Oct 13, 2023
In an increasingly interconnected world, the advent of mobile banking has ushered in a transformative era. Access to mobile banking can have a profound impact on people in developing economies. The impact of mobile banking extends beyond individual lives – it ripples through entire communities, affecting aspects such as migration and inequality. Dr Saravana Ravindran, Assistant Professor at the Lee Kuan Yew School of Public Policy, studies challenges and potential policy solutions relating to migration, including mobile banking and digital financial services. He has conducted research into whether mobile technology can reduce inequality by modernising traditional ways to transfer money. He joins us to explore the socio-economic impact of mobile banking on communities.
David Austin: Thank you very much for joining us.
Saravana Ravindran: Thank you for having me.
David Austin: Could you share some insights on the impact of mobile banking on migration for work that's based on the research that you've done? What are the challenges of mobile banking adoption in rural areas? What are some ways of overcoming these challenges?
Saravana Ravindran: Certainly. I'm happy to share this work that we've been doing in Bangladesh and now hoping to scale across a number of countries in South Asia.
The idea behind this project was that if we just take a step back and think about earlier theories of economic development that traditionally has been this focus on rural development or bringing resources to rural areas. So, we can think about our microfinance programs, our ultra-poor graduation programs, and a lot of economic development policy in the past has focused on this idea that, okay, these are areas where they are rural, they're very poor and in need of resources. And let's think of policies that are targeted at these areas.
The complementary area of work has focused on this idea of harnessing the opportunities and the returns that are available in the cities, in urban areas to help with this rural poverty.
It's also in the urban areas, there's in sort of relative terms much greater abundance of employment opportunities, which presents itself with higher incomes. And over time, there has been a lot of migration that has been taking place, rural urban migration.
There's this on the one hand, a movement of people from rural areas to these urban centres. Now, can we then also think about how we can facilitate the movement of resources and money from the urban areas to these rural communities?
In the past, in many of these settings, migrants have found ways to remit money home. And they do this through a variety of different means. So, we have very traditional agents that they use to remit money. They may use bus drivers and other people who may be passing through those rural areas back home. Or they may send it through a family or a friend who may be making that journey, or they may even just take time off to make a trip back home to their families. In a lot of this, you already see that the traditional ways for these migrants to send money home can be costly, can be inefficient and also could be risky.
So, there is a risk that some of this money may be lost if the bus driver or acquaintance were to run away with that money, say, not deliver the money to the intended recipients.
We document that in our work in Bangladesh. This is the setting that we are going to conduct our study in. So, the backdrop is one of, there's been a lot of policy thinking into how we can bring resources to rural areas.
Maybe less attention has been paid to this idea of, okay, can we harness patterns of rural urban migration and the complementary movement of money back to the rural areas as an alternative policy to really harness the opportunities that cities offer, but then think about ways in which we can overcome these challenges associated with remitting money home.
That's the broad frame for the work that we do.
Our study tries to harness a number of trends that have been happening over time, across the developing world.
One has been this rapid adoption of phones and more recently smartphones. So, the penetration rates of mobile phones have been rising very rapidly in developing countries across the world.
Now associated with that, there has also been this growth in mobile money services. The more familiar or the very popular product globally would be that of M-Pesa, in Kenya. So, this really started off a revolution in many other countries, but I think as one of the very early pioneers in this space when M-Pesa really took off, it really showed how this type of transformation can be beneficial.
So, the appeal of this is that for many of these communities and these individuals who, for various reasons, are not able to access and formally have a bank account. Mobile money offers an alternative. So, you can be digitally financially connected. Even if you don't have a bank account, you are still able to open and have access to a mobile money wallet.
So, this wallet is something that sits on your phone, and you can transact through by interfacing with a mobile money agent. But you have access to a mobile money account, which then enables you to do a number of different transactions. Now the use cases may differ from place to place, but the one that we are focusing on in our study is really that of this remittance channel. So, P2P or peer to peer transfers, that's the primary use case that we're going to be thinking about and analysing in our study.
So that's the appeal of this technology. What we do in our study is really harness this movement, this trend that has started in Kenya and Africa, but also spread to parts of Asia and South Asia.
At the time we did this study, so this was done in 2014 to about 2016. So, at the time we did this study, the Adoption rates of mobile money in the country, in Bangladesh, was really low. Even at the time we were wrapping up this work and doing our end line surveys, in areas where we did not do our intervention, the adoption rates organically were just about 20 per cent.
Of the population and this is we are working with bKash the largest mobile money provider in Bangladesh at least at the time, the organically the rates at which people were deciding to sign up and adopt this technology was about 20 per cent.
It was a new product at the time. It was slowly gaining traction. But the rates of adoption were low. What we then wanted to do is to say, okay, this is our setting where we are interested in these migrant populations, we are interested in seeing how we can better facilitate remittances from these urban communities back to the rural families.
