Explaining UPI

I just paid my cook his salary for November. Given the cash crunch, I paid him through a bank transfer, using IMPS. Earlier today, my wife had asked him for his account details (last month I’d paid him on his wife’s account).

An hour back he sent me his account details (including account number and IFSC) via WhatsApp. I had to wait till I got home and got access to my laptop (Citibank app doesn’t let you add payees on mobile banking).

I get home, log in to Citibank Online. Add payee, which includes typing his bank account number twice. Get SMS asking me to confirm payee addition. I authorise payee. And after all this I am able to finally do the transfer – and I expect him to have got his money already.

For a long time I was wondering what the big deal with UPI was, given that IMPS is already fast enough. Having finally tried UPI earlier this week (it’s finally coming to iOS, but only available on ICICI now. And the implementation so far sucks, since you need to pull out your debit card for two factor authentication – defeating the point of UPI. I’m told it’s better on Android), I realise how much easier and safer the transaction would’ve been.

Firstly, the cook needn’t have sent me his account number. All I would need was his virtual payment address. I would then open my UPI app (in my case, iMobile) and click on “send money”. And then I’d add his virtual ID there, following which his name would appear. Two or three more clicks, and entering my PIN code, the transfer would be done.

No bank account number. Not even a mobile number or an email ID. Just a random string of characters would allow me to transfer money to him! And later I could give him my UPI ID, and next month onwards he could simply send me a request via UPI for his salary. And two clicks later it would be done!

Mint has reported that there are massive delays in merchants installing point of sale devices in response to the cash ban. Banks should instead seek to acquire merchants to accept money via UPI. It’s simple, it’s quick and it protects privacy.

In fact, if the bank sales staff now have bandwidth, it can be argued that all the planets have aligned for UPI to take off for merchant payments – people have less cash, point of sale devices are not available, and both merchants and shoppers have shown openness to cashless payments, and there is a push from the government.

If only the banks can bite…

Using my cook as an ATM

This happened ten days before high value notes were withdrawn, and suggests nothing about my cook’s political opinions or views. 

On 30th October 2016, I paid my cook his salary for October. As it was the usual practice, I paid him in cash. He asked me if I could do an online transfer instead.

It was the first day of Diwali, and he needed to send money to his wife in Bihar. And it being Diwali, all banks were closed, and there was no way he could send money to her. So he asked me if I could do that. And if I were anyway transferring money to his wife’s account, could I send her a bit more, he asked – he would compensate me for the extra amount in cash.

And so like that I used my cook as an ATM. He gave me his wife’s account details (it was such an obscure branch that I’d to google it to find the IFSC code – wasn’t in citibank’s lookup list). I added her as a “payee” and immediately IMPSd the amount to her. And my cook gave me the extra funds I’d transferred in cash.

Later on, I told him to install his bank’s app on his newly acquired fancy phone (with a Reliance Jio sim). I’m not sure he’s done that but considering how resourceful he is, it wouldn’t be long before he does that. And more of the Bihari cooks network in Bangalore do likewise.

Nandan Nilekani, in his championing of the UPI, likes to talk about how “anybody can be an ATM” with the new technology. This was an exemplary example of that.

The only fly in the ointment was that I didn’t need cash that day – after all I’d been to the ATM earlier that morning just so that I could get cash to pay my cook – so I ended up with a lot of cash that I didn’t need. Thankfully I was able to spend it productively before the ceased to be legal tender.

Following the withdrawal of high currency notes, I told my cook I would pay his subsequent salaries by bank transfer. He gladly agreed.

Cash on delivery

One of the big problems in the Indian e-commerce industry is that a lot of business happens through the “Cash on delivery” model, where the customer pays for the goods upon receiving it rather than at the time of ordering. According to sources, nearly three fourth of e-commerce in India is paid for using this model.

The problem with Cash On Delivery (COD) is that it leads to higher product returns and non-deliveries, since the recipient is not pre-committed to accepting it. While e-commerce vendors might try methods such as blacklisting customers in order to cut their losses, there is no clear solution in sight. COD is also a massive source of fraud, especially given that currently e-commerce platforms are more likely to subsidise rather than take a cut of transactions on their platforms.

One of the reasons given for the development of the COD model is the low credit card penetration in India (compared to other markets), and Indians’ unease with transmitting money online. Research (which I can’t be bothered to find and link to right now) shows that the Indian e-commerce industry actually took off once CoD was introduced.

Given that India is developing some new and innovative payment systems (the Immediate Payment System (IMPS) is one. There is a Unified Payments Interface (UPI) which is even better which is coming up), it will be interesting to see how the e-commerce industry in India shapes up from a payment standpoint.

There are two factors that drive CoD – one is the ease of payment transaction – you just hand over the cash to the courier when the goods are delivered. This is not seamless, of course, since it could involve problems involving change, and handling of large amounts of hard currency which makes it unsafe.

The other factor is trust – Indians don’t seem to trust vendors enough to pay for their goods before they receive them. While not prepaying gives the option to change mind at a later date, this can lead to significant friction in the system resulting in costs that are likely to get added to the customer (this doesn’t happen right now since platforms are still in heavy subsidy mode).

By paying for goods on delivery, the customer hedges against fraud by the vendor, and the transaction is smoother from the customer’s perspective.

