[HN Gopher] Payment fees matter more than you think
___________________________________________________________________
Payment fees matter more than you think
Author : dxs
Score : 86 points
Date : 2026-03-03 18:27 UTC (4 hours ago)
(HTM) web link (cuencahighlife.com)
(TXT) w3m dump (cuencahighlife.com)
| ezfe wrote:
| 4% seems high. Quick googling in the US (which has high rates)
| shows 1.5-3.5%, avg of 2%
| Groxx wrote:
| You might be surprised to hear that my small business sometimes
| sees fees at _11%_.
|
| We blocked that card processor, obviously. But the % is very
| much not constant across e.g. Visa, often every purchase in a
| day is slightly different, and we can't even tell people what
| the rate is for their purchase due to a couple layers in
| between (still figuring out if we can fix that). It's vile, and
| probably should be illegal to not pass through the cost
| visibly.
| ezfe wrote:
| Stripe is 2.9%+30C/ right, and that's the advertised rate. So
| I assume any business seeing averages higher than that can be
| avoided by using a platform like Stripe.
| Spooky23 wrote:
| Small merchants are people, and people vary in their
| intelligence/savvyness and may have other issues like poor
| credit or high chargebacks that they usually forget to
| mention. Some people roll the cost of the terminal into a
| higher fee as well.
|
| I helped out a friend who owns a deli when he took over
| from his parents. His dad saw cash as a way to avoid
| taxation and had some awful payment processor where they
| paid a high fee and was renting a POTS based terminal -
| $60/mo to Verizon and $30/mo for the terminal.
|
| Now he keeps one set of books, and raised his average sale
| by about 10%. Their catering business, which drives
| profits, are up significantly with online ordering through
| the POS.
|
| Ditto with a non-profit I was on the board of. Pushing
| Venmo and Square for donations increased donations by like
| 30% and reduced shrink at fundraisers. Anyone who claims
| they can't afford a 3% fee is going out of business anyway.
| havaloc wrote:
| The Stripe rate is a careful blend. There are many cards
| that are cheaper to process and there's probably a few that
| are more expensive to process at that rate, it works out in
| the wash in favor of Stripe. Also, that 30 cents is a large
| percentage if you're looking at a $5 transaction, for
| example (6%!)
|
| See the "raw" rates here: https://www.mastercard.com/conten
| t/dam/mccom/us/business/doc...
| fragmede wrote:
| Why not just pay Stripe/Adyen/Braintree to smear the fees to
| 2.9% + $0.30?
| MengerSponge wrote:
| There are a few cards that offer 2% cash back with no annual
| fee. No chance their fee is 1.5-2%
| wccrawford wrote:
| Cards don't make money from their fees. They make money from
| people who fail to pay and then pay the ridiculous interest.
| apparent wrote:
| Do people who pay ridiculous interest qualify for 2% cards?
| Honest question; I don't carry a balance so have no idea
| what is advertised at other types of consumers.
| SoftTalker wrote:
| Why not? I'd gladly pay you 2% of $1,000 if you pay me
| 21%
| apparent wrote:
| I was just under the impression that the cards with the
| best benefits were somewhat harder to get. I do
| understand that credit card companies make money on
| interest and late fees, so they should find consumers to
| be attractive so long as they ultimately pay the
| bill/interest.
|
| I guess the question is whether they can distinguish
| between people who are going to carry a balance but
| ultimately pay and people who are a true
| default/bankruptcy risk.
| bumby wrote:
| Interchange fees seem to be a sizable portion of revenue.
| Discover has listed them as 29% of revenue, BoA at ~$10B
| annually...
| drdec wrote:
| Revenue does not equal profit
| bumby wrote:
| If you could identify where fees get decoupled from
| profit in finance, I'd be open to the position that they
| aren't related but you didn't share anything along those
| lines. Considering the costs like cash-back programs that
| would erode the profitability of those fees are largely
| in the banks control, I don't think that is a strong
| position.
| konschubert wrote:
| In the EU, we now have instant, free SEPA bank transfers.
|
| I know that the banks are trying to build a payment solution on
| top of this technology but it's not really getting traction.
|
| I am wondering if there is a way to bootstrap something bottom-up
| by offering something to merchants that has a clear value prop.
| HoldOnAMinute wrote:
| There is always opportunity in FinTech.
|
| How did Venmo and Cash app get any traction? After all, we
| already had PayPal. There was already a way to transfer money
| to your friends.
|
| How did Robinhood get any traction? We already had Etrade and
| other online brokers.
| yobbo wrote:
| Usually by giving away free money to reach critical mass.
| mrweasel wrote:
| Probably not worth it, given that the EU caps the credit card
| fees at 0.3%, 0.2% for debit cards.
| carlosjobim wrote:
| You're putting the cart in front of the mule. The important
| thing is not what merchants want, but what customers want.
