Latest entries

Eight things we learned in China driving Fintech fortune



In May, the FiDA Partnership took 26 leaders from Africa’s digital finance industry to China to learn about their unprecedented FinTech boom. The week was immersive and intense, designed to allow participants to escape the boardroom and experience Chinese digital life firsthand on the streets of Beijing, in the mom-and-pop stores of Hangzhou, and the on factory floors of Guangdong. In this blog, we reflect on the key lessons.

1. Government played a critical role in providing an enabling environment

There is no doubt that China’s protectionist policies allowed local players to flourish, free from competition from global incumbents. The government also invested considerably in establishing the ecosystem to support digital financial services. They mandated state-owned banks to provide widespread access, resulting in 80% of adults having a bank account and rural branch and ATM penetration on par with high income countries. This eliminates the need for FinTechs to develop expensive cash-in-cash-out networks as they have been required to do in sub-Saharan Africa. This has been underpinned by prioritization of rural development, providing broadband connectivity, universal IDs, and road infrastructure which lowers the cost of servicing remote areas.

Regulators’ “wait and see” approach sparked the FinTech boom in China. They began with an incredibly open framework, allowing models to evolve before tightening regulation, reminiscent of the approach that enabled MPesa to take off in Kenya. When regulation was eventually introduced it created space for FinTechs to flourish – over 200 licenses were recently granted to non-bank institutions. Companies we spoke to said regulators encourage frequent interaction with them. This means FinTechs know about upcoming rule changes and regulators are up-to-speed with new innovations, which is also assisted by regulatory sandboxes.

2. Business model innovation is happening in a big way

Laissez-faire regulation and growing tech expertise gave rise to innovative new players entering the financial services space with fresh business models. One such model is peer-to-peer (P2P) lending. This fulfils a demand from middle income savers seeking higher returns and greater flexibility than was offered by banks, and provides credit to the mass-market which was neglected by traditional financial institutions in favor of state-owned big business. We met China’s largest P2P lender, Yirendai, which offers prime borrowers access to unsecured credit by connecting them to investors through its online marketplace.

Digital finance business models in China are increasingly data-driven. Therefore, companies with large data pools and expertise in deriving insights from them have led the way. These tend to be tech giants such as JD.com, Alibaba and Tencent, so we are seeing profits in the financial services value chain shift from financial institutions to tech companies. The fact that these companies’ core business is not finance, but social networking or e-commerce, means they are not dependent on financial services revenues for their existence. This promotes experimentation and therefore product sophistication as companies can be patient for profit. It also means companies can subsidize entry-level products – usually payments – to encourage uptake. Alibaba’s rural e-commerce entity, Cuntao, explained how diversified revenues also help increase financial services’ reach: although bank business models lack incentives to serve rural customers who are low value and expensive to acquire and service, these clients hold value for e-commerce business models, permitting the financial services which support them to reach these communities as well.

3. The power of AI

China’s FinTechs have developed enviable know-how in artificial intelligence (AI), which has a visible presence in China: we shopped at JD’s unmanned stores which use smart shelves and facial recognition. We also saw AI in action at Weijun Grocery, a mom-and-pop shop in Hangzhou which uses Alibaba’s retail management platform to digitize inventory management. Weijun receives advice from the platform on what to stock based on historical sales, a trove of data about neighbouring communities, and timely information such as weather forecasts – for example to ensure ice-creams are in stock during a heatwave. We saw how cameras track customer movements to create a heat map showing where they spend most time in-store to help the owner optimise layout and merchandising.

AI is also used extensively on the back end. JD, Yirendai and Bairong told us how they leverage AI to process large datasets for risk management. JD uses more than 30,000 variables and 100 models to credit score over 200M users. Some of the variables used are astoundingly nuanced such as how hard a customer presses the screen and at what angle when they make a purchase. Their system is over 10x more efficient than traditional models and enables them to offer differentiated pricing. AI is also used to increase efficiency and efficacy of customer acquisition: Yirendai’s robo-adviser, Yiri, has resulted in a 50% hike in assets under management. WeBank, China’s first online-only bank, showed us their impressive live dashboard – like a banking version of NASA’s mission control! By using AI and blockchain they cut annual per user IT cost to $1 – a tenth of the industry average.

4. Partnerships are key

In the world of African digital finance, commercial partnerships are most eagerly sought by smaller companies hoping to leverage assets of larger incumbents – typically banks and MNOs. When collaborations do happen they are often tense, the imbalance of power played out in battles for revenue share and responsiveness. Rarely do we see strategic partnerships between giants, except out of regulatory necessity.

Not so in China. There, even the largest tech titans acknowledge that they can’t do it all. We heard about partnerships addressing a range of needs including risk management, scale, capital constraints, technology, expertise, and regulatory compliance. Two of the biggest players, JD Finance and Tencent, told us how they partner to share user data, enabling JD Finance to develop a deeper understanding of their customers for credit profiling. JD Finance explained how they partner with banks to access low-cost loanable funds in exchange for their large de-risked customer base. Yirendai revealed how they forward prospective clients to partners if they are unsuited to their own business model. In China, partnerships abound.

5. Social media & entertainment are powerful for customer acquisition

Tencent’s core business is social networking and entertainment, channeled largely through it’s messaging app, WeChat. WeChat offers a vast array of services – in Beijing we ordered taxis, hired Ofo bikes, ordered meals, checked movie times, and shopped online – all within the app. This grants WeChat relevance in almost all aspects of millions of users’ everyday lives, enabling Tencent to integrate financial services in meaningful ways with immediately applicable use cases. By designing experiences around lifestyles, financial transactions are minimized into the background of services that matter more to users. This is in contrast to what we see even in more mature digital finance markets in Africa where financial services are often offered in isolation from existing behaviors, limiting uptake and breadth of use.

