Joep Roest is a Financial Inclusion Specialist with the Pacific Financial Inclusion Programme (PFIP). PFIP is a Pacific-wide programme helping to provide sustainable financial services to low income households. It is a joint project of the UN Capital Development Fund (UNCDF) and the United Nations Development Programme (UNDP) and has received additional funding support from the Australian Agency for International Development (AusAID) and the European Union. The programme is based at the UNDP Pacific Centre in Suva Fiji.

Papua New Guinea (PNG) is a country so complex it defies easy description. A place of such diversity it hosts 850 distinct languages for a population of about 7 million. The population figure, mind you, is only a guess as nobody really knows. The landscape is so rugged that the capital, Port Moresby has no road link to any other city. To get anywhere in PNG, you walk, fly or take a boat. As such, it was only in the 1930’s that the first Western gold miners started to penetrate the interior and “discovered” the densely populated highlands where the bulk of the population lives. The miners were a sign of things to come as PNG rides the crest of a gargantuan resource boom. Over the last few years, PNG has become notorious for its high levels of crime which is a major preoccupation of all who live and work there.

State of play in PNG

Against the backdrop of PNG’s security issues and difficult geography, a means to save, store and send money is desperately needed. Mobile money looks to be the answer and typically for PNG, things have developed in their own unique way. In the last half year three providers have launched very distinct offerings, all on the same Telepin platform. Interestingly, it is not a line-up of the usual suspects as both the postal service and an MFI have entered the fray. PNG now plays host to Post’s Mobile SMK, Nationwide Microbank’s MiCash and Digicel’s (of Haiti fame) Cellmoni, with more rumoured to be waiting in the wings.

Both Post and Digicel have opted for a virtual wallet product while Nationwide has developed a real-time linkage into their core banking system. To make matters even more interesting, both Post PNG and Nationwide Microbank have partnered with Oceanic Communications Limited (OCL) to manage their agent network. Confused yet? It gets better; OCL does most of Digicel’s airtime distribution yet does not provide agency services for Cellmoni. That is, at least, where it all stands now.

Airtime distributor

As we all know, mobile money is hard, especially in places like PNG. A successful mobile money operation has to be excellent at everything, all the time. Building and maintaining an excellent agent network may be hardest part of all. It takes tremendous investment in time, resources and energy for it to work. Unfortunately there is no quick fix or technological silver bullet that ensures success. It is a long, inglorious slog. Who wouldn’t want to farm it out?

That is where OCL comes in. As an airtime distributor they seem ideally placed to play a role. They make money by getting airtime (electronic and scratch) out to the furthest reaches of this challenging country and take cash in return. Airtime distributors are operationally minded businesses where solid processes are the cornerstone of profitability. In OCL’s case, they have established relationships with 12,000 resellers. From their interactions with these resellers, they have years of data that can help predict liquidity needs and identify resellers who are prime candidates for becoming successful agents. Admittedly, airtime distribution is a far cry from mobile money agent management, yet many of the same capabilities come into play.

The opportunity

OCL is now doing much of the hard stuff on behalf of its two partners. They manage agent recruitment and training, agent monitoring and liquidity management. This frees up the mobile money operators to concentrate on their offerings. There is also the compelling possibility that OCL could drive agent interoperability and standardization of the agent experience across partners. It would make their management of the network simpler, drive down their costs and make it easier to recruit agents. This seems especially likely, as all the current products are on the same platform. Both agents and customers stand to benefit. Agents won’t have to maintain separate balances. Customers will benefit from a broader agent network, unified customer experience and an even playing field for all competitors, ensuring competitive offerings.

Will we see OCL shape mobile money in PNG due to their central role? More generally, will third-party providers become a force for standardization and interoperability in other markets?

So far…

It seems to be working. OCL’s two partners have been able to quickly expand far beyond their brick & mortar footprint. Nine months in, customer growth is strong and accelerating. If developments so far are anything to go by, there will be a lot more to write about over the coming months.

- Joep Roest -

This is a guest blog by Ignacio Mas, an independent consultant who has written extensively on mobile money.  He is former Senior Advisor with the Financial Services for the Poor team at the Bill & Melinda Gates Foundation and at the Technology and Business Model Innovation Program at CGAP. 

Mobile money feels right for mobile network operators (MNOs): it is an extension of the basic prepaid platform and distribution networks they already operate. Mobile money does require greater surveillance against fraud and money laundering measures, but it’s all fundamentally about secure messaging.

