 |
| 24-April-2025 |
Solar power is expanding rapidly.[1] In 2023, the total capacity of the world’s solar photovoltaic (PV) panels was four times its 2017 level, and nearly 40 times its 2010 level. And it is not slowing down.
The charge is led by China, which installed 253 gigawatts (GW) of new solar panels in 2023, bringing its total to 656 GW, or 40% of the world total of 1642 GW. Big capital elsewhere is investing heavily, struggling to catch up with China’s manufacturing capacity and supply chains.
 |
| Rooftop solar installation in India. Photo from Nikhil Technochem |
Big solar farms, owned by corporations (utility-scale, ground-mounted solar in the industry language) account for the largest chunk of world solar capacity – but rooftop solar comes a close second.
The new rooftop solar installed in 2023, 196 GW, was not far behind the 251 GW of new utility-scale solar. Total rooftop solar capacity more than doubled in South Africa, where households and businesses use it as a defence against frequent power cuts, and in Austria, Germany and Italy.
Rooftop solar kept expanding in 2024 too, and is projected to do so this year, Solar Power Europe, an industry association, says. High prices of fossil-fuel-generated electricity are a driver.
Some national and local governments now require solar to be fitted on new buildings, Solar Power Europe notes. And:
residential and commercial power consumers are evolving into prosumers [who supply electricity they do not use to the grid], PV modules are starting to become building materials, and smart cities are embracing small-scale distributed solar, combined with storage and digital solutions.[2]
Capital’s efforts to control and profit not only from renewables expansion in general, but especially from the use of renewables to supply electricity to people without grid access, have been extensively researched.[3] So have the heavy social, material and ecological burdens imposed by solar, wind and other “renewable” technologies, including by the supply chains for solar and wind farms. The very term “renewable”, used by corporations anxious to mask those burdens, has rightly been challenged.[4]
This research into the exploitative, hierarchical, economic and social relationships in which solar technologies develop is vital. Here I shift the focus, to ask: can rooftop solar be developed at least partly free of those relationships? Can it, potentially, play a part in our collective struggle to supersede capital-dominated society with a socially just society? My answer to both these questions is a tentative yes.
Much depends on ownership, which is mostly now in the wrong hands. Another crucial factor is whether people in communities, as opposed to states and corporations, can turn rooftop solar and other decentralised technologies into their own tools. Potentially, they can do, just as, potentially, they can take control of other aspects of their lives.
This article is in two parts. The first is an overview of how rooftop solar is expanding now; the second considers its technological and social potentials.
Part 1. Rooftop solar expansion
China
China’s 254 GW of rooftop solar panels in 2022 generated 120 TWh of electricity – about the same as the Netherlands’ total electricity output. Rooftop solar accounted for about 28% of China’s total solar capacity: almost all of the rest is huge solar farms.[5]
Rooftop solar is installed as part of the state’s renewables construction drive, associated with China’s stated policy of peaking greenhouse gas emissions by 2030 and reaching “carbon neutrality” by 2060. Provincial governments play a key role: most rooftop solar is sited in just five of China’s 22 provinces, with Shandong the runaway leader.
More than 80% of China’s rooftop solar panels belong to state-owned power companies, PV manufacturers or other energy companies. They lease roof space for the panels from households and businesses: ownership of panels is usually transferred to the rooftop owner only after 20-25 years. Typically, each county in a province is assigned to a particular developer. State-owned banks provide finance.
The remaining one-fifth of China’s rooftop solar is owned by households, who sell electricity they do not need themselves back to the grid. (See Note below on the terms of trade.)
China’s rooftop solar expanded at breakneck speed in 2021-22, but since then has slowed down. Big solar farms have proliferated, especially on inland deserts.[6]
Electricity distribution companies have failed to adapt, or resisted adapting, to decentralised generation. A quasi-official research body reported that, since distribution companies’ performance targets are based on electricity sales, they “lack enthusiasm” for installing panels they do not own. In Shandong, decentralised solar had been constrained by “overloaded grid integrations”, a lack of transformer capacity, excess voltage in parts of the network, disparities in equipment quality and “chaotic oversight and maintenance management practices”.[7]
China’s undoubted progress in installing rooftop solar – as well as its dominance of supply chains internationally – must be seen in the context of its continuing expansion of coal-fired electricity generation. In 2024, China started building 94 GW of new coal-fired power stations, after a permitting boom in 2022-23.
