On July 8, 2026, the International Monetary Fund (IMF) released its World Economic Outlook (WEO) Update and gave it a telling subtitle: "Global Economy in Crosscurrents of War and Technology" [source: IMF World Economic Outlook, 2026]. The title alone captures where the world economy now sits. On one side, war in the Middle East is pushing energy prices up; on the other, an investment boom led by artificial intelligence (AI) is holding demand up. The two flows run in opposite directions. That is what makes this a year in which a single number struggles to sum up the global economy.
The IMF's latest outlook compresses into three figures: global growth of 3.0 percent in 2026, headline inflation revised up to 4.7 percent, and the stalling of the disinflation — the fall in prices — that had been under way since early 2024 [source: IMF World Economic Outlook, 2026]. Growth is slowing gently but not collapsing, and inflation has turned sticky again. Behind those numbers sit exactly those two crosscurrents: the war shock and AI-driven demand.
That makes now a good moment for a mid-year checkup on the global economy. This article sets out the figures the IMF presented, then separates the two forces that produced them — the energy shock made by war and the demand made by technology — before turning, in order, to stalled inflation, growth that is splitting apart country by country, and the points worth watching ahead. One premise up front: most of the numbers here are IMF projections, not settled actuals. That distinction runs through the whole piece.
A word on sourcing follows from it. The figures below come from the IMF's own July 2026 release wherever possible; where the IMF's pages were not directly reachable, the same numbers were checked against outlets reporting them, among them Bloomberg, The Korea Times, and The National. Where second-hand accounts disagreed — country-by-country estimates of war damage are the clearest case — the figures were left out rather than averaged. Three layers therefore run through what follows: outturns already recorded, projections the IMF publishes with a date attached, and the assumptions those projections rest on.
Table of Contents
- The 2026 numbers — inside 3.0 percent growth
- The war shock — Hormuz and the shadow of energy
- The technology pull — the other half AI holds up
- Disinflation that has stalled
- A world pulling apart — what the crosscurrents really mean
- Conclusion — what to watch
The 2026 numbers — inside 3.0 percent growth
The headline path
The IMF projects global growth of 3.0 percent in 2026 and 3.4 percent in 2027, a step down from the 3.5 percent averaged in 2024–25 [source: IMF World Economic Outlook, 2026]. Yet the IMF adds that, compared with its April 2026 WEO, this outlook is broadly unchanged on a cumulative, two-year basis. In other words, the core message is that the world economy is slowing gently, not snapping.
That phrase — broadly unchanged on a cumulative basis — is doing careful work. A cumulative, two-year comparison adds 2026 and 2027 together and asks whether the total has moved; the total can hold still while the individual years shift, if a downgrade in one is offset by an upgrade in the next. The IMF is therefore making a narrow claim: the level of output it expects by the end of 2027 has barely changed since April. It is not claiming that nothing happened in between.
The country picture
Here is the country picture hidden behind the aggregate.
- Advanced economies overall: 1.7 percent in 2026, 1.8 percent in 2027 [source: IMF World Economic Outlook, 2026]
- United States: 2.3 percent in 2026, 2.2 percent in 2027 — virtually unchanged from April [source: IMF World Economic Outlook, 2026]
- Euro area: about 1.2 percent in 2026 — weighed down by energy dependence and manufacturing competitiveness [source: IMF World Economic Outlook, 2026]
- China: 4.6 percent in 2026 — revised up from 4.4 percent in April [source: IMF World Economic Outlook, 2026]
Three things make that list readable: every figure covers the same calendar year, comes from the same July 2026 vintage, and is produced by one institution on one method — which is what makes them comparable at all. A fourth matters more: all of them are projections. Growth numbers from different bodies, or from the same body in different months, are not interchangeable, so each figure here is quoted with the outlook it came from.
Even this short list is wider than the aggregate suggests: about 1.2 percent for the euro area against 4.6 percent for China, with the world average between them at 3.0 percent [source: IMF World Economic Outlook, 2026]. Two further figures — Korea's and the Middle East's — appear later, and they stretch the range at both ends.
