What will the consequences be of expanding air conditioning loads predicted for the future?
Simon Trace, EEG’s programme director, discusses the need for research
Air conditioning is currently a luxury, available largely to the relatively wealthy. However, as incomes improve this will change rapidly – especially in developing countries in the hotter regions of the world. According to the International Energy Agency (IEA), by 2050, global energy demand from air conditioners is expected to have tripled, with the global stock of air conditioners in buildings growing to 5.6 billion, up from 1.6 billion in 2018.
However, despite predictions that demand will surge, there has been little research on the consequences of expanding air conditioning loads, how fast this expansion is likely to happen, and the consequences for policy.
It is vital to understand the impact increased air conditioning use will have on electricity markets, including how it will affect power generation and distribution capacity. Air conditioners consume vast amounts of electricity – a typical unit uses 20 times as much electricity as a ceiling fan and 100 times as much as an LED lightbulb. Their adoption can therefore dramatically change a household’s electricity consumption profile – and potentially raises severe concerns about grid reliability, particularly in developing countries.
For example, the entire electricity system in Sierra Leone has a capacity of only 240MW, and the grid already suffers from frequent blackouts. The country currently has low adoption of air conditioning (less than one per cent in 2011), but it would take just a small increase to outstrip supply during high-demand hours, exacerbating reliability issues. To meet increased demand without blackouts, large economic investments in electricity generation and transmission infrastructure will be required.
The IEA suggests that by 2050, new electricity capacity equivalent to the current (2018) combined capacity of the United States, the EU and Japan will be required to meet the energy demand from air conditioners.
One potential solution recently investigated was whether the air conditioning devices expected to come online this century can be powered by clean photovoltaic (PV) electricity. The research paper Meeting global cooling demand with photovoltaics during the 21st century from Finland’s Aalto University, published in Energy and Environmental Science and recently covered in PV Magazine, demonstrated that a clear majority of the rapidly increasing cooling demand could be met with PV and small-scale distributed storage. The team found that even without any energy storage, PV could directly power approximately 50 per cent of the cooling demand.
The paper also highlights that Sub-Saharan Africa and South Asia are expected to see increasing demand for cooling in the second half of the century – and that the geographic shift of cooling demand closer to the equator (where the seasonal differences in air conditioning are lower and the temporal dynamics of PV and cooling are better aligned) significantly improves the synergy of cooling, PV and small-scale distributed storage.
There is indeed enormous air conditioning potential in Sub-Saharan Africa and South Asia – but we need to develop an understanding of how and when households in these regions will adopt these cooling devices. If demand is going to be met efficiently, energy suppliers need accurate predictions, and there is currently little reliable information.
While there is some quantitative evidence from the United States and Mexico, describing how air conditioner adoption increases as household incomes rise and how they rise differentially across climates of differing temperatures, little work has been carried out to estimate this same relationship in Sub-Saharan Africa and South Asia.
The lack of direct empirical evidence and analytical modelling of potential policy impacts is being addressed through an EEG-funded research project being led by the University of California, Berkeley. The research team is focusing on Sierra Leone, where evidence is particularly scant, but will also include all EEG priority countries in Sub-Saharan Africa and South Asia.
The project team aims to assemble and disseminate the most comprehensive dataset ever compiled on air conditioning in the selected countries. It will perform a thorough inventory of available household microdata on air conditioning, and will combine this with other data, including temperature, population, income, geographic and future energy consumption information, along with estimates/forecasts.
The use of household-level microdata is critical both for incorporating accurate measures of climate, and for describing the true relationship between household-level income and air conditioning adoption (households don’t adopt air conditioners when country-level average income reaches a certain threshold, but rather when their income reaches that level). Most previous studies about air conditioning potential are based on aggregate data – aggregate analyses are very limited, and are less accurate for forecasts.
The research will enable the team to describe empirically the relationship between income, temperature, and air conditioner adoption in Sierra Leone and other EEG priority countries in Sub-Saharan Africa and South Asia.
The study also aims to generate predictions about the future environmental impacts of increased air conditioning adoption under different policy scenarios; again, there is a lack of reliable research in this area. Little is known about how minimum energy efficiency standards, energy price reform, and other policy interventions could mitigate these impacts, by, for example, encouraging the adoption of high-efficiency equipment and/or air conditioning alternatives like evaporative cooling, cool roofs and passive cooling systems.
Increased air conditioning use will potentially have large economic and environmental impacts, and several knowledge gaps need to be addressed. The EEG research will represent a significant advance in academic literature, providing tools and evidence that can be used directly as the basis for policy development.