Background, challenges, and context

Air conditioning brings relief on hot days, makes people more comfortable, increases productivity, and, during extreme heat, can even make the difference between life and death. However, it is a luxury which, for the moment, is mostly available to the relatively wealthy. Air conditioning is perhaps the single most consequential form of adaptation to climate change, and yet is unavailable to large parts of the global population.

Sierra Leone currently has low adoption of air conditioning (less than 2% of households) and only 20% electrification, but, as incomes increase, this will change rapidly – potentially raising severe concerns about grid reliability.  

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), so their adoption can dramatically change a household’s electricity consumption profile. The entire electricity system in Sierra Leone has a capacity of only 240 MW, and the grid already suffers from frequent blackouts. It would take just a small increase in air conditioning adoption 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 – and understanding where and when is crucial if investments are to be made efficiently.  

Air conditioning also raises major environmental challenges. The energy efficiency of air conditioners currently for sale around the world varies significantly – some of the worst-performing products consume more than twice as much electricity per unit output. However, minimum energy efficiency standards work, in large part, by removing the most energy-inefficient air conditioners from the market – but these are also the cheapest units, thus creating a trade-off between efficiency and equity. 

In addition, the refrigerants used in air conditioners are potent greenhouse gases known as hydrofluorocarbons (HFCs). Therefore, understanding the pace of growth in air conditioner sales is crucial in terms of the Kigali Agreement, which seeks to significantly reduce the use of HFCs.  

Despite the enormous air conditioning potential in Sierra Leone and other countries in Sub-Saharan Africa, as well as South Asia, and the large potential economic and environmental impacts that increased use may have, there is little direct empirical evidence on this topic, or analytical modelling of potential policy impacts.  

Little work has been carried out to estimate how air conditioner adoption increases as household incomes rise, and how it rises differentially across climates of differing temperatures, and there is limited reliable information about where and when air conditioning adoption will occur. There is also a lack of reliable research on the impact of air conditioning adoption on the environment, and 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.

Research overview and objectives 

This project aimed to fill several important knowledge gaps, including a lack of data, while generating predictions about future environmental impacts under different policy scenarios. 

The research team focused on Sierra Leone, where evidence is particularly scant, looking at how air conditioning demand materialises in the context of planning for the development of the country’s power sector.

The research team also aimed to characterise empirically the relationship between climate, income, and residential air conditioner adoption in 16 countries (including Sierra Leone). The potential impact for minimum efficiency standards, electricity tariff design, and related policies was assessed. In addition, the team projected energy consumption and environmental impacts out to 2030 and beyond for a variety of alternative policies.  

Research methodology

In Sierra Leone, a thorough inventory of available household microdata on air conditioning was performed, using sources such as the country’s Integrated Household Survey, which was carried out with assistance from the World Bank, and CLASP, a non-profit organisation focused on making appliances more energy efficient with experience in designing and implementing household-level energy and demographic surveys.

Household-level microdata was combined with other data, including temperature, population, income, geographic, and future energy consumption information, along with estimates/forecasts.

Members of the research team together with Innovations for Poverty Action Sierra Leone (IPA-SL) also conducted focus groups with households, interviewed government and electricity utility officials, and obtained and reviewed official documents regarding energy policies.

Furthermore, using household-level microdata from 16 countries (including Sierra Leone), the team compiled one of the most comprehensive household-level datasets ever assembled on air conditioning, with data including representative samples from over 50% of the total global population. This microdata was matched with rich climate data.

A model was developed to forecast future adoption and show how patterns vary across countries and income levels.

Research results, key messages, and recommendations

• It is forecast that residential air conditioning penetration in Sierra Leone will increase from less than 2% of households today to around 3.5% by 2030 and 10% by 2050.

• The effect of increased air conditioning uptake on total electricity demand in Sierra Leone is highly uncertain, but ‘back-of-the-envelope’ calculations suggest it will play a large role. Despite low adoption, air conditioning could dramatically increase total electricity demand in the country in the coming decades. Projections of energy consumption from air conditioning by 2050 are, for example, 636 GWh in a Low Energy Efficiency Scenario or 595 GWh in a High Energy Efficiency Scenario. Sierra Leone’s current annual electricity consumption is roughly 270 GWh.

• The study of 16 countries implies large growth in residential air conditioning, particularly among low- and middle-income countries.

• Across all 16 countries, it is predicted that the fraction of households with air conditioning will increase from 35% in 2020 to 55% in 2050. With the total population in the 16 countries expected to increase from 3.9 billion to 4.7 billion, this implies that the total number of air conditioning units across the 16 countries will increase from 295 million in 2010 to 691 million in 2050.

• Starting from a low base, Sierra Leone, Ghana, Nigeria, and South Africa experience several-fold increases in air conditioning. India experiences dramatic growth, from below 10% in 2010 to 50% in 2050.

• The results imply that income growth is far more important than the speed of global warming for determining future adoption of air conditioning. While climate/global warming matter, income matters much more, explaining 85% of the growth in air conditioning by 2050.

• In relatively cool areas, air conditioner adoption is near zero for all income levels. In warmer areas, however, there is a close relationship between household income and air conditioner adoption, with sharp increases in air conditioning starting at annual household income of about $10,000 USD.

• Not only do richer countries have much more air conditioning than poorer countries, but within countries, both current and predicted future use of air conditioning is overwhelmingly concentrated among high-income households – this is particularly stark in relatively low-income countries. For example, the study suggests that in Pakistan, by 2050, 38% of households in the high-income tercile will have air conditioning, compared to only 5% in the low-income tercile. A similar story is evident in many African countries, with air conditioning penetration for the lowest-income tercile remaining below 4% in South Africa, Nigeria, and Sierra Leone. Low-income households being substantially less likely to have access to air conditioning has important implications for productivity, education, and health.

• The predicted air conditioning saturation levels imply 550,000 fewer deaths per year by 2050 and 9.2 million fewer cumulative deaths by 2050, with a disproportionate number of avoided deaths occurring among the highest-income tercile. Differential adoption of air conditioning puts vulnerable households at a growing disadvantage, threatening to exacerbate existing disparities in society. A growing number of poorer households will succumb to heat-related mortality by 2050.

• Many of the world’s poorest households will be exposed to increased heat associated with climate change without the important mitigating influence of air conditioning.

This research represents a significant advance in academic literature. It has provided tools and evidence that can be used directly as the basis for policy development, and the team anticipates that other researchers will use the data to continue to drive the air conditioning agenda forward.  

Local partners

Innovations for Poverty Action Sierra Leone (IPA-SL)

CLASP, formerly known as the Collaborative Labelling and Appliance Standards Program