key: cord-0779134-tvc39eg7
authors: Yang, Ming; Yu, Xin
title: China’s rural electricity market—a quantitative analysis
date: 2004-01-22
journal: Energy (Oxf)
DOI: 10.1016/j.energy.2003.12.002
sha: b1758122645fe23b60639235c9400e99485739e9
doc_id: 779134
cord_uid: tvc39eg7

The objective of this paper is to quantify the development of the rural electricity market at county level and below in China. A sectorial energy demand analysis and forecasting model was developed to analyze six Chinese provinces with different economic backgrounds. Historical data for over 20 years were collected on rural economic development, households, population, per capita income, community infrastructure development, capital investment, electricity consumption, output values in agriculture sector, and township and village enterprises (TVEs). This paper concludes that by 2010, annual electricity demand will increase at a rate between −1.40% and 15.60% (depending on the sectors and provinces). It also recommends a preferred order for future rural electricity investment: Jiangsu, Hebei, Henan, Shaanxi, Liaoning and Xinjiang, i.e. from the most to the least developed provinces, if the investment objectives are to find the best market return and the greatest impact on rural market development.

Over the past few decades, China has successfully implemented economic system reforms in rural areas, and as a result, the rural electricity market is expanding quickly. From January 1998 to September 2003, the Chinese government invested 380 billion Yuan (US$46 billion) in urban and rural electricity market development [1] . Its goal was to stimulate domestic demand and push economic development in China, given that about 950 million people (75% of the country's population) are living in rural areas.

However, very little of China's rural electricity market is known to the public due to a lack of research and a rapid development of the market. Many questions remain unclear, such as ''what are the shares of electricity consumption in rural areas?'', ''how and where will these parameters change in rural China?'', ''how much electricity will a rural household consume?'', ''which part of China will incur the greatest rural electricity expansion by 2010?'', and ''what is the priority order of investment in rural electricity markets? The objectives of this paper are to establish a model to identify and assess rural electricity market development in China as well as answer the above questions.

In this research, we collected a large amount of rural, social and economic data for six provinces in China. We grouped the provinces according to their economic and geographical conditions. Jiangsu and Liaoning were selected and grouped together because (1) they both have good economic development conditions in rural areas; and (2) they are both located in the eastern coastal area. Liaoning is in northeast and Jiangsu in east China. These two provinces represent China's well-developed areas. Hebei and Henan, located in central China, are grouped together to represent China's mid-developed provinces. Finally, Xinjiang and Shannxi, located in the western region of China, represent China's least developed provinces. These three groups will also represent six electricity markets that the Chinese government plans to establish for power retail competition by 2005. We forecast electricity demand in the rural areas of the six provinces and tried to identify the best investment market in terms of high growth rate of electricity demand and greatest impact on rural electrification and economic development.

The paper is comprised of five parts. In Section 2, we present literature reviews on the electricity market. This includes brief assessments of China's electric power sector and provides visions of the rural electricity market that leads to the focus of this study. Our model, methodology and approaches are discussed in Section 3. In Section 4, we analyze data, describe variables, and state scenarios used in the study. Section 5, the key section of this paper, presents our results. Section 6 shows the weakness and limitations of this research and directs further research in this topic. Conclusions and recommendations are presented in Section 7. Finally, we have included a summary of electricity market information for the six provinces in Appendix A. The information presented in this paper will be useful to policy makers in China and other developing countries when planning for rural electrification and electricity full retail competition.

Very few articles and documents are available in detailed quantitative analysis for the rural electricity market. Historical papers in this field mainly focus on electricity market reform, urban electricity market, and qualitative analysis. Yang [3] presents an international literature review on rural electricity market development with case studies for the USA, Thailand and Lao People's Democratic Republic (Lao PDR). The paper summarizes the following:

Investing in rural electricity market in general is not financially viable but economically sound and socially friendly. Private sectors have little incentive to participate in rural electricity market development. Instead, governments should initiate and/or invest in rural electricity development programs.

(ii) The main objectives of the government in rural electricity market development include poverty reduction, rural economic development, sustainable development in rural areas, and mitigating the migration from countryside to city.

Yang and Yu [2] reviewed and examined the evolution of China's power sector for the period 1949-1995. The paper covers the areas of institutional development and reform of China's power industry, power shortages in the 1980s, power supply by different resources, capital investment structure, tariff policies, demand side management, energy efficiency policies and practices, laws in the power industry, and environmental impact from power generation.

