First of all, let me reaffirm my belief that digital financial services (DFS) really do have the potential to provide financial inclusion for many (but perhaps not all) of the 2 billion un-banked. Its potential to reduce the costs of providing a range of financial services is an exciting prospect for those of us who have long critiqued the monoculture of microcredit. And the potential of DFS is already being hinted at in recent studies – most recent, “The Long-run Poverty and Gender Impacts of Mobile Money” by Tavneet Suri and William Jack, is particularly encouraging:
“We estimate that access to the Kenyan mobile money system M-PESA increased per capita consumption levels and lifted 194,000 households, or 2% of Kenyan households, out of poverty. The impacts, which are more pronounced for female-headed households, appear to be driven by changes in financial behavior—in particular, increased financial resilience and saving—and labor market outcomes, such as occupational choice, especially for women, who moved out of agriculture and into business.” – Tavneet Suri and William Jack
However, DFS is currently facing two big challenges that have the potential to undermine all the progress made to date:
1. Digital Credit
Access to small amounts of credit with a few key strokes can be immensely important and valuable for people facing short-term cash-flow problems or emergencies. So digital credit meets important demand – as the enormous uptake of such products in East Africa have demonstrated. But it does carry important risks for consumers and the industry as a whole.
With a few honourable exceptions, notably Equitel’s Eazzy Plus loan (which draws on Equity Bank’s customer transaction data) and the KCB-M-PESA loan, most digital credit products currently on offer bear all the hallmarks of consumer or pay-day lending. Their annualised percentage interest rates contain high risk premiums. This is understandable in that, for all the hype, low-income people leave very shallow digital footprints because of their limited use of mobiles – for calls, messages, or sending money. But few digital credit providers reduce interest rates for those customers who do build up deep digital footprints through high levels of activity on the network and/or regular on-time payment their loans. Instead, they are rewarded with offers of higher loan amounts at the same, risk premium-inflated, interest rates.
For the leading players at least, these interest rates do compare (sometimes even favourably) with the 10% per month interest rates typically charged by the informal sector lenders in East Africa. But I am not sure that this is the right benchmark to which we should aspire.
CGAP and others have also been raising important questions around customer protection measures as well as the opaque terms and conditions surrounding digital credit. The Competition Authority of Kenya (CAK) has taken important steps to ensure that: (1) borrowers are presented with key elements of the terms and conditions as part of the process of accessing their loan on their mobiles; and (2) lenders provide both negative and positive reports to the reference bureaus active in the country.
Nonetheless, the growing numbers of clients blacklisted for outstanding loans of >90 days on the credit bureaus is a matter for profound concern. After alarming newspaper articles on this, MicroSave called the leading credit bureau in Kenya and was told that the reports were correct. The bureau has about 10.6 million listings for digital credit borrowers, out of which about 2.7 million are negative. We do not have the full data set to corroborate these numbers, but there seems little reason to doubt them, given their source.
This would suggest that digital credit has already resulted in the effective financial exclusion of 2.7 million people in Kenya. And that, if this trend persists, providers of digital credit will soon run out of borrowers.
It is intriguing to speculate as to why so many people should fail to repay their typically small digital credit loans. Part of the problem may lie in borrowers’ lack of ability to pay, and part in lack of understanding (despite the efforts to clarify terms and conditions and use of SMS reminders to encourage people to repay on time) … but I would suggest that an important part lies in the ease of accessing these loans. And that some digital credit providers actually push these loans at potential borrowers.
I wonder if providers (or CAK) need to encourage user-defined friction to address these latter problems, in an attempt to avoid the ill-considered, non-essential loans that are taken out, particularly in bars on Friday and Saturday evenings. Users could, as part of the initial registration process, define additional criteria such as the credit being only available after a day’s cooling off period (during which the application could be cancelled). To provide for emergencies, when the one-day cooling off period could present a real problem, a user-defined, second PIN could be added to override this provision. To be effective, this type of product engineering must be developed through careful research into consumers’ needs, perceptions, aspirations and behaviour – as outlined in The Helix’s training on Product Innovation and Development.
We need to understand the consumer experience and what behavioural levers might be used to nudge them towards better self-protection much better than we currently do. In addition, as AFI and CGAP have already flagged, additional regulation and oversight is needed, given the burgeoning of digital credit and the challenges it presents to regulators. It was, therefore, excellent to see the ITU working group on Consumer Protection in DFS make such a robust set of recommendations.
2. The Trust Deficit
DFS depend on customers’ trust in the systems and agents used to deliver the service even more than in traditional financial services. Protecting the customer and minimising the risks as he/she uses the service is essential to build and maintain that trust. And yet, in most of the 20 markets where MicroSave has worked on DFS, a combination of service/system downtime, agent illiquidity and churn, complex USSD/SMS-based user interfaces, and poor customer recourse means that there is limited trust in DFS. As a result, many users prefer to perform over-the-counter (OTC) transactions rather than take the risk of keeping money in their wallets or sending it to the wrong recipient. We often hear that financial education or marketing will overcome these barriers – but I suspect that this belief is over-optimistic.
Many of these issues require providers to invest more in their platforms and agent networks in order to build a comprehensive and thus profitable digital ecosystem, beyond an OTC-based and payments-dominated one. This is a “hygiene factor” pre-requisite for transforming DFS into real financial inclusion and rolling out a range of services.
Smart phones may provide the key to realising the full potential of DFS ecosystems. Smart phones allow us to build interfaces that are much more intuitive for the end-user, and to reflect the mental models that poor people use to manage their money.
Zooming out to see the big picture
Ultimately we need an over-arching strategy to address the two inter-related challenges of trust and digital credit. We need to encourage low-income people to maintain and use digital liquidity, thus deepening their digital footprints and allowing digital credit providers to reduce their interest rates. G2P and other bulk payments will be key to creating digital liquidity – some important work is underway on this already. Digitising a wide variety of value chains and procurement/distribution platforms as well as retail merchant acceptance ecosystems will also be essential to keeping money digital and circulating it.
But the success of these efforts will depend on addressing the hygiene factors that currently undermine trust in digital financial services. Only this will lead to a trusted, convenient and efficient digital ecosystem, with increased volumes, reduced costs, increased providers’ profitability and, ultimately, deeper digital footprints to inform digital lending operations.