such as the govern- ment's Vision 2030 which places ICT as one of the 3.3 Will Kenya be the Home of Africa's key development pillars, also encourages growth "Silicon Valley"? and market deepening. In addion, the govern- Internet access has also achieved a significant, ment has acvely expanded electricity connecons though less dramac, growth rate. In 2004, there to urban and rural areas, providing a plaorm for were an esmated 500,000 acve internet users, private sector investment in IT enabled products and services. December 2010 | Edion No. 3 16 Special Focus ­ The ICT Revoluon and Mobile Money applicaons that the broad penetraon of ICT in Kenya can make it to the short list of possible off- the populaon has enabled. The recent explosion shore ICT Hub locaons. A comparison between of mobile money, described in the next secon, al- Kenya and India's ICT sectors shows that Kenya has ready has fueled a significant increase in the growth a cost advantage, but of the financial services sector. ranks poorly on talent compared to India (see 4. Mobile Money Annex 5). India has a Kenya can now significant advantage 4.1 What is Mobile Money? leverage the larger in terms of qualified Mobile money systems consist of electronic mon- manpower, thanks to EAC common mar- ey accounts that can be accessed via mobile phone the large populaon. ket to increase its (see box 3). There are currently four mobile money Kenya can now leverage talent pool and, systems in Kenya, each run by a mobile phone op- the larger EAC common thus, could evolve erator: Safaricom's M-PESA, which was introduced market to increase its into an ICT hub for in March, 2007; Zain's Zap, iniated in January, talent pool and, thus, 2010; yuCash, started in December, 2009, by Essar; the region. could evolve into an ICT and, Orange Money (Iko Pesa), which was launched hub for the region. In- in November 2010 by Telkom Kenya. M-PESA is by frastructure (mobile & IT) is well developed within far the largest system, accounng for more than 90 Kenya's urban areas, but lack of supporve infra- percent of mobile money subscripons. The genesis structure, e.g. in electricity, limits technology up- of mobile money is described in Annex 8. take in the rural areas. 4.2 Exponenal Growth There are a number of other ICT innovaons where Kenya is already leading the way. Whether it is the Since 2007, mobile money usage grew rapidly creaon of BPOs as Kenya becomes a regional or a through 2008 and 2009. Extrapolang from data global ICT hub, or the use of smart phones connect- through April 2010 (figure 20), we esmate that as ing the internet to the mobile user, it is possible to of December 2010, the value of person-to-person envisage important innovaons happening in a vari- transacons alone will exceed Kshs 38 billion per ety of sectors--from manufacturing to agriculture, month, more than 20 percent of GDP. from construcon to tradeall fueled by cung Due to the rapid growth of mobile money usage, edge developments and innovaons in ICT. Kenya has received parcular aenon for developing the portrait of mobile money users is a moving and applying "crowd-sourcing" technology which target. Figure 21 summarizes the trend of mobile allows for real-me monitoring of social, polical money users combining M-PESA customers, Zap and economic events (see Annex 6 for a descripon subscribers, and yuCash subscribers, based on data of other innovaons that are happening in Kenya on Figure 20: Mobile money: expected to exceed 20% of GDP three providers from each of the at the end of 2010 through August the mobile phone plaorm). 25% Kenya appears poised for another growth spurt 20% Monthly Value of Person-to-Person Transfers as % of GDP with ICT again making a healthy contribuon. The 15% % of GDP Trend with Forecast massive investment that went into powering Ken- 10% ya's ICT revoluon contributed significantly to the 5% economy's last growth spurt in 2004-07. The future significance of ICT to Kenya's growth will not be a 0% 2007 2008 2009 2010 result of direct investment in the sector, though a sustained level of investment will remain impor- Source: World Bank analysis of data from Safaricom, Zain, and tant. Rather, economic growth will come from the the CBK See Annex 7 for data sources on the stascs and projecons used in this report for mobile phones and mobile money. December 2010 | Edion No. 3 17 Special Focus ­ The ICT Revoluon and Mobile Money Box 3: How Does Mobile Money Work? To have a mobile money account and to make a deposit, a customer must own a cell phone SIM card from the mobile operator and register for a mobile money account. The customer then can make cash deposits at the physical offices of one of the operator's mobile money agents. (It is now also possible for customers to make direct electronic transfers to their mobile money from certain bank accounts.) These cash deposits create electronic money credit in the customer's account. Using their mobile money, customers can make person-to-person transfers of their mobile money to the accounts of other mobile money users on the same network. They can also use mobile money to pay bills and to buy phone air- me. Withdrawals (conversion of mobile money to cash) can be made at the offices of the network's mobile money agents. It is also possible for a mobile money customer to make a transfer to someone who is not registered with the same network. In this case, when noce of the transfer is received in the form of an SMS the text message, the recipient can also receive the cash at a mobile money agent at an addional fee. 1 2 3 Customer A Customer A sends electronic (sender) money to Customer B deposits cash Customer B (receiver) with Agent X withdraws cash from Agent Y Electronic money Electronic money transferred from Agent X's transferred from account to Customer B's Mobile Money Bank Account account to Agent Customer A's account Y's account Flow of real money Flow of electronic money Flow of electronic money Person-to-person transfers between Agents and between Agents and between Agents and of electronic money commercial bank commercial bank customers between customers Source: World Bank staff December 2010 | Edion No. 3 18 Special Focus ­ The ICT Revoluon and Mobile Money Figure 21: By end 2010, Kenya will have 15 million 2010. By mid 2010, the number of mobile money mobile money customers customers exceeded 13 million, and based on recent growth rates, we project that in December 2010 the 16 15 Mobile Money Customers 14 13 12 11 number of customers will reach 15 million, equiva- 10 lent to more than 2 out of every 3 Kenyan adults. (millions) 9 8 7 6 The number of financial agents has grown to ex- 5 4 3 2 1 0 ceed the tradional banking outlets by a wide mar- 2007 2008 2009 2010 gin. The expansion of the M-PESA agent network, close to 17 thousand agents in mid-2010, relave Source: World Bank staff analysis of data from Safaricom, Zain, to coverage by brick and mortar bank branches, and Yu. ATMs, and post offices (figure 22) demonstrates the Note: The doed lines indicate projecons based on recent growth rates. A large jump in mobile money customers in July 2010 