Between 1995 and 2001, two additional hot topics have been added to China's power literature: (1) continued reform of the power industry aiming at establishing a full retail market for power sector competition; and (2) rural electricity market development. Yang [3] reviewed the Chinese government strategies during 1993-2001 on power system reform towards a market based power system. The paper summarizes that China's rural market has been initially formed and can provide 98% of the rural households with electricity by 2001.

In 2002, the national government (the State Council) issued a national policy document (Document No. 5, 2002) and outlined four key issues in China's power reform [4] . Rural electricity market development is one of the key issues in China. The Chinese government also established ''Four Principles of Facilitations'' to the development of the rural market. These are (1) facilitating economic development and poverty reduction; (2) facilitating the merger of urban and rural power systems (any separation is not allowed); (3) facilitating power supply cost reduction and alleviating farmers' financial burdens; and (4) facilitating further power system reform in China.

China's current electricity situation can be described in three key statements: (1) short of power supply; (2) electricity tariff reforming; and (3) electricity market development. The State Electricity Regulatory Commission said that power supply and demand in 2002 was generally balanced. In 2003, the demand exceeded the supply and 18 provinces, autonomous regions and municipalities (including Shanxi, Jiangsu and Shanghai) faced severe power shortages last summer that affected local economic and social development, particularly rural households' daily lives [5] . The Commission attributed the shortages mainly to (i) the lack of electricity market analysis, (ii) incorrect power demand forecasting, (iii) the slow investment in power supply, (iv) climate change (hot summer), and (v) higher production by enterprises after the severe acute respiratory syndrome (SARS) outbreak. China's policy makers are pondering responsive market strategies, hoping to accelerate electricity facility construction and balance power supply and demand by ''breaking monopoly and introducing competition'' during the coming years.

Following the splitting up of the State Power Corporation at the end of 2002, China plans to create six regional competitive power markets within 3 years in a determined move to break up the traditional province-based electricity monopolies. These comprise: north, northeast, east, west, central and south electricity markets. When established, the markets will be able to sup-port a power plant to sell electricity within the local market as well as export to other markets as long as the transmission line capacities are sufficient. Similarly, an end-user will be able to choose a power supplier either inside or outside of his own market system. This is called a ''Full Retail Competition Market'' in China. Evidently, rural electricity market development plays a very important role in China's power reform.

By the end of 2002, China still had about 20 million poor people, most of whom could not access electricity. According to the China Statistical Press [6] , one rural family has 4.25 people. Thus, 20 million rural residents would be equivalent to about 4.7 million rural households. If US$3300 is required today to electrify one rural family [3] , about US$14.6 billion is needed. Developing the rural electricity market, providing power for these people and raising them above the poverty line by 2010 (with limited funds) is a great challenge to the Chinese government and the power industry.

In the coming decade, China's rural electrification has multiple tasks. The first task should be rural electricity demand forecasting and rural market development planning. The results will assist the government in arranging capital and formulating policies. The second task should be the implementation of institutional reforms to strengthen the rural electrification results. In the long run, the reforms will encourage full retail competition in the rural electricity market. The third task would be the continuation of the ''Brightness Program'' and the ''Developing Rural Power with Wind Program''. These programs are very important to electrify remote households with renewable energy technologies, especially in the western region of China.

In this study, we focused on electricity market analysis and power demand forecasting for the rural areas-county level and below. A sectorial electricity demand model was developed. On the basis of available data, we divided the rural electricity market into several sectors: township and village enterprises (TVEs), agriculture and other family business, and households. Value added, energy intensity and market share of the economy are the main variables in the study. The model and variables are discussed in detail.