6. Trust is a big pull factor

Professor Long Chen, Alibaba’s Chief Strategy Officer who joined us for a lakeside chat in Hangzhou, noted that tech platforms’ ability to create trust has been a significant growth factor. Micro-entrepreneurs in Panjiayuan Market, Beijing told us how digital payments took off because they eliminated the risk of fake notes. A Calligraphy Brush Seller noted: “I even use Taobao to sell to customers I already know because it creates trust. Payment only happens if the customer is happy & there is a process for returning goods if they are not”.

7. Rural e-commerce is creating new opportunities for job creation

Cuntao has extended e-commerce into rural areas by setting up 30,000 service centers at village convenience stores, enabling villagers with poor internet access to access goods previously unavailable to them. Centers receive goods ordered through Alibaba’s ecommerce sites and handle last mile logistics to customers’ homes, in return for commission. They are also education centers, teaching villagers how to shop online and helping them place orders.

Cuntao also provides two-way distribution infrastructure, enabling rural producers to access new markets. Clusters of rural online entrepreneurs who have spontaneously opened shops on Taobao Marketplace to sell their produce have emerged and have been termed Taobao villages, We went to one of them – Huidong – where we visited a factory started 6 years ago by a young man who worked in shoe factories his whole life, first as a labourer then working his way up to manager, designer, partner and finally starting his own business. He started the factory with $1,600 and a single room. He now employs 200 people and sells 1 million pairs of shoes annually on Alibaba platforms. It was impressive to see the role that e-commerce has played in creating jobs. Cuntao, told us their rural ecommerce solution has created more than 1.3 million new jobs nationwide and brings brings RMB 180,000 benefit annually to each village.

Olga Morawczynski from the Mastercard Foundation commented:

Tracking the economy of jobs that have been created because of Alibaba has been fascinating. Everything from people sorting the shoes to manufacturing the shoes to sorting the post-production material. We hear the story that technology destroys jobs but I think here we’ve seen the ability for it to create jobs and increase the rate of creation of jobs.

8. The future is offline

Alibaba believes that a foothold in traditional retail is the path to growth and has developed an online-to-offline (O2O) plan called “new retail”, which redefines commerce by enabling seamless engagement between online and offline worlds. New retail melds the best of in-shop and online experiences in new and unexpected ways, Using data such as purchasing history and store visiting habits, Alibaba intends to personalize product offerings, purchasing experiences and marketing campaigns. Alibaba’s Hema supermarkets incubate these innovations; we experienced how shopping there is a smartphone powered experience. We scanned QR codes to get product information and payment is cashless and almost touchless, using facial recognition combined with Alipay embedded in the Hema app. Each store also serves as a fulfillment center for online sales, delivering customer orders within 30 minutes to anywhere within a 3km radius.

This is all part of an effort to tap into the huge offline commerce sector in the world’s largest retail market. Despite the success of e-commerce in China, 85% of purchases still happen at brick-and-mortar stores, amounting to a $3.9trn opportunity. Venturing offline also enables Alibaba to access customers who are excluded from it’s online offerings owing to lack of digital literacy or phone ownership. Using offline experiences to bring people’s transactions online presents clear learnings for Africa.

Eye in the sky, maize on the ground



Harvesting Inc. maps millions of miles of cropland across various regions. This image is from the state of Maharashtra in India.

Richard is a smallholder maize farmer in Western Kenya. He used to borrow money from his family or friends to buy essential inputs to grow maize—seed, fertilizer, and equipment—because he couldn’t access a formal loan from a bank. Most banks consider him a high risk client because of his low-income and lack of credit history. Recently though, Richard obtained a $50 loan for his farming business. By processing and analyzing satellite imagery on his farmland and cropping cycle, in combination with other alternative data, a lending organization was able to assess his creditworthiness.

Richard’s story is not unique. Remote data captured through earth observation technology (i.e., satellite imagery) could be a game changer for insurance and lending products by reducing organizational and logistical costs. High resolution remote sensing can guide ground observations and enable organizations to estimate yield at the village level. Satellite imagery—in combination with demographic, financial, agronomic, geospatial, and psychometric data—provides sufficient detail on “thin file” clients like Richard (clients without an established credit history) to make lending decisions. Innovative FinTechs like Apollo Agriculture and Harvesting Inc. are generating credit scores using algorithms that rely on mobile phone and alternative data such as earth observation imagery as well as machine learning of farmers’ needs, incentives, behaviors, and agricultural activities.

Earlier this year, FiDA spoke with Apollo and Harvesting to learn about the journeys they took to integrate satellite imagery (i.e., images captured with earth observation technology) into their business offerings and how they implemented earth observation technology. The findings highlighted in FiDA’s case study, “Launching into space: using satellite imagery in financial services,”  offer financial service providers (FSPs) interested in leveraging non-traditional, alternative data two relevant paths for using earth observation technology.

What does it take to utilize satellite imagery?

The FinTechs’ findings indicate initial predictive power from satellite imagery in terms of generating features relevant to credit scoring. However, for Apollo Agriculture and Harvesting to leverage this technology, they had to have the right expertise on board to process and analyze satellite imagery. Ideally scientists should have experience working with a combination of:

  • software engineering,
  • machine learning/data science,
  • remote sensing science, and
  • agricultural and/or environmental science

These skills, neither easily available nor inexpensive to recruit, are key to building the infrastructure necessary to process thousands of images and develop the algorithms and tools that turn raw data into meaningful insights. FiDA’s case study also delves into the other components that Apollo and Harvesting had to have in place before embarking on their journeys, such as investment capital and training data.

