Tips on Mobile Money Product Design from Uganda

31 May 2011

How much would you like to pay to save? Zero, most likely. In fact, you probably hate the fees your bank charges for the privilege of holding your funds and disbursing when you ask for them.

So why was it so easy to find people paying huge negative effective interest rates on a recent trip to Uganda?

One man – let’s call him Richard – used an ingenious method to save for a new motorcycle. Whenever he had a bit of extra cash he wrapped it in heavy plastic, waterproofed it with industrial-grade resin, and dropped it into the petrol tank of his current motorcycle. After a year when he thought he had enough to buy a bigger, better vehicle, he had the petrol tank removed and cut open with an acetylene torch to recover the currency. The motorcycle cost USD 750. He paid USD 110 to have the petrol tank cut and buy a new one to make his old motorcycle re-sellable. That equates to negative 14.66% annual interest on his savings.

Why would he do this? Same reason a woman — let’s call her Jennifer – dropped coins into her jerry can of cooking oil. She wants to save money to buy some clothes for her children at Christmas (at which time she cut open the jerry can to get to the money).

For both Richard and Jennifer, the requirement to destroy something they owned created an effective barrier against temptation. For the poor, the most salient feature about their incomes, after the low level, is its variability. They may live on an average of USD 2 per day, but rarely see exactly that much. Some days they earn USD 10, other days nothing. But expenses for food are constant, and emergencies arise no matter how much they earned that day. The temptation to raid savings is continual and almost irresistible. Unless big barriers are erected.

Can we productize these insights to improve branchless banking services? The wallet in most mobile money is rarely marketed as a product in its own right. Most providers have styled the wallet as nothing more than a temporary holding zone for money before you send it, or the landing spot when you receive some. Wallets are designed with zero features except liquidity: if you pay the requisite fee, you can deposit or withdraw money whenever you like.

What if this liquid wallet had a twin: a store of value that was explicitly illiquid? What if Richard or Jennifer could name the point in time when they wanted their money? What if they could not touch it until reaching their goal? There might be an emergency safety valve to get at the funds, but from looking at the savings instruments that Richard and Jennifer devised, consumers might very well want the barriers to be very high indeed.

Moreover, consumers might be happy to pay for the service. Perhaps not the effective 15% which Richard paid, but 2% to 3%. That’s the same average price many people are paying already on an average mobile money transfer. P2P helps consumers alleviate the pain point of moving money over distance. Is it so crazy to think poor people might pay for a service which helps them effectively move money over time, allowing them to move current income forward to the future to finance much-desired purchases? Maybe not so crazy if we understand that a better motorcycle helps Richard earn more as a motorcycle taxi, or if we could see the smiles of Jennifer’s children when they open Christmas presents.

For those of us working to design better financial services for the poor, it may pay to spend some more time understanding their lives.


- Mark Pickens



Submitted by Ggombe Kasim Mu... on
This is an insightful blog, thanks! Indeed, the poor are willing and read to save for their development and providing commitment incentives could enhance their ability save for emergencies and the acquisition of critical assets. A purely saving product in the mobile money ecosystem is desired exactly for this purpose.

Submitted by Anonymous on
Mark, a very interesting blog post! Two thoughts: (1) A second reason for the cumbersome saving methods of Richard and Jennifer, besides creating a barrier against temptation, might be safety: Money kept at home obviously is at higher risk to theft than money kept in a bank account. But if you don’t have a bank account, your best option is to find innovative ways of hiding your money. This is what Richard and Jennifer did. Clearly, a mobile-phone-based savings account would provide the safety that Richard and Jennifer may be looking for. (2) You propose a mobile-phone-based savings account that would discourage withdrawals in the same way as Richard’s and Jennifer’s savings methods did. The implementation of such an account might, in fact, be straightforward: Enable convenient deposits via a mobile phone, but require a visit to a bank branch in order to make a withdrawal. This solution would make it reasonably difficult to withdraw money, while still allowing emergency withdrawals.