The theoretical study of electricity market started in the middle of the 20th century with the development of computer science. However, progress in theory and practice was slow during the initial period. It was not until the 1980s that the theoretical study of medium-to-long-term electricity demand forecasting began. A series of forecasting methods, such as top-down econometric modeling and bottom-up end-use accounting modeling, was successively developed and widely accepted in the electricity demand forecasting of power systems [7] . Having reviewed the literature of energy modeling, Lin made an analysis of energy demand forecasting for China by using an econometric model. The article states that econometric models using GDP and electricity tariff as main driving variables may have at least one disadvantage. Since the economic variables employed in the econometric model are likely to be endogenous, estimating electricity demand by a single equation may produce simultaneous bias and lead to unreliable forecasts. There could be many other factors that require proper attention in determining electricity demand in China's rural area. An important one, for example, is the climate. Dry climate in a year means that more electricity is needed for irrigation. Cold or gloomy days mean that more electricity will be generally used for heating and lighting. Electricity demand also varies according to the time of day. Therefore, the demand for electricity in rural areas depends on climate changes and peak demands during cold and hot seasons. However, it is difficult to incorporate such climate change factors in an econometric model.

For our research, we used SEDA 2 V2, a well-developed mathematical computer model for sectorial energy demand analysis. We developed the first version to analyze the strategies of 10 Asian oil importing countries and economies during 1992-1996 [8] . In the current study, to describe energy consumption in rural households and capture the variable of ''population'', we updated the model with a special module. In this section, we present part of the mathematical equations of SEDA 2 V2.

The total electricity demand (ED) in the rural electricity market is the sum of electricity demand for all sectors in a rural area. Mathematically, it can be expressed as follows:

where ED t is the total electricity demand, ED it is the electricity demand in sectors not including rural households

, ED Ht is the electricity demand in rural households in period t.

Eq. (1) can be further formulated as follows:

where VA it is the value added by sector i (i ¼ 1, 2,. . .n) at period t (t ¼ 1, 2, 3,. . .m) (billion 1999 Yuan); EI it is the electricity intensity in sector i (i ¼ 1, 2,. . .n) at period t (t ¼ 1, 2, 3,. . .m) (GW h/million Yuan); S it is the economic shares of the rural economy in sector i at period t (%); GDP t is the gross domestic product of rural economy at period t (billion 1999 Yuan). Since rural GDP is not available, we use rural economic output in this study; NH t is the number of households at period t (t ¼ 1, 2, 3,. . .m); EIH t is the electricity intensity in rural household at period t (kWh/per household); INH t is the inverse of the number of people per household at period t (household/number of people); POP t is the rural population at period t (million persons). By differentiating Eq. (2), we derived

where dEC t is the electricity demand change in period t in rural market, P n i¼1 ½EI it  GDP t  dS it is the electricity demand change in rural market in period t due to structural effect, P n i¼1 ½S it  GDP t  dEI it is the electricity demand change in rural market in period t due to sectoral effect, P n i¼1 ½EI it  S it  dGDP t is the electricity demand change in rural market in period t due to activity effect, EIH t  INH t  dPOP t is the electricity demand change in rural households in period t due to population change in rural area, EIH t  POP t  dINH t is the electricity demand change in rural households in period t due to the change of number of people per household, IN H t  POP t  dEIH t is the electricity demand change in rural households in period t due to the change of electricity intensity per household.

On the basis of historical data and scenarios, we used the above model to analyze the rural electricity market year by year from 2001 to 2010. In determining the changes of the parameters (dS it , dEI it , etc.), we analyzed the historical values of the variables and used scenario analysis on the basis of China's 10th Five-Year Economic and Social Development Plan, and interviewed with Chinese government officers on social and economic development policies. In the next section, we will discuss the data, variables, and their processes for fitting the model.

A panel data set including six provinces over the period from 1978 to 2000 was constructed from various governmental sources [9] [10] [11] [12] [13] [14] [15] and on-site interviews with many experts in China. In addition, we projected data for the period 2001-2010 for electricity market analysis. In the following example, we briefly describe the methods used in preparing several historical data in the research.

The rural economy was divided into five sectors: TVEs, agriculture, county level industry (industry), household, and others. Given that we only had output data for the county industrial sector in 2000, we estimated industrial outputs below county level and presented the results between 1978 and 1999 by using provincial GDP for the period. Since we did not have any output data for other sectors, we simply assumed growth rates and projected the electricity increase in the period 2000-2010. In the following, we first analyze historical electricity market.

Electricity consumption by rural households, rural irrigation, agriculture, and TVEs in each of the six provinces for the period 1978-2000 is collected to analyze the rural electricity market.

On the basis of the research objectives and the available data, we used the primary variables defined in Table A .1 for this study.

We collected data of 2001-2005 from the Chinese government [16] . The following list presents the main collected data.