“If not curtailed, the wave of new coal plants could undo President Xi’s pledge to strictly limit growth in coal consumption”, Global Energy Monitor stated. Coal accounts for more than half of China’s rising primary energy consumption, and so its greenhouse gas emissions also continue to rise. Climate Action Tracker rates China’s policies “highly insufficient” to meet the government’s own targets.[8]
India and Pakistan
India’s central government, like China’s, has set solar expansion targets linked to climate policies, and more than 90 GW of panels have been installed in the last few years.
Four-fifths of this capacity is utility-scale solar. Rooftop installations have been slower: by March 2024 they added up to 12 GW, 60% of which was in the commercial and industrial sector.[9]
The solar construction drive began after the 2015 climate talks in Paris, boosted by a $1 billion loan to the government from the World Bank and Asian Development Bank. From 2017, a 40% upfront capital subsidy was offered to businesses and households that bought and installed panels.
But banks responsible for managing it could not be bothered to help, or credit-check, even middle-class household customers – and so “only the richest C&I [commercial and industrial] consumers were found to have access” to the subsidy, Mohua Mukherjee, who worked on the scheme, wrote in a retrospective review. Panels were installed only by large businesses, or by energy companies offering “rooftop solar as a service”.[10] (See Note on terms of trade.)
Private capital has also invested in microgrids, primarily to sell electricity to small businesses in areas served, but poorly served, by the grid. Tata Power, India’s largest energy company, in 2019 launched a renewable microgrid company that by 2023 had installed such 200 microgrids.[11]
The government, having initially failed to stimulate household ownership of solar panels, in 2019 adjusted the terms of the scheme. The 40% capital subsidy was limited to small household systems of between 1 and 3 kilowatts (kW), with a 20% subsidy for 3-10kW systems. This effectively excluded businesses, whose systems are usually larger. The renewable energy ministry also sought to persuade electricity distribution companies to cooperate with banks to provide finance. Still, uptake was slow. (The Covid pandemic was also a factor.)
In 2024, the scheme was revamped yet again. A subsidy was offered to households of 60% for solar panel systems up to 2kW, and of 40% for systems up to 3 kW, and to housing associations, for systems up to 500kW. A mechanism to cover payment risk borne by electricity companies providing “solar as a service” was introduced.
The authors of one research report reckoned that while the target demographic was “small to medium scale electricity consumers”, the terms “still favour the wealthier and creditworthy residential consumers”.[12] State subsidies, on top of federal subsidies, help; high electricity prices put on pressure, but the outcome is still unclear.
Mohua Mukherjee concluded that the challenge of making decentralised solar “affordable and relevant to the 800+ million poor people in a country of 1.3 billion” still faces India. In my view, as long as electricity is treated a saleable commodity, instead of a service, and as long as private sector finance is seen as the main instrument of solarisation, businesses and well-off households will fit solar panels. Lower-income households may not.
In Pakistan, the same dilemmas have unfolded, in the absence of any significant policy initiative, indeed in the context of constant political instability. In 2023-24, the falling prices of Chinese solar panels combined with the rising price and persistently poor quality of grid electricity to create a perfect storm for electricity distribution companies. Businesses and middle-class households fitted panels en masse: at least 5 GW off-grid and 3 GW grid-connected was installed in 2024 alone. Poor households were left still further behind. (I wrote about this in an earlier blog post.[13])
Solar inequalities across the global south
The way solar is expanding in India and Pakistan, and its social significance, is conditioned by the limits to electricity provision, common to these countries and many others outside the rich world. There are 750 million people who have no electricity access at all. But also, 2 billion or more other people are constrained by supply that is constantly interrupted and unpredictable.[14]
Development agencies, and private capital, have funded rooftop solar and off-grid microgrids to provide electricity where there is none. But rooftop solar is also being adopted by millions of people who are counted in statistics as having electricity access, but who despair of ever getting a reliable supply.
