Windear Consulting | 29 October 2026 | Outlook to 27 January 2027

World trade is having a better year than forecasters expected. A business buying diesel, moving goods or financing inventory could be forgiven for wondering whose recovery it is reading about.

In October, the World Trade Organization raised its forecast for merchandise trade volume growth in 2026 to 3.9 per cent, from 1.9 per cent in March. It simultaneously lowered the forecast for commercial services trade volume growth to 3.3 per cent, from 4.8 per cent. The WTO’s explanation connects the divergence to strong AI-related investment, adaptable supply chains and the uneven consequences of the Middle East conflict.

These are different operating experiences inside one world economy. For the next 90 days, the consequential question is how much adaptation an institution can afford before its own commitments become difficult to keep.

Windear’s assessment is that the pressure deserves attention at four points: fuel and logistics, market access, financing, and concentrated technology dependencies. Two opportunities remain worth preparing for: productivity improvements that can be demonstrated, and better trading conditions that could make previously unattractive options workable.

The growth figure conceals a concentrated engine

The IMF’s October World Economic Outlook projects global growth of 3.1 per cent in 2026 and 3.4 per cent in 2027. Those are annual projections, not predictions for this 90-day planning window. Their value lies partly in explaining what is supporting activity and what could weaken it.

AI is an important part of that explanation. The WTO’s executive summary reports that AI-enabling goods accounted for 47 per cent of the increase in the value of global merchandise trade in the first half of 2026. That figure concerns growth in trade value, not half of all goods traded or half of trade-volume growth.

The distinction matters for an institution outside those supply chains. Strong demand for servers and semiconductors does not establish stronger demand for its products. Nor does growth in nominal trade values necessarily mean an equivalent increase in the quantity of goods moving.

Before borrowing confidence from the global outlook, identify the source of your own expected revenue. Which customers are expanding? What evidence supports their orders? How exposed are they to costs or financing conditions that could change their purchasing decisions?

Risk one: crude supply can recover while the fuel you need remains constrained

The IEA’s October Oil Market Report describes a fragile recovery in Gulf crude exports alongside continuing pressure on refined products. It reports that global diesel exports in September were about 700,000 barrels a day below their level a year earlier. Its assessment anticipates tight gasoil and diesel conditions extending into 2027.

For transport operators, manufacturers and organisations dependent on backup generation, the relevant exposure may therefore be the delivered cost and availability of a specific fuel. A reassuring crude-production headline does not answer that question.

Windear’s practical recommendation is to test the commitment closest to the operation. Examine fuel-adjustment clauses, delivery terms, inventory coverage and the time needed to qualify another supplier. A second supplier using the same constrained route may offer less diversification than the procurement list suggests.

The review trigger should reflect the organisation’s tolerances. Decide which increase in delivered cost, deterioration in stock coverage or change in delivery time would require action. Use actual operating requirements to set it; a generic percentage threshold can create false precision.

Risk two: trade is adapting, but access remains specific

The WTO finds that the gap between trade within and across geopolitical blocs narrowed through 2025 and into 2026, even as US–China decoupling accelerated. The evidence supports a more selective account than the claim that every supply chain is separating into rival camps.

For a company choosing a market or production location, that means examining the restriction that applies to the product, technology or counterparty in question. A broad geopolitical label is an inadequate substitute for the relevant tariff, export-control, origin or licensing requirement.

Announcements belong on the watchlist. Implemented measures belong in the operating model. Establish the effective date, responsible authority and contractual consequences before treating the two as equivalent.

This is also where opportunity can emerge. Firms may need additional suppliers or routes as trading relationships change. An attractive alternative still has to meet standards, delivery requirements and full landed-cost expectations. Geography alone does not make it a better option.

Risk three: a financing shock can arrive before the sales decline

The IMF’s October Global Financial Stability Report identifies AI-related asset corrections, sovereign-bond yield shocks and cyber threats as possible catalysts for tighter financial conditions. It also examines vulnerabilities that could amplify stress, including leverage and liquidity mismatches.

For an operating institution, the first consequence need not be a cancelled order. It could be a refinancing offer on worse terms, a customer taking longer to pay or a counterparty requesting cash earlier than expected. Those are transmission channels to examine, not predictions that each will materialise.

Review the calendar of commitments rather than relying only on an annual funding total. Which obligations fall due before customer receipts? Which facilities must be renewed? Does a public-sector contract depend on a payment schedule that the business cannot comfortably finance?

A plan that works under the expected timetable may fail under a modest delay. Identify that point before deciding how much additional business to accept.

Risk four: AI exposure extends beyond the tool on your screen

In its analysis of rising financial risks, the IMF connects technology concentration with financing and operational vulnerabilities. It also describes how disruption at a payment system, cloud provider or software platform could spread through interconnected institutions.

A business can therefore have little direct investment in AI and still depend on the same providers as many others. Map the services whose failure would interrupt payments, access to records or delivery to customers. Ask what can continue during an outage and whether the recovery process has been tested.

For proposed AI adoption, separate the appeal of the technology from the economics of the task. What error rate is acceptable? Who reviews consequential outputs? What do integration, staff training and oversight cost? How difficult would it be to retrieve data or change providers?

Two opportunities worth earning

The first is measured productivity. In The World Can’t Afford to Get AI Wrong, Kristalina Georgieva distinguishes the immediate demand generated by AI investment from the productivity gains that adaptation could deliver. Buying access to a tool does not establish that the second has happened.

Choose a bounded task, record current performance and compare the new process on quality as well as speed. Scale when the evidence supports the additional commitment. That approach leaves room for gains without putting unearned savings into the budget.

The second opportunity is improved trading conditions if geopolitical and trade tensions ease. Prepare a shortlist of decisions that could become viable under better conditions. Specify what must improve, such as delivery reliability, financing terms or an applicable trade requirement. This preserves the ability to move without assuming that improvement has already arrived.

Make the next review capable of changing something

Over this planning window, distinguish three possible paths: continued adaptation at elevated cost; renewed disruption that consumes cash or interrupts delivery; and selective easing that improves particular opportunities. These are planning scenarios, not probability estimates. An organisation should decide which evidence would distinguish them in its own operation.

Our analysis of turning research into workable decisions makes the implementation problem explicit. A useful review needs someone authorised to act, enough time to act and an option that has been checked before it becomes urgent.

At the next leadership meeting, take one material commitment through that test. Identify the assumption keeping it viable, the evidence that could overturn it and the decision that would follow. Then examine who must approve that decision and whether they would receive the information soon enough. If those answers remain unresolved, that is the research brief. Bring it to Windear: the value of an outlook is the decision it helps an institution make while there is still room to choose.