This is a guest blog by Justin Oliver & Dan Radcliffe. Justin is Executive Director of the Centre for Micro Finance in India and Dan is Program Officer with the Bill & Melinda Gates Foundation.
Imagine you’re a migrant laborer living entirely in the cash economy. How do you send money home to your wife and kids? How do you buy supplies from the next town over? How do you pay utility bills? In short, how do you move physical cash over distances? Without access to systems that permit transferring money conveniently, safely, and cheaply, hundreds of millions of domestic migrants face these dilemmas regularly.
To better understand just how costly making remote payments can be for poor households, the Bill & Melinda Gates Foundation commissioned the Centre for Micro Finance at the Institute for Financial and Management Research (IFMR) and the Reserve Bank of India’s College for Agricultural Banking to survey 274 domestic Indian migrants and their families living at opposite ends of four domestic remittance corridors. The researchers sought to understand how these households move money and to quantify the total end-to-end cost of these transactions. You can find the full study here.
The key take-aways:
57% of respondents most recently used an informal payment channel – most commonly hawala couriers – to transfer money. As shown in the table below, the average cost of making a typical domestic remittance of Rs. 2,000 (US $44) through a hawala agent was 4.6% of the transfer amount, including formal fees and indirect costs, such as paying a bribe or traveling to the nearest payment outlet.
Contrary to popular perception, sending and receiving money through India Post is expensive: in addition to paying 5% of the total transfer amount in formal fees, respondents spend on average another 1% in bribes, tips, and other indirect costs. The Post, used by 13% of respondents, was by far the most expensive payment option used by respondents.
Banks offer the cheapest method for sending money, costing on average 3% of a $44 transfer. But only 30% of respondents use banks to send and receive money, largely due to the difficulty of traveling to the nearest branch, obtaining the documents needed to open an account, and waiting in line to send or receive a payment. And while banks may be cheap, they’re by no means convenient. As shown in the table below, making a transfer through a bank requires an average of 15 minutes of travel and 45 minutes of wait time for senders, plus another 40 minutes of travel and 50 minutes of wait time for recipients, for a total of 2.5 hours.
The upshot is that poor domestic migrants in India tend to be excluded from the formal payment system and are left with informal transfer mechanisms which tend to be inconvenient and expensive. Pent-up demand for safe and convenient e-payment services fueled the growth of successful branchless banking deployments in Kenya and Brazil. Could the need for e-payments also provide the “hook” to drive take-up of branchless banking in India?
Dear A.K. Rao,
Thanks for your comments! We decided to keep this blog posting short and sweet rather than include all of the results here. But in the larger study this blog post was based upon, you can find the answers to some of your questions — I encourage you to go through it. It’s at http://ifmr.ac.in/cmf/publications/wp/2010/42_Putting_Money_in_Motion.p….
To your specific question about the percentages of respondents who most recently used informal mechanisms, hawala was 26%, cash courier 19%, and friend 11%. (See Figure 1 in the study). It’s worth noting here that we only found cash couriers along one of the four routes we studied, between Orissa and Surat, where over 50% of migrants we spoke with used cash couriers.
You’re also absolutely right to point out the differences in mechanism used for the transfer according to the transferred amount. In fact the economics of this decision can change a lot if someone is transferring Rs. 500 versus Rs. 10,000. What we’ve reported here is the median transfer amount for our sample — Rs. 2000. The 25th percentile amount is Rs. 1500, at which point the costs don’t change considerably. But for migrants sending only Rs. 800, the (small) fixed cost of a bank transfer means that hawala couriers and cash couriers are indeed a cheaper option, and the bank transfer option costs roughly the same as the post office. At the 75th percentile, or Rs. 5000, the differences become stark: bank transfers cost only 1.4%, whereas hawala and cash couriers cost twice as much and post offices cost 3.6 times as much. Accordingly, there’s a clear relationship in the analysis: people who transfer through banks tend to transfer larger amounts, while those using informal means or post offices transfer smaller amounts. But we can’t conclude that there’s a causal relationship necessarily — it could just be that wealthier people tend to have better access to bank accounts, not that having a bank account necessarily leads to people transferring larger amounts. This is discussed more on page 9 of the study.
Regarding the statement that “informal transfer mechanisms…tend to be inconvenient and expensive,” the analysis clearly suggests that informal transfers and post office transfers are indeed more expensive than transferring through banks (subject to the caveat described above, when amounts are quite small). And half of the cost of transferring through a bank is opportunity cost — not formal fees, but the cost of the 2.5 hours spent to make a transfer. (We calculated opportunity costs at the reported rate for each respondent). But you may be right that the use of the word “inconvenient” here was inexact. “Unsatisfactory” may have been a better word in this case. You’ll see on pages 4-6 of the study that migrants overwhelmingly would prefer to make transfers through a bank, but many are constrained from doing so. Nearly half of those who use hawala couriers would prefer not to use them if they could avoid it. This clearly highlights the pent-up demand for improving the options for those who want to send money from one part of the country to another. Current systems in India, including informal systems that have sprung up to fill gaps where formal systems were impractical, don’t meet the need. I think there’s an enormous opportunity for electronic payment systems to develop and meet the need, but it will require innovation, flexibility, and effort by banks and the Indian central bank to make this work for the poor, not just the wealthy.
While the study reinforces the already realized need for an affordable and efficient money tranfer/remittance mechanism, it would be interesting to delve deeper into the analysis. It would be good if those are also published sometime soon. Especially the percentage break up of 57 into what percentage used Hawala, Cash Courier and Friends respectively. While the sample of 274 may not be able to do justice for further break up, it would certainly give some idea on the nature of transactiions within the informal transfer modes.
It would also be interesting to see how the behaviour is impacted given the variations in amount, which could be a prime determinant for the mode of transfer. All of these modes of transfer are used by all segments of economy (Migrant-Stagnant and Low-Medium-High incomes) at some point or the other. So the behaviour of using transfer among migrant labourers should be contextualised to the general transfer behaviour, whihc would actually be able to establish clearly the “Pent-up need”.
One statement that I found difficult to understand was how “…left with informal transfer mechanisms which tend to be inconvenient and expensive.” was derived! The analysis just proved that except for Hawala couriers, the other modes are as cheap or even cheaper than formal tranfer mechanisms, and add to that the time taken for transaction, which to some extent contributes to the ‘convenience’ aspect.
It would also have been great to see the service delivery aspect published..especially on the time it takes for the money to reach the intended recipients hands from the time it leaves the remitter’s hands.
Hi Justin and Dan - great post. I was wondering if you came across the phenomenon of folks circumventing the banks transfer fees in an ingenious way? When I was at MicroSave we found an ingenious way of people using banks without using formal transfers. The sender (often the husband, say in Delhi) would open an account, but would give the recipient (often the wife in say Bihar) the physical ATM/debit card. The sender in Delhi would the make a normal deposit into his account and the recipient would make a normal withdrawal as needed. It was a small percentage but was growing - what I love about it was how the poor were extremely creative in figuring out and using the system to their advantage (although it certainly annoyed the banks). Thanks!