| aleph_minus_one wrote:
| > The important thing is not what merchants want, but what
| customers want.
|
| What many people in Germany want is a payment system that is
| as anonymous and is as hard to control by some untrusted
| entitity (both government and banks are very distrusted) as
| possible and what cash offers. That's basically cash.
|
| Not without reason, in Germany there exists the well-known
| phrase "Bargeld ist gelebte Freiheit" ("cash is lived
| freedom").
| carlosjobim wrote:
| Agreed. Customers are benefitted either by paying in cash -
| for the reasons you described - or by paying with cards,
| for fraud protection and the ability to make purchases
| online.
|
| Any other payment method will not give customers any
| benefits over those methods. Unless banks are willing to
| take responsibility for fraud like with card purchases.
| tdi001 wrote:
| There is already the EPC QR code, which contains all the data
| required to initiate a SEPA credit transfer. This code is
| supported by practically all banking apps (at least in
| Germany). The standard is public and free (see
| https://en.wikipedia.org/wiki/EPC_QR_code)
|
| The merchant's system displays this code, you open your online
| banking app, scan the code, select "SEPA INST" (here's the
| usability catch!) to make the payment instantaneous, and
| confirm. Within 10 seconds, the money is transferred to the
| merchant's account. Either the merchant's bank or a third-party
| Open Banking API immediately informs the merchant's system
| (e.g. by push notification or webhook), and a receipt is
| issued.
|
| Everything is already here, but since this system would be
| virtually free to use, nobody really has an incentive to push
| it. It costs money to educate the public, and there is no money
| to be made. Instead, everyone gets paid handsomely by the card
| mafia.
| kro wrote:
| In general I'm all for free and European systems, but SEPA
| payments imo still have pain points:
|
| - you can send money to companies and individuals alike. It's
| easier to trick people into fake shop payments, a card
| payment provider requires at least a bit it
| verification/registration
|
| - it's really hard to dispute/call back sepa payments. The
| card companies often step in there afaik
| konschubert wrote:
| Yea, but I think that there is still a business model, if
| only in consulting or building software solutions to make
| this easier.
| ralferoo wrote:
| Wechat pay in China is interesting. It costs nothing to add money
| to your balance from your bank, or to pay someone from your
| balance. It only costs the end merchant who wants to withdraw it
| from their balance back into their bank. If they can keep it in
| Wechat pay and spend it on other things (which is very easy as it
| and Alipay are the primary payment methods for everything), then
| there's no charge.
|
| I guess Tencent are making their profit from the interest they
| earn on the money that was transferred into them that just stays
| in people's Wechat wallets in effectively a parallel currency.
| mihaelm wrote:
| And no doubt they'll find a way to spend it in the app
| considering you can manage almost all aspects of your life
| within it.
|
| Revolut works similarly. You don't pay any fees on transfers to
| other Revolut accounts, but you do for other bank accounts.
| johneth wrote:
| > Revolut works similarly. You don't pay any fees on
| transfers to other Revolut accounts, but you do for other
| bank accounts.
|
| Does it? I'd be surprised if it does in the UK at least, as
| all banks do free transfers to every other bank in the UK via
| Faster Payments. I thought it was the same in the EU?
| mitthrowaway2 wrote:
| Well, that definitely creates a powerful pull from customers to
| make vendors accept Wechatpay for transactions.
| touwer wrote:
| It's time for Europe to process the own money. Strange that the
| dominance of Visa/Mastercard/Maestro was left for so long. Of
| course there is a lobby from them to attack the digital Euro
| Finnucane wrote:
| One factor that the author glosses over a bit is that highest
| swipe fees are for credit cards with benefits. The interbank
| system he describes is a debit system, no credit is being
| extended. Even in the US, debit swipe fees tend to be less.
| saharshpruthi wrote:
| In India there is UPI (Unified Payment Interface), which works
| with all bank accounts, it's facilitated by the Government and it
| comes with i. QR Code (Used with strangers and at Merchants) ii.
| UPI ID iii.And links to phone number.
|
| Anyone can pay to anyone instantly free of Charge. Only limit is
| it's limited to ~ $1000 payment. The QR code can also be
| dynamically created by POS terminals containing the total bill
| amount as well, so upon scanning the amount is auto populated in
| the payment app, you just have to enter the security pin.
|
| And since it's a Govt. Project, its not limited to just one app,
| there are lots and lots of apps working on the same system. There
| is even a VISA/Mastercard credit alternative : RuPay that works
| within the system.
| 0x5FC3 wrote:
| Its limited to about $1000 _a day._
|
| The QR is a URI with the ID, amount and maybe other stuff. It's
| a client-side implementation.
|
| RuPay sure "works within the system" but is pretty much useless
| for international payments/subscriptions. Not really a
| VISA/MasterCard replacement.
