By: Travis J. Lybbert and Oscar Barriga Cabanillas
Introducing mobile money-based Digital Financial Service (DFS) products to the unbanked working poor raises a host of challenges. These challenges – once understood – would seem daunting and discouraging if they weren’t outweighed by the staggering potential of DFS as a gateway to more productive financial management and broader improvements in livelihoods and well-being.
In a new project in Haiti, we are tracking the launch of a new mobile platform that will grant access to small, short-term digital loans known as nano-loans. Our objective is to better understand both the barriers to uptake and usage of DFS and their potential benefits to the working poor. Digicel-Haiti has designed nano-loans to operate on their Mon Cash mobile money platform. These small 30-day loans ranging from $2 to $20 are disbursed and repaid via Mon Cash.
Nano-loans are the first major DFS product in a pipeline of other products planned for the future and are therefore positioned to serve as a gateway to financial inclusion, and improved financial management among households that are currently unbanked. Given this lead-out position, however, the stakes are high: subsequent access to and benefits from DFS hinges on the success of this initial nano-loan product.
A Foundation for Digital Loans in Haiti
We conducted an exploratory study to better understand the likely nano-loan clients. In July 2017, we convened a series of 10 Focus Group Discussions (FGDs) and mini-surveys that targeted potential nano-loan clients. Since initial diffusion of DFS is unlikely to penetrate isolated rural areas without Mon Cash agents, we restricted these FGDs to urban and semi-urban areas around Port-au-Prince and the Artibonite valley.
Digicel-Haiti has offered top-up advances to their customers for the past few years. These advances, called KrediPam and accessible through the Digicel phone menu, are effectively hyper-nano loans (average size of $0.33) in the form of pre-paid phone credit to be repaid a day or two later.
Given their top-up habits and liquidity constraints, KrediPam is familiar to many Digicel-Haiti customers. Nano-loans will likely target many of the same customers, so this familiarity provided a useful point of departure for the FGDs, which aimed to recruit Digicel customers from working poor neighborhoods with some familiarity with Mon Cash.
Typical Nano-loan Consumers
Those who participated in our FGDs tend to be food insecure: over half worry about food and frequently reduce meals or portions in response. Most were self-employed with some access to informal credit sources, including family and friends (especially popular with women) and informal pawnbrokers (especially popular with men). These were treated largely as stop-gap credit for food, transportation or school fee expenses. Crucially, financial literacy among participants was on average fairly limited.
The concept of nano-loans was appealing to many of our participants who anticipated tapping such a credit source regularly as an alternative to covering stop-gap expenses in a pinch. They quickly understood that the effective collateral for such loans was their continued use of their Digicel phone number. Many would do just about anything to keep the number given it is a critical link to their social world.
Potential Nano-Loan Challenges
The FGDs revealed several potential concerns and challenges that require careful, ongoing attention. Two stand out:
1. Financial literacy is low among the target population. To serve as a gateway to financial inclusion, nano-loans must attract those with little or no experience with formal financial services. Financial restraint and fiscal discipline in the face of what might be perceived as easy money develops only slowly for some.
2. Target clientele – many of whom are self-employed or day laborers – are vulnerable to high daily and weekly income volatility. This presents a particular challenge to the success of nano-loans when fixed repayment dates are strictly enforced. Since their informal alternatives often have some built-in flexibility, this rigidity was alarming to many of our FGD participants
These focus group discussions and short interviews provided a glimpse into the potential usage and demand for a nano-loan product among unbanked and low-income Haitians. Once the product rolls out in the spring of 2018, we’ll continue to work with Digicel to understand the impacts of nano-loans on users’ consumption patterns, short-term food security, and debt.
Travis Lybbert has conducted research in applied microeconomics in Haiti, Africa and India for 15 years. He has designed and conducted rigorous evaluations of many interventions in many locations with for-profit, non-profit and government partners. In Haiti, Travis has worked in the agriculture and microfinance sectors, including extensive work with the emerging MonCash mobile money platform and with various partners, including Digicel-Haiti.
Oscar Barriga-Cabanillas is pursuing his PhD in Agricultural and Resource Economics at UC Davis. He traveled to Haiti in Summer 2017 to conduct focus group discussions for the DCO-funded pilot study “Nano-Loans and CDR-Based Credit Scores: A Gateway to Financial Inclusion for Unbanked Haitians?”.
