Perspectives
July 12, 2026

Macro Pulse for July 10: Chipflation: The Bottlenecks Driving a More Volatile Leg of Tech Leadership

AI is still the theme. But markets are now pricing the constraints. The AI buildout is no longer just about compute. It increasingly depends on memory, bandwidth, data centers, and power: the physical infrastructure needed to scale AI. For investors, that means opportunity, but also higher volatility.

By Global Market Strategy Team

In 2024, we argued that AI’s first investable chapter would be in the physical “infrastructure layer”: semiconductors (computer chips), data centers, power capacity, and related materials. That framework continues to play out: strong and growing capital expenditures by hyperscalers; soaring rental rates in data centers; a boom in commodities tied to tech; and more.

Semiconductors have always been central to these dynamics, but in recent months, chip producers have become the core source of tech leadership in equity markets. The chip moonshot has expanded beyond that of NVIDIA and logic chips to include memory chip giants, where scarcity has created powerful pricing and export momentum. Now, the same “chipflation” leading returns can pressure hyperscaler margins and drive volatility within equity markets.

What makes today’s semiconductor leadership different: memory

The semiconductor giants now dominating market leadership have distinct roles in the tech supply chain. Computer chip capacity can be divided into two high-level realms: logic and memory.

Logic chips are the brains behind any AI model, in addition to the brains behind our vehicles, appliances, and smartphones. NVIDIA is the world’s leading designer of chips suited to AI compute. Taiwan Semiconductor, or TSMC, manufactures roughly 90% of the world’s logic chips on the island of Taiwan (including those designed by NVIDIA). Up until this year, logic chips were the key beneficiary of the AI trade, as demand for sophisticated compute capacity to train generative AI models surged.

Memory chips are the storage capacity for all that compute. Korea is the world’s memory specialist, with giants Samsung and SK Hynix together comprising nearly half of the Korean KOSPI equity index; in the U.S., Micron is the predominant memory supplier. Historically, memory has been a highly commoditized product, subject to consistent over- and under-supply cycles. Memory is what makes this year of market leadership different: in recent months, the demand for high-bandwidth memory to store the data generated by AI models has surged. South Korea’s exported memory value is growing at 370% YOY, a far cry from historical cyclical peaks of ~100% YOY growth.

Iran conflict has reversed nearly two years of disinflation

Our 2024 AI framework emphasized that AI adoption is constrained by physical infrastructure. Back then, the market focused on compute. Three years later, the bottleneck has broadened: compute still matters, but high-bandwidth storage capacity increasingly influences costs across the AI supply chain.

Chipflation’s market manifestation: narrow leadership

South Korea is the top performing country-level equity index in 2026, up over 71% in USD terms year to date (total return as of July 10 close). It’s also hovering near bear market territory, down nearly 18% in the past two weeks. Meanwhile, the Magnificent 7 hyperscalers are lagging, flat on the year.

This performance divides the companies footing the bill for chipflation from the suppliers benefiting from it. The Magnificent 7 hyperscalers are seeing input costs rise – for both compute and memory – and these costs are factoring into investors’ questions around margins, how sustainable the growth rate of capex may be, and when these capacities will show an acceptable return on investment.

A step-change in pricing does not make for durable market leadership

While the relative “winners and losers” of chip pricing are an important piece of the fundamentals, chips are just one segment of the tech supply chain and represent a segment of market leadership that comes with distinct risk. On 90-, 180-, and 360-day tenors, the major semiconductor indices show higher volatility than the Magnificent 7 over the past decade.

This volatility reflects greater cyclicality, commoditization of memory capacity, and monopolistic industry structures. Accordingly, we see increased risks around a memory-led market. We believe a more holistic set of strong fundamentals are better reflected in leadership by the hyperscalers, which – for all the “live” questions around capex and input costs – continue to deliver world-class earnings growth and strong cash-flow generation that fuels ongoing physical investment rates.

Portfolio strategy

The chip moonshot may well have several more months or quarters to run, and we expect to see ongoing volatility as the pricing cycle matures. But over a medium-term horizon, we believe leadership will retrench in strong profitability found in the hyperscalers. Our conviction in this view strengthens in a more hawkish interest rate scenario (more likely if the Iran conflict re-escalates), as a more restrictive rates environment may pressure more cyclical segments of the market, including memory.

Leadership within the tech segment has shifted rapidly this year, leaving investors looking for areas of respite across asset classes. Within U.S. equities, finding true diversification requires looking under the surface.

As tech’s influence permeates markets, even small caps offer less diversification than appears on the surface: the Russell 2000 benchmark is having its best run in decades, up nearly 24% since the end of March, but driven by tech (the NASDAQ small cap tech technology index is up nearly 42% over the same period).

We do see greater diversification potential within the financials sector. All systemically important U.S. banks passed the Fed’s stress test last month with flying colors, then followed the passing grade with widespread announcements of dividend increases and buyback program expansions. Though a flatter yield curve has not favored profitability of bank lending, the cash flow generation outlook is robust, suggesting financials have lagged this year more due to sentiment than fundamentals.

Recent From the Desk responses

Recent deep dive research

This material represents an assessment of the market environment as at a specific date; is subject to change; and is not intended to be a forecast of future events or a guarantee of future results. This information should not be relied upon by the reader as research or investment advice regarding any funds or any issuer or security in particular. The strategies discussed are strictly for illustrative and educational purposes and are not a recommendation, offer or solicitation to buy or sell any securities or to adopt any investment strategy. There is no guarantee that any strategies discussed will be effective.