occurred potenal of mobile payments and banking services during the month when the government iniated the requirement to dramacally expand formal financial inclusion in that all SIM cards be registered. Because registered mobile money Kenya. customers are not required to go through any addional SIM card registraon process, it is likely that the registraon requirement spurred many people to sign-up for mobile money. The mobile banking network connues to expand. Considering just M-PESA, the number of agents and Figure 22: M-Pesa stores outnumber bank branches the number of customers (figure 23), as well as the by a factor of 20 cumulave value of transfers are all growing rap- 100,000 idly. 20,000 Crical to the effecveness and reliability of the 1,510 1,030 mobile money network is the interface between 800 agents and customers. By mid-2010, each M-PESA agent was serving an average of 600 clients. As 440 more sophiscated banking services are added on top of the M-PESA plaorm (such as banking through M-KESHO), the cash reserves, equipment quality and literacy levels of these agents will need Note: This figure is not drawn to scale Source: Mas, Ignacio and Dan Radcliffe, 2010, Mobile Payments to improve to sasfy regulatory requirements. The Go Viral: M-PESA in Kenya, Bill & Melinda Gates Foundaon mobile money network will see constant innovaon as suppliers compete for an increased share of the Figure 23: By August 2010, M-Pesa had enlisted 12.6 million customer base. customers and nearly 20,000 agents countrywide Customers (Millions) Agents (Thousands) 4.3 Users 14 25 While mobile money has achieved penetraon 12 across age groups, older customers are much more 20 10 Customers (Millions) Agents (Thousands) likely to have used mobile money exclusively for receiving transfers. Mobile money usage is high- 15 8 est among those aged 25-29 and falls with age, but 6 10 even among the oldest Kenyans--those over age 65--about half use mobile money (see figure 24). 4 5 Usage paerns vary greatly by age group. In par- 2 cular, among older mobile money users nearly half 0 0 have only received transfers via mobile money and have never sent money. In contrast, among younger Source: World Bank staff esmates December 2010 | Edion No. 3 19 Special Focus ­ The ICT Revoluon and Mobile Money users, less than a quarter are "receivers only." Figure 25: The rich use mobile money dispropoonately, but poor Kenyans are catching up Majories of both men and women have used mo- bile money. Women are slightly less likely than men 100 90 Jan-Mar 09 to use mobile money, and much more likely to only Aug-Sep '09 80 receive and not send funds. Data from August 2009 70 Dec 2010 (Projecons) show nearly 59 percent of men to have used mobile 60 pe r c e nt 50 Figure 24: Mobile money usage is highest among the 40 younger, but 45% of those above 50 years also use it 30 20 100 10 90 Jul-Aug 2009 0 80 Dec 2010 (Projecons) Poorest 2nd poorest Middle 2nd wealthiest Wealthiest 70 Quinles of wealth index 60 pe r c e nt 50 Source: World Bank staff esmates 40 30 Mobile money usage is highest among urban Ken- 20 10 yans but also substanal among rural residents. As 0 of August 2009, 47 percent of rural Kenyan adults 15-17 18-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65+ and 69 percent of urban Kenyan adults had used Age groups mobile money. For December 2010, we project that these figures have grown to 59 percent of rural Source: World Bank analysis of Kenya 2009 AudienceScape Survey. Projecons to December 2010 are based on AudienceScape data com- Kenyans and 86 percent of urban Kenyans. In Nai- bined with subscriber growth rates and are described in Annex 5 robi mobile money usage is nearly universal (figure 26). Majories of the adult populaon in Nyanza, money, compared to 51 percent of women. Extrap- Western, Ri Valley, Central, and Eastern provinc- olang from aggregate subscriber growth trends, es are mobile money users. Usage rates in Coast we esmate that in December 2010, 64 percent of and Eastern provinces are lowest and low relave Kenyan women and 73 percent of Kenyan men are to their mobile phone ownership rates (figure 27). mobile money users. Among female mobile mon- This suggests that these areas have high potenal ey users, one-third (34 percent) only used mobile for growth in the use of mobile money. money to receive and not to send. Only 23 percent of men are in this category. Figure 26: Usage rates are higher in Central and West of Kenya, and men are slightly more likely to use mobile money Mobile money was inially concentrated among the wealthiest Kenyans but has grown rapidly Percent among the poor. Figure 25 shows the fracon of Percent adult Kenyans who have used mobile money at dif- 73 ferent points in me by quinle (groups of 20 per- 64 cent, ranked from poorest to richest). At the begin- EASTERN 53% RIFT VALLEY ning of 2009, 70 percent of the wealthiest quinle 70% WESTERN 73% NORTH and just 15 percent of the poorest Kenyans had used NYANZA 68% CENTRAL 83% EASTERN 37% mobile money. Six months later, mobile money had NAIROBI 95% connued to grow rapidly among the wealthiest-- reaching 83 percent--while it skyrocketed among COAST 42% the poorest, nearly doubling. Based on overall sub- scripon growth rates, we project that by the end of 2010, 4 out of 10 of the poorest 20 percent of Ken- Source: World Bank projecons yans will have used mobile money, along with more than 90 percent of those in the top two quinles. The mobile money usage figures by age group are based on World Bank staff projecons described in Annex 7. The breakdown of usage paerns among users are based on early 2009 FinAccess survey data. December 2010 | Edion No. 3 20 Special Focus ­ The ICT Revoluon and Mobile Money Figure 27: Mobile money usage trails mobile phone senders, mobile money is overwhelmingly the first ownership in Coast and Eastern provices choice of approaches. Two-thirds of respondents in a survey, in early 2009, reported sending remianc- 100 Nairobi es via mobile money rather than alternave meth- ods. As can be seen in figure 29, data from a similar 80 Central survey in 2006 shows that by 2009 mobile money % Households with phones Coast Rift Valley transfer had almost enrely displaced transfers via 60 the post office and via bus/matatu (minibus), which Eastern Nyanza Western were previously popular methods. A substanal 40 North Eastern minority, however, sll send transfers via friends and family. 