Photo Credit: Oscar Barriga-Cabanillas
By: Carly Trachtman with contributions from Ethan Ligon and Ketki Sheth
It was a hot day in Jaipur and my colleagues and I stopped in a store to get a juice in Barkat Nagar market. Barkat Nagar is a typical Indian market brimming with activity from motorcycles, auto rickshaws, cars and pedestrians lined by stalls on all sides boasting everything from books to clothes to electronics. When it came time to pay, instead of pulling out credit cards, everyone in my group started searching their wallets to the right bills with which to pay. In fact paying with credit cards would not have been an option, as this merchant was one of many in Barkat Nagar that only accepts cash payments. This reliance on cash is not a phenomenon specific to the market, but a broader trend throughout many parts of India.
Over the last two years, the Indian government has been trying to change this trend and has been engaged in shifting its cash-based economy into a digital one, but their success to date has been relatively limited. Hence, in an attempt to better understand the Indian digital payment landscape, I traveled to Jaipur last summer as part of a DCO-supported investigation, headed by CEGA affiliates Ethan Ligon and Ketki Sheth.
Given the economic diversity of India, I visited as many different markets as I could and was astounded by the differences I saw from market to market -- especially depending on the socioeconomic classes served by each. In an upper class shopping area, most merchants were able to accept credit cards and other mobile wallets for payment. Even the chai cart located outside of one of the shops had a prominent sticker reading “PayTM accepted here” (PayTM is one of the major mobile wallet providers in India). In the same market, I talked to a merchant in a larger store who sold children’s clothing. When I asked him if he was able to accept credit cards, he scoffed. “Of course I do!” When asked if it would be useful for him to receive loans based on his digital transaction history, he informed me that he had already used a lending service like this in the past and had indeed found it very helpful.
However, when I visited markets that weren’t catering to the elite, the cash economy dominated and digital payment options were severely limited. One such market I explored was in a low income area of the Shastri Nagar neighborhood. Small stalls selling everything from bags of rice to school supplies to women’s formal shoes were packed together on the sides of the street. A portion of produce sellers were organized together under a tent, whose roof was part formal structure and part cloths tied together with string. I walked up to a small stall where a young man was selling cell phones, an item that seemed to me expensive enough to not warrant payment in cash. When asked if he accepted digital payments, he told me he did not accept credit cards, but did accept payments from several mobile wallet services; however, almost all of his customers paid in cash. He explained that people in the area got paid in cash and so they paid for everything in cash. When asked about where he got loans for his business if he needed them, he replied that he would mostly borrow money from friends and family, because getting a bank loan was a long and tedious process.
These two interactions highlighted how just within a few miles of each other, in the same city, there are merchants who are fully integrated into digital finance and its related benefits (including credit), and those who are excluded. While the average penetration of digital finance across India is known to be low, this average statistic hides significant variation across markets serving different socioeconomic classes. Many of the merchants who lacked credit expressed a strong desire to get formal loans for working capital and business expansion, if there was a more accessible process. So how do we bridge this gap?
Digital credit may be part of the answer. If low consumer demand for digital finance is driving low adoption of digital payment systems, then informing merchants about the benefit of digital credit tied to digital finance may be an alternative driver for adoption. For merchants who have limited formal credit histories, a digital payment transaction history can provide proof that they run a viable business. So in this sense, digital credit can help simultaneously with two distinct goals: to increase financial inclusion for small-scale merchants and to facilitate the transition to a digital economy.
However, switching away from a cash-based economy may be a slow, tenuous process, as consumers, merchants, and intermediaries all adapt to new ways of doing business. Despite these difficulties, our research shows a potential silver lining. A recent survey of over 6,000 fixed store merchants conducted by People Research on India’s Consumer Economy (PRICE) in conjunction with the USAID funded initiative Catalyst, found that while only about 3% of merchants had reported ever receiving a formal business loan, around 96% of merchants reported having a formal bank account and around 80% reported having a smartphone. This suggests a potential need for digital credit, and that many merchants possess the technological literacy and some of the necessary infrastructure to manage digital loans. Hence, perhaps with the right push, merchants will find it worthwhile to adopt digital payment devices. This will likely make them eligible to access digital starter loans. Our team is hoping to explore this avenue further to improve financial inclusion for merchants in Jaipur.
Carly Trachtman is pursuing her PhD in Agricultural and Resource Economics at UC Berkeley. She traveled to India during Summer 2017 to conduct qualitative interviews as part of the DCO funded pilot study “Mobile Payments and Inclusive Credit: A Silver Lining in India's Cash Crisis?” This study received a second round of pilot funding in November 2017.
