By Heorhi Tratsiak
There is a moment when the abstraction becomes concrete.
For me it arrived not as a dramatic event but as a series of small operational changes that, taken individually, could each be explained as routine compliance procedure. A correspondent bank requesting additional documentation on a category of transactions that had processed cleanly for years. A processing time that extended by a day, then two. A form that had not previously existed, now required for a class of transfers that had previously required no such form.
Each change, on its own, was manageable. Each had a plausible administrative explanation. Taken together, over weeks, they described something else: the moment when a financial system I had worked inside for years revealed that it was not what it had always presented itself as being.
It was not neutral. It had never been neutral. And the mechanisms I had spent years learning to operate were also, I was learning, mechanisms that could be operated against you.
This chapter is about those mechanisms. How they were built, how they evolved, how they work at the operational level that most accounts of financial sanctions never describe, and what it means that they exist.
The Myth That Needed to Exist
The global financial infrastructure was built on a premise that was partly accurate and partly necessary fiction.
The accurate part: the systems genuinely were designed as technical infrastructure, as a set of standards and protocols that would allow financial institutions across the world to interoperate efficiently. SWIFT’s founding documents describe it in exactly these terms. A shared utility. A communications network. A service to the global financial community.
The necessary fiction: that technical infrastructure operates in a political vacuum. That a system built by institutions headquartered in specific countries, governed by rules set by those institutions, clearing through the central banks of those countries, denominated primarily in the currency of one of those countries, would be equally available to all participants under all circumstances.
This fiction was useful because it encouraged participation. Every country that connected to SWIFT, every bank that established correspondent relationships, every institution that certified for Visa and Mastercard was integrating itself into the global financial system. Integration created economic benefits. It also created dependencies. And dependencies, over decades, accumulated into leverage.
The leverage was always there. What changed, gradually and then rapidly, was the willingness to use it explicitly.
The Thirty-Year Education
The use of financial systems as tools of policy did not begin in the twenty-first century. Economic sanctions are as old as statecraft. Athens blockaded Megara in 432 BC. Napoleon’s Continental System attempted to strangle Britain through trade restriction. The League of Nations, in its brief and ineffective life, experimented with collective economic pressure.
What changed with the modern financial infrastructure is the precision of the instrument and the breadth of its reach.
The first systematic use of the modern SWIFT-era financial system as a sanctions tool came with Iraq following the Gulf War in 1990. The United Nations Security Council imposed comprehensive sanctions that effectively cut Iraq off from the international financial system. The mechanism was blunt, multilateral, and enforced through the direct prohibition on transactions rather than through the infrastructure itself.
The more sophisticated evolution came with Iran.
Between 2006 and 2012, the United States and European Union developed an increasingly comprehensive sanctions regime targeting Iran’s nuclear program. The early measures were conventional: asset freezes, travel bans, trade restrictions. But the architects of the policy understood that these conventional tools had limited effect on an economy that had adapted to decades of pressure. What would have real effect was disconnection from the financial infrastructure itself.
In February 2012, SWIFT announced that it would disconnect Iranian banks from its network, following European Union regulations that made providing such services illegal. Approximately thirty Iranian banks lost access. The immediate operational consequence was the near-impossibility of conducting international financial transactions. Wire transfers became extremely difficult. Trade finance, which depends on letters of credit and documentary collections processed through the SWIFT system, became severely constrained. Foreign reserves held offshore became harder to access and repatriate.
The economic effect was not immediate, because economies are resilient and find workarounds, but it was substantial. Iranian oil exports, which required payment in currencies that could only move through the international financial system, fell by roughly forty percent within a year. This was not an estimate or a projection. It was a measured consequence of removing access to the communications infrastructure through which oil payment instructions traveled.
The lesson was absorbed by those who had designed the measure. Financial infrastructure disconnection worked. It worked faster, more comprehensively, and with less political cost than most alternative coercive tools. It required no military deployment. It did not directly harm civilian populations in ways that generated visible sympathy. It was, from the perspective of those wielding it, a remarkably clean instrument.
The Iran case became the template.
How the Instrument Actually Works
I want to be precise about the mechanism, because the way sanctions are discussed in most public accounts obscures more than it reveals.