| bigfishrunning wrote:
| So people scan a QR code, and then enter a secure banking
| pin? this sounds like a security problem waiting to happen...
| wiradikusuma wrote:
| The QR code doesn't open a link. It's just "gibberish" text
| only usable by app that can understand it (e.g. banking
| apps).
|
| (I don't know anything about UPI, but in Indonesia we use a
| similar system)
| Imustaskforhelp wrote:
| I am Indian and I think what you are saying is correct.
| It opens up the banking app or in our case UPI providers
| app so like Google pay, Phonepe,paytm, Bhim UPI and other
| such apps.
| porridgeraisin wrote:
| Its not gibberish text.
|
| Its just a URI.
| upi://pay?pa=payeeID&pn=payeeName
|
| You can add things like &am= to prefill the amount.
| Merchant txns have reference IDs and all that stuff.
| Egor3f wrote:
| In russia there is SBP (translated as FPS = "fast payment
| system") using the same mechanism, also free for individuals
| and relatively cheap for businesses
| montenegrohugo wrote:
| Shameless self-promo: We've been working for the last two years
| on making money movement free. We use ethereum (and stablecoins)
| for that, and integrate with the novel real time payment networks
| like PIX, Wechat, Yape, Mercadopago and so on.
|
| We're still orders of magnitude smaller than Visa and Mastercard,
| but I do believe products like ours (and competition is red hot
| here, theres so much good choice!) will be good for consumers.
|
| Money should be like a message: free and instant
|
| We're open source btw, happy to show off codebase and review PRs
|
| https://github.com/peanutprotocol/peanut-ui
|
| https://peanut.me/
| Imustaskforhelp wrote:
| What I want is being able to pay stablecoin from my side and be
| able to esentially get a virtual card of payment.
|
| The only way which has worked for me was actually using g2a
| with their crypto provider to get a 10$ gift card of rewarble
| and activate it.
|
| Does your app allow something like this? can I convert any
| crypto to payment provider which can be accepted from the other
| side or am I reading it wrong because you do mention PIX, if it
| can work with PIX, does it work with VISA/Mastercard cards too?
| intrasight wrote:
| Why not FedNow?
| kwanbix wrote:
| In Argentina we can transfer using our account number (or account
| Alias, for example my alias could be kwanbix) directly, account
| to account, instantly, it costs 0.
| wagwang wrote:
| Let's just call it for what it is.
|
| Visa and mastercard innovated in an era where payment settlement
| was notoriously difficult and expensive but they've used their
| monopolies to entrench themselves in (by negotiating deals with
| merchants and bribing consumers with points) while the rest of
| the world moves on towards "layer 2" payment systems that are
| much cheaper and efficient.
| jen20 wrote:
| The real thing consumers get is fraud protection, and the
| ability to charge back when merchants become intransigent.
| Let's not pretend other systems (even the ones prominent in the
| US like Zelle) offer the same protections.
| mikepurvis wrote:
| I've noticed recently indie (non-franchise) merchants being
| much more brazen about charging extra fees for accepting a
| credit card payment. This includes counters at my local
| farmer's market, two local cafes and a sushi restaurant, and my
| city's public electrical utility.
|
| All of them are happy to receive cash or interac (Canadian
| debit infrastructure) or even e-transfers in some cases
| (Canadian venmo). But they'll say an extra $1-2 charge if you
| want to pay by credit card.
|
| Maybe I'm just remembering badly, but I don't remember
| encountering this twenty years ago; back then the rules were
| clear that you either didn't accept credit payments, or you did
| and it was the same price as cash.
| bigthymer wrote:
| > Maybe I'm just remembering badly, but I don't remember
| encountering this twenty years ago; back then the rules were
| clear that you either didn't accept credit payments, or you
| did and it was the same price as cash
|
| My memory is in accordance with yours
| mixmastamyk wrote:
| Because credit card companies mandated that you couldn't
| raise prices to pay their fees. Believe this was later
| outlawed in the US and perhaps elsewhere.
| stevehawk wrote:
| in the US it was a class action lawsuit + Supreme Court
| decision
| carlosjobim wrote:
| Small business owners ("indies") are notorious for getting
| hung up on fees and costs which do not matter, while ignoring
| important savings and revenue sources. That's why they
| haven't grown to be bigger.
|
| A sensible business owner increases the base price a little
| to offset card fees instead of bothering customers with these
| details and losing sales.
| havaloc wrote:
| If a restaurant runs on a 9% net margin and pays around 3% in
| card fees, then roughly one-third of its net profit is going
| toward payment processing.
| matsemann wrote:
| It's weird how this works. Saw something similar when working
| for a bus company. After reaching a minimum amount of sales for
| a bus route, everything after that is basically pure profit.