By: Zenan Wang
Wenge Chang is a renowned crab farmer in Sihong county. His farm has bred multiple “the king of crabs”, the biggest crab of the year in national contests. Sitting in an office full of trophies, he complained to us how he struggled getting loans to expand his production. Despite a business with CNY￥200,000 to CNY￥300,000 annual revenue and an excellent credit history, Mr. Chang found that few banks would give loans to help him expand his production because he could not provide collateral. Farmland in China technically belongs to the government and farmers are only allowed to operate on the plot assigned to them. They can rent land from others if they want to expand production, but can neither sell nor use the farmland as collateral. But for Wenge Chang that’s all the valuables he’s got, a lot of land and a lot of crabs.
For people living in the rural part of China, getting a loan from banks is not easy. Because of the national policy to support rural areas, big national banks are required to charge a very low interest rate for farmers. But evidently, banks do not make much profit from such low rates. As a result, it’s nearly impossible to get a loan from the four big national banks. Most of those banks do not have branches in rural areas. Even Agricultural Bank of China, one of the four banks whose mission was to support agriculture when it was established, pulled out of most rural markets.
Rural borrowers usually turn to regional commercial banks and credit unions. Those banks typically charge a competitive market interest rate, which is much higher than those charged by the national banks. But regional banks’ top priority is often lending to local private enterprises, which also have low chances of getting credit from a national bank, and are willing to pay high interest rates. In contrast, lending to small holders without collateral is extremely costly, plus charging high interest rates may be politically dangerous. As a result, it is no surprise to see that regional banks aren’t swarming to lend to small holders.
Chinese internet giants have identified this niche market and their fintech subsidiaries are racing to expand financial services in rural areas. By 2016, almost all the tech giants in China have declared that developing a rural finance service is one of their company’s long-term strategies. Despite pressure from their peer fintech companies, the competition from banks is almost non-existent. When a company manager told me that they are planning to provide a new type of loan that takes chicken coop equipment as collateral, I asked him how they are going to compete with traditional banks with this product (it seemed to me that banks should have being doing this and would have more of an advantage than this company). The company manager responded that they thought the same thing at first. But after they did their research and ironed out the details, he was very surprised at how easy it was to implement the idea and that no bank was doing it. It seemed that the traditional banks really lack an appetite for giving out small loans, he concluded.
However, the fintech companies will not be satisfied with profiting from a small niche market. They want to disrupt the entire financial industry. Managers at many fintech companies like to use buzz words such as “Big Data”, “Cloud Service” or “Artificial Intelligence” to inadvertently remind us that they are a technology company rather than simply another financial service provider. They are indeed bringing a new perspective to an old industry. During my conversation with managers of such fintech companies, it became clear that their ambition goes beyond making profit off small loans, but to make everything “digital”, which in turn could lead to more innovation. For example, a company provides production management software for free to its husbandry loan recipients, and subsidizes farmers to install smart environmental control systems that constantly transmit real-time data to farmers’ cell phones and the company’s data server. What might the company be able to do with such data, one might wonder? For one thing, this kind of production data could be used to detect fraudulent behaviors from borrowers, to provide technical support to farmers, or to design and offer actuarially fair insurance products. Data is the new gold of the fintech era, and Chinese fintech companies are already mining it.
While some companies are trying to collect enough data to generate a digital profile for everyone in the country, others are already thinking “why not get a profile for every chicken, every pig in the country as well?” The rise of financial technologies and data mining for digital financial services marks the beginning of a new era in which fintech companies could increasingly wield more and more power over consumers and their financial decisions.
Zenan Wang is pursuing his PhD in Economics at UC Berkeley. He traveled to rural China in Summer 2017 to interview recipients of a husbandry loan product for his DCO funded study “Access to Digital Credit and Its Spillover Effects in China”.
By: Shazeda Ahmed
Imagine if receiving a credit score were as easy as using some of the most popular mobile apps—Uber, Venmo, Amazon, Facebook Messenger. In China, fintech firm Ant Financial (a spin-off of tech giant Alibaba) has created a mobile payments app, Alipay, whose credit rating feature has issued over 200 million “social credit” scores using exactly these types of data inputs. Think of Alipay as a web browser attached to a mobile wallet: within the app, anything users can pay for digitally, from medical appointments to tuition, or car rentals to airfare, is factored into a Sesame Credit score. Sesame Credit takes traditional financial information such as bill repayment into account, alongside behavioral data (e.g., donations to charity), social network information (who one’s friends are within the app and how well they score), and demographics including education level, gender, and occupation. One common use of the scores is to qualify for consumer credit through the Alipay feature Huabei (literally “just spend”). Additionally, high-scoring users can share their scores with third parties for deposit waivers and expedited services, with examples including bicycle and car rentals, hotels, dating websites, and even visa offices at certain foreign consulates.
