UPI is quite revolutionary. Launched in 2016 [0], UPI handles 882.3 million transactions with ~USD 21 billion in transaction value [1], as of July, 2019.
UPI is designed to be a payment utility by National Payments Corporation of India (NPCI). NPCI is not a governmental organisation but is formed by a collective of 56 Indian Banks including 10 core banks [2].
NPP is on it's way to being really handy, but it still sucks that I have to hack things to force NPP for some payments
(ie. CommBank will only use NPP for PayID payments, but banks like Up will use NPP for transfers for a BSB/Account number if the receiving banks supports it, not all banks are fully on the platform yet)
The downside is my teenagers can now shake me down for cash remotely
That transaction limit is only a bank imposed limitation - while out of the ones I checked, Commonwealth (and I assume some other) does have a $1K daily limit for personal accounts, various other banks appear to have higher limits or just use the regular transaction limit (which you can usually choose yourself).
If you regularly need to move 250k sums electronically inside the Australian economy, using this kind of mechanism, I'd love to know what you're doing. Only because I am nosy: I cannot conceive of transactions on this level which demand NPP class behaviour.
> If you regularly need to move 250k sums electronically inside the Australian economy.
Paying sub-contractors, settling property transactions, the list goes on.
When buying a house, I remember having to withdraw $60k in cash from my savings bank account and walking over and depositing it in another because it would take the other bank 3 working days to clear a bank-cheque. Some banks do offer an over-night inter-bank mechanism, but surprisingly CommBank does not.
I don't think it was designed for what you want to do. Interesting problem: fast efficient payment of GST-incurring and other taxable money in a low doc manner.
I'm fairly sure this is to ease it into use. I would assume that as it becomes more ubiquitous (many banks don't even have it yet) the limit will increase.
> Guidelines for the Brazilian Instant Payments Ecosystem: The Brazilian instant payments ecosystem will have a flexible and open participation structure and will be similar to the existing interbank electronic funds transfer (TED)—with greater availability, speed, and simpler addressing of payments. The transactions will be performed in a simple and fast way to different beneficiaries. Moreover, the BCB (Brazilian Central Bank) will develop a single settlement infrastructure.
While BACS' 24h clearing of interbank transfers was normally ok, FasterPayments really makes having multiple accounts split across bank's super convenient.
My wife and I both have our own accounts with Barclays and joint accounts (and mortgage) with Nationwide, and I have a Monzo account for small day to day expenses. It takes me less than a minute to open the Barclays app, transfer some money to the joint account, and then open the Nationwide app to confirm the balance and do whatever I needed to do from there. Total money management game changer. I've done a four hop transfer from my savings to overpay the mortgage, in about three minutes.
“...In its simplest form, a completed payment through the FedNow Service involving two participating banks would have the following steps.103 To start, a sender would initiate a payment through its bank, by submitting instructions to it using an end-user interface outside the FedNow Service. After the sender’s bank authenticates the sender and validates the payment, it would submit a payment message to a Reserve Bank using the FedNow Service. The FedNow Service would authenticate the sender’s bank and validate the payment message, for example, by verifying that the message meets the FedNow format specifications. Before the Reserve Bank executes the payment message, the service would place a provisional hold on funds in the master account of the sender’s bank and would then send an inquiry message to the receiver’s bank seeking confirmation that the receiver’s bank, among other things, maintains a valid account for the receiver included in the payment message received by the Reserve Bank. If the receiver’s bank sends a positive response to the inquiry, the FedNow Service would execute the payment for the Reserve Banks by sending a payment message forward with an advice of credit to the receiver’s bank and nearly simultaneously processing a final debits and final credit to the master accounts of the sender’s bank and receiver’s bank, respectively.104 The banks are responsible for debiting and crediting their customers’ accounts and providing further notification to their customers that the payment has been completed. The entire process would take place within seconds...”
Is the primary role of the Fed in this case as an arbiter for disagreements? Why can't the banks achieve the same directly between each other using an agreed upon protocol?