The phrase “disconnection from SWIFT” implies a single action: a switch is thrown, and a country or institution disappears from the network. The reality is considerably more complex, which is both why the tool is powerful and why it is difficult to fully defend against.
SWIFT itself is one layer. Removing an institution from SWIFT prevents it from sending and receiving the standardized messages through which international financial instructions travel. This is significant but not immediately fatal, because institutions can attempt to route messages through other channels, through bilateral correspondent relationships, through intermediary institutions, through alternative messaging systems. These workarounds are inefficient and expensive, but they exist.
The more devastating layer is correspondent banking.
When a major financial institution decides, for compliance reasons, that it can no longer maintain correspondent relationships with institutions from a sanctioned jurisdiction, the effect is the loss of access to settlement in the major currencies. Dollar-denominated transactions must settle through the Federal Reserve system, which means they must pass through institutions that maintain accounts there. If no such institution is willing to process your transactions, you cannot settle in dollars. If you cannot settle in dollars, a very large portion of international trade becomes operationally impossible, because the dollar remains the primary invoicing currency for global commodity markets.
The enforcement mechanism is not a direct prohibition issued to every bank in the world. It is more elegant than that. The United States Office of Foreign Assets Control, a bureau of the Treasury Department, maintains lists of sanctioned entities and jurisdictions, and imposes penalties on any institution under US jurisdiction that conducts transactions with listed parties. Since virtually every significant international financial institution has operations in the United States, has correspondent accounts at US banks, or processes dollar transactions that touch the US financial system at some point, virtually every significant international financial institution is subject to OFAC jurisdiction to some degree.
The result is that US sanctions, formally applicable only to entities under US jurisdiction, effectively apply globally through the compliance behavior of international banks. A German bank, a Japanese bank, a Singaporean bank: none of them are required by German, Japanese, or Singaporean law to comply with US sanctions against a particular country. But each of them has dollar-clearing relationships that depend on their standing with US regulators. The cost of maintaining relationships with sanctioned entities is the potential loss of US market access. The math is simple, and it produces the same result across jurisdictions.
This is the mechanism. It does not require global political consensus. It does not require multilateral agreement. It operates through the rational compliance decisions of private institutions, each protecting its own most valuable market relationships.
The Dollar’s Structural Position
To understand why this mechanism has the reach it has, you need to understand the structural position of the dollar in global finance.
The Bank for International Settlements surveys the foreign exchange market every three years. In its most recent comprehensive surveys, the US dollar appears on one side of roughly eighty-eight percent of all foreign exchange transactions globally. This does not mean the US is party to eighty-eight percent of global trade. It means that when a Japanese company buys Brazilian soybeans, the transaction is typically conducted in dollars. When a Turkish company buys German machinery, the invoice is often in dollars. When a central bank in Southeast Asia holds foreign exchange reserves, a substantial portion is held in dollar-denominated assets.
This position was established over decades, through a combination of the dollar’s role in the postwar Bretton Woods system, the 1974 petrodollar arrangements that linked oil pricing to the dollar, and the simple network effect of a currency that everyone uses because everyone else uses it. Each of these historical decisions reinforced the others, creating a system where the dollar’s dominance was self-perpetuating.
The consequence is that any actor who wishes to participate in the global economy at scale requires access to dollar-clearing infrastructure. And dollar-clearing infrastructure is controlled, ultimately, by the Federal Reserve and the US banking system. This gives the United States a structural lever over the global economy that has no peacetime precedent in history.
It is not a lever that requires any individual decision to use. It is embedded in the architecture. It is activated whenever the US government decides to instruct its financial regulators to treat a particular jurisdiction or entity as a compliance risk, because the downstream compliance behavior of institutions everywhere will do the rest.
The Consumer Layer: Cards and the Last Mile
Below the wholesale banking layer, there is a second mechanism that affects individuals rather than institutions, and its operation is visible in ways that the correspondent banking layer is not.
Visa and Mastercard process transactions for individual cardholders. When a country or a class of institutions is sanctioned in ways that include the card networks, individuals lose the ability to use their cards internationally, and in some cases domestically if the card is issued by an affected institution.