(i) Yearly national economic growth rate: 7%. This scenario will guide the assumptions of TVE output and other sectors' output in the six provinces. (ii)

Creating new jobs and transferring rural residents to urban residential areas in 5 years: 40 million. This information tells that rural population might keep decreasing or stable in the future years due to migration. During a field trip to China, the prime author interviewed a number of Chinese experts and government officers regarding economic and social development trends for the selected six provinces 2 . The following information was collected from the trip:

Electricity intensity for rural irrigation will increase slightly after rural electrification. Lower electricity prices reduce labors and increase machines in agriculture production. (ii)

Electricity intensity for TVEs industry will continue to decrease because TVEs will change their structure from high electricity intensity industries to low electricity intensity industries. (iii)

Electricity intensity for ''other sectors'' will decrease slightly because of the electricity conservation effect and increasing outputs from newly developed low electricity intensive enterprises. The number of people in a rural household will slightly decrease at the trend during 1995-2000 due to the government policy of birth control, economic development in rural areas, urbanization and migration. Annual rural population growth rate will remain low. (vi) Net rural per capita income growth rates in real prices: Jiangsu: 7-8%; Liaoning: 5%; Henan and Hebei: 4%; Shaanxi: 6-7%; Xinjiang: 4-5% 3 . (vii) Inflation rate may keep between 2% and 5%. (viii) The growth rates of TVEs will be the same as those of individual provincial industries. (ix)

The Chinese government will continue to invest in rural electrification. During 2002-2004, the government will invest an additional 100 billion Yuan.

The government's rural network investment during 1998-2000 (189 billion Yuan) did not have significant impact on electricity consumption of rural agriculture production because the objective was to retrofit the power network for household electricity supply only. However, reduced tariffs will in some way stimulate electricity consumption in agriculture production.

(xi)

Electricity intensity in TEV and other sectors might decrease due to electricity conservation and business structure change. However, it will be at a slower rate than during 1995-2000 because electricity conservation will be more expensive. (xii)

Rural market reform will continue to aim at full retail competition. In the first section, we mainly considered the historical data of 1995-2000 to set growth rates. In the second and third sections, we took into account both historical trend, future scenarios and findings.

In order to keep the article within a reasonable length, we have described the case of Jiangsu in detail and combined the discussion and results of the other provinces. Table A .3 in Appendix A also presents the projected results of the future electricity market by sector in Jiangsu:

Electricity markets in all sectors will continue to grow; however, growth rates will decrease over the years. 

In addition to Jiangsu, we have quantitatively analyzed electricity markets for Liaoning, Hebei, Henan, Shaanxi and Xinjiang. The analysis for the five provinces followed the same procedure as that of Jiangsu. Tables A.3 Electricity market development ranked by growth rates from the highest to lowest are as follows: Xinjiang, Hebei, Jiangsu, Shaanxi, Henan and Liaoning. The fact that Xinjiang might have the highest growth rate in electricity demand during 2000-2005 is due to its historically small base of electricity consumption. Hebei ranked second in both electricity quantity and growth rate, and Jiangsu ranked third in growth rate. This indicates that the electricity market will be reasonably sustainable and stable in both Hebei and Jiangsu in the near future (see Fig. 2 ). Table A .2 and Fig. 3 comparatively highlight the analysis results of electricity markets for the six provinces. It is observed that Jiangsu, Hebei and Henan have good rural electricity development potential with both high volumes of electricity demand bases and fast growing rates in the coming 5-10 years.

Among the strengths of the model and the study include conciseness, flexibility, and understandability. The model is developed on the basis of basic theory of energy economics. The study approach can be operated by a professional if he has experience in running computer models and undertaking such energy market analysis assignments.

Since SEDA 2 is a bottom-up accounting model, there are a couple of weaknesses or limitations in the research. A bottom-up accounting model requires a large amount of data. If more detailed data is available, the energy sector can be further broken down, and more accurate model results can be achieved. However, data shortage is one of the main limitations in China. Household business income and county industry output, for example, are not statistically avail- able. We had to estimate the data. In addition, due to the reform of the statistical system and a change in methodology, some historical data are not consistent throughout the history. For example, the statistical methodology for rural population and economic sectors in Liaoning has been changed in the early 1990s. Many people living in rural areas have been statistically treated as urban residents. As a result, the rural population according to the Chinese official statistical data has been decreasing very quickly. By 1999, there were about 4.3 million people in Liaoning's rural areas. This figure may not be true. As another example, energy consumption by county small industries has been statistically presented in the industry sector. In recent years, the State Power Corporation has treated electricity consumption at county level and below as rural electricity consumption.

