When we picture a product crossing a border, we often imagine a port, a truck, an aircraft, or a customs gate. Yet a product does not travel alone. Trust in a contract, acceptance of a standard, and the expectation that a payment system will work travel with it. The invisible infrastructure of world trade is the reasonable predictability of these expectations.

The World Trade Organization’s outlook released on October 8 projects 3.9 percent growth in merchandise trade volume in 2026, compared with its March forecast of 1.9 percent. It projects 4.1 percent for 2027. These are forecasts, not outcomes already achieved; they may change with circumstances. The improved outlook matters, but it does not by itself mean that all the problems of a shared trading order have been solved.

The organization’s 2026 World Trade Report, published in September, also considers how the multilateral system can adapt to changing distributions of economic power and new conditions. I think these two conversations belong together. Trade flows can remain strong for a time even as confidence in the institutions and rules governing them erodes.

Connections remain, but their meaning is changing

Declaring the end of globalization seems to me too easy an explanation. A country may diversify suppliers for some goods while becoming more connected to the rest of the world in other fields. Alongside physical products, software, finance, data, and knowledge move across borders. A changing map should not immediately lead us to conclude that the world has come apart.

At the same time, it is insufficient to build every economic connection on a calculation of cost alone. Countries want to weigh security, access to energy, critical production capacity, and social resilience together. Dismissing these concerns as irrational from the outset makes it harder to understand people’s genuine anxieties about the economic order.

The real question is where a legitimate security need ends and an indefinite, ever-expanding exception begins. If no one knows when a rule will apply, an actor across the border cannot plan. An expectation that a decision will not be enforced against the powerful can also weaken the system’s legitimacy.

We cannot build common rules as a system that the less powerful must obey and the powerful may bypass when convenient. Trust in a rule does not require every result to satisfy everyone equally. It grows when reasons are given, appeals are possible, and application is predictable. We may not eliminate differences in economic strength, but we can narrow the space for arbitrary decisions.

Strengthening a country’s productive capacity and remaining committed to a shared order with other societies need not be an absolute choice. Resilience may call for recognizing excessive dependence and widening options more than severing every connection. We cannot have a sound debate by concealing either the cost of complete separation or the risk of complete openness.

New rules should not deepen old inequalities

Digital trade makes this debate more complex. The place where a service is produced, the place where its data is processed, and the institution responsible for it may be different. If we try to govern new services only with concepts designed for older goods, we will encounter gaps in the system. This is precisely where computer science meets international commerce.

Rebuilding common rules should not simply preserve the historical advantages of developed countries. The technology and finance needed to meet a new standard are not available to everyone in equal measure. Requirements imposed without discussing transition periods, capacity building, and clear criteria may deepen inequality even when their stated goals are sound. The quality of cooperation shows in how we manage those differences.

Governments have more to do than protect their own companies. They must also attend to fair competition, consumer rights, and workers preparing for change. When trade policy substitutes the expectations of a single industry for the interests of society as a whole, the distribution of gains and costs becomes a larger issue.

The legitimacy of the global economic order is tested here, too. Telling people that total trade has grown does not automatically explain a loss they experience in their own lives. If jobs change in a region, skills become outdated, or incomes come under pressure, we should not describe that only as a failure of individuals to adapt. Institutions have a responsibility to manage change.

Defending the benefits of trade and taking job loss, wages, and regional inequality seriously can go together. Strong education, transition support, and transparent decisions can strengthen the social foundation of economic openness. Without a path that lets people take part in change, even an impressive growth figure may fail to earn trust.

I do not find it sufficient to read international developments only as one country’s victory or loss. A decision that appears to favor one side in the short term may raise the long-term cost of the shared system. What matters is building institutions that defend their own society’s interests while recognizing other societies’ need for predictability.

Globalization’s future will be determined by more than the number of percentage points trade grows. It will also depend on how we protect the trust that makes that growth possible. In an interconnected world, lasting strength cannot be measured by keeping others perpetually unprepared. The ability to build common ground on which people, businesses, and countries can reasonably plan for tomorrow is also a form of strength.