Developing China’s National Emission Trading Scheme: Experiences from Existing Global Schemes and China’s Pilot Programs

Market-based emission trading schemes (ETSs) are widely used in the developed world to reduce greenhouse gas (GHG) emissions which are perceived as the source of global climate change. China, as the largest GHG emitter in the world, is committed to introducing an ETS to reduce emissions. Here we reviewed existing ETSs and sustainable energy policies worldwide as well as China’s pilot programs. These studies were conducted in order to propose recommendations for national initiatives and strategies to be implemented in China in relation to climate change adaptation and mitigation. It has been shown that setting emission caps in the context of a national emission intensity target is difficult. However, implementing reliable systems for measurement, reporting, and verification of emissions are essential. A two-level management system (by central and provincial governments) for carbon trading is beneficial to ensure uniform standards and compliance while maintaining flexibility. Persistent political support from, and effective coordination of, policies by the government are crucial. In addition, strengthening of institutional innovation, and the establishment of a national GHG emissions information system, are of equal importance. This vital information could provide a great opportunity for China to re-define its economic growth and take global leadership in combatting climate change.


Introduction
Market-based emission trading schemes (ETSs) represent the most efficient economic policy towards climate change. ETSs facilitate emission reductions in a costeffective manner (Egenhofer, 2007). Within an ETS, allowances or permits are freely sold and purchased on the international market, resulting in the development of the market 'carbon (C) price'. If the market price is lower than the cost, companies are allowed to purchase the allowances from the market, since making its own investments would be more costly than purchasing the necessary allowances. In such a case, a company may benefit from an additional abatement that exceeds the allowances required by allocation (Böhringer and Rosendahl, 2009).
Nearly all developed economies have existing or proposed ETSs, such as the European Union Emissions Trading System (EU ETS), the New Zealand Emissions Trading Scheme (NZ ETS), the 2007 Canada Regula-tory Framework for Air Emissions, and the Tokyo C Trading Initiative (Ranson and Stavins, 2016). Some developing countries are also attempting to introduce a trading market, such as the tradable permit system for total suspended particulates in Chile (Montero et al., 2002). There are two predominant types of trading schemes: 'cap-and-trade systems' and 'baseline-andcredit systems' (Betsill and Hoffmann, 2011). In a cap-and-trade system, the upper limit on emissions is set as fixed, while the emission permits are auctioned out or distributed for free according to specific criteria. In contrast, under a baseline-and-credit system, there is no such fixed limit on emissions. However, when polluters reduce their emissions more than they are obliged to, they can earn 'credit'. This credit can then be sold to those who need to comply with regulations. The capand-trade approach has been the most widely used ETS in the developed world. On the contrary, the baseline-and-credit approach has been adopted more frequently in developing countries. In this way, companies earn emission reduction credits for emissions below their baselines.
Most of the existing schemes have been introduced in a step-by-step manner; phases and different parts of emission entities were gradually put into the scheme (Betz and Schmidt, 2016). The initial phases were used to gather emissions information, undertake experimentation of C market efficiency, obtain feedback from different sectors, and seek resolutions of emerging disadvantages (Chaabane et al., 2012). Monitoring, reporting, and verifying (MRV) systems are equally important in the implementation of emissions reduction programs. A national or regional online registry system is always in effect in advance of the scheme start date. Some schemes also consult with industrial representatives and economic organisations about regulations such as the threshold of liable entities and the level of assistance to heavily affected participants. Nearly all ETSs are roughly deficit-neutral in their budget. This is because money gained by the government from the auction is always spent on supporting strongly affected industries and households, thus establishing a low-C technology fund (Chapple et al., 2013).
As the largest contributor of greenhouse gas (GHG) emissions, and the largest developing economic entity, China been committed to reducing its C intensity by 60%-65% of the 2005 level by 2030 under the Paris Climate Change Accord (Mu et al., 2018). Acknowledging that it will be significantly affected by global climate change, China's central government attacheds great importance to this issue and has already adopted a series of actions. China has announced its intention to establish a national emissions trading market. In fact, in 2013 and 2014 (Zhang, 2015) pilot ETS programs were developed in seven municipalities and provinces (Beijing, Shanghai, Tianjin, Chongqing, Shenzhen, Guangdong, and Hubei). Although playing an active role in the international emissions trading market through the Clean Development Mechanism, China's national ETS has not yet been developed. Learning from the experiences of other countries in introducing an ETS would be helpful in achieving its ambitious goals while posing minimum damage to the economy. This paper reviews existing ETSs and sustainable energy policies worldwide, especially in the context of international treaties and initiatives on climate change adaptation and mitigation strategies. The status of China's pilot program is also discussed. Recommendations on effective ETSs are developed in order for China to implement national schemes and strategies for climate change adaptation and mitigation.

