Brazil's central bank operates their instant payment network Pix [1] [2] [3] for ~$10M/year [4]. Its not that these are small contracts, but that large, inefficient, unnecessary contracts have become the norm (I argue). Similar example from India's UPI payment system [5]. The US has FedNow to move instant payments for pennies, but banking and payment system participants in the US ecosystem are avoiding it to continue to private payment system rake [6] (cc networks, Zelle commercial bank network, private wallets, etc).
The evidence is clear you don't need to skim 3% off of an economy to provide instant payment capabilities. The enterprise value of US payment companies is a function of how long they hold onto this volume for, when competition is ramping up. You're just pushing ISO 20022 XML messages around a bus.
> This makes the American dispute more sophisticated than it may first appear. Pix certainly puts pressure on private payment models, card networks and acquirers. It also reduces friction for consumers, small businesses and person-to-person transfers. But its deeper effect is institutional. It turns the bank deposit into an even more efficient payment instrument — and, by doing so, changes the role of banks in liquidity intermediation.
> There is an irony here. For decades, the United States built the narrative of private financial innovation. Brazil, through a public, interoperable and massively adopted system, produced one of the world’s most efficient payment infrastructures. The study notes how unusual Pix adoption was: more than 150 million users in its first year, use by nine out of ten small businesses, and daily volumes capable of reaching about 1% of annual GDP on a single peak day.
> The reading should not be triumphalist. Pix is a powerful innovation, but it is not cost-free for the financial system. It improves the user experience, reduces transaction costs and increases competition in payments. At the same time, it requires banks to hold more liquidity and may reduce the transformation of deposits into credit. For the United States, Pix appears as a digital-trade issue. For Brazil, it is a question of financial sovereignty. For banks, it is a question of liquidity. Pix began as a button inside an app. It became a piece of financial policy — and now, of geopolitics.
The BCB in Brazil does very little to operate Pix. It's effectively a P2P system, where the BCB forces all the banks to interop with one-another (and all the banks directly call eachother). They can operate it that cheaply because they do close to nothing technically (they host the main discovery endopints). The only place the BCB actually ingests data is via their reporting mechanisms.
UPI is a bit more centralized, where the NPCI does the top-level routing between banks, so their operating budget is likely much higher than Pix. It also is drastically more simple to be a participant in UPI compared to Pix.
For Pix adoption: you can thank Covid for that. The Brazilian government said if you wanted to get free money from the government, you had to set up and use Pix.
US Financial Innovation: I'd say the hard thing here is that the government is extremely strict (lots of regulation) when you start looking like a bank. Lots of companies have tried to innovate here, but regulation makes it really hard to do. There's a lot of regulator capture going on.
This is incorrect . The system is not P2P, while it appears to be for customers. The BCB keeps both the directory that ties a chave pix to an actual account (DICT), and a centralized ledger (SPI - Sistema de pagamentos instantaneo)
│
(2. Send order)
▼
[ SPI - Central Ledger] ──────(3. Real time settlement over Financial Institutions account: Debt/Credit)
│
(4. Notification Message)
▼
[Receiver Bank]
The ledger doesn't keep individual accounts, but a Instant Payments account for each institution. This account is not the master bank account in the Reserve Transfer System (Sistema de Transferencia de Reservas) which is the system of record for banks funds, so the banks need to allocate funds from the STR to the SPI every day to be able to honor PIX transfers. The STR system doesn't work out of normal banking hours, so the banks need to predict how much money they will need for PIX transfers and move that money from the STR to the PI (pagamento instantaneo) account during defined liquidity transfer windows, to avoid banks double-spending the same funds over different rails (traditional vs PIX). If a bank finds itself without funds on its PI account in a saturday night, it can loan the funds from another bank who still have excess liquidity on his PI account).
As you can see, between the SPI, STR an DICT, it is a very centralized system.