The initial impetus for Pakistan’s solar boom came in 2023 from businesses seeking more reliable and cheaper electricity supplies; they use panels to supplement, not replace, grid electricity. A similar trend is reported by South Africa’s solar industry association: the main driver for the recent expansion of rooftop solar, mostly combined with batteries, is mounting power cuts (load shedding), that in 2023 reached a high of 6907 hours, nearly double the 2022 level. Residential solar PV has “unfortunately been limited to higher income households”, the association says.[15]
Capital finds other ways into spaces at the edges of grids. A recent World Bank report mentioned a privately-funded microgrid in Nigeria that buys from the grid for seven hours each night, to supplement its own daytime solar generation, and is moving into the business of supplying electricity currently served by the grid.[16]
In urban environments with unreliable grid electricity, rooftop solar and other technologies combine in patchworks of access arrangements.
Researchers in Cotonou, Benin and Ibadan, Nigeria – both cities where the grid is “hard put to supply seven hours’ electricity a day” – found (i) grid connections, (ii) “spider webs” (unsanctioned extensions of grids by residents, often without safety precautions), (iii) off-grid market solutions such as generator sets and solar panels, and (iv) energy storage technologies. Residents with grid connections, or solar panels, resell electricity to neighbours. Richer households import panels directly from China; solar lamps and other off-grid equipment is available; some households rely on non-grid fossil-fuelled equipment such as generators.[17]
 |
| An engineer inspects rooftop solar at a business premises in Jimo district, Shandong, China. Photo by cnsphoto |
Rural areas have other patchworks. Researchers in Haryana state, India, found that, while urban households decided according to “cost effectiveness” whether to buy solar panels, and if so which ones, rural residents sought small, technically simple systems to moderate the impact of limited hours of grid electricity. Solar is often treated as a temporary solution. Systems were used with old, reconditioned batteries. “Most of the systems we saw were less than one kiloVolt and were installed on roofs, with no pillars or solid stands”, the researchers wrote. As grid supply improves, rural households are less likely to opt for solar.[18]
These examples – from places that supposedly have electricity access, but where the grid falls short – show that the human need met by solar panels varies widely from place to place.
A research project covering six countries in the global south concluded, among other things, that “solar energy has difficulty fitting into the practices of every day energy use”; equipment is often designed to offer specific services and limited amounts of power (this is particularly true of e.g. solar lamps and solar home systems); and that in many contexts “it does not replace existing traditional energy use, but operates somewhat intermittently, alongside these alternative forms”, such as cooking and heating with wood, charcoal or kerosene.[19]
Rooftop solar will continue to expand in these huge spaces, economically and geographically at the edges of commodified energy systems. There is ample evidence of capital’s efforts to control this process; there is also potential for autonomous practices.
The USA and Europe
The USA has the world’s second-largest solar capacity after China.[20] As in other rich countries, rooftop solar markets differ from those in the global south, firstly because tens of millions of households have invested in rooftop solar with no subsidies or a much lower level of subsidies. A second difference between the rich world and the rest is that, generally, the socially formed, or perceived, need is for uninterrupted 24/7 electricity supply. For this reason, off-grid solar is negligible. Among households with grid-connected rooftop solar, the proportion who also use batteries and smart meters is rising.
In the USA, with its relatively mature market, a battle is raging between middle-class households with panels and electricity utility companies, over the terms on which households trade electricity with grid suppliers.
The principle of net energy metering – that the grid pays the same price for electricity sold to it by households as those households pay for grid-supplied electricity – is under attack from the companies. For them, solar expansion carries a double threat: (i) that they lose customers for their core business, selling grid electricity, and (ii) that they lose control over the market – a key issue especially in the USA, where most states’ regulators grant supply monopolies to privately-owned utilities. (See also Note on terms of trade.)
California, where the US solar industry pioneered rooftop generation and where 1.2 million households (nearly one in ten) have solar panels, in 2023 shifted the terms of trade in utility companies’ favour. The amended scheme, NEM 3.0, cut average payments for electricity supplied to the grid by 75%.[21]
In a fierce public debate before the changes, utility companies had argued that a lower rate would take account of “avoided costs” – that is, that households generating their own electricity were not contributing sufficiently to the cost of maintaining the grid. This false logic only makes sense if rooftop solar is considered an economic burden on the existing (out of date) system, rather than as a socially and economically desirable innovation.