20 20 40 60 80 100 Figure 29: Within 3 years, mobile money has replaced % Mobile -money (individuals) tradional means of reming funds Source: World Bank staff esmates 2006 2009 4.4. Uses Sending and receiving transfers are the dominant uses of mobile money, and substanal numbers use mobile money for other purposes. Figure 28 shows that in addion to receiving and sending money individuals are also using mobile money to buy airme on their cell phones and as a secure place to store or save money. Figure 28: Sending and receiving transfers are the dominant uses of mobile money Source: World Bank staff esmates Remiances are largely transfers from customers in wealthier urban areas to family and friends in poorer rural areas. This phenomenon explains why of mobile money users in rural areas are receivers only, i.e. they report having used mobile money but never having sent funds via mobile money. An increasing number of mobile network subscrib- ers are purchasing airme directly from their mo- bile money accounts, rather than buying top up scratch cards from approved agents and dealers. Among the reasons why customers would prefer to purchase air me from their accounts is because it offers flexibility and convenience. Mobile network Source: World Bank analysis of FinAccesss survey conducted Janu- operators too could potenally gain through re- ary-March 2009. The survey quesons referred solely to M-PESA. Because Zap had just been introduced at the me of the survey and duced commissions to agents, airme resellers and yuCash did not yet exist, the responses correspond closely to mobile retail outlets, who nevertheless play a crucial inter- money usage in general. mediary role in the mobile money business. The overwhelming use of mobile money is to By mid 2010 nearly Kshs 4 out of every Kshs 10 of send and receive person-to-person transfers. Part Kenya's mobile money transfer (deposits plus with- of these transfers are remiances sent to friends drawals) took place within the capital city. The aver- and family at distant locaons. Among remiance age monthly value of mobile money transfer within December 2010 | Edion No. 3 21 Special Focus ­ The ICT Revoluon and Mobile Money Nairobi is roughly equivalent to that of five prov- Figure 31: Richer quinles deposit substanally more than they withdraw inces - Eastern, Nyanza, Coast, Western and North Eastern (see figure 30). Month-on-month growth 80 rates in value of mobile money transfers are high 70 Billions of Kenyan Shillings Deposits to M-PESA agents throughout the country, 118 percent on average, 60 Withdrawals from M-PESA agents but most remarkable in North Eastern province. 50 40 Figure 30: Nairobi accounts for the largest share of mobile 30 money flows, but the fastest growth is in remote regions 20 (Shares by Province) (Apr ­ Aug 2010) 10 Nairobi 35.8% 0 Richest 2nd Middle 2nd Poorest Unknown Ri Valley 16.5% Richest Poorest Quinle Quinles of Districts Ranked by Poverty Rate Central 12.3% Eastern 10.4% Source: World Bank analysis of data from Safaricom and the Kenya Integrated Household Budget Survey (KIHBS). Data on deposits Nyanza 9.4% and withdrawals at individual agents was aggregated up the level of districts and matched to poverty esmates by districts based on Coast 8.6% the 2005-06 KIHBS. "Unknown Quinle" reflects the total for agents Western 5.2% that could not be matched reliably to a district. North Eastern 1.7% The me paern of flows most likely may reflect Percentages are based on average monthly data for the period April the fact that remiances are being used to pay ­ August, 2010 school fees. Net oulows for the poorest two quin- Source: Safaricom, Zain and World Bank analysis les were close to zero in March and April 2010. A sense of the extent to which remiances via However, in January, February, May, and June 2010, mobile money generate a net flow of funds from net flows to the poorest quinles were posive. wealthier to poorer areas can be seen in figure This paern may reflect remiances to poorer rural 31. The figure shows deposits and withdrawals to areas to pay school costs in those months. and from the M-PESA system for Kenya's districts, grouped into quinles (groups of 20 percent) by Up unl recently incenves to save money via mo- poverty rate during the period January-June 2010. bile money have been low because the accounts Thus, the richest quinle consists of the districts do not pay interest. Thus, the chief advantage to with the lowest poverty rates, and the poorest quin- keeping savings in mobile money rather than in le consists of the districts with the highest pover- cash has been the increased security associated ty rates. The overall net flows through agents are with having the funds in mobile money. Savings via posive (meaning that net funds are entering the mobile money are expected to grow, now that Safa- M-PESA system) because some transacons do not ricom and Equity Bank have introduced a new form take place through agents. In parcular, purchases of account called M-KESHO that can be accessed via of airme and bill payments are oulows that do M-PESA and pays an interest rate of 1 percent. not take place through agents. These data provide an imperfect but nonetheless revealing portrait Mobile money is also used for a variety of consum- of mobile money transfers. Net inflows happen er transacons. Electricity and water ulity bills can overwhelmingly in the wealthier quinles and, in be paid via mobile money and a late 2009 survey parcular, in the richest quinle of districts, which shows 11 percent of mobile money users reporng includes Nairobi. Net oulows for the six month they use it to pay such bills. Survey data shows that period took place in the poorest two quinles of use of mobile money to pay bills is chiefly among districts. These districts encompass 40 percent of wealthier, urban customers. It is less clear how the populaon. much mobile money is used for other consumer transacons, such as market purchases. December 2010 | Edion No. 3 22 Special Focus ­ The ICT Revoluon and Mobile Money Of those who reported sending or receiving money They said there was more remiances flowing into for business purposes in the 2009 FinAccess survey, rural areas, which was perceived to have increased half said they used mobile money. For small busi- local economic acvity in those areas. Preliminary nesses, mobile money has the aracon of conven- results from Mbi and Weil (2010) also suggest that ience, support, cost, sasfacon and security. the growth of M-PESA in communies has been as- sociated with increases in local farm employment. 4.5. Benefits It is not possible to gauge the overall trends in do- Much aenon has been given to the queson of mesc remiances using exisng data, and it is the impacts of the introducon of mobile money. unclear to what extent the growth of remiances Evidence for different effects comes from various via mobile money truly consists of new remiance surveys as well as from Plyler et al. (2010). The lat- flows, rather than simply transfers that took place ter provides analysis of a qualitave study on M- beforehand via other means. But it is likely that the PESA conducted in late 2009 at two rural sites and decline in the effecve cost of sending remiances has spurred greater flows. one urban site. The rise of mobile money also has made possible The clearest and most direct benefits of mobile "just-in-me" remiances. The Plyler et al. study money are the greater convenience, far greater included this remarkable finding: "Many parci- speed, and generally lower cost of transferring funds. This is parcularly true in the case of remit- pants noted that agricultural producvity has gone up, even in areas experiencing drought, because tances, i.e., the transfer of funds from urban to M-PESA enabled the fast transfer of capital when