Live data from Hacker News

Bank transfers as a payment method

bam.kalzumeus.com

51–60 of 335 posts

Re: Bank transfers as a payment method

#51
post #43
post #18

Earlier quoted context omitted.

I could do without #5 to be honest. Should be working towards a more decentralized approach.

Why this obsession with doing away with a central authority? What is so evil about a central bank? Decentralization is not cheap. I buy a pack of bread from a door-to-door bread and eggs salesman (Yes that's a thing in India). It costs INR 40 ($0.53). I pay him via UPI, he gets money instantly in his bank account with zero fees. He gets full INR 40 like I had paid him in cash. I don't think any decentralized system w…

The GP's point isn't about RBI being a central authority, but about NPCI being granted a quasi-monopoly on retail payments despite being a non-government body. One of the important recommendations from the Watal Committee report[x] was to restructure the NPCI-relationship:

1. Change ownership structure. It was 75% owned by 10 banks in India, with no seats for other players in the industry.

2. Allow competition to NPCI. This got envisioned in the recent push for a NUE[0], but with a lot of caveats. RBI ended up deferring the plan[1]

3. NPCI also is pushing heavily to move from a non-profit to a for-profit model: https://www.livemint.com/news/india/npci-turning-for-profit-.... This will significantly change the current incentives it holds.

As an example of this conflict, RBI runs NEFT and RTGS for free as near-realtime transaction platforms for free, and now 24x7 (The 24x7 was a recommendation in the report). RBI does not charge banks for interchange on NEFT, but NPCI does charge for IMPS.

[0]: https://www.bloombergquint.com/business/npci-its-nue-competi...

[1]: https://inc42.com/buzz/rbi-defers-its-plans-to-distribute-li...

[x]: https://dea.gov.in/sites/default/files/watal_report271216.pd...

Re: Bank transfers as a payment method

#52
post #6

UPI can potentially be scaled to all countries. I think the biggest USP of UPI is the interoperability between payment apps(Google Pay, Paytm, PhonePe, etc), VPAs(alice@icici, bob@axis, dan@upi, 9848123456@paytm) and banks. One could have an account with bank A, create a VPA with bank B and use that VPA to pay a business through virtually any of the payment apps(GPay, Paytm, PhonePe, BHIM, etc). The main pain point U…

Actually the India-Singapore link will start in 2022 [1] [1] https://techcrunch.com/2021/09/13/india-and-singapore-to-lin...

But it's going to be an overlapping mesh of 1-to-1 agreements. Singapore is also negotiating with Malaysia to allow cross-border payments, but it's unlikely a Malaysian user would be able to chain these together to pay someone in India.

Re: Bank transfers as a payment method

#53

Fun Fact: UPI also allows certain other interesting kinds of transactions based on "special VPAs". These include: - IFSC code and account number combination, resolved directly by NPCI, is represented as account-no@ifsc-code.ifsc.npci (e.g. 12345@HDFC0000001.ifsc.npci) - RuPay card number, resolved directly by NPCI, is represented as card-no@rupay.npci (e.g. 1234123412341234@rupay.npci) The following were proposed, bu…

This is a very useful list. I recently found out that every FASTAG (highway toll payments) has its own VPA handle for convenient top-us. https://www.npci.org.in/PDF/npci/netc/Bank-UPI-Handels.pdf

Oh yeah, forgot about this! NETC is also linked closly with UPI now. netc.vehiclenumber@bankupihandle

Re: Bank transfers as a payment method

#54

Earlier quoted context omitted.

If it could work, it would have worked three decades ago. Bank transfers are just a pain in the ass. I still remember all the fees I paid just to transfer money to another bank/another country. Cryptocurrency was considered a way to simplify the problem, but probably didn’t work in this aspect

This seems out of touch with modern offerings e.g. wise.com. Easier than PayPal with none of the crazy bank transfer fees.

Wise fees are still too high for international transfers. They are doing currency arbitrage to sweeten the deal they get.

Re: Bank transfers as a payment method

#55

About UPI: The standardisation of payment interface is done by a corporation called NPCI. It’s an initiative from India's central bank (RBI) and it is non-profit. [1] This means that a private player's implementation does not become defacto standard. Multiple factors are aligning to make UPI successful: 1. Massive Mobile (smart phones) & Internet penetration 2. Cheap mobile data from telcos 3. Low cost of payment inf…

Indeed, UPI is a culmination of several years of ground work laid before it (IMPS, Adhaar etc.,)

Let me add one thing here though.

