The Next Decade of Global Projects

What will the world be building over the next ten years, and does it have the capacity to build it? The evidence points to a pipeline that is large, broad and increasingly interlocked: power generation and grids, data centres, mines and processing plants, factories, transport corridors and the social infrastructure of fast-growing cities. The International Energy Agency (IEA) expects global energy investment alone to reach US$3.4 trillion in 2026 (IEA, World Energy Investment 2026). McKinsey estimates that about US$106 trillion of cumulative investment will be needed in infrastructure through 2040 (McKinsey & Company, The infrastructure moment, 2025).

The second half of the question is the more consequential one. Our reading of the evidence is that, for much of the coming decade, the binding constraint on global projects is less likely to be the availability of capital than the capability to turn capital into working assets: long-lead equipment, grid connections, skilled people, capable contractors, permits and community consent, and leaders able to decide, sequence and stop. The Project Management Institute (PMI) estimates that organisations could need up to 65 million project professionals by 2035, against almost 40 million in 2025 (PMI, Global Project Management Talent Gap, 2025).

That shifts the centre of gravity of project leadership. The question for boards and governments is no longer only which projects to fund, but which ones they can realistically deliver, in what order and with whose capacity.

In brief

• The global project pipeline to the mid-2030s spans energy, grids, digital infrastructure, critical minerals, industry and transport. The IEA expects energy investment of US$3.4 trillion in 2026, nearly 60% of it electricity-related.
• Digital infrastructure is moving on a faster clock than the rest of the pipeline. The IEA’s April 2026 update projects data-centre electricity use roughly doubling to about 950 TWh by 2030, and its estimates imply about US$3.9 trillion of data-centre investment between 2026 and 2030.
• Long-range infrastructure estimates vary with method and date, from the G20 Global Infrastructure Hub’s US$94 trillion to 2040 (2017) to McKinsey’s US$106 trillion (2025). They are estimates of need, not committed projects.
• Execution capacity is the pressure point: turbine and grid-equipment supply is tight, investment in critical minerals fell in 2025, and PMI’s high-growth scenario shows a gap of up to 29.8 million project professionals by 2035.
• Our reading: organisations that treat deliverability as a capital-allocation criterion, secure scarce inputs early and manage interdependent projects as a portfolio are likely to convert more of the pipeline into working assets.

Reading the pipeline: what the numbers are, and what they are not

Pipeline figures deserve careful reading because they mix three different things: money already being spent, estimates of what will be spent this year, and estimates of what would need to be spent to meet a goal. Each is useful. None is a forecast of what will actually be built.

The firmest numbers are for energy, because the IEA tracks spending every year. Its 2026 edition expects capital flows to the energy sector to grow by 5% to US$3.4 trillion in 2026, with around US$2.2 trillion going to renewables, nuclear, grids, storage, low-emissions fuels, efficiency and electrification, and some US$1.2 trillion to oil, natural gas and coal. Electricity-related spending makes up nearly 60% of the total; investment in electricity supply and infrastructure is expected to reach US$1.6 trillion, or US$2 trillion once end-use electrification is included (IEA, World Energy Investment 2026).

The same report makes a point that matters for anyone planning the next decade: much of this year’s energy investment was effectively committed by decisions taken well before the current conflict in the Middle East began. Projects carry earlier decisions forward. The pipeline of the early 2030s is being shaped by approvals being taken now, under today’s assumptions about prices, policy and security.

New build is visible across technologies. The IEA counts 78 GW of nuclear capacity under construction in 15 countries, with annual nuclear investment above US$80 billion. Orders for new gas-fired power plants reached 130 GW in 2025, a 25-year high, and more than 100 billion cubic metres of new LNG export capacity was approved in 2025, a record. Spending on electricity networks is expected to reach about US$550 billion in 2026, up nearly 20% on the previous year (IEA, World Energy Investment 2026).

Beyond energy, the long-range numbers are estimates of need. In 2017 the G20’s Global Infrastructure Hub, working with Oxford Economics, forecast global infrastructure investment needs of US$94 trillion to 2040 across 50 countries and seven sectors, plus US$3.5 trillion to meet the Sustainable Development Goals for drinking water and electricity. It estimated that almost 19% of the combined US$97 trillion would be unfunded if spending trends continued, and that closing the gap would require annual investment to rise from 3% to 3.5% of global GDP (Global Infrastructure Hub, Global Infrastructure Outlook, 2017). McKinsey’s more recent estimate spans seven verticals, including digital, social infrastructure, agriculture and defence, and puts cumulative needs at about US$106 trillion through 2040: US$36 trillion for transport and logistics, US$23 trillion for energy and power, US$19 trillion for digital and US$16 trillion for social infrastructure, with Asia accounting for about US$70 trillion (McKinsey & Company, The infrastructure moment, 2025).

Two conclusions follow. First, the definition of infrastructure is widening: fibre networks, data centres and charging networks now sit beside roads, ports and grids. Second, these estimates describe a scale of ambition. Whether ambition becomes assets depends on funding decisions and delivery capacity that headline numbers do not capture.