From a telecoms regulation point of view, mobile money is another instance of a value added service and those tend to receive a light regulatory treatment. All the specific regulations that pertain to the safety and soundness of mobile money –who can issue accounts, conditions of service, data security and privacy standards, supervisory treatment, consumer protections, etc.— should be the domain of the banking regulator.

But MNO participation in retail payments presents competitive challenges which banking and telecoms operators will need to monitor closely and perhaps address jointly. The problem is that MNOs are both component suppliers and direct competitors to banks wanting to offer mobile financial services. There is a risk that MNOs transfer market power from their core market to the emerging retail mobile payments market, in such a way as to effectively shut banks out of mobile payments. In the future, most financial services can be expected to have a mobile component, so such a situation would have severe implications for competition in the financial inclusion space more broadly.

This is quite understandably spooking many banking regulators into preventing MNOs from playing directly in the payments/banking space. That’s unfortunate, the fact is that we need to enlist MNOs to push the frontiers of financial access, while guarding against any potential abuses of market power on their side. We need to allow MNOs to contest the market without dominating it.

Authorities need to identify specifically those components of mobile communication services over which MNOs have bottleneck control and which are essential for the provision of mobile financial services. One asset MNOs control is the SIM card – a smartcard which identifies every mobile user. Access to the SIM card provides benefits in terms of security, since SIM cards may contain pre-loaded security keys which can implement end-to-end data encryption from the mobile handset all the way to the transaction authorization server. Access to the SIM card might also enhance the usability of services, since the SIM card controls the on-the-phone menu onto which mobile money can be incorporated directly.

However, there is no precedent worldwide for establishing equal access rules to SIM cards, as unbundling the SIM card might have severe implications for the security of mobile networks. In any case, the problem of proprietary control over the SIM card is mitigated if banks can build an equivalent service through other means.

MNOs also control the phone’s communications (or bearer) channel. The more common channels, such as voice, SMS (the protocol underlying text messaging) and packet-data (under various flavors such as GPRS, EDGDE, 3G or HSDPA), are broadly made available by MNOs under standard commercial offers. In this case, it should be fairly easy to establish and monitor a requirement of non-discriminatory access by banks and third-party providers of mobile money services to these channels. Since the market for these bearer services is sufficiently large and lucrative for the MNOs, we can expect MNOs not to over-price their voice and SMS services or to degrade the quality of their service specifically to lock out banks from using them to construct competing mobile money services.

The challenge is with a less common bearer service called USSD (Unstructured Supplementary Service Data), which is not widely commercialized by most operators. (You may recognize it as the service you use when you are asked to dial a sequence of numbers starting with star and ending with hash after you buy a scratch card.)  The session-based nature of USSD presents two strong advantages over SMS as a channel for mobile financial services: it lends itself to implementing network-based menus which makes it easier to use, and it entails no storage of messages anywhere which makes it more secure. It may also be more feasible than using voice in countries where voice tariffs are still expensive, or packet-data in countries where most people still use simple phones. For banks without access to the SIM card, USSD may be the only realistic option.

In this case, simply stating non-discriminatory access to the USSD service may not suffice as long as USSD is used primarily for banking services, since MNOs may price it specifically to preclude competition from emerging in mobile financial services. Therefore, the telecoms regulator ought to monitor USSD pricing to see if it bears a reasonable relationship with the price of alternative access channels such as voice and SMS. Moreover, an operator might be forced to offer USSD to other mobile financial service providers provided that: (i) the operator has an installed USSD capability (even if it is only used for the MNO’s own purposes, such as airtime top-ups), and (ii) the operator is offering its own mobile financial service through a proprietary channel (e.g. SIM-based) not available to others.

These actions –monitoring MNOs for potential discriminatory pricing and service quality on voice and SMS, and monitoring USSD offers to prevent undue denial of service or pricing that does not bear sufficient relationship with the pricing of other channels— fall in the first instance in the domain of the telecoms regulator, since these are purely about the channel and not higher-level financial services. Of course, the banking regulator could play a vigorous role in helping banks present their case to the telecoms regulator if they feel discriminated against, or to competition authorities if the situation warrants escalation or the telecoms regulator lacks the necessary powers to intervene.

Authorities ought to pass a clear and strong message that MNOs can play at the financial services layer if they play fairly at the communications layer. This might be expressed around a protocol or memorandum of understanding signed jointly by the banking and telecoms authorities, stating how the various authorities will work together and laying out the competition standards to which they will hold MNOs.

