Adding capacity to the electricity grid is not a simple task

An inability to provide enough grid connections means that all around the world new generating projects face lengthening queues before they can get onto the grid (Photo: Reuters)
An inability to provide enough grid connections means that all around the world new generating projects face lengthening queues before they can get onto the grid (Photo: Reuters)


  • It is even more difficult if the power comes from renewable sources

Chill Sun Solar, a facility in Nevada with a generating capacity of 2.25GW, should produce enough electricity to meet 1.7% of California’s annual demand. The 4.1GW Berwick Bank wind farm, off the east coast of Scotland, will provide more energy over a year than could Britain’s two largest gas-fired power stations running full tilt. But none of this will matter until these renewable heavyweights get hooked up to their respective grids.

The grids used by developed countries are not accustomed to rapid change. At the turn of the century a couple of power plants a year might be connected to meet new demand driven by demographic change, to replace plants at the end of their lives or, as in shifts between coal and gas, to compete on price. But the overall rate was typically slow, with net capacity changing little and new plants often using the same connections as old ones.

Unsurprisingly, the business of supplying the highly specialised components big grids require was paced to match. Take the transformers needed to step electricity up to, and down from, the highest voltages on the grid—massive machines made with bespoke techniques. Gleaming fettuccine strands of copper bundled into sapling-thick, paper-sheathed cable are wound tightly around bus-sized wooden frames by patient hands. Such windings, perfect to the millimetre and precisely tailored to the buyer’s requirements, are then brought together in huge structures of wood and glue—any metal would disrupt the electromagnetic fields around the copper. The whole assembly is then submerged in a tank filled with mineral oil which rises all the way to its brim.

To see one of these titans being made at the Siemens Energy factory in Nuremberg, Germany is to witness a remarkable mixture of scale, strangeness and craft skills (Siemens’ attempts to do the winding by robot have so far failed). The finished products typically weigh 300 tonnes, cost €30m ($32m), and are ready for delivery about three years after being ordered.

Expanding the world’s capacity to make such exotica is vital if grids are to expand quickly. But it is not easily done. There are few providers and they are cautious. Siemens is working on making its existing plants more productive, as are its competitors, but it is reluctant to build lots of new capacity for what it sees as a one-off spike in demand. So are its competitors.

The expectation that lots of new equipment will never be needed very quickly is one the legacies of steady-state grids. Another, linked legacy is a low appetite for investment.

For most of the past half century the governments which owned many national grids and the bodies which regulated grids in the private sector focused on two main goals. The first was that grids be stable and secure; that the lights be kept on and industry humming. The second was that grids should add as little as possible to the price of the electricity they supplied.

For the most part, the grids met these goals. Consumers in most of the rich world can rely on the fact that, except under the most adverse of circumstances, the flick of a switch will illuminate their homes and the food in their freezers will not thaw. This certainty has been provided quite cheaply. Ben Wilson of National Grid, which owns and runs Britain’s transmission infrastructure, says that in 2019 the cost of the grid accounted for just £20 ($25) of the average consumer’s electricity bill of £1,300, a mere 1.5%.

But conservative regulation offered no incentive, or scope, for strategic investment. Grids could spend only what was needed to provide connections for projects which were guaranteed to get built or to alleviate chronic congestion. Connection requests were typically dealt with on a first-come-first-served basis, to make things fair, rather than by some other mechanism such as auction or a strategic plan.

That is not good enough for a world which wants to replace the vast majority of the plants currently generating its electricity, and which will often need to build the replacements in different places. An inability to provide enough grid connections means that all around the world new generating projects face lengthening queues before they can get onto the grid. Those queues do not just slow the rate at which new capacity becomes available. They also add to uncertainty and thus increase the cost of investment. Zoisa North-Bond, chief executive of Octopus Renewables, a British energy firm, has called grid access the “biggest limiting factor" to increasing the market share of renewables.

