# Why high-ticket B2B payments are still running on 1970s infrastructure SWIFT, correspondent banking, and wire transfers date to 1973. Here's what that costs deal professionals moving serious money today. --- ## Why high-ticket B2B payments are still running on 1970s infrastructure The infrastructure moving millions of dollars between businesses right now was designed before the fax machine was common. Before the personal computer existed. Before the internet was a concept anyone outside a defence laboratory had entertained. SWIFT — the Society for Worldwide Interbank Financial Telecommunication — launched in 1973. The correspondent banking model it runs on is older still, with roots in the mid-nineteenth century. The wire transfer, as a concept, predates the automobile. And yet here we are: a broker closes a seven-figure deal, a consultant invoices across borders, a commercial agent splits a commission across three jurisdictions — and every single dollar moves through the same chain of intermediaries, the same message-passing protocols, the same manual reconciliation windows that were designed when "processing time" meant something typed on a telex machine and handed to a human being to act on. This is not a technology gap. It is a structural one. And it costs deal professionals — closers, brokers, agents, advisors — real money, real time, and real risk every single time a high-ticket transaction clears. ## The Infrastructure That Refuses to Die ### What SWIFT Actually Is — and Isn't There is a common misunderstanding about SWIFT that makes it worth addressing directly, because the misunderstanding explains why reform is so difficult. SWIFT does not move money. It never has. SWIFT is a messaging network — a standardised language that banks use to instruct one another. When a bank sends a SWIFT message, it is sending a promise, an instruction, a notification. The actual movement of funds happens through a separate, parallel set of correspondent banking relationships — bilateral credit lines, nostro and vostro accounts, and settlement systems that operate on their own schedules, in their own time zones, with their own liquidity constraints. This distinction matters enormously for anyone moving high-ticket payments. When your bank tells you a wire has been "sent," it means a SWIFT message has been dispatched. What happens next is a chain of events that the sender cannot control, cannot monitor in real time, and cannot reverse once set in motion. ### The Correspondent Banking Chain For a domestic wire between two accounts at the same bank, the process is relatively contained. But almost no significant B2B deal happens that cleanly. A commercial transaction crossing a border — a licensing payment, a deal commission, a property settlement, a distribution fee — typically travels through what is called the correspondent banking chain: a sequence of intermediary banks, each holding accounts on behalf of the next, each taking a slice of the transaction, each applying its own processing windows and compliance checks. Here is the anatomy of a standard international wire for a high-ticket B2B payment: **Step 1: The originating bank accepts the payment instruction.** The sender's bank receives the wire details — beneficiary name, account number, SWIFT BIC, amount, purpose. This step alone requires the sender to have obtained the correct details from the recipient, typically by email or document, with no verification layer built into the process itself. A transposed digit, a mismatched name, or an incorrect SWIFT code at this stage can send a six-figure payment into a resolution queue that takes days to unwind. The bank validates the instruction internally and prepares a SWIFT MT103 message — the standard format for a single-customer credit transfer. **Step 2: The SWIFT message is transmitted to the correspondent bank.** The MT103 leaves the originating bank and arrives at the correspondent bank — which may or may not be in the same country as the sender, and almost certainly is not in the same country as the recipient. Correspondent banks are the backbone of international settlement. Large institutions — JPMorgan, Deutsche Bank, Standard Chartered, Citibank — hold accounts for thousands of smaller banks worldwide. When a regional bank in one country needs to move money to a recipient whose bank has no direct relationship with the sender's bank, it routes through one of these correspondents. Sometimes two. Sometimes three. **Step 3: The correspondent bank processes the instruction.** This is where time disappears. The correspondent bank applies its own compliance screening — OFAC lists, AML checks, transaction monitoring rules calibrated to its own risk appetite. If anything in the transaction triggers a review flag — an unusual amount, a flagged jurisdiction, a beneficiary name that partially matches a watchlist — the payment enters a manual review queue. The correspondent bank is not obligated to notify the sender in real time. The payment simply stops moving. **Step 4: Funds are credited to the beneficiary's account — eventually.** If the chain is clean, if no correspondent along the route requires manual intervention, and if the receiving bank's cut-off times align, the beneficiary sees funds in one to three business days. In practice, for high-ticket transactions with multiple parties in multiple jurisdictions, five business days is common. Longer is not unusual. At no point in this chain does anyone have a unified, real-time view of exactly where the money is, what fees have been deducted, or whether the payment will arrive in full. The sender does not have it. The recipient does not have it. The banks themselves often do not have it across the full chain. ## What This Costs in a Real Deal ### The Fee Structure No One Fully Discloses Every participant in the correspondent chain charges for the service. The originating bank charges a wire fee — a flat amount or a percentage, depending on the institution and the relationship. The correspondent bank charges a handling fee, typically deducted from the principal, which means the amount that arrives at the other end is smaller than the amount that left the sender's account. If there is more than one correspondent — and on less common currency routes, there frequently is — each charges independently. The receiving bank may charge an incoming wire fee. For deal professionals, the practical consequence is that the recipient cannot be certain how much they will receive until the funds actually land. A commission payment of $250,000 may arrive as $248,400. The $1,600 difference represents fees extracted by parties the recipient never agreed to work with and cannot negotiate with. In a multi-party deal — where a broker is splitting commission between two agents, a legal advisor, and an introducer — each party may receive a slightly different shortfall. Reconciling those shortfalls back to the original payment is administrative work that no one budgeted for. This fee opacity is not accidental. It is structural. The correspondent banking model predates modern fee disclosure norms, and while regulators in some jurisdictions have pushed for greater transparency, the practical reality for cross-border B2B transactions remains that full fee disclosure across a correspondent chain is rare and often not available until after the money has moved. ### The Float Problem Float is the period during which money has left the sender's account but has not yet arrived in the recipient's. In high-ticket B2B transactions, this period can represent significant carrying cost. A $500,000 payment sitting in transit for four business days is $500,000 that cannot be deployed, cannot be reinvested, cannot be used to fund the next stage of a project. For deals with tight timing — a property settlement with a completion date, a licensing agreement where a signature is conditional on cleared funds — float is not just inconvenient. It can be deal-breaking. The originating bank earns interest on float. So does every correspondent handling the transaction. The sender earns nothing. The recipient earns nothing. The float is, in effect, a silent tax on urgency — one that scales directly with the size of the transaction. ### The Amendment and Recall Problem When something goes wrong — wrong account number, wrong beneficiary, wrong amount — the process for correction is not digital. It does not happen through a dashboard. It happens through a sequence of SWIFT MT199 and MT299 messages, bank-to-bank correspondence, and in many cases, manual telephone calls between compliance officers. The originating bank initiates a recall request. The correspondent bank must agree to freeze the funds — assuming they have not already been credited to the final beneficiary's account. If funds have already been credited, the bank must attempt to recover them from the recipient, which is a request, not a right. In practice, if a $300,000 payment reaches the wrong account, full recovery is not guaranteed. Partial recovery is common. The process takes weeks. For a deal professional, the cost is not just the money at risk — it is the relationship damage, the legal exposure, and the time spent navigating a system not designed for human oversight. ## Why It Persists: The Structural Lock-In ### Network Effects at Scale SWIFT connects over 11,000 financial institutions in more than 200 countries. The correspondent banking relationships built on top of it represent decades of bilateral agreements, credit lines, and legal frameworks. No single institution can unilaterally exit and still serve its clients. The network effect that makes SWIFT valuable is precisely the same force that makes it impossible to reform quickly. Every major bank depends on the system. Every major bank has compliance infrastructure, technology stacks, and operational processes built around it. The switching cost is not financial — it is existential to the business model. ### Regulatory Layering The correspondent banking model has also become the primary delivery mechanism for international AML and sanctions compliance. When regulators need to monitor cross-border money flows, they look to SWIFT data and correspondent bank reporting. This has created a feedback loop: the heavier the regulatory burden placed on cross-border payments, the more valuable the incumbent infrastructure becomes — because only large correspondent banks can afford the compliance operations required to remain in the network. Smaller banks have been derisked off of correspondent relationships entirely over the past decade, concentrating the system further into fewer, larger nodes. For deal professionals, this means higher costs and slower processing on the routes that matter most — emerging markets, smaller currency pairs, newer jurisdictions. ### The Trust Deficit Between Deal Parties There is a dimension of this problem that sits upstream of the infrastructure itself. High-ticket B2B transactions frequently involve parties who do not have a pre-existing trust relationship sufficient to justify sequential payment — where one party pays, another receives, and only then does the next obligation trigger. This is the deal structure that wire transfers were designed to support: send the money, confirm receipt, proceed. But in complex deals — multi-party arrangements, commission splits, co-broker agreements, joint ventures — sequential payment creates a principal problem. Whoever receives first holds all the leverage. Whoever receives last bears all the risk. The traditional solution is escrow: a neutral third party holds funds until conditions are met, then distributes. Escrow is slow, expensive, and introduces its own counterparty risk. The escrow agent must be trusted, must be licensed, must hold the money in a segregated account, and must execute the release manually, on their schedule, within their processing window. For deals moving quickly, escrow adds friction at exactly the wrong moment. For deals involving multiple currencies or multiple jurisdictions, escrow can introduce tax and regulatory complications that exceed the complexity of the original payment. The result is that deal professionals routinely default to sequential wire transfer — not because it is the right structure, but because it is the familiar one. And they absorb the trust risk, the timing risk, and the relationship strain that comes with it. ## The Real Cost Is Not the Wire Fee ### What Professionals Actually Lose The direct costs of the correspondent banking model — fees, float, amendment charges — are real but calculable. The indirect costs are larger and almost never measured. Consider a commercial broker closing a deal that requires payment to be split between four parties across three countries. The options available are: One party receives the full amount and redistributes manually. This means the other three parties are dependent on the first party's speed, accuracy, and goodwill. If any relationship in that group is strained, the redistribution may be slow. If the receiving party encounters a cash flow problem, it may be delayed. The others have no legal recourse