Who Should Pay for Banking Infrastructure in G2P Programs?

The transition from cash to electronic G2P payments has been getting a lot of attention lately, and rightly so. Conventional wisdom tells us that cash is more costly, less reliable, and more susceptible to fraud. The evidence base to support these claims is slowly growing, but that hasn’t stopped governments and donors alike to push for this switch.

Cocadas Vendor in Colombia

Photo Credit: Ingrid Bonilla Rodriguez

From the recipient standpoint, though, an electronic payment is only as good as the location where the electronic value can be converted back into cash. Most developing countries are operating in an ecosystem that is far from cash-lite, especially in the communities where G2P recipients tend to live. While research shows that recipients value the convenience of receiving an electronic payment, by and large they still want to feel the cash in their hands at some point along the way. This is largely because they are unable to conduct their day-to-day transactions in anything other than cash due to the lack of infrastructure necessary for electronic transactions.

In those countries where distribution networks to convert electronic value to cash already exist (as in Brazil before Bolsa Familia began its large transition to electronic payments), this is all well and good. But what about countries that have no existing agent distribution network or lack a strong banking infrastructure of ATMs and point of sale devices?

The question then arises, who is responsible for building up this infrastructure? Is it the government? If so, then which ministry in the government? The Ministry of Social Development, the Ministry of Finance, the central bank? Or should the provider be responsible? What incentives will push the provider to make the necessary up-front investments of this kind?

To begin to answer these questions, CGAP along with Marulanda Consultores carried out a study of the experience in Colombia of promoting an agent network through a government incentive program of subsidies. In August 2006, over 300 of Colombia’s 1,100 municipalities had no bank presence. At the same time, the government was expanding its conditional cash transfer program Familias en Acción to all municipalities. The dilemma was clear: how should these payments be delivered to rural and hard to reach municipalities that lacked any banking infrastructure?

To begin to meet this need, the policy body Banca de las Oportunidades (BdO) designed three subsidy schemes to incentivize payment providers to build out their agent networks in these challenging locations. The target of the first tender was the 128 municipalities with fewer than 50,000 inhabitants. Banca de las Oportunidades guaranteed a minimum number of transactions at each agent location over a period of 3 years to give the agent time to build up transaction volume on the path to sustainability. The subsidy was provided on a decreasing basis whereby in the first year BdO paid the difference between the transactions made at the agent and those guaranteed at 100% of the price, at 50% the second year, and at 0% the third year. A simple auction system was used and the incentive was awarded to the bank that requested the least number of guaranteed transactions. In the case of the first tender, only the state bank Banco Agrario presented a bid and therefore won.

The second tender targeted the remaining 67 municipalities that were even smaller and more remote than those covered in the first. The same system was used of guaranteeing transactions, although the guaranteed number covered by BdO was increased due to the more challenging locations. The subsidy also spanned 4 years instead of 3 since more time was assumed to be needed by agents to reach break-even on their own.

Finally a third tender was launched to cover marginal neighborhoods in 5 of the major cities in Colombia. A different subsidy model was used this time based on co-financing between BdO and the bank to cover the costs of installing an agent, instead of using the guaranteed transaction model. Co-financing was also set on a decreasing basis over 3 years.

So were all of these incentive schemes successful? Well in one sense, yes. Today 99% of all municipalities in Colombia have access to financial services. The investment of public resources totaled $2.5 million and increased coverage to 187 rural municipalities and to 25 marginal neighborhoods in the country’s largest cities. Familias en Acción payments were able to be made more efficiently and closer to where the beneficiaries lived and today account for almost 40% of the transactions over the agent channel.

But the research also shows that incentives alone were not enough to stimulate private financial institutions to develop business models to reach remote rural communities. The lack of participation in the tenders by private banks indicated that even with the offer of subsidies, many banks were still not ready to make this investment in building an agent network. It is possible that the incentive scheme came too early in the development of the branchless banking market in Colombia. Over time as the subsidy programs were implemented, banks became increasingly interested in building their own agent networks, and now multiple banks have increasing networks of agents throughout the country.

But the Colombian case does provide an informative example of how G2P programs can be leveraged to build the banking infrastructure of the country, and how the government might think about supporting such a roll-out.  You can read a lot more detail about this incentive scheme in the report in Spanish here and in English here.

Beatriz Marulanda heads Marulanda Consultores in Colombia and Sarah Rotman leads CGAP’s work on G2P.