Submitted by Anonymous on
I agree with Alex that there are certain “must haves” in any mobile money service: your platform must be stable on the back end and usable on the front, agents must be aggressively recruited, money must be spent on marketing, obviously regulations must be complied with. A sizable portion of mobile money plays which have not achieved volume are guilty of poor execution on one or several of these fronts. But, how do we explain those implementations that do get the prerequisites right, and still don’t see big uptake from consumers? Increasingly I suspect there is weakness in the product. Put differently, basic P2P and liquid wallet is not proving interesting enough to a sufficiently large slice of consumers. This is my leading candidate for why 78 out of the 104 branchless banking services live in the world today have fewer than 1 mil registered users (not active, just registered). To attract enough consumers, providers might need to appeal to several segments by offering 2 or 3 initial products. This might include the basic P2P/liquid wallet, but also a saving product, or other that market analysis shows demand for. Should this be surprising? It would be just on more thing about M-PESA’s success that’s peculiar to Kenya and needs to be adapted to make mobile money work in other markets.

Submitted by Anonymous on
Mark, I like the way you have articulated the notion of moving money over time relative to moving money over distance. Good challenge you raise. Sadya, my first thought when I read this was that they were already intrinsically motivated to save so the answer would be a reasonably priced, sustainable instrument that enables them do so. What I mean is that the future reward for these particular folk seems clearly defined, and what they want, most of all, appears to be that the barrier that stands between the press of immediate needs and that future goal. Which suggests to me that what they need most is a trustworthy accountability partner in working towards a goal that they are already very determined to achieve. My two cents.

Submitted by Anonymous on
A nice illustration though let me point out some few observations. A nice platform, aggressive agent acquisition and a successful go-to-market strategy are the key things to ensure the successful launch of any mobile financial inclusion project. Mobile money transfer and mobile banking are two different things. Mobile money transfer has evolved into mobile financial services, but many people don’t distinguish between MMT/MFS and mobile banking. The amount of money/time wasted in funding the development of new platforms, if channelled to funding existing mobile money schemes we could have made very big strides in bringing the under-banked population closer to accessing financial services. Rollout technology is very critical if the target market is to reach and win the less banked population. Surely telling my grandmother to download a java applet on her phone so as to enable her receive menu which will make her receive the $3 I sent to her… Other key things to consider are how friendly the regulation is. So next time before one jumps onto that plane to Dubai/New York/Miami or Amsterdam for a mobile money conference, one should stop and think who really is our target customer?… Perhaps one should wear mud boots and venture into Kibera slums in Kenya/ Koko village in Delta state in Nigeria and witness how a simple and easy to use STK based platform can change millions of people lives…

Submitted by Anonymous on
hanks, everyone, for the first crop of comments. Kevin, I definitely thought of “Nudge” after talking to these clients in Uganda. Sadya, definitely worth looking at incentives vs. penalties. For mobile network operators, airtime has an effective marginal cost close to zero. I wonder about matching client deposits with a % of airtime? Jake, your comment points at something quite important: branchless banking providers don’t have to go it alone with product design. Nonbanks can partner with banks better positioned to offer deposit or credit products, for instance. But we find that often other players which might have more regulatory permission to offer other products can also be missing insights about what low-income consumers want. There is definitely work to be done generating insights about consumers and converting those into innovative products to test.

Submitted by Anonymous on

Submitted by Anonymous on
Hi Mark, Interesting post. I think your examples are nice illustrations of what people will do/pay in order to constrain their own spending and sock away value for a later date. It will be interesting to see how mobile money will evolve to meet client needs, whether they be needs to constrain spending or other needs. Mobile moeny players may try to develop commitment and other more tailored product features themselves. I would think that mobile money providers will eventually begin to see their role as a platform provider, linking insitutions offering these kinds of services to clients in a way that benefits both sides. As our recent paper documents, this may already be happening in Kenya. Thanks again for the post, Jake

Add new comment