rural or richer to poorer areas. The most common it was most needed. In instances of a farmer run- methods of sending funds before the introducon ning out of seed or a family needing to hire casual of mobile money were through friends and family labor, M-PESA facilitated the fast and safe transfer (by hand), via a bus or matatu , or through the Post of funds to deal with expenses." Likewise, focus Office's Postapay service. These methods posed group parcipants "frequently cited M-PESA as a risk that the funds would be stolen or lost along a means for helping them to pay school fees and the way and took me. Mobile money transfer is get money for medical procedures. This appeared generally cheaper than to help in school aendance and retenon, and to Postapay, and both the seek medical consultaon faster than they would sender and receiver are have otherwise." Evidence for the importance of given instant informa- The most common mobile money for paying school fees come from the on that the transfer has been made. methods of send- fact that net transfers of M-PESA from wealthier to ing funds before poorer areas are larger in months when school fees the introducon of are due. The spread of mobile money may have in- mobile money were While it is difficult to quanfy these benefits, re- creased the volume through friends and miances transferred through M-PESA, Zap, and of remiances and family (by hand), yuCash seem to have generated large benefits for spurred local econo- mies in poor areas. via a bus or matatu, rural households that receive the transfers. For or through the Post example, the combinaon of mobile money and Plyler et al. (2010) found Office's Postapay communicaon by cell phone has reduced the me that focus parcipants required for a request and subsequent transfer of reported a number of service. remiances from days or weeks to mere minutes. observaons about the This huge drop in the costs of coordinaon may well flows and impact of re- have increased not just the volume of remiances miances. They ranked "money circulaon" as the but also their meliness. most important effect of mobile money and said that mobile money had boosted local consumpon. December 2010 | Edion No. 3 23 Special Focus ­ The ICT Revoluon and Mobile Money Small businesses are another beneficiary of mobile money. Half of all small business respondents in the 2009 FinAccess survey indicated they had used mo- bile money to receive payments or pay expenses. Micro-business operators in another survey (Mbo- go 2010) in Nairobi expressed high levels of sas- facon with the accessibility, low cost, security, and convenience of mobile money. Mobile money has been a parcular ben- efit to women, giving them an independ- The combinaon ent place to store and of mobile money manage funds. Evi- Evidence from focus group interviews suggests that mobile dence from focus group and communica- money may have empowered women by giving them an in- interviews suggests on by cell phone dependent place to store and manage funds. that mobile money has reduced the to turn it over." Addionally, "the ability to keep may have empowered me required for a such transacons private was especially important women by giving them request and subse- to women and many female parcipants claimed an independent place quent transfer of M-PESA was the main way to keep their money se- to store and manage funds. Besides facili- remiances from cure. They said that privacy is important both for days or weeks to their ability to store money, and in controlling how, tang trade and inter- what and when the money would be spent." personal transacons, mere minutes. mobile money trans- Mobile money has also improved personal secu- fer empowers certain rity. Plyler et al. note that in their focus group in- household members who have tradionally had less terviews, many respondents highlighted the great- bargaining power, in parcular women. Plyer et al. er security they had because they are able to keep reported, "Women commented oen on M-PESA's funds as mobile money, rather than at home in cash. ability to keep money safely stored on a cell phone, Men in Kibera parcularly focused on the physical because it kept them safe from pickpockets, but danger. Respondents said that safety had improved even more importantly from their husbands. Wom- due to M-PESA because thieves have learnt that few en in all three study areas complained that when people now carry large amounts of cash. Likewise, they had cash in pocket, their husbands (and some- other respondents in the same study indicated that mes other family members as well) would take local businesses and street vendors oen convert their money. With M-PESA, women could either their cash to M-PESA at the end of the day for safe- claim they did not have any money or could refuse keeping. December 2010 | Edion No. 3 24 Special Focus ­ The ICT Revoluon and Mobile Money 4.6. The Future of banking" systems like M-KESHO, large numbers of Mobile Money Kenyans will have not just the financial transacons mechanism of mobile money but also access to a The recent introduc- Mobile money basic suite of modern financial services. on of M-KESHO bank accounts by Equity reduces the im- Many of the effects of mobile money are only be- Bank and Safaricom portance of physi- ginning to be understood, and future work will be have the potenal to cal distance, thus needed to understand its broad implicaons. In the vastly expand the im- integrang millions broadest sense, mobile money does just two things: pact of mobile mon- of Kenyans into the it vastly reduces the costs of the transfer of funds, ey. These are inter- modern economy. parcularly across distances, and also reduces the est-earning accounts cost of the storage of funds. Like the ICT revolu- that can be accessed on as a whole, mobile money reduces the impor- via M-PESA. The M-KESHO system also includes an tance of physical distance, thus integrang millions emergency micro-credit accessed through M-PESA of Kenyans into the modern economy. and a micro-insurance program. With "branchless Bibliography Mas I and Radcliffe D (2010). Mobile Payments Go Viral: M-PESA in Kenya. Bill and Melinda Gates Foundaon. hp://www.microfinance- gateway.org/p/site/m/template.rc/1.9.43376/. Plyler M, Haas S and Nagarajan G (2010). Community Level Economic Effects of M-PESA in Kenya: Inial Findings, Execuve Summary, (Maryland: Iris Center, University of Maryland College Park). Mbogo M (2010). The Impact of Mobile Payments on the Success and Growth of Micro-Business: The Case of M-Pesa in Kenya, Journal of Language, Technology & Entrepreneurship in Africa 2, no. 1 (2010): 182. Jack W and Suri T (2010). Mobile Money: The Economics of M-PESA, available at hp://www9.georgetown.edu/faculty/wgj/papers/Eco- nomics-of-M-PESA.pdf December 2010 | Edion No. 3 25 ANNEXES Annexes Annex 1: An Update on the KEU Edion 1 (December 2009) In the first edion of the Kenya Economic Update we had a special focus on the food crisis and, in parcular, grain markeng and the implicaons of resistance to reform of the Naonal Cereals Produce Board (NCPB) on food security, and producer