Executive Summary

The world is building electricity-generation capacity at extraordinary scale.

IRENA reports that 692 GW of renewable power capacity was added globally in 2025, taking total renewable capacity to approximately 5,149 GW. Renewables represented 85.6% of global power-capacity additions during the year. [1]

At the same time, the electricity actually produced by the world's power system tells a more complex story. According to the IEA, renewables supplied around 34% of global electricity generation in 2025, approximately the same share as coal, while natural gas supplied around 21%. Global electricity generation increased by more than 850 TWh during the year. [2]

These figures are not contradictory.

They measure different parts of the power system.

GW measures installed power capacity. TWh measures electricity generated over time. Neither measure, by itself, tells us how much dependable electricity will be available when and where the system needs it.

That distinction becomes increasingly important as electricity demand grows and the composition of generation changes.

Renewable technologies now dominate new capacity additions according to IRENA's dataset, but the operating generation fleet remains diverse. Coal, gas, hydro and nuclear continue to perform material roles, and those roles differ significantly across regions. At the same time, every generation resource—including conventional dispatchable technologies—faces some combination of resource, fuel, technical, climatic or operational constraints.

The central argument of this chapter is therefore not that generation has become secondary.

It is the opposite:

Generation remains fundamental. But installed generation capacity alone cannot determine electricity security.

Electricity security ultimately depends on whether sufficient electricity can be produced and made available at the required location and time. That involves generation, dependable capacity, resource and fuel availability, networks, flexibility, finance, institutions and the ability to deliver infrastructure.

And the relative importance of those factors differs materially across power systems.

01 — The Generation Build-Out

Global electricity demand is expanding, and the generation system is expanding with it.

The IEA estimates that global electricity generation increased by more than 850 TWh in 2025. Renewables accounted for the vast majority of the increase, while the combined growth of renewable and nuclear generation exceeded the total increase in global electricity production because fossil-fired generation declined slightly overall. [2]

Solar PV was particularly important. Using the IEA's Global Energy Review 2026 as the historical reference for this chapter, solar PV generation increased by approximately 600 TWh during 2025. [2]

Capacity expansion was equally significant.

IRENA reports:

• 692 GW of renewable power capacity added during 2025;
• approximately 5,149 GW of renewable capacity installed globally by year-end; and
• renewables representing 85.6% of global power-capacity additions during the year.

These numbers establish something important.

The generation transition is not hypothetical. Large amounts of new infrastructure are already being deployed.

But they also introduce the first analytical problem.

Capacity being built is not the same thing as electricity being generated.

02 — The Capacity Illusion

Installed capacity is one of the most widely quoted measures of electricity-system development.

It is also one of the easiest to misinterpret.

A gigawatt (GW) measures power capacity.

A terawatt-hour (TWh) measures electricity produced over a period of time.

Two power systems can therefore install the same number of gigawatts and receive very different quantities, profiles and system value from those assets.

A generating unit's contribution depends on more than its nameplate rating.

It can depend on:

• resource availability;
• fuel availability;
• technical availability and outages;
• capacity factor;
• time of production;
• location;
• correlation with demand;
• network access;
• flexibility elsewhere in the system; and
• system conditions during periods of stress.

Consider five hypothetical fleets, each rated at 100 GW:

100 GW solar

100 GW wind

100 GW hydro

100 GW gas

100 GW nuclear

They do not necessarily produce the same annual TWh.

They do not produce electricity at the same times.

And they do not necessarily provide the same contribution to resource adequacy.

This is why nameplate capacity cannot be treated as a direct proxy for electricity security.

GW ≠ TWh ≠ dependable capacity

This is not an argument against any generation technology.

It is a measurement discipline.

03 — What Still Generates the World's Electricity?

The technologies dominating new investment are not yet identical to those dominating the operating power system.

According to the IEA's 2025 global generation estimates: [2]

Renewables — approximately 34%

Coal — approximately 34%

Natural gas — approximately 21%

Wind and solar combined — approximately 17%

Low-emissions generation, including renewables and nuclear — approximately 43%.

This reveals an important distinction between two views of the transition.