Projection is not outturn
One important layer here. These numbers are all projections. Actual outcomes can miss them, and recently they have. According to the IMF, in the first quarter of 2026 the growth of several technology-exporting economies came in an average of 4.4 percentage points above the original forecast [source: IMF World Economic Outlook, 2026]. In macro forecasting, 4.4 percentage points is no small miss. Put differently, 3.0 percent is not a fixed trajectory but the central scenario the IMF draws at this moment.
The basis of that surprise matters. The 4.4-percentage-point gap is an average for the first quarter of 2026, measured on a seasonally adjusted annualized basis — the quarter's growth expressed as if it ran for a full year [source: IMF World Economic Outlook, 2026]. Quarterly rates stated that way swing much harder than annual ones, so a miss of that size does not carry into the annual number one for one. It is evidence about direction, not a ready-made revision to the year.
It also explains why the IMF republishes its numbers every few months. Each outlook is a reading taken at a moment, using what was known then; the April edition and the July update differ precisely because the quarter between them contained new facts. A projection quoted without its date and issuer is missing half of what it says.
The war shock — Hormuz and the shadow of energy
What happened, and where
One of the crosscurrents is war. In late February 2026, hostilities between the United States, Israel, and Iran escalated, and a period of closure at the Strait of Hormuz followed [source: IMF World Economic Outlook, 2026]. Roughly one-fifth of the world's crude oil and liquefied natural gas (LNG) supply passes through Hormuz. When that narrow waterway is disrupted, the shock spreads straight into the entire energy market.
A chokepoint matters out of all proportion to the cargo sitting in it at any moment. Once a strait carrying that share of the world's crude and LNG turns uncertain, the effect travels through routes, schedules, and the cost of covering risk — which is why energy prices can move on the possibility of interruption, not only on interruption itself.
The price assumption behind the forecast
Prices show the shock. The IMF assumes a global oil (petroleum index) price averaging about 89 dollars a barrel in 2026 — well above the roughly 62 dollars assumed in the pre-conflict January 2026 WEO [source: IMF World Economic Outlook, 2026]. When oil rises, the countries that buy energy feel it first.
That 89-dollar figure is an assumption rather than a market print: it is the average price the outlook carries for 2026, about 9 percent above the assumption used in April [source: IMF World Economic Outlook, 2026]. Every growth and inflation figure in this update rests on that oil path. If energy settles well below it, the downgrade for importers and the upgrade to inflation are both too gloomy; if it settles above, the reverse. Assumptions are where a forecast is most exposed and least discussed.
Same shock, different report cards
As a result, the same war left different report cards in different places. Growth in the Middle East and North Africa fell to about 0.7 percent for 2026, sharply lower than the 1.9 percent projected in April — though the IMF expects the region to rebound strongly in 2027 [source: IMF World Economic Outlook, 2026]. The United States, by contrast, is a net energy exporter, so it absorbed less of the oil-price blow, and its growth outlook held nearly steady at 2.3 percent [source: IMF World Economic Outlook, 2026]. Energy-importing economies such as the euro area stand on the opposite side. The same shock becomes a cost for some and an income for others.
The size of that regional cut is worth stating plainly: about 1.2 percentage points came off the April projection in the space of one quarter [source: IMF World Economic Outlook, 2026]. The recovery expected on the far side is just as steep — The National, reporting the IMF's regional figures, put the 2027 rebound at 6.5 percent [source: The National, 2026]. A V of that shape is the usual signature of a supply interruption rather than a demand collapse: output that is blocked, not destroyed, comes back when the blockage clears. It is also, for now, a projection about a year that has not begun.
Underneath the different report cards lies a simple asymmetry. For a net energy exporter, a higher oil price arrives partly as income; for a net importer it arrives as a bill, payable in foreign currency before anything else is bought. That is why the United States held near 2.3 percent while the euro area — importing its energy, competing in manufacturing — sits at about 1.2 percent [source: IMF World Economic Outlook, 2026]. One event, entered on opposite sides of the ledger.
The technology pull — the other half AI holds up
The offset, in the IMF's words
So why did global growth not bend more deeply? Because of the crosscurrent flowing the other way — technology. The IMF explains that this slowdown reflects "the effects of the war in the Middle East being partly offset by accelerated demand-driven momentum in the global technology cycle thanks to advances in artificial intelligence (AI) and its adoption" [source: IMF World Economic Outlook, 2026]. As much as the war eats into growth, the AI investment boom has filled part of the hole.