Cost-effective customer acquisition is a challenging aspect of the business

There are potentially half a billion farmers who are not served by financial institutions and a decent chunk of them could be if we overcame information gaps. There is a huge demand for data and increasing availability of supply. That makes sense for a business case

Harvesting

Apollo and Harvesting strongly believe there is a business case in leveraging satellite imagery, even more so when an organization is utilizing satellite data at scale. But, in its business-to-consumer (B2C) model, Apollo has had to contend with  the challenge of serving rural customers who are difficult and costly to reach. They have found that is it more difficult to recover customer acquisition costs for farmers with low-value loans.

Nevertheless, Apollo has demonstrated that with the use of satellite imagery, there is “ample room” for profitability at “imminently” achievable repayment rates. Harvesting points out that it’s a numbers game: using satellite imagery is ultimately cost-effective because the majority of the costs are fixed and, as an organization scales (with customers), those fixed costs remain, by and large, fixed.

FSPs must have a clear vision of why they want to leverage satellite imagery as well as realistic expectations of what the technology can achieve

[Satellite imagery] is sexy to put on a slide deck but if it’s not solving a problem, it’s a hammer in search of a nail. It’s hard and expensive and requires very specialized skills. Build your business without it if you can! Don’t do it because you think investors want to see it or conferences want to talk about it.

Apollo

The most fundamental question an organization must ask is: what do we actually want to do with satellite imagery? The value of earth observation data depends on the existing data an organization utilizes and the additional value satellite imagery can provide given the high human capital cost of utilizing this technology. It would be prudent for organizations to first be sure of the benefit of satellite imagery and reflect on how the technology will enable them to reach their specific objective.

FiDA is confident that the journeys presented in this case study will provide a critical perspective on both the challenges and the promises of leveraging satellite imagery in financial services.

Virtue and value in mobile operator big data



This guest blog was jointly authored by Marissa Dean (FiDA) and Jake Kendall. Jake Kendall is the Director of the Digital Financial Services Lab (DFS Lab), ​an early-stage accelerator delivering innovative fintech solutions to the developing world.


Mobile network operator (MNO) data has been the first bloom of big data’s potential on the African continent. Widespread adoption and use of MNO services underpin the belief that patterns in data can reveal ability and willingness to pay for financial services, as well as demographic and other segmentation signals.

With this in mind, the Mastercard Foundation Partnership for Finance in Digital Africa (FiDA) interviewed 30 leading financial services organizations to understand how they think about big data and analytics and how they use (and don’t use) big data. Of the 30 organizations that use big data, more than half agreed to share details about the types of big data they used in their business. This blog post focuses on how these organizations use and perceive MNO data in conjunction with lessons learned from the DFS Lab’s three cohorts of FinTech startups, many of which also use this kind of mobile data.

The key learnings from the interviews with the organizations and startups are:

  • The risk of fraud likely outweighs credit risk in unsecured lending;
  • Mobile behavior data has been heavily hyped but has significant limitations (discussed in detail below)
  • More specifically, mobile operators don’t share behavior data freely, and for good reasons;
  • While credit reference bureaus are adapting, they are nevertheless creating an unintended consequence of blacklisting many people for trivial default amounts.

Fraudsters, not farmers, are the real risk when it comes to unsecured lending.

Can data points like the patterns in the frequency and amount of mobile voice or data top-ups actually demonstrate an applicant’s ability to pay? Arguably, providers that use mobile behavior data this way are not actually looking at hundreds of micropayments to assess whether the noise indicates this individual can pay back 100 Kenyan Shillings, but rather performing Know Your Customer (KYC) analytics to suss out whether the applicant acts like a real person or like a fake account created by a fraudster. Indeed, premeditated fraud (especially “at scale fraud” conducted by criminal syndicates) may be a greater threat to the business model of digital lending than default by applicants who had intended to pay but run into financial difficulties after taking the loan, particularly for small denominations. As discussed in FiDA’s Focus Note, one digital lender relying on MNO KYC data encountered an instance of fraud wherein more than 100 loans were taken out by a fraud ring using the same name of a popular fictitious character for every SIM card registration. Another research participant noted that as more people become aware that mobile money, data, and voice behaviors are factored into credit decisions, the data becomes distorted as people figure out ways to game the system.

As algorithms based on behavioral data become more prevalent, providers will find themselves constantly playing a game of cat and mouse with fraudsters who want to take advantage of the all-digital nature of these products in order to scale successful fraud schemes to high volume (this is the flip side of scale, which is normally thought of as a way to scale successful products to high volume). There is money to be made and it’s likely that  fraud can’t be eliminated entirely.

Despite the excitement, there are many limitations to mobile behavior data.

There are a number of reasons why mobile behavior data does not give a holistic picture of an individual’s spending or device behavior.

One factor is that many low-income customers carry multiple SIM cards to take advantage of the different promotions offered and reduce their expenses from making off-network calls. They also turn off the data plan, wifi, or GPS to save battery power; delete heavy apps to conserve limited space; or use feature phone versions of apps like Facebook Lite that capture less data. Further, low-income people, especially women in rural areas, often share phones, further limiting the data’s usefulness. For example, although 93% of Kenyans had access to a mobile phone in 2016, only 78% owned a phone implying many were sharing phones owned by others.

Second, mobile behavior offers only a limited view of someone’s financial life. Most transactions in the informal sector are still conducted using cash, even in countries with successful mobile money schemes like Kenya. Therefore, relying solely on mobile money data means missing significant information about an individual’s income and expenditures. Daniel Goldfarb from Lendable mentioned these challenges in an interview with FiDA conducted in August 2017: Just looking at M-PESA only shows you a small percentage of someone’s total transactions. So M-PESA only gives you a small band of information. This works for MNOs who are offering very small loans, but the second you try to give out larger, meaningful sized loans you can’t just rely on M-PESA data to understand someone’s cash flows.

