Trade shocks redraw the global carbon map but do not shrink it:product-level evidence from integration and fragmentation

About a quarter of the carbon dioxide humanity releases into the atmosphere comes from making goods that are shipped across borders before they are ever used. This means that when a country buys a car or a shirt from abroad, it is also buying the pollution created to make it. As nations try to fight climate change, they are increasingly using trade rules to do it. Some countries are placing taxes on imports that are made with dirty energy, while others are cutting off trade entirely with nations that rely heavily on fossil fuels. The hope is that these moves will force the world to produce less pollution. But a critical question remains unanswered: when trade barriers rise, does the total amount of carbon pollution actually go down, or does it simply move to a different place?

A new study by Florian Pothin, an economist at the University of Rennes, answers this question by looking at the actual movement of goods over nearly three decades. Instead of guessing or running computer simulations, the researcher built a massive, detailed map of the world's trade in carbon. By tracking millions of specific products moving between countries, the study reveals a stark reality. When trade barriers are removed, carbon-intensive production slowly moves toward the open market. But when trade is blocked by sanctions or tariffs, that production does not vanish. It is pushed out quickly and reappears elsewhere, often in countries with even dirtier production methods. The total amount of pollution in the global system stays roughly the same; only the map of where it happens changes.

To understand how this works, one must first understand what the researcher measured. Most previous studies looked at broad categories of goods, like "steel" or "agriculture," and assigned an average pollution level to everything in that group. This approach is too blurry to see what happens when a specific trade shock occurs. Pothin went much deeper, breaking trade down to the level of individual products, such as a specific type of steel pipe or a particular kind of fertilizer. He then attached a precise "cradle-to-gate" carbon footprint to every single item. This means he calculated the emissions created from the moment raw materials were extracted until the product left the factory, but before it was shipped. By matching this data to real trade records from 1995 to 2023, he created a high-definition view of how carbon flows across the globe.

The study focused on three major events that acted like natural experiments. First, it looked at what happened when China joined the World Trade Organization in 2001, which opened its markets to the world. Second, it examined the trade war that began in 2018 when the United States placed heavy tariffs on Chinese goods. Third, it analyzed the sanctions imposed on Russia in 2022 after its invasion of Ukraine, which effectively cut off Russian energy and materials from Western markets. These events provided a clear way to see how the global carbon map reacts when borders are opened or closed.

The results show two very different speeds for how carbon moves. When a country opens its doors to trade, as China did, it takes a long time for carbon-intensive factories to build up there. Over ten years, the carbon flowing into China from the West grew by 31 percent, and after fifteen years, it had nearly doubled. This is a slow, gradual process of building new supply chains. However, when a market is suddenly closed, the reaction is immediate and violent. After the 2022 sanctions, Russia's carbon exports to the countries that imposed them dropped by 31 percent in the first year alone. By 2023, that drop had reached 74 percent. The production did not disappear; it was simply rerouted.

The study tracked exactly where that missing Russian carbon went. Instead of vanishing, it flowed to China, India, and Turkey. In just two years, shipments to these three countries more than doubled, rising by 236 percent compared to the pre-sanction levels. The same pattern appeared with the United States and China. When the US placed tariffs on Chinese goods, China's carbon exports to America fell by 44 percent by 2023. But that carbon did not evaporate; it was redirected to other countries, often passing through intermediate hubs before reaching the West. The researchers found that for every ton of carbon that disappeared from a sanctioned market, a nearly equal amount appeared in a new market. In the case of Russia, the study estimates that the displaced carbon was retained at a rate of nearly 100 percent, meaning almost none of it was actually eliminated from the global system.

Perhaps the most concerning finding is that this rerouting often makes the problem worse. When Russian oil and gas were forced out of Europe and into Asia, the study found that the new shipments carried 27 percent more carbon per dollar of value than the old ones. This suggests that the new buyers were getting dirtier energy, or that the discount Russia offered to its new buyers made the carbon intensity look even higher. Similarly, when China's exports to the US were blocked, the remaining supply chains shifted in ways that did not reduce the overall pollution. The study concludes that trade-driven decarbonization is, to a large degree, just a redistribution of pollution. A country might look cleaner because it stopped importing dirty goods, but the world as a whole is not cleaner because those goods are still being made and shipped elsewhere.

The research also tested the effectiveness of a popular policy idea called a "border carbon adjustment." This is a tax that countries like the European Union are planning to put on imports based on how much carbon was used to make them. Many models predict that such a tax would significantly cut global emissions. However, Pothin's study used real-world data from the US-China trade war to estimate how responsive trade actually is to price changes. The results showed that trade is much less sensitive to price than the models assume. When the researcher applied a realistic price of 100 dollars per ton of carbon to the sectors covered by the proposed tax, the reduction in pollution was only between 2 and 4 percent. Furthermore, because the supply of goods is often locked in place, between one-third and all of that small reduction would simply leak out to other countries that are not taxed.

This finding challenges the idea that trade barriers alone can solve the climate crisis. The study argues that while trade policy can change who buys what, it does not inherently shrink the total amount of carbon produced. The geography of pollution follows the barriers, but the total volume does not follow them down. If a country blocks imports from a dirty producer, that producer will simply sell to someone else, often with less efficient technology. The study suggests that unless policies address the production methods themselves, rather than just the flow of goods, the global carbon map will continue to be redrawn without ever being shrunk. The total amount of pollution remains stubbornly constant, moving from one corner of the globe to another, leaving the climate problem unsolved.