> [Interoperability between all banks has] been one of the arguments for blockchain-based payment systems like Facebook's (FB) Libra, which would circumvent the need for third-party payment settlement. During two Congressional hearings on Libra last month, some lawmakers called on the Fed to create a real-time payment system itself rather than having to rely on a network developed by Facebook.
It's a reminder that blockchain solutions aren't as useful if there actually is a trusted central entity, upon whose sterling reputation the global economy relies in the first place, to record transactions. And it does speak to a mission to regulate and facilitate interstate commerce. If they get the branding and execution right this could be huge for enabling micro-entrepreneurship. Of course, they won't get it right, but one can dream.
I expect banks to start charging a $65 per transaction premium to interface with this system.
FedACH, or "direct deposit", costs pennies wholesale [0]. The value proposition of Venmo/Zelle is not just that they're instant, but that the fees are not insane. Compare that to retail banks which charge upwards of $35 for a "next-day ACH".
The Fed should impose fee caps to encourage adoption.
The real cost to ACH is fraud, not the electronic processing, per se. The faster a payment settles the easier and more profitable fraud becomes as it makes it easier to cash out and run.
Low fees, fast transfers, high volume/high value--you only get to pick two of the three. (Okay, maybe that doesn't quite work, but you get the idea.)
Once upon a time I heard a quote from a PayPal CEO that was something along the lines of "we used to have a lot of competitors, but they all failed because they thought we were all in the business of transferring money. We're actually in the business of fraud prevention."
I'm not exactly an expert on clearing, but isn't the reverse also true? FedACH's 24+ hour turnaround makes a whole slew of scams (e.g. fraudulent cash-in, legit cash-out[1]) possible that wouldn't be with an instantaneous system.
[1]: I send you, good samaritan, $500 "by accident" and ask you to return it. You return it, my original transaction never clears, and I'm $500 richer.
That's how check kiting works, and AFAIU banks long ago stopped permitting withdrawal of funds that haven't cleared, unless they have other reasons to trust you or to claw back their money.
Perhaps there still are modern equivalents to kiting. I dunno. But, yeah, in any event fraud is a real problem. It's why we can't have nice things, at least without paying an insurance premium.
I think reversibility/non-immutability is one of the sides of this iron triangle, as Bitcoin currently supports all three you listed, but is a fraud risk because while you can transfer a practically unlimited amount globally for $1 in an hour, there is no “undo”, which is, in various ways, a blessing and a curse.
Bitcoin holder and user for ~5 years here. As a comparison of two of those features, against Faster Payments (which is what the Fed is proposing creating a similar system to):
Bitcoin is slow - it takes minimum 10 mins with high fees, or in times of congestion, hours or days with low fees. Faster payments takes minimum a few seconds, maximum 2 hours, for processing, no fees involved.
Bitcoin is expensive - fees are variable depending on congestion on the network, and current value of BTC. Faster Payments charge nothing to transfer 0.01 - 100,000.00 GBP (in reality, 10k - 100k, depending on your bank. I have banks that allow 25k and 50k) in a single transaction.
Bitcoin is a wonderful experiment, wonderful proof of trustless immutable transactions, but as far as a payment system goes, is only better in one way, and worse in many others.
The fees are variable with congestion, sure, but in practice are very low, and even when “high” are less than what some people are mentioning the retail cost of a single one of these new fed transactions might cost (routinely, not congested).
Also, you cite speed, but Bitcoin transactions confirm sufficiently in the usual case in 20-40 minutes, where two hours is the upper bound you mentioned for the new fed system.
Bitcoin is also functioning today, isn’t censorable, and is transnational. The new fed system isn’t due to be operational for years, and will only serve a single country with fewer than 4% of humans living in it.
I think you are exaggerating the downsides of code that is working and providing value today.
Faster Payments exists today, costs nothing to the end user, and usually takes seconds.
Bitcoin processes 200-500k payments per day[0].
Faster Payments processes 230k payments per hour[1].