consumer welfare in Kenya. Notably, the weak governance at NCPB saw maize prices in Kenya skyrocket to US$ 400 per mt in 2009, double the global maize prices. Following the food crisis in 2009, the government and development partners implemented an input subsidy scheme to boost food supply. The scheme saw 123,000 small scale farmers provided with vouchers for inputs, another 33,000 farmers/traders were able to access inputs through credit supported by the project, and a further 80,000 were sup- ported with inputs. Addionally, the government rehabilitated hitherto dysfunconal irrigaon schemes. A combinaon of these downstream policy intervenons and good rains resulted in a bumper harvest in 2010 and producon increased by 0.2 million mt. Consequently, maize prices declined drascally in the first seven months of 2010, and have remained in line with global prices at about US$ 185 /mt. Food inflaon declined to a two year low, below 4 percent. However, upstream resistance to comprehensive reforms sll remains, and limits the success of downstream supply side intervenons. Afflotoxin contaminaon of the bumper harvest in 2010 from marginal areas is a good example which, unfortunately, due to weak polices in post harvest management and grain markeng resulted in major losses for small scale farmers from marginal producon areas. Some reforms have now taken place at NCPB which guarantee farmers a market determined price for their produce. The NCPB has introduced a Warehouse Receipng System (WRS) where farmers deliver their grain for storage and get a receipt which they can use as collateral in the credit market. Through this mechanism the farmer has the freedom to choose when to sell their produce, and reduce post harvest losses. The WRS demonstrates some progress towards link- ing grain markeng and agriculture financing. As droughts become more frequent, a consequence of climate change, and a bulging populaon increases demand for food, reforms in grain markeng will remain a priority. The agriculture sector can learn from the ICT revoluon: when the incenves are right, the private sector will respond, innovang and providing a plaorm for growth. December 2010 | Edion No. 3 27 Annexes Annex 2: An Update on the KEU Edion 2 (June 2010) The special focus on the port of Mombasa detailed the significance of the port to the economy of Kenya and to East Africa. It detailed a number of operaonal and efficiency problems facing the port and stressed the need for urgent acon to alleviate congeson and to facilitate the faster clearance of import and export goods. Specifically, it recom- mended a number of short and medium term port reform measures to be undertaken on which there has been some progress: In the next 3 to 6 months · Implement the IT Port-Based Community System: A decision was taken by cabinet in September 2010 to implement a Naonal Single Window and to set up an independent body to implement it. This development will in me see the import and export process transformed into a paperless operaon enhancing efficiency and transparency. · Decide on a route for the Mombasa by-pass and start implementaon together with the link road from the port: This decision is expected shortly. · Appoint the managing director of the Kenya Ports Authority (KPA) on a 3-year performance contract: Mr. Gichiri Ndua has been appointed as the new Managing Director of KPA. Mr. Ndua is the current President of the Internaonal Associaon of Ports and Harbours (the first African to hold that post). · Approve the concessioning of berths 11-14: The concessioning process is sll awaing approval by cabinet. It is impor- tant that this process moves ahead without delay to prevent potenally serious capacity problems which are likely to impact negavely on the economy over the next 12 to 18 months. At the same me new regulaons being developed by Kenya Marime Authority on foreign ownership of port operaon need to be resolved to enable a successful and compeve bid process. · Idenfy a coordinang body with a mandate for reform: The cross-ministerial commiee on Mombasa headed by the Ministry of Transport is currently not meeng. · Set up a system of incenves to enable full 24 hour port operaons: There is evidence that some agencies at the port commied to work 24 hours are not doing so, slowing down the clearance of goods and frustrang cargo owners. In the next 12 months · Clarify the metable for a full landlord port status as well as the role of KPA: A metable has yet to be established. · Clarify roles and responsibilies of public and private port stakeholders: Yet to be clarified. · Undertake reforms of customs collecons: Despite implementaon of the Naonal Single Window which will begin shortly, cargo owners sll face on-going problems with Kenya Revenue Authories SIMBA system which connues to experience technical problems. · Strengthen port operaonal capacity at the Ministry of Transport: The Ministry of Transport has yet to strengthen its capacity for port operaons. December 2010 | Edion No. 3 28 Annexes Annex 3: Global Commodies Prices Trends 1200.0 Global commodity prices 1000.0 800.0 600.0 DAP $/MT 400.0 Coffee, C/kg Tea c/Kg 200.0 Crude oil $/bbl 0.0 2008 2009 2010 450.0 Global commodity prices 400.0 Coffee, C/kg 350.0 300.0 250.0 Tea c/Kg 200.0 150.0 100.0 Crude oil $/bbl 50.0 0.0 2008 2009 2010 Source: World Bank staff esmates December 2010 | Edion No. 3 29 Annexes Annex 4: Key Assumpons to the Growth Forecast GLOBAL ASSUMPTIONS The internaonal outlook is broadly posive and will boost the demand of Kenya exports notably horculture and tour- ism. The global performance will also boost the performance of migrant remiances to boost domesc consumpon. Aer the sharp decline in growth in 2008 and the contracon in 2009, global GDP growth re-entered into posive terri- tory in 2010, growing by an esmated 3.5 percentage points. The rebound is expected to connue in 2011 and 2012--al- beit with a slight deceleraon in 2011--at respecvely 3.2 and 3.6 percentage points. As before and during the "Great Recession", growth will be led by the low and middle income countries, which posted a growth of 6.6 in 2010, and are esmated to grow by 5.9 and 6.1 in the next two years. This is more than twice the respecve figures for high income countries, at respecvely, 2.5, 2.3 and 2.8 for 2010-2012. There is a considerable degree of heterogeneity in the distribu- on of gains, as some regions namely, East and Central Asia were among the worst affected by the crisis and are among the slowest in the recovery process. Nevertheless, growth is expected to decelerate during the forecast horizon in the developing economies, reflecng the expected slowdowns in the East Asia and the Pacific, and the Lan America and the Caribbean regions, which, in turn, will be driven by the reducon of the speed of expansion in the two largest economies in those regions, namely, China and Brazil. KENYA SPECIFIC ASSUMPTIONS Domesc Demand In the low case scenario the dry weather could be more severe and prolonged than ancipated, pushing inflaon in the range of 10 percent. Government response to such a weather-related crisis would crowd out