Of course simple greed plays a part, too. The utilities and their advocates claimed that households with solar panels do not pay their fair share, and thereby place a bigger burden on low-income households without panels. This conveyed the absurd impression that they themselves care about low-income households, belied by their notorious, ruthless practice of disconnecting thousands of those who struggle to pay their bills. “Utilities are simply trying to protect their monopolies over how energy is generated, consumed and paid for”, a prominent opponent argued.
The dispute is not limited to California. In Georgia, proposals to expand net metering were rejected; in Nevada and Florida, plans to scrap it were approved but then withdrawn in the face of popular opposition; elsewhere, the battle continues.[22]
The changes in California squeezed renewable electricity companies, that installed and operated rooftop solar as a service, out of the market. In 2023-24, 32 companies closed and the workforce shrank by more than one-fifth. New solar installations continued, but with a much higher share of households also putting in batteries. Energy researchers reported a shift to lower-income areas, with solar – supplied as a service by third parties, i.e. not a capital investment by households – being fitted to new homes under construction.[23]
Total solar capacity in Europe (EU-27) in 2023 reached 264 GW, one-and-a-half times the USA’s. Almost one-third of this is in Germany, where government policy has supported rooftop solar with subsidies (feed-in tariffs) since the early 2000s. Auctions of long-term supply contracts support utility-scale solar, which is now growing faster than rooftop solar – although, in response to the sharp increase in fossil fuel prices in 2022, the government improved rooftop feed-in tariffs. Climate policy advocates are calling for regulatory changes to support energy self-consumption, and energy sharing in multi-apartment residential blocks. Across Europe, rooftop solar – the capacity of which across the continent is twice that of utility-scale solar – is increasingly being backed up with battery storage.[24]
As a rule, the share of solar on rooftops, as opposed to utility-scale solar parks, is higher in the rich world: two-thirds in Europe and Australia, and a little more than two-fifths in the USA. In China, it is little more than a quarter, and in India, one fifth. (There are outliers, e.g. Brazil with two-thirds on rooftops.) See the Table.[25]
Forms of ownership
Who owns the world’s rooftop solar? The Chinese state, through its energy companies, is the runaway leader. In India, big capital is concentrated on utility-scale solar farms, and moving into microgrids. Businesses of all sizes who use electricity appear to be the dominant owners of India’s rooftop solar. Panels on people’s homes are mainly owned by renewable electricity supply companies – but a small number of households, and an even smaller number of co-ops, have taken ownership. In Pakistan, where raw market dynamics rather than government policy has driven solarisation, the share of households and small businesses who own their own panels is probably greater.
This pattern – big capital and businesses (whether electricity suppliers or users) dominant, with a small contingent of (mostly well-off) household owners – is repeated in many places outside the rich world. In rich countries, households who own their own panels play a bigger part. Some of these, notably in the USA and Australia, are a strong enough social force to fight political battles with electricity supply companies, who fear losing control over networks.
Direct subsidies (e.g. in India), and advantageous terms of trade with the grid (e.g. in Germany or California), have boosted household ownership of solar panels. If and when subsidies are withdrawn, or market rules tightened against households, solar expansion slows or stops.
In capitalist economies, profitability is the paramount concern in investment decisions. So solar and other renewables technologies – despite their low operating costs and low-carbon electricity production – will always struggle to compete with incumbent, politically powerful fossil fuels and market regulations that favour them.[26] The speed at which renewables expand will therefore depend heavily on political and social action.
The potential of rooftop solar in the context of social change, and in the fight to deal with climate change, is discussed in the second part of the article.
Go to part 2 of this article here. / Download both parts as a PDF here
Note. Terms of trade between households and electricity companies
These are the most common electricity trading arrangements I have come across while researching this article.
Net metering. The household owns and operates the solar panels. Excess electricity it exports to the grid, and additional electricity imported from the grid, is charged at the same rate. Usually, the electricity supply company bills the household for the net amount (imports minus exports).
This arrangement is economically most favourable to households, and thus disliked by electricity companies the world over. In the rich world, households with these agreements seek to increase panels’ effectiveness, including by acquiring batteries and smart meters.
