Enhancing Returns
The Primacy of Asset Selection
November 3, 2026 5 Minute Read Time
Overview
As markets normalized after the pandemic, the defining features of the fourth era of enhanced returns became clearer. Returns increasingly depend on entry discipline, asset selectivity, and a credible path to net operating income (NOI) growth.
The macro backdrop increases the cost of getting asset selection wrong. Sovereign bond yields moved materially higher across major markets in Q3 2026, and in several markets reached multi-decade highs.1 Persistent inflation, energy-price volatility and uncertainty over the path of monetary policy have contributed to higher borrowing costs.1, 2 CBRE IM’s House View also identifies long-term bond yields as central to its outlook for capitalization rates.3 At the same time, modest trend growth and inflation have raised the importance of active leasing and asset-management strategies. Assets need inflation-linked clauses, contractual escalators or sufficient pricing power to recover higher operating costs and protect real cash flows. These conditions shift more of the burden of return generation toward NOI growth and credible leasing catalysts.
With elevated energy costs, persistent supply-chain disruption and ongoing geopolitical conflict, assets with contracted income and pricing power carry a premium. Managers must work harder to identify potential income-growth opportunities, assess operational risks, define strategies to overcome them and assemble teams with the competencies required to execute.
Two structural forces sit at the center of this new real estate investment environment. First, accelerating asset bifurcation has widened performance dispersion within individual sectors and made broad-brush allocations less reliable as a consistent source of outperformance. Second, artificial intelligence (AI) has created a powerful source of potential demand, but not a guaranteed source of investment returns. The opportunity lies in identifying assets whose physical, contractual and operational characteristics can convert AI-driven demand into durable, contracted and growing income, without paying upfront for all of that future upside.
Performance bifurcation
In earlier enhanced-return eras, sector allocation and structural demand tailwinds were reliable sources of return. Both remain viable but weaker in isolation. With interest rates higher for longer and cap rate compression fading, investors need a new edge.
Aggregate market trends now conceal growing return dispersion within the same sector. Assets in the same city and sector can show sharply different fundamentals because the dimensions that drive returns are increasingly specific to their use case. In logistics, those dimensions may include automation readiness, robotics, power availability and supporting digital infrastructure. In affordable housing, they may include energy efficiency, operating costs, resident retention and the quality of community engagement. In offices, they may include location, amenity quality, flexibility and the ability to attract occupiers amid hybrid working and an AI-enabled economy.
The gap between future-ready and obsolete assets is widening, creating significant performance dispersion even within the same sector. CBRE IM’s long-run U.K. analysis, covering 1981-2024, shows how wide that gap can become: at its widest, the annual total-return difference between upper-quartile and median assets averaged about 15 percentage points across sectors. The analysis suggests that dispersion may widen again as occupier demand becomes more selective, reinforcing the need to identify asset-level NOI catalysts rather than rely on sector labels or thematic demand tailwinds.4
Even a best-in-class label does not guarantee durable returns. A modern building in a favored sector, for example, can disappoint if its income-growth catalysts are already exhausted, its purchase price is too high or debt-servicing costs have surged. A secondary asset in a strong location, conversely, can generate leasing demand and rental growth where credible NOI catalysts are identified, and a capable asset-management team executes. Those catalysts may include a change in ownership or operating control, mark-to-market repricing, asset repositioning, redevelopment, improved tenant engagement, refinancing or a more effective operating model. The challenge is whether income can be grown relative to the price paid, the capital expenditure (capex) budget required to deliver and the risks assumed.
The same principle applies in thematic markets. A large structural theme does not create uniform investment opportunity. Market capacity determines whether demand can be converted into returns. In the advanced-manufacturing buildout, for example, available specialized labor, building materials, permitting and digital infrastructure are stronger signals of favorable development outcomes than market narratives are.
Sourcing outperformance in bifurcated markets therefore requires more than asset selectivity alone. It also requires disciplined entry pricing, income visibility and a credible path to NOI growth that justifies capex and financing costs.
