ENERGY – Air conditioning: a climate enemy or misplaced environmental guilt?
Every new heatwave triggers the same knee-jerk reaction in Europe: equipping one’s home with air conditioning is seen as environmentally irresponsible. The reasoning appears sound, but it confuses several time scales and types of impact: the physical operation of an air conditioner, the history of refrigerants that are no longer in use, and its actual effect on the electricity grid. On all three counts, the situation in 2026 bears little resemblance to the concerns of the 1990s.
An air conditioner does not produce cold: it moves heat
An air conditioner does not ‘produce’ cold: it removes heat from a space and discharges it elsewhere. In physics, cold is not a substance in itself, but the result of a lower level of thermal energy. This process is based on a very simple cycle: the refrigerant absorbs heat from inside the room as it evaporates, then releases it outside as it condenses. The principle is exactly the same as that of a fridge, but applied to an entire room.
The compression-expansion cycle of an air-conditioning system, simplified diagram
This gives rise to a paradox that is rarely highlighted: this same logic of heat transfer also applies to heat pumps, which are now widely favoured and subsidised for their energy efficiency. The main difference lies in the direction in which the cycle is used: in winter, a heat pump extracts heat from outside to heat the interior; in summer, a reversible heat pump reverses the process to cool the home. In other words, what is praised as a low-carbon heating solution often comes under scrutiny when used for cooling.
Air conditioning does indeed discharge heat into the immediate outdoor air: a landmark study (Tremeac et al., Applied Energy, 2012) estimates that, during widespread use of air conditioning during a heatwave comparable to that of 2003 – characterised by temperatures reaching 44.1 °C in the south and 39.5 °C in Paris – this phenomenon could increase the average temperature of Parisian streets by 0.25 to 0.75 °C, with localised rises potentially approaching 2 °C. This is a real effect, which contributes to the urban heat island phenomenon, but is distinct from the overall climate impact, which depends on the electricity consumed and any refrigerant leaks. The former relates to urban planning and energy efficiency; the latter to the electricity mix and the type of gas used, which explains the importance of the history of refrigerants.
From gases that depleted the ozone layer to today’s refrigerants
Early refrigeration systems used hazardous refrigerants – ammonia, sulphur dioxide and methyl chloride. Chlorofluorocarbons (CFCs), adopted as early as the late 1920s due to their apparent safety, were found to be responsible for the hole in the ozone layer. The Montreal Protocol (1987) introduced their phasing out, before extending this approach to hydrochlorofluorocarbons (HCFCs), which were used as interim solutions.
Hydrofluorocarbons (HFCs), which replaced them, have eliminated the risk to the ozone layer, but have a global warming potential that can be several thousand times greater than that of CO₂. Their use is therefore being phased out globally under the 2016 Kigali Amendment1 and, even more rapidly, within the European Union under the F-Gas Regulation2. From 2026, this regulation restricts the use of gases with a high global warming potential (GWP) for maintenance purposes, and then their presence in new equipment from 2027 onwards, with the aim of phasing them out almost entirely by 2050. Modern air-conditioning units already use new-generation refrigerants (HFOs, CO₂, propane) that are in a completely different league to the gases used in the 1990s: the climate case against air-conditioning therefore largely concerns a technology that no longer exists.
Cooling in summer, heating in winter: the same struggle for survival
The strongest argument in favour of air conditioning is health-related. A study published in Nature Medicine attributes 61,672 heat-related deaths in Europe during the summer of 2022 – a figure comparable to the 2003 heatwave (around 70,000 deaths) – and 47,690 deaths in 2023 (Gallo et al., 2024), with older people being the hardest hit. According to Santé Publique France, the heatwaves of summer 2026 resulted in a provisional figure of nearly 8,000 excess deaths, based on figures published in mid-September.
No one would describe winter heating as a luxury: it keeps homes at a temperature suitable for living, particularly for older people. Air conditioning plays the same role in summer. Low uptake, rather than air conditioning itself, is the real public health anomaly in Europe: in the United Kingdom, for example, barely 5 per cent of households live in air-conditioned homes.
Can the electricity grid cope with the rush for air conditioning?
The most legitimate concern, however, is whether the electricity grid can cope. Europe is starting from a very low base – the share of households with air conditioning stood at 19 per cent in 2022, compared with 90 per cent in the United States (IEA), far below the levels in Japan (91 per cent) or South Korea (86 per cent).
Share of households with air conditioning by country — IEA (2022) and press estimates (2025–2026)
In France, the gap is closing rapidly: according to the French Electricity Transmission Network3 (RTE), the proportion of air-conditioned homes could reach 50 per cent as early as 2035, compared with nearly 25 per cent at present.
Projected proportion of air-conditioned homes in France — RTE and IEA
This increase will indeed have an impact on electricity consumption, but the scale of this impact remains limited. According to ADEME/MICO4 and Enedis5 , air conditioning could add between 4 and 13 TWh of additional consumption by 2035–2050, depending on the scenario – significantly less than the electrification of heating or transport. Above all, its climate impact depends heavily on the carbon content of the electricity used: with electricity at 19.6 gCO₂ /kWh in 2025 (RTE) – one of the lowest carbon levels in Europe, thanks to nuclear power – compared with 178 gCO2 /kWh on average across the European Union, the same increase in consumption has around nine times less impact on the climate in France than elsewhere in Europe. The key issue is therefore not just how much electricity will be consumed, but when it will be consumed and with what carbon intensity.
It is precisely on this point that the risk of electricity peaks comes into play, rather than on annual consumption itself. The demand for cooling, moreover, coincides with the solar peak: in summer, solar power covers on average 8 to 10 per cent of French consumption, with instantaneous peaks reaching up to 35 per cent around midday. This temporary surplus of supply regularly causes wholesale prices to fall to zero or even become negative. Two measures can further limit the risk of peak demand: firstly, the new equipment being installed consists mainly of reversible heat pumps – the same unit, when used in winter, heats or cools depending on the season, without requiring the addition of separate equipment to the existing stock; secondly, the energy efficiency of buildings and demand-side management can smooth out power demand during heatwaves. The rise in air-conditioning use is therefore an issue the grid needs to anticipate, but it need not become either a climate or a power-system crisis.
Three charges, one real issue
The criticism levelled at air conditioning conflates three things: a fluid-based technology that is already regulated and largely obsolete; a public health issue where Europe’s lack of adequate facilities costs far more lives than air conditioning threatens; and a real but foreseeable electricity challenge, already partly resolved by reversible heat pumps. Doing without air conditioning changes neither the emissions from a coal-fired power station nor the global warming potential of a refrigerant that has been banned for fifteen years: it merely shifts the risk to the health of those who remain exposed to the heat.
Find all Economic Studies publications on our website.
1. Kigali Agreement (2016) and the Ministry for Ecological Transition
2. EU Regulation 517/2014 and the F-Gas III revision (EU Regulation 2024/573, OJEU of 7 February 2024)
3. RTE, 2023–2035 Forecast and Energy Futures 2050
4. ADEME / MICO Observatory (2024)
5. Enedis, French Observatory for the Ecological Transition