Blog

The Human Touch Required to Evolve Digital Finance

Posted on by MSC

Service Offerings at Agents are Static & Rudimentary

Across East Africa The Helix Institute’s research (2013) is showing that even after seven years of market development in Kenya, six years in Tanzania and five in Uganda, agents are still providing the same very rudimentary services that they did from the beginning.  The graph below shows that just about all of them provide both cash-in and cash-out services for customers, but that those are the only services which are provided in any uniform manner across countries.

In Kenya, 79% of agents also report offering account opening services for new customers, but both Tanzania and Uganda lag far behind here.  In Uganda, significantly more agents are involved in airtime top-up and bill pay compared to its East African counterparts, however, even in Uganda, 83% of agents do not offer these services.  Money transfers (also referred to as direct deposits) are not another service, but just a way that customers try to circumvent paying a P2P transfer fee by having agents send the money for them and is actually something most East African providers try to eliminate.

Commissions Structure & Desire for Speed to Scale Drives This Pattern

This often confuses people who correctly understand that airtime top-ups and bill-pay are major drivers of volumes and values on digital finance platforms.  The difference is in East Africa they are executed on the handset and not at the agents.

There are two major factors driving this trend, the first financial and the second related to scale.  The first issue is the manner in which most East African providers structure their pricing and commissions, best explained here by Ignacio Mas.   Basically, providers lose money when customers cash in, because the customer does not pay a fee, yet the agent earns a commission.  Further, they have to share the revenue they make on cash-outs with the agent, as the agent is actually physically conducting the transactions and needs an incentive to do so.  However, providers earn and retain 100% of the revenue from transactions made over the handset. Therefore, in order to limit the amount of commissions they have to pay to agents, and increase the amount of revenue that they earn directly, services are designed to incorporate the agent as little as possible.

The second major reason is speed to scale.  If a service can be offered digitally, then it is easier for it to grow virally.  People can enroll and use it anytime, anywhere, and all the impediments of paperwork and having to deal with a person are eliminated.

Agent Banking Growing, but Just in Kenya, and is Still Relatively Small

From both a business model perspective, and a financial inclusion perspective, we should be most interested in the absence of the sophisticated financial services on the right side of the chart (savings, credit and insurance).  These are services that are likely to create more transactions per customer (revenue), and also play a greater role in supporting customers to manage their money (financial inclusion).

Banks like Equity Bank, Kenya Commercial Bank (KCB), and Co-operative Bank are now building agent networks in Kenya that allow people to access banking services at the agent level, but relative to the telecom’s mobile money agent networks, the banks’ efforts are still eclipsed in the country figures, and therefore only really show-up as the 2% in the deposit column for Kenya on the above chart.  Uganda is still waiting for agent banking regulation that would allow banks to enter the market, and in Tanzania, the banks are far behind the developments in Kenya.

Channel Detachment for Next Generation Services

The next generation services like M-Shwari and M-Benki in Kenya and M-Pawa in Tanzania are examples of the more sophisticated banking services that have been lacking in the industry.  All three services provide savings and credit to customers.  They offer the potential to increase revenue for the providers, as well as provide more useful services for the mass market customer.  However, like P2P, airtime, and bill pay that came before them, they are all being offered exclusively over the handset, detached from the agent channel.

Customers enroll on their handset, and then mostly just move e-value from their mobile wallets to and from these services.  To cash-out (withdraw) from the services, agents are still involved, but the key is that they are not incorporated in the enrolment process or the subsequent support services.  Therefore, this new generation of exciting new services is detached from the agent channels that have historically been defining features of successful roll-outs.

M-Shwari was the first of these next generation services to hit the market, and about a year later, after a marketing campaign that was largely detached from the channel, InterMedia data shows that only 10% of Kenyan adults report having used it. This is far from the viral growth which we all hoped to see and the channel detachment that characterized its launch is likely a key factor in the lower than expected results.

Channel Involvement to Support Growth & Uptake

During this launch and initial growth phase, the most important activity to focus attention on is the detachment of the marketing campaign from the channel.  Most marketing for these products has been above-the-line, focusing on mass media advertising, billboards and banners.  While that was also a crucial component of the marketing when mobile money was launched, the big difference was that it was complemented heavily by below-the-line strategies which in the case of M-PESA in Kenya involved spending 10-15 minutes with each customer explaining the service, and paying over a dollar per customer acquired in commissions.

The current approach seems to assume that now that this expensive time consuming customer acquisition has been done, customers will now register for new products directly over the handset given some advertising to make them aware of the opportunity.  However, savings is different from other transfers and payments in that the person who is saving must have an order of magnitude more trust in the provider offering the service.