"Partly offset" is arithmetic, and deserves to be read as arithmetic. Two large forces can pull hard in opposite directions while the number surviving their collision stays small. A modest slowdown in the aggregate is therefore consistent with a severe shock in one part of the world economy and a boom in another; it does not mean either was mild. The average is the residue of the contest, not a description of it.
The four hardware exporters
At the front line of this flow are Asia's hardware-exporting powers. The IMF names South Korea, Taiwan, Thailand, and Malaysia as the world's largest net exporters of AI-related hardware and the standout beneficiaries of the current technology cycle. The 4.4-percentage-point first-quarter growth surprise mentioned above is precisely the average across these four economies [source: IMF World Economic Outlook, 2026].
Two clarifications keep that number honest. It is an average across the four, so it says nothing about any single one of them; and it measures outcomes against the forecast rather than the previous quarter, so it records how wrong the forecast was rather than how fast those economies grew [source: IMF World Economic Outlook, 2026]. What it establishes is that the technology channel was strong enough to move economies of that size faster than the people modelling them expected.
Korea, the clearest case
South Korea is the clearest case. The IMF raised its 2026 growth forecast for Korea to 2.6 percent, up 0.7 percentage points from April — the largest upgrade among 30 major economies [source: IMF World Economic Outlook, 2026]. What stands out is that Korea is an energy importer. Even carrying the war-driven oil burden, its strong external demand for semiconductors and other technology goods dominated that negative effect. Some of the warmth of the technology cycle shows up in China's small upgrade as well. But let us be clear on one thing: "partly offset" is the IMF's additive reading of the net effect of two shocks, not a causal claim that the war produced the AI boom. The two flows exist at once, each for its own reasons.
Korea comes close to a natural experiment for the crosscurrents thesis, which is why it carries so much weight here. The country sits on the losing side of the energy channel and the winning side of the technology channel at once, and the technology side won by enough to produce the largest upgrade in the group — a figure reported alike by Bloomberg, The Korea Times, and the Seoul Economic Daily [source: Bloomberg, 2026]. When two channels can be told apart inside one economy, the net result says something an aggregate never can.
The same case shows what would falsify the story. If the upgrade rests on a hardware cycle, it lasts exactly as long as that cycle does; a slowdown in AI-related investment would remove the very thing now outweighing the energy burden. For these economies the optimistic case and the central risk are one fact seen from two sides.
Disinflation that has stalled
The projected path
Behind the growth story sits prices. The IMF projects global headline inflation rising from 4.1 percent in 2025 to 4.7 percent in 2026, before easing to 3.9 percent in 2027 [source: IMF World Economic Outlook, 2026]. The 4.7 percent figure for 2026 is an upward revision from April, while the projection for core inflation was left broadly unchanged.
The split between those two measures is the most informative part of that sentence. Headline inflation includes energy and food; core strips them out. Raising the headline while leaving core broadly unchanged is a statement about where the pressure originates — in the supply of energy and food rather than in demand running hot [source: IMF World Economic Outlook, 2026]. It is also why the revision is tied to the aftermath of the Hormuz disruption rather than to an overheating world economy.
Why the descent stopped
The trend matters more. The IMF judges that the disinflation under way since early 2024 has now stalled [source: IMF World Economic Outlook, 2026]. For two years the world had been gradually bringing down the high inflation that followed the pandemic; that descent has come to a halt. Much of the stall traces back to the energy and food prices seen earlier — the supply shock out of Hormuz.
Energy shocks do not arrive all at once. Fuel prices move first, then the cost of moving goods, then food, and each step takes months to work through contracts and price lists. A two-year disinflation can therefore be halted by a shock already in the past, because its consequences are still travelling through the price level. "Stalled" describes the shape of the path, not a fresh burst of inflation.
What it does and does not imply for policy
What this means should be read with care. If prices cannot fall further toward target, the room for central banks to cut interest rates may narrow accordingly. But this is a direction of travel, not a settled outcome. The actual policy path depends heavily on how fast energy prices stabilize, how core inflation moves, and the state of each country's economy. The fact that inflation has "turned sticky again" and any verdict that "monetary policy will go this way" are claims at clearly different layers.