The new-build system

This tells us which technologies are dominating capacity additions today.

The operating system

This tells us which technologies are actually producing electricity from the accumulated generation fleet.

Infrastructure turnover takes time.

Power plants can operate for decades. Demand continues to grow while new assets are being constructed. Existing plants retire at different rates across countries. Fuel availability, economics, policy and system requirements affect utilisation.

Consequently:

The technology dominating new capacity additions need not immediately dominate total electricity generation.

This distinction helps explain how renewable deployment can be exceptionally rapid while coal and gas simultaneously remain major components of global electricity supply.

04 — Conventional Generation Has Not Disappeared

The growth of renewable capacity should not be interpreted as evidence that conventional generation has already ceased to matter.

The global picture is considerably more heterogeneous.

Coal

Coal remained the world's largest single fuel source of electricity in 2025, supplying approximately 34% of global generation, even though global coal-fired output declined slightly during the year. [2]

Its role varies dramatically by region.

In 2025, coal supplied approximately: [2]

55% of electricity generation in China

71% in India

48% in Southeast Asia.

These figures should not be interpreted as evidence that these systems will remain unchanged.

They demonstrate something narrower and more important:

The starting generation mix differs materially across regions.

That starting point influences the scale, sequencing and infrastructure requirements of future change.

Natural Gas

Natural gas supplied approximately 21% of global electricity generation in 2025. [2]

Its role is not uniform.

In some systems gas provides substantial baseload or mid-merit generation. In others it provides peaking capacity, balancing capability or security during periods when other resources are unavailable.

But dispatchability should not be confused with unconditional firmness.

Gas generation can itself depend on pipeline capacity, LNG availability, fuel prices, storage, competing fuel demand and infrastructure resilience. [9 — SYNTHESIS]

Hydropower

Hydropower can provide large quantities of low-emissions electricity and, where reservoir configurations allow, valuable system flexibility.

But hydro availability can also depend strongly on rainfall, snowpack, reservoir conditions and competing water requirements.

Drought therefore illustrates a broader principle:

Installed capacity does not remove resource risk.

Nuclear

Nuclear power provides another example of why generation discussions require more than technology labels.

At the end of 2024, 63 nuclear reactors representing approximately 71 GW were under construction globally. Around three-quarters of this capacity was located in emerging economies, with approximately half in China. [7]

Policy interest in nuclear has strengthened in many countries, and nuclear generation can provide large quantities of low-carbon electricity with high utilisation.

But construction activity remains geographically concentrated, while financing requirements, long development periods and project-delivery performance remain material constraints in some markets.

It is therefore more defensible to describe strengthening nuclear momentum than to assume a universal nuclear renaissance.

05 — Every Technology Has an Availability Profile

Electricity-system discussions are often reduced to two categories:

Variable renewables

versus

firm or dispatchable generation.

That distinction can be useful operationally, but it becomes misleading when treated as absolute.

Solar output depends principally on irradiance, weather and time of day.

Wind output depends on wind conditions and weather patterns.

Hydropower depends on hydrology and reservoir conditions.

Gas generation depends on technical availability and secure fuel supply.

Coal depends on plant availability, fuel supply and logistics.

Nuclear depends on technical availability, planned refuelling and unplanned outages.

The relevant question is therefore not simply:

Is this technology dispatchable?

It is:

What contribution can this resource reliably make to this particular power system under the conditions that matter?

Resource-adequacy value is therefore system-dependent.

A universal firmness percentage should not be assigned to a technology without defining the power system, demand profile, resource mix, interconnection, storage and adequacy methodology being evaluated.

Editorial principle

Dispatchable does not mean unconstrained. Variable does not mean valueless.

06 — From Generation Capacity to Electricity Security

The previous sections lead to a broader framework.

Installing generation capacity is the first physical step.

But several additional conditions determine how much electricity-security value that capacity ultimately provides.

07 — Eight Regions, Different Generation Problems

Global averages can identify scale.

They cannot describe every regional power system.