Also, the system operates in a dedicated zero-trust networks which is completely isolated from the internet. The messages between banks and the central bank follow the ISO 20022 format. IBM MQ is used to route messages back and forth between banks and the BCB.
In both Brazil and the US, making something like Pix mainstream requires regulation. Brazil did its part, but it’s unclear whether the US will do the same anytime soon, I don't think so, unfortunately. Technically, it would benefit the American population (the process of making pix payments is so smoothless), but payment companies may have little incentive to support it, and could lobby against it.
In Brazil, the Central Bank overpowered the coordination problem. In the US, it may be the opposite: the government seems to have less power over the payment lobby, or at least less willingness to confront it directly.
Thanks for that. I am wondering if anyone have written up a high level technical comparison between different payment systems including Pix, UPI and others.
Feels odd that you exclude mentioning the EU, which has had instant transfers for more than a decade. More than two decades, if you include things like iDeal from The Netherlands.
Wasn't intentional, I mention SEPA and Wero in other comments, not intended to be an enumeration of all instant payment systems currently active globally. My apologies!
> not intended to be an enumeration of all 54 instant payment systems currently active.
Even then, not mentioning those who pretty much started / invented instant transfers still seems odd :) but no need to apologize haha, maybe I was a bit too abrasive.
I get why you prioritized to mention those though. The Chinese and Indians have leapfrogged us. No more fussy legacy (digital) cards, just scan a QR and go. Even illicit food stalls and street wanderers have accounts, when they wouldn't be able to get a 'real' bank account.
And the Chinese and Indians don't have to pay tribute to the Mastercard-Visa overlords either. Although Wero and the digital Euro might eventually change that for Europe too.
Can you imagine how cumbersome conversations would become if people felt obligated to qualify ad-hoc statements with what amounts to a historical ledger?
Every new entry would open up an opinion around “if you included that, why didn’t you include this?”
The comment literally mentioned how unusual Pix adaption is and how innovative it is. It is neither of those things. Because of aforementioned history.
don't centralized payment systems like this reduce the overall resilience of the ecosystem and prevent future innovation? You hint on those lines with the possible future transformation of deposits into credit.
Why doesn't the US private ecosystem manage to lower costs similarly? (Zelle comes to mind). It is interesting that this has happened in more highly regulated countries where the free market likely could not have come up with a cheaper solution on their own due to the same overbearing system that effectively forces adoption of this centralized solution.
All payment systems are centralized. Zelle is owned by the largest US commercial banks ("Early Warning Services"), Congress directed the Federal Reserve to build and offer FedNow as a utility so smaller banks would not be excluded from offering instant payments. It costs $~30/month (last I checked the rate sheet) to plug into it. The instant payments are the utility, your opportunity to innovate is using this as a component of your user experience.
Propose some innovation here, I am interested, as someone adjacent to payments in financial services. Besides instant payments, the most we've seen is closed wallets (Venmo, Cash App) no longer needed with broad instant payment access from most demand deposit accounts and Buy Now Pay Later (BNPL) (and I argue BNPL is simply dressing revolving credit card debt up as innovation).
> Why doesn't the US private ecosystem manage to lower costs similarly? (Zelle comes to mind). It is interesting that this has happened in more highly regulated countries where the free market likely could not have come up with a cheaper solution on their own due to the same overbearing system that effectively forces adoption of this centralized solution.
Because it is a grift ("regulatory capture") [1] [2]. The "overbearing system" is the result of regulation to bring the consumer excess of cheap payments to an entire country's financial user population. Why does Jamie Dimon not like stablecoin yield [3]? Because JPMC makes almost $100B/year in interest income taking customer deposits and lending against them, which stablecoins would compete against by operating as a form of narrow bank, parking the underlying deposits in risk free US Treasuries [4].
As a US financial services consumer, it is hard for you to avoid the rake of the machine built to skim off of you as you hold onto fiat or move it, but the rest of the world can avoid being captured by it (as this piece demonstrates). Also, Europe can't regulate Stripe as easily as they can Adyen. You don't have to be the biggest or the greatest, it just has to work "good enough".