Read strictly, the policy claim is conditional on three things at once: that energy prices stay near the assumed path, that core inflation does not fall fast enough to offset the headline, and that each central bank weighs price stability against a labour market and a growth outlook that differ country by country. What can be said is that the room to cut has narrowed inside the IMF's projected world; what cannot be said is what any particular committee will decide.
A world pulling apart — what the crosscurrents really mean
The dispersion inside the average
What the word "crosscurrents" really points to is the dispersion an average hides. Within the single figure of 3.0 percent global growth, economies plugged into the AI value chain and economies that rely on importing energy and commodities are moving in opposite directions [source: IMF World Economic Outlook, 2026]. The aggregate looks like a gentle slowdown, but the gaps inside it are, if anything, widening.
Set the projections for one year side by side and the range is the story: about 0.7 percent for the Middle East and North Africa, about 1.2 percent for the euro area, 2.3 percent for the United States, 2.6 percent for Korea, 4.6 percent for China — all for 2026, all from the same July outlook, all averaging out at 3.0 percent [source: IMF World Economic Outlook, 2026]. Not one of them is growing at the world average. That is what an average does: it names a number that describes nobody.
Two readings, held at once
Two opposing readings of this picture are possible. Optimistically, even the war-hit Middle East is projected to rebound strongly in 2027, and AI-led demand still looks firm — the so-called V-shaped recovery. Cautiously, the risk that the Hormuz situation deteriorates again, the uncertainty over how long the AI investment cycle can last, and the stalled disinflation all remain as downside risks. The IMF itself, in presenting this outlook, stressed that risks and uncertainty are large [source: IMF World Economic Outlook, 2026].
The two readings are not evenly evidenced. The optimistic case rests mostly on projections — the 2027 rebound, the continuation of AI-led demand — while the cautious case rests on uncertainties nobody can measure yet: whether Hormuz stays quiet, how long an investment cycle runs. Neither side is quoting outcomes. The one hard data point nearby, the first-quarter surprise, cuts the optimists' way while showing how far reality can sit from any of these numbers.
The April-to-July reframing
The clearest evidence that the IMF changed its own frame is on the cover. The April 2026 outlook carried the subtitle "Global Economy in the Shadow of War"; three months later the July update reads "Global Economy in Crosscurrents of War and Technology" [source: IMF World Economic Outlook, 2026]. In April one dominant force stood over everything else. By July there were two, running against each other. Nothing in the war had ended; something else had grown large enough to be named beside it.
So we return to the distinction between projection and actual. As the first-quarter miss of 4.4 percentage points shows, today's crosscurrents could flow harder in either direction. Neither 3.0 percent nor 4.7 percent is a fixed ending — each is only the mid-game score that the tug-of-war between the two forces has produced so far.
Conclusion — what to watch
The scoreboard, restated
The mid-2026 report card on the global economy is hard to sum up in one line. Growth has slowed gently to 3.0 percent, inflation has climbed again to 4.7 percent, and disinflation has stalled. War has pressed down on energy importers, while AI has lifted technology exporters [source: IMF World Economic Outlook, 2026]. That two flows running in opposite directions move through one and the same global economy at once — that is the true meaning of the word "crosscurrents."
Four things to watch
The points to watch ahead are clear. First, the Hormuz situation and the path of oil prices; if the energy shock calms, room opens for disinflation to resume. Second, how long the AI investment and export cycle can hold demand up; the durability of that boom will decide the depth of the slowdown. Third, the gap between the two flows — whether the growth divide between economies inside and outside the AI value chain widens further. Fourth, whether all these projections are confirmed by actual outcomes. What is needed now is not a verdict for one side, but the discipline to watch the two crosscurrents of war and technology together, at the same eye level.
One habit is worth carrying out of this update. Every figure here belongs to an issuer and a date — the IMF, July 2026 — and the next edition will move some of them, exactly as this one moved April's. When those numbers land, the useful question will not be whether 3.0 percent was right, but which of the two crosscurrents the revision came from.