Global supply chains have spent much of this decade absorbing shocks. A pandemic was followed by war, trade restrictions, extreme weather, shipping disruptions, energy-market upheaval, and growing geopolitical tension. The lesson for businesses is no longer simply that disruption can happen. It is that companies need supply networks capable of continuing to operate when conditions change unexpectedly.

Recent events also show that resilience does not necessarily require dismantling global supply chains. In the first half of 2026, the World Trade Organization estimated that crude oil exports from the Middle East fell by roughly 24 percent, while liquefied natural gas exports fell by 47 percent. Yet global exports declined by only around 6 percent for crude oil and 1 percent for LNG as other suppliers increased shipments. Trade was also rerouted through alternative ports and corridors.

The adjustment helped the WTO raise its forecast for world merchandise trade volume growth in 2026 to 3.9 percent, from 1.9 percent in its March forecast. These figures are forecasts of trade volumes rather than the dollar value of trade, and the revision partly reflected stronger AI-related investment as well as the ability of supply networks to adapt to disruptions in energy and fertilizer markets.

The experience points toward a broader principle. A resilient supply chain is not one that avoids every disruption. It is one that can absorb a shock, find alternatives, and continue functioning.

Modern supply chains can stretch across suppliers, factories, ports, logistics companies, warehouses, technology platforms, and customers in several countries. That structure can reduce costs and give businesses access to specialized producers, but it can also transmit disruption from one part of the world to another.

Transportation remains particularly important. UN Trade and Development reported in September 2026 that more than 80 percent of world merchandise trade by volume moves by sea. When major maritime routes are disrupted, the effects can spread through freight costs, delivery schedules, energy markets, and inventories.

The Red Sea disruptions provided an earlier example. UNCTAD's 2025 Review of Maritime Transport found that rerouting ships around the Cape of Good Hope helped increase global ton-miles by 5.9 percent in 2024, nearly three times the growth in the volume of seaborne trade. Ton-miles measure both the amount of cargo transported and the distance it travels, so the increase reflected longer voyages rather than an equivalent increase in the quantity of goods being shipped.

For companies, this means supply-chain risk cannot be assessed solely by asking where a supplier is located. Transportation routes, ports, energy supplies, subcontractors, digital systems, and the availability of replacement suppliers can all matter.

One response to supply-chain uncertainty is to move production closer to customers or bring it back into the domestic economy. That may make sense for some products, particularly where national security, extremely long lead times, or limited alternatives make dependence unusually risky. But broad reshoring carries its own costs.

The OECD's 2025 Supply Chain Resilience Review modeled a substantially more localized global trading system using its METRO economic model. Compared with an interconnected system, the modeled relocalization scenario reduced global trade by more than 18 percent and global real GDP by more than 5 percent. Individual economies experienced modeled GDP losses ranging from 1.1 percent to 12.2 percent. The exercise is a scenario analysis rather than a forecast of what will happen. It is designed to estimate the consequences of a more domestic production structure under the model's assumptions.

Importantly, the localized system was not consistently more stable. The OECD found that GDP stability deteriorated in more than half of the economies it modeled.

More recent evidence also suggests that wholesale reshoring has not become the dominant response. In a 2026 nowcasting exercise covering 41 economies, the OECD estimated that the export-weighted domestic value-added share of gross exports increased only modestly, from about 77 percent in 2022 to 77.6 percent in 2024. The estimates are experimental and measure the domestic value embedded in exports rather than the physical location of every stage of production. The OECD concluded that the results provided little evidence of widespread reshoring during 2023 and 2024.

Instead, firms appear to be making more selective changes, including diversifying suppliers and strengthening inventories.

Not every supplier represents the same level of risk. A component available from dozens of producers is different from one purchased almost entirely from a single country, factory, or company.

OECD research shows that import concentration has been rising. Its analysis covers roughly 4,800 internationally traded products and uses the Herfindahl-Hirschman Index, a standard measure of market concentration, at the six-digit Harmonized System product level. The OECD classifies a product as having significant import concentration when a country's supplier concentration is more than twice the corresponding concentration in global exports, subject to additional concentration thresholds.

Using that methodology, the average number of products with significant import concentration was about 50 percent higher in the early 2020s than in the late 1990s. The increase was largely driven by non-OECD economies, while the average level among OECD members remained comparatively stable.

For a business, the practical equivalent is to identify products that are both important and difficult to replace. Mapping should extend beyond the immediate supplier where possible. A company may purchase from several vendors yet still be exposed if all of those vendors depend on the same upstream manufacturer, raw material, logistics route, or technology provider.

Adding suppliers can reduce dependence, but diversification is not free. Additional vendors can create qualification costs, smaller purchasing volumes, more complex logistics, and greater administrative burdens.

That makes selective diversification more useful than trying to duplicate every source.