> Propose some innovation here, I am interested, as someone adjacent to payments in financial services. Besides instant payments, the most we've seen is closed wallets (Venmo, Cash App) no longer needed with broad instant payment access from most demand deposit accounts and Buy Now Pay Later (BNPL) (and I argue BNPL is simply dressing revolving credit card debt up as innovation).
UPI for instance only works with a physical SIM. Your phone number on the account must match the physical SIM on the device. This indirectly relies on India's insistence on KYC (for accounts naturally) on issuance of physical SIMs. "Innovation" here would be a player who can support VOIP based phone numbers (maybe by complying with phone number KYC in some other way).
UPI also makes it quite confusing to deposit money to a particular account you own. You could share a specific identifier (string or qr) based on your account but the other party generally assumes they can send you money using your phone number, and sometimes follows through with that.
(I don't have a finance background.) There any multiple instances of a one-size fits all user experience decision which strikes me as a result of the centralization and removal of competition (in efforts to drive up adoption).
I don't disagree with most of your reply (thanks for the thoughtful citations too). But i wonder why the free market cannot lower cost/settlement time similarly.
> UPI for instance only works with a physical SIM. Your phone number on the account must match the physical SIM on the device. This indirectly relies on India's insistence on KYC (for accounts naturally) on issuance of physical SIMs. "Innovation" here would be a player who can support VOIP based phone numbers (maybe by complying with phone number KYC in some other way).
The Indian government has mandated this for strong identity assurances. Your only hope at "innovation" (ie violating financial services regulators and laws) here is cash or something like Monero.
> UPI also makes it quite confusing to deposit money to a particular account you own. You could share a specific identifier (string or qr) based on your account but the other party generally assumes they can send you money using your phone number, and sometimes follows through with that.
I haven't used UPI recently, but I imagine this is a UX issue around aliases (phone numbers, email, and other human identifiers that associate to an underlying account).
TLDR People problems cannot be fixed with tech (in this context, regulatory requirements or alias UX, submit a public comment to the regulator if you can).
> I don't disagree with most of your reply (thanks for the thoughtful citations too). But i wonder why the free market cannot lower cost/settlement time similarly.
Because without regulation, it turns into Monopoly (the board game). Sometimes, competition can be encouraged, but in some cases (broad, shared infrastructure) it cannot and regulation must fill this gap to ensure the target outcome. This is why we regulate electric utilities similarly. Happy to help, I am very interested and curious on this topic.
> Propose some innovation here, I am interested, as someone adjacent to payments in financial services.
Well, as a brazilian who is used to pix and has also faced the bad payment ux in american, I think a possible solution should be adding the missing Pix-like layer above the existing US rails.
One of the best part that makes Pix an incredible experience, It’s that the app almost doesn’t matter. I can use one bank, you can use another, a merchant can use a different provider, and it still works through the same basic language: QR code, Pix key, payment request, confirmation screen. I can even transport my "pix key" to another app/bank provider.
So maybe the US opportunity is a agnostic interoperability layer on top of FedNow/RTP/Zelle/bank APIs: universal aliases, QR payments, routing, fraud checks, receipts, and reconciliation. Making instant account-to-account payments feel universal before the government forces a universal standard.
I don't think consumers broadly should be the main goal first, the best initial path should be small-business, marketplaces, rent, etc.
If someone could own that neutral UX/addressing layer, that seems much closer to the useful part of Pix than just another closed wallet.
None of which are in production at scale. I admit crypto is optimal for less than legal transactions and speculation, but the volume for legal payment transactions is negligible.
Solana mainnet was shown to have capability to handle 100k transactions per second [1]. Granted, it was a synthetic benchmark of noops, but it shows the capability is there.
For reference, Visa claim 83k TPS [2] for their system. It also has the advantage of being vastly cheaper than tradfi payment networks at ca. $0.0005 per transaction [3] irrespective of transaction amount.
