Rising global electricity demand pushes annual grid expenditure to $970bn by 2050
Agency report
Global electricity demand is set to double by 2050 as the world’s reliance on fossil fuels diminishes, according to DNV’s New Power Systems report.
This is also pushing global grid capacity to grow 2.5 times its current size, with annual expenditure on grids more than doubling to $970 billion by 2050.
The report finds that the pathway to a decarbonised energy system requires significant grid expansion, solutions for grid congestion, and new business models to accommodate rising electricity demand and generation from wind and solar.
The report also concludes that grid expansion is affordable, and this is due to growing efficiencies in grid technology and the increased electricity load, with DNV expecting global grid charges passed to consumers to remain stable or decline in the long term.
Global economic growth and the electrification of transportation, heating, and industry are primary drivers of the anticipated surge in global electricity demand.
Additionally, the development of data centers, notably for AI applications, is also expected to contribute to this increase while at the same time, AI holds considerable potential to introduce efficiencies in both the provision and use of power.
By mid-century, electricity will constitute 37 per cent of global final energy use, a substantial rise from 20 per cent in 2023.
This upswing is accompanied by a dramatic shift towards renewable energy, with wind and solar anticipated to generate half of the world’s electricity by 2040 and 70 per cent by 2050.
The decarbonisation trajectory predicts that nearly 90 per cent of electricity will be sourced from non-fossil sources by 2050.
The growing share of renewable power requires flexibility and a robust system of demand-response: as variable renewable energy sources (VRES) expand ninefold, the need for short-term flexibility will double.
Fluctuating demand from sectors like heat and transport calls for new ancillary services such as synthetic inertia products and fast frequency response and adapting market and regulatory frameworks to support these technologies is critical for a sustainable energy future.
In this new model, advanced technologies such as AI and automated activation of demand response will assist grid operations and market predictions: “Deep digitalisation, including the application of AI, is crucial for managing the increased complexity of a renewable-dominated power system,” explains Remi Eriksen, Group President and CEO at DNV. “So there will be a rapid growth in both information technology and operational technology in the coming years which must be supported by a robust approach to risk management, particularly on cyber security, to reap the benefits of the new power systems.”
Energy storage is another core system component needed to address the intermittent nature of renewable energy sources, provide grid balancing, and manage load. Lithium-ion battery technology is set to play a dominant role in this segment, offering three times more storage capacity than hydropower and pumped storage by 2050.
However, achieving the necessary level of flexibility will require innovative market designs and advanced tariff schemes to incentivise automated demand-response, vehicle-to-grid (V2G) and behind-the-meter storage systems. Power-to-hydrogen value chains are a critical market element for renewable generation, and need to be scaled through concerted investment efforts by all stakeholders.
A key bottleneck in the energy transition is the expansion of the electrical grid. Global grid capacity needs to grow 2.5 times its current size, with annual expenditure on grids more than doubling to USD 970 billion by 2050.
While Grid Enhancing Technologies (GETs) offer potentially significant temporary relief, the long-term solution lies in accelerating the construction of new grid infrastructure and advanced controlling systems, which are currently hampered by permitting timelines, the scarcity of human and material resources, and financial constraints
