The Data Center Investment Supercycle: Financial and Governance Implications for 2026

Executive Summary

The global data center market is set to nearly double by 2030, with approximately 100 GW of new capacity and up to $3 trillion in combined investment. Beyond its significance for the real estate and technology sectors, the expansion carries deep implications for corporate finance, financial reporting, capital markets, and governance.

Introduction

According to JLL Research, the global data center sector is expected to expand by 97 GW between 2025 and 2030, effectively doubling current capacity to roughly 200 GW. This growth rate, a 14% compound annual increase, marks the beginning of one of the largest infrastructure investment supercycles of the modern era. For finance executives, this is not simply a real estate phenomenon; it is a capital allocation event that will influence corporate balance sheets, risk frameworks, and accounting standards for years to come.

Financial Context

The scale of capital deployment is comparable to major national infrastructure programs. JLL estimates that 100 GW of new capacity could require $1.2 trillion in real estate asset value creation and approximately $870 billion in new debt financing. In addition, tenants are projected to spend between $1 trillion and $2 trillion on IT infrastructure, including GPUs and networking equipment. Total data center expenditures over the next five years could therefore approach $3 trillion, positioning data centers as a defining investment theme of the late 2020s.

Main Analysis

AI Demand and the Shift to Inference

AI workloads represented approximately one-quarter of global data center demand in 2025, with model training dominating. JLL anticipates that AI could represent half of all workloads by 2030 and expects inference workloads to overtake training as the primary driver by 2027. Inference workloads, which generate ongoing revenue through application usage, require decentralized locations to reduce latency. This will redistribute capacity away from centralized clusters to regional hubs and edge facilities, altering capacity planning and asset-level financial projections.

Energy Constraints and Capital Costs

Power availability has emerged as the critical bottleneck. In major data center markets, grid connection wait times can exceed four years, prompting operators to pursue behind-the-meter generation, battery storage, and 'bring your own power' arrangements. In the United States, natural gas is expected to play a major role, both as bridge power and for permanent on-site generation, although some hyperscalers are resistant due to sustainability concerns. In EMEA and APAC, renewable energy combined with private wire transmission is gaining traction; JLL notes that such projects can lower power costs by 40% relative to the grid. These choices directly affect operating expenses, capital expenditures, and carbon accounting.

Construction Costs and Development Economics

JLL estimates the average global data center construction cost rose from $7.7 million per MW in 2020 to $10.7 million per MW in 2025, a 7% CAGR. The firm forecasts a further 6% increase in 2026 to $11.3 million per MW. These figures cover shell and core only; tenant technology fit-out can add as much as $25 million per MW for AI infrastructure. Rising costs, extended lead times, and limited skilled labor mean that speed to power, rather than construction price, is the primary site-selection criterion. However, as projects scale, construction cost differentials may become more decisive.

The $3 Trillion Infrastructure Investment Supercycle

The projected 100 GW of new capacity includes hyperscale, colocation, and on-premises facilities. With $1.2 trillion in real estate value and $870 billion in debt financing, the capital intensity is rising sharply. JLL compares the current period to a "generational investment supercycle." The development pipeline is consolidating, as higher barriers to entry and technical sophistication favor credible developers with access to capital. For these groups, debt markets are expected to remain open, though lenders are likely to demand enhanced due diligence around power arrangements and tenant commitments.

Business & Market Impact

The data center expansion will affect financial reporting and enterprise finance in several measurable ways.

First, companies constructing or acquiring data centers must assess whether assets qualify as property, plant and equipment or leases under ASC 842 or IFRS 16. Self-built power generation and battery storage add layers of complexity, including asset retirement obligations and the potential for impairment when technology changes or energy markets shift.

Second, long-term power purchase agreements and, in some cases, direct ownership of generation assets could create off-balance-sheet or consolidated obligations. Enterprises will need to apply robust controls to ensure liabilities and commitments are transparently disclosed. The move toward natural gas in the U.S. introduces exposure to carbon pricing and climate-related reporting requirements. Under IFRS S2 and the SEC's climate rules, companies may have to disclose scopes 1, 2, and 3 emissions, making energy sourcing decisions material to financial statements.

Third, tenant fit-out costs, ranging up to $25 million per MW, will be capitalized and depreciated over the hardware's useful life. CFOs and controllers must evaluate whether GPU and server equipment should follow ASU 2020-23, which aligns depreciation to technical obsolescence, and how these charges affect segment profitability and performance metrics. Investors will increasingly focus on cash flow dynamics, such as funds from operations and adjusted EBITDA, where data center entities may exclude meaningful items.

Governance Insights

Boards and audit committees should treat data center investments as a high-risk, high-reward strategic decision. Oversight must cover capital allocation policies, risk tolerance, and consistency with long-term sustainability commitments. Given the speed of AI adoption, the governance process should incorporate cyclical scenario analysis to stress-test assumptions around demand growth, power availability, and technology lifecycles.

Transparency is essential. Management should disclose not only financial metrics but also the assumptions behind energy procurement, construction cost inflation, and tenant creditworthiness. The financial reporting architecture must differentiate between regulatory requirements, industry best practices, and internal estimates. Clear communication with investors about how "speed to power" decisions affect long-term costs supports corporate credibility and reduces uncertainty around asset valuations.

Future Outlook

Over the next three to ten years, the intersection of data centers and finance will evolve considerably. Real-time energy monitoring and AI-driven facilities management may reshape asset utilization and cost allocation models. Accounting standard-setters will likely continue to refine guidance for cloud services, GPU leases, and digital infrastructure. Energy regulations, including new pricing and carbon tax mechanisms, will increase the complexity of operational planning.

The shift from AI training to inference will encourage investments in edge and regional facilities, which could diversify the current concentration in primary markets. As enterprise migration to cloud continues, on-prem data center capacity is expected to decline at around 6% annually in APAC, suggesting a divergence in growth across asset classes and geographies. Financial leaders should embed flexibility in long-term contracts and capital plans to respond to changing demand patterns.

Conclusion

The data center supercycle is not solely a technology or real estate event. It is a broad financial transformation that tests the capacity of companies to allocate capital, manage risk, and report transparently. With up to $3 trillion of investments at stake, CFOs, boards, and regulators must prioritize rigor in financial analysis, accounting treatment, and governance. Those who integrate evidence-based decision-making with disciplined oversight will be better positioned to navigate the coming era of digital infrastructure.

Key Takeaways

• Global data center capacity is projected to nearly double by 2030, with a 14% CAGR and around 100 GW of new supply.
• AI workloads could reach 50% of all data center use by 2030, and inference is set to overtake training by 2027.
• Energy constraints are reshaping capital planning, with behind-the-meter generation and battery storage gaining prominence.
• Construction costs have risen 7% CAGR from 2020 to 2025 and are forecast to increase another 6% in 2026.
• The sector requires approximately $3 trillion in total capital, including tenant fit-out, with $870 billion in debt financing projected.
• Companies must address accounting, disclosure, and governance implications of power procurement, asset retirement, and technology obsolescence.