Published on: 2026-08-04
Updated on: 2026-08-04
The HALO trade favours companies built around heavy, hard-to-replace assets, from power grids and factories to transport and distribution networks. An irreplaceable asset can still be a poor investment when debt, maintenance and expansion costs consume the cash it generates.
LOHA has now turned the idea into a formal strategy. US data centres could consume 11.8% of national electricity by 2030, raising the stakes for the physical systems that support AI.
Goldman Sachs’ European capital-intensive versus capital-light trade gained about 20% by July 7, 2026, reducing the scope for another advance driven mainly by valuation.
Expected 12-month earnings per share rose 45% in 2026 for Goldman Sachs’ emerging-market capital-intensive basket and remained flat for its capital-light counterpart.
More than 2,500 gigawatts of generation, storage and large-load projects are waiting in grid queues, exposing the speed gap between AI expansion and physical infrastructure.
Consumer discretionary and staples represented 31.44% of LOHA, while utilities accounted for only 1.02%, showing that physical distribution and franchise networks can qualify alongside grids and factories.
The HALO thesis weakens when capital expenditure and debt rise faster than operating cash flow.
Heavy Assets, Low Obsolescence measures economic durability, not company age, sector or valuation. A business qualifies when its physical assets or operating network are difficult to reproduce, remain useful across technology cycles and earn more than they cost to maintain.
AI cannot transport freight, collect waste or recreate an established distribution network. Railways, waste services and consumer supply chains resist technological substitution. Grid equipment, cooling systems and semiconductor machinery qualify for a different reason. Some HALO companies supply the infrastructure AI cannot expand without.
US data centres could consume 11.8% of national electricity by 2030. The resulting demand extends to transformers, transmission lines, cooling systems and generation capacity.
The supply response moves far more slowly. More than 2,500 gigawatts of generation, storage and large-load projects are stalled in grid queues worldwide. A data centre can be completed in one to three years, while major grid infrastructure can require five to 15 years.
Those delays increase the strategic value of cables, switchgear, transformers and permitted power capacity. Strong orders create little value when labour shortages, financing costs and project delays absorb the resulting cash.
Prysmian’s $3.8 billion agreement to acquire Atkore shows how that scarcity is reshaping the electrical supply chain. Atkore adds conduit, cable-management products and roughly 30 manufacturing and distribution centres serving data centres, utilities and industrial construction.
Power demand is visible. Profitable execution is harder to see.
Goldman Sachs’ European pair trade, expressed as long capital-intensive companies and short capital-light companies, was up about 20% year to date by July 7, 2026. A separate emerging-market basket gained 115% from late 2025 to June 5, compared with 7% for its capital-light counterpart. The emerging-market capital-intensive group still traded at a 20% discount, but gains of that scale leave less room for another broad advance driven by valuation alone.
Earnings have begun to support the gap. Expected 12-month earnings per share rose 45% during 2026 for the emerging-market capital-intensive basket and remained flat for the capital-light group. Revenue growth, margins and free cash flow must now justify the prices already paid.
A physical moat creates value only when the asset remains necessary long enough to recover its costs, higher expenses can be passed through, orders become cash and new investment earns more than its funding cost.
Low obsolescence protects demand, not margins or balance sheets. The thesis weakens when debt and capital spending rise faster than operating cash flow, utilisation falls or new capacity arrives after demand peaks. Stranded assets remain heavy assets.
LOHA applies the same test to companies as different as Owens Corning, O’Reilly Automotive and Domino’s Pizza.
The Roundhill HALO ETF began trading on May 14, 2026. It tracks a 100-stock, equally weighted US index that screens for physical asset intensity, exposure to AI displacement, valuation and business quality.
Industrials represented 39.23% of LOHA on June 30, 2026. Consumer discretionary and consumer staples together accounted for 31.44%, while utilities represented only 1.02%. HALO therefore extends beyond power networks, pipelines and heavy engineering.
The portfolio includes four distinct forms of physical advantage.
| HALO example | Why it qualifies | Earnings test |
|---|---|---|
| Owens Corning | Manufacturing scale and building-material capacity | Pricing offsets input and construction costs |
| O’Reilly Automotive | Dense stores and distribution centres | Inventory and expansion produce free cash flow |
| Yum Brands | Franchised restaurants and established supply chains | Unit growth lifts earnings without excessive spending |
| Domino’s Pizza | Delivery infrastructure and franchise density | Sales growth improves margins and cash returns |
All four appeared among LOHA’s 10 largest holdings on June 30, 2026.
The common thread is dependence on physical networks that software cannot reproduce quickly. Index inclusion identifies HALO characteristics, not disciplined spending, pricing power or an attractive valuation.
Yes. Foundries, equipment makers and specialist materials suppliers can qualify when scarce factories, complex engineering and long customer approval cycles protect the business. Asset-light chip designers fit less naturally.
No. LOHA offers diversified exposure through a defined index. Individual infrastructure, industrial, transport and essential-service companies offer more targeted exposure but carry greater company-specific risk.
No. Value investing focuses on whether a stock is cheap relative to earnings, assets or cash flow. HALO focuses on whether a company owns hard-to-replace physical assets or networks that can remain useful through technological change. A HALO stock can still be expensive, and a cheap capital-intensive company may own obsolete assets.
Lower rates could make growth stocks more competitive while reducing financing costs for utilities and infrastructure. HALO would weaken only if earnings failed to support current valuations.
Josh Brown introduced Heavy Assets, Low Obsolescence on February 8, 2026. Goldman Sachs later expanded it into a broader capital-intensive equity framework.
The Q3 2026 reporting cycle will show whether infrastructure demand is lifting margins and free cash flow across industrial, utility and essential-service companies. Prysmian’s October 29 results offer one concrete checkpoint through revenue growth, project margins and cash generation.
In HALO’s next phase, backlog without cash flow is only expensive optimism.