At WIRED, journalist Mara Hvistendahl has written a fantastic article about her experience signing up for Sesame Credit while living in Beijing. Some consumer protection concerns she raises include opacity about which transactions might raise or lower a score, and how data previously collected for a purpose unrelated to credit evaluation might later be drawn into one’s rating. An example is the list of students who were reported for cheating on the national college entrance exam (similar to the SAT), which Ant Financial has discussed using as an additional data source for downgrading scores. The company already draws from government-issued blacklists of debtors to decrement users’ Sesame Credit scores as an incentive to repay their debts. Ant Financial has even boasted that in some instances after debt repayment, there are users whose low scores plateau and never recover.
Another concern that Hvistendahl raises, and that an upcoming research project of mine will address, is the question of what data third parties exchange with Sesame Credit. Several hundred companies have entered into such agreements. As Hvistendahl notes of one car rental firm, if a user “crashes one of the rental company’s cars and refuses to pay up, that detail is fed back into his or her credit score.” Given that the number of Sesame Credit’s third party partners is on the rise—most recently, eight Chinese cities have announced that landlords will be allowed to use the scores to evaluate potential tenants—it is critical to establish security standards for these data exchanges. What other data are third party partners sharing with Sesame Credit, and how aware are users that this is occurring?
While in Beijing on a Digital Credit Observatory pilot study grant this summer, I spoke to Sesame Credit users to get a sense of their consumer protection concerns. While a few people raised issues including data security and privacy, the most interesting complaint I heard from users was that they disliked the way the app tries to mimic social media. Whereas the app wants more data about people’s social networks to assess correlations, users expressed annoyance that the app encouraged them to add complete strangers (e.g., street food vendors) as “friends” in the app. Even people who like Alipay and Sesame Credit told me that they prefer to keep their social interactions removed from the app’s assessments. Yet they might not be able to maintain this separation for much longer, as China’s dominant social media and messaging company, Tencent (owner of WeChat), is currently experimenting with its own social credit product.
While the system is relatively under-regulated, it is time for the regulators, tech firms, researchers, and users to discuss much-needed consumer protections such as limits on both the types of data that can be used for social credit scoring, and applications of these scores. Credit repair mechanisms and more detailed user-facing descriptions of the types of data that factor into the score may be the first steps toward demystifying the process.
Shazeda Ahmed is pursuing her PhD at UC Berkeley’s School of Information. During Summer 2017, she traveled to Beijing to conduct qualitative interviews with users of Sesame Credit to understand consumer protection concerns. Her study “Consumer Protection Oversights in the Chinese Social Credit System” received funding through a DCO travel pilot grant in Spring 2017.
To learn more about WeChat Pay and Alipay's plans for advancing rural financial inclusion, check out this CGAP brief "China's Alipay and WeChat Pay: Reaching Rural Users".
By: Alexandra Wall
Sitoyo Lopokoiyit’s energetic voice breaks through the conference line’s elevator music. It’s 6:00pm in Tanzania and Sitoyo kindly reminds me that he needs to run by 7:00pm sharp to attend an Iftar staff event (an evening meal ending the daily Ramadan fast at sunset). With Sitoyo’s enthusiasm and the ubiquitous background noise of an international call, I have a feeling our conversation about digital credit will be anything but dull.
The Digital Credit Observatory (DCO), launched by the Center for Effective Global Action (CEGA) last year, generates rigorous evidence on the impacts of digital credit products and the effectiveness of related consumer protection measures in emerging markets. The DCO aims to bring academic researchers and private industry players, such as Sitoyo, together to collaborate on rigorous research.
Sitoyo, the Director of M-Commerce at Vodacom Tanzania, is responsible for managing and growing M-PESA in Tanzania – which is no small task considering M-PESA has been hailed as an unprecedented mobile money success story in neighboring Kenya. But for Sitoyo, a man who has always been passionate about finding creative ways to accomplish new things, this is an exciting challenge. “I enrolled in a Masters in Information Technology Management which opened my eyes to the concept of leveraging technology for businesses. In 2006, while I was working for Chevron, I was approached about a collaboration project with M-PESA. I worked on this until the role of the Head of Department of Financial Services at M-PESA became available and I applied. The funny thing is that I had never used M-PESA personally before I started working for Safaricom.”