other essenal develop- ment spending. External Sector If a food crisis occurs, food imports will increase the import bill widening the current account deficit. But if the EAC common market remains robust, Kenya's exports to the region could increase, narrowing the structural current account deficit. Polical Developments The low case scenario assumes the implementaon of the constuon could be slow, with or no clear road map and with the execuve signaling counter reform signals. However, fast tracking of legislaon and implementaon of essenal reforms could send the right signals to the rest of the economy with a much higher growth rate. Source: World Bank staff December 2010 | Edion No. 3 30 Annexes Annex 5: Key Factors Determining the Compeveness in the ICT Sector ­ Comparison of Kenya with India There is a broad agreement that several key factors determine compeveness in IT/BPO: (i) availability of employable skills (including IT skills), (ii) compeve costs, (iii) quality of infrastructure relevant to the IT/BPO industry, and (iv) an overall envi- ronment that is conducive to business. Of all these factors, countries can substanally increase their internaonal compeve advantage if they execute smart strategies to increase their skills offering for the industry. Given these developments, the lack of skills is now the most important binding constraint to the growth of the IT/BPO sector in Kenya. The country currently produces around 30,000 university graduates and about 250,000 graduates from high school annually. However, very few of these graduates, whether at school or university level, are immediately suitable for employment in the IT/BPO industry. According to the recent McKinsey Report (2008), the talent pool for the BPO sector in Kenya currently is very limited. Only about 5,000 graduates are suitable for employment in the industry. The report has projected the skills required for BPO sector to be 70 percent for voice and data operators, 5 percent for managers, 10 percent for engineers and 15 percent for technicians. Locaon Readiness Index: Comparing Kenya and India Talent pool (data) 10.0 Maturity 8.0 Talent pool (IT) 6.0 4.0 2.0 Risk - Cost (data) Environment Cost (IT) Infrastructure Kenya India Source: World Bank 2008, Mckinsey & Company December 2010 | Edion No. 3 31 Annexes Annex 6: Other Sector Innovaons on the Mobile Plaorm In addion to mobile money, services that rely on 4-digit SMS short codes and GSM USSD (simple, menu-driven interfaces that allow users to work through mulple opons along a decision tree on basic mobile phones) are now ubiquitous in Kenya. These mobile interfaces, combined with open source technology plaorms such as RapidSMS and Ushahidi that facilitate bulk informaon distribuon and crowdsourcing, provide simple and effecve tools for innovaon across a variety of products and services. Some of the recent innovaons in other sectors Agriculture Price discovery and disseminaon in commodity markets (both local- ized and naonal) to allow farmers and traders to make beer deci- sions; delivery of extension services (DrumNET, SMS Sokoni, Livestock Informaon Network and Knowledge System (LINKS), RATIN SMS, Na- onal Agriculture Informaon Service (NAFIS) Public Informaon Mobile phone technology is being leveraged as a low-cost and reliable channel for providing public awareness and disseminang structured informaon in campaigns ranging from educaon to public health; par- cularly important in post-crisis situaons and for facilitang access to emergency response services (school /university exam results via SMS, Games4Life, Health Tips, My Queson, VCT Online) Health Disease tracking, demographic informaon collecon, knowledge dis- seminaon and consultaon services to allow communies isolated from healthcare infrastructure to diagnose and treat diseases, which is crical in emergency and natural disaster situaons; administraon and monitoring of healthcare services; behavior change ­ encouraging HIV tesng, maintenance of ARV regimen (ChildCount+, WelTel, EpiSur- veyor, BloodbankSMS) Civil Society and Governance Civil society organizaons have effecvely used mobile technology to monitor social unrest and human rights violaons, mobilize voters and disseminate elecon results, and track the management of local budgets; the Kenyan government is implemenng a comprehensive e-government strategy to improve accessibility and transparency of government services and reduce scope for gra (Uwiano Plaorm for Peace/PeaceNet, www.e-government.go.ke, BungeSMS, Budget Track- ing Tool, Martus) Search and Social Media Localized informaon search through Google SMS Kenya, specialized trading markets (Mosoko, SMSoko), job informaon (Kazi560) Source: J. Hellström (2010). "The Innovave Use of Mobile Applicaons in East Africa." (SIDA) December 2010 | Edion No. 3 32 Annexes Annex 7: Data Sources and Projecons for Mobile Phone and Mobile Money Stascs The analysis of mobile money presented here draws on mobile subscripons and some SIM cards may only be several different data sources: used to receive calls and thus not be counted in the of- ficial figures. · The 2006 and 2009 Access to Finance Surveys in Kenya. The 2006 survey was conducted August - September Overall, extrapolang linearly from recent growth rates 2006. The 2009 survey was conducted January - March (see figure 18), we esmate that in Kenyan in Decem- 2009. These are naonally representave surveys of ber 2010 there are 22 million acve mobile phone sub- adults. scripons, approximately 9 for every 10 Kenyans adults · The 2009 AudienceScapes survey, which was conduct- (age 18+). Similarly, we esmate that at the end of 2010, ed July - August 2009. This is a naonally representa- roughly 3 out of 4 Kenyan households own a mobile ve survey of adults. phone, and 9 out of 10 have access to a cell phone--in- · Data on M-PESA agent deposits and withdrawals pro- cluding those using a phone outside their household. vided by Safaricom. Data on mobile money also comes from both surveys · Data on Zap transfers provided by Zain. and official data from providers. Survey data provide es- · Stascs on mobile phone usage from the Communi- mates of the percentage of adults who have ever used caons Commission of Kenya. mobile money: 38% (early 2009 FinAccess), 55% (mid- · Populaon and demographic figures from the World 2009 AudienceScapes), and 69% (late 2009 Jack and Suri Development Indicators. survey). Separate data is provided by mobile providers · Addional data taken from the websites of Safaricom, on the number of current mobile money customers. Yu, and Equity Bank. As with the acve mobile phone subscripons, mobile money customers refers to SIMS that are acve on ex- Data on both mobile phones and mobile money comes isng networks and also registered with a provider's from two broad types of sources: surveys (and the cen- mobile money service. This figure exceeded 13 million sus) and data from mobile providers. The prevalence of in August 2009. Based on a linear extrapolaon from mobile phones can be measured in different ways. For recent growth rates, we project that that in December recent years, various