Digital infrastructure: a faster clock in the pipeline

Data centres have become a distinct class of capital project, and their pace is setting the tempo for adjacent sectors. The IEA reports that the capital expenditure of the largest technology companies exceeded US$400 billion in 2025 and is expected to rise by a further 75% in 2026. The capital spending of just five technology companies is now larger than global investment in oil and gas production (IEA, Key Questions on Energy and AI, 2026).

The IEA’s updated central projection sees data-centre electricity consumption roughly doubling from 485 TWh in 2025 to 950 TWh in 2030, around 3% of global electricity demand, with consumption by AI-focused data centres tripling over the period. Its estimates imply cumulative investment in data centres of US$3.9 trillion between 2026 and 2030, which the agency notes is too large to be funded from company balance sheets alone and will require debt and equity markets. The pace of the build-out will therefore be sensitive to investor expectations of returns on AI (IEA, Key Questions on Energy and AI, 2026). The energy spending attached to data centres is already substantial: the IEA estimates that energy-sector investment for the build-out of data-centre infrastructure exceeded US$100 billion in 2025, more than the total invested in Africa’s energy sector that year (IEA, World Energy Investment 2026).

What makes digital infrastructure a project-economy issue, rather than a technology-sector story, is a mismatch of clocks. The IEA’s 2025 analysis found that new transmission lines can take four to eight years to build in advanced economies, and estimated that around 20% of planned data-centre projects could be at risk of delay unless grid risks are addressed (IEA, Energy and AI, 2025). Its 2026 update adds that planning and regulatory systems are being stretched by the wave of data-centre applications, and that some communities are pushing back against projects. A server hall can be specified and financed quickly; the power, permits and public consent it depends on cannot.

For executives, the implication is that technology investment decisions now carry capital-project risk. Choices about AI capacity, cloud strategy and digital transformation depend on physical assets whose delivery sits outside the technology function, which is why transformation and technology investment increasingly belongs on the capital-projects agenda.

Resources and industry: the pipeline behind the pipeline

Every grid, battery plant and data centre draws on a second pipeline of mines, refineries and factories, and that pipeline is behaving differently. The IEA reports that capital spending on critical minerals by major mining companies fell about 9% in 2025, the first substantial decline since 2020. Companies focused on battery metals cut spending by more than 20% and lithium specialists by about 40%, while copper-focused companies increased investment by 8%. Exploration spending for critical minerals fell 10% (IEA, World Energy Investment 2026). The IEA also notes that financing constraints continue to hinder supply diversification, and that capital costs for mining and refining projects are higher in emerging regions than for incumbent producers.

Manufacturing capacity for energy equipment is concentrated. China accounted for around 75% of total clean-energy manufacturing investment in 2025, including 80% of production capacity in the lithium-ion battery supply chain and 95% for solar PV wafers (IEA, World Energy Investment 2026).

Our reading is that industrial and resource projects have become strategic infrastructure in their own right. A downstream programme that assumes an uninterrupted flow of copper, battery materials or components is making an implicit bet on upstream projects it does not control, whose investment cycles respond to commodity prices rather than to the downstream schedule.

From capital to capability: where the constraint now sits

If the pipeline is large, the practical question is whether the delivery system can absorb it. The evidence points to four pressure points.

Equipment and factory slots

Long-lead equipment makes the constraint visible. The IEA reports a 70% surge in gas turbine orders in 2025 (IEA, Key Questions on Energy and AI, 2026), and notes that strong demand from the United States and the Middle East is limiting the availability of turbines for near-term deployment elsewhere. Part of the recent rise in grid investment reflects higher prices for key components such as transformers and cables in tight supply chains, not only more physical build (IEA, World Energy Investment 2026). Equipment is now allocated globally: a project in one region competes for factory slots with projects in another. The consequences for power systems are examined in our analysis of energy infrastructure projects.

People

The workforce constraint is broader. PMI estimates that there were almost 40 million project professionals in 2025 and that organisations could need up to 65.4 million by 2035 in its high-growth scenario, a shortfall of up to 29.8 million. Even in its low-growth scenario, demand rises 48% to 58.5 million, and a further 4.4 million professionals are expected to leave the workforce through attrition and retirement (PMI, Global Project Management Talent Gap, 2025). Because PMI projects demand from gross capital formation, the gap can be read, in our view, as a measure of investment at risk of being under-managed.

Craft and engineering skills are under similar strain. McKinsey’s report notes that more than half of US construction firms report project delays due to worker shortages, and that the renewables sector alone must add about 2.8 million jobs globally by 2030 (McKinsey & Company, The infrastructure moment, 2025). The IEA observes that electricity system operators often lack sufficient resources and that the industry faces a shortage of skilled labour to deliver grid connections (IEA, Energy and AI, 2025).