Changes in the environment or in MNOs’ own systems or marketing schemes will change the relationship between the real variables of interest that lenders want to predict (e.g., income or free cash flow) and information gathered from mobile behavior that might be used to build such models (e.g., top up behavior or numbers of contacts or calls). Thus, models need continual rebuilding.

Finally, machine-learning algorithm-based models can employ data on mobile activity, airtime top-ups, and online behavior to score, for example, the probability of credit default on a statistical basis. However, there are some instances where well-intentioned machine learning algorithms may discriminate against a particular group,such as women or an ethnic minority. These specific algorithms are trained to recognize and leverage statistical patterns in the data. So, if the data used contains bias or historical discrimination, the model will incorporate it into future predictions, thus potentially creating an unfair system that perpetuates the same bias and discrimination.

Machine learning in financial services is a nascent but growing field, and identifying algorithmic unfairness should be a priority in the development of future models. Recently, the Federal Reserve System in the US offered FinTechs an evaluation framework to guide their early thinking on the use of alternative data and fair lending risk. They argue that alternative data that doesn’t have an obvious link to creditworthiness may have a higher probability of fair lending risk, and unfairness (in algorithms) can only be detected by very careful analysis of the data and outcomes.

MNOs haven’t been proactively sharing data with the ecosystem.

The Cambridge Analytics data fiasco combined with Facebook’s complicity and complacency regarding how its software treats personal data have brought concerns around privacy and personal data usage to living rooms and boardrooms alike. What was legal yesterday is considered unethical today. The spectre of being sued or shutdown is looming globally as consumers become more aware of the reality of what using free apps and services actually means.

To date, MNOs have been very careful about sharing data. In interviews, they expressed that they were prohibited from sharing any kind of raw data due to regulation: their interpretation was that they could not give individualized information to third parties without consent from both the individual and the regulator. Further, they believe that selling raw customer data would erode hard earned trust with the customer,  especially in light of recent events.

Often the primary generators of data (i.e., customers) don’t know what kind of data they are generating and how it is being used, as the Facebook episode demonstrated. Moreover, the proliferation of relatively accessible digital credit providers, with more than 20 in Kenya, coupled with the advent of digital data trails created by smartphones, has given rise to consumer protectionist movements, such as the SMART Campaign, to advocate for  practices that protect consumer data by providing clear customer opt-in and -out services for data usage, mining, and reuse of data by third-party services. Such campaigns also push regulators to establish and enforce legal frameworks that safeguard financial consumers’ welfare.

Credit reference bureaus are starting to receive information from alternative digital lenders.

Whether credit reference reporting by digital lenders is helping or hurting low income individuals is debatable. Certainly, if managed properly,positive reporting particularly could help previously excluded people build credit records. However, anecdotal reports indicate that many digital lenders are not reporting positive data to CRBs, although it is not always easy to determine which are reporting and what they are reporting.

Additionally there is a growing risk that many individuals will be blacklisted for very small defaults. Yet, arguably, such defaults should be treated differently and forgiven more quickly because they  usually stem from circumstances other than intentional default or fraud. A case in point is that over 400,000 Kenyans are blacklisted with the CRBs for outstanding mobile loans of less than 200 Kenyan Shillings (about $2).

Where does the digital finance community go from here?

Despite obvious privacy concerns, the sector shouldn’t be over-regulated. Digital lending on the back of alternative data is still a relatively new experiment, and many are still hopeful that the business incentives will drive providers to target farmers and small businesses with attractively priced credit. A light-touch approach, such as personal data privacy standards and recourse for people who have been blacklisted for low value loans, would help to make the market more predictable and tell providers exactly what they need to do to comply.

For MNOs, it would be interesting to see more providers exploring how they might employ differential privacy techniques, so that the learnings that come from mobile behavior patterns can be shared ethically and responsibly with the DFS community for product development or market research, and with the development sector more broadly, without exposing the privacy of customers. While mobile behavior data will never reveal a complete picture of life, there still is utility in generating learning that can be used to better tailor products. Moreover, MNOs should focus on transparency with customers, particularly clear terms and conditions for customers who do want to share individual data with credit providers. More broadly, the sector would benefit from MNOs agreeing to standards for data access and data sharing with third parties to assess credit risk.

For Fintech digital lenders, there are several considerations: first, the amount of energy expended to use complex techniques to determine default risk due to lack of income versus fraud; second, finding the right threshold and pattern for reporting positively and negatively to credit bureaus; and third, building algorithmic models without statistical bias.

What are the opportunities for and threats to the growth of digital financial services?



The digital ecosystem—a collection of organizations, individuals, and policies that enable and deliver digital financial services—is rapidly evolving due to increased smartphone penetration in emerging markets and innovation in mobile technologies. How will this shift help or hinder innovation and development of digital financial services?

FiDA’s Snapshot 13, The opportunities for and threat to the growth of digital financial services, discusses three principal areas where opportunities for or threats to the growth of digital financial services may arise, namely: regulation, technology, and partnerships. For instance, regulation has spurred innovation in digital finance in some markets but hampered it in others. In 2015, the UK’s Financial Conduct Authority became the first to launch a regulatory sandbox (a means to develop regulation that can keep up with the fast pace of innovation) so that FinTech startups can test their services without normal constraints and thereby flourish. However, recent regulation on lending and data flows has negatively impacted the growth of alternative lending FinTechs in some emerging markets. Financial Inclusion on Business Runways (FIBR) found that—unlike regulations in the developed markets of China, the UK, or the US—regulation in Ghana and Tanzania did not allow non-financial institutions to lend, potentially preventing innovative FinTechs from providing credit to the unbanked. For digital finance to thrive, regulators need to effectively manage risks to stability and integrity, such as by protecting privacy and shielding against fraud, without stifling future innovation, as further  discussed in FiDA’s Snapshot 14 (forthcoming).