Bitcoin may be superior to what currently exists in the USA as provided by banks, but it isn't compared to existing private sector payment solutions, and it definitely isn't compared to what other countries have already implemented successfully in their banking system.
Again, long time bitcoin holder and user, I'm painfully aware of the downsides having paid absurd fees and still ended up waiting hours for a transaction to finally get accepted into a block.
Bitcoin's great as a speculative asset, but it's not so great as currency.
Venmo is free for non-commercial uses and there are weekly or monthly limits.
If you want to accept venmo for commercial purposes or more cash than is usually allowed, they charge you as if it were a credit card transaction. 2.9% + 0.30.
I think they’re talking about wire transfers. Those are $35 at one bank I use (which automates them) and free at another (which makes you call someone and fill out PDFs in triplicate).
FedACH is already widely used- this makes perfect sense. All US Banks are part of the Federal Reserve system, so it makes sense for the FRB (which is the entity being discuessed here) to drive this with the help of each Federal Reserve branch.
I'm still hoping for something similar where I currently live, here in Switzerland (not in the EU). Payments generally arrive the next business day but that isn't good enough for most in-person transactions.
Is Switzerland working on something along those lines now? I'm surprised, given the outsized Swiss position in global banking, that they're not among the leaders on this front.
Swiss banks are not leaders on anything now... They are just expensive. I pay 10 swiss francs per month to have an account and card there. And they still want 5 more so I can make more than 10 transactions a month (which I want since I started paying contactless in shops). And there aren't really any good alternatives, any bank is like that.
And I pay 0 swiss francs for an account + card (you need more than 25k of assets, else its 5...But any product such as third pillar retirement funds count!) with unlimited transactions and cash-withdrawal at any ATM. You probably have to shop around.
The reason why banks are charging now for simple accounts, is that since we have negative interest rates (the lowest in the world), its actually damn difficult for banks to make a profit now with normal accounts since the traditional business model of the difference in interest rates is now actually reversed (the target rate is negative, yet on my account I have 0%...).
If I lived here and wasn't eligible for a student account as I am now, I might honestly consider just using one of the US accounts which allows for free ATM withdrawals anywhere. It seems absurd to have to pay a monthly fee just to have an account.
There do at least seem to be a few free (with no minimum balance) app-based checking/private accounts.
Worst-case Congress can pass a law imposing onerous requirements on digital currencies backed by a basket of something. Best-case Libra has to comply with a bunch of existing money transfer regulations that were never written with cryptocurrency in mind and thus are open to interpretation; various agencies may choose stricter or looser interpretations depending on which way the winds are blowing.
Which in itself is one of the biggest rationales for developing such a system. The Fed really has a third unspoken directive: maintain dollar supremacy.
Maybe in the United States, but I would still wager there would be interest for a Facebook's digital currency in countries with more volatile, native currencies.
Libra can still provide hassle-free currency access from within supporting apps and Web sites. Also it can provide "free" transactions whereas FedNow won't be free.
For 99% of legal money senders this removes the need to consider Bitcoin or most shitcoins.
The only use case left for crypto, which Bitcoin itself has somewhat neglected, is anonymous uncensored payments. And that's the one that doesn't work because it's extremely hard to get in and out without being noticed.
The are actually/ could actually be talking about xrapid blockchain inter ledger protocol from Ripple, to settle within 3-5 seconds, they just can't say it yet.
Coming from a country that has had a modern payments system like this for a while - you generally still use a credit card for anything online or at stores, in an Uber or taxis, but use the real time bank payment system for paying invoices and bills, or buying a car from somebody, or sending $12 to your friend in real time because he paid for your lunch etc.
I was lumping in debit cards/EFTPOS with credit cards (and of course cash is also an option in stores). If not paying with cash, you use cards/contactless in stores because the terminals are integrated into the POS systems, have instant approval feedback etc. (and literally every card terminal here has contactless that works with pretty much any card and Apple Pay etc. too, which is nice).