And can we harness this new technology that's emerging, this mobile money technology and these new players in the market. Let's try to leverage all of this to see if we can boost mobile money adoption rates, usage rates, and through that, remittances. And then, in doing so, we can then measure various socio-economic impacts and think about how the lives of the poor have changed as a result of what we are doing.
So that's the sort of idea behind the study. And so, what did we do then? So, we're working with these communities in Bangladesh. These are migrant communities that, so we worked in a particular migrant corridor, migration corridor. So, the rural families were mostly in this area called Gaibandha.
And they had someone in the family, a migrant, who was working in Dhaka, in the city, to remit money home. And they were working in a mix of different industries, many in garment factories but also other sectors. The distance is such that if you were to take a bus ride from Dhaka to Gaibandha it's not an easy journey. It's probably an overnight bus ride at the least. These are communities that are not very close in distance.
What we did is we worked with a sample of these families and their migrants, and we worked with slightly over 800 of these family migrant pairs. And this was recruited in partnership with an organisation that was on the ground that had been working with these communities, but then we also snowball sampled and added more of these families and their migrants to the study to ensure that we had a large enough size for our analysis.
We're working with these communities and what we did is a randomised control trial. And the idea behind this is that we are going to do our intervention in a randomly selected half of the study sample and the idea behind this randomisation is that because we are doing this randomly, there should not be anything systematically different between The treatment group and the control group, other than the fact that we went in, and we randomised half of them to receive our training treatment.
So, what we did was very simple information as well as a training intervention. With the randomly selected half of the study sample, we did a number of things.
First we provided information on bKash, this mobile money platform. So, we told them, “This is a service that you can use very easily to remit money, you don't have to make the trip yourself or send it through someone, you can do it through this platform, right?
The second thing we did is we provided them with very basic training. So, we taught them how to use the platform and the sort of key barrier at the time when we did the study is that the menus that the bKash was using at the time were all in English.
For the communities that we are working with, that was a big challenge. If they want to do a transaction and have to click through different menu options to get to the right page and enter the right phone number and all of that that was a challenge.
We helped them in all of this. We presented training instructions, which said, if you want to send money, first hit one. Then hit two to go to this second page and so on and so forth and guided them through the steps that they would need to do in order to do such a transaction.
We also said, “Let's try sending some money back and forth. With our team, let's practise and see if you got it and know how to do the transfer”. So that was the second component, a sort of training component.
And then the third thing that we did was a very basic onboarding facilitation. bKash, in order for the new customers to sign up there's some basic paperwork that they need to fill out and if these families needed some help with filling out these forms and such, we would provide some very basic assistance.
So that was our package intervention that we delivered, and we delivered that to both the families that were in Gaibandha, as well as the node that was also in Dhaka, so to the migrant as well, we provided this training.
After this, we're going to track the families for about a year plus, and then go back after a while to conduct further surveys to assess how these individuals are doing.
So, we're going to survey everyone, we're going to survey the control group, we're going to survey the treatment group, we're going to, we're surveying the families in Gaibandha, we're surveying the migrants in Dhaka. And what we are finding as a result of our intervention was really stark.
So, in terms of the mobile money adoption rates, we're able to boost these adoption rates by about 50 percentage points. So earlier when I mentioned that in the control group, by the time we are wrapping up this work in the end line service, the organic sort of rates of adoption was about 20 per cent.
We are in the treatment group with the group in which we did this intervention, we're able to boost that to about 70 per cent. So, about a 50 per cent point increase in adoption of these bKash services. So that's a very important first step. We're able to boost the adoption.
The next question is what about active use of these accounts and use for the purposes that we are really interested in? And so, then we track remittances. So, when we look at remittances, what we are seeing is that compared with the control group, the treatment group is really sending more in terms of value of remittances to the order of about 26 per cent.
So, in terms of the value of remittances, the treatment group is these migrants in Dhaka in the treatment group, they're sending a lot more in remittances back home to the rural families. In the study we go into much more detail on why this is, but you can, at a very high level, think of this as a price and convenience effect.
In what I had set out, the alternatives are much more costly, inefficient, and risky. As an alternative, bKash charges much lower fees for doing these remittance transfers. It's an instantaneous transfer, so there's no time delay in sending the money home, but once you push it on your account, the recipient account would receive it and it's a much safer way to send money. It's not subject to the loss and fraud and the other issues that I talked about at the beginning. So, because of all of this there's an increase in just the value of remittances that's being sent to the families.
And then in the third step, what we're going to say is, okay, we've boosted adoption, we've boosted remittances. Now let's think about What are the socioeconomic impacts of this for both the families as well as the migrants?
And for the families, what we see is that as a result of this money that's coming in, they're doing much better. When we look at the rates of say Rural consumption increased by about 7.5 per cent. Extreme poverty fell and in the treatment group in these rural communities in terms of their financial behaviours, they borrowed less, they were saving more. And so, we see improvements in the financial behaviours as well.
Interestingly, they were also more likely to send additional migrants. And so, we see this sort of very interesting, induced migration effect and various reasons for that. But that is also driven by the remittances that are coming into the families and overall, the picture in the rural household is looking much better on all of these different dimensions.
