Mobile money agents who register customers and handle cash deposits and withdrawals are central to most digital financial service networks throughout developing countries. As such, their role is crucial for progress on digital financial inclusion, including providing financial services to some of the world’s 2 billion ‘unbanked.’
So what happens when mobile money agents don’t check customers’ identification documents properly? How do agents respond when a customer wants to send a transfer over the prescribed limit? And how transparent are the fees when conducting a transaction with an agent?
These questions have important implications for regulators and supervisors charged with overseeing digital financial services.
Identification requirements and transaction limitations help prevent money laundering and terrorism financing risks, while fee transparency is a basic consumer protection to ensure consumers have all the information needed to make an informed choice about financial services available to them.
To address these issues and more, I worked with the telecommunications regulator in Zambia, ZICTA — and on behalf of the ITU’s Focus Group on Digital Financial Services— to conduct a ‘mystery shopping’ study in order to help assess agent behaviors and regulatory compliance in the Zambian market.
Mystery shopping is a powerful tool for regulators and supervisors to use for monitoring their markets and enforcing regulations. Attaching fines for providers onto the results can encourage provider compliance, and if mystery shopping is done regularly, regulators can measure improvements. Using mystery shopping and other consumer research techniques is one of the key recommendations to come out of the Consumer Experience and Protection Working Group, as part of the ITU’s Focus Group on DFS.
For the study in Zambia, we sent mystery shoppers into agent shops with specific instructions to test seven different scenarios, such as registering without an ID or sending an OTC transaction over the limit, across 300 mystery shopping visits. The results provided insights into areas in need of greater monitoring and enforcement, and helped ZICTA and the Zambian Central Bank think through new regulatory approaches needed in the market.
Out of 20 shoppers attempting to register for a digital wallet without the needed identification, 14 were able to do so. And this appears to be a common problem outside of our mystery shopping study: when we interviewed agents, separately from the mystery shopping visits, many said they have customers daily trying to register without an ID. This has serious implications for the effectiveness of the very-important ‘know your customer’ (KYC) process.
In addition, Zambia has a rule that no individual may have two wallets with a single provider (to prevent bypassing wallet transaction limits). In our mystery shopping, 11 out of 15 shoppers who already had a wallet with the same provider were able to sign up for a second wallet. One agent got around the rule by using a wrong name and ID number: “The agent used abbreviations when entering my name and used a wrong ID number, because I already had a mobile money account with the provider.”
All of the shoppers attempting to send OTC transfers over the prescribed transaction limit were able to do so successfully. Agents recommended a variety of creative techniques to help shoppers bypass the rules. One shopper was told to have a customer behind him in line conduct a transaction for him, and many others were told to make two separate transactions. One mystery shopper was told to accompany the agent to the bank where the agent would get additional e-float to complete the large transaction.
66% of agents visited had fee charts displayed in their shop, though only a fifth had printed brochures that a shopper could take with them (something that is especially helpful for illiterate users who could have a family member explain the fees at home).
Verbal disclosures of fees were provided inconsistently. For transactions conducted with an agent (such as withdrawing money or sending money over the counter), a third of shoppers were never informed of the fees associated with the transaction. For another 20%, the fees were disclosed only after the transaction was completed. In addition, far more agents quoted the wrong fees than quoted the correct fees. According to one mystery shopper, “I was initially informed the transfers attract zero charges, but I later incurred charges after the transfer was complete and the agents were unable to explain this.”
This indicates either poor training of agents, such that they do not know the correct fees to disclose, or poor oversight by providers to ensure their agents are consistently providing accurate information to customers.
One of the most common challenges mobile money users face is sending money to the wrong recipient (e.g. inputting the wrong number in the mobile money menu) and trying to retrieve the funds. In Zambia, we tested how well agents were trained on dealing with this common issue.
37% of agents with whom we tested this scenario referred the mystery shopper to the mobile money provider’s customer care number, an appropriate response to let the trained customer care staff address the issue. Another 26% of the agents proactively tried to contact the (incorrect) recipient and ask them to return the funds, meaning nearly two-thirds tried to help in some way. A quarter of the agents told the shopper there was nothing they could do (the remainder gave other various responses). The helpfulness of agents is a positive sign, while the varied responses could indicate a need for further training, so that agents know the correct way to direct a customer in need of help.
Having problems resolved becomes more complicated when multiple parties are involved in the transaction. Paying a bill through mobile money, for example, may involve the mobile money provider, the mobile network operator (if different from the mobile money provider), the agent (if conducted over the counter), the utility company receiving the bill payment, and additional third parties that facilitate the transaction processing. When something goes wrong with a transaction, it can be unclear to customers where to turn for help and, in some cases, even unclear to the providers who is responsible for resolving a problem.
Of 16 mystery shoppers who sent a bill payment to the “wrong” account as part of their scenario, none were able to retrieve the funds. Twelve were told to contact the mobile money provider’s customer care, 6 were told to contact the utility company customer care, and the overlap were told to contact both. In addition, three shoppers were told their money would be returned within hours or days, and none actually received the promised reimbursement. Clearer, more effective recourse channels are clearly needed as more complex products come to market.
These findings point to areas of immediate concern for regulators, such as poor enforcement of ID requirements when registering. They also point to areas, such as agent training and recourse options, where regulators may want to consult with providers or strengthen regulatory requirements. These kinds of insights, which can be gathered through mystery shopping, are essential for regulators to understand what is happening on the ground in their markets.