Sitoyo’s chameleon-like ability to work for a Kenyan supermarket chain, a multi-national energy corporation, and now a mobile network operator allows him to adapt to different business models and strategies, while constantly innovating to enhance the consumer experience. This was the motivation behind M-PAWA, a now three-year-old mobile savings and loan product in Tanzania. “The credit component of M-PAWA wasn’t the original vision; it was actually designed with the intention to drive consumers’ savings rates, and not credit uptake. We need to prioritize assisting customers with getting their money to a safe place where they can save and earn interest and then, when there’s a need, be able to provide credit. One of the main objectives of our grassroots education campaign is to grow customers’ savings behavior and to educate them on the linkages between savings and credit."
Which brings us to the next challenge: the task of rolling out a new product and educating a rural, and often financially inexperienced, population on its use. Even for M-PESA, which now has 18 million active users in Kenya, it took a lot of initial time and company effort to explain the concept of “sending money over a phone.” Customer education can often be one of the main obstacles for product growth and adoption. “When you’re explaining to a first time user how a personal financial account works, it’s all about teaching money management. There’s a list of concepts that you can’t effectively teach over a TV or radio ad. You need in-person discussions and trainings which are very expensive to deliver nationwide.” It’s a well-coordinated tango of TV, radio, and in-person campaigns mixed with the natural osmosis of product use and knowledge moving from urban to rural areas.
But the alternative of poor customer education can prove disastrous. “We’ve had cases of customers discovering that they’ve taken out a loan and were never aware of, or customers who take out a loan just because they’re curious or have heard about their friends taking out loans.”
Nonetheless, M-PAWA seems to be hitting its stride. With nearly 6 million customers, there’s been a growing demand for the product which boasts roughly 200,000 loans disbursed per month. “We’re seeing a lot of individuals who use the credit component of the product to generate personal businesses. But we’re also seeing some customers who only want to save and have no interest in the credit offerings. To this end, we’re trying to better understand our customers in order to provide more customized products. We’ve been in Tanzania for three years and we’re in the midst of building up data sources to better predict customers’ behaviors.”
“Big data”, “algorithms”, “machine learning”, and “alternative data” are the buzz words around credit scoring these days. So how exactly does M-PAWA determine if a customer is “creditworthy”? “The customer must be an M-PESA client for at least six months in order to give us enough time to generate a network history and to assign a credit score. We also evaluate the customer’s account usage and activity, airtime, payment patterns, etc. We use over 80 parameters on the Vodacom network to determine creditworthiness and we review customers’ credit scores every month.” (See M-PAWA’s FAQ section for a full list of loan eligibility requirements).
Given the hype, can digital credit and mobile money, as it’s currently used in emerging markets, ever become as common as global payment systems such as credit cards and PayPal? “I think products like M-PAWA will become very widespread primarily because the traditional banking sector excludes a large population of people from gaining access to small amounts of credit. I assure you there is no formal bank in Kenya or Tanzania that has done something close to the magnitude for the types of people we’re reaching. For me, I’m focused on targeting those 2.2 billion people who don’t have access to formal financial services - and not the middle and upper income population who are using traditional credit cards and e-payment services like PayPal.”
Sitoyo’s focus on the financially excluded is what draws him to collaborate with researchers and academic research organizations such as CEGA, whose work aims to improve the impact and effectiveness of programs and policies for the poor. In 2013, Sitoyo participated in a mobile money conference hosted at UC Berkeley (at the time he was with Safaricom’s M-Pesa division).
Since then, he has been eager to understand more about the knowledge gaps that exist in mobile technology and the importance of focusing on the financial health of the consumer. “When I first visited UC Berkeley a few years ago, it was great to meet and engage with smart researchers who enjoyed bouncing ideas off each other in such an open forum. Private telecommunications companies, like Safaricom and Vodacom, don't have all of the answers to our own questions and we want to work with open-minded people who can conduct rigorous research in areas of business, opportunity, and financial inclusion. Having academic researchers in the conversation, who aren’t focused on company profit, helps us to focus on the consumers. It’s great to meet researchers who are doing really impressive work with data and who are interested in collaborating with us.”
At the DCO’s recent matchmaking workshop in May, Sitoyo presented Vodacom’s latest product and possible research questions. Lipa Kwa M-Pesa is a new service which allows merchants, retailers, and distributors to settle payments by using M-Pesa and also allows businesses themselves to receive a credit score. “At the moment we have a lot of customers who are taking out loans to run their businesses. But this is difficult to pinpoint because the customers are borrowing from us as individuals – so we have data and records for the individual but not on the business for which they’re using the loan.” Sitoyo emphasized that Vodacom is interested in having DCO researchers help the company design, build, and test the credit proposition of this innovative product developed for small and medium enterprises. We look forward to seeing what types of collaborations ensue.