Customers are no longer just trusting the provider to quickly transfer value to someone they can verify with immediately. Saving money or buying insurance, the customer gives the provider money to hold, and must trust the provider over the entire duration of the policy cover, or life of the deposit.  This is a tough sell for a digital system most people still do not understand, and tougher sell for a digital system without a human face that can assure mass market customers it will work for them, and actually help solve some of their specific problems.

Concluding Thoughts

While offering simple transfer services digitally over the handset has taken-off in East Africa, scaling them digitally has not.  The lynchpin is likely in finding the right people on the ground to sell these more complex products.  Not all agents in the existing East African networks will be able to do this as it is fundamentally a different skill than conducting transactions, but there definitely will be an overlap, and given the trusting relationships many have with customers, this seems like a good starting point.  For the agents that cannot evolve to sell as well, they can be supplemented with sales agents who roam around professionally pitching the product.

After several years of development in the industry we do seem ready to make the leap to more sophisticated services, however, given the business models the telecoms have developed that encourage channel detachment, it might just be the banks – slow and steady – that invest correctly in the human touch that brings the needed trust.

InterMedia (with funding from the Gates Foundation) and the The Helix Institute of Digital Finance (a joint partnership between MicroSave, The Gates Foundation, the IFC and The UN Capital Development Fund) together, have collected the largest datasets in the world on digital finance (mobile money and agent banking), and the results are beginning to challenge some of our long held beliefs about how these systems work and what people are using them for.  The research focuses on eight digital finance markets around the world, with InterMedia responsible for the demand data from the customer perspective, and The Helix Institute complimenting it with supply data from the providers offering these services.  

Market readiness for mobile money interoperability

Posted on by MSC

Interoperability of mobile financial services potentially offers great benefits for the wider ecosystem. The value to consumers is obvious. This, in turn, leads to wider adoption; higher transaction volumes; greater velocity of money in the ecosystem; all of which are advantageous to service providers. It is now well established from both MicroSave’s Helix Institute of Digital Finance and CGAP studies that non-exclusive agents transact and earn more than exclusive agents. For the regulators, this translates to the reduction in cash; expansion of the formal financial economy and a direct impact on advancing financial inclusion.

The launch of M-PESA in Kenya in 2007 catalyzed a worldwide ‘mobile money movement’ and as services have proliferated, the pressure to create interoperable mobile money systems has mounted. Today, the development of an interoperable mobile money ecosystem is a prevailing need. During 2014, 9 mobile network operator groups (Bharti Airtel, Etisalat, Millicom, MTN, Ooredoo, Orange, STC, Vodafone, and Zain), have pledged to offer interoperable mobile money services across Africa and the Middle East. The question of interoperability is no longer of if, but of when.

Interoperability crosses many levels. For example, access or channel interoperability, service interoperability, and cross-sector interoperability. The below table provides examples of levels and their enablers:

From the consumers’ perspective, interoperability means more convenient and efficient services. Interoperability also has a key role to play in advancing financial inclusion.Furthermore, one of the main barriers to mobile money adoption cited amongst customers in Africa and Asia is related to the inability to send and receive money irrespective of provider. Many factors, including market conditions, will dictate how and when interoperability graduates from the very basic to eventually reach a state of full interoperability. Agent or channel interoperability can be introduced relatively easily, with low investments and mainly through regulations. Although, the key barrier is to bring competing market players together to offer even the simplest form of interoperability. Higher levels of interoperability(often referred to as account-to-account interoperability) need bilateral or multilateral agreements; investments in technology capabilities to integrate services across providers; and implementation of common risk management practices. These are more complex to implement.

Even the simplest kind of channel interoperability in the form of agent sharing has a significant bearing on consumer experience and agent income. Apart from the convenience of improved access and networks effects, it increases market competition and therefore enables the resultant benefits in terms of lower tariffs, better service quality, and greater consumer centricity. In most markets, the dominant player would tend to resist even a rudimentary form of interoperability, for fear (often unfounded) of losing out, even though the strategy is rather expensive and difficult to execute. The introduction of the non-exclusivity of agents in Kenya is a welcomed step and early signs of impact are already evident. Reserve Bank of India, the central bank in India, has been toying with the idea of white labeled agents (business correspondents), that can be a channel for any service provider.

Account to account interoperability is more complex and has major implications for the service providers. There are multiple models through which account-to-account interoperability can be implemented. The easier forms are bilateral arrangements between providers. The more complex forms are multilateral arrangements; common processing entity with or without commercial interests; or arrangements through automated clearing houses (ACH).  India, through IMPS system, provides a great example of infrastructure built to allow payments to be made across multiple MNOs and banks.  Interoperability between platforms and services can be a costly endeavor and could, in fact, make mobile money services more expensive to the consumer. This can be offset once network effects come into play and volumes pick up, leading to economies of scale.