Overview of existing and emerging schemes
About 40 national and over 20 sub-national jurisdictions are putting a price on carbon. Together these carbon pricing instruments cover almost six gigatonnes of CO 2 or about 12% of the annual global GHG emissions (World Bank Group, 2016). Various types of ETS programs have been implemented around the world, including multi-national and national schemes, such as the EU ETS and the NZ ETS; regional schemes, such as the trading schemes in North America, Japan and Chile; legally binding schemes, such as the EU ETS and NZ ETS; voluntary schemes, such as the Chicago C Exchange (CCX) and the UK ETS; and schemes that only cover specific GHGs, such as the US sulphur dioxide (SO 2 ) trading scheme. Regardless of the levels and the kinds of GHG the schemes are based on whether the scheme is in action or still being implemented, examining and summarizing the key features of these schemes should afford a clearer picture of the basic elements that are present in a classic ETS.
In operation from 2005, the EU ETS has faced some challenges resulting from the establishment of the largest emissions trading market in history (Convery, 2009). Currently, the EU ETS is operating in 31 countries -including 28 EU Member States and Iceland, Liechtenstein, and Norway. The EU ETS covers CO 2 emissions from the power sector, combustion plants, oil refineries, and the iron and steel industry, as well as installations producing cement, glass, lime, bricks, ceramics, pulp, and paper. More than 10 000 covered entities account for around two gigatonnes or 40% of the total EU GHG emissions.
National emission trading programs have been discussed in the US and Canada, but have so far failed to receive the necessary political support. Instead, North American C trading schemes have emerged at the regional level: nine US states have joined forces in a joint trading system named the Regional Greenhouse Gas Initiative, and Quebec (Canada) linked with California's emissions trading program in January 2014 (Ranson and Stavins, 2016).
Recently, Asia has seen a strong momentum towards emissions trading. South Korea's emissions trading program entered into force in January 2015 and covered over 60% of the country's emissions (Park and Hong, 2014). Over 500 companies are covered in the power and industry sectors, as well as waste and domestic aviation. In other areas of the Asian Pacific region, the situation is much varied. The Tokyo Metropolitan Government has been operating a trading scheme for indirect CO 2 emissions since 2010 (Ascioglu et al., 2011). Australia abandoned a long-planned national ETS (C tax) in 2013 after a change in government. The Kazakhstan ETS started with a pilot phase in 2013 and covers CO 2 emissions. Sectors including energy, mining and metallurgy, chemicals, cement and the power industry are included (Ranson and Stavins, 2016).

Main features of existing emissions trading schemes
Existing and emerging trading schemes vary in terms of design features such as coverage and scope, or allocation methods. Large-scale, multi-national or national ETSs tend to have long-term targets and cover all industrial sectors and all GHGs (Perdan and Azapagic, 2011). However, regional and local GHG plans have a relatively short action period and only include specific sectors. The EU ETS focuses on large energy producers while ETS schemes in the US and Canada seem to have similar coverage (with the exception that the linked California/Quebec program also includes the transport sector). However, some of the emerging schemes in Asia involve smaller facilities, buildings, and indirect emissions from energy consumption. The NZ ETS is the only national ETS to cover forestry activities. The key criteria for the assessment of coverage and scope options in an ETS include the effectiveness of meeting the environmental objectives with less uncertainty, such as the efficiency of meeting the environmental objectives at the least cost, and the determination of the political acceptability (Chevallier et al., 2011).
Few ETSs worldwide have a significant share of auctioning. However, emerging schemes prefer to auction the emission allowances rather than their free allocations, or at least adopt a year-by-year increasing auction (Hepburn et al., 2006). In addition, project-based offset and banking are allowed in most of the internationally linked schemes. For example, California's scheme has a price floor that is implemented by way of reserve price at auction. Most ETSs have an implicit price ceiling in the form of a compliance penalty.
Important indicators of the more efficient and prosperous C trading markets include a full inventory of sources and emissions, establishment of a national registry system, better measurement of emissions, and more certain definition over permits. In addition, it is critical to include appropriate allocation of allowances, clear penalties, banking permission, low transaction costs, and improvement of monitoring and enforcement .