This regulatory vigilance will be essential until such time when everyone has a smartphone and the phone (rather than the SIM card) can perform locally all security and menu presentation services. At that point, banks will have credible options to build their own mobile financial services with minimal control by the MNOs. Banking will just be an app on your phone, which anyone can download. That competitive nirvana is now more imaginable than ever, but it is still a ways off in developing countries.

- Ignacio Mas

This is the fifth and final post in our series on remittances and branchless banking.  You can read the first four posts here.  So far, we have highlighted the emerging success factors and challenges featured in our 2012 landscaping exercise.  Paolo Baltao from Globe’s GCASH shared with us the lessons learned over a period of eight years during which time he led one of the first services to link remittances to a mobile wallet.   Subsequently, Stefan Staschen explored the untapped opportunities in leveraging the large flows of remittances between Russia and Tajikistan by linking them to other financial products.  This week, we revisit the study conducted by CGAP and Dalberg Global Development Advisors, by focusing on the role of bigger players such as Western Union and BICS Homesend in consolidating existing corridors and accelerating further adoption. 

Since we started this series, new remittance data from several countries has been released by the World Bank. This newly available dataset reveals that officially recorded remittance flows to developing countries reached $372 billion in 2011, an increase of 12.1 percent over 2010. This is higher than our earlier estimate of $351 billion. While there is clearly some market potential there, so far we have seen that uptake for mobile enabled remittances services has been anaemic to say the least.

Source: CGAP & Dalberg Global Development Advisors

As our understanding of the factors that lead to customer adoption of branchless banking expands, there is a growing consensus that for international remittances services to reach a significant level of scale, they will require an existing mobile money ecosystem that allows for downstream transactions which give users access to a wider array of cost-effective services and products such as payments and access to savings.  This will provide not only value-added for consumers but also the much needed transaction revenue for providers.  It is evident that these recalibrated strategies will no longer place remittances as the core driver for adoption, but factor them in as one of the many financial services which can be provided to a customer once a branchless banking ecosystem has reached a certain level of maturity and depth.

One thing is certain: although some new innovative models have emerged, traditional remittance providers or MTOs like Western Union still have a huge advantage through the benefits they offer to partners. These include fast access to a broad range of sending countries as well as significant brand recognition and regulatory compliance, though often at the expense of their partner’s pricing power (an expense which could end up being passed on to consumers, ultimately reducing the demand for these types of services).   Nevertheless, Western Union has already launched mobile money transfer services in nine countries: Bangladesh, Burkina Faso, Canada, Kenya, Madagascar, Malaysia, the Philippines, Tanzania and the U.S, and in the last couple of months, has announced strategic alliances with MTN Uganda, Roshan in Afghanistan and Tigo in Paraguay.  This move will allow senders to remit funds directly into the recipient’s mobile wallets from any of Western Union’s agent locations around the world.  It is evident that MNOs remain optimistic about deploying international remittances through mobile money, yet they are increasingly aware that the full benefits will only be realized in the long-term.  As mobile money ecosystems become more mature in these markets, these flows could play a pivotal role in consolidating corridors and accelerating adoption.

Alternately, hub providers such as BICS Homesend are now making it possible to integrate mobile wallets or money transfer systems of two different providers. Besides facilitating services that are mobile centric from sender to receiver, HomeSend can provide access to other service providers, such as MTOs and banks which can offer technical solutions to streamline domestic interoperability between systems and competing regulatory frameworks. Given their structure, Homesend is a cheaper alternative to giants such as Western Union and can provide a more flexible partnership without having to cede too much control from the side of the operator or bank.  However, they don’t have the same instant global reach and given their lack of direct interface with mobile wallet users, they are completely reliant on their partners to market and push transactions which make the issue of consumer education even more critical.

Nevertheless, operators seem to remain optimistic over the long-term opportunity to deploy international remittances through mobile money, and hope it will eventually contribute to the economic viability of their deployments through the added revenue opportunities for them and their agent networks.  It remains to be seen how these remittance flows are actually tied to specific financial products and services.  Only then will the real impact of mobile-enabled remittances on the unbanked be uncovered.

- Camilo Tellez

Pablo García Arabéhéty is an independent consultant who focuses on business model innovation in the mobile and environmental industries.  He has previously worked at the Organization of American States and the Innovation Lab at the Inter-American Development Bank.

