Mobile money in emerging markets: is the operator-led model broken?

The runaway success of early operator mobile money products in East Africa held out a future for mobile network operators (MNOs) eager to generate new revenue streams, beyond voice and data by boldly grabbing the public mind and private purses of previously low and UN-banked emerging market society. These services have become a much-referenced ‘gold standard’ of not only what mobile technology and financial services can achieve in emerging markets.

Four years on, with over 200 mobile operator mobile money services around the world, how successful have these services been?

Research by the Mobile Money Unit of the GSM association (GSMA MMU) suggests that while the MNO’s have enjoyed varied levels of success with mobile money in emerging markets, their overall impact and success remains a bit of a mixed bag. The survey sampled 52 MNO mobile money services from the 150 services around the world. It included 11 in East Africa (including Safaricom), 13 in West Africa, 15 in Asia/Pacific and 8 from Central Asia and the Middle East.

I would strongly recommend reading the survey for yourself — it makes for fascinating reading because it brings into sharp relief several elements that should cause every mobile money executive to sit back and take a serious look at what they are doing. It highlights three issues I have long suspected — first, there is a vast difference between the number of ‘registered’ customers and ‘active’ customers — of all the services surveyed, only two could claim more than a million active customers (active defined as a transaction on that account in the past 30 days). Second, a staggering 80% of all transactions processed worldwide, according to the survey sample, were processed in East Africa. The survey further identified 8 of the fast-growing mobile money services enjoying a similar (if less spectacular) growth path to Mpesa. Six of these eight services are in East Africa, and after Kenya, Tanzania, and Uganda, the mobile money wave is now reaching Rwanda. Third, 68% of all mobile money transactions were airtime top-up transactions.

So what does this tell us (if anything)? Well, aside from Mark Twain’s cautionary aphorism about lies, damned lies and statistics, we can broadly see that East Africa’s contribution to global mobile money figures are vastly disproportionate, and has skewed perceptions of global mobile money “success”. It also tells us broadly that that airtime recharge, and not actual financial services (i.e. payments, transfers and purchases) is what drives most of the activity on those mobile money accounts.

But what about Mpesa and the other fast growing services in East Africa? Can’t they tell us something? Four years after its launch, Mpesa’s runaway success has yet to be repeated (although in fairness some have come close). The odds of repeating an Mpesa-clone that enjoys similar success outside of East Africa is unlikely because of the specific market conditions, so referencing and modeling Mpesa when comparing what is happening in, say, Fiji is unhelpful.

So why are so few operator mobile money services shooting out the lights, given the scale and reach to customers that most operators enjoy? The answers are many and varied depending on who one speaks to, but can best be summarised by a few observations: firstly, operators mobile money services are not bank accounts and operators are not banks. The observation may appear trite and obvious, but sometimes, like Occam’s razor, the most obvious explanation is often the right one.

While operator stored-value accounts may emulate some elements of a bank account, with some ‘bank-like’ functionality, (eg. the ability to pay), an operator stored value account and a bank account are very different beasts, defined by different rules and regulations that govern what they can and can’t do.

As a consequence, the business model for operator led mobile money does not, in my view, necessarily stack up for operators. The reason is that legislative compliance has ensured that all operator mobile money services (including Mpesa) keep an equivalent value of the operator stored value, on deposit in a trust account at a bank or banks. For example, if an operator holds US $10 of account holders money in a operator hosted wallet, it must have US $10 on deposit in a licensed bank’s trust account (similar to an attorneys’ or accountants’ trust account).

This means the operator cannot access (or leverage) those funds either as working capital, nor can it derive any benefit in the form of interest or access to surplus. Very different from a bank whose core business is taking in deposits and lending money out.

Regulation also doesn’t generally permit operators to earn interest or charge operating costs or expenses to those funds (other than perhaps off-setting a few bank-charged fees, and so on), as they are held in a trust account.

Consequently the mobile network operator derives little benefit from holding those funds, other than as a ‘value-added service’ from which the operator derives fees and commissions (a so-called ‘clip-of-the-ticket’ on transactions) for providing the service.

For some operators, this still manages to be a large sum. This transaction revenue contributes over 30% to Safaricom’s bottom line. For most, however the situation is less glamorous. At the lower end of the scale, an emerging market MNO in Asia operating a mobile money service with a registered base of just under 400 000 customers barely generated $100 000 in account activity, over 18 months (activity, not revenue for the operator, mind).

Could new technologies such as NFC move the dial and increase take-up? I sincerely doubt that the issues are technology, marketing or environment related. Its fair to say that customers in an emerging market don’t buy the technology — customers buy what technology can do for them.

This is not to suggest that existing mobile money models are fatally flawed or that operators should not provide these services, or work actively to grow this space (various flavours of large and profitable mobile money services within the current ecosystem are testament to this). On the contrary, mobile operator money services are good for operators, driving operator airtime sales, lowering airtime distribution costs and driving incremental revenue — and that is good for customers.

However, given that we have yet to see a single, repeatable operating model for emerging market mobile banking that can be rolled out successfully across markets, we can reasonably conclude that the time has come for a fundamental rethink on the role each ecosystem players holds in delivering services to the underbanked and unbanked in emerging market economies.

In my next article in this series, we will look at whether bank and operator interoperability is the next step that will change the face of mobile money, and whether there is sufficient compelling incentive for banks and mobile network operators to move closer together.