surveys and the census provide es- 2009 this figure has reached 15 millionmore than 2 mates of the simplest figure, the percentage of Kenyan out of every 3 Kenyans over age 18. households that own a mobile phone: 13% (2003 Demo- graphic and Health Surveys), 26.9% (2006 FinAccess), The profiles of mobile money users by age, gender, and 60% (early 2009 FinAccess), 61% (2008/09 DHS), 63.2% locaon in this report are based primarily on the mid- (mid-2009 census), and 71% (mid-2009 AudienceScapes 2009 AudienceScape data. The profiles are shown for survey). Notably, these figures understate the number the point in me of the survey as well as a projecon for of people who have access to a cell phone. In the early December 2009. For most of the projecons presented 2009 FinAccess survey, 20 percent of adults surveyed in this report, the projecons assume 25 percent growth did not have a phone at home but used a phone outside across all categories. However, for the projecons by their own households. quinle (figure 25), the overall growth of 25 percent is al- located among the five quinles proporonal to growth A separate data series is the number of mobile phone rates esmated for each quinle based on two earlier subscripons, usually reported relave to the number of surveys. An upper limit of 95 percent is also assumed adults (age 15+ or age 18+). These figures are assem- for the top quinle. The total size of the populaon cor- bled month-by-month, collected by the Communica- responding to the projected populaon is approximately ons Commission of Kenya from mobile operators, and equal to our esmate number of current mobile money compiled for many countries of the world by the Inter- customers in December 2009. Thus, although the origi- naonal Telecommunicaons Union. In Kenya, these nal AudienceScapes data is for those who had ever used figures generally reflect the number of SIM cards that mobile money at the me of the survey, the projecons have been acve and generated revenue during the pre- can be taken as an approximate profile of the extent of vious 90 days. These figures translate only roughly into mobile money customers in December 2010. cell phone access because some Kenyans have mulple December 2010 | Edion No. 3 33 Annexes Annex 8: The Genesis of Mobile Money in Kenya The story of how mobile money was launched in Kenya The system was given the name M-PESA, with "m" for is told by two people involved in its development, Nick mobile and "pesa" meaning money in Swahili. It was Hughes and Susie Lonie. What is perhaps most notable launched on a pilot basis in October, 2005, with just about the creaon of the first mobile money system is 500 customers, all of whom were microfinance clients. that it was not the product of a top down plan. No one The system faced a number of challenges inially, and ever set out to create mobile money. Rather, it emerged modificaons were planned to simplify the service. A out of a process of private iniave which evolved over month or two into the pilot, Safaricom allowed custom- me in response to consumer demand, aer an inial ers to buy prepaid airme with their M-PESA credit. nudge from a donor innovaon fund. Then Safaricom discovered that M-PESA transacons were taking place among their microfinance clients for In 2000, the UK government established the Financial a wide variety of purposes not imagined by its crea- Deepening Challenge Fund (FDCF). The fund made tors: available £15 million for joint investments with the · People repaying the loans of others in return for serv- private sector to improve access to financial services. ices. In mid-2003, Nick Hughes, a Vodafone execuve de- · Payment for trading between businesses. veloped a proposal and built up support within the · Some of the larger businesses using M-PESA as an company. Hughes notes that the "entrant of a telecom overnight safe because the banks closed before the company into a funding compeon for the financial agent shops. services sector took a few of the FDCF proposal team · People journeying between the pilot areas, deposit- by surprise," but nonetheless the company was award- ing cash at one end, and ed funding of nearly £1 million, which was matched by · withdrawing it a few hours later at the other. Vodafone. · People sending airme purchased by M-PESA direct- ly to their relaons up country as a kind of informal Even at the point the funds were awarded, the concept remiance. of mobile money sll had not taken shape. Hughes · People outside the pilot populaon being sent mon- explains that the proposal consisted largely of "a pre- ey for various ad hoc reasons; for example, one lady's liminary needs assessment and not a funconal speci- husband had been robbed, so she sent him M-PESA ficaon of a new product." Vodafone then conducted a to pay for his bus fare home. series of open workshops in Nairobi and Dar es Salaam · People repaying loans in return for cash on behalf of to consult with banks, microfinance organizaons, a few colleagues who had not mastered the use of technology service suppliers, nongovernmental or- the phone--or simply sold it. ganizaons involved in micro-credit, and telecom and financial sector regulators to think about ways in which The innovave uses customers made of M-PESA during they could increase access to financial services. Aer the pilot, which ran through October, 2006, convinced much brainstorming, the idea emerged of a partner- Vodafone and Safaricom that there was a market for ship between Safaricom (a member of the Vodafone M-PESA far beyond its narrow use for microfinance group and a Kenya telecommunicaons company), a transacons. Given the experience of the pilot, Vo- Kenyan micro-finance instuon called Faulu, and a dafone execuves recognized in parcular that a mo- Kenyan commercial bank. Specifically, they decided to bile money system would have widespread aracon develop a plaorm to allow customers to receive and for sending remiances, a common phenomenon in re-pay micro-finance loans using a phone handset, with Kenya. As a result, with a number of modificaons, M- payments made via Safaricom's airme resellers. The PESA was launched naonally in March 2007, and had service was conceived of enrely as a way to improve 20,000 customers signed up within the first month of the efficiency of Faulu and extend its reach to remote operaon. It was followed by Zap in January 2009, yu- locaons. Cash in December 2009 and Orange Money in Novem- ber 2010. The source material for this Annex came from an arcle in Innovaons (winter/spring 2007) by Nick Hughes and Susie Lonie. Hughes is the Vodafone execuve who championed the mobile money concept and obtained the inial seed capital in 2003. Lonie is the e-commerce expert that went to Kenya in 2005 to develop the pilot operaon. December 2010 | Edion No. 3 34 Annexes Annex 9: Background Stascs Key Macroeconomic Indicators 2007 2008 2009 2010 2011 Macro aggregates Real GDP 7.1 1.7 2.6 5.0 5.3 GDP Per Capita 4.4 -1.0 0.0 2.3 2.6 Private Consumpon 7.3 3.0 3.8 5.1 3.9 Government Consumpon 7.2 3.7 5.5 4.1 3.4 Gross Fixed Investment 13.3 8.9 0.6 13.8 12.6 Exports, GNFS 6.0 3.6 -7.0 