On the death of credit cards

An article that was recently recommended to me on Medium talks about the death of credit cards (among other things that are currently incumbent in the banking system). As someone who has worked a fair bit in “FinTech”, I broadly agree with what he says. As someone who has worked a fair bit in “FinTech”, I’m also not sure how easy it is to disrupt.

The article says:

The two primary use cases for a credit card today could be illustrated thus:

  1. I’m at the grocery store, swiped by debit card and the transaction was declined because my salary hasn’t yet hit my bank account. I need to buy these groceries for the family today, so I’ll use my credit card and worry about why my salary hasn’t hit the account later, or

  2. I really want this new iPad Pro, but I can’t afford it based on my current savings. If I use a credit card I can pay it off over the next few months

And proceeds to explain why each of the above situations can be unbundled to some kind of an instant credit scenario, rather than the bank having extended a lien to you through which you can borrow.

While the idea of instant credit (on the lines of Affirm) makes intuitive sense, the problem is with transaction costs. Irrespective of algorithms significantly slashing the time required and marginal cost of underwriting loans, the fact remains that the marginal cost of underwriting and extending new credit can never be brought down to zero.

There are costs to updating the information the bank knows about you. There are costs to creating any kind of legal documentation, and insuring that. If you were to list down all such costs, you would find that even if the cost of actual underwriting itself were to be zero, the marginal cost of issuing a loan is significant.

It is for this reason that banks have traditionally settled down on a model of “approve once, borrow multiple times”. For retail borrowers, this translates to a credit card, where they can borrow up to a predetermined limit, with no questions asked for each borrowing. For corporate borrowers, this translates to something like a “working capital lien” or “overdraft”.

The article I’d linked above talks about one of the solutions being an “overdraft”. In that sense, what it says is that the physical credit card might disappear, but not the fundamental principle, which is “approve once, borrow multiple times”.

In fact, as companies come up with new and innovative ways of slashing marginal cost of underwriting to enable “on-demand approval” (I’ve been involved in such efforts with a couple of companies), the question is whether such costs can actually be brought to zero, and if not, whether the model can be sustainable.

As long as the marginal cost of underwriting remains even mildly positive, it is not profitable for lenders to lend out small amounts with “on-demand approval”. How this problem can be solved will determine how well “FinTech” lenders can disrupt banks (on the lending side).

Payment systems

I had lunch today at a rather fancy Japanese restaurant here in Barcelona (I’ve forgotten if I wrote that blog post last year on how you get fantastic East Asian food of all kinds here). I didn’t pay a fancy price – this concept called “Menu del dia” (menu of the day), one of the very few good things instituted by General Francisco Franco meant that you can get cheap weekday lunches at most restaurants in Spain.

The above (Katsudon and beer), along with some noodle soup and two sushis and a cup of coffee, set me back by €13, which isn’t too bad by Barcelona standards (most weekday lunch platters at restaurants cost ~€10).

While eating I noticed that other patrons at the restaurant were walking up to the bar to pay the owner directly, rather than asking for the bill at the table.

So once I was done with eating and drinking, I went up to the bar to pay. The owner had seen me coming and had prepared my bill, which he presented to me. As I reached into my pocket, he got out the card swiping machine.

It might have been a shock to him when I presented a €20 bill instead, and he had to scramble to produce the change from somewhere inside the kitchen (the other patrons before me had all paid by card).

While this is one data point, it’s interesting how the economy here has moved to a situation where the default method of payment is through credit/debit card, rather than by cash (though my favourite bakery refuses to accept card for payments less than €5). The ease of card payments (most debit cards nowadays come enabled with NFC, though a fair number of merchants still insert the card to read the chip) combined with ubiquity of cards has meant that card usage has started trumping cash.

It will be interesting to see how the payments ecosystem will develop in India, which is still largely a cash economy. My belief (and hope) is that India will leapfrog credit/debit cards (as it has leapfrogged landline telephones and big box retail, moving directly to mobile phones and e-commerce) and take up electronic payments in a big way.

IMPS (immediate payment service) is already a fantastic protocol for bank-to-bank transfers, and the costs are extremely low. In April, the Unified Payment Interface (UPI) will be rolled out, which makes transfers to hitherto unknown people even easier! If our banks do a good job of implementation, there is a good chance it might get adopted widely (long back I’d made a case for the RBI to subsidise such payments).

The RBI does a Ramanamurthy

This is the second time in a few weeks I’m referring to this scene from Ganeshana Maduve. Please watch it first.

To repeat the story:

Ramanamurthy the owner of the “vaTaara” (a kind of apartment that was popular in Bangalore till the 1980s, with lots of small houses in the same compound) wants to whitewash his house. The residents of the vaTaara  demand that if he whitewashes his house he should whitewash the entire vaTaara. After a long and protracted negotiation, Ramanamurthy agrees to their condition – he doesn’t whitewash his house!

It is a similar story with taxi operators in India. Uber (the Ramanamurthy) figured out a way to bypass RBI’s two factor authentication system and offer seamless payment options for their taxi services. Soon other taxi operators like TaxiForSure and Ola started crying foul saying they too wanted their houses painted, i.e. they too wanted to locate payment servers abroad to accept one factor authentication credit cards.

And now RBI, like the rent controller ubiquitous (in mention only) in movies of the late 80s has stepped in and stopped Ramanamurthy from painting his house, too – they’ve barred Uber from charging in US dollars for Indian rides. It would be interesting to see how the market will develop now.

My personal opinion is that RBI’s insistence on two factor authentication is half-assed. They should make every effort possible to increase the number of credit card (or account-to-account) transactions. On one hand it decreases flow of black money but more importantly it means that people will keep more cash within the banking system (rather than as hard cash) which will have a multiplier effect on money available for lending and all that.

It’s fine to have regulations in place such that credit card fraud is minimized but that doesn’t mean cutting credit card transactions altogether! Hopefully the RBI will see the light of day on this one soon.


We live in an era of unprecedented liquidity. Think about the difference from just about ten years ago. Back then, there was a much larger amount of cash reserve that one had to keep in one’s home, or on one’s person. There were no ATMs. There were no credit cards. All purchases needed to be meticulously planned, and budgeted for.

Now, because we don’t need to carry as much hard cash, there is so much more money in the banking system. While that gives depositors the nominal daily interest rate (at some obscenely low rate), there is much more money available with the banks to lend out, which increases the total amount of economic activity by nearly the same amount.

Just think about it. It’s fantastic, the effect of modern finance. And I don’t disagree with Paul Volcker when he says that the most important contribution of modern finance has been the ATM.

PS: My apologies for the break in blogging. I was in and around Ladakh for a week (yes, I was there when the cloudburst happened) and there were some problems with my laptop when I returned because of which I wasn’t able to blog. Hopefully I’ll be able to get back to my one-post-a-day commitment. And I have lots of stories to tell (from my Leh trip) so hope to keep you people busy.