Page references are to each chapter's own page numbers (Ch. 1, Ch. 2, Ch. 3).
A split view of the global economy
The report opens with two forces. Geopolitics is the first: the war in the Middle East, plus hostilities elsewhere and trade tensions. Its main route into the economy is a supply shock, working through higher and more volatile energy prices and the inflation they feed (Ch. 1, p. 1). Technology is the second. Spending on AI, hardware and infrastructure is acting as a positive demand shock. Whether it becomes a positive supply shock, through faster productivity growth, "remains to be seen" (p. 1). Fiscal support to cushion energy costs, rising defence spending and swings in financial conditions add to the picture.
The IMF says policymakers must do two things at once: contain the near-term risks from inflation, debt and financial fragility, while building the infrastructure, skills and competitive environment that let technology lift growth more broadly (p. 2).
Recent developments: resilient but uneasy
Growth in the first half of 2026 held up better than expected, but unevenly (p. 2). US growth slowed to 2.2% at an annual rate in the second quarter after 2.5% in the first. The euro area rebounded to 2.6% in the second quarter, driven mainly by Ireland and Spain. India's growth stayed robust despite higher oil prices. China's second quarter was weaker than expected, at 4.3% year on year, with weak consumption and property strains only partly offset by exports. Global headline inflation rose in August as energy prices bounced back after a short-lived ceasefire. Core inflation stayed moderate, though in some advanced economies core goods prices began to reflect the rising cost of memory chips.
Sources of resilience
Since 2023 the IMF's forecasts have underestimated global growth three years in a row, although the errors were small (p. 3). The IMF points to household and company cash cushions built during the pandemic, long-term fixed-rate debt taken on when rates were low, central banks' credibility, and better policy frameworks. It also notes that the world relies less on oil than it did: energy use per dollar of global output has fallen 32% over two decades, and renewables account for 60% of the rise in electricity production since 2010 (p. 4).
The report estimates what would have happened without the flexibility in supply. A closure of the Strait of Hormuz with no supply response would have pushed oil to about $110 a barrel for the year and cut 2026 global growth by 0.9 point. Rerouted production, extra output elsewhere and stock drawdowns kept the average near $85 and the hit to 0.4 to 0.5 point (p. 4).
The AI offset, and its limits
AI-related investment is the main offset to the war's drag. Net trade in AI hardware has grown sharply as a share of world output (p. 5). But the gains are concentrated. Taiwan Province of China and Korea source 62% of net trade in AI computing hardware, and China and the United States account for more than 75% of net imports of it and 57% of newly built data centres since 2024 (p. 6).
On productivity the evidence is mixed. Studies find large task-level gains, for example 14-15% in customer support (30-35% for less experienced staff) and 26% in software output, but experienced open-source developers were 19% slower with AI, and economy-wide data show only modest effects so far (p. 6).
The boom also has costs. It is a large, front-loaded demand shock that strains capacity in capital goods, construction, land, energy and skilled labour before any broad productivity gains arrive. It has also stretched equity valuations and concentrated the market in AI-related firms (p. 7). Energy inventories are running down, and public deficits remain larger than before the pandemic. In most large economies, stabilising debt at 2025 levels would need more tightening than current policies imply, with India an exception (p. 7).
The baseline and its assumptions
The forecast rests on stated assumptions (pp. 8-9). The Strait of Hormuz is assumed to reopen from March 2027, with prewar conditions broadly back by October 2027. Policy and geopolitical uncertainty stay elevated through 2027. The AI cycle is assumed to moderate, with no extra productivity boost from AI. Oil is up 41% in 2026 relative to 2025 and natural gas 48%; fertiliser is up 10% and food 5%. Monetary policy is expected to be less supportive than the IMF assumed in April, with policy rates in the euro area and the US rising in real terms. Public debt keeps climbing: in advanced economies from 108% of output in 2025 to 115% in 2031, and in emerging markets from 74% to 84%. Current trade policies, including measures described as temporary, are assumed to stay.
