CGAP CEO Speech on Fintech and Financial Inclusion at Afro-Asian Fintech Festival


Greta Bull, CGAP CEO
Address to Afro-Asian Fintech Festival
July 15, 2019
Nairobi, Kenya

Good afternoon. My job here is to kick off a discussion of fintech and financial inclusion along with this illustrious panel. I want to start by putting some definition around what I mean when I use these two terms, as people often use them in very different ways. My definition of fintech is straightforward and broad: it refers to the use of technology to deliver financial services in a better way – and can be done by anyone from a tiny start-up to an MNO to a bank to Facebook. Financial inclusion in my world is quite simply about reaching poor and excluded groups who have never had access to formal financial services (sorry, rich world millennials!). At CGAP, we have been working on the nexus between financial inclusion and technology for decades. Our mission is to understand the ways that better financial services, delivered through technology, can help people to improve their lives. In our thinking, financial services are not an end in themselves; they are a means to a greater end. With all the excitement going on in the fintech space, it is really important for us to hold on to the meaning of this term, financial inclusion.

Because it IS an exciting time to be working on technology and financial services. It is pretty clear globally that fintech is playing an outsized role in disaggregating the financial services value chain in fundamental ways – making it more customer centric, nimble and efficient. I like to think of my financial services toolkit as a Swiss Army knife. It is a tool I use to get something done. With technology, the pieces of that toolkit are being taken apart by providers attacking the maddening inefficiencies in financial services delivery – finding vulnerabilities around risk, cost and customer experience that have made banks such an easy target. What it means for me is that I can now piece together the best set of services for me, in a new and convenient way. I still have a current account and mortgage with my credit union but get a better return on my savings through the robo-investing app, Betterment. I send money to my account in the UK with TransferWise – it is a lot cheaper and faster than my bank. I collect payments for renting my house on Airbnb through PayPal. I insure that house through a specialized on-line broker. I buy my coffee using a Starbucks app, because I get loyalty rewards. And so on.

But I come from a high-income country: I have a wallet that is worth financial services providers competing for. And there are plenty of ways for fintech providers to get those services to me in the United States. That is not the case in many emerging markets. Although there are certainly people with means in emerging markets, for the most part, the mass market is low income. Here in Kenya, for example, per capita income is $1,620 per year, which means that the average Kenyan earns only $4.33 a day. Almost 37 percent of the population lives on less than half that amount, which is the official poverty line. That kind of baseline changes the equation considerably for providers. And yet we have seen huge innovation here in Kenya in the last twelve years, based on novel ways of delivering financial services to the mass market. And that early fintech revolution in Kenya has echoed across the world and contributed to the rapid pace of innovation in the financial services space that we see today.

So, what are the forces that are delivering this transformation in financial services in emerging markets? I would argue that it is being driven by three factors: distribution, digital connectivity and data. Distribution is the foundational element in this – the one that makes it all possible. And not coincidentally, it was the first one that was cracked – by pioneers like M-PESA here in Kenya, who realized the need for a cash-in cash-out interface between the analog and digital economies. Alibaba in China solved for distribution in a different way, building an e-commerce business on a cash on delivery basis, and then reverse-engineering a digital payment system to support it. More recent entrants like Gojek in Indonesia are solving distribution by bundling it with other services, like ride hailing. These distribution businesses are important because of their reach and their ability to onboard people to new accounts, but they are expensive to run and therefore need scale. And to really build scale, you need not only large numbers of customers, but also deep relationships with those customers. And that depth can only be gained through product diversification. So those early distribution businesses are busy turning themselves into platforms so they can deliver more to their broad customer base. It is this “platformization” of integrated services where the other two factors – APIs and data - have an increasingly important role to play.

Let’s start with APIs. Although large DFS providers across Africa have proven their ability to scale distribution, they are often not best placed to develop that deeper range of services for their many customers. They are too busy managing core operations – registering customers, managing a huge distribution network, and dealing with regulators. The relatively slow pace of innovation across these vast distribution networks is a contributing factor to high rates of inactivity in many emerging markets. At the same time, there is a range of 3rd parties who are developing really interesting products and services. Many of these start-ups are small and nimble with creative ideas that range from leasing tractors to PAYGO solar to financial health apps to social networks and e-commerce. But they need a way to get their product to market. And the large DFS players offer that channel. Problem is, integration is complicated, painful and time-consuming, and small companies have neither the resources nor the bargaining power to deal with the MNOs. What ends up happening is that large DFS providers prioritize a handful of larger partners with a clear business case where profits outweigh the high cost of one-off integrations. These tend to be companies like utility providers who integrate for bill payments. Sometimes they offer an innovative solution, but often as not, we are looking at established, volume-based businesses. Result?