Snapshot 13 discusses these opportunities and threats in more detail, presents notable new learning in this field (such as the opening of APIs and increasing tech accelerator programs), points to implications for consumer protection and transparency, and lists the top 10 reads in this space this year.

Building a healthy digital ecosystem



Snapshot 12: “Building the infrastructure for a healthy digital financial ecosystem” addresses one of the key questions of FiDA’s Learning Theme 12. Each Learning Theme addresses a range of topics within the digital finance space. The FiDA Partnership synthesizes the digital finance community’s knowledge of these learning themes as “Snapshots” that cover topics at the client, institution, ecosystem, and impact levels to present “What we (in the digital finance community) know” about the topic in question, highlight “notable new learning,” and call attention to “implications” for future research and investment.

One of the most disruptive innovations in digital finance has been the development of mobile money agent networks. This model enabled financial services to shift away from the costs of  “brick and mortar” banks to a more accessible, trusted, and pervasive network of retailers that act as the ambassadors of mobile money. However, as noted in FiDA’s Snapshot 8, “What is the commercial landscape of digital finance?,” building and managing an agent network is a heavy investment, and, unfortunately, the growth and uptake of mobile money has been hampered, in part, by a lack of adequate agents in hard to reach, rural areas (i.e., the last mile). Nevertheless, as the industry evolves, new and more inclusive delivery models will emerge.

The agent network is simply one component of the larger digital finance “ecosystem” that needs to be both reliable — to build trust with consumers — and flexible — so that it can adapt to new consumer demands and technologies. Three core infrastructure pillars collectively define, at least from a supply-side perspective, the digital finance ecosystem: operational, technical, and regulatory. Digital finance will only thrive if the ecosystem infrastructure is robust, reliable, and dynamic. FiDA’s Snapshot 12 focuses on the challenges that the ecosystem currently faces and the infrastructure improvements that need to take place to deliver digital finance successfully. Although digital finance has evolved at a remarkable speed over the last ten years, its growth is at risk due to some persistent challenges across all three aforementioned pillars as discussed in the Snapshot.

Read Snapshot 12 to learn more about the infrastructure challenges digital finance faces, the implications for the ecosystem as digital finance evolves, and the top 10 must reads in this space this year.

Getting off the methods pedestal—how can we assess the impact of Digital Finance on clients?



Snapshot 15:’ “Approaches to determining the impact of digital finance programs” addresses one of the key questions of FiDA’s 16 Learning Themes. The FiDA Partnership synthesizes the digital finance community’s knowledge of these Learning Themes as “Snapshots” that cover topics at the client, institution, ecosystem, and impact levels.

Spoiler: Snapshot 15: “Approaches to determining the impact of digital finance programs,” is not the key to a cheap, fast impact method that works with any program. Blame complexity — and the digital finance community’s many innovations. If we replicated similar products with similar clients, in similar markets, then, maybe. But reality is more complicated than that.

Instead of one method to rule them all, Snapshot 15 explores different approaches to client impact measurement in order to expand the digital finance impact community and encourage the inclusion of various methods to advance the state of knowledge about the impacts of digital finance.

There is an impact conversation and we need you to join it

At FiDA we see impact assessment as a process more akin to a conversation than to the rendering of a verdict. One need not (and should not) rely on “capstone” documents shared only at the end of a successful project. Instead, impact conversations are broad, ongoing, multi-voiced dialogues about what is and is not working. Different study designs inform impact conversations in different ways at different stages.

But why should you, as a program manager or a product developer, pay attention to and share non-experimental research? The digital finance community needs to continue innovating to push the industry forward, and such innovation may be hindered if there is a two-year wait for the results of an impact study. We need to build based on the insights available. Individually, an impact insight may appear insignificant but collectively they are necessary pieces of this crucial  conversation. See the Digital Finance Evidence Gap Map (EGM) for an illustration of the power of collective impact insights.

Impact insights from diverse sources are valuable and advance the community

Certain experimental methods have been put on a pedestal. While experimental designs with control groups can produce powerful insights, the various setups of digital finance programs mean that experimental methods are not always an option. That is OK. Experimental methods are not the only way to gather impact insights. For instance,  if the FiDA EGM only included Randomized Control Trials (RCTs), our opportunity to learn would be vastly reduced. Including studies that used an array of methods have added appreciably to the digital finance sector’s knowledge at a point in time where evidence is limited.

What approaches have been used to explore the impact of digital finance

In Snapshot 15 we describe a variety of approaches that have been used in digital finance impact research to get a sense of the possible approaches from which researchers and developers can draw. The methods we highlight represent primary approaches that can be coupled with other tools. For example, bolstering a panel with Most Significant Change stories to provide clients’ versions of impact. The evaluation toolbox is full of approaches wherein primary methods can be enhanced by other tools to confirm, refute, explain, or enrich the findings.

So what can you do to measure impact?

Invest in a strong theory of change and test it.

A theory of change describes how a program is expected to contribute to change and in which conditions it might do so. That is, “If we do X, Y will happen because…”. In Snapshot 15 we also describe the attributes of a robust theory of change. Impact research based on robust theories of change guides choices about when and how to measure outcomes. This simplifies decisions on the choice of tools in the research toolbox.