Absorptive capacity

Scaling up spending does not automatically scale up delivery. IMF analysis found that the cost of an individual project can increase by 10% when public investment in the country is high (IMF, Fiscal Monitor, October 2020). When many programmes draw on the same contractors, engineers and supply chains at once, prices rise and supervision thins. Every national or corporate pipeline has a delivery speed limit, and pushing past it costs money.

The price and location of capital

Capital is not evenly available either. The IEA’s exploratory scenarios suggest that emerging market and developing economies outside China, which accounted for 10% of the growth in energy investment over the past decade, would account for more than 50% over the next ten years. Yet their financing costs are already at least double those of advanced economies and China (IEA, World Energy Investment 2026). The same report warns that if borrowing costs stay higher for longer, capital-intensive projects will be disproportionately affected. The geography of need and the geography of cheap capital do not coincide.

Interdependence turns projects into portfolios

The defining feature of the coming pipeline is not only its size but its coupling. A data centre needs grid capacity; the grid needs transformers, cables and permits; the cables need copper; the copper needs mines that take years to develop. McKinsey notes that lagging development among the assets of one infrastructure vertical can create bottlenecks across the system, citing insufficient electricity supply as a constraint on data-centre construction (McKinsey & Company, The infrastructure moment, 2025).

Coupling changes the management problem. When projects depend on each other, a delay in one propagates to many, and the unit of control moves from the individual project to the portfolio. Sequencing, interface management and the willingness to pause or defer become as important as selection. This is the ground covered by portfolio leadership beyond project selection and by the wider discipline of project portfolio management.

What the decade asks of project leadership

For senior executives, five disciplines follow from the evidence.

1. Treat deliverability as a capital-allocation criterion. Before approval, test whether the equipment, people, grid access and permits exist to deliver on the proposed schedule. Many failures are set in motion before construction begins, as our analysis of front-end definition argues.
2. Secure scarce inputs as strategic positions. Reservation agreements for turbines and transformers, framework contracts with capable contractors and early grid-connection applications are now part of project strategy, not procurement administration.
3. Build owner capability, not only contractor capacity. Sponsors who cannot specify, integrate and challenge will pay for it in claims and delay. The talent gap PMI describes falls on owners as well as on suppliers.
4. Plan against scenarios, not single forecasts. The IEA updated its data-centre outlook a year after first publishing it and says it will keep revising it. The value of an asset depends on which scenario arrives.
5. Measure delivery confidence continuously. Portfolios need integrated cost, schedule and risk information to see where one project’s delay is about to become another’s. That is the territory of project controls and of measuring capital project performance as cost, time and value together.

How the pipeline redraws responsibilities

The pipeline also redraws responsibilities across functions. Portfolio leaders must balance a larger number of interdependent commitments. Leaders of capital projects and CAPEX programmes face longer lead times and higher equipment risk. Finance and governance leaders must stage funding against delivery confidence rather than headline need, as discussed in our analysis of project funding and capital allocation. Technology leaders must plan AI capacity with the physical assets it requires in view.

Research and innovation belong in the same conversation. The IEA notes that China now accounts for over one-third of global public investment in energy R&D (IEA, World Energy Investment 2026). The technologies that reach the project pipeline in the 2030s are being selected in research portfolios today.

Decide, fund, deliver, prove: the decade in four stages

The full life of a project in the project economy, Decide → Fund → Deliver → Prove, offers a practical way to read the decade ahead.

• Decide. Choose projects against realistic delivery capacity and a range of scenarios, not against headline need alone.
• Fund. Match capital structures to risk, recognising that the cost of capital differs sharply between markets.
• Deliver. Secure equipment, people and interfaces early, and manage interdependent projects as one system.
• Prove. Judge the pipeline by the assets that operate and the services they provide, which is the subject of our companion analysis of the real-world impact of global projects.

Project Economy Forum’s Dubai edition, on 27–28 January 2027, is organised around the same sequence: its second day follows a single project on one stage, from first decision to final proof.

Conclusion: the number no outlook publishes

The figures that dominate discussion of the next decade are, in the end, claims on the future: trillions of dollars of intended investment, gigawatts of planned capacity, millions of needed professionals. The delivery system decides which of those claims are honoured. The organisations and countries that gain most over the coming decade are unlikely to be those with the largest announced pipelines. They will be those with the highest rate of conversion from decision to working asset. That rate is a leadership variable: it can be measured, managed and improved, and it may prove to be the most important number that no outlook publishes.

Sources

• International Energy Agency, World Energy Investment 2026 (2026). IEA
• International Energy Agency, Key Questions on Energy and AI (2026). IEA
• International Energy Agency, Energy and AI (2025). IEA
• McKinsey & Company, The infrastructure moment (2025). McKinsey
• Global Infrastructure Hub, Global infrastructure investment need to reach USD97 trillion by 2040, release on the Global Infrastructure Outlook (2017). Global Infrastructure Hub
• Project Management Institute, Global Project Management Talent Gap (2025). PMI
• International Monetary Fund, Fiscal Monitor: Policies for the Recovery, executive summary (October 2020). IMF