Carbon trading: a scheme for reducing emissions?

The rapid economic growth of many developing countries has resulted in extremely high carbon emissions in recent decades. This is exacerbating the already high and in some cases growing emissions of more economically developed countries – resulting in significant harm to environmental and human health. For example, it was estimated that in 2017 deaths due to long-term exposure to air pollution reached nearly five million globally, with approximately 1.2 million of these in China. Rising levels of carbon dioxide in the atmosphere, comprising 74.4% of total greenhouse gas emissions in 2016, is the largest contributor to global warming. This warming causes extreme weather events, such as tropical storms, wildfires, severe droughts and heat waves, as well as other forms of climate change. The climate crisis requires worldwide pursuance of environmentally sustainable development. One potential solution is carbon credit trading. 

What is carbon trading?

Carbon trading involves a market-based approach to reducing emissions by putting a price on pollution. Such schemes are currently in operation in China, the EU, and other markets.

For example, a cap-and-trade system is an emissions trading scheme (ETS) in which total allowable emissions are set, with emission rights allocated out to polluters accordingly and then traded. If a firm’s emissions exceed its emissions quotas, it can purchase them on the market, and conversely, if emissions are below quotas, firms can sell unused ones. This creates a monetary incentive for firms to reduce their emissions. Theoretically, the price of carbon should automatically adjust under the quantitative limit of carbon emissions, offering an advantage compared to a fixed carbon tax. The idea is to gradually reduce these allocations each year, which makes carbon credits more valuable and drives up the price of polluting.

The potential advantages of an ETS are plentiful: including cost-effectiveness, emissions reduction capacity, equity, and technological improvement incentives. Despite clear theoretical underpinning, the practical effectiveness of such schemes is less certain. Success may depend upon both macro conditions, such as socioeconomic development, governance mechanisms, and market environments, as well as specific design elements such as emission caps, industry coverage and quota allocation¹.

Carbon trading in China

As the largest developing country and biggest carbon emitter since 2006, accounting for 30.9% of total global carbon emissions in 2021, China's emission trading policies have the potential to significantly reduce global emissions; both directly through a reduction in China’s own impact and by providing a reference point for the implementation of similar policies in other countries.

Figure 1: Annual CO2 emissions from fossil fuels

_{Source: Grant Thornton's own analysis using data from Global Carbon Project 2022.}

In 2011, seven Chinese provinces and cities were identified as pilot emissions trading locations, with a cap-and-trade system introduced across these locations throughout 2013-2014. We conducted an empirical study into the mitigation effects of this pilot system, which is now being expanded nationally. The study used a panel dataset of carbon emission inventories for 28 industries in 30 provinces from 2007-2016, obtained from the China Emission Accounts and Datasets. This includes the seven pilot provinces and all covered industries of the scheme.  

To evaluate the mitigation effect of the pilot scheme, it's possible to use a difference-in-differences (DiD) model. DiD is a statistical technique that attempts to mimic an experimental research design using observational data, by comparing the effects of a treatment (typically a policy intervention) on a ‘treatment’ group relative to a ‘control’ group.

This involved observing the carbon emissions of regions that were exposed to the pilot ETS scheme (ie, treated) and the regions that weren't exposed to the pilot scheme (ie, control), both before and after the policy intervention. By comparing the average change over time in carbon emissions for the treatment group to the average change over time for the control group, the treatment effect of the pilot carbon emissions trading scheme can be estimated.

Figure 2 plots the trends in average total carbon emissions of a pilot province, Guangdong, compared to the non-pilot provinces. Guangdong was chosen for further analysis as it represented the pilot region where both the parallel trends assumption holds, and provincial characteristics (total CO2 emissions, GRP/capita and ratio of coal in total CO2 emissions) don't differ significantly from the non-pilot provinces. This reduces the likelihood of obtaining a biased estimate of the effect of the pilot that arises through differences in provincial characteristics and a violation of the parallel trends assumption. There's evidence of a parallel trend in the three years prior to the reform announcement, indicating that in the absence of the reform, the carbon emissions trends in the treatment group would have evolved as in the control group. The parallel trends assumption is necessary for the difference-in-differences estimator to reveal an unbiased estimate of the treatment effect associated with the pilot.

Figure 2: Guangdong vs. Non-pilot CO2 trends

_{Source: Grant Thornton's own analysis using data obtained from CEADs: Carbon Emission Accounts and Datasets for emerging economies.}

Regression analysis was run using 2011 onwards as the policy intervention period, to account for any expectation effect through which firms began changing their operation methods before the policy was implemented. Controlling for factors such as population size, GDP and ratio of coal in total carbon emissions, the pilot scheme in Guangdong was associated with a reduction in provincial carbon emissions by 78.8 Megatons- a significant reduction of 51.8%. The synthetic control method was also employed and estimated a decline in carbon emissions of 45%, consistent with the findings from DiD analysis.

According to the US Environmental Protection Agency, 78.8 Megatons of CO2 emissions is equivalent to taking 21.1 coal-fired power plants offline for a year or the electricity use of more than 15 million homes for one year. Furthermore, it's estimated that every 4,434 metric tons of carbon dioxide emissions in 2020 results in one additional excess death globally between 2020 and 2100. This implies that a 78.8 Megaton reduction in emissions would result in approximately 18,000 fewer excess deaths globally between 2020 and 2100².

The need for regulation

Evidence suggests that carbon trading schemes could significantly reduce carbon emissions and may therefore be an effective tool for tackling the climate crisis. Indeed, the World Resources Institute has suggested that a global trading market could work by allowing countries that may struggle to meet their emissions reduction targets to buy emissions reductions from other countries¹.

However, regulation is key to carbon trading schemes. Without well-structured rules that are appropriately regulated, it's possible that carbon trading schemes could be ineffective or even increase emissions. For example, carbon offset schemes which involve firms buying an ‘offset’ that finances emission reduction projects, have significant risk of ‘infra-marginality’ issues. This is a situation in which offsets may be issued to projects that would have existed anyway, potentially increasing emissions.

Even in emissions trading schemes, it's important that firms are well regulated to ensure compliance, for example, by asking polluters to show evidence that they have bought enough allowances to cover their annual emissions. It's also important to consider whether governments require regulation that involves tracking firms’ actual emissions or if it can be assumed that actual emissions are equal to reported emissions.

For more insight and guidance, get in touch with Schellion Horn.

Footnotes

1 Gao, Y., Li, M, Xue, J and Liu, Y, (2020) "Evaluation of effectiveness of China’s carbon emissions trading scheme in carbon mitigation". Energy Economics, Vol 90

2 Bressler, RD (2021) “The Mortality Cost of Carbon,” Nature Communications, 12, 4467