In 2018, Boston Consulting Group studied the economics of mobile money agent networks and their role in advancing financing inclusion. This report addresses the following key questions:

• What are the economics of mobile money networks for both individual agents and the providers such as banks and MNOs who develop them?
• Where is the agent model sustainable—and what factors limit its ability to reach the rural poor?
• What actions can private sector, governments, and philanthropies take to cultivate and support these agent networks for the purpose of financial inclusion?

Drawing on a range of research and analyses, we explore why so many people still lack CICO access. Looking across seven countries (Bangladesh, India, Indonesia, Pakistan, Kenya, Tanzania, and Uganda), we find that about ~85% of the coverage gap is in rural areas and ~40% is in remote “rural frontier” areas. Geospatial analysis suggests that about 500,000 new agents would be required to deliver universal CICO access in these countries, with 90% located in rural areas (and two-thirds specifically in the rural frontier).

However, challenges such as the limited local consumer base and distance from bank branches mean that few new agents would currently find it economically or operationally viable to operate successfully. Achieving universal financial inclusion objectives will therefore require government and financial-inclusion stakeholders to make a broad range of interventions, from direct economic incentives to regulatory reform and investments in infrastructure.

Looking at the seven focus countries, significant variations and nuances emerge. Populations in Kenya and Uganda have significant CICO access, but large gaps remains, mainly in rural frontier areas. India and Indonesia have a particular challenge in rural frontier areas, where the gap is immense (respectively 65% and 73% of the population in rural frontiers lack access to CICO services in these countries). For other countries (Bangladesh, Pakistan, Tanzania), opportunities exist to expand in “dense rural” and “rural oases” areas in addition to “rural frontier” areas.

However, while some countries (such as Bangladesh) have high population-to-agent ratios in rural areas, which could provide a viable market for CICO models, the low population-to-agent ratios in other markets (Tanzania and Kenya), especially in the rural frontiers, make long-term economic viability uncertain without external intervention or the development of alternative CICO distribution models such as roving agents.

Achieving rural CICO coverage will demand a broad range of interventions from government and key stakeholders, including regulations that support agent network expansion, large-scale investments in infrastructure, direct economic incentives, and better data to improve CICO networks’ efficiency and effectiveness.

Exploring direct economic incentives governments can use in the short term to support rural growth of CICO agent networks, we have identified 12 specific levers that fall into 3 main categories. These include: (1) inclusion guidelines to “push” providers (including banks, mobile money agents, and financial services companies) to expand into rural areas, (2) mechanisms that can increase agent profitability (such as preferred loans, research and innovation grants, favorable government contracts, and tax rebates), and (3) demand generation levers that can build long-term consumer demand for digital financial services.

Based on 55+ global case studies and 30+ expert interviews across regions and sectors, our research reviews initiatives that aim to expand: (1) CICO services in developing markets, (2) other ‘basic’ services – such as utilities and telecoms – also in developing markets, and (3) financial services to low-income/remote populations in developed markets.

These case studies reveal that a combination or sequencing of levers can be effective, and that collaboration between government and providers will also be important to ensure that incentives are operationally feasible, cover on-the-ground costs for providers and agents, and align with providers’ strategic priorities.

While CICO economics are viable in Nigeria’s urban, peri-urban, and rural oases – where transaction volumes provide agents with sufficient revenue – in rural frontiers, transaction volumes are below breakeven point. Based on current conditions, the country’s existing agent model could be scaled up to reach 51% of the adult population.

Further expansion would, however, require interventions such as the digitization of government-to-person (G2P) payments, subsidies for agents, and the implementation of roving agent models. However, avoiding any unintended consequences arising from these interventions will require greater stakeholder engagement and robust evaluation of feasibility, impact, and costs.

In Indonesia, significant agent network expansion has helped deliver CICO access to 87% of the population. However, coverage does not reflect the network’s viability. Low transaction volumes mean agents make little profit and providers generate insufficient fees to cover their costs. Meanwhile, many factors constrain demand, and efforts to increase demand (such as digitization of government-to-person payments) are not changing the economics. Other challenges include agent inactivity and account dormancy issues (i.e., lack of use).

These operational and economic challenges offer little incentive for agents or providers to invest in expansion, leaving about 34 million people without access to CICO services – especially in islands outside Java. This gap may widen as the divide between those with and without access to the digital economy expands. Regulators could help address this by improving the viability of existing agent models and by supporting the extension of high-potential models, such as e-commerce and e-money models, into under-served areas.