12.0 8.9 Imports, GNFS 12.7 5.3 -0.2 14.5 8.6 Prices Inflaon 5.7 15.4 11.0 5.9 6.0 GDP Deflator 5.1 13.5 9.2 5.0 5.0 Exchange rate (Kshs/US$) 67.3 69.2 77.4 76.6 78.1 Revenue and Grants 22.0 21.6 22.6 23.1 23.2 Expenditure 23.3 28.3 30.3 29.2 28.1 Budget Deficit -5.9 -6.3 -6.6 -6.0 -5.2 Gross Domesc Investments 19.1 19.4 19.4 19.3 19.5 Government Consumpon 14.5 14.6 15.7 14.9 14.6 Private Consumpon 77.3 77.5 78.5 79.1 79.3 Exports 26.0 26.3 25.2 25.8 26.2 Imports 36.1 40.1 39.4 42.1 43.4 Current account balance -3.8 -6.6 -5.5 -5.8 -6.0 General government balance -3.0 -4.9 -5.6 -6.0 -5.6 Memo Nominal GDP (US$) 27124.4 30354.9 30944.9 33523.1 35940.8 Source: KNBS and World Bank staff esmates Sub-Saharan Africa Forecast Est. Forecast (annual percent change unless indicated otherwise) 2007 2008 2009 2010 2011 GDP at market prices (2005 USD) 6.5 5.0 1.6 4.4 5.1 GDP per capita (units in USD) 4.0 3.0 -0.3 2.5 3.0 PPP GDP -4.4 -10.1 3.6 4.2 3.0 Private consumpon 7.6 3.3 1.1 3.7 4.5 Public consumpon 5.7 7.4 5.3 5.2 5.0 Fixed investment 16.7 12.0 5.0 6.8 7.6 Exports, GNFS 3.7 4.7 -6.9 9.3 6.5 Imports, GNFS 11.1 6.5 -4.2 9.8 6.5 Net exports, contribuon to growth -2.6 -0.8 -0.8 -0.6 -0.6 Current account bal/GDP (%) -0.4 0.4 -2.5 -1.5 -1.9 GDP deflator (median, LCU) 7.0 9.2 7.7 4.9 5.1 Fiscal balance/GDP (%) 0.5 0.7 -5.6 -4.8 -3.3 Memo items: GDP SSA excluding South Africa 7.1 5.8 3.7 5.3 6.0 Oil exporters 8.0 6.2 4.1 5.7 5.8 CFA countries 4.4 4.1 2.6 3.7 4.1 South Africa 5.5 3.7 -1.8 3.1 3.4 Nigeria 6.4 5.3 5.6 6.1 5.7 Kenya 7.0 1.7 2.6 4.9 5.2 Source: World Bank staff esmates December 2010 | Edion No. 3 35 Annexes Sub-Saharan Africa Forecast (percentage change from previous year, except interest rates and oil price) 2008 2009 2010 2011 Global Condions World Trade Volume (GNFS) 3.2 -11.6 6.2 6.7 Consumer Prices G-7 Countries 3.1 -0.2 1.5 1.6 United States 3.8 -0.3 2 2.2 Commodity Prices (USD terms) Non-oil commodies 0.0 -21.6 16.8 -4.0 Oil Price (US$ per barrel) 97 61.8 78.1 74.6 Oil price (percent change) 36.4 -36.3 26.4 -4.5 Manufactures unit export value 5.9 -4.9 0.0 -3.7 Interest Rates $, 6-month (percent) 3.2 1.2 0.8 2.2 , 6-month (percent) 4.8 1.5 1.0 1.5 Real GDP growth World 1.7 -2.1 35 3.2 Memo item: World (PPP weights) 1.3 -0.4 4.2 4.0 High income 0.4 -3.3 2.5 2.3 OECD Countries 0.3 -3.4 2.2 2.3 Euro Area 0.4 -4.1 0.7 1.3 Japan -1.2 -5.2 2.5 2.1 United States 0.4 -2.4 3.3 2.9 Non-OECD countries 3.0 -1.7 4.2 4.2 Developing countries 5.7 1.7 6.6 5.9 East Asia and Pacific 8.5 7.1 8.7 7.8 China 9.6 8.7 9.5 8.5 Indonesia 6.0 4.5 5.9 6.2 Thailand 2.5 -2.3 6.2 4.0 Europe and Central Asia 4.2 -5.3 4.1 4.2 Russia 5.6 -7.9 4.5 4.8 Turkey 0.7 -4.7 6.3 4.2 Poland 4.8 1.7 3.0 3.7 Lan America and Caribbean 4.1 -2.3 4.5 4.1 Brazil 5.1 -0.2 6.4 4.5 Mexico 1.8 -6.5 4.3 4.0 Argenna 7.0 -1.2 4.8 3.4 Middle East and N. Africa 4.2 3.2 4.0 4.3 Egypt 7.2 4.7 5.0 5.5 Iran 2.3 1.8 3.0 3.2 Algeria 2.4 2.1 4.6 4.1 South Asia 4.9 7.1 7.5 8.0 India 5.1 7.7 8.2 8.7 Pakistan 2.0 3.7 3.0 4.0 Bangladesh 6.2 5.7 5.5 5.8 Sub-Saharan Africa 5.0 1.6 4.5 5.1 South Africa 3.7 -1.8 3.1 3.4 Nigeria 5.3 5.6 6.1 5.7 Kenya 1.7 2.6 4.9 5.3 Memorandum items Developing countries excluding transion countries 5.7 3.0 6.6 6.2 excluding China and India 4.3 -1.8 4.5 4.4 Source: World Bank staff esmates December 2010 | Edion No. 3 36 Annexes New Wage Employment Numbers Sectors 2000 2001 2002 2003 2004 2005 2006 2007 2008 Agriculture and forestry (270) 1,584 1,000 2,526 4,533 6,531 7,427 5,264 862 Mining and quarrying 55 42 (62) 160 156 221 291 222 283 Manufacturing (1,708) (1,310) 13,260 9,909 3,868 9,204 6,435 5,550 (717) Electricity and water (389) (926) (96) (208) (203) (602) (742) (586) 327 Construcon (612) (1,310) (217) 131 710 833 1,705 1,410 3,498 Wholesale and retail trade 1,745 1,525 626 5,278 5,208 7,397 10,484 9,935 6,554 Transport and communicaon (410) 889 1,201 1,313 11,820 15,351 17,128 22,819 3,739 o.w Communicaon 2,126 2,530 1,656 2,453 7,685 11,194 12,662 17,047 5,074 Financial intermediaon n real estate 123 (885) (599) 554 1,509 3,702 3,393 989 2,747 Other services 4,671 739 7,243 8,134 9,876 4,200 (126) 6,584 16,411 Total 3,205 348 22,356 27,797 37,477 46,837 45,995 52,187 33,704 Source: Kenya Naonal Bureau of Stascs Average price of voice calls (Kshs per minute) 2004/05 2005/06 2006/07 2007/08 2008/09 2009/10 Post -paid On-net 24 22.12 14.5 10.98 7.05 6.6 Off-net 27 26.53 23.75 17.68 12 8.22 Charges to fixed network 27 25.65 23.75 17 11.43 8.22 Pre-paid On-net 20.67 16.2 10 8.58 5.34 5.62 Off-net 27.33 26.53 19.5 13.28 12.19 12.31 Charges to fixed network 26.78 26.53 19.5 12.51 11.34 12.31 Average On-net 17.5 19.16 12.25 9.78 6.195 4.78 Off-net 24 26.53 21.625 15.48 12.095 10.265 Average Charges Mobile Network 20.75 22.845 16.9375 12.63 9.145 7.5225 Charges to fixed network 24 26.09 21.625 14.755 11.385 10.265 Source: Communicaon Commission of Kenya Mobile cellular subscripons per 100 inhabitants 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Cameroon 0.7 2.6 4.2 6.3 8.8 12.6 17.2 24.3 32.3 37.9 Ethiopia 0.0 0.0 0.1 0.1 0.2 0.5 1.1 1.5 2.4 4.9 Ghana 0.7 1.2 1.9 3.8 7.9 13.1 23.3 33.2 49.6 63.4 Kenya 0.4 1.9 3.6 4.7 7.3 12.9 20.0 30.1 42.1 48.7 Nigeria 0.0 0.2 1.2 2.3 6.7 13.2 22.4 27.3 41.7 47.2 Rwanda 0.5 0.8 1.0 1.5 1.6 2.5 3.4 6.7 13.6 24.3 Senegal 2.5 3.0 5.3 7.3 10.2 15.3 25.8 30.5 44.1 55.1 South Africa 18.6 23.7 29.7 36.0 43.9 70.6 81.5 86.0 90.6 92.7 Tanzania 0.3 0.8 1.7 5.3 5.1 8.7 14.4 20.2 30.6 39.9 Uganda 0.5 1.1 1.5 2.9 4.2 4.6 6.8 13.7 27.0 28.7 Zambia 0.9 1.1 1.3 2.1 4.0 8.1 13.8 21.4 28.0 34.1 Source: Internaonal Telecommunicaon Union December 2010 | Edion No. 3 37 Kenya at the Tipping Point? with a special focus on the ICT Revoluon and Mobile Money Kenya may again be at a "pping point", the theme of the third Kenya Economic Update which has a special focus on the transformave impact of Informaon and Communicaons Technology (ICT) and mobile money. Over the last decade, ICT has outperformed all others sectors growing at an average of 20 percent per year. The benefits of ICT are starng to be felt in other sectors, and have contributed to the condions for Kenya to reach this pping point. Kenya has entered the new decade with renewed and stronger than expected growth. The passing of the new constu- on, connued strong macroeconomic policies, and a favorable regional environment have created a new posive economic momentum. Kenya may again be posioned to experience high growth. Over the last three decades Kenya has experienced only two short episodes when economic growth exceeded five percent and was sustained for at least three consecuve years: 1986-88 and 2004-2007. Is Kenya again at the verge of experiencing another growth spurt? Will it last longer and go deeper than the previ- ous two episodes? The photograph shown on the cover (middle) is from Robert Waiharo and that on page 24 is from Valeria Straziota. All the other photographs displayed by the report are from Safaricom's collecon. Design by Robert Waiharo www.worldbank.org/kenya/ke Hill Park, Upper Hill Road P. O. Box 30577-00100 Nairobi, Kenya Telephone: (020) 3226000 Fax: (020) 3226382