A 2025 IMF working paper examined the trade-off using a multi-country, multi-sector general equilibrium model that incorporates frictions limiting how quickly supply chains can adjust. The researchers found that diversification can improve resilience but can also reduce efficiency. The potential benefits were strongest for imports that are particularly exposed to shocks, sit upstream in production networks, and are difficult to replace quickly.

The implication is that businesses should concentrate redundancy where disruption would have the greatest effect. Critical components with few substitutes may justify multiple qualified suppliers, while widely available inputs may not require the same investment.

Diversification can also mean geographic diversification rather than simply increasing the number of suppliers. Two suppliers exposed to the same port, electricity network, political risk, or natural hazard may provide less redundancy than their number suggests.

For years, companies pursued leaner inventories to reduce working capital and storage costs. Those economic benefits have not disappeared. But recent disruptions have made the cost of running inventories too close to zero more visible.

OECD analysis of recent global value-chain adjustments indicates that companies have responded partly by strengthening inventories alongside diversifying suppliers, rather than systematically relocating production.

The appropriate level of buffer stock depends heavily on the product. Holding months of inexpensive but essential components can be economical if a shortage could shut down a high-value production line. Carrying the same level of inventory for perishable products, rapidly depreciating technology, or easily sourced commodities may make far less sense.

Resilience therefore does not require abandoning efficient inventory management. It requires recognizing that inventory has an insurance value as well as a carrying cost.

Supplier diversification has limited value if every shipment depends on the same vulnerable transportation corridor.

Companies can reduce this exposure by assessing alternative ports, carriers, transport modes, warehouses, and distribution routes before they are needed. The purpose is not necessarily to use every alternative during normal conditions. It is to understand which options are operationally realistic when the preferred route becomes unavailable or uneconomic.

The events of 2026 showed how quickly trade can adjust when alternatives exist. Despite severe disruption to Middle Eastern energy exports, the WTO reported that additional supply from elsewhere and the rerouting of trade helped limit the decline in global shipments. Global container throughput was 3.9 percent higher year to date through July 2026, even as geopolitical disruptions affected major trade routes.

Flexibility can therefore come from having more than one route to market as well as more than one source of supply.

A business cannot respond quickly to a disruption it cannot see.

Digital systems can help companies track shipments, monitor inventories, exchange information with suppliers, and identify delays earlier. In its June 2026 report on supply-chain resilience and AI, the OECD emphasized the importance of machine-readable information, interoperable systems, digital trade documentation, and secure cross-border data flows. Such infrastructure can make it easier for companies and authorities to process goods and respond to changing supply conditions.

Technology, however, introduces another form of concentration risk. Supply chains may depend heavily on a limited number of cloud platforms, software providers, communications networks, or data services. The OECD has warned that increased digitalization can therefore improve resilience in some areas while creating new vulnerabilities, particularly to cyberattacks and outages.

Digital resilience needs to develop alongside physical supply-chain resilience. More visibility is valuable, but companies also need cybersecurity, reliable data, backup procedures, and an understanding of their dependence on critical technology providers.

Many of the factors determining how easily a company can switch suppliers are outside its direct control. Customs procedures, transportation infrastructure, border delays, data rules, and regulatory requirements can either support or restrict supply-chain flexibility.

The OECD argues that streamlined customs procedures, efficient logistics services, interoperable digital systems, and international regulatory cooperation can make supply chains more agile. These measures reduce the friction involved when companies need to shift sourcing or transportation arrangements.

Governments are also increasingly using explicit diversification policies in strategically important markets. The European Union's Critical Raw Materials Act, for example, establishes a 2030 benchmark under which no more than 65 percent of the EU's annual consumption of each strategic raw material should come from a single third country at a relevant processing stage. The legislation also sets benchmarks for EU extraction, processing, and recycling capacity. These are policy benchmarks rather than forecasts of future production.

Such measures show how supply-chain resilience has moved beyond corporate purchasing departments and become an issue of industrial and economic policy.

Building resilience inevitably involves trade-offs. Additional inventory ties up capital. Multiple suppliers can raise purchasing costs. Alternative transportation routes may be less efficient. Greater visibility requires investment in technology and data systems.

The challenge is therefore not to eliminate every possible vulnerability. Doing so would be prohibitively expensive and may create new vulnerabilities elsewhere.

The strongest evidence instead supports a more targeted approach. Companies can identify concentrated dependencies, develop alternatives for the most critical inputs, maintain appropriate buffers, improve visibility across their supply networks, and prepare transportation options before disruption occurs.

Global supply chains are unlikely to become risk-free. They also show little evidence of disappearing. OECD estimates indicate that global value-chain activity remained historically high in 2024, while its more recent analysis points to gradual reconfiguration rather than a wholesale retreat from international production.

Resilience is ultimately less about predicting the next crisis than about preserving enough options to respond when it arrives.