The forecast by region
Advanced economies are projected to grow 2.1% in 2026 and 1.8% in 2027. The US is at 2.3% in both years, the euro area 1.1% and 1.2%, the UK 1.3% and 1.2%, Japan 0.7% in both, Korea 3.3% and 2.8%, and Canada 1.1% and 1.7% (pp. 10-12). Emerging market and developing economies slow from 4.5% in 2025 to 3.7% in 2026, then recover to 4.4%. China is at 4.5% and 4.0%. India is at 7.0% in both years on a fiscal-year basis, upgraded in the October report and described as among the fastest-growing major economies (pp. 12, 15). The Middle East and Central Asia contract 0.3% in 2026 before rebounding 6.3%; Bahrain, Iraq, Kuwait and Qatar face double-digit contractions in 2026 and double-digit rebounds in 2027, and Saudi Arabia is projected at -3.3% then 7.9% (p. 13). Sub-Saharan Africa holds at 4.5%. World trade volumes grow 4.6% in 2026 and 4.7% in 2027 (p. 14).
Risks
The IMF says risks are tilted to the downside (p. 15). The main ones are a prolonged conflict or escalation in the Middle East or Ukraine, with cyberattacks and extreme weather (including the current El Niño) adding to commodity stress (pp. 15-16); fiscal and sovereign strains, where banks' holdings of government securities are linked to lower private lending, especially in high-debt emerging markets (pp. 16-17); and renewed trade tensions (p. 17).
Upside risks are a longer-lasting AI boom (a scenario in which a bigger AI build-out lifts world output by 0.6% by 2029), new trade agreements such as the EU's with India and Mercosur and the UK's with India, and faster diversification of energy routes and sources (pp. 17-18). The IMF's October financial stability report cites a WEO scenario in which an unwinding of AI investment cuts global output by 0.4%, investment by 1.3% and consumption by 0.6% by 2028.
A prolonged-conflict scenario keeps key shipping routes largely shut to the end of June 2027, takes oil to $130 for a few months, and raises long-term borrowing costs. In it, advanced economies' output sits 0.7% below baseline by 2031 and their public debt rises by more than 3 points of output (pp. 26-27).
Policy: agility under constraint
On fiscal policy, the IMF says support should be sparing, targeted and temporary. Broad fuel subsidies, generalised tax cuts and price controls "should be avoided"; hardship support should go to vulnerable households, preferably as cash transfers, with eligibility rules, sunset clauses and offsets set up front. The bar for helping firms should be high (p. 19). On monetary policy, the right stance differs by economy. Energy importers can look through the direct price effect only if the shock is temporary and expectations stay anchored, and where inflation is already above target the IMF sees a bias toward tightening (p. 20). On financial policy, supervisors should watch sovereign-bank links, foreign-currency exposures, non-bank lenders and liquidity mismatches, and stress tests should combine commodity spikes, currency falls, higher term premiums and an asset-price correction (p. 21). Structural reforms should raise business dynamism and energy security (p. 21). Emergency IMF liquidity is a backstop but "not a substitute" for orderly debt restructuring (p. 22).
Chapter 2: cost-of-living crises
Chapter 2 examines surges in the prices of food, energy and other basic necessities. They hit lower-income households hardest and risk unsettling inflation expectations (Ch. 2, p. 1). The first line of defence, it says, is monetary policy aimed at low and stable inflation plus social protection systems that act automatically. Past episodes show advanced economies leaning on price measures near the consumer, such as temporary VAT or excise cuts, and on income support, while emerging and low-income economies relied more on producer subsidies, price controls, customs cuts and broad wage and pension rises (pp. 17-18).
The key finding is the cost. Giving vulnerable households the same protection through price-suppressing measures can take between 3 and 22 times the fiscal resources of targeted income support. Producer subsidies are especially inefficient because part of the benefit leaks abroad, and price controls work less well when many countries use them at once (p. 18).
Chapter 3: corporate tax spillovers
Multinationals now account for more than 20% of global output and about 15% of global corporate profits, and cross-border investment has nearly quadrupled since the 2000s (Ch. 3, p. 1). The chapter finds that tax competition has eased on headline rates after the reforms of the 2010s but persists, most strongly among economies at similar stages of development. Anti-avoidance reforms have made profit shifting less sensitive to tax differences while making real investment more sensitive, so profits and real activity are lining up more closely (p. 17). Debt-financed tax cuts can push up global interest rates and crowd out investment abroad, and cuts paid for by lower public investment can weaken productivity over time. Stronger anti-avoidance rules can protect revenue and support output, especially in emerging markets that rely on corporate tax (pp. 1, 18).