The Poor Need Better Payment Services

Poor people’s need for domestic payments is often large, whether it is spurred by migrant labor remittances, informal support networks of friends and family, government welfare payments, or utility payments. It’s also something they are willing to pay for, partly because the alternatives for moving cash over distance – e.g., informal bus networks and hawala couriers – tend to be slow, expensive, and risky. Hence, the ability to transfer money securely and instantaneously can provide a big enhancement to a household’s financial toolkit.

Over the past few years the Financial Services for the Poor team here at the Bill & Melinda Gates Foundation has launched multiple studies to assess existing payments and money transfer options available to poor people. Results from studies of remittance senders conducted in India, Kenya, Tanzania, the Philippines, and in post-earthquake Haiti, show:

  • The poor are moving money intensively and in some countries many more people are sending and receiving domestic remittances than international. Preliminary results from questions we had added to the 2011 Gallup World Poll in Tanzania and Kenya show that while only 5-6% of the adult populations in either country had sent or received an international remittance, 57% of Kenyans and 24% of Tanzanians sent or received a domestic remittance from family or friends in the last 30 days.
  • Available money transfer options are often slow and inconvenient to use. In India, customers spend an average of more than two hours to send money at a bank, including travel and waiting time for sender and receiver, while post offices and the more common informal hawala courier took around 45 to 50 minutes. In Haiti, researchers found that money transfer service customers can spend 50% of the value of the money transfer on transportation costs getting to and from the transfer services provider. Researchers also found that the services were slow in making the money available on the other end. In the Philippines, customers said that they choose money transfer service providers based on their ability to deliver payments quickly.
  • Money transfer services often have high direct costs. In India, transaction fees varied from 1% at banks to 3-4 percent at informal money carriers, to over 5% at the post office. These fees were often further exacerbated because of requests for bribes and tips. In Haiti, the researchers documented one typical transaction that incurred a fee of 12%. In the Philippines, clients sending domestic remittances to family or friends expected to pay 3-5% of the transaction amount.
  • Risks are more often associated with robberies traveling to and from a provider than risks from the provider stealing the money. In India, 72% of clients cited security of the money as the most important feature they looked for in choosing a money transfer service. They reported feeling risk of robbery around money transfer points, and even greater risk linked to informal money transfer options. In the Philippines, the second most prominent driver of choice of a money transfer service after delivery speed was trust. Despite the perceptions of risk and high value placed on trust in the Philippines, researchers did not document many actual losses in sending money.
  • Demand for digital payments appears to be strong even at the bottom of the pyramid. A nationally representative household survey of 2,016 Kenyan households conducted in fall 2008 and repeated a year later found that M-PESA is rapidly propagating down market, reaching a majority of Kenya’s low income, unbanked, and rural populations, just 28 months post-launch (See Figure 1). A related survey among users of Telenor Pakistan’s Easypaisa service both found a similar result (Figure 2). The Kenya and Tanzania data from the Gallup World Poll cited above show that in Kenya 80% of transfers were mobile phone based (i.e., M-PESA). More surprising is that in Tanzania 50% of transfers were mobile based, showing just how much progress has been made by mobile money in that market (others were made by sending physical cash via bus or with a traveling relative, or through banks). The data also show that the poor are conducting these domestic remittances at nearly equal rates to higher income people (international remittances, in contrast, are concentrated in the top two income quintiles).

Figure 1: M-PESA Kenya downmarket penetration¹

Figure 2: Telenor Pakistan’s Easypaisa 

What are the benefits to the poor when they are provided with high quality payment service? Two studies show real welfare gains from better payment options. A new randomized control trial (RCT) from Jenny Aker of the Center for Global Development looks at the effects of delivering cash assistance to drought stricken families through mobile money. The intervention took place in Niger using the ZAP service there, which is still at the pilot phase and not achieving high levels of activity or service quality yet. Nevertheless, the delivery of the payments through the mobile channel had measurable benefits for clients. There were clear time and cost savings for the recipients and for the program (on average it was a 4km walk for cash recipients, 0.9km for ZAP) and more direct targeting to women. This appears to have led to more diverse crop varieties, food purchases, and diet diversity in the group given the ZAP product and less asset depletion.

More striking, new evidence from a Gates Foundation-funded, nationally representative household panel survey in Kenya finds that access to a digital payment system can increase the size and efficiency of these informal insurance networks. The researchers find that M-PESA users were able to fully absorb large negative income shocks (such as severe illness, job loss, or harvest failure) without any reduction in household consumption. In contrast, statistically comparable households who weren’t connected to M-PESA experienced, on average, a 7% reduction in consumption in response to similar shocks. The researchers also find that following a shock, households with access to M-PESA are more likely to: (i) receive a remittance; (ii) receive a larger total amount of remittances; (iii) receive funds from a larger network of senders; and (iv) receive funds from senders who are located further away. Digital payments thus appear to improve risk sharing across households (Suri and Jack (2011) Risk sharing and transaction costs: Evidence from Kenya’s mobile money revolution).

