Kaleigh Harrison

Air conditioning is no longer a side conversation in energy policy. It is emerging as both a frontline adaptation tool and a fast-growing source of emissions.

A new study in Nature Communications, led by the University of Birmingham with research partners across Europe, China, and Australia, finds that electricity demand for cooling could rival the current emissions footprints of some of today’s largest economies by midcentury. For utilities, real estate operators, manufacturers, and policymakers, cooling is shifting from a marginal end-use to a defining variable in the energy transition.

Using established Shared Socioeconomic Pathways (SSPs) and Representative Concentration Pathways (RCPs), researchers modeled a range of futures—from aggressive decarbonization to fossil-fuel-intensive growth. The results point to sustained, structural growth in cooling demand across scenarios.

By 2050, global air-conditioning use is projected to more than double. Under mid-range development and climate assumptions, annual electricity demand for cooling could reach 4,493 terawatt-hours. In higher-emissions pathways, air conditioning alone could generate up to 8.5 gigatons of CO₂-equivalent per year—exceeding the United States’ current annual emissions of roughly 5.9 gigatons.

The researchers estimate that the additional cooling demand could contribute between 0.03°C and 0.07°C of global warming by midcentury. In a climate system operating within tight margins around the 1.5°C target, that increment is material.

Income Growth, Not Just Rising Temperatures, Drives Demand

While more frequent and intense heatwaves are increasing the need for cooling, the study identifies income growth as the dominant long-term driver of emissions from air conditioning.

As urbanization accelerates and household incomes rise in emerging markets, air conditioners are transitioning from discretionary purchases to baseline infrastructure. The modeling suggests that adoption scales more closely with purchasing power than with temperature alone.

Depending on income thresholds, additional unit deployment could reach:

Approximately 94 million units at medium-income levels
Around 150 million units at high-income thresholds
More than 220 million units at the highest-income tiers

For utilities and grid operators, this implies that cooling demand will track economic development curves rather than climate variables alone. Peak load planning, transmission investment, and distributed energy strategies will need to reflect that shift.

The study also compares the additional warming impact to the emissions equivalent of 74 to 183 billion transatlantic return flights, underscoring the cumulative scale of incremental appliance growth.

A Development Gap with Climate Consequences

The most complex dimension of the research lies in its inequality analysis. Regions facing the highest exposure to extreme heat—particularly South Asia and large parts of Africa—currently have the lowest access to air conditioning. Conversely, higher-income regions in North America and Europe have comparatively lower cooling needs but significantly higher penetration rates.

Closing that access gap is essential for public health, labor productivity, and economic stability. However, if low-income regions were to reach high-income access levels using today’s carbon-intensive technologies, the emissions impact would be substantial. Even under the most climate-aligned scenario modeled, equalizing access could add up to 0.05°C of additional warming.

This creates a structural tension between adaptation and mitigation. Higher temperatures increase cooling demand. Rising incomes expand appliance ownership. If electricity systems and refrigerant technologies remain carbon-intensive, the result is a reinforcing emissions loop.

To quantify these outcomes, researchers combined population-weighted temperature and humidity exposure with the Global Change Analysis Model (GCAM) to project adoption and energy use. Emissions outputs were translated into warming impacts using the MAGICC climate emulator.

The policy implications extend across sectors:

Accelerated grid decarbonization to limit emissions intensity of additional load
Faster transitions to low-global-warming-potential refrigerants
Strengthened appliance efficiency standards
Building design strategies that reduce thermal loads through insulation, shading, and passive cooling

Operational and behavioral shifts also matter. Modest thermostat adjustments and demand management during peak hours can reduce both emissions intensity and grid stress.

For the private sector, cooling now sits at the intersection of market expansion and climate risk. Equipment manufacturers, utilities, developers, and financiers face a dual imperative: expand access to essential cooling while redesigning the underlying systems to operate within carbon constraints.