The World Trade Organization (WTO) estimated that new tariffs and other import-related measures applied by G20 economies in the year to mid-October 2025 covered about USD 2.6 trillion of trade, or 14.3% of G20 merchandise imports. Yet international trade has not stopped growing: the United Nations Conference on Trade and Development (UNCTAD) estimated that global trade would exceed USD 35 trillion in 2025, a rise of about 7% on the year before. Trade is still expanding, but along different routes and under different rules. UNCTAD also notes that nearly two-thirds of world trade takes place within global value chains, so almost every buyer is exposed to a policy decision somewhere upstream, whether they have mapped it or not. The system has survived; the assumptions built into it have not. Rising geopolitical tensions are a large part of the reason, as governments increasingly treat trade as a matter of security as well as commerce.
These pressures reach a buyer through different channels, and treating them as a single problem leads to the wrong response. Three matter most: tariffs, limits on exports, and conditions attached to public spending.
Tariffs are the most visible. A tariff raises landed cost directly, and a buyer can model it. The difficulty is that rates change with little notice and can differ by country of origin or product classification, so a quotation that is sound on Monday can be unreliable by quarter-end. Anti-dumping and countervailing duties, which follow formal investigations, can also sit on top of the standard rate.
The second type sits on the supplier’s side of the border. Export controls, licenses, quotas, and bans can cut off access to inputs whatever a buyer is willing to pay. The Organization for Economic Co-operation and Development (OECD) found that export restrictions on industrial raw materials grew more than fivefold between 2009 and 2023. The burden is uneven: between 2021 and 2023, restrictions affected 67% of trade in cobalt and 46% of trade in rare earth elements. A manufacturer that never imports these directly may still depend on them several tiers down, which is why mapping source materials beyond the first-tier supplier has become a core task rather than a nice extra.
Third, many governments attach conditions to their own spending and incentives. Procurement policies that favor locally made goods, or incentives that require a share of local content, can change who is eligible to bid and which suppliers qualify for support. Rules like these are one half of a larger story. The other half is what governments do to attract the factories themselves.
Where barriers restrict flows, industrial policies try to shape where production happens in the first place. Semiconductor incentive programmes in the United States and the European Union, India’s production-linked incentive schemes, and the EU’s critical raw materials legislation all pursue a similar goal: building domestic production in sectors that governments consider strategic. Two things follow for buyers. First, subsidies change the supply side. Plants that were uneconomic a few years ago may now be viable, and suppliers in supported regions can price differently from rivals elsewhere. New production capacity in more places is a real opportunity for anyone who has relied on a single region. Second, incentives come with strings, such as conditions on ownership, local sourcing, and delivery timelines. A supplier’s cost position is therefore tied to decisions its government makes outside any contract. That creates a risk a quotation will not show: a subsidy in force today can be revised, and the cheapest supplier in a supported region may be the one most exposed to that revision. Buyers should ask suppliers in supported regions whether their price depends on a subsidy, when it ends, and what happens to the price afterwards.
Governments also shape sourcing through the deals they sign with one another. Trade agreements offer a counterweight to barriers, but only for goods that qualify. Preferential duty rates apply when a product meets the agreement’s rules of origin, which set how much of its value or processing must come from member countries. A component assembled in a partner country from mostly outside inputs may not qualify, and buyers who assume a deal automatically lowers their duties can find the benefit disappears at the border for lack of documentation. The practical lesson is to treat origin evidence as a procurement requirement: written into the contract and checked before goods ship. Because trade policies move faster than most contract cycles, origin evidence needs periodic re-checking.
All of these pressures end up in the same place: the cost comparison between suppliers. A unit price captures only one part of what a sourcing decision costs. Five further items now belong in that comparison:
1
Tariff exposure
current duties by country of origin, plus the likelihood they change within the contract term.
2
Compliance cost
licenses, origin documentation, screening, and audit time.
3
Inventory cost
extra safety stock absorbs delays but ties up working capital and storage space.
4
Switching cost
qualifying a second source takes testing, tooling, and management attention.
5
Disruption cost
the expected loss if a production line stops because a part cannot cross a border.
None of these is new. What has changed is their size and volatility. A global supply chain model built around the lowest quote, with thin inventory, treated them as rare events; today they are recurring line items. A simple, hypothetical example shows how the ranking can flip. Supplier A quotes 100 per unit, and Supplier B quotes 108, so on price alone, A wins by eight. A’s country of origin attracts a 12% import duty, one of the trade taxes that now vary by origin, which adds 12; paperwork and extra safety stock add another 5, for a landed cost of 117. B’s goods qualify for a zero preferential rate under a trade deal, so there is no duty; paperwork and stock add 2 for a landed cost of 110. The supplier that looked eight units cheaper is in fact seven units dearer. The result holds only while B keeps meeting the rules of origin and duty rates stay where they are, which is exactly why those conditions belong in the contract.
With these costs visible, the question is how to respond. The sensible answer is not to abandon global sourcing. Cross-border purchasing still offers scale, skills, and variety that no single country can match; what has to change is how selectively it is used. UNCTAD observes that firms are relocating some production closer to end markets and taking more control over key inputs, a sign that risk is now weighed alongside cost. This is where strategic sourcing earns its name: it prices policy risk the way mature buyers already price currency risk, as a measurable factor with a likely range rather than a surprise. There is also a case against overreacting. Duplicating every supplier relationship across several regions is expensive, and many items do not need it. A tiered approach works better. Items with one or two qualified suppliers, long qualification times, and a high impact on output deserve a second source in a different jurisdiction. Commodity items with many substitutes can stay with the lowest-cost option, with a watch on duty changes. The aim is to buy insurance where a failure would hurt, not everywhere. Contracts are the second lever. Clauses worth negotiating before conditions change include:
Tariff-sharing terms that state who bears new duties and at what threshold the price is reopened.
Change-in-law provisions that cover new licenses, export limits, or origin requirements.
Origin warranties with documentation duties, so preferential rates are not lost.
Alternate-source rights that let the buyer qualify a second supplier without breaching exclusivity.
Notice periods long enough to move volume if a supplier’s government changes its support.
Some readers will see all of this as an expense with no return, which misses what the cost buys. In many cases, supply chain disruptions of recent years were made worse less by high input prices than by the lack of a tested alternative when a route closed. A second source that is already qualified turns a crisis into a negotiation. Resilience costs money, but so does its absence; the difference is that one cost appears in the budget and the other arrives as a surprise. Supply chains designed around policy risk may cost slightly more to run, yet they tend to cope far better when conditions turn. The best buyers quantify their exposures and revisit each category’s protection as policies change.
Turning these ideas into decisions is the hard part, and it is where Cognitud's procurement consultants start, at the point where a unit-price comparison stops. For each category, we build a tariff-adjusted landed-cost model that also carries compliance, inventory, and switching costs, so the true cost of every origin is visible before a decision is made. We then map suppliers several tiers down to find hidden single points of failure and score each category on criticality to show where a second source is worth paying for. Finally, our team drafts the contract language, from tariff-sharing and change-in-law terms to origin warranties, and runs scenario reviews with procurement and finance together, so a policy change triggers a prepared response instead of an urgent meeting.
