Per the National Renewable Energy Laboratory (NREL), cooling today is responsible for “10% of the global electricity use and nearly 4% of the annual greenhouse gas emissions,” a number poised to skyrocket as millions of people enter the global middle class in hot countries such as India. This rapid growth has major implications for the climate. So long as the power grid is not decarbonized, this rapid expansion in electrical demand can result in massive emissions, about 2 billion tons annually according to the NREL, a number they project to increase fivefold by 2050 due to cooling demand growth around the world, especially in India. Further compounding the climate challenge is the use of coolants which, if leaked into the atmosphere, can have far more damaging effects than CO2 (though they last much shorter in the atmosphere). Estimates differ on the share of emissions stemming from energy demand versus refrigerant leakage, with the NREL calculating roughly 57% of all cooling-related emissions originating from energy demand, while the IEA places the number around 70%.
Harms to the climate are hardly the only costs to a massive expansion in air conditioner use. According to Colelli, Wing and de Cian (2023), cooling imposes large, correlated demand in electricity usage, often putting strain on weak electrical grids and potentially “leading to blackouts precisely that can ultimately leave populations without power precisely when cooling, and the electricity it relies on, are most needed.” Moreover, the increase in electrical demand leads to further burning of fossil fuels, contributing to noxious local air pollution. According to the IEA, as of 2015 (when A/C ownership in India was substantially lower than today), “power plant emissions due to space cooling accounted for 9% of sulfur dioxide (SO2), and 8% of nitrogen oxides (NOx) and particulate matter (PM2.5) emissions.” These particular emissions can have disastrous effects on human health, with total particulate emissions (not just those caused by A/Cs) shortening the average resident of India’s life by over 5 years on average.
To be clear, the solution cannot be to reduce the number of air conditioners. Cooling provides enormous benefits as well–from enhanced quality of life to improved health to improved economic productivity. Per Somanathan et. al 2021, “climate control significantly mitigates productivity losses. In a national panel of Indian factories, annual plant output falls by about 2% per degree Celsius.” Globally, Romanello et. al 2021 further estimated that “use of air conditioning… averted an estimated 195 000 heat-related deaths among people aged 65 years or older in 2019.” But, in India, where only 6% of households had A/Cs in the period examined, the authors estimated air conditioners averted only 2400 heat deaths (5% of total). In other words, access to air conditioners has enormous benefits. The key is to make sure this valuable expansion of A/C availability does not come at the cost of substantially higher pollution.
Market failures in cooling
Important work has been done to improve the efficiency of A/C units over time. Stricter standards and regulations may have played a major role as well. Indeed, a report from Lawrence Berkeley National Laboratory estimates that “market average efficiency of room ACs by about 35% between 2006 and 2016 (3% per year).” Furthermore, the Kigali Amendment to the Montreal Protocol–the international treaty governing the use of coolants–specifies that countries like India have to substantially reduce the use of popular polluting coolants called HFCs by 80% by 2047. However, there are several reasons to believe that private economic incentives are insufficient to properly incentivize research and development into creating a new air conditioner.
First, the effects of air conditioning on climate change, local air pollution and electrical grid reliability are externalities, or effects that occur to others and in which the purchaser does not themselves incur. The buyer captures some of the benefits of more efficient units–lower monthly electrical bills–but those private benefits are only a small fraction of the total social benefits. Since a producer cannot charge extra for all of the social benefits their more efficient unit produces, they have little incentive to invest large sums of money attempting to develop such a unit.
Second, electrical subsidies in India mean that consumers may not even be paying the resource cost of the electricity they use. Per the International Institute for Sustainable Development, direct electricity tariff subsidies and cross-subsidies totalled roughly USD $25 billion in 2019, with residential consumers paying less than 80% of supply costs. When electricity is subsidized, the private incentive to pay more for an electricity-efficient unit further falls.
Third, a combination of credit constraints and behavioral biases may result in consumers optimizing primarily on the upfront sticker price of an A/C unit rather than the lifetime cost. If one has limited cash reserves and limited ability to borrow, then one may not want to choose a unit with a greater sticker price, even if it saves money over the long term. Indeed, 3/4s of surveyed households in India cited cost as a barrier to purchasing an energy-efficient A/C unit.
Given these market conditions, the commercial incentive to invest in A/C R&D to develop a more efficient unit may be incommensurate with the social benefits of doing so. Market shaping may thus be a valuable tool to align these incentives. To this end, the MSA has selected clean cooling as one of its seven finalist topic areas for its inaugural Innovation Challenge.
How market shaping can help
The Global Cooling Prize launched in November 2018 with the goal of producing an air conditioning unit that was substantially more efficient than existing market models. The $1 million Grand Prize was ultimately split in 2021 between two firms, who delivered on the promise of delivering a unit with 5x lower climate impact than conventional units. Unfortunately, three years later, these new units do not exist on the market, and there seem to be little imminent plans to make them a reality, with winning manufacturer Daikin citing supply chain barriers as a principal issue. As a result, further work may be necessary.
The beauty of a pull mechanism–which conditions payment on outputs–is that it does not require the sponsor to pay unless results are produced. Many pull mechanisms can also be firm- and technology-agnostic–that is, they do not require specifying in advance which firm or technology is desired. They only require specifying the desired result (an affordable A/C unit with high energy efficiency) and are flexible to support whatever company or technological approach that can produce such a result. A well-designed pull mechanism may also have a market test, whereby payments are linked to actual purchases (and thus consumers must bear some of the cost) which may be especially important in this context. A large number of failure cases cannot be reasonably anticipated in a target product profile specified ahead of time. What if the unit breaks easily? What if it is hard to fix? What if fixing the cooling unit requires specific technicians that are not readily available? What if installation is difficult, or does not work with commonly available apartment dimensions? What if supply chain issues unexpectedly raise the price of the unit? It may be impossible to anticipate how much different consumers weigh each of these factors, let alone their interactions/combinations of these factors. While some of these can be anticipated ahead of time, it may not be desirable to be prescriptive about each of these problems–it is possible that a product fails one of the criteria but consumers still like it because the unit excels in other dimensions. The market test provides the correct balance between being over- and under-specified. It is a flexible mechanism that will weed out subsidies to products that are not useful in the real world (false positives) while minimizing cases where useful products are excluded for failing unnecessary criteria (false negatives).
There are other pitfalls to avoid as well. On the one hand, it is important to avoid subsidizing innovations that were likely to happen regardless. Considering that consumers do have an incentive to seek out more efficient A/Cs (because it saves on electricity), firms have some–though inadequate–incentive to compete on efficiency. As a result, there is always a risk that an insufficiently ambitious demand-pull program merely subsidizes A/Cs that likely would have been invented without the extra encouragement. On the other hand, designers need to be careful about setting too ambitious a target, not suited to India’s market, that results in manufacturers or consumers ignoring the incentive. This is not to say that pull mechanisms cannot be technologically distant– in fact they can be a useful way to incentivize the development of technologies where we are uncertain about their feasibility.
MSA is supporting a team from the Clinton Health Access Initiative advised by Professor Paul Novosad of Dartmouth to tackle all of these questions and more, in ultimate pursuit of a pull mechanism that can make more climate-friendly cooling a reality.
