The War Wiped $7 Trillion Off Stocks


GOOD MORNING, The first quarter is ending with markets in a very different place than where they started. The Iran war has erased roughly $7 trillion from global equities, pushed crude prices up about 70% this year and turned expected Fed cuts into a meaningful chance of hikes. At the same time, Big Tech is preparing to spend around $630 billion on AI infrastructure — but the next constraint may be less about demand than whether enough data centers can physically be built.


MARKETS | TLDR


WAR PREMIUM

The War Wiped $7 Trillion Off Stocks

The Iran war has erased roughly $7 trillion from global equity markets since it began, according to Reuters, while oil and natural-gas prices have risen around 70% and 85% respectively this year. U.S. stocks ended Friday with a fifth consecutive weekly decline, and the Dow joined the Nasdaq and Russell 2000 in correction territory.

The important part is how broadly the shock is transmitting. Oil is feeding into gasoline, transportation costs and inflation expectations; Treasury yields have risen as markets rethink central-bank policy; and consumer and business confidence are beginning to soften. The S&P 500 is now down more than 7% since the U.S.-Israeli strikes on Iran began in late February.

That leaves markets in a difficult regime. A ceasefire could remove part of the energy premium quickly, but words alone are no longer moving the tape the way they did earlier in the conflict. Investors increasingly want physical evidence that Hormuz is reopening and supply is normalizing. Until that happens, the war is not just a geopolitical story — it is the macro story.


AI CAPEX

Big Tech Has $630 Billion to Spend. Building It Is the Problem.

Amazon, Microsoft, Alphabet and Meta are projected to spend about $630 billion on AI chips and data centers in 2026, according to Morgan Stanley estimates cited by Reuters Breakingviews. That number has become a shorthand for bubble risk: what happens if companies build enormous amounts of infrastructure and demand fails to justify it?

The more immediate risk may be the opposite. Big Tech may struggle to spend the money fast enough. New data centers require power connections, transformers, cooling systems, networking equipment, construction labor and permits — all of which can become bottlenecks long before customers stop asking for AI compute.

That distinction matters for the AI trade. Suppliers can keep seeing strong orders even if hyperscalers miss their original deployment timelines, while revenue from completed capacity may arrive later than investors expect. The question is therefore shifting from “will they spend?” to “how much of the spending can actually become productive infrastructure on schedule?”


HEADLINES


UPCOMING

  • JOLTS arrives March 31: Job openings and hiring will show whether the war-driven confidence shock is beginning to affect labor demand.
  • Consumer confidence is due March 31: The survey will provide another read on how higher gasoline prices are affecting household expectations and spending plans.
  • ISM manufacturing arrives April 1: Prices paid and new orders will show whether the energy shock is feeding into factory costs before it reaches broader inflation data.
  • The March jobs report lands April 3: Payrolls and unemployment will be the biggest near-term test of whether the Fed is facing only an inflation shock — or a simultaneous growth slowdown.

DEEP INSIGTHS

Markets in Q1: Everything, Everywhere, All at Once

Reuters' quarter-end review puts the scale of the regime change in perspective: roughly $7 trillion erased from global equities, oil up around 70%, natural gas up 85% and rate-cut expectations replaced by renewed tightening risk. It is useful context for understanding why even strong company fundamentals are struggling to dominate the tape.

How Big Tech's $630 Billion AI Splurge Could Fall Short

Reuters Breakingviews argues that investors may be focusing on the wrong AI risk. The immediate problem is not necessarily weak demand, but the physical challenge of turning enormous budgets into functioning data centers quickly enough — a constraint that links AI growth directly to power, construction and industrial supply chains.

background

Subscribe to Markets in Minutes