The operational mechanics of a card network suspension are different from a SWIFT disconnection. Card networks make decisions at the issuer level: they can suspend the ability of cards issued by specific banks to process international transactions, suspend their ability to process transactions at all, or in more extreme measures, require acquirers everywhere to decline cards from affected institutions.
For individuals, the effect is immediate and personal in a way that wholesale financial sanctions are not. A sanctioned government’s ability to settle international trade is an abstraction for most of the population. The inability to use your card at a foreign hotel, to withdraw money from an ATM abroad, to make a payment through an international platform: these are concrete losses, felt immediately.
I watched this layer activate during the period following the geopolitical escalation of early 2022. The speed was striking. Within days of the announcement of sanctions by major Western governments, the card networks had suspended services for Russian-issued cards. ATMs in foreign countries stopped accepting them. International online merchants stopped processing them. People found their purchasing power truncated overnight, not because their bank accounts had changed but because the network through which those accounts connected to the world had withdrawn its service.
The consumer card layer is, in a sense, the most politically legible part of financial sanctions. It is the part that ordinary people experience directly, that journalists can photograph and write about in accessible terms, that produces visible queues and visible frustration. It is also, operationally, the layer most amenable to workaround through cash and alternative systems. But its symbolic and psychological effect is significant, in both directions: for those experiencing the loss of access, and for those observing that access has been withdrawn.
The Compliance Cascade
There is a concept in sanctions policy that receives far less public attention than it deserves: secondary sanctions.
Primary sanctions are the straightforward case. An entity under US or EU jurisdiction is prohibited from transacting with sanctioned parties. If an American bank processes a transfer to a sanctioned entity, it has violated the law and faces penalties.
Secondary sanctions extend the logic extraterritorially. They threaten non-US, non-EU entities with the loss of access to US or EU markets if those entities maintain significant business relationships with sanctioned parties. The Countering America’s Adversaries Through Sanctions Act of 2017, and similar legislation, formalized this principle: a Chinese company, a Turkish company, an Indian bank, can each face US penalties not for violating US law, which they are not subject to, but for conducting business with parties the US has designated.
The legal theory behind secondary sanctions is contested. Many jurisdictions do not recognize their validity under international law. But legal contestation does not change practical behavior, because the practical question is not whether secondary sanctions are legally valid. It is whether an institution is willing to risk US market access in order to maintain a relationship that has become regulatory. Almost universally, the answer is no. US market access is worth more than almost any alternative relationship.
The result is what I think of as a compliance cascade. The US Treasury designates an entity. OFAC places it on a restricted list. US banks immediately cease transactions. International banks, regardless of their own jurisdiction’s legal position, review their own exposure to the designated entity and to parties that do business with it. Each institution’s compliance function makes a rational calculation: the cost of US regulatory scrutiny exceeds the value of the relationship. They reduce or eliminate exposure. Their correspondents do the same. The cascade moves outward through the network, through every institution that has any connection, however indirect, to the US financial system.
The designated entity does not need to be directly cut off from every institution in the world. It only needs to be cut off from the institutions at the nodes of the network. Once that happens, the cascade takes care of the rest.
I watched this cascade reach the edges of my working environment. Not as a primary sanctioned institution, but close enough to the perimeter that the edge effects were operationally real. The compliance behavior of correspondent banks was not driven by specific legal requirements applied to our specific institution. It was driven by their institutional risk assessments about entire jurisdictions and classes of relationship. We existed, in the eyes of some of their compliance functions, in a category of elevated risk. That category determination changed the texture of every interaction.
This is the underappreciated sophistication of the financial infrastructure weapon. It does not need to aim precisely. It radiates through the compliance behavior of private institutions, each acting rationally in its own interest, producing field effects that extend far beyond the formally designated targets.
What I Watched From the Inside
I want to be specific about what this looked like from the position I occupied, because the operational detail matters in ways that the high-level descriptions miss.
I was not at an institution that was directly and comprehensively sanctioned. The range and severity of what we experienced was different in kind from what a primary-sanctioned institution faces. But the edge effects of a large sanctions regime reach further than its formal scope, and I worked within those edge effects for an extended period.
What I observed was a gradual transformation of the friction profile of normal operations.
