For a country with a large economy like the United States, trade sanctions have obvious attractions. They offer what looks like a muscular pursuit of foreign policy goals–that is, more than giving worthy speeches or recalling ambassadors from foreign embassies. They don’t involve declaring war or sending troops to fight. They can seem cost-free, in the sense that declaring that certain kind of trade will be blocked doesn’t involve a direct budgetary cost. Trade sanctions appeal to protectionists who view trad with suspicion, anyway.
So what’s the available evidence trade sanctions and how they work? Two papers take up this issue in the just-published Winter 2023 issue of the Journal of Economic Perspectives. T. Clifton Morgan, Constantinos Syropoulos and Yoto V. Yotov provide an overview in “Economic Sanctions: Evolution, Consequences, and Challenges,” while Marco Cipriani, Linda S. Goldberg and Gabriele La Spada focus on financial sanctions and the workings of the SWIFT system in “Financial Sanctions, SWIFT, and the Architecture of the International Payment System.” (Full disclosure: I’ve been the Managing Editor of JEP for 37 years, and thus am probably predisposed to think the articles are of interest. But no money is being made here. These articles, like all JEP articles back to the first issue, are freely available online courtesy of the American Economic Association.)
Syropoulos, Yotov, and a group of co-authors have created the Global Sanctions Data Base, which provides systematic data about 1,325 sanction cases during the period 1950–2022. A figure from the JEP paper shows the sharp rise in sanctions in the last couple of decades. In particular, financial and travel sanctions have risen especially quickly.
How well do the sanctions work? Here, Morgan, Syropoulos, and Yotov point out an interesting disjunction in how economists and political scientists answer this question. They write: “[E]economists have tended to interpret ‘effectiveness’ in terms of the economic damage that sanctions cause, while political scientists have considered sanctions ‘effective’ only if they achieve their declared political objectives.” For example, economists tend to judge the recent sanctions on Russia in terms of how much they hurt Russia’s economy, while political scientists tend to judge sanctions in terms of whether they lead Russia to stop its war in Ukraine. In addition, the success rate of trade sanctions is doubtless linked to factors like whether they are internationally coordinated, how well they are targeted, and whether they are reinforced with other policies. Or perhaps sanctions in the past have only been adopted when they seemed especially likely to work, which means that their past performance cannot be casually extrapolated to future scenarios.
Morgan, Syropoulos, and Yotov work through these various issues and offer this thought: “[E]ven in their worst light, sanctions have been shown to be effective in a modest fraction of cases. Even a 25 percent success rate for sanctions may be considerably higher than doing nothing, and the costs may be substantially lower than
other alternatives, like overt military interventions. Perhaps the `sanctions glass’ should be viewed as one-quarter full, not three-quarters empty.”
What about financial sanctions? In particular, the policy decision to cut off Russia from the SWIFT system received considerable attention, although even among economists, a fair number could not tell you in any detail–at least pre-2022–just what the SWIFT system actually did. Cipriani, Goldberg and La Spada discuss the history of financial sanctions, with some prominent examples of how they have worked or not in the past, and with some emphasis on the Society for Worldwide Interbank Financial Telecommunication, more commonly known as SWIFT.
When two banks need to transfer funds inside a country, they can do so through the central bank for that country–which is why the Federal Reserve is sometimes called the “bank for banks.” But most central banks (Switzerland is an exception) don’t facilitate transactions between domestic and foreign banks. Instead, there is a network of “correspondent” banks that operate in more than one country. If a Russian bank wants to transfer or receive funds with a bank in another country, it must typically operate through one of these correspondent banks. Messages must be sent back and forth
Back in the 1950s, 1960s, and 1970s, those messages were typically sent by Telex (for those of you old enough to remember that term). But Telex messages had no special formatting and were comparatively high-cost. Thus, banks and bank regulators around the world set up SWIFT as a nonprofit financial institution based in Belgium–and thus directly under financial laws of Belgium and the European Union. The idea was to have a set of computer protocols for a wide array of financial transactions. The correspondent banks were still needed, but messages about international financial transactions could be sent quickly and efficiently. Here’s a figure showing the number of financial institutions connected to SWIFT and the number of messages sent over the system.
The key point to remember here is that SWIFT doesn’t actually move any money or hold any money. It is just a messaging system between banks. In addition, the computer protocols that SWIFT has created for financial transactions are in the public domain. If a financial institution wants to use those protocols to move money, but to send the messages outside of the SWIFT system, it is quite possible to do so. Thus, cutting off Russia’s access to SWIFT surely raises the cost to Russia of making international financial transactions, but it does not block the transactions themselves.
The SWIFT system is by far and away the most dominant for sending messages between international financial institutions. But countries around the world have noticed that SWIFT is being used for trade sanctions–for Iran for a period of time in 2012, for North Korea in 2017, and now for Russia–and some of them are setting up alternative systems using the SWIFT protocols that don’t do much now, but could be ramped up if needed. As Cipriani, Goldberg, and La Spada point out, Russia has set up such a system:
Russia developed its own financial messaging system, SPFS (System for Transfer of Financial Messages). SPFS can transmit messages in the SWIFT format, and more broadly messages based on the ISO 20022 standard, as well as free-format messages. More than 400 banks have already connected to SPFS, most of them Russian or from former Soviet Republics. A few banks from Germany, Switzerland, France, Japan, Sweden, Turkey, and Cuba are also connected. By April 2022, the number of countries with financial institutions using SPFS had grown from 12 to 52, at which point the Central Bank of Russia decided not to publish the names of SPFS users. Due to its limited scale, SPFS mainly processes financial messages within Russia; in 2021, roughly 20 percent of all Russian domestic transfers were done through SPFS, with the Russian central bank aiming to increase this share to 30 percent by 2023 (Shagina 2021).
And so has China:
In 2015, the People’s Bank of China launched the Chinese Cross-Border Interbank Payment System (CIPS)
with the purpose of supporting the use of the renminbi in international trade and international financial markets. In contrast to SWIFT, … CIPS is not only a messaging system but also offers payment clearing and settlement
services for cross-border payments in renminbi. … [A]t the end January 2022, there were 1,280 participants from 103 countries. Among the direct participants, eleven are foreign banks, including large banks from the United States and other developed countries. The system is overseen and backed by People’s Bank of China. Similarly to Russia’s SPFS, CIPS uses the SWIFT industry standard for syntax in financial messages. Indirect participants can obtain services provided by CIPS through direct participants.
India is developing an interbank messaging system based on the SWIFT protocols as well, and there are some reports in the business press of plans for Russia, China, and India to merge these systems, although at present SWIFT remains far and away the dominant system for sending these international interbank messages.
Economic sanctions can impose costs. But whether they can achieve the desired political goals is a more complicated issue. The answer seems to be “sometimes,” but depending heavily on specific circumstances and the possibilities for evading the sanctions.