Goldman Sachs reintroduces "HALO trading": the second phase has just begun.
Heavy asset stocks have outperformed light asset stocks, a trend that has lasted for over a year. However, in its latest report released on July 7, Goldman Sachs Europe Equity Strategy Team stated: the first phase of the HALO trade—valuation repair—has basically concluded, and the second phase has just begun, with the driving force switching from “repricing” to “profit realization.”
What does this switch mean? Simply put, “HALO” was previously bought because it was cheap; going forward, real earnings will have to support stock prices. HALO stands for Heavy Assets, Low Obsolescence.
Up 20% year-to-date, geopolitical shocks have strengthened the logic
The HALO pair trade—long heavy asset stocks (GSSTCAPI), short light asset stocks (GSSTCAPL)—has risen about 20% year-to-date.
Geopolitical shocks briefly disrupted this trade but failed to reverse it. In the early days of the US-Iran war, Goldman heavy asset stocks (GSSTCAPI) fell about 7%, mainly because these stocks have larger cyclical exposure to manufacturing and global trade. However, the decline ended quickly, and they are now about 2% higher than before the conflict.
More importantly, the geopolitical shock has actually strengthened HALO’s underlying logic. Narratives around energy security and industrial autonomy are clearer; oil and gas (about 10% of the basket) and utilities (about 10%) benefit directly, while infrastructure provides resilience via inflation-linked mechanisms.
The only exception is aerospace and defense (about 10%)—despite favorable geopolitical conditions, this sector has underperformed expectations. Goldman attributes this to profit-taking after four straight years of strong gains, but maintains a positive medium-term view for aerospace and defense.
Valuation gap nearly closed, but the story isn't over
Heavy asset stocks currently have a 12-month forward PE of about 14.5x; light asset stocks about 16x—the valuation discount between the two has narrowed substantially.
This means the first phase of "valuation re-rating" is basically complete. But that does not mean the trade is over.
There are three reasons:
First, there is a clear positive correlation between capex and valuations—the capex cycle is highly correlated with the relative valuation of heavy asset stocks; every step up in the capex-to-sales ratio expands the heavy asset stocks’ relative premium. This ratio is now at multi-year highs, suggesting further support for valuations.
Second, high interest rates continue to suppress light asset stocks that rely on distant cash flows for valuation, especially software and services companies, whose valuations largely come from terminal values far in the future. Meanwhile, AI disruptions make it hard to determine which business models will be overturned, which technology paths will win, and how long current competitive advantages will last.
Third, the book value of many heavy asset companies may not yet fully reflect the true replacement cost of their assets in a high inflation environment.
By contrast, heavy asset companies have physical networks and infrastructure, long build cycles, high regulatory barriers, and are hard to replicate quickly, which naturally forms a moat against disruption.
Earnings-driven phase begins: 15% vs 10%, the gap is widening
The HALO trade is shifting from "valuation-driven" to "earnings-driven."
For a long time, light asset companies' EPS growth consistently led, which was the fundamental reason for their long-term outperformance. But this is reversing.
Year-to-date, heavy asset stocks (GSSTCAPI) have seen the strongest positive EPS revisions in all European baskets; during the same period, light asset stocks’ EPS revisions were near zero or even slightly negative.
Market consensus expects heavy asset stocks’ EPS growth in 2026 to be about 15–16%, light asset stocks about 10%—this is the first clear reversal in years.
More importantly, analysts say: "The next phase does not require further valuation expansion, or even significant earnings upgrades. It just needs to deliver."
This means the coming divergence will be sharper—not the entire heavy asset sector rising together, but only companies that can actually deliver on earnings expectations will outperform, while those that cannot will lag.
AI capex supercycle: 13% growth, more than 6x historical median
The underlying driver supporting the HALO logic is an unusually large capex expansion wave.
Global capex is expected to grow about 13% in 2026, while the historical median is only about 2%; 2027 expectations are also climbing. The capex-to-sales ratio has reached multi-year highs, reversing the systemic underinvestment of the past decade.
