Very small loans/ very big ideas: microfinance in developing markets

I’ve been lucky enough to have been on great business trips, but one of my most memorable was for a project in Ghana – not least of all because the fact it abutted an already-booked family trip means I must be one of very few people who’ve ever flown into Accra with a suitcase full of skiing gear!

I was there with Experian and Felix Duku, helping a local commercial bank to prepare for its first foray into retail lending. Although I’d already delivered projects across Africa at that stage, this was the first time I got to work on a lending business from day not-yet-launched and that was exciting.

A key lesson I took home with me, after that skiing holiday, was that once you strip away the complicated maths of scorecard building, the logic behind consumer lending strategies remains stable in almost all conditions. Albeit sometimes the order of priorities changes.

it's really not that the scoring or anything would be different, right? It's not that you use different variables... but when it comes to responsible lending, the whole credit underwriting process has to have that mindset.

Which is a sentiment that Joffre Toerien of Credit Insight Analytics echoed when I spoke to him in episode 5 of How to Lend Money to Strangers. Unlike me, though, Joffre is very familiar with the challenges of building models for the microlending industry, so we chatted about his career as well as the consumer lending ecosystem in Georgia.

My project in Ghana must have been a decade ago when useful lending data was scarce, so it was great to hear that situation has improved — in part by embracing alternative data — that situation has improved. That said, data is only as useful as a lender’s capacity to use it, and that isn’t always as easy to scale up. As a result, Joffre likes to bring a lot of pragmatism to his projects, favouring projects that are implementable above those with claiming grand but impractical upsides.

Sometimes you have to put in place the boring foundations now, so that in eighteen months time you are actually in a position to implement the exciting project you wish you could be implementing now.

And one of the most common forms he sees this taking is with lenders who overlook the valuable data that comes with existing customers. If microlenders don't have at least a separate process for new and existing customers, that's almost always the place he starts. And I’ve seen this play out, too, with lenders investing in scorecards for new-to-bank applicants while ignoring the fact that they have a large portion of their customer base who are dormant but also far-easier-to-model and with an evidenced desire to interact.

And it's difficult to test new clients, but for existing clients you can just go with the Chief Operating Officer to a branch, talk to the loan officers about the clients, you'd be surprised how many they know by name, and test the score’s suggestions with them.

If you’d like to hear more about this, you’ll find that conversation and all other episodes of How to Lend Money to Strangers on just about any podcast player, or via the links on my page.

#microfinance #creditrisk #scorecardbuilding