Yet, interoperability is not such a straightforward issue. Some would argue that if market interoperability happens too early, there is a risk of stalling mobile money movement before it even starts. At inception, it is a matter of significant investments with limited returns. Unless global standards evolve and network effects occur quickly, interoperability can be a barrier for the first movers in the market to invest. We have witnessed with Safaricom in Kenya, MTN in Uganda and Vodacom in Tanzania that return on investment typically takes between 5-6 years from the launch of service. These operators became profitable without interoperability. It might be argued that the lack of interoperability did, in fact, enable profitability. Without interoperability between service platforms, the cost of offering mobile money and running out of e-money is enhanced because non-exclusive agents have to hold a separate float for each provider. In Uganda, agents turn away, on an average, 3 transactions per day for want of float; in Tanzania, they turn away 5 transactions each day, dissatisfying customers and negatively impacting their bottom lines. A layer of complexity does get added to the agents’ business model as they will be managing e-money floats between different providers, instead of just one.

Kenya, Uganda, and Tanzania are advanced mobile money markets. Interoperability makes sense in these countries, where there are multiple providers, most of whom have experienced the business benefits of offering mobile money products; and enough customers and transactions to not consider interoperability as a threat but an enabler to commercial business.

Even though Kenya is one of the most mature markets, interoperability has only recently come to pass in the form of agent sharing. For years, Safaricom reigned supreme over the mobile money kingdom, despite there being 5 other providers. Safaricom’s M-Pesa service still controls circa 67% of the Kenyan mobile money market, partly due to its early agent exclusivity arrangement, which is no longer in place and was formally outlawed in July. The Central Bank of Kenya ordered Safaricom to open up the M-PESA agent network to other operators in a bid to improve fair competition and encourage lower fees for customers. It will certainly be easier to achieve interoperability when the providers’ market shares are more even as in Tanzania (Vodacom 35%, Airtel 31%, Tigo 31% and Zantel 12%) than in markets with one predominant player (as in Uganda or Kenya).

Safaricom did have the first mover advantage in Kenya, but invested significantly in building and maintaining an extensive and robust agent network, so it’s no wonder that they may feel they are being treated inequitably now that competitors can simply use any of the agents they nurtured. For that reason, pioneers in other markets may feel disadvantaged too.

In any case, the initial cost and lower revenues that may result from interoperability are short terms. Taking a long-term view, interoperability can prove to be very beneficial to all the stakeholders. Historically, when interoperability has been introduced in the world of payments, be it with ATMs, credit or debit cards, huge growth and adoption have followed suit. Global standards for interoperable mobile money services will go a long way in presenting greater value to first movers and early adopters in the emerging markets. Just as standards like the ISO 8583 enable debit and credit cards at any ATM to transact with any bank around the world, it is desirable that similar standards are established for connecting mobile money services and making them interoperable.

Bangladesh Pioneering Unique Models & Innovations for Agent Networks

Posted on by MSC

Earlier this year, The Helix Institute of Digital Finance conducted a nationally representative survey of 2,800 mobile money agents in Bangladesh, coupled with qualitative interviews across the country. The 2014 Bangladesh Country Report provides insights into some of the unique models and innovative techniques players in this country have designed to develop an agent network of approximately 160,000 agents in under four years (given  bKash reports 80,000 agents and our findings are that  50% of agents in the market serve bKash).

Unique Leadership in the Market: No Telecoms

The report finds that bKash —a third party provider majority owned by BRAC Bank Ltd, managed by Money in Motion LLC with equity investments from The International Finance Corporation (IFC) and the Gates Foundation—dominates the digital financial services space with 50% of the agents offering their services.  They are followed by the Dutch Bangla Bank Ltd. (DBBL—28%), United Commercial Bank Ltd. (UCash- 14%) and Islami Bank Bangladesh Ltd. (mCash – 6%).

Beyond the burgeoning competitive landscape in Bangladesh, it is intriguing that none of the above players are telecoms.  Thus far we have seen the majority of the success in the digital finance space lead by telecoms who have large marketing budgets, national networks of retailors already serving them, and usually tens of millions of customers they can entice to register for digital finance.  In Bangladesh the regulation stipulates that telecoms are not allowed to brand their own digital finance services, which has given the opportunity to banks and third party providers like bKash.  This is strong evidence that players other than telecoms can scale agent networks in digital finance.