Last December, Starbucks announced that during 2011 it processed 26 million transactions in the US through its mobile payment application. While this news was anecdotal for traditional financial service providers such as banks and credit card companies, it showed mobile network operators (MNOs) the speed with which they can be left out of the business. Their only income in this case was the data traffic generated to complete the transactions.  It is not news that the Average Revenue per User (ARPU) continues to decrease and mobile financial services are a great opportunity to reverse this trend. In mature markets such as Western Europe, the decline in ARPU has already led to a reduction of revenue. In Latin America the continued expansion of the subscriber base still enables revenue growth, but this trend will not last forever.

Source: Strategy Analytics 2012

I recently met with Tom Elliot from Strategy Analytics, a Boston-based consulting firm to discuss these issues.  Tom stressed that nowadays it is hard to find an MNO that is not developing some kind of financial service, but that nonetheless the business model is still uncertain, and what works in certain contexts is hard to replicate successfully in other markets.  Strategy Analytics’ forecasts for 2016 (see charts) do not show many signs of innovation in the industry. Their outlook is rather an inertial one where the aggregate income of the industry will flatten or decrease according to the region. These figures are more or less within the consensus of the mobile industry consulting world.  However, the promise of financial service provision is enticing for MNOs when properly implemented. M-Pesa, Safaricom’s mobile money service in Kenya, contributes 17% of total ARPU, which represents 53% of non-voice ARPU. While Kenya has its own particular market characteristics, we can use this as a best case indicator of the potential of mobile financial services. A 17% increase in ARPU in 2016 in the case of Western Europe, for example, would push income levels above those of 2007.[1]

The threat to this promise is the model à la Starbucks, where MNOs become dumb pipes. The model for obtaining significant revenues must be one in which the carriers are efficient players of the ecosystem, beyond the mere provision of connectivity to mobile phones.  Consequently, the construction of a model that avoids treating carriers as dumb pipes in the developing world requires important definitions of the core variables of the ecosystem. Depending on the definitions of these variables, very different business models can be shaped: from a scheduled savings product for house improvement targeting the unbanked, to the Starbucks model mentioned above. In each of these models, the players of the ecosystem have different roles: banks, MNOs, retailers, credit cards, etc.   In Latin America, at the time of shaping this ecosystem of mobile financial services, carriers have decided to split the risk and the investments by partnering with banks, credit card companies or both. The two most significant initiatives in the region are Wanda, a joint venture between Telefónica and Mastercard and Transfer, another JV between America Móvil and Citibank, which officially launched in Mexico this past month.

Source: Strategy Analytics 2012

The uncertain viability of the different business models explains much of the reasoning behind this decision. However, these partnerships have direct implications when defining the basic variables mentioned above, which need to be negotiated and agreed with the partners.  Banks for example, can be very good partners for cash management and identification of customers, but not so effective for other tasks. A report published this year by the World Bank, puts Latin American banks among the most expensive in the world. Expensive partners might be reluctant to embark in low margin/high volume business models. Experience shows that banks have yet to reach out with a value proposition to the 50% or 60% of unbanked households in Latin America.  On the other hand, credit card companies can be great allies for mobile payments and short-term loans, but their record in offering other financial products such as savings products is lean.

There is still little evidence in LAC to establish the conditions under which these associations can be functional to the carriers’ need to supplement their declining ARPU.  But already some points are clear. Carriers are those with the most to gain (and to lose) in this bet on mobile financial services. Their partners do not have as much at stake. Some results that would be catastrophic for MNOs, such as the Starbucks model, would leave their partners relatively well positioned.  In an industry with the current volumes exhibited in Latin America, there is ample space for a wide array of players to explore the business opportunity to provide mobile financial services for the base of the pyramid, which will hopefully result in a more tailored provision of services to those who most need it.

 

---

[1] Strategy Analytics 2012

Photo Taken by Adam Jones

In a meeting last week with some colleagues from a branchless banking provider in Africa, we came across data that indicated that one particular provider (a competitor) was capturing a large portion of the payments market. The response around the table was clear, “Oh, that’s because this player is seen to be close to the people.”  The perception of this provider’s role in the market had given it a significant edge.

We can all think of specific examples of how perception influences behavior in our own markets. Some banks are seen to focus just on the corporate segment, while others are thought to reach further down market. Some MNOs are perceived to target their services to the upper class, while others appear to be geared more toward the “average consumer.” But how does a company, or in this case – a new player in the financial services market – change the way it is perceived among a new customer segment it is trying to reach with new products and services?

This is a tricky question because as behavioral economics tells us, people often think and act irrationally. And while perceptions are usually based on some level of reality, they can also be completely inaccurate. I read Blink by Malcolm Gladwell a few months ago, and one of the many examples in the book of the power of perception showed that just by changing the label on a can of soda, people’s perception of the quality of the drink changed even though the product inside stayed exactly the same. Add to this the fact that perceptions tend to be very personal, and we’ve got quite a mess on our hands.