> Low cost of payment infrastructure to the banks

The cost is low primarily because it's is borne by tax payers [1] and the issuing banks. Issuing banks aren't too happy with forgoing a good chunk of their profits that they earn through issuing charges and also having to maintain the infrastructure free of cost.

A new for-profit payment consortium [2] has been proposed to compete with NPCI. Will be interesting to see how this plays out.

[1] https://www.business-standard.com/article/economy-policy/mdr...

[2] https://m.rbi.org.in/scripts/bs_viewcontent.aspx?Id=3832

Re: Bank transfers as a payment method

#56
post #26

Let's adopt patio11's criteria to a programmable public blockchain, like ethereum - Customer UX: crypto wallets built into browsers and phones. Connect your wallet to any webpage to pay. Or, tap your phone to pay via a crypto payment - Certainty: crypto transactions are, by default, instant and irreversible. Any kind of chargeback scheme, refund policy, or time-delay may be built into the crypto app layer - Settlemen…

- Customer UX: This really isn't as straightforward as you put it. The main issue for customers is that if they loose their wallet, they loose everything. This has to be accounted as part of UX, even making transactions is simple and straightforward. - Settlement/Certainty time: Blockchains combine these into one. They depends heavily on the specific chain. For some networks, its basically instant. Eth seems to be ar…

> Customer UX: This really isn't as straightforward as you put it. The main issue for customers is that if they loose their wallet, they loose everything. This has to be accounted as part of UX, even making transactions is simple and straightforward.

At scale people will want to use custodians for large amounts of money. So basically banks will always be a thing. Only the backbone is changing.

Re: Bank transfers as a payment method

#57
post #35
post #26

Let's adopt patio11's criteria to a programmable public blockchain, like ethereum - Customer UX: crypto wallets built into browsers and phones. Connect your wallet to any webpage to pay. Or, tap your phone to pay via a crypto payment - Certainty: crypto transactions are, by default, instant and irreversible. Any kind of chargeback scheme, refund policy, or time-delay may be built into the crypto app layer - Settlemen…

You keep saying “instant”, but in practice ethereum takes a few minutes for finality and can only handle a maximum of 50 tps. I’ll grant that both of those may improve with as zk tech matures, but even with rollups there’s no way of having better finality than ethereum, which is unlikely to ever be better than 60s

In practice, rollups can offer better finality than ethereum's base layer by using a bonding scheme. Basically, a rollup provider can sign a message that says "I promise that this transaction will become finalized, otherwise I forfeit my bond." Depending on the implementation, the receipt of that message can give the payment recipient pretty good confidence that the transaction will be finalized. I believe some rollup providers are either doing this already or currently researching it.

Re: Bank transfers as a payment method

#58
post #47

Earlier quoted context omitted.

If it could work, it would have worked three decades ago. Bank transfers are just a pain in the ass. I still remember all the fees I paid just to transfer money to another bank/another country. Cryptocurrency was considered a way to simplify the problem, but probably didn’t work in this aspect

India, a third world country has worked it out. Apart from UPI, we have other bank transfer mechanism that doesn't involve writing a check. There's RTGS (Real Time Gross Settlement) for large payments. It's what we use when we buy property or do large business transactions. There's NEFT (National Electronic Fund Transfer) for instant transfers upto INR 500K. There's IMPS (The underlying mechanism of UPI) for instant…

1. India is a single country, cryptocurrencies are global

2. Nobody uses proof of work anymore (besides bitcoin)

Re: Bank transfers as a payment method

#59

Earlier quoted context omitted.

If it could work, it would have worked three decades ago. Bank transfers are just a pain in the ass. I still remember all the fees I paid just to transfer money to another bank/another country. Cryptocurrency was considered a way to simplify the problem, but probably didn’t work in this aspect

I'm in the lucky situation that locally I can transfer instantaneously or nearly so across all of Europe and internationally I use bitcoin for recipients in places where banking fees are too high or the transfers too slow. It's a good combination.

Why not a stablecoin with lower fees?

Re: Bank transfers as a payment method

#60
post #4

This is all great if the currency between the buying and business is the same (ie, USD to USD), but still does not fix transactions between entities of different currency types. In my opinion, this is where cryptocurrency as a payment method for B2B, B2C will shine. No more dealing with foreign transaction fees and conversion fees. Plus the added benefit of not converting your money into an unstable currency (ie, Tur…

You could also just have a bank account with money denominated in USD or Euros or something.

You can have a cryptocurrency wallet with money denominated in USD or Euros, too (asset-backed stablecoins).
Post reply on HN