China

China is simultaneously expanding renewable generation, thermal capacity, networks and flexibility resources. [2 — SYNTHESIS]

Coal still supplied approximately 55% of Chinese electricity generation in 2025, even as renewable deployment expanded rapidly. [2]

This is not evidence that generation has been “solved.”

Rather, China's scale increasingly creates a combined challenge involving generation, resource geography, transmission, system operation and flexibility.

India

India faces a different problem.

The IEA forecasts average electricity-demand growth of approximately 6.4% per year between 2026 and 2030, while coal supplied approximately 71% of generation in 2025. [3][2]

For India, rapid renewable deployment does not remove the need for substantial generation growth and dependable capacity.

Generation, networks, storage, flexibility and resource adequacy must develop together.

United States

North American reliability assessments point to renewed pressure from load growth.

NERC's 2025 Long-Term Reliability Assessment projects more than 224 GW of additional summer peak demand and more than 245 GW of additional winter peak demand over its ten-year assessment period. [5]

Data centres and other large loads contribute materially to projected demand growth in several areas, but the challenge is not uniform across the continent.

Resource adequacy, generation additions, transmission development, project delivery and the location and timing of new loads interact differently by region.

Europe

Europe demonstrates why large renewable additions and resource adequacy are not mutually exclusive issues. [6 — SYNTHESIS]

ENTSO-E's European Resource Adequacy Assessment 2025 identifies approximately 6–8 GW of capacity at risk of decommissioning during 2028–2030 under its economic viability assessment. [6]

This does not establish that renewable deployment causes inadequacy.

It demonstrates that:

Renewable capacity growth does not remove the need for resource-adequacy planning as the wider generation fleet changes.

Middle East

Many Middle Eastern power systems combine rapid electricity-demand growth, high cooling loads, substantial gas-fired generation and accelerating solar deployment. [7 — SYNTHESIS]

The region therefore should not be reduced to a simple “solar replaces gas” narrative.

Generation diversification, fuel availability, peak demand, networks and flexibility can all matter simultaneously.

Africa

Africa presents perhaps the strongest challenge to any claim that generation itself is no longer central.

In many African power systems, expanding generation, extending transmission and distribution, improving electricity access, mobilising finance and strengthening utilities and institutions remain simultaneous priorities. The IEA estimated in 2025 that around 600 million people in Africa lacked electricity access and identified financing, grid extension and decentralised supply as essential parts of the response. [8]

The continent also contains enormous differences between national power systems.

It is therefore inappropriate to frame Africa primarily as a renewable-integration or grid-bottleneck problem.

In many locations, basic generation and electricity access remain fundamental infrastructure requirements.

Latin America

Latin America's large renewable resource base does not eliminate electricity-security challenges. The IEA documents the region's major reliance on hydropower and the value of greater interconnection; IDB research identifies transmission planning and investment as major integration challenges. [9][10]

Hydropower exposes some systems to hydrological variability, while renewable resources can be geographically distant from demand. The IEA has documented climate-related risks to hydropower availability in Latin America. [11]

Transmission, interconnection, flexibility, project finance and changing weather patterns can therefore interact with generation adequacy.

Southeast Asia

Southeast Asia combines rapid electricity-demand growth with continued dependence on coal and gas, accelerating renewable deployment and increasing concern around fuel security and infrastructure requirements. [3 — SYNTHESIS]

The IEA forecasts electricity demand in Southeast Asia to grow by approximately 5.3% annually over 2026–2030, while coal supplied around 48% of regional generation in 2025. [3][2]

The region therefore faces several challenges at once:

more generation, cleaner generation, secure fuel supply, stronger networks, greater flexibility and sufficient investment.

Interpretation note: This is a qualitative author synthesis, not a numerical regional ranking. Sources [2][3][5][6][8][9][10][11] support selected inputs; they do not independently validate every matrix characterization.

08 — The Counter-Case: What If Generation Really Is the Bottleneck?

The title of this chapter requires an explicit qualification.

In some power systems, generation itself remains a primary constraint.

Countries with inadequate electricity access need additional generation.

Rapidly growing economies need substantially more electricity production.

Systems retiring dispatchable resources may need replacement dependable capacity.