The AI and data center underwriting challenge
AI is driving one of the most capital-intensive infrastructure buildouts of the current investment cycle, spanning data centers, semiconductors, power generation, transmission and cooling systems.5, 6 Bain projects that between $5 trillion and $6.5 trillion could be spent by 2030 to build approximately 150 gigawatts of new compute capacity, almost tripling global capacity in five years.5
The scale of the opportunity is clear, but the economic visibility is not. The data-center buildout is advancing before investors have full visibility into where AI will generate durable revenues, productivity gains and enterprise adoption. Bain’s analysis therefore raises a central investment question: whether the AI economy can generate sufficient recurring revenue to support the level of infrastructure capital being deployed.7
Investors must distinguish between capacity supported by economically durable, mission-critical workloads and demand that depends on more uncertain assumptions around model development, AI monetization and future compute requirements. The two may appear similar in headline capacity terms but carry different implications for tenant credit, renewal probability and residual value.
Occupancy alone is therefore an insufficient measure of resilience. The underlying workload, tenant economics, renewal probability and continuing facility relevance matter more than the initial lease term alone. A fully let AI data center may not deliver durable returns if the workload it supports becomes less economic, the tenant’s monetization remains unproven or the facility cannot accommodate the next generation of computing requirements.
Data-center demand also varies by workload. Cloud computing, AI training, inference, sovereign workloads and regulated workloads have different compute intensities, latency requirements, power profiles, cooling needs and location sensitivities. This makes data centers increasingly difficult to underwrite as conventional real estate. The relevant analysis must combine the real estate characteristics of the asset with the infrastructure and technology supporting its use.8
At the real estate level, investors must assess location, rent, contracted income, tenant credit, lease structure, renewal economics and residual demand. At the infrastructure level, they must assess power availability, power cost, grid interconnection, cooling systems, equipment requirements, connectivity, permitting, construction risk, operator capability and technical adaptability. The underwriting challenge is therefore whether the asset controls scarce inputs, serves durable workloads, has credible contractual and tenant economics and is adaptable as technology evolves, leases come up for renewal and an exit is eventually sought.
The underwriting advantage of an integrated platform
These conditions increase the value of integrated investment and operating platforms. Their advantage is not scale alone. It is the ability to combine market knowledge, sector expertise and asset-level judgment when assessing whether a property can produce durable NOI growth.
That judgment must extend beyond the asset’s position within an attractive theme. It must address the physical, contractual and operational factors that determine whether demand can be converted into income. In infrastructure-linked assets, this includes power deliverability, grid interconnection, operating capability and the terms on which capacity can be contracted. Across data centers, logistics and powered land, these factors can determine whether a high-demand location produces resilient cash flow or only an attractive development narrative.9
The same discipline applies as real estate and infrastructure become more closely connected. Alternative and hybrid assets are often more operationally intensive than traditional core property, which makes specialist knowledge and execution more important. The third article turns to how managers convert these underwriting judgments into durable income growth.
2 European Central Bank. Economic Bulletin, Issue 6, 2026. September 2026. https://www.ecb.europa.eu/press/economic-bulletin/html/eb202606.en.html
3 CBRE Investment Management. Macro House View Q3 2026: Resilience Through Persistent Shocks. August 10, 2026. https://www.cbreim.com/insights/articles/macro-house-view-q3-2026
4 CBRE Investment Management. Understanding Allocation Benefit: The U.K., sector or stock? June 3, 2025. https://www.cbreim.com/insights/articles/understanding-allocation-benefit-the-uk-sector-or-stock
5 Bain & Company. AI Data Center Boom: Can We Build It If They Come? September 29, 2026. https://www.bain.com/insights/ai-data-center-boom-can-we-build-it-if-they-come-technology-report-2026/
6 International Energy Agency. Energy and AI. 2025, “Energy Supply for AI.” https://www.iea.org/reports/energy-and-ai/energy-supply-for-ai
7 Bain & Company. New Innovation Is Required to Fund AI’s $6 Trillion Buildout. September 29, 2026. https://www.bain.com/insights/new-innovation-is-required-to-fund-ais-6-trillion-buildout-technology-report-2026/
8 CBRE. North America Data Center Trends H1 2026. August 27, 2026. https://www.cbre.com/insights/books/north-america-data-center-trends-h1-2026
9 CBRE Investment Management. Powered Land. June 29, 2026. https://www.cbreim.com/insights/articles/powered-land
Related Insights
-
How enhanced-return real estate strategies evolved over three decades, from scaling core to sector and thematic plays, and why asset selection now leads.
-
Net lease real estate offers higher yields, resilient performance and diversification across a large and growing opportunity set.
-
Why execution now outweighs allocation in real assets: controlling scarcity, choosing the right route to market and driving NOI growth for durable outperformance.