Look for comparison groups when you can—but it’s OK if you can’t find them

Strong theories of change coupled with comparison groups and/or other causal inference methods is an ideal standard. However, a strong theory is still useful in the absence of a comparison group.

Impact research should determine if a program contributed to an observed change. A comparison group is one of the many approaches to inferring causality. But even without comparison groups, you can connect the dots. Consider the way that justice systems establish causality beyond reasonable doubt. Theories are developed, information is gathered, and evidence is built to create a case for the initial theory and for alternative explanations. Evaluators have been inspired by this and we highlight alternative approaches to understanding causality.

You might never be certain but you can get close

In realist research there is no such thing as final truth or knowledge. Nonetheless, it is possible to work towards a closer understanding of whether, how, and why programs work, even if we can never attain absolute certainty or provide definitive proof. Moreover, to approach the truth, we must acknowledge not only what worked but also what did not work. Insights into the positive, negative, and null effects all serve to refine theories of change, the implications of which are broader than any single program.

To conclude, our best sources of evidence come from numerous methodologies in dialogue with each other. But we need the digital finance community to to lead with rigorous theories of change, find the comfortable research method, and join the impact conversation. To learn more about how you can add to the conversation, read Snapshot 15.

Is digital finance changing the lives of the “excluded” for the better?



Snapshot 4: “How do advances in digital finance interact with dynamics of exclusion?” addresses one of the key questions of FiDA’s Learning Themes. The FiDA Partnership synthesizes the digital finance community’s knowledge of these Learning Themes as “Snapshots” that cover topics at the client, institution, ecosystem, and impact levels to present “current insights” about the topic in question, highlight “what (the digital finance community) can do,” and call attention to “implications” for future research and investment.

Grace is 55 years old; she has a primary school education and a small business selling produce in her village. Her husband owns a mobile phone, but she does not. Benson is 30 years old; he has a secondary school education, and works in the city as a taxi driver. He owns a mobile phone. Grace and Benson learn  about a digital credit product that provides small loans through a mobile phone.

Do you think Grace and Benson will have the same experience accessing and using digital credit?

Why we wrote this Snapshot

When digital financial services are designed for broad populations, it’s easy to assume that different demographics within a population will use them in the same way and experience the same impacts. Yet, factors such as age, literacy, gender, geography, and language shape the nature of the adoption, use, and impact of digital finance. That is,  a digital credit product’s effect on an outcome like “growth of business” may be greater or less depending on the client’s age or education. A recent analysis by CGAP and FinMark Trust highlighted the distinct variables that can compound exclusion from financial services.

Snapshot 4, “How do advances in digital finance interact with dynamics of exclusion?” discusses persistent issues surrounding exclusion and how digital finance research measures exclusion. We offer these  observations to sensitize researchers and practitioners to the importance of these dynamics and encourage exploration of how distinct variables of exclusion determine how marginalized groups experience outcomes.

Don't assume effects are universal

The prevailing narrative around the impact of technologies on a given population is a variant of the idea that a rising tide lifts all boats. Yet, research rarely presents dis-aggregated insights. Instead, studies report the average effect which suggests that every individual in the study experienced the reported effects equally. A review of digital finance impact studies from FiDA’s Evidence Gap Map found that 78% of studies did not report even a  basic variable: gender dis-aggregated data. While, studies with just women result in useful insights, if we do not look at the way digital finance is used by women in comparison to men, it is impossible to argue that a digital finance product or service has been more or less helpful to those women than it would have been to other segments of the population.

Sometimes the marginalized benefit more

When the design of a digital financial service syncs with the needs of an excluded group, the benefits of using it accrue disproportionately to them. Snapshot 4 highlights digital finance studies that present instances in which being female, less educated, lower income, or from rural areas was associated with greater effects than being male, more educated, or from urban areas. For example, an impact study in Burkina Faso highlighted that while mobile money made no difference in the savings behavior of relatively advantaged groups (urban, male, and highly educated), it increased the probability of saving for disadvantaged groups (rural, female, and less educated).

Sometimes the marginalized benefit less

Unfortunately, there are many cases in which the benefits of a technology accrue mostly to high status, high skill individuals, rather than to the marginalized populations we often wish to serve. Snapshot 4 highlights a study that found, for higher income households, that an increase in savings services was associated with less reliance on asset depletion to cope with economic shocks. However, an opposite effect was observed for those with lower incomes.

These observations can provide the digital finance community with a more refined understanding of impact by determining the conditions under which impact applies or is stronger or weaker. They also underscore the need to examine how a digital finance product may interact with and affect various excluded groups.

Build and evaluate strong theories of change

What to do when faced with these dynamics of exclusion? We foreground that the first step is to develop theories of change that allow for impacts to accrue differently to different user groups. This is fundamental to understanding the potential impact heterogeneity and, ergo, to designing impact research. A heightened awareness of these challenges will help practitioners plan appropriate digital finance services that their underserved clients will want , and be able, to use regularly. 

Read Snapshot 4 for more details on our findings, implications, and a list of the top-10 reads in the space.

The state of the Mobile Money industry in 2017



This guest blog was written by Francesco Pasti, Senior Manager of Mobile Money Services at GSMA.

Numerous new trends emerged in mobile money throughout 2017 – from the accelerated growth of bank-to-mobile interoperability, to the emergence of South Asia as the fastest growing region, and a raft of innovations designed to reach the most underserved. The mobile money industry is now processing a billion dollars a day and generating direct revenues of over $2.4 billion. With 690 million registered accounts worldwide, mobile money has evolved into the leading payment platform for the digital economy in many emerging markets. The 2017 State of the Industry report on Mobile Money from the GSMA sheds light on several factors underpinning the success of a growing number of mobile money providers: a sustained focus on activity rates, the digitisation of platforms and measures to reduce the net cost of the agent network. On each of these fronts, the trends in 2017 were positive.