For example, I remember hearing in one African country that having an expat as the CEO of a local company was a good thing because the population knew that s/he wouldn’t be dragged into the ethnic politics that often swayed the way businesses were run. This distance was perceived to be a good thing by some. On the other hand, other people in the same market thought that a company with a local CEO showed that the company was committed to the country and to its people. This closeness was perceived to be a good thing by others.  How is one to win?

This is frustratingly complex, and yet it is incredibly important to figure out. Why? Because perception is very closely linked to trust, and trust is fundamental for any sustained behavior change in the way people manage their money, and therefore for any branchless banking service to succeed. In some recent data that we gathered in Ghana, mobile money was perceived by the non-user to be more costly than the alternatives. Yet the reality is that mobile money is cheaper, if not the cheapest, way to move money in Ghana, especially at low values. On the other hand, users of mobile money understand that one of its greatest values is not just the convenience, but the price.  And users also have an accurate perception of the way mobile financial services can solve concrete financial pain points in their lives.

And herein lies the key – inaccurate perceptions only begin to change as customers are exposed to the accurate reality through repeated interactions. Get someone to use mobile money a few times, and they will soon see that it is cheaper than what they’ve been using. Their perception on cost will eventually change.  Guarantee (through platform reliability, agent liquidity, etc.) that the first few user experiences are positive and then other negative perceptions about trust and security will begin to disappear as well. For providers, it is critical to first understand the customer perceptions that may be holding their businesses back and then to tackle them head on.

Dr. Susan Johnson is a Senior Lecturer in International Development at the Centre for Development Studies at the University of Bath.  Her primary research interest is the means through which social and cultural factors influence the economy and in particular how these factors influence the operation of markets in developing countries.

The rapid uptake of mobile money transfer in Kenya has ignited enthusiasm globally over the potential to bank poor people via the platform of mobile phone technology. On the basis of research undertaken for Financial Sector Deepening Kenya, I argue that the evidence suggests an alternative explanation which means that formal service provision for poor people needs to be thought through in a very different way.  It means going beyond the expectation that mobile technology can adequately lower transactions costs to produce a revolution in inclusion, to recognising that managing financial resources has important social dimensions.

The research examined the reasons behind use of the whole range of services and so explores how mobile money transfer fits into the financial landscape as a whole.  For years the popularity of informal financial groups in the form of ROSCAs and ASCAs has been evident.  Indeed, many of those who are banked also use these mechanisms. Mobile money transfer has now overtaken informal financial groups as the most used service.  In our survey, based in three more rural towns and chosen to cross-cut poverty levels but particularly focus on the low-income group, some 61% were registered mobile money transfer users, 51% were using informal financial groups and 36% were using banks (higher than the last FinAccess 2009 survey figures of 22%).  So how can we explain why banks lag so far behind when from an objective perspective they appear to offer a safe and secure place to save?

The reasons people give for using mobile money transfer have now gone a long way beyond the original “send money home” remittance rationale.  Mobile money seamlessly facilitates inter-personal transfers to their close and extended family and friends for school fees, investments, celebrations and funerals, “assistance” and “help”, borrowing and so on – that is, any reason that people might need to send money to each other.

These interpersonal transfers operate within social networks that involve relationships of ‘give and take’ that can operate over long periods of time and in which resource transfers may be given in one form, for example, cash and returned in other, for example, support with resources of many different kinds or social connections to a job and so on.  Hence mobile money transfer has brought a range of financial transactions that involve a reciprocal dimension.

Informal financial groups offer, first, discipline and commitment in saving through the regularity of the contribution; second, the ability to access small but useful lumps sums; and third, proximate liquidity in the event of particular emergencies or needs. The latter is achieved through the social connections groups offer which allow people to negotiate access to funds directly from the group or indirectly from other members. These groups can also be characterised as operating through a reciprocal dimension. Hence they operate with some similar characteristics to the transfers being captured by mobile money transfer:  there is reciprocity and negotiability over how funds are borrowed (or ‘saved’ with others) and returned.

Our evidence indicates that the logic behind bank account use is often related to the need to receive payments rather than make voluntary savings and this helps explain high levels of dormancy.  Access to loans from banks is limited and they can be hard to manage when they are received.  Since interest on small amounts of savings is effectively irrelevant and loans are hard to get, putting funds in the bank secures neither access to financial support nor useful social connections. That is, banks lack an attractive reciprocal dimension and there is little negotiability involved.

