Making China’s Electricity Markets Work for Clean Energy


As part of the Chinese central government’s power sector reform effort, several provinces have been developing electricity markets. The next stage of this undertaking—design and implementation of “spot markets”—could bring many benefits, including system flexibility to support wind and solar generation, a better-functioning carbon market, rational prices to guide better investment decisions, and corresponding incentives to reduce coal-fired generation. In turn, this can help achieve China’s goals of cost-effectively reducing emissions while maintaining high levels of reliability. However, implementation of these kinds of markets is no easy task—and can stall out (as happened in China more than a decade ago) or result in crisis (of which California in the early 2000s is the best-known example). Here we’ll recap a few practical ideas to reduce implementation risks and avoid major setbacks.

The story so far

The effort to implement electricity markets has focused on the first stages envisaged in the government’s 2015 power sector reform blueprint: the gradual introduction of competition between generators for allocations of operating hours, in the form of annual or monthly contracts between generators and large end users or retailers. This is often referred to in China as the medium-long-term (MLT) market or direct trading and has been evolving at the provincial level. These provincial MLT markets, although incomplete, can be seen as a step forward—at least in the sense that comparatively dirty and inefficient coal generators now have a harder time winning hours than their more efficient counterparts in China’s massive coal-fired fleet. That is, it appears the new markets are helping to change the long-standing and highly inefficient approach to generator dispatch. That approach traditionally guaranteed each coal-fired generator an administratively decided allocation of annual operating hours, largely without regard to relative operating costs, fuel efficiency, or emissions. These MLT markets do little, however, to ease the hour-by-hour inflexibility that contributes to wind and solar curtailment.

Next steps

More recently, the central government has asked several provinces to quickly roll out what it refers to as spot markets. These are expected to comprise day-ahead and real-time markets as well as mechanisms to compensate ancillary services. Few details have been revealed publicly, despite a requirement that some of these markets begin operating by the end of this year. Guangdong province is expected to release a new plan soon.

Such spot markets, if well designed, can serve as the basis for economic merit order dispatch. This should help reduce costs, improve system flexibility, and facilitate clean energy resources. However, design and implementation of spot markets and corresponding dispatch reform are challenging tasks, with many potential pitfalls.

Practical recommendations for spot markets

Much discussion has occurred among policymakers, experts, and stakeholders in China about international experience with these types of markets, including the various ISO/RTO markets in the United States. Many international examples are complex, however, and may be imperfectly suited to China’s situation. In addition, given that policymakers in the U.S., Europe, and elsewhere are debating how to best update policy and market design to support new challenges of renewables integration and decarbonization, it makes sense to think about how Chinese decision-makers might leapfrog the inapt elements of these international approaches. Our suggestions for spot markets in China include:

  • Go for broad cross-provincial market footprints. Broader footprints help foster renewables integration and reduce system costs. Most of the spot market efforts in China are proceeding province by province, although the central government has asked the five provinces in the Southern Grid region to eventually implement a regional market. It would make sense to design regionally integrated spot markets from the outset.
  • Develop long-term regional resource adequacy and reliability planning processes in a way that complements the markets. Even after implementation of markets, planning studies will still play an essential role in evaluating resource adequacy, informing the need for adjustments to market design, and helping to coordinate investments in generation with those in transmission and demand-side resources. The National Energy Administration’s 2016 Power Sector Planning Regulation (in Chinese) is a good basis on which to build.
  • Implement measures to monitor and deal with the risk of inadequate competition in spot markets. A robust market power monitoring and mitigation framework will be important, given the risks of market manipulation by generators in parts of the country with highly concentrated ownership. It may even be worth considering (at least in initial years) conducting spot markets on the basis of estimations of each generator’s operating costs. This could help avoid market manipulation while affording most of the benefits of dispatch based on an economic merit order. In any case, such data on operating costs can be used for monitoring market performance. Guangdong’s proposal to design a system to collect this data appears promising.
  • Transition long-term contracting toward a financial model. The practice of dispatching according to each generator’s annual allocation of operating hours has long been a source of inflexibility. Even in provinces where the MLT market has been gradually replacing that annual allocation, the dispatch centers continue to consider MLT contracts in operational decisions throughout the year. Any such “physical” treatment of contracts is best phased out as spot markets are phased in, to avoid constraining flexibility and hampering the ability of the system to respond in real time to changes in grid conditions such as fluctuating weather. A better model would allow generators and consumers to hedge electricity prices through financial contracts, while the dispatch center focuses on optimizing the system according to hour-by-hour conditions and needs.
  • Look ahead to participation of distributed resources. Demand response, storage technologies, and other distributed energy resources have the ability to provide valuable flexibility services. But markets need to be designed to recognize, reveal, and reward these abilities and services. It is worth thinking ahead in this regard, particularly given the parallel effort in China to unbundle local distribution networks from the grid companies that own and operate the bulk transmission system.

These are complex and difficult topics. There is much room for continued discussion as decision-makers in China and other parts of the world grapple with the challenges of market design, especially in the context of meeting goals for clean energy.

For more on these recommendations and related issues, see this Toolkit of Global Insights. In addition, this month the Lawrence Berkeley National Laboratory released a report outlining suggestions for several of China’s power sector reform pilot programs.  

A version of this blog is also available in Chinese.