Finite substitution capacity and the limits of diversification: a multi-agent model of U.S. seafood trade-corridor resilience
This paper employs a multi-agent model of U.S. seafood trade to demonstrate that while diversifying supply corridors effectively restores resilience during isolated bilateral shocks, it fails under simultaneous multilateral disruptions due to the exhaustion of finite shared substitution capacity, thereby highlighting the critical limits of diversification strategies.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Imagine the global food supply as a massive, bustling city of delivery trucks. Most of the food we eat travels by sea, riding on giant ships that act like the city's main highways. But what happens when one of these highways gets blocked? Maybe a storm closes a port, a new tax makes a route too expensive, or a factory runs out of product. For a long time, experts thought the solution was simple: just send the trucks down a different road. This idea is called "diversification." The logic was that if one road is closed, you have plenty of other open roads to choose from, so the food keeps flowing, and prices stay steady.
However, this paper asks a tricky question: What if all the roads get blocked at the same time? Or what if the "open" roads are already so crowded that they can't take any more trucks? The authors use a computer model to simulate how the United States, the world's biggest seafood buyer, handles these traffic jams. They treat each shipping route as a smart robot (an "agent") that has to decide: should I try to fix my own broken truck, or should I try to squeeze my cargo onto a neighbor's truck? The study looks at a concept called "finite substitution capacity," which is a fancy way of saying that there is a limit to how much extra food a country can suddenly start eating. If everyone tries to switch to the same few open roads at once, those roads get jammed, and the plan fails.
This paper builds a digital simulation of the U.S. seafood trade, treating twenty major shipping routes as twenty smart robots. Each robot has a "resilience score" based on five pillars: can we get the fish (Availability), can we legally move it (Accessibility), is it affordable (Affordability), is it fresh when it arrives (Utilization), and is the supply steady (Stability). Every month, these robots face a choice. If their route is hit by a shock—like a tariff war or a stock collapse—they can either spend money to fix their own route (hedging) or try to auction off their extra fish to other routes that have space (switching). The model runs through history from 2010 to 2026, covering different eras like the trade war with China, the pandemic, and a simulated future where tariffs hit everyone at once.
The simulation reveals two very different worlds, or "regimes." In the first world, which happened during the 2018–2019 trade war with China, only one major partner was blocked. In this scenario, the robots successfully switched their fish to other countries like Vietnam or Latin America. The "open roads" had plenty of space, and the system worked perfectly. Diversification saved the day. But in the second world—a simulated 2025 scenario where tariffs hit almost every major partner at the same time—the story changes completely. Suddenly, every robot is trying to dump its extra fish onto the same few remaining routes. These routes, like a small town trying to host a massive festival, get completely overwhelmed. The "switching" plan collapses because there is no spare capacity left to absorb the shock.
The paper finds that when the system is overwhelmed by simultaneous shocks, the usual advice to "just diversify" stops working. Instead of finding a new road, the robots are forced to spend their money on fixing their own trucks and holding onto their existing inventory. The study suggests that in these crowded, high-stress situations, resilience doesn't come from finding new partners, but from having strong, pre-existing relationships with a few trusted partners (like Canada and Mexico) who are insulated from the chaos. The model shows that while smart robots can improve the system's stability by about 1% and reduce the worst price drops by roughly 8%, they cannot fix a broken system if the entire network is clogged. The key takeaway is that diversification is a great plan when things are calm, but when a global crisis hits, having a few deep, reliable connections is far more important than having many shallow ones.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.