Hence this argument suggests that the rise of mobile money transfer is evidence of extensive informal financial behaviour which has characteristics similar those to informal financial groups.  Hence rather than suggesting that mobile money is a short cut to formal sector financial inclusion, this analysis suggests that mobile money transfer has revealed an alternative underlying logic to which the formal sector needs to respond if it is to attract savings – it must offer an acceptable reciprocal relationship.  In order to provoke discussion an alternative approach might be for banks to pursue a “credit-led savings” strategy in which they offer easily accessible small personal finance loans in order to demonstrate that they can enter into valuable relationships with their customers.

For further information click the following links for the summary and the full report.  Also, for more information, check out Dr. Johnson’s recent post at Accion’s Centre for Financial Inclusion blog.

For those of you in Washington D.C, Susan Johnson will be discussing the challenges that formal services face in the search for financial inclusion in Kenya at an event hosted by CGAP on April 25^th.

To attend this event, please register here by April 23. 

Nick Hughes and Gautam Ivatury are two of the founding members of Signal Point Partners, a company created in 2009 to build innovative mobile services in emerging markets. Nick was previously at Vodafone, where he started M-PESA, taking it from a concept to a multi-million-dollar business in five years. Gautam’s previous role was leading the technology program at CGAP, where he focused on branchless and mobile banking.

When we launched Jipange KuSave – a mobile-only savings product – in Kenya in early 2010, our goal was to out-compete the mattress. Back then, Safaricom’s M-PESA service was in hyper-growth phase and ramping up to become the de facto national retail payment system. But even more exciting was M-PESA’s potential as a pervasive and low-cost delivery channel for a wider set of financial services.

With this in mind, we decided to attempt for savings what M-PESA had done for money transfers – get millions of Kenyans to abandon informal mechanisms and instead become our paying customers. But if Kenyans were going to save with us instead of the mattress, we’d need to solve two challenges.

First, a ‘traditional’ bank-type savings proposition would never work. Poor people have never abandoned the convenience and enforced discipline of informal savings services for a couple of percent interest.  In Jipange, the combination of micro-loans and savings in a structured program met several customer needs, notably the need for cash when cash flow is low (liquidity) and steady progress towards a lump sum (a savings goal).

Second, our costs would need to be radically low. As ING Direct had shown, “pure” mass-market savings plays can make money, but only at high volumes and low margins. And that was in developed markets with larger account balances. For us to succeed, we would need to “throw out the rulebook” and design from scratch the most efficient and lowest-cost processes to manage relationships and transactions.

With our two “first principles” in mind, we gathered the essential ammunition for an attack on the mattress:  a radical product design, drawing heavily from Stuart Rutherford’s work; a set of web-based processes to run the product solely via M-PESA (limited physical contact with customers); a stellar project lead to manage implementation; and passionate, risk-seeking funders in CGAP and FSD Trust Kenya.

Interested readers may find it useful to read more about our product development and trials here in MIT Innovations. Also, this evaluation produced by FSD Kenya. In short, the Jipange KuSave product gave customers small amounts of credit at zero interest, while placing a portion of the credit into a “forced” savings account. As customers repaid the credit at whatever speed and in whatever amounts they wished, they became eligible for a bigger zero-interest loan. By borrowing multiple times and being forced to save a portion of each loan, they gradually accumulated savings.

The short version of our battle report is this:

1. Customers are hungry for better ways to save. They deal with cash flow complexity everyday and use a range of high cost / high risk methods to achieve liquidity. Some product designers would consider blending credit and savings as too complex – that was not our experience.  Clear, structured program, yes – but too difficult for customers to grasp, no.

2. Silicon Valley-style discipline and lean startup principles are keys to success. This starts and ends with customers. We quickly acquired a first trial cohort and modified and iterated the ‘offer’ on the back of real evidence from users.

3. A brand-new, mobile-oriented deposit-taking institution has the best chance of beating the mattress. This is perhaps the most difficult stumbling block on the way to scale. Only a regulated institution can take deposits — but hungry, highly innovative regulated institutions are rare beasts.

Chrissy Martin is a Senior Project Manager at MEDA, a non-profit organization that is partnering with Mobile Transactions on product development and agent network expansion, with a specific focus on reaching rural clients through services designed for the agricultural market.  Azalea Carisch is a Rural Microfinance Intern with MEDA who is currently based in Lusaka supporting Mobile Transactions on agent network monitoring and compliance. 