The Chinese pilot trading schemes
President Xi Jinping has urged that China should pursue a new mode of growth to promote 'more efficient, equal, and sustainable economic development'. This demonstrates the role of climate change mitigation in redefining China's economic growth. In 2011, the National Development and Reform Commission (NDRC) announced its plan to initiate seven ETS pilot programs in five cities, Beijing, Shanghai, Tianjin, Chongqing, and Shenzhen, as well as two provinces, Guangdong and Hubei (Fig. 1). The implementation of this plan began in 2013. In 2015, all programs were enforced. Table 1 pre-sents an overview of Chinese ETSs and Table 2 provides background figures for the seven pilot regions.
Covering an area of 480 000 km 2 with a total population of 250 million, the pilot regions have different industrial structures and economic development levels, producing 27% of China's gross domestic product (GDP) in 2010 (Zheng, 2014). Each pilot was designed to combine the strenth of the provincial and municipal Development and Reform Commissions, local emissions trading exchanges, and academics in universities and think tanks.
These pilot programs are different in terms of caps and targeted sectors (Munnings et al., 2016). Beijing is the only pilot that requests absolute annual emission reductions (quantifying the amount of CO 2 and other reduced GHGs) for existing facilities in the manufacturing and service sectors. The other ETSs require reductions in emission intensity, a ratio of GHG emissions per unit of production (Liu et al., 2017). Shenzhen and Tianjin allow investors and entities not covered by the ETS, such as financial organisations, to participate in trading. This results in higher trading frequency and potentially larger price fluctuations. By choosing regions with different features for policy experiments, pilot programs can help policy makers to avoid risks associated with the direct implementation of a one-size-fits-all policy. In this way, they assist the government in tailoring an eventual national program to reflect the diversity of China's regional circumstances. The increased performance of the pilots has encouraged other regions to develop C trading programs (Liu and Lu, 2015).

Approaches and challenges towards a national ETS in China
In December 2014, the NDRC released interim measures for C emissions trading, providing some legal basis for a national ETS. However, more specific details still need to be worked out to develop a fully operational scheme and present a clearer picture of the exact nature of the national market. There are two options to move into this direction. The first is to establish a nationwide ETS by linking those existing pilot C trading schemes that meet all of the qualification conditions, and to integrate them into a national linked system. The second is  Firms in the electricity, cement, iron, steel, and chemical industries with annual CO 2 emissions during 2011-2014 higher than 20 000 t