that moves faster than the problem itself. Or four separate wires are sent simultaneously. This requires the payer to have the correct banking details for all four parties, to issue four separate instructions, to pay four separate wire fees, and to track four separate transactions through a system that provides no unified visibility. If any of the four wires fails or is delayed, the discrepancy must be identified, investigated, and corrected — again, manually, again, through the correspondent banking chain. Neither option is acceptable for deals where timing, accuracy, and relationship equity matter. And in high-ticket B2B transactions, they always matter. ### The Reconciliation Tax After the money moves — however it moves — comes reconciliation. Every party needs to confirm receipt. Every party needs to account for what they received against what they were owed. Where fees were deducted mid-chain, the math does not reconcile cleanly. Someone has to chase the discrepancy. Someone has to issue an explanation. In some cases, someone has to issue a follow-up payment to cover the shortfall. In professional services and deal-making contexts, that follow-up payment is itself another wire, another fee, another float period, another reconciliation cycle. The administrative burden of post-payment reconciliation in high-ticket B2B transactions is consistently underestimated. Finance teams at professional services firms report that cross-border payment reconciliation consumes disproportionate hours relative to domestic transactions — not because the transactions are more complex, but because the infrastructure provides so little transparency that reconstruction after the fact is often required. ### The Relationship Cost This is the cost that never appears on a ledger. When a commission payment arrives late, the recipient does not always know why. When it arrives short, the explanation — "fees were deducted by the correspondent bank" — is accurate but unsatisfying. When a deal closes cleanly on the legal and commercial side but the payment side takes ten days and requires three emails to reconcile, it colours the professional relationship. Not fatally, usually. But persistently. The infrastructure creates ambiguity at the exact moment that deal professionals need clarity. And ambiguity, in business relationships, compounds. ## The Architecture of the Problem ### Why "Faster Payments" Does Not Solve This Domestic faster payments systems — the UK's Faster Payments, the US's RTP and FedNow, the EU's SEPA Instant Credit Transfer — represent genuine improvements within their jurisdictions. For domestic B2B transactions, they reduce settlement time dramatically. But they do not address the cross-border problem, because they do not interoperate with one another at scale. A wire from a US business to a European counterpart still travels through correspondent banking. A commission split across three jurisdictions still requires multiple wires or a manual redistribution. The domestic improvements are real, but they are islands — and most high-ticket B2B deal activity is not island-shaped. SWIFT GPI — the Global Payments Innovation initiative launched in 2017 — has improved tracking transparency for participating banks and reduced settlement times on many corridors. But it is an improvement to the existing architecture, not a replacement of it. Correspondent relationships, deducted fees, cut-off windows, and manual compliance queues remain. GPI makes the existing system slightly more visible. It does not make it structurally sound for the demands of modern deal-making. ### The Timing Mismatch Every other element of a high-ticket B2B deal has accelerated. Due diligence that took weeks now takes days. Document execution that required couriers and wet signatures now happens on e-signature platforms in hours. Communication that required scheduled calls now happens in real time across time zones. The commercial and legal layers of deal-making have converged on speed. The payment layer has not. The wire transfer remains the single slowest, least transparent, and most friction-heavy step in a transaction that professionals have otherwise optimised to move quickly. This mismatch is not neutral. It creates pressure at closing. It creates anxiety for recipients. It creates administrative overhead for payers. And it creates recurring vulnerability — every deal that relies on sequential trust in the payment step is a deal where the relationship is tested at its most sensitive moment. ## The Resolution Is Structural, Not Incremental Incremental fixes to the SWIFT correspondent model — faster messaging, better tracking, improved fee disclosure — address symptoms, not causes. The cause is architectural: a system in which multiple intermediaries must each process, verify, and forward a transaction sequentially before funds arrive, with no mechanism for simultaneous multi-party settlement and no transparency into the chain as it operates. The only resolution that changes the structural dynamic is one that eliminates the chain itself — where payment settlement is simultaneous rather than sequential, where the distribution logic is embedded in the transaction rather than managed by a human after the fact, where every party in a deal receives their share in the same instant that the payment is made, and where the record of that payment is public, permanent, and requires no reconciliation. Shaka operates on exactly this architecture. A deal creator sets the payment split — broker, agents, advisors, any number of parties — generates a single payment link, and the buyer pays once. The smart contract distributes to every party simultaneously, in the same transaction, with no one holding the money in between. The infrastructure does not have a correspondent. It does not have a cut-off window. It does not deduct fees mid-chain. The payment is final the moment it confirms. The reason high-ticket B2B payments are still running on 1970s infrastructure is not that better architecture does not exist. It is that the incumbent system has been too embedded, too regulated, and too familiar to displace through incremental reform. The architecture of simultaneous, programmable, onchain settlement is not an upgrade to SWIFT. It is a different answer to the same problem — one that treats multi-party payment distribution as a first-class function rather than an afterthought. For deal professionals who move serious money and cannot afford the ambiguity that the legacy system generates, that difference is not academic. It is the whole point.