  • Only large companies and aggregators integrate
  • Smaller, higher risk innovators are shut out because it is difficult and expensive to gain access to the channel
  • Development of new and niche products is held back

Recognizing this, many large providers are beginning to use Open APIs to adopt a ‘payments-as-a-platform’ approach, and CGAP has worked in partnership with a number of them to figure out how to make this work. If an MNO opens up its APIs to third parties, as operators like MTN in Uganda and Wave Money in Myanmar have now started doing, then a whole new range of options becomes possible – the innovators come to you, because you can make customer acquisition possible for them. Open APIs are not only good for giving consumers more choice, they are also good for businesses. Customers stay within the provider’s ecosystem, using their services. Because consumers have more options, they buy more, increasing revenues. The integration costs for large providers come down due to economies of scale. And smaller start-ups have lower integration costs and can bring their products to the market more quickly, testing customer demand. It’s a win-win. And there is good evidence that operating as a platform enhances profitability. In its state of the Industry Report for 2017, the GSMA reported that profitable mobile money deployments were associated with high levels of integration with third parties. In fact, this group was 42% more profitable on average than other providers in the survey. Open APIs are driving a lot of the innovation we are seeing in the fintech today – from emerging platforms in Africa to Open Banking in the United Kingdom. They are enablers for innovation.

APIs also enable a freer flow of data, which is our third driver of fintech innovation. Although it is widely understood that data is the fuel that will drive the digital economy, we are still in the early stages of figuring out how to build the engine in most emerging markets. Business models trying to harness data in Africa today face a few challenges. Firstly, although there is an abundance of data and that abundance is growing exponentially, it is unevenly distributed. Low income consumers tend not to be well-represented in digital data trails. Secondly, the providers who are gathering data on low income consumers are generally inclined to hoard that data, understanding that it is potentially a valuable resource. And this leads to a third challenge – those who have the data don’t necessarily know what to do with it. This may be because they lack the skills to effectively structure and extract insights, but they may also lack clear use cases. The data may in fact be more valuable to another provider, who may have greater ability to do something with it. But we don’t yet have many good examples of business cases, protocols and commercial agreements that make sharing data a viable proposition. Beyond this is a challenge around fragmentation of data: data is valuable when it comes in large quantities and is structured such that it can observed over time. Today, data comes in pieces that make it a perfect illustration of the old metaphor of blind men feeling an elephant. To some of them, the elephant seems to be a trunk, to others, a tail, and to others, a leg. No one has the complete picture. The need for large data pools also means there are at the same time emerging risks around concentration of data. Big data-driven tech companies like Facebook, Google and Ant Financial pool enormous amounts of data under one roof, and they will eventually bring those capabilities to the financial services market in Africa. But these big tech players raise important questions of market concentration and competition. And underlying this are important issues of who controls a consumer’s data; who has the right to benefit from it: the consumer or the provider; and who is responsible for protecting a consumer’s data and privacy? Although we can see the immense potential of data, and some providers, like digital lenders, are making commercial use of it, the data journey is only just beginning, and we still have much to learn.

We know what factors are driving financial innovation and inclusion around the world, but what are the special considerations needed to serve the poor? CGAP spent two years working with 18 early stage fintechs to get a better understanding of how providers are overcoming the challenges of working with this segment. Not surprisingly, they are the result of the absence of the very thing driving rapid change: poor people don’t have full access to distribution networks, they are less likely to be digitally connected, and they don’t leave much by way of data trails.

Our fintech partners focused on solving for this in three ways: building visibility, lowering cost and improving the user experience.

Visibility. Low-income people operate in the informal, analog economy, and therefore don’t have the visibility needed to access financial services, even if they have capacity to use them well. For example, low-income people cannot easily access cheap, timely credit because they or their businesses are not considered creditworthy. That is not necessarily because they are not creditworthy – providers simply don’t have the information to make an informed judgement. We know from financial diaries that the poor have complex financial lives and can be expert money managers. But if they remain informal and analog, these customers are invisible and therefore, ineligible. MaTontine in Senegal and Patasente in Kenya are making digitally visible the transactions that occur in existing networks like a women’s savings group or a fast-moving-consumer-goods (FMCG) value chain, helping to improve these customers’ creditworthiness by digitally recording payments within those networks. Another firm, Tulaa, also here in Kenya, aims to make input financing less risky by applying a digital loan directly towards input purchase and tying loan repayments to the sale of a farmer’s crop.