In 2022 about 54% of Britain’s electricity was generated without emitting carbon; its grid is one of the most deeply decarbonised in the world. But that does not mean it has solved the queue problem. On average, the projects connected to the grid in 2022 were doing so about four years after the date they had requested. The company developing Berwick Bank, SSE, should have all the wind farm’s capacity installed by 2030—but at least one part of the farm is due to connect to the grid only at the end of the following year, according to published data. A supplier asking for a grid connection today can expect to be offered one for some time between 2030 and 2038.

There is a real limit to how fast connections can be provided. Expanding the grid changes its properties and thus risks its stability. But today’s delays are much longer than necessary. Mr Wilson of National Grid says the queue has grown to its inordinate length because, until very recently, Britain’s grid regulator, Ofgem, allowed no investments in upgrades before a connection request had been signed on the dotted line. It might be obvious that a country betting big on offshore wind as a new source of power would need strong grid connections to the North Sea and the Atlantic. But they could be invested in only once National Grid had specific and detailed requests from individual developers. As well as slowing investment, this sort of bottleneck also increases permitting time. Only after the generator’s planning and permitting is completed can the equivalent process for the grid connection start. In engineering terms, what could be done in parallel is being done in series.

But perhaps the biggest cause of queue length is speculation. In 2008, in an effort to incentivise the connection of small-scale renewables, National Grid made applying for a connection much easier. That made it possible for speculators to file “paper projects" in the hope of flipping their spot in the queue to a project behind them. Such projects may make up as much as 80% of the queue; many have no designs prepared nor engineers on staff that might carry them out. Weeding them out would seem an obvious step. But the regulatory regime does not give National Grid the tools with which to discriminate between builders and flippers.

Much the same is true in many other countries. Thomas Egebo, the boss of Energinet, the state-owned firm which owns and operates the Danish grid, says the state of affairs is “exactly like in the UK. There needs to be a project and we need to sign things and then we can apply to build [new grid capacity]."

The situation in America, which has many grids and even more lawyers, is dire. A study published last year by Lawrence Berkeley National Laboratory in California found that the average grid connection in America in 2021 had taken three years to be completed, and that, as elsewhere, the lag was getting longer. By some estimates there is at least 1,400gw of new renewable capacity in America’s queues for connection.

Here comes the sun, traffic permitting

Chill Sun Solar, which shares the queue to connect to the Californian grid with about 220GW of other solar plants and energy storage facilities, may be in luck. A new transmission line called Greenlink Nevada, designed to serve solar generation in that state, is currently being planned by the Bureau of Land Management and is due to be in service by 2026. But like the rest of America’s planned energy infrastructure, it will be at the mercy of a permitting process which is vulnerable to court cases and delays.

As the Greenlink plan shows, there is some progress, and regulators are beginning to understand that there needs to be more. In December 2022 Ofgem published new rules, drawn up over eight frenetic months of consultation, allowing National Grid to go ahead and expand in advance of grid-connection requests. It is also reforming its procedures so that projects in the queue can be required to hit milestones or lose their place, though it is leery of making the rule retroactive (which would be needed to shorten the queue quickly).

In America the Federal Energy Regulatory Commission has expressed concern that grid operators “may not be planning transmission on a sufficiently long-term, forward-looking basis to meet transmission needs driven by changes in the resource mix and demand", and last year gathered feedback on proposals designed to push grid operators to invest according to longer term plans as a way to end today’s “piecemeal and inefficient development". Its new proposals are expected this summer.

There are also plans afoot to reduce the delays planning permission and building permits can cause. In November the European Commission, the executive of the European Union, agreed to a new set of temporary regulations giving renewable-energy projects a “presumption of overriding public interest". The commission also agreed that “environmental-impact assessment for grid reinforcements should be limited" in order to “facilitate the integration of renewable energy".