When we look at the migrants who are sending the money, though, it's a mixed picture.
On the one hand, they are also experiencing less poverty, they're saving more. But on the other hand, we find this interesting result that, when we look at their self-reported health on a number of different dimensions, their physical health, emotional health, mental health they are consistently self-reporting to be in slightly worse health.
Those in the treatment group -- who were sending these remittances -- were self-reporting to be in worse health. That's a really interesting result, and we dig into this a little more. One story that seems to be consistent with this is that now that we've made it convenient, easy and quick to send money, they are under greater pressure from their families to send money.
So, their families are calling them up and saying, “Hey, I need some money. Can you please send money?” And so, they are under a little more stress from the families to remit more. As a result, in some of the industries and some of the heterogeneity analysis that we do, we see that some of the migrants are working long hours which could also help explain the worse health that they are experiencing, but they're doing this in order to raise more money, to send this money because we've now reduced the price of remittances and made it so cheap and easy to send money. But they're then under great demand from their families. And so that we think of that as being a consistent story with what's going on the health side.
So overall, it's not a completely positive picture for everyone. We are able to, through our intervention, boost adoption and usage rates and remittances as a result of which I think, in a different dimension, things are looking better for the rural families receiving the money, but for the senders, it's a mixed picture.
And I think we need to be careful and look out for these sorts of unintended or unexpected results when we think about policy and how this can be scaled up.
David Austin: Really interesting. Some unexpected results there as well. Now looking ahead though, what do you foresee will be the key challenges and opportunities in shaping policies relating to mobile banking, let's say in the next five to 10 years?
Saravana Ravindran: Another big area that I want to highlight. This is really that of consumer financial protection.
So, I think this is now going to be a big issue as we go forward. In these communities now. There's been for various reasons this idea that, over time as the many are more and more aware of these platforms and what can be done, they also come across stories or maybe even with their interactions with agents If they start to lose trust in the platform itself, then that can further hurt adoption rates, usage rates and we don't then benefit from the great sort of socioeconomic impacts that I talked about at the beginning with our study.
As I noted, these agent networks form an important part of the ecosystem and many have to interface with these agents to deposit money, withdraw money, some also send money just using the agents. These are over the counter transactions. So, the agents form a very important part of this whole ecosystem. And so, when the customers interact with the agents, How, trusting the customer is of that agent becomes really important. What kinds of fees the agents are charging the customers is important. Are the agents charging the sort of truthful rates or are they overcharging their customers? Are they adding some hidden fees or additional charges that they want to? These become important issues.
And so right now there's been a lot of work on the ground in different settings to really understand how these customer interactions with the agents have been taking place and very interesting ways of measuring this as well. There are studies that are currently using, these are called mystery shopper experiments. So, what they do is they basically send people who are trained to pose as shoppers and take down all sorts of very detailed notes.
They pose themselves as customers to mobile money agents and say, okay, I would like to make a deposit and make a transfer of this certain amount. They are then secretly recording all sorts of details about the transaction, exactly how much was charged for each step of this and so on and so forth.
And so that, as an audit through the consumers it's really providing information on what agents are charging, are they charging the truthful rates or are they charging additional rates and so on. So, part of the consumer interaction with the agents is really key.
That's one where I think we need to pay a lot more attention to because if you have a mistrust and customers feel that they are being overcharged or they can't trust the agent then that becomes a huge problem for adoption and use of these services.
So, in terms of the potential solutions. This is a very tricky way forward because there's only so much that information can do. There's been interesting work that really tries to provide consumers with information on how to identify scams.
This is work that the group of researchers did with the Innovations for Poverty Action in Kenya, where they really tried to provide tips to customers on how do I identify scam messages, right? So, if you see an SMS message with an offer or what have you, how can you identify what's a scam and what's not a scam, right? And these are very standard tips that are being pushed to these consumers.
Interestingly what their preliminary work has been showing is that, as a result of this, the intervention group with which they were providing the tips actually were not much more likely to correctly identify the scam messages.
If anything, they were just becoming very cautious overall. That they were becoming much more cautious, but they weren't much more likely to correctly identify scam messages as a result of the kind of scam prevention tips that were being given. Which is actually concerning because then we worry about, as we are pushing out a lot of information about how to detect scams and fraud. Are we just making people more scared overall?
And then that discourages adoption and use of these mobile money platforms. So, I think in the information provision space there's maybe issues to work out, but then also maybe there are some technological innovations that can be done in this space.
So, I don't think they are very easy answers going forward, which is why I think this is going to be a big issue in the next five to 10 years as the mobile money landscape changes.
David Austin: Thank you very much for sharing all that with us. It's really clear that, as you said, mobile banking and the whole payments ecosystem has such a huge impact on the lives and economic condition of people, both the developing markets and mature markets. So, I'm sure there will be a lot for you to continue to study and learn and hopefully share with us in the future.
Saravana Ravindran: I hope so too. Thank you so much.