But this capex cycle has a defining feature: it is highly concentrated.
Data centers, semiconductors, utilities, and defense are expected to account for over 40% of global capex in 2026, up from about 25% in 2022. Meanwhile, mid-term capex expectations for traditional sectors like chemicals, construction equipment, and airports are relatively weak.
This concentration is itself a structural signal—AI investment, energy transition, and reindustrialization are the main drivers of this cycle, not simple inventory restocking or cyclical rebound.
This is not just Europe’s story; it’s a shift in global market leadership
Some may ask: Is HALO only applicable to Europe?
The answer is no. Goldman has constructed similar stock baskets in the US, Asia-Pacific, Japan, and emerging markets, and observed consistent patterns: heavy asset stocks outperform, light asset stocks lag, valuation gaps narrow globally.
Since 2022, heavy asset stocks in Asia-Pacific are up 112%, light asset stocks down 15%; US heavy asset stocks up 71%, light asset stocks up 25%; Europe heavy asset stocks up 92%, light asset stocks down 2%; Japan heavy asset stocks up 37%, light asset stocks down 7%.
However, regional differences do exist. The light asset sector (especially large tech stocks) has a much higher weight in the US market than in Europe, Japan, or some emerging markets. This partly explains why non-US markets have relatively stronger performance this year.
This also means that markets with lower light asset weights may have more room for HALO’s re-rating.
Funds have not yet arrived, allocation space remains
Tactically, there is a risk to note: correlation between HALO and the momentum factor is at a historic high. This means if markets rotate factors or make systemic position adjustments, heavy asset stocks may face short-term pullback pressure.
But from a longer-term perspective, fund reallocations are far from complete.
In the last 12 months, net inflows into European value funds were about 3% of assets under management, while growth funds saw net outflows of about 15% of AUM (Source: EPFR). Industrial, infrastructure, and commodity funds continue to attract flows, while healthcare and consumer funds continue to see outflows. However, total flows into European value funds relative to growth funds are still at about -30% historical lows—in other words, despite recent shifts, long-term under-allocation has not been fixed.
In terms of stock baskets, about 50% of heavy asset basket constituents currently have a buy rating, about 10% have a sell rating. Buy ratings are mainly concentrated in five themes:
1. Infrastructure: Operators of power grids, pipelines, fiber networks, and transport assets, hard to replicate.
2. Basic materials: Holders of key resources and industrial facilities needed for construction, manufacturing, and electrification.
3. Aerospace and defense: Highly specialized manufacturing capabilities, long product cycles, rising geopolitical demand.
4. Manufacturing and consumer platforms: Large-scale production assets, distribution networks, and persistent consumer demand.
5. Technology physical layer: Hardware and infrastructure supporting the digital economy, from semiconductor equipment to telecom networks.
These five directions appear quite different, but share commonality: scarce physical assets, high entry barriers, rising replacement costs, limited technological obsolescence risk.
Light assets won't disappear, but aren’t uncontested winners anymore
One last question: are light asset stocks now out of the game?
The answer: No, but their situation has changed.
AI’s biggest impact may not be declining profit margins, but declining investor confidence in “terminal value.” No one can be sure which business models will be disrupted, which technologies will win, or how long current competitive advantages will last. This uncertainty hits light asset companies harder, because a substantial portion of their valuation comes from cash flows far in the future.
By contrast, the physical assets, networks, and infrastructure held by heavy asset companies are hard to replicate, and so harder to disrupt. "Capital intensity itself has become a moat against disruption."
Moreover, in areas where demand is strongest—power infrastructure, data centers, semiconductor equipment, defense—supply is highly inelastic, new capacity requires years of investment, approvals, and construction. When demand rises and supply can't keep up, pricing power, returns, and earnings visibility for existing assets all improve.
Risk Warning and DisclaimerThe market has risks, investments need caution. This article does not constitute personal investment advice and does not take into account the individual investment goals, financial situation, or needs of specific users. Users should consider whether any opinions, views, or conclusions in this article are suitable for their circumstances. Invest accordingly, at your own risk.