Cost. The single biggest barrier for providers serving the poor is the high cost to serve relative to the amounts available from consumers, so providers serving this segment have a need to bring unit costs down to a level that supports a viable business model. Think about yield-based crop insurance. Traditional methods of assessing or sampling yields at a community level, physically, makes this kind of insurance very costly to provide and therefore unattractive for exactly the segment that needs it most: smallholder farmers. Pula uses satellite data to create predictive models for the yield of specific geographic areas, which will in theory drastically reduce the cost of assessing a farmer and pricing insurance for them. The biggest challenge here is data: the statistical models that drive these services require many years of varied data and application in different climatic regions to ensure reliable conclusions.

Experience. We often don’t think about building an engaging, interactive customer experience for low-income customers, because it is too expensive, and they often don’t have the smartphones or data plans that enable such interfaces. Traditional inclusive business models are built on standardized, lean, basic delivery, which can often leave a poor customer hanging when they most need answers, leading to low-trust and a correspondingly low uptake of services. Some B2B firms like Juntos and Arifu are trying to address this, by engaging with customers in real-time, on behalf of banks and MNOs, and often with very basic technologies like SMS. In markets where smartphone penetration is rising, the experience can be even further enhanced. Payment apps like NALA build intuitive, engaging payment experiences on a smartphone, but make sure to ride on USSD technologies, which are still dominant in many markets, keeping costs down for their customers, who may have a smartphone but do not use their data capabilities for cost reasons.

So we know that technology can help us serve these clients. What we don’t yet know is whether there are sustainable business models that can reach these families at scale. To serve the poor – both broadly, in other words reaching everyone, and deeply, providing a well-rounded set of financial services that meets their needs - we need to find ways to bring the power of distribution, connectivity and data TO the poor, making them visible and therefore eligible; bringing down the unit costs so we can build sustainable business models; and improving the customer experience in ways that crowd the poor into the digital economy. This is a challenge, but the good news is that there are companies out there working hard to crack the code.

This is doable. But it will take time and continued investment. Here are three things we can do to make it happen:

Keep building the rails. Access to channels and other market infrastructure is critical to the survival of fintech companies trying to serve the poor. They need to be able to plug into larger platforms that enable them to leverage distribution, connectivity and data for their clients on affordable and easily accessible terms. Ideally, that should be a win-win for fintechs and platform providers, and there is a growing recognition that this is the case. What hasn’t happened yet is self-service Open API models that make this work at scale. This is a high-tech, low-touch way to make partnerships work, and is a proven model in many developed countries.

Keep data open. Digital finance easily lends itself to a winner takes all attitude, and this is especially important in data. For data to be useful, and for markets to remain competitive, data needs to be available, and we need to invest in the skills to use it, and in protective guardrails to make sure it isn’t misused. Ultimately, we should be aiming to put control of data in the hands of consumers, so businesses aren’t competing on the basis of their control over consumer data, but rather on offering a service that is competitive and of value to those consumers. Data aggregation models are emerging that can help make this a reality, but there is a lot of work to do to make sure this evolves in a carefully considered way.

Finally, we need smart investments. Serving the poor is difficult and requires investors who are in it for the long haul. Money is flooding into fintech right now, but it isn’t distributed evenly. Africa receives less investment than many other regions and, when it does come into Africa, it tends to chase a few established players. The Inclusive Fintech 50 recently announced by the MIX, in partnership with the MetLife Foundation, Visa, Accion and the IFC, found that of more than 100 fintechs headquartered in Africa, 70% of the funding went to only 7 companies. We need to look for ways to make sure a broad range of investment capital is available to fintechs in Africa from angel through to venture rounds. We also need to make sure that donor resources are focused in the right way – sustaining grant funding so that critical market infrastructure can continue to be built, while at the same time ensuring that blended finance facilities don’t crowd out, but rather de-risk and extend the funding available to earlier stage companies. Beyond this, local currency debt finance remains a huge constraint among more mature fintechs.

These are without a doubt very exciting times to be in the business of fintech innovation, but my plea to all of you is to make sure that fintech innovation, and the forces that are driving it, help develop economies and produce opportunities for all, including the poorest among us. Thank you.

 - END -