A growing number of mobile money services are achieving activity rates of over 50 per cent

While average industry activity rates grew modestly to 36 per cent in December 2017, a closer look reveals significant variation among providers. Our analysis shows that these providers all have a strong distribution network, enjoy enabling regulation, and rely more on an account-based business model.

More funds are entering and leaving the mobile money ecosystem in digital form

Use cases such as bulk disbursements, bill payments and bank-to-mobile transactions have been the main drivers of this trend. As mobile money becomes more digital, it is connecting the wider economy and, in turn, becoming more profitable for providers and more useful to consumers.

Many successful providers are decreasing the net cost of the agent network

Agents remain a crucial and distinguishing asset of mobile money providers. In recent years, we have seen growth in the number of active agents and average values processed by agents. At the same time, the inflow of digital funds is reducing provider costs, by alleviating the need for subsidised cash-in agent commissions.

Amidst a changing landscape that sees the spread of smartphone and fintech companies and an increased digitisation of new sectors of the economy, mobile money providers serving as a payment platform for a broad range of entities appear to be best placed to thrive.

The persistence and scale of the cash economy in emerging markets means that complex distribution networks remain crucial for digital services to interface with physical lives. By leveraging these enduring assets and finding new ways to connect scale with innovation, mobile money providers can serve as a gateway to the widening array of digital services in emerging markets.

Policy objectives will play an increasingly important role, as the scope of mobile money regulation broadens. While the pace of core regulatory reform is slowing, this masks two important emerging trends: the extension of new areas of regulation to mobile money and the rapid spread of financial inclusion policies. As regulators confront questions around data protection, regulatory sandboxes, and more, the policy end game of greater inclusion must remain at the fore.

Read the full report for the detailed analysis of these levers to growth and sustainability, and for spotlights on success stories and examples of innovation from mobile money providers around the world.

What makes a successful commercial partnership?



Snapshot 10: “What makes a successful commercial partnership?” addresses one of the key questions of FiDA’s Learning Theme 10. Each Learning Theme addresses a range of topics within the digital finance space. The FiDA Partnership synthesizes the digital finance community’s knowledge of these learning themes as “Snapshots” that cover topics at the client, institution, ecosystem, and impact levels to present “what we (in the digital finance community) know” about the topic in question, highlight “notable new learning,” and call attention to “implications” for future research and investment.

In 2016, Paypal, a US based online payments system, partnered with two of the largest global credit card networks, Mastercard and Visa, and concurrently invested in a FinTech platform, Acorn. In doing so, Paypal became a bridge between traditional financial institutions and emerging FinTechs. Collaboration may extend the value of Paypal’s services to more people, expedite innovation, and future-proof Paypal in this era of rapid mobile-based innovation and the rise of large internet players.

On the other side of the Atlantic, digital finance players appear to have an increasing appetite for collaboration. For example, Safaricom’s M-Pesa, the Kenyan mobile money giant, and the Commercial Bank of Africa (CBA) in Kenya, partnered to deliver the savings and microloan product M-Shwari. Leveraging Safaricom’s dominant position in the marketplace allowed the product to successfully scale: as of 2016, M-Shwari accounted for approximately 15% of CBA's total revenue. The partnership has succeeded because each partner has a clear understanding of their respective roles and how the product benefits the interests of each. CBA, a corporate bank that targets higher net worth individuals, benefits from the large pool of savings without diving into the operational challenges. Further, CBA does not necessarily want to brand within M-Shwari’s target market, and thus only Safaricom brands the product. In turn, Safaricom benefits from the banking infrastructure of CBA without which it would not be able to provide loan services.

These and other examples suggest that there is a business case for dominant players in financial services markets to collaborate at the ecosystem level. However, a number of factors determine whether a digital finance provider will collaborate in a financial services market as Snapshot 10, “What makes a successful commercial partnership?” explores in detail.

What influences players to collaborate? 

In essence, power dynamics in a financial services market influence how digital finance players engage with each other. Market conditions, such as a fiercely competitive market or a quasi monopoly environment, determine players’ actions. In some markets it may be neither necessary nor fruitful for players to collaborate. Providers who have highly dominant positions—such as b-Kash in Bangladesh which accounted for about half of the market presence in digital finance in 2016—may be reluctant to share scale advantage with smaller competitors. They may prefer instead to forego the advantages of interoperability — which provides interconnection, payments aggregation, and infrastructure sharing — to lock in their market position.

However, in fragmented markets where competition is greater,  digital finance providers stand to benefit from working together to pool their customers into one interoperable network in order to enable interconnection and payments aggregation. For instance, mobile money providers in Cote D’Ivoire collaborated to provide a universal and accessible digital school registration and fees payment solution along with a streamlined user experience. The program worked because its services were attractive to each of the stakeholders: the MNOs benefited from increased revenue flows and the government benefited from the cost savings and reduction in lost payments.

Moreover, the increasing success of mobile insurance services and the tangible benefit it brings to all stakeholders justifies collaboration between specialist providers and digital finance providers. In 2014, 64% of mobile insurance services were launched by MNOs in partnership with specialist solution providers. An MNO might use such a partnership strategically by offering the insurance product under its own brand. Or, in a purely transactional  partnership, the MNO might only provide the platform. For new launches in 2015, 57% of services collected premiums through airtime deduction;the remaining 43% relied on mobile money as the primary payment option.