Gross metering. The household owns and operates the panels, but all the electricity produced is bought by the electricity supply company, which resells what the household needs to it at a higher rate. A similar arrangement in China is named “self generation and self consumption with full grid feeding”.
Feed-in tariff. The household owns and operates the panels. Any surplus electricity is sold to the grid at a pre-determined tariff. FiTs were widely used in Europe and the USA: in the early years of solarisation they were set higher than the retail tariff, to encourage households to invest in solar. Generally, regulators and electricity companies reduce FiTs over time. In China it is named “self generation and self consumption with surplus grid feeding”.
Solar as a service. A third party (usually, a renewable energy company) owns and operates the panels. It sells the household what it needs, and sells the remainder on the market.
In India, a variant of solar-as-a-service, third-party financing, is used. A renewable energy company raises a loan to finance panels installed at a business, and contracts with the business either to build, own, operate and maintain the panels, or to build, own, operate and transfer the panels to the business after e.g. ten years.
Rooftop leasing. A third party (usually, in China, a state-owned electricity company) leases rooftop space for panels, and generates the electricity. The household buys electricity from the company.
References
[1] With thanks to members of the Red Green Study Group, where a presentation based on this text was discussed. All opinions expressed, and mistakes made, are my own. SP
[2] Global Market Outlook for Solar Power 2024-2028 (Solar Power Europe, 2024), page 51. The statistics in the introduction of this article are also from this report
[3] On off-grid solar, an overview of the research is: B. Radley and P. Lehmann-Grube, “Off-grid solar expansion and economic development in the global South: a critical review and research agenda”, Energy Research & Social Science 89 (2022) 102673. See also: Lucy Baker, “New frontiers of electricity capital: energy access in sub-Saharan Africa”, New Political Economy 28:2 (2023), pages 206-222; Kirsten Ulsrud, “Access to electricity for all and the role of decentralised solar power in sub-Saharan Africa”, Norwegian Journal of Geography 74:1 (2020), pages 54-63
[4] On material and ecological impacts, see Dustin Mulvaney, “Embodied energy injustice and the political ecology of solar power”, Energy Research & Social Science 115 (2024) 103607; A. Dunlap and B. Sovacool, “‘It’s just another factory’: a political ecology of solar panel manufacturing in Perrysburg, Ohio”, Journal of Political Ecology 32 (2025); B. Sovacool, A. Dunlap and B. Novakovic, “When Decarbonisation Reinforces Colonisation: complex energy injustice and solar energy development in the California desert”, Annals of the American Association of Geographers 115:3 (2025). On corporate control of supply chains and markets, see e.g. Sean Sweeney, “Sustaining the Unsustainable: why renewable energy companies are not climate warriors” (New Labor Forum, August 2021); Dexter Whitfield, Challenging the rise of corporate power in renewable energy (Spokesman Books, Nottingham, 2023). On the term “renewable”, see: A. Dunlap, “Does renewable energy exist? Fossil fuel + technologies and the search for renewable energy”, in: S. Batel and D. Rudolph (eds.), A critical approach to the social acceptance of renewable energy infrastructures (SpringerLink, 2021)
[5] Embarking on a New Era: Rural Residential Photovoltaics are Driving China’s Rural Revitalisation (Energy Foundation China / Asian Infrastructure Investment Bank, 2024), pages 26-27
[6] “China’s growth in large-scale solar exceeded rooftops in 2023”, The Straits Times, 1 November 2024
[7] Embarking on a New Era, pages 12, 42-45, 50 and 64; “China’s blistering solar power growth runs into grid blocks”, Reuters, 22 May 2024; “China’s rooftop solar boom faces grid capacity challenges”, Bloomberg, 10 January 2024
[8] China’s primary energy consumption in 2023 was 170 exajoules (compared to 122 exajoules in 2013, a decade earlier). Of this, 92 exajoules were from coal, compared to 11 Ej from hydro, 16 Ej from non-hydro renewables, and 48 Ej from oil and gas. See Energy Institute, Statistical Review of World Energy 2024. On coal, see Global Energy Monitor, Boom and Bust Coal 2025: tracking the global coal plant pipeline (April 2025). On greenhouse gas emissions, see Climate Action Tracker