The most transformational benefits will occur once large numbers of poor people are connected to a payment system with a convenient way to get money in and out of the system. This will engender large public sector savings from more efficient payments and receipts, and could have large market creation effects in the private sector as well. In a paper I wrote with co-authors Clara Veniard, Philip Machoka, and Bill Maurer (2011) Emerging Platform: From Money Transfer System to Mobile Money Ecosystem, we document an emerging ecosystem of start-ups and existing players launching new products that leverage the payment system embodied in M-PESA in Kenya. If this ecosystem really takes off, we can expect to see a broad basket of financial services and products competing to fill the various financial needs of poor households. Getting similar digital payment platforms up and running in other countries could be a major game changer for financial inclusion.

¹ Unbanked, poor, and rural are not mutually exclusive – clients can fall into more than one category. Low-income Kenyan households are those below median (50th percentile) income. Unbanked, poor, and rural are not mutually exclusive – clients can fall into more than one category.

A recent Financial Times article entitled, “Innovators don’t ignore customers” argued that the rapidly dropping share price of Netflix, a DVD rental and online film service could be explained by the fact that the company lost touch with what its customers wanted. Keeping a sharp eye on client demand is thus not only the responsible or developmental thing to do–it simply makes good business sense.

This special Clients at the Center blog series has lined up a broad range of voices to delve into what we mean by understanding client demand and how developing a much more profound understanding can further responsible financial inclusion. We encourage you to read and jump into the conversation.


24 August 2012 Submitted by Dr V.Rengarajan (not verified)

Dear Jake Kendall
In the midst of too much focused deliberation on Micro credit services under MF platform as if it is only panacea for poverty cure, it is very much refreshing to read on ‘better payment services to the poor’. This neglected service assumes more importance particularly for migrant poor who sacrifice lot of entitled welfare benefits due to their inevitable mobility for making both the ends meet. Thanks for bringing this subject in the surface for further brainstorming for the cause of the poor.
Keeping them as target group, three issues ‘ safety’, ‘risk’, and ‘cost’ associated with payment mechanism merit attention here as it facilitates for intensive inclusion on one hand and enhances their financial management capability towards their up-liftment on the other.
a). Regarding existing payment mechanism like post office two concerns are safety and timeliness. Despite time taken in the process of transferring money , safety assumes more vital than time element in the given risk infested informal mechanism for the said purpose.
b).In the case of risk associated mechanism, isn’t possible to cover the transaction under micro insurance for protecting against loss due to robberies /theft ? On similar account of DICGC(Deposit Insurance and Credit Guarantee Cover in India) covering for deposit and credit , is it possible to protect Payment mechanism also under such cover against any risk.? While poor has been innovatively made ‘credit worthy’, will it be difficult to innovate means to make the poor ‘insurance worthy’ also in respect of their risk free cash movement.?
c) In the case of ‘cost’ while many welfare measures and ‘free bees’ are provided to the poor and disadvantaged/marginalized by the state government, why not the provision for concessional cost for the target group (migrant poor) for movement of cash be considered as a kind of social welfare good in the formal sector post office? ( As in micro credit, concessional cost with just 4% interest rate for the selected weaker sections under Differential Interest Rate-DIR in Indian banking priority sector lending) In such case both safety and low cost are ensured for the bottom
Thank you for sharing my views.
Dr. Rengarajan

In fine, albeit mobile tech (M-PESA)has advantages in out reaching the unbanked, rural and poor, the bottom most in the pyramid appears without much beneficial impact . The data presented in fig 1 & 2 do not much support to the view that poorest. at bottom are benefited ( only 5% less than 1.25$ category in Fig 2). Does this situation indicate the difficulty in outreaching them from supply front, or no effective demand from bottom layer for the said purpose ? In the case of former, still ‘tech’ innovation is necessary for bottom inclusion . Otherwise there will be tech divide widening the intra poor inequity gap.
Dr Rengarajan

24 August 2012 Submitted by Jake (not verified)

Dr Rengarajan

Thanks for your comments. Please excuse my late reply as I was out for the Holidays.
You make some good points, RE risk and costs. I think that a lot of these issues will cease to be issues when high quality mobile money systems are rolled out in more countries. Mobile money has shown lower risk, greater speed and efficiency, and lower costs than many of the alternatives (which may excell in one of these categories by rarely all at once).

You also make a good point that the Easypaisa product may not have yet reached the poorest of the poor. That said, the data shows that in Kenya where mobile money has been up and running longer, the M-PESA service has steadily crept down market each year penetrating the lower economic segments more and more (sorry, I don’t have this in a graph form to present). Thus, I think the Easypaisa deployment is showing great signs by being so far down market already and will likely make even more progress.

Again, thanks for the comments.

Add new comment