To use a bank transfer you need to enter details (phone number or a bank number and account number) and then need either a description or reference number to match it to at the other side. And even with the New Payments Platform (fast payments), it can still take 30 seconds or so to come through. That's fine when paying an invoice or sending money to friends, but too cumbersome at a point of sale.
What do you mean? There’s usually no way to pay without credit card / debit card / cash in most countries. Contactless payments are still not as popular as they could be.
This system will need to interface with other domestic and international settlement and payments networks.
There is thus an opportunity for standards, a need for federation, and a need to make it easy for big players to offer liquidity.
As far as I understand, e.g. Ripple and Stellar solve basically exactly the 24x7x365 RTGS problem that FedNow intends to solve; and, they allow all sorts of assets to be plugged into the network. Could FedNow just use a different UNL (Unique Node List) with participating banks operating trusted validators and/or offering liquidity ("liquidity provisioning")?
Notably, Ripple is specifically positioned to do international interbank real time gross settlement (RTGS) and remittances. Ripple could integrate with FedNow directly. Most efficiently, if it complies with KYC/AML requirements, FedNow could operate an XRP Ledger. Or, each bank could operate XRP Ledgers.
https://xrpl.org/become-an-xrp-ledger-gateway.html
Getting thousands of banks to comply with an evolving API / EDI spec is no small task. Blockchain solutions require API compliance, have solutions for governance where there are a number of stakeholders seeking to reach consensus, and lack single points of failure.
Here's to hoping that we've learned something about decentralizing distributed systems for resiliency.
>> In contrast, the XRP Ledger requires 80 percent of validators on the entire network, over a two-week period, to continuously support a change before it is applied. Of the approximately 150 validators today, Ripple runs only 10. Unlike Bitcoin and Ethereum — where one miner could have 51 percent of the hashing power — each Ripple validator only has one vote in support of an exchange or ordering a transaction.https://news.ycombinator.com/item?id=19195050
So, you want to get banks onboard with only one s'coin USD stablecoin; but you don't want to deal with exchanges or FOREX or anything because that's a different thing? And, this is not just yet another ACH with lower clearance time?
> Interledger provides for secure payments across multiple assets on different ledgers. The architecture consists of a conceptual model for interledger payments, a mechanism for securing payments, and a suite of protocols that implement this design.
> The Interledger Protocol (ILP) is the core of the Interledger protocol suite. Colloquially, the whole Interledger stack is sometimes referred to as "ILP". Technically, however, the Interledger Protocol is only one layer in the stack.
> Interledger is not a blockchain, a token, nor a central service. Interledger is a standard way of bridging financial systems. The Interledger architecture is heavily inspired by the Internet architecture described in RFC 1122, RFC 1123 and RFC 1009.
[...]
> You can envision the Interledger as a graph where the points are individual nodes and the edges are accounts between two parties. Parties with only one account can send or receive through the party on the other side of that account. Parties with two or more accounts are connectors, who can facilitate payments to or from anyone they're connected to.
> Connectors provide a service of forwarding packets and relaying money, and they take on some risk when they do so. In exchange, connectors can charge fees and derive a profit from these services. In the open network of the Interledger, connectors are expected to compete among one another to offer the best balance of speed, reliability, coverage, and cost.
Why should we prefer an immutable, cryptographically-signed blockchain solution over SQL/BigTable/MQ for FedNow?
Keep in mind that what the Fed is building will just be a backend service for facilitating instant money transfers, not an end-user product. It will be up to the banks to develop the front-end for their customers to use.
FedLater℠ Service.. Joking aside, this area is ripe for an overhaul/redesign, and I'm glad they're working towards it.
> The rapid evolution of technology presents a pivotal opportunity for the Federal Reserve and the payment industry to modernize the nation's payment system and establish a safe and efficient foundation for the future.
UK - Faster Payments Service (2008) https://en.wikipedia.org/wiki/Faster_Payments_Service
EU - SEPA Instant Credit (2017) https://en.wikipedia.org/wiki/Single_Euro_Payments_Area#SEPA...
Australia - New Payments Platform (2018) https://en.wikipedia.org/wiki/New_Payments_Platform