While the industry claims that CoD is primarily due to lack of credit card penetration, my hypothesis is that it is more due to the trust factor. So far there has been no method (apart from possibly surveys which are internal to e-commerce platforms and which will never get disclosed) to understand which of the two it is.

With the development of new and innovative payment platforms, and the ability for a large number of people in India to transact online (willingness is another matter), this hypothesis can be tested. Once people have access to mobile apps that let them make instant and secure inter-bank payments (we are already on our way there), the low credit card penetration is unlikely to be a constraint against pre-payment for goods. If my hypothesis is true, the proportion of CoD will not fall despite the growth of these new payment methods.

There are flies in the ointment of course – platforms, driven by losses, are investing in moving customers away from CoD, so the data might not be very clear. Also, over time, people may develop more trust in e-commerce companies and start pre-paying, which will not tell us anything about their confidence levels right now.

We are in for interesting times!

PS: Like telecommunications (where most of India skipped the landline) and retail (where India skipped the “walmart step”), the payments industry in India is also likely to “leapfrog”, with a large part of the country set to bypass credit cards altogether.

Expanding IMPS

I hate carrying and transacting with cash. I find it extremely inconvenient and ineffective. The only place where I’m happy carrying and transacting with cash is Spain, where there is a high rate of pickpocketing, and carrying cash puts a floor on your downside.

There are several reasons to this. Cash is messy and dirty. Cash is prone to mutilation. Change is a massive problem. Even from the point of view of the central bank, printing currency costs significant money. When splitting bills at dinners I’m usually the guy who uses his card and “friendTMs”.

Recently (much belatedly, as I figured), I discovered IMPS. This service by the National Payments Corporation of India allows you to transfer money realtime. I used it once to transfer money between two bank accounts held (at different banks) by me. The “funds received” SMS arrived before the “funds transferred” SMS. It’s actually real time.

I had to make a payment to someone else last week and I had a problem with my ATM Card. Using the Citibank Mobile App, I discovered I could pay him up to Rs. 1000 without a second factor authentication, and only knowing his account number and bank IFSC code. The transaction took less than a minute. If he has a “MMID”, I could do the transfer using that ID and his mobile number, without him giving me his bank details. Again instantaneously.

So I’ve started wondering what prevents the tender coconut guy down the road (with whom I have a perennial change problem since a coconut costs Rs. 25, and 5 rupee notes/coins are hard to come by) putting up a board with his mobile number and MMID so that I can pay him through IMPS. I wonder the same about other vendors that I encounter in daily life.

The problem is one of product management and pricing. One reason credit cards haven’t taken off as much in India is that many vendors are concerned about the (~2%) interchange fees they pay on every transaction. So far I haven’t been charged for IMPS (at either end). Popularising and marketing it needs funding, though, and some kind of transaction fee structure needs to be figured out.

Currently, you have apps like Pockets, PingPay or Chillr that allow IMPS transfers. The beauty of these apps is that they eliminate the need for sharing MMID (which recipients have shared with the app on registration), and money can be transferred using the recipient’s Mobile Number only. The problem, though (as I had mentioned in this LinkedIn piece), is that these apps are currently building walls around banks, not permitting interoperability.

Since transactions take place on IMPS, there is no technical constraint. It’s about the war between these apps which prevents inter-bank integration. Given the network effects, though, it makes eminent sense for these platforms to merge and consolidate (or for one to “beat” the other), since this will unleash the “2ab term”.

Having watched the payments sector in a while now, I’m fairly bullish that electronic and mobile payments will take off in a rather large way here. What I’m not so clear about is what kind of pricing model will emerge, who will pay for it, and who will ultimately make money from it.

Exponential increase in uptake of IMPS

We had dealt with exponential increases on this blog once before. We revisit the topic, and this time this is in the context of the inter bank mobile payment system that came into place sometime last year. I’ve never used it so I’m not sure how it works, but going by the data put out by the National Payments Corporation of India, the volume of transactions is increasing at an exponential rate.

How do we determine this is an exponential rate? First, let us look at the time series of total volumes of transactions:

Source: http://www.npci.org.in/impsVolumes.aspx
Source: http://www.npci.org.in/impsVolumes.aspx

Notice that after remaining flat for a couple of months (maybe even decreasing) the number of transactions has really taken off (March is probably an aberration – but given that it’s the month of financial closure the higher volumes can be expected). Increased exponentially, you say? How can we test that?

We can test that by using a logarithmic scale for the y-axis. Here is the same plot again, except that this time the Y-axis is logarithmic.

Source: http://www.npci.org.in/impsVolumes.aspx
Source: http://www.npci.org.in/impsVolumes.aspx

Notice that apart from the part with the aberration and the initial two months, the graph is now linear. In other words, we can describe this graph by a line of the form

log y = a + b x

or y = exp (a + bx)

Thus, exponential!

Coming back from the geekery, it is really good to note that IMPS has taken off. However, this should not be taken as proof of the fact that mobile payments are easy, for IMPS is anything but easy. New RBI Governor Raghuram Rajan has said in his inaugural speech that he hopes to make it simpler to make payments via mobile. Hopefully this will take off soon. Till then all we can do is to contribute to the exponential growth in the update of the IMPS!