Red queens racing

Such moves are vital when it comes to building new long-range connections, which may be required to relieve congestion as the grid expands, or to connect generators and loads across large distances. The construction of Ultranet, a series of high-voltage cables that will link wind power from the North Sea to industrial users in southern Germany, requires its developers to obtain some 13,500 building permits, says Tim Holt, who sits on Siemens Energy’s board. “The tech is ready, but the permits are not." In February massive transformers destined for Ultranet were sitting ready for shipping at the company’s factory in Nuremberg.

In the developing world, where grids are typically too small to meet today’s demand, let alone the future’s, some of these problems look less fraught. Expanding a growing grid further will often be easier than booting a mature grid out of its slumber.

India, for instance, has built out its grid at a furious pace in the past decade. Its Ministry of Power reports the construction of 161,000km (100,000 miles) of transmission lines since 2014, when Narendra Modi became prime minister. The percentage of the population which has access to a grid connection has shot up from 76% in 2010 to 99% in 2020, according to World Bank data.

The existence of connections does not in itself mean a stable supply. Keeping supply and demand balanced often requires load to be shed, leading to blackouts across India’s cities. But according to the Ministry of Power the amount of kerosene used to keep generators running in rural parts of India fell from 9bn litres to 2bn litres between 2015 and 2021. That suggests that people were able to rely on the grid to provide much more of the energy they needed.

As the demands of grid decarbonisation have become apparent, India has simply kept building. In December 2022 the ministry announced that it would spend 2,440bn rupees ($30bn) to build another 50,000km of electricity transmission lines by 2030 to connect renewable generation, often in fairly remote places, to growing demand. It is planning to connect 500GW of renewable capacity of all kinds by the same date.

India’s huge grid expansion might seem to run counter to a popular narrative about developing-country electrification. It has often been suggested that building big, centralised grid infrastructure was too slow, expensive and wasteful in these markets. One alternative might be a more rapid growth in distributed generation, specifically solar panels, that could be connected into local microgrids. Each microgrid would be capable of running as an independent island, but also able to exchange energy with its neighbours. There would be no need for any big grid backbone, because solar panels would be plentiful and distributed widely enough that there was always enough electricity being generated nearby.

In truth, though, this is something of a false opposition. Distributed solar panels are capable of pumping out lots of electricity. But as Brent Wanner, the grids guru at the International Energy Agency, an intergovernmental think-tank, points out, making such resources your mainstay is a costly undertaking. Economies of scale make small-scale or rooftop solar “several times more expensive" than the same capacity would be if installed at a grid-connected solar farm. Distributed solar panels can reduce the demands on an existing grid by supplying some energy locally, but the economic case for a wholesale move to decentralised power has yet to be made. Developments of both sorts are needed.

China inclines towards big grids. State Grid, which runs the synchronous grid that spans most of the country, is the world’s largest employer and runs its largest machine. Like India, China is benefiting from the momentum of recent grid expansion. The country only reached 100% electrification in 2012, at a point when the immense demands of grid decarbonisation were already becoming apparent and investment in infrastructure was being used to drive the economy.

As a result it has simply continued to invest. That said, it too is seeing increasingly long queues for connection, according to Chongqing Kang, who studies electricity systems at Tsinghua University in Beijing. He thinks the lack of real-time pricing is part of the problem. Those trying to transform the grid are left asking the government to update its policies.

Even with all the planning reform and queue-management wizardry in the world, the waits will not all go away. Construction of grid infrastructure cannot be done on a purely laissez-faire basis. Each grid is a single, cohesive machine, and new connections must be made carefully so as not to upset their delicate balance. The same is true of the new transmission capacity needed. But there things are being made easier by a 19th-century technology that is only now coming of age.

© 2023, The Economist Newspaper Limited. All rights reserved. From The Economist, published under licence. The original content can be found on


Catch all the Industry News, Banking News and Updates on Live Mint. Download The Mint News App to get Daily Market Updates.


Switch to the Mint app for fast and personalized news - Get App