Hubei 2013
Industrial companies producing iron and steel, chemicals, cement, automobiles, electricity, non-ferrous metals, glass, and paper that consume more than 60 000 t of standard coal equivalents of energy per year Tianjin 2013 Heat and electricity production, iron and steel, petrochemical, chemical, and oil and gas exploration firms that emit over 20 000 t of CO 2 /year Chongqing 2014 Firms in the power sector and electricity consuming sectors with a threshold of 20 000 t CO 2 /year Note: Data were collected from http://www.icis.com/energy/C-emissions/china-ets-news-analysis/ that, based on experience and lessons learned in the pilots, China establishes a national ETS, and until a fully functional ETS is established, the regional ETSs continue to function in parallel. However, those entities covered in the existing regional C trading pilots will be unconditionally incorporated into a nationwide ETS if they meet the standards set by a nationwide regime. These standards are expected to be much stricter than those set in most of the existing regional C trading pilot programs. Both options have advantages and disadvantages that must be weighed against a range of criteria including administrative expenditures. If worths further investigation to determine which option fits better in China's case. Initially, China plans to include six sectors in its national ETS: power generation, metallurgy, nonferrous metals, building materials, chemicals, and aviation (Zhang, 2016). The threshold for an emission source to be covered is set at 26 000 tonnes CO 2 per year (Zhang, 2016). This suggests that the national ETS will initially include about 10 000 entities and estimate China's C market to cover three to four billion metric tonnes of CO 2 emissions (Zhang, 2016). This means that China's ETS will be the world's largest scheme, twice the size of the current largest ETS, the EU ETS. A national legislation of ETS needs to be established to authorise emissions trading at the national level. It is important to provide united guidelines and methodologies for the design and operation of the ETS and enforcement of MRV, incorporate penalties for non-compliance, ascribe allowances as financial assets, and define their valid duration with the aim of generating economically valuable and environmentally credible reductions. These factors will serve to provide a solid basis for developing a sound national ETS. The recently released interim measures for C emissions trading are a move in the right direction. Different timing, coverage, and scope suggest a twolevel management system of C trading. The central government should be in charge of setting national rules to ensure, among other things, that there are the same rules regarding coverage and scope, uniform standards for MRV, and the allocation of allowances, as well as standard rules of compliance across provinces or their equivalent. In the meantime, provincial governments should be assigned to take responsibility for implementing rules. This includes identifying the entities covered and determining their relevant emissions, calculating the amount of allowances to the entities covered, distributing these allowances to the entities, and implementing compliance rules once approved by the central government. Provincial or municipal governments should be allowed to set even stricter rules than the national rules. For instance, they could increase the coverage of the sectors and the scope of entities, and could have even stricter allocation rules for allowances.
Price uncertainty and market stabilisation are expected to become even larger issues in a nationwide ETS. Although the pilots reserve some allowances for cost-containment purposes, the difficulty lies in setting aside an appropriate level of allowances for this purpose. This is related to triggering conditions that have not yet been disclosed for most pilots. It would be helpful to introduce both a price ceiling and a price floor in the pilot trading scheme, which will remove downside risks for investors while delivering the objective of efficiently cutting C emissions (Wood and Jotzo, 2011). Meanwhile, setting a price ceiling is very helpful in limiting the potential market power of a given large player in a small, fragmented market. Detailed sectoral, regional, and countrywide studies on C abatement can provide some basis for selecting a figure. Given that the cost of abating C emissions varies significantly among sectors, a price floor needs to be set higher than the lowest abatement cost projected for the trading sectors. This will encourage C abatement for some sectors that find it relatively hard to meet their emission targets through their own actions.
It is difficult to determine whether this central and provincial government, two-level management system, is aligned with the status quo in China. In order to establish a nationwide ETS in the future, adequate coordination between central and provincial governments is required in the initial stage. Provincial governments should build an exchange and cooperation mechanism based on relevant information from their own pilot programs, seeking a common ground while recognising differences. Central government should collect such information and unify all of the key issues about the C trading scheme as soon as possible.