New players will bring new business models

Digital finance providers are looking to the future. The imminent threat of large internet players is driving partnerships between new types of players in the digital finance space, such as that of  M-Kopa, a pay as you go solar energy provider, and Safaricom. These types of collaboration can facilitate cross-network mobile payments which encourage a larger population to use digital financial services, and, in turn, help digital finance achieve its social and commercial potential. Snapshot 10 discusses the necessary ingredients of a successful partnership: such as  a long-term vision of the partnership and each partner working from the position of their strengths and competitive advantages. Successful partnerships have demonstrated — such as in the case of the Kenyan savings and loan product, M-Shwari — that they have the potential to reach a large number of unbanked but mobile enabled customers and thus extend financial inclusion.

Read Snapshot 10 to learn more about what it takes to successfully collaborate, the new types of partnerships that are evolving, and the top 10 must reads in this space this year.

Can big data shape financial services in East Africa?



Every minute of every day millions of users in Africa create digital data. This new data is creating opportunities for alternative “big data” to catalyze an expansion of financial services to low-income and hard to reach populations. FiDA’s Snapshot 9, “Best Practices in Big Data Analytics” and FiDA’s Focus Note, “Can Big Data Shape Financial Services in East Africa? discuss the potential of big data analytics in financial services in greater detail.

Snapshot 9 focuses on the opportunities created by big data  — such as the number of lending and insurance products that have launched across sub-Saharan Africa. Indeed, mobile credit services are leveraging usage and behavior data to determine which customers can be granted small amounts of credit (cash or airtime) over their mobile phones. However, the extent to which companies are using mobile operator data and other big datasets is not well known. The return on investment in big data analytics is another unknown quantity still. What types of data, particularly what types of big data, are being used? Are the benefits tangible — how do the results from big data compare to those of traditional underwriting methods? What costs and risks exist? Is a market for data developing?

The FiDA Partnership’s Focus Note, “Can Big Data Shape Financial Services in East Africa? sets out to answer some of these broader questions. The Focus Note shows how 30 leading organizations — including Kenyan and Tanzanian banks, microfinance institutions, mobile network operators, and FinTech organizations — think about big data and analytics and how they use (and how they don’t use) big data.

FiDA’s  research found that players do not exchange and profit from big data directly. Rather companies have focused on developing relationships on a case-by-case basis to provide analytics of traditional or big datasets. This has led to a slow (but steady) uptake of big data and analytics and, as a result, a more measured expansion of services leveraging alternative methods. According to the interviewed organizations, four factors constrain rapid growth in this sector:

  1. Lenders have a limited use case for third-party data. Traditional underwriting practices still dominate credit decisions and repayment data and are seen as the most accurate predictor of risk. Big data is mainly used to determine a client’s ability to pay, typically with reference to low-denomination, short-term loans.

    Increasingly, lenders are exploring the use of big data to assess willingness to pay, however, there are limitations, including the ability to validate quality and authenticity. Marketwide data is typically neither structured nor detailed enough to be used in a meaningful way. However, one type of marketwide data — satellite data — is proving useful for several FinTechs.
  2. Organizations are pursuing partnership models in lieu of transactional relationships. For a variety of reasons, including compliance with laws and regulations, preserving customers’ trust, and maintaining an early competitive advantage, organizations are treading cautiously with customer data. As a result, relationships between players working in this field resemble partnership structures rather than transactional marketplaces. While the financial services ecosystem is evolving slowly in East Africa, the strategy behind partnerships allows players to explore what is possible with data and analytics on a case by case basis. Moreover, this partnership trend is likely to continue because MNOs need to rapidly build an ecosystem around digital wallets and establish partnerships with banks and FinTech players that can deliver a wider variety of digital financial services. In turn this ecosystem will generate more data that will enable the creation of better targeted products.
  3. Most of the organizations interviewed are still testing, refining, and experimenting with which datasets are most predictive as well as validating their analytical models. This has  led some FinTech organizations — such as Apollo Agriculture, FarmDrive, and Tala — to become customer facing, at least temporarily. By working directly with customers these FinTechs are able to capture data with which  to train and prove the predictive power of their models. The implication of this is that they have to fund the loans from their own balance sheets; as a result, the amount of capital available for lending limits the volume of data they can collect. Further, even in situations in which customer data has been shared, data analytics organizations have learned that integrating with a provider’s system  requires tremendous time and effort and hampers the ability of FinTechs to scale. Banks and MFIs need to improve readiness as much as FinTechs need more runway to improve models and offerings.
  4. Most banks and FinTechs believe a business case that is strategy- and leadership-led is essential to the uptake of big data and analytics. To expand financial services to low-income or hard to reach customers, banks need to be part of the solution. However, for this to happen, banks will need to shift their perspectives on the breadth of the customers they would like to reach.

The absence of a marketplace — and the pivoting of pioneers who initially paved the way to develop the market — could indicate that it’s too early for big data plays. More likely, however, it demonstrates that a market will never emerge in the same way that it has in the United States. Progress will come gradually, and partnerships among mobile network operators, banks, and FinTechs will be crucial to success. Organizations will need to weigh the value of being a pioneer against the current market challenges outlined above.

As a complement to  the Focus Note, FiDA has developed Profiles of Digital Finance Organizations Leveraging Data and Analytics—brief profiles of how the organizations that participated in the research are using data. The profiles categorize data sources (big data as well as traditional) across four functional categories:

  • Individuals’ financial services use or history.
  • Individuals’ digital interactions using a device.
  • Other individual data, such as psychometric survey responses.
  • Marketwide data, such as crop prices or satellite imagery.

We hope that readers will benefit from the candid details shared by the participants in this study, without whom these findings and knowledge outputs would be impossible.