[NB: this post was written before I was affiliated with FrontlineSMS:Credit. Just goes to show that blogging can occasionally land you a job!]
I’ll be candid: I am really quite excited about the idea behind FrontlineSMS:Credit. Microcredit has seen a lot of innovation on the funding side, and some on the product design side, over the past 30 years, but service delivery has remained virtually static throughout the world: loan officer, 10-30 clients, group repayments, wash rinse repeat next week. A secure, reliable, open-source platform for making mobile payments would be the first game changer in service delivery in the history of the industry.
What I find particularly intriguing here are questions of how product design might have to change in order to accomodate mobile payments. Specifically,
- Group structure. The predominant group lending methodology in microcredit cuts both ways: some of FINCA’s clients in the DRC found the social support (or pressure) of group lending very helpful, whilst others considered it burdensome. Mobile payments would obviate the logistical necessity of weekly repayment meetings, but MFIs will have to think carefully about how to maintain the social benefits of the old group structures for those who valued them, whilst also seizing the opportunity to expand services to new clients who were perhaps turned off by the group method in the past. (For some insightful observations on group vs. individual loan structures, see this paper by Karlan & Gine, who find that group loan clients in the Philippines who are randomly switched to individual loans maintain the same repayment rate.)
- Repayment schedules. The challenge of managing an irregular low income is one of the main themes of Portfolios of the Poor, and the authors aptly point out that inflexible repayment schedules prevent many potential microloan clients from accessing this type of formal financial service. Mobile payments have obvious potential for facilitating flexible repayments. The question, then, is how one might design repayment plans that are sufficiently flexible to attract customers with varying cash flow management needs, yet sufficiently rigorous and verifiable for the MFI’s own cash flow management. (Karlan again has some good ideas for this. My suggestion: let customers pick among repayment schedules varying from daily to monthly.)
- Predictability & transparency. Another key takeaway from both Portfolios and the FINCA clients I spoke to this summer concerned the desirability of predictable and transparent financial services. FINCA consistently won praise from clients for being “serious” – that is, for having loan officers show up on schedule, for reliably disbursing funds, and for not stealing clients’ money. They knew what to expect, and they got what they expected. Mobile payments bode very well for the transparency front – probably much better than the paper-based systems that many low-budget MFIs are still using, in fact – but implementing MFIs will have to do some serious work to prove to clients that mobile payments & disbursements are more reliable than the in-person variety. (I think the question of theft is also an interesting one. Several FINCA clients said that they had been robbed whilst coming out of the bank after a loan disbursement – an obvious risk of dealing in cash – but, anecdotally, cell phone theft seemed much more common among clients than theft of cash on a day to day basis. You’d have to build a system that’s robust to these risks before it would truly be useful to clients.)
But what a fascinating set of challenges overall! I’m very much looking forward to following the development of FrontlineSMS:Credit.
Congrats on the new job, looks very interesting. Mobile transactions are the future!
Thanks! I’m pretty excited. : )
Hi Mark! Donor perception is a good question. I’m not working for CreditSMS, so I don’t know anything about what their interactions with donors or other funders have been like, but I would imagine that the combination of the popular approaches of appropriate tech & microfinance, as well as the real potential for increased efficiency of transactions, is really drawing some attention. It’s a good historical moment for moving into widespread electronic transactions & record keeping in microfinance, I think. M-Pesa’s proven it’s possible, and the variety of MISes coming out recently are an encouraging step towards better records – and CreditSMS could bridge the gap between these two technologies nicely.
(Also, your comment about it being an uphill battle to introduce tech into some MFIs is right on. FINCA-DRC had a whole bank of computers used by loan officers to write their reports, but they weren’t connected to the internet and had had their USB ports removed to prevent anyone from introducing virus-filled flash drives, so they were basically expensive typewriters. Just having the tech doesn’t guarantee that it’ll improve productivity!)
I would be interested to hear something about donors’ perspectives of credit SMS. As an individual donor I specify that MFIs maintain detailed accountability for all loan activities. The model of loan officers managing peer groups, face to face, although costly, is a minimal level of supervision. Once we get past trusting the loan officers we can go forward with funding MFIs. With SMS introduced into the model we have a entirely new scenario for verification and due diligence. If SMS can be implemented reliably, both in reality and in perception, the cost reduction could be a really good thing. The MF clients that I have met all live analog lives and it seems like an uphill battle to radicalize the process with high technology. Do you feel that the donors are on board with credit SMS?