A further limitation of the bottom-up accounting model is that it does not explicitly apply the variables of energy price and income. However, these variables have been acknowledged implicitly in the model. For example, a survey on willingness to pay for electricity at different prices and different income levels has been undertaken in the research and the information has been incorporated in setting the growth rate of household energy consumption.

The model can be used to undertake detailed cross-sectorial analysis of similarities and differences among provinces. Part of the mathematical formulas of the model for such analysis was stated in Yang [8] . However, conducting such analysis might become the scope of another paper 6 .

Nevertheless, the main significance of this study is that we used a computer-supported model to analyze the electricity market and forecast electricity demand for the Chinese rural areas. The availability of this type of analysis is scarce in the history and literature. 6 The impact of China's rural economic reform on rural sectoral energy consumption-an article in preparation. 

There is a great electricity market development potential in China's rural areas. Even if the Chinese rural residents are far above the poverty benchmark, their electricity consumption is very low. For example, in Jiangsu-the most developed province among the six ones, the average per capita income of a rural resident was more than 4300 Yuan/year/capita (US$530) in 2000. If there are four people in a rural family in Jiangsu, the family income will be 17,000 Yuan/year (US$2100/year) 7 . With this income, the family can purchase all the basic family electric appliances such as a TV, a refrigerator, a washing machine in 1 or 2 years. However, our findings show that electricity consumption per family in Jiangsu rural area was about 1.3 kWh per day. This amount of electricity can only run a refrigerator. It is evident that electricity demand by rural households will be very high. After retrofitting the rural power network (rural electrification) and cutting down the rural electricity prices, electricity consumption in rural households will increase quickly.

In the agriculture sector, electricity consumption will also increase during 2000-2005, even though the sector has declined in electricity consumption in some provinces. After the mass investment in the rural power network since 1998 and rural institutional reform, electricity tariffs in China's rural areas have been reduced and this has facilitated electricity substitution for labor and diesel in agriculture production.

TVEs and county industry have been increasing consumers in the rural electricity market. Given that TVEs and county industry will be the most important economic sectors in the rural economy between 2000 and 2010, electricity demand will continue to grow. The average annual growth rates of TVEs between 2000 and 2005 in the six provinces are in the range from 2.87% in Liaoning to 12.60% in Xinjiang, while the rates for county industry are between 3.00% in Liaoning and 0.90% in Jiangsu.

Electricity demand for other sectors will continue to grow but the shares are not significant in the near future. The reason is that the service industry in rural areas has not been developed well and the shares of electricity demand in total are small, ranging from 3.90% in Henan in 2005 to 13.90% in Xinjiang in 2005 (see Table 1 ).

On the basis of the above electricity market analyses, we would recommend the preferred future rural electricity market investment in the following order: Jiangsu, Hebei, Henan, Shaanxi, Liaoning and Xinjiang. This implies the following order of investment for the development of six national electricity markets: East and South China electricity markets, North and Northeast China electricity markets, and Central and Western electricity markets. 

China's potential of power blackouts. Investment and Construction News

China's power management

China's rural electrification and poverty reduction

The reform is not just for reform. China Power News

China faces power shortage

Beijing: China Statistical Publishing House

Structural change, efficiency improvement versus energy demand forecasting-case study of China's power sector

Beijing: China Statistical Publishing House

Beijing: China Statistical Publishing House

Beijing: China Statistical Publishing House

Beijing: China Statistical Publishing House

Beijing: China Statistical Publishing House

Beijing: China Statistical Publishing House

State Power Cooperation-SPC. China rural electrification integration statistical book

China People's Congress. The planning outline of China's 10th five-year national economic and social development. A government policy paper approved by the Fourth Meeting of the Ninth People's Congress of China

Acknowledgements are due to the anonymous referees for their excellent comments on the previous versions of the article. The authors are indebted to the editor-in-chief of Energy-The International Journal for his constant encouragement and advice on refining this article. The