Mobile Transactions Zambia - courtesy of Chrissy Martin

Until recently, Mobile Transactions could have been considered the best kept secret in Africa.  Operating in Zambia on a shoe-string budget, they have been developing their own unique business model for electronic financial services slowly and with little media attention.  Now, as of February 2012, this small company has secured investments from three big investors, Omidyar Network, ACCION Frontier Investments, and Sarona Asset Management. All three are banking on the fact that Mobile Transactions’ experience and innovative approach to serving a range of consumers situates them to fill crucial gaps in the mobile money transactions and payments market in Africa.

Mobile Transactions offers services to both individuals and institutional customers.  For individuals, they offer both mobile wallet peer-to-peer transfers and over-the-counter money transfers through their agent network (a mobile wallet with a stored value is available but not compulsory.)  For institutions, they are providing e-voucher services to donors, governments and private corporations (including Dunavant, FAO and WFP).   They are also testing a variety of bulk payments products, including microfinance loan payments and utility payments.    Finally, they are working with Zambian Breweries on supplier payments, often referred to as business-to-business (B2B) payments.   These products are at a variety of different stages, with money transfers and e-vouchers reaching the most customers so far.  This product mix is extensive, but is not in and of itself enough to distinguish Mobile Transactions from the many other mobile payments start-ups.  So the question is what makes Mobile Transactions different?

• Mobile Transactions is an independent operator:  Readers of the CGAP Technology blog know that the market has developed past a simple MNO-led or bank-led model. Increasingly, the market is facilitating new business models which do not partner exclusively with any one mobile network operator or commercial bank.  Mobile Transactions is one of these independent companies, due in part to the regulatory environment, which allowed them to become an independent, certified financial transactions company.  This allows them to think creatively about their technology and go beyond the mobile phone when developing services. For example, the e-voucher service mentioned previously can by delivered to end-users either through a mobile wallet or via paper scratch-cards.  Paper-scratch cards are printed with a unique transaction code that is linked to the recipient’s identity and is redeemable at certain retailers.  The paper-based distribution method works better in rural communities in Zambia, where many people do not have a mobile phone but everyone trusts and understands scratch-cards, which are extensively used for mobile airtime top-up.  A mobile phone company, on the other hand, is interested in promoting their core business, airtime, and would not have an incentive to offer a service that does not require the end-user to make a transaction via a mobile phone.  Of course, there are drawbacks to the independent model, mainly the lack of an existing capital base, brand recognition, or distribution network, all of which are crucial to any branchless banking business.  Yet, Mobile Transactions has managed to find alternative solutions to provide quality service for people and institutions needing to move cash within Zambia.

• Mobile Transactions relies on independently owned and operated agents: Mobile Transactions’ original market-entry strategy looked much like M-Pesa in Kenya, which meant that they recruited and trained existing retail outlets to grow their agent network as quickly as possible.  However, they quickly saw that this strategy was not sufficient to drive customer acquisition.  The retail outlets weren’t motivated to act as sales people for a company with little brand recognition or to push a new product that few Zambians understood.   As a result, Mobile Transactions decided to recruit, train, and set-up their own “Champion Agents”.  Champion agents operate in much the same way as franchises do: each store is independently owned and operated by a trained individual who receives marketing and commercial support from MTZ.  This model has driven brand recognition and has provided Mobile Transactions with a backbone of core agents who are devoted entirely to selling their products and services.  Although this model is much more expensive than the retail agent model and therefore results in slower agent network growth, the benefits are accrued in the quality of the network, which benefits from the support and oversight that Mobile Transactions provides to agents, who are the customer-facing entity of any branchless banking operation.

• Mobile Transactions’ products are derived from a service-led approach. Mobile Transactions does not experience the low activity rates of many other mobile money operators.  Most other operators offer one product, the mobile wallet, and then build value-added services on top in order to drive higher customer use of the m-wallet.  Mobile Transactions, on the other hand, provides a variety of services (business to business payments, payments to farmers or microfinance loan payments) that respond to a specific customer need, and these services may leverage the mobile wallet, a paper-voucher, or simply an agent-based transaction (for example, the scratch card voucher mentioned previously.)  The commonality is in the central Mobile Transactions IT platform, where each type of transaction is processed regardless of the delivery channel. The challenge of this approach is scale, which is reliant on the agent network and the agents’ ability to adapt quickly to new services and to serve multiple customers segments.  As they deal with these challenges, Mobile Transactions is continually growing services through a learning-based approach to product development that allows it to develop services based on Zambia’s market realities, rather than success from other markets.