Transactions that had processed in two days began taking four. Not because any formal rule had changed. Because the correspondent banks through which they traveled were applying more careful review to anything associated with our jurisdiction. The review was not capricious. It reflected their own compliance obligations, their own assessments of where they wanted to concentrate regulatory risk, their own judgment about which relationships were worth maintaining and which were generating more compliance cost than commercial value.
The documentation requirements expanded. The questions became more detailed. The informal conversations with compliance contacts at correspondent institutions acquired a texture of careful distance that had not been there before.
None of this was surprising, once I understood the mechanism. Correspondent banks are private institutions with their own shareholders and their own regulators. Their compliance functions exist to protect the institution, not to facilitate the operations of their correspondents. When the regulatory environment signals that a class of relationships is higher risk, the rational response is to increase scrutiny of those relationships, reduce exposure, potentially exit them.
The rational compliance behavior of institutions everywhere, each acting in its own interest, produces a system-wide effect that no individual institution decided to create. This is the distributed nature of financial sanctions at the operational level. There is no single decision-maker. There is a regulatory signal at the center, and an emergent response at the periphery, and the cumulative effect is the tightening I described at the start of this chapter.
The Doctrine Takes Shape
The systematic use of financial infrastructure as an instrument of foreign policy did not emerge from a single strategic decision. It developed through a sequence of uses, each building on the lessons of the last.
After Iran, the tool was refined. The secondary sanctions concept, which extended US sanctions extraterritorially by threatening third-country institutions with loss of US market access if they maintained relationships with sanctioned parties, dramatically increased reach. What had been a tool that required multilateral consensus became a tool that could operate through unilateral US decisions with global effect.
The measures taken toward Russia following 2014 were, in retrospect, a calibration exercise. Targeted sanctions on specific individuals and entities. Restricted access to Western capital markets for state-owned banks. The message was a demonstration of capability and a warning, not a full deployment.
The full deployment came in 2022. The speed and comprehensiveness of the measures that followed the geopolitical escalation of that year represented the most complete use of the financial infrastructure as a policy tool in the modern era. Multiple major Russian banks were removed from SWIFT. Visa and Mastercard suspended operations in Russia within days. The Central Bank of Russia’s foreign currency reserves, approximately three hundred billion dollars held in Western financial systems, were frozen. The entire architecture of the global financial system was deployed, simultaneously, against one of its larger participants.
The financial consequences were significant. The long-term consequences, for the architecture itself, may be more significant still.
The Fracture Line
Using a system as a weapon changes the system.
Every country that watched what happened in 2022, every central bank governor and finance minister who observed the speed and scope of the measures, updated their assessment of the risk embedded in integration with the Western financial system. The integration had always carried implicit risk: that the rules governing access could change. What 2022 demonstrated was that the rules could change very fast, very comprehensively, and without meaningful advance warning.
The speed mattered as much as the scale. The freezing of the Russian Central Bank’s foreign currency reserves, approximately three hundred billion dollars, was accomplished within days. Central banks hold foreign exchange reserves precisely for crises. The reserves are supposed to be the backstop, the resource available when everything else fails. Discovering that the backstop can be rendered inaccessible in seventy-two hours changes every central bank’s assessment of what foreign exchange reserves held in Western institutions are actually worth.
The rational response to an updated risk assessment is to reduce exposure to the risk. For countries and institutions that have reason to be concerned about their own standing in the Western-dominated financial system, this means: reduce dollar reserves, develop alternative payment channels, accelerate the infrastructure that would make it possible to conduct significant financial activity outside the SWIFT-correspondent banking-dollar settlement complex.
China’s Cross-Border Interbank Payment System, launched in 2015 and expanded since, now connects over one hundred and sixty countries and provides an alternative clearing channel for yuan-denominated transactions. Russia developed its own financial messaging system following its partial exclusion from SWIFT in 2014. India has expanded rupee-denominated trade settlement with specific partners. The Gulf states have been developing payment infrastructure that reduces their operational dependence on dollar-clearing channels.
None of these alternatives currently approaches the scale, liquidity, or global reach of the existing system. The dollar’s structural position has not fundamentally changed. SWIFT still carries the vast majority of international financial messaging. The correspondent banking network still dominates international settlement.