[9] “India Surpasses 90 GW of installed solar capacity”, PV magazine, 22 October 2024; Prabakhar Sharma and Joti Gyulia, Unleashing the Residential Rooftop Solar Potential (IEEFA, October 2024), page 9
[10] M. Mukherjee, India’s Ongoing Rooftop Solar Journey 2017-2022 (Oxford Institute for Energy Studies paper ET18, November 2022). See also M. Mukherjee, “The evolution of rooftop solar energy in India”, New Energy World, 10 July 2024
[11] Bernard Tenebaum et al, Mini Grid Solutions for Underserved Customers: new insights from Nigeria and India (World Bank Group, 2024), page xix; “Solar microgrids address power shortages in rural areas”, Energy World, 4 December 2024
[12] Sharma and Gyulia, Unleashing the Residential Rooftop Solar Potential, page 8
[13] S. Pirani, “Pakistan’s rush for rooftop solar brings dreams and nightmares”, People & Nature, January 2025
[14] The IEA monitors electricity access on its web site, but does not collect statistics on people without reliable electricity supply. Researchers at the Energy for Growth Hub found in 2020 that 3.5 billion people live below a threshold of “reasonably reliable” supply, defined as no more than one outage, or one hour of outage, per month. The IEA estimates there are 2 billion people without access to clean cooking (i.e. usually they rely on solid biomass, coal or kerosene) and these categories overlap. See: 3.5 billion people lack reliable power (Energy for Growth Hub, 2020); John Ayaburi et al, “Measuring ‘reasonably reliable’ access to electricity services”, The Electricity Journal 33 (2020), 106828; Lauren Gilbert, “To Drive Structural Transformation, Stop Counting Connections — Focus First on Ensuring Reliable Electricity” (Energy for Growth Hub, 2025)
[15] Global Market Outlook for Solar Power 2024-2028 (Solar Power Europe 2024), page 112. In 2023, 85% of newly installed systems included batteries
[16] Tenebaum et al, Mini Grid Solutions for Underserved Customers: new insights from Nigeria and India, page xix
[17] Melanie Rateau and Sylvy Jaglin, “Co-production of access and hybridisation of configurations: a socio-technical approach to urban electricity in Cotonou and Ibadan”, International Journal of Urban Sustainable Development 14:1 (2022), pages 180-195
[18] Marie-Helen Zerah and Sarada Das, “Solar rooftop systems and the urban transition: shall the twain ever meet? Interrogations from Rewari, India”, Journal of Urban Technology 30:2 (2023), pages 103-125
[19] Ankit Kumar et al, “Solar energy for all? Understanding the successes and shortfalls through a critical comparative assessment of Bangladesh, Brazil, India, Mozambique, Sri Lanka and South Africa”, Energy Research and Social Science 48 (2019), pages 166-176
[20] The Solar Energy Industries Association reports solar installed, 2024, as: total 165 GW, including residential 40.9 GW, commercial 21.1 GW, community 8.4 GW, utility 94.6 GW.
[21] “How a policy change disrupted California’s solar panel market”, CNET, 23 May 2024; “Understanding NEM 3.0 in California”, Greenlancer.com, December 2024
[22] Jeff St John, “The battle over net metering 3.0 in California: energy equity and the future of rooftop solar (Canary Media, 12 April 2021); H. Crystal, R. Lin and J. Su, Rooftop Solar Justice (Center for Biological Diversity, March 2023)
[23] Galen Barbose, One Year In: Tracing the impacts of NEM 3.0 on California’s residential solar market (Lawrence Berkeley National Laboratory, May 2024). “How a policy change disrupted California’s solar panel market”, CNET, 23 May 2024
[24] EU Market Outlook for Solar Power 2024-2028 (Solar Power Europe, 2024); Climate Action Network, Germany’s Solar Rooftop Country Profile (2024); Climate Action Network, Engaging citizens and local communities in the solar revolution. An update (April 2024)
[25] Solar Power Europe reports that in 2023, 65% of solar in the EU-27 was rooftop, as opposed to utility-scale. (Global Market Outlook 2024-2028, page 89.) Other countries: see note to Table
[26] This subject is discussed in: Brett Christophers, The Price is Wrong: why capitalism won’t save the planet (Verso, 2024)
No comments