Mobile Transactions’ model is not without its challenges, as any member of the company’s small and dedicated staff will tell you.  However, the experience of this independent operator with a service-led approach challenges many of the commonly held assumptions about mobile money implementation, and their patience is starting to pay off.  Our guess is that it won’t be a secret for much longer.

This is the fourth post in a series on remittances. The previous blog in this series discussed the scale of the remittance flows between Russia and Tajikistan and the huge potential for linking them with other financial products. Stefan Staschen works regularly as a consultant for CGAP’s Government and Policy Team and is an Associate with Bankable Frontier Associates. 

You can probably understand our excitement when CGAP learned about the initiative of two banks to leverage the large remittance flows between Russia and Tajikistan. Agroinvestbank in Tajikistan offers three such linkage products in cooperation with Russlavbank in Russia, which operates the CONTACT money transfer system.

The following three products are offered:

1. A “Road Loan” where migrants (before they leave Tajikistan) can receive loans up to US$1,000 mainly to cover travel expenses for their trip to Russia. The migrant, once in Russia, uses the CONTACT system to make repayments from Russia at a preferential commission rate (i.e. a lower rate than charged for remittances). In April 2011, Agroinvestbank had 6,000 road loans in its portfolio with an outstanding loan balance of US$3.2 million. (This is the only product which is also offered by a number of other banks.)
2. “Hamvatan” (meaning “compatriot”) is a basic deposit account offered by Agroinvestbank, which can be opened remotely at Russlavbank branches in Russia and can be replenished through the CONTACT system. Only the migrant can withdraw money from this account (and only once he has returned to Tajikistan). So in a way it’s a savings account in Tajikistan replenished by a remittance from Russia. As of mid-2011, only 38 accounts had been opened with an aggregate balance of US$30,000. Most clients close their account once they return to Tajikistan as they would rather deposit the money in an account earning a higher interest rate.
3. “Family Card” is a current account with two debit cards – one for the migrant and one for his family. The migrant worker opens the account in Tajikistan before he travels and he can replenish it from Russia through the CONTACT system at a preferential commission rate. Unlike the Hamvatan account, family members can withdraw money in Tajikistan even before the migrant returns home. (No data on number and volume of accounts was available for this, but I assume the banks would have touted the success if there was anything to tout.)

What are the reasons for the low take-up of these products, even though the potential seems huge, and what can be done to make them more attractive? The following are a few indications, although more detailed research would be required to provide definite answers to these question Read the rest of this page »

This is the third post in a series on remittances. Stefan Staschen works regularly as a consultant for CGAP’s Government and Policy Team and is an Associate with Bankable Frontier Associates. 

Tajik migrants, courtesy of Stefan Staschen

My colleague Olga Tomilova and I recently were in Tajikistan and Russia to learn more about the large remittance flows between these two countries. We were most interested in the potential to link remittances with other financial products, such as loans and savings accounts in order to increase access to finance on both ends of the remittance corridor.

Having lived and worked in Kenya for many years, I inevitably started comparing Tajikistan with Kenya, and I realized that this is actually quite an interesting comparison.

For example, taking one point of comparison, in Kenya there is a lot of talk about the high aid-dependence of the economy (5.7% of GDP). But it turns out that it is still small in comparison to the “remittance-dependence” of Tajikistan, which peaked at 50% of GDP in 2008 (i.e. just before the effects of the global financial crisis could be felt) and stood at 40% in 2010 (according to the World Development Indicators). Obviously remittances and donor funds are not the same, but if we are concerned about the changing mood of donors and its potential effect on Kenya, how much more should we be concerned about fluctuations in remittance flows to Tajikistan!

Such large remittance flows do not only benefit the recipients, but can also be great business for banks. (In the case of one bank, the revenues from commissions corresponded to 46% of its net interest income in 2009.) But they create the risk of precarious living conditions for Tajik labor migrants and expose the recipients and the Tajik economy as a whole to the whims of Russia’s immigration policy, which has already been tightened several times.

There is actually another analogy between Kenya and Tajikistan: in both countries remittances boom, only that in Kenya it is mostly domestic remittances exemplified by the run-away success of Safaricom’s M-PESA mobile money service. While every Tajik receives about $325 in remittances annually (of which at least 90% originates from Russia), domestic transfers flowing through the M-PESA system alone amounted to about $200 per capita in 2010. Mostly as a result of M-PESA, the percentage of the population excluded from formal financial services dropped substantially between 2006 and 2009.

Read the rest of this page »