| However, how do we get those last sales? Well, by bidding
| higher on people searching for transfer between those two
| cities. Let's say the ticket was $20. We could end up for
| instance accepting to bid $10 for an ad that would lead to a
| sale. So for every $10 of pure profit we then got, Google also
| got $10. In a sense it was a good deal for both parties, but
| it's also kinda insane that in the end, Google made as much
| profit on our busses as we did.
| carlosjobim wrote:
| If they're running on 9% net margin, then card fees are pretty
| down on the list of problems. They're vulnerable to any kind of
| fluctuation.
| crazygringo wrote:
| This article is missing some important aspects/context:
|
| > _If the card processing cost is 4 percent of the sale price,
| the fee amounts to $6. That $6 is not 4 percent of the profit; it
| is 12 percent of the merchant's margin._
|
| Sure, but merchants are raising prices overall together with all
| their competitors, or charging more when using a card. Credit
| cards aren't taking away 12% of merchants' profits that they'd
| keep otherwise.
|
| Also, credit card fees are not 4%, they're 1.5-3.5% with an
| average of around 2.3%.
|
| > _The merchant may pay little or nothing per transaction, the
| funds usually arrive immediately and no physical terminal is
| required._
|
| But what's missing here is fraud protection. It's more like debit
| cards than credit cards, and debit card transactions are much
| cheaper in the US too (more like 0.7%).
|
| Now that it's increasingly common for local merchants to
| implement a credit card surcharge (so non-CC users don't have to
| pay more), and a large percentage of credit card fees come
| straight back to the user as rewards (e.g. a 2% cash reward),
| it's not really clear that payment fees matter all that much in
| the end. See:
|
| _Fed Data Shows Economics of Interchange: 86% of Fees Fund
| Rewards Programs_ : https://www.pymnts.com/news/loyalty-and-
| rewards-news/2025/fe...
| Reason077 wrote:
| > _"Also, credit card fees are not 4%, they 're 1.5-3.5% with
| an average of around 2.3%."_
|
| And they're much lower than that in the UK and EU. Even the
| smallest UK retailers can pay as low as 1.6% or 0.8% + PS0.02.
| The big guys who are running billions of PS are paying 0.5% or
| less.
|
| (These apply to in-person transactions, accepting cards on the
| internet typically costs more).
| AnthonyMouse wrote:
| > Sure, but merchants are raising prices overall together with
| all their competitors, or charging more when using a card.
| Credit cards aren't taking away 12% of merchants' profits that
| they'd keep otherwise.
|
| That depends on the market. Suppose you're selling something
| which has a substitute available for the same price, so the
| customer will buy whichever one costs less by any amount, but
| the substitute is in a different market where the payment is in
| the form of a bank transfer or it's a tax deduction etc., or
| the substitute is paid for in time rather than money.
|
| Then none of the sellers accepting credit cards can raise the
| price because otherwise the buyer would go to the substitute.
|
| And all markets are like that _to some extent_. For some the
| seller can 't raise the price by a cent without losing 100% of
| their sales. For others, raising the price would "only" cost
| them 5% of their sales... but that's still 5% of their sales
| gone, and all the same fixed costs to cover. And, of course, in
| the cases where the price goes up, now it's the customer eating
| the fees, which is fairly nefarious when it's being subtly
| hidden from them.
|
| > Also, credit card fees are not 4%, they're 1.5-3.5% with an
| average of around 2.3%.
|
| Stripe is 2.9% + $0.30. For a $10 transaction, that's 5.9%. For
| a $3 transaction it's 12.9% -- of revenue, not profit.
|
| > But what's missing here is fraud protection.
|
| "Fraud protection" is is independent of the fees. You're making
| the case for that yourself -- if 86% of the fees go to rewards
| programs then they could be reduced by 86% without affecting
| "fraud protection", and the rewards programs are a wretched
| thicket of dark patterns and unpaid interest scams taking
| advantage of people who are bad at math or too short on time to
| configure them efficiently, on top of a tax on lower income
| people who don't qualify for them.
|
| It's also not clear why "fraud protection" should cost
| _anything_. The bank has no meaningful way to investigate a low
| dollar amount he-said she-said and the high dollar amount ones
| should be going to the actual court system, so why should they
| have anything more than a set of rules (e.g. which transactions
| are eligible, how long you have to file a dispute) under which
| you can make a request to have them flip the bits back in their
| computer for free?
| flerchin wrote:
| Payment processing networks are not free to build or operate.
| There are necessarily fraud controls and transaction reversals
| that require human oversight. This all costs. Nations can and
| should build this infrastructure, but in the absence, a payment
| processor is going to charge interchange. Otherwise why would
| they bother.
| Retric wrote:
| Uninvolved companies wouldn't bother, but outlaw such fees and
| lenders would still build this kind of infrastructure so they
| could make money charging people interest.
___________________________________________________________________
(page generated 2026-03-03 23:00 UTC)