Hemant Mishra/Mint
Hemant Mishra/Mint

The only thing that’s cheaper in rural areas is rent

BTI Payments' K. Srinivas talks about traction in rural areas, and whether wallets and money apps could replace cash

Bengaluru-based BTI Payments Pvt. Ltd is the third largest white label ATM deployer in India. Considering the focus on spreading banking services, banks are looking at partnerships with ATM companies to reduce the cost of setting up the machines. K. Srinivas, managing director and chief executive officer of the company, speaks about traction in rural areas, and whether wallets and money apps could replace cash. Edited excerpts.

The Reserve Bank of India (RBI) is looking to increase the number of white label ATMs in unbanked areas. What has been your company’s contribution?

When RBI started granting white label ATM licences, it was with a fundamental view to increase ATM density, especially in unbanked areas. Areas are divided into tiers. Tier I is urban, Tier II and III are semi-urban and Tier IV, V and VI are called rural. White label ATM companies were given three options. The first was that for every ATM you put up in a rural area, you are allowed to put one in Tier I and II location. But the minimum commitment for a span of three years was about 50,000 ATMs. The second option was one for every two with a commitment of 15,000. The third option was one ATM for every three in rural areas, with 9,000 ATMs’ commitment.

Like many, we opted for the third option. We were supposed to complete 3,000 ATMs by March 2016 but we have already done this. While technically we should be putting up about 67% ATMs in rural areas, as we stand today, almost 92% of our ATMs are in smaller towns. Another point is that out of the 3,000 machines, 60% are in areas that are ‘bank dark’, which neither had a bank branch nor an ATM.

In market share, we are about 25% in the white label space but in terms of transaction market, we are 28%. We are among the top 20 ATM brands in India.

How does the system work?

We deposit money in a commercial bank and take currency notes. We give that to cash management agencies who load ATMs. When a customer uses the ATM, she uses our switch to talk to the bank’s switch; once the bank approves, the cash is dispensed. The day after the transaction, the bank from which money was drawn pays us the money withdrawn plus 15 (per transaction) as fees. Our sponsor bank will get the money. A sponsor bank is used for settlements and for front-ending customer disputes. We work with 7-8 banks for currency notes and three sponsor banks.

Is it cheaper to set up a white label ATM as compared to a bank ATM? Is your model low-cost?

What we have done is rip off all the frills. So, the operating model is low-cost. We operate with 50-60 sq. ft for every ATM. A bank machine would be typically larger. Other things, like machine, security and power backup, are identical. The (other) difference would be that this is our only business. For a bank, ATMs is not priority.

Also, banks have not adopted the concept of shared infrastructure in a big way. Banks like to own their switches. We borrow a switch from another operator that is running 10,000 ATMs and pay per use. We have outsourced central monitoring of all the ATMs to Mphasis. So, the cost per ATM comes down. It might cost a bank 50,000 per month to run an ATM; we are able to do it in about 35,000.

Is it cheaper to set up ATMs in rural areas?

The only thing that’s cheaper in rural areas is rent—3,000-4,000. In a city it could be 10,000-15,000. But there are some other costs that are higher in rural areas. One is that you have to travel long distances, say, 100 km, every day or every alternate day to fill cash. So, we have a higher amount of cash sitting in ATMs in terms of number of days; typically, 3-4 days. In a metro, we can operate with one day’s cash. Another issue is availability of power. In some places, power is not there for 20 hours a day.

What do you think of payments banks? Will usage of wallets and money apps reduce ATM usage?

About 20 new banks—payments and small banks—are coming up. It would cost them a lot to put up ATMs. We wouldn’t mind cobranding with them. For a small sum, we can give them all the services of an ATM, including visibility. We could even cobrand with 2-3 banks.

There are countries where credit cards are widely used. ATM transactions in those countries haven’t dropped in the past five years. In India, 90% payments happen through cash. Even if this comes down to 50%, there is still enough room. Wallets are great instruments for money transfers but not for payments. They are not interoperable and people can have only a limited number of wallets.

Is running ATMs profitable?

It is an ongoing process. We wait for 6-7 months. If we think it doesn’t make commercial sense, we don’t mind taking it (the ATM) off. Even though there might be a loss in closing down the ATM, we can reuse the stuff. Profitability is still some distance away. To break even, we need 90-100 transactions per day. If the ATM is 7-8 months old, it may get 70-75 transactions (a day). But three big issues here are: high interest cost of cash, managing power, and availability of cash. Often, when we need 100 rupee notes, they are not available with the banks. We are going back to the regulator to show them how much it costs to run an ATM.

Apart from bank fees, is there any other revenue source?

The ATM is a computer with connectivity. We are looking at utilising this connectivity. For instance, can we sell mobile recharges, partner with ecommerce companies, sell insurance products. When you have 80-100 people walking into an ATM, they can be the ideal persons to talk to, for any brand. We are just starting to advertise.

What do you think about the 100% foreign direct investment (FDI) announced in white label ATM space?

It is good. It reflects the seriousness RBI is attaching to this space. Today, we have about 210,000 ATMs in the country. If you want to take the banking penetration to the 70-80% mark, the country would need at least 400,000-450,000 ATMs.

Currently, the deployment is 20,000 a year. So, it would take another 20 years to reach the required level. That’s why the government wants more people in this space. I’m sure global companies would look at India as a potential market.

Close