But there is a difference between infrastructure that has no serious alternative, and infrastructure that has emerging alternatives under active development with strong political backing from large economies. The global financial system moved from the first category to the second category in a period of months.
Infrastructure that took seventy years to build does not fracture overnight. But the incentive structure that had previously supported it, the widely shared belief that integration was beneficial and the rules were stable, has been altered in ways that will shape investment decisions, policy decisions, and infrastructure development decisions for decades.
The fracture, if it comes, will not be dramatic. It will be the accumulation of bilateral payment arrangements, of de-dollarization in specific trade relationships, of reserve diversification decisions, of parallel systems built to handle specific high-priority corridors. Each individual change will be described as a marginal adjustment. The aggregate will be a restructuring of the global financial architecture more significant than anything since Bretton Woods.
Whether that outcome is good or bad depends entirely on where you sit. But understanding that it is underway, and understanding the mechanism that triggered it, is not optional for anyone who works in international finance.
The Weight of What Has Changed
I want to say something directly about what it means to have worked inside this infrastructure during the period when it became, unmistakably, a weapon.
I spent years learning these systems. Learning how SWIFT messages were formatted, how correspondent relationships were maintained, how card network certifications worked, how international payment flows moved through the global economy. I learned them as technical disciplines, as operational knowledge, as the professional foundation on which I built my career.
I did not learn them as the mechanics of coercion. That was not the frame in which the knowledge was acquired. But it is the frame that subsequent events imposed.
This is not a complaint. The systems are what they are, and understanding what they actually are is better than misunderstanding. But there is something worth naming in the experience of learning that the infrastructure you operate is also an instrument of power, that the neutral technical skills you developed are also, in certain contexts, the skills of operating something that can be used to cause significant harm to populations and institutions that find themselves on the wrong side of it.
The operational person does not make the policy decisions. The person processing the transfer does not decide whether the correspondent bank applies more scrutiny to a particular jurisdiction. The person certifying a card network integration does not decide whether that network will eventually suspend services to a country’s banks. The operational layer and the policy layer are separated by enormous institutional and geographical distance.
But they are connected. The policy works because the operational layer functions. Every correctly processed transaction, every maintained correspondent relationship, every certified payment service is, in some small sense, part of what makes the tool available to be used.
This is not a reason to stop working in financial systems. It is a reason to understand what financial systems are.
What the Weapon Reveals
Here is the thing about weapons: they reveal the values of those who build them and those who use them, and they reveal the vulnerability of those against whom they are used.
The financial infrastructure weapon reveals that the global economic system is not governed by neutral rules applied equally to all participants. It is governed by rules shaped by those with structural power, and the neutrality is real during periods when that power is not exercised, and fictional during periods when it is.
This is not a cynical observation. It is a structural one. All systems of rules are shaped by those with the power to write and enforce them. The financial system is not an exception. Understanding this does not mean the system is illegitimate or that participation in it is wrong. It means that participation comes with dependencies that are not always visible, and that those dependencies can become liabilities under the right conditions.
What this means practically for any institution, any country, any actor in the global financial system is the question that everyone with access to the policy lever, and everyone potentially subject to it, is now working through.
The answer is not obvious. The alternatives are costly and immature. The existing system, for all the leverage it embeds, still provides genuine economic benefits to its participants. Exiting it entirely is not a realistic option for most actors.
But the comfortable assumption that the infrastructure is neutral, that the rules will apply equally, that access is a technical matter rather than a political one: that assumption is no longer available. It was the assumption I operated under for most of my early career. The events I witnessed dismantled it, transaction by transaction, documentation request by documentation request, until what remained was a clearer and colder picture of what the system actually is.
A clearer picture is always better than a comfortable one. Even when what it shows you is harder to live with.
And here is what it shows: the global financial system is simultaneously the most powerful engine of economic cooperation in human history, and the most powerful instrument of economic coercion ever devised. These two things are not in contradiction. They are the same thing. The features that make it so effective at connecting markets and enabling trade are the same features that make it so effective at isolating and pressuring those who fall outside its tolerance.
You cannot have one without the other. The power to connect is the power to disconnect. The infrastructure that enables everything also enables the removal of everything.
I spent seven years learning how to make it work. The more important education was learning what it was.
