India’s UPI is being hailed as a success. With billions of transactions per month it is indeed a great system. However, we must also understand the system has many challenges which will only become more serious with time and scale.

Economics

UPI is a socialist scheme. The Indian government pretty much funds it today one way or another which means while people using UPI think it is free, in reality it is fully funded by the taxpayer money. This shields the system from market forces and hence hampers innovation and efficiency.

National Payments Corporation of India is well aware of this and has floated all sort of proposals but had to shelve them because of backlash. India is an extremely price sensitive economy and hence even a small fee would kill the entire UPI model.

No recourse from fraud

Unlike credit cards, UPI is instant which means once the transaction has happened there is no reasonable way to get money back. This makes UPI a prime target of financial fraud. Once the phone and pin is in the hands of attackers the account is pretty much at the mercy of the attacker. There is no way to prevent the money transfer from happening.

Scaling for money

UPI transaction limits are very low. Today UPI can not be used for large transactions. Clearly the technology is simple but the real problem is fraud prevention. What UPI needs going forward is a smarter risk assessment and challenge flows. This would mean that every possible transaction above certain limit could be subject to challenge flow which would include OTP, phone call verification or tap from a debit card.

Path for future innovation

Economics

This problem can be solved in multiple ways. One way is to by creating a multi tiered system of UPI where masses could use the UPI system as it is but the “pro” users or users with significant amount of money in their bank being moved to higher tier plan with a monthly fee. Say an INR 50 per month if you have more than 1L in your bank account could be charged as UPI fee only if you do more than 10 UPI transactions a month. At higher levels you will get better fraud prevention, fraud insurance, NFC based payments etc.

On the merchant side, NPCI should charge a flat monthly service fee. Having some revenue stream is better than none but for people adjust to this, it has to be done in a way that does not disrupt the flow of transactions drastically.

In short NPCI has to be innovative here.

Fraud prevention

NPCI has taken some good steps here already. One is to create a UPI lite system where you can spend small amounts of money without pin and asking PIN only for large sum. This is ostensibly simplifies the payment flow for nearly most transactions.

What we need now is the higher level where there could be flexible challenge that the banks, merchants and NPCI all could use.

Further more any account with significantly higher number of incoming transactions must be subjected to additional scrutiny and delayed transaction commit. This gives enough time for a victim to approach authorities if there has been a fraud.

Scaling operations

Chances are UPI will evolve into credit cards pretty soon. Credit cards have a lot of institutional history and very deep knowledge of all these issues and they have solved it pretty well and in a very distributed way so that different companies compete to give consumer an advantage.

Compared to this UPI is government enforced centrally planning system that has no notion of competition.

NPCI should encourage private companies to come up with their own alternative of UPI if they so desire. That alone should solve a lot of problems for us.