Until it is in prod, it is a proof of concept. Blockchains solve for low trust; if you have trust, you don't need the efficiency loss of a decentralized ledger or blockchain. Central banks provide trust.
As mentioned in one of my other comments, Pix in Brazil costs ~$10M/year. They process ~6-8 billion monthly transactions and roughly $6.7 trillion in payment volume a year [1]. That's roughly ~$0.0015/transaction based on the math in this comment, and we don't know what the ceiling is based on existing capacity (which would drive per transaction costs down further). Choose boring technology, when possible [2].
The innovation in this context is nuking the profits of Visa and Mastercard (their margins are ~45-50% [3] [4] [5]), replacing them with central bank instant payment systems run at cost. The reduction in their revenue is money back in the pockets of everyone paying unnecessarily to move value around. I highly recommend the book "The Innovator's Dilemma" on this topic [6].
Pix is really good but I don't think it beats Solana or Ethereum L2s.
Most recent indicator of peak Pix transaction volumes I could find [1] was 227M/day (=2700 TPS). You can see yourself that Solana does 130-140M/day consistently. Pix fees you quote are still triple Solana's.
Not to mention there is the entire decentralization aspect, which means the government does not control your money as with other blockchains.
Fraud can be managed at instant payment rails, you don't need credit card rails to manage it; fraud must be managed on any rails utilized, so it is somewhat agnostic. If you as a consumer want credit card chargeback protection insurance, push the fee onto the consumer with a surcharge to cover the cost. I believe the evidence is robust credit card rails are unnecessary in today's world, plain and simple. They are a bloated legacy holding on for relevance (and their grossly excessive margins) imho.
Common misconception - it's the merchant who pays for fraud and the 3% is pure bank profit. If there's a fraud transaction, the merchant has to pay the whole amount back and ends up negative the transaction fee and the chargeback fee. The banks just want you to think they're earning the 3% but actually it's pure profit.
Most blockchain systems are fairly or even extremely centralized as well and amount to little other than decentralization theater. Bitcoin and Ethereum are arguably more of an exception than the norm.
The evidence is clear you don't need to skim 3% off of an economy to provide instant payment capabilities. The enterprise value of US payment companies is a function of how long they hold onto this volume for, when competition is ramping up. You're just pushing ISO 20022 XML messages around a bus.
[1] https://en.wikipedia.org/wiki/Pix_(payment_system)
[2] https://frontierfintech.substack.com/p/55-send-pix-brazils-i...
[3] https://brazilstockguide.com/behind-the-lines/the-cost-of-pi...
> This makes the American dispute more sophisticated than it may first appear. Pix certainly puts pressure on private payment models, card networks and acquirers. It also reduces friction for consumers, small businesses and person-to-person transfers. But its deeper effect is institutional. It turns the bank deposit into an even more efficient payment instrument — and, by doing so, changes the role of banks in liquidity intermediation.
> There is an irony here. For decades, the United States built the narrative of private financial innovation. Brazil, through a public, interoperable and massively adopted system, produced one of the world’s most efficient payment infrastructures. The study notes how unusual Pix adoption was: more than 150 million users in its first year, use by nine out of ten small businesses, and daily volumes capable of reaching about 1% of annual GDP on a single peak day.
> The reading should not be triumphalist. Pix is a powerful innovation, but it is not cost-free for the financial system. It improves the user experience, reduces transaction costs and increases competition in payments. At the same time, it requires banks to hold more liquidity and may reduce the transformation of deposits into credit. For the United States, Pix appears as a digital-trade issue. For Brazil, it is a question of financial sovereignty. For banks, it is a question of liquidity. Pix began as a button inside an app. It became a piece of financial policy — and now, of geopolitics.
[4] https://news.ycombinator.com/item?id=44753626
[5] https://en.wikipedia.org/wiki/Unified_Payments_Interface
[6] https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...