Lessons and experiences from existing ETSs worldwide
The EU ETS is now characterised by unexpectedly low prices and reduction efforts (Brink et al., 2016). This is due to the decline in the economic growth rates and the growth outlook of the EU from the 2010s. Strong regulatory and subsidy policies in some European countries have contributed as well (Brink et al., 2016). As De Perthuis and Trotigno (2014) stated, the EU ETS was 'undermined variously by the weakness of its regulation, an undesirable overlap with other public policies and the far-reaching economic and financial crisis that caused the market price of allowances to plunge'. Notably, one important aspect of the EU experience is the interaction of the ETS with other policies that also aim to reduce emissions. This can in turn help to reduce efforts required to stay within the cap prescribed by an ETS. Flues et al. (2014) assessed the effects of scenarios for electricity demand, and interactions between the EU ETS and other renewable energy support policies. As has been suggested, there can be severe and high-cost consequences of policy interaction at a time when the aggregate electricity demand is low. Consequently, fixed renewable energy support policies result in a situation where ETS permit prices are more sensitive to changes in economic activity.
In a trading scheme, a designed future trajectory of caps can assist market predictability. Meanwhile, a fixed trajectory of caps may result in the program requiring unexpectedly larger mitigation efforts, at higher than expected prices, when the underlying emission growth is higher than expected, or unexpectedly lower efforts and prices when underlying growth is weak. If China's cap were set according to an intensity standard, the economic growth related uncertainty about permit prices could be reduced. Hübler et al. (2014) assessed the policy design options for a Chinese ETS while considering the uncertainty about future growth. As suggested by their modelling results, higher economic growth in China increases the mitigation costs under the intensity target for 2020. Meanwhile, under a fixed emission cap for 2030, the results are sensitive to predictions of economic growth in China.
There are distinct views on whether unexpectedly high or low prices are an issue to be rectified or not. Some suggest that a hybrid approach between price and quantity control is optimum (Convery and Redmond, 2007). This could be achieved through a combination of a price floor and a price ceiling. California's trading program has a price floor implemented by reserve prices at the auction, and a price floor was planned for the Australian scheme before it was abandoned in 2013. Most ETSs have an implicit price ceiling in the form of a compliance penalty. Given the large uncertainties about China's future underlying emissions trajectory, the effects of other policy measures and the response of the economy to a C price can be made for price corridors (price floors and ceilings) in a national ETS. Another approach is to adjust the caps in light of the market, economic growth, and technological developments. For example, Australia had a system of 'rolling caps', in which a trajectory of caps was fixed for the first five years, with indicative caps beyond this time, with an independent authority that would advise the government in setting the cap each year beyond that (Jotzo and Betz, 2009).
South Korea, as a close neighbour of China, can also provide information relevant for establishing a national ETS. Park and Hong (2014) discussed the consequences of low-price electricity controlled by the government in the implementation of a Korean ETS. However, the strictly regulated electricity price blocks the cost pass-through not only in the electricity trading market but also in the C market. This leads to the suggestion that intensity measures apply exclusively to the power sector, while an absolute emission cap applies to the remaining sectors yet to be regulated. In order to achieve the market efficiency even in the presence of market imperfections and price and quantity controls in the electricity market, Kim and Lim (2014) recommended a particular policy mix. They argued that a cap-and-trade regulatory system for indirect emissions combined with relative (intensity) targets for direct emissions can achieve market efficiency. This provides price incentives for consumers to adjust their electricity consumption in line with an efficient allocation of other types of energy. While intensity regulation provides mitigation incentives to power producers, the output subsidy effect of a rate-based updating allocation helps prevent the double burden of C costs on electricity consumers.

Conclusion
China has announced to start its nationwide ETS on December 19th, 2017. This move is highly ambitious and demonstrates its commitment as a responsible county. Challenges lie ahead for the regulated energy sector in the context of a host of regulatory policies that may constrain emission growth. This is a great opportunity for China to explore a mechanism where environmental policies are based on and adjusted to market instruments.
From existing experience, it has been shown that setting emissions caps with a national emissions intensity target may bring about specific difficulties. Relevant price developments are uncertain and largely depend on underlying emission growth rates. Implementing reliable systems for MRV of emission remains a major task. A two-level management system for C trading is beneficial. The central government would set emissions caps to level the playing field and avoid in-country C leakage. The central government would be responsible for setting national rules to ensure, among other things, the same rules of coverage and scope, uniform standards of MRV, the allocation of allowances, and the rules of compliance across provinces or their equivalent. Likewise, provincial governments would be assigned to take responsibility for implementing rules, but would be given the flexibility to go beyond national requirements. However, the central and provincial government's two-level management system may not be the best one for China. This two-level management system requires adequate coordination between central and provincial governments.
The coordination of many policies-energy, climate, and broader economic (pricing) and fiscal measures that bear on the energy system-will be essential. The creation of an expert assessment team appointed by multiple agencies could help policymakers identify the interaction between an ETS and existing policies, and recommend ways to avoid redundancy. An ETS broadly supports China's goals of reducing its environmental footprint while upgrading its production structure and reduc-ing dependence on trade-exposed and energy-intensive industries. A sustained effort to develop an ETS in China is likely to deliver greater national and global benefits in the future.