Hydro-dependent systems can face energy shortages during severe drought.

Gas-dependent systems can face fuel-security constraints.

And networks or storage cannot compensate for an absolute shortage of primary electricity generation indefinitely.

Therefore:

“Only the beginning” does not mean generation is solved, secondary or easy. It means generation is necessary—but not sufficient—for electricity security.

This distinction is fundamental to THE GLOBAL GRID RACE.

The purpose of examining networks, flexibility and integration is not to diminish generation.

It is to understand the whole power system.

09 — Looking Toward 2030

Evidence classification: Forward-looking / model-based

The next several years are likely to deepen rather than eliminate the differences between regional power systems.

The IEA's Electricity 2026 forecasts average annual electricity-demand growth during 2026–2030 of approximately: [3]

China — 4.9% [3]

India — 6.4% [3]

Southeast Asia — 5.3% [3]

European Union — 2.3% [3]

United States — close to 2% [3].

The same outlook expects renewable generation to expand strongly, while nuclear generation also grows. [4]

At the same time, the IEA forecasts global gas-fired generation to increase by around 2.6% annually over 2026–2030, while coal-fired generation declines by around 0.9% annually. Coal is nevertheless expected to remain the largest single fuel source of electricity through 2030. [4]

Renewables and nuclear together are expected to provide around half of global electricity generation by 2030. [4]

These are forecasts—not observed outcomes.

They depend on assumptions about demand, economics, fuel markets, policy, project delivery, technology costs and infrastructure development.

The appropriate conclusion is therefore not that one generation technology has “won.”

It is:

The generation system is changing rapidly, but the direction, speed and system consequences of that change differ materially by region.

10 — What We Still Do Not Know

Energy-system outlooks necessarily contain uncertainty.

Several factors could materially change the generation trajectory described in this chapter.

Electricity demand

Electrification could accelerate or slow. Industrial demand could change. Data-centre growth may be concentrated differently from current expectations.

Fuel markets

Gas, coal and LNG availability and pricing can alter dispatch patterns and investment decisions.

Nuclear delivery

Projects under construction do not automatically become operating capacity on their original schedules or budgets.

Hydrology and climate

Changing weather patterns can materially affect hydro, wind, solar output and electricity demand.

Project execution

Permitting, supply-chain constraints, financing costs and construction delays can change when announced capacity actually reaches operation.

Infrastructure integration

Generation projects can be completed before the networks, storage, flexibility or market arrangements required to use their output efficiently.

Policy

Governments can change energy policy, market design, subsidies, permitting rules, emissions requirements and security priorities.

For these reasons:

A 2030 electricity outlook is a structured assessment of possible trajectories—not a predetermined outcome.

11 — What This Chapter Establishes

The evidence supports six conclusions.

1. Generation expansion remains fundamental.

Global electricity demand cannot grow without substantial additional electricity production.

2. Renewables increasingly dominate new capacity additions.

But capacity additions should not be confused with actual generation or dependable capacity.

3. The operating generation system remains diverse.

Coal, gas, hydro and nuclear continue to play material roles.

4. Every generation technology has an availability profile.

No technology's nameplate capacity alone describes its electricity-security contribution.

5. Generation adequacy remains a real issue.

For several regions, additional generation and dependable capacity remain major requirements.

6. There is no universal generation pathway.

Different starting points, resources, demand trajectories, infrastructure and institutions produce different regional challenges.

Together they support the central thesis:

Generation is fundamental. But electricity security is a system outcome.

Conclusion — Generation Is Only the Beginning

The world will require enormous investment in electricity generation.

That conclusion should not be weakened.

Renewable technologies are being deployed at unprecedented scale. Solar and wind are becoming increasingly important sources of electricity. Nuclear interest is strengthening in several markets. Coal remains deeply embedded in major power systems. Gas continues to play important roles. Hydropower remains fundamental in many regions.

But counting gigawatts cannot, by itself, tell us whether a power system is secure.

GW does not tell us TWh.

TWh does not tell us whether electricity is available during periods of system stress.

Dispatchability does not eliminate fuel or technical risk.