Distinctive Agency Demographics: Non-dedicated and Male

The different types of institutions leading market growth in Bangladesh are also making distinctive decisions about the demographics of their agent network.   In Bangladesh almost all agents (96%) have pre-existing, parallel businesses in addition to the digital finance service they provide (they are “non- dedicated” to the DFS business). As the below chart shows, this is very different to the leading DFS markets in East Africa, where many more agents are completely dedicated to the DFS business.  Generally, these types of agents can only exist in markets where transactions per day and therefore profits are relatively high, which therefore sustain the entire business.  Hence we might see a move towards more dedicated professional agents, if profits increase in Bangladesh in the future.  Another difference worth highlighting, is that while the majority of agents across East Africa are female, in Bangladesh 100% are male.  More research will have to be done to uncover both the drivers and the implications for customer uptake and usage of this gender difference.

Different Business Model Viability: Low Transactions and Profits

Median monthly profitability ($51) as compared to the leading East African Countries is low, and is a result of low transactions per day for agents.  However, 96% of agents are profitable, driven by very low median operational costs. When asked what the biggest barrier is to conducing more transactions, agents most commonly cited there are too many other agents competing for business, which is often an indication that focus must be shifted towards acquiring more customers, and encouraging them to transact more often.

Innovation: Liquidity Management

Bangladeshi service providers have created an innovative system to tackle the prevalent issue of liquidity management.  Most agents have their cash and electronic float delivered to them at their outlets, mainly by a ‘runner’ who is an employee of master agent (referred to locally as a distributor or aggregator).  As a result, the frequency of rebalancing (both cash deposits and withdrawals) is higher in Bangladesh than in East Africa. In Kenya, for example, agents do a median of just four cash deposits and three cash withdrawals per month as compared to a median of 12 cash deposits and ten withdrawals in Bangladesh.  As a result, Bangladeshi agents report denying a median of zero transactions daily due to lack of liquidity, in comparison, Tanzanian agents deny a median of five transactions each day, which is equal to 14% of their median daily transactions.

Summary: The Market View

Bangladesh is showing impressive results, and is finding unique ways of achieving them given the different operating environment there compared to those of the pioneers in East Africa.  There are definitely some challenges ahead in terms of increasing transactions and therefore profits at the agent level.  However, it is also important to note that with a liquidity management system that rebalances on demand, and agent demographics where almost everyone has a core business operating in parallel to the digital finance services they are providing, this might be much less important than it is in East Africa.  Further, many transactions in Bangladesh are done over the counter (OTC) and therefore only partially captured by the above statistics.  While this means the transactions and profits might be higher in Bangladesh this OTC system usually does not involve the required KYC verification, is not permitted, and therefore represents much more of a risk to the growth and functionality of the system than profits.  This topic will be discussed in future blogs.

Agent Network Accelerator Survey – Bangladesh Country Report 2014

Posted on by MSC

Bangladesh has created many unique systems for agent network management which are yielding world-class results especially with regards to liquidity management.  However, transaction volumes and profits are low compared to East Africa and support structures are still developing.

The research is based on 2,490 nationally representative agent interviews, carried out between March and April 2014. This is the largest research into agent networks carried out in the country, and reveals many exciting, cutting-edge operational insights into agent network management in the country.

To read through the report, please click here.

Insurance Product for Contractual MSE Workers of India – Behavioural Insights

Posted on by MSC

Taking cue from the earlier Focus note on need of insurance for contractual MSE workers of India, a team from MicroSave conducted behavioural research around the preference and choice of the clients for insurance products and services. This note details the research process, assumptions and broad behavioural insights generated through the research. The note culminates into conceptualisation of a low fidelity product concept designed using user centred design approaches.

Taking cue from the earlier Focus note on need of insurance for contractual MSE workers of India, a team from MicroSave conducted behavioural research around the preference and choice of the clients for insurance products and services. This note details the research process, assumptions and broad behavioural insights generated through the research. The note culminates into conceptualisation of a low fidelity product concept designed using user centred design approaches.

Insurance for Contractual Workers of Micro and Small Enterprises in India – A Conscience Call

Posted on by MSC

We applied a behavioural lens to examine why many clients do not save in accounts that MFIs open as business correspondents (BCs) of a bank. We found that a typical MFI is positioned (viewed in the market by their clients) as a credit service provider and, as such, MFIs do not fit into clients’ mental models of where to save. We discuss how MFIs can turn this situation around using client’s demand for credit and desire to accumulate lump sums as triggers to induce active savings behaviour through MFI-BCs. Such a change in product strategy will require MFIs to focus on branding themselves as savings service providers and to highlight their relationships with respected commercial banks to build trust. MFIs will have to be cautious not to position these savings services as a compulsory requirement, as part of loan insurance or at risk of offset against unpaid credit balances.