Resource abundance does not guarantee network access.

Generation capacity does not guarantee deliverability.

The relevant question is therefore broader:

Can sufficient electricity be produced and made available reliably, at the time and place it is needed?

Answering that question requires us to look beyond generation assets themselves—to resource adequacy, fuel security, networks, flexibility, finance, institutions and infrastructure delivery.

But it also requires discipline.

Networks cannot replace missing primary generation.

Storage cannot eliminate the need to produce sufficient electricity.

And there is no evidence that one infrastructure constraint dominates every power system.

That is why the title matters.

Generation is only the beginning—not because generation matters less, but because the power system matters as a whole.

Next in THE GLOBAL GRID RACE

CHAPTER 03 — THE GRID BOTTLENECK

Generation may exist.

But can it be connected, transported and delivered to demand?

And more importantly:

When does the network become the binding constraint—and when does it not?

That is the question Chapter 03 must test.

Evidence & Editorial Note

This chapter distinguishes between:

Evidence-backed — historical observations directly supported by cited sources.

Evidence-backed / model-based — forecasts, scenarios and resource-adequacy assessments.

Author interpretation — synthesis developed from the evidence.

The analysis deliberately avoids treating installed capacity as equivalent to generation, generation as equivalent to dependable capacity, forecasts as historical outcomes, or regional observations as universal global pathways.

Where authoritative datasets differ materially, the source and dataset used should be explicitly identified rather than silently harmonised.

References

[1] IRENA (2026). Renewable Capacity Statistics 2026, and press release, 1 April 2026. Historical installed capacity and additions for calendar year 2025. https://www.irena.org/Publications/2026/Mar/Renewable-capacity-statistics-2026 ; https://www.irena.org/News/pressreleases/2026/Apr/Near-700-GW-Surge-in-2025-Proves-Renewable-Energy-Resilience

[2] IEA (2026). Global Energy Review 2026 — Electricity Supply. Historical 2025 generation estimates. https://www.iea.org/reports/global-energy-review-2026/electricity-supply

[3] IEA (2026). Electricity 2026 — Demand. Forecasts for 2026–2030, not observed outcomes. https://www.iea.org/reports/electricity-2026/demand

[4] IEA (2026). Electricity 2026 — Supply. Forecasts for 2026–2030, not observed outcomes. https://www.iea.org/reports/electricity-2026/supply

[5] NERC (2025). 2025 Long-Term Reliability Assessment. North American assessment areas, 10-year peak-demand forecasts; not a US-only forecast. https://www.nerc.com/globalassets/our-work/assessments/nerc_ltra_2025.pdf

[6] ENTSO-E (2025). European Resource Adequacy Assessment 2025. 6–8 GW is a modelled decommissioning risk in the 2028–2030 horizon, not announced closures. https://www.entsoe.eu/eraa/2025/

[7] IEA (2025). The Path to a New Era for Nuclear Energy — Status of Nuclear Energy. Confirms 63 reactors (71 GW) under construction as of end-2024; this is a dated historical snapshot. https://www.iea.org/reports/the-path-to-a-new-era-for-nuclear-energy/status-of-nuclear-energy

[8] IEA (2025). Financing Electricity Access in Africa. Supports ~600 million without electricity access and the role of grid expansion, decentralised solutions and financing. https://www.iea.org/reports/financing-electricity-access-in-africa/executive-summary

[9] IEA (2023). Latin America Energy Outlook 2023 — Executive Summary. Regional hydropower reliance, interconnection and flexibility. https://www.iea.org/reports/latin-america-energy-outlook-2023/executive-summary

[10] IDB (2025). Unlocking the Grid: How to Ensure Reliable and Sustainable Energy in Latin America and the Caribbean. Transmission, planning, regulation and financing. https://publications.iadb.org/en/unlocking-grid-how-ensure-reliable-and-sustainable-energy-latin-americaand-caribbean

[11] IEA (2021). Climate Impacts on Latin American Hydropower. Climate and hydrological availability risks; historical report, not a 2025 generation statistic. https://www.iea.org/reports/climate-impacts-on-latin-american-hydropower