Efficient saving targets of electricity and energy for regions in China

Efficient saving targets of electricity and energy for regions in China

Yao-Chun Lee

, Jin-Li Hu

, Chih-Hung Kao

a

Department of Marketing & Distribution Management, Ching Yun University, 229 Chien Hsin Rd., Jung-Li 320, Taiwan

b

Institute of Business and Management, National Chiao Tung University, Taiwan

c

Bureau of Energy, Ministry of Economic Affairs, Taiwan

a r t i c l e

i n f o

Article history:

Received 18 December 2007 Accepted 1 January 2011 Available online 11 April 2011

Keywords:

Sustainable development Data envelopment analysis Energy consumption

a b s t r a c t

This paper computes the three major types of efficient electricity, coal, and gasoline oil savings for 27 regions in China during the period 2000–2003. The data envelopment analysis (DEA) with a single output (real GDP) and five inputs (labor, real capital stock, coal consumption, gasoline oil consumption, and elec-tricity consumption) is used to compute the energy-saving targets of each region for each year. The effi-cient energy-saving ratios of each region in each year are obtained by comparing the actual energy inputs to target energy inputs. Our major findings are as follows: 1. The east area contains most of the efficient regions with respect to the three major types of energy in every year during the research period. 2. The east, central, and west areas have 2000–2003 average target saving ratios of coal consumption at 18.58%, 44.00%, and 59.80%, gasoline consumption at 13.43%, 22.70%, and 45.04%, and electricity consumption at 8.55%, 16.42%, and 43.70%, respectively. 3. Compared to the cases of gasoline oil and electricity, coal con-sumption saving is China’s most urgent task.

Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction

Energy changes and transformations make things happen. We buy energy, sell energy, eat energy, waste energy, talk a bit about conserving energy, and fight over energy. Energy, a motive force of economic activity, includes all natural resources that can be used after refining. International Energy Outlook[20]predicts that world energy consumption will increase by 60% from 1999 to 2020. Energy demand in developing Asia is projected to more than dou-ble by 2020.

Overusing energy will cause energy shortages, energy crisis, the price of energy to go up, and environment pollution. Production costs of energy rise, and this raises manufacturing costs. For the consumer, the price of energy goes higher, leading to reduced con-sumer confidence and spending, higher transportation costs, and general price increases. In particular, environment pollution endangers an organism’s health and life indirectly through the food chain. Therefore, energy saving has been a crucial issue for sustain-able development. Before new and substitute fuels become avail-able, energy saving is a must in order to make economic growth possible.

The causes of rapid Asian economic growth and its sustainabil-ity have generated considerable debates since the early 1990s (e.g., [2,4,10,24–27,40,43,44]). China, India, and other developing coun-tries are considered the largest energy consumers and are also the

largest emitters of greenhouse gases. As such, they should be in-volved in the efforts to solve these global problems[21].

China has abundant energy mines, but the per capita usable vol-ume of energy is relatively low. Kambara [23] and Dorian [9] showed that the aggregate demand for energy increased corre-spondingly, yet the aggregate supply of energy was relatively insufficient. The inefficient energy use results from uneven mineral distribution, unbalanced regional development, and insufficient infrastructure. In order to satisfy sustainable economic develop-ment, social advancedevelop-ment, population growth, and increased en-ergy demand, the enen-ergy supply must suffice the enen-ergy demand. Therefore, how to guarantee steady energy sources forms the en-ergy topics in security, diplomacy, and trade[38].

China’s energy consumption accounts for approximately 58% of East Asia’s (excluding Japan) total energy consumption. All forms of energy are on the increase, and as result energy demand and use are both up. This paper presents the consumption status of the three main types of energy in China: coal, oil, and electricity. In order to avoid repeated calculation, this paper only regarded final consumptions as energy inputs.

First, coal use steadily increased in China until 1995, then declined for a few years, but now continues to rise. Coal consump-tion in China makes up 70% of energy use and China is the biggest consumer and producer of coal in the world. The development and production of the coal industry has provided stability in China’s economic growth, but since 1949, China has suffered mostly from a shortage of coal. China’s coal consumption in 2003 was 1.64 bil-lion tons, but total coal available for consumption was 1.58 bilbil-lion

0142-0615/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.ijepes.2011.01.015

⇑ Corresponding author. Tel.: +886 3 4581196x7513; fax: +886 3 4683994. E-mail address:yaochunlee@gmail.com(Y.-C. Lee).

Contents lists available atScienceDirect

Electrical Power and Energy Systems

tons. A shortage of coal has limited the growth of China’s steel industry, and therefore coal imports there went up to 11.09 million tons in 2003.

The major pollutants generated from coal burning are carbon dioxide and sulfur dioxide. These pollutants cause acid rain and global warming, and cause health deterioration in China’s popula-tion. China’s coal use discharges 19 million tons of sulfur dioxides into the atmosphere annually and impacts 30% of the economy’s territory with acid rain. China is not the only country suffering from acid rain problems. Other Asian countries, such as Japan, Tai-wan, South Korea, and the Philippines have all reported acid rain originating from China’s coal burning pollution.

In summary, coal burning in China is having a significant impact on the physical environment, the population in China, as well as the overall world atmosphere. Due to the rapid increase of health problems, government actions are being taken up to reduce the burning of coal and move toward cleaner technologies and renew-able energies. Zhijun and Kuby[47]enhanced the model by adding investment variables for improving efficiency on coal and electric-ity. They found that these two energy demands in the year 2000 could be satisfied with less cost and pollution than in the supply-side-only results.

Second, China is the world’s second largest oil consumer, be-hind the US, but its oil consumption is growing 7.5% per year, seven times faster than that of the US Growth in China’s oil consumption has accelerated mainly because of a large-scale transition away from bicycles and mass transit toward private automobiles. Conse-quently, by the year 2010 China is expected to have 90 times more cars than in 1990. With automobiles growing at 19% a year, projec-tions show that China could surpass the total number of cars in the US by 2030. Another contributor to the sharp increase in automo-bile sales is the very low price of gasoline in China. Chinese gaso-line prices now rank among the lowest in the world for oil-importing countries and are a third of retail prices in Europe and Japan, where steep taxes are imposed to discourage gasoline use [30].

At current production rates oil is likely to last for less than two decades. In order to deal with more and more oil demand, China imported up to 95.80 million tons of crude oil in the first 8 months in 2006, up 15.3% over the same period of last year[33]. This prob-lem has put a strain on the world’s current oil contracts, and the issue has become so serious that China’s president took a trip to Gabon to secure a deal for oil with Total Gabon. China’s new energy plan reflects Beijing’s concern about the rising cost of energy and the country’s growing dependence on imported oil[34].

China’s expectation of a growing future dependence on oil im-ports has prompted it to acquire interests in exploration and pro-duction in places like Kazakhstan, Russia, Venezuela, Sudan, West Africa, Iran, Saudi Arabia, and Canada. Despite efforts to diversify its sources, China has become increasingly dependent on Middle East oil, as 58% of China’s oil imports come from the region. By 2015, the share of Middle East oil will stand at 70%.

Third, total electricity consumption in China has also increased due to a growing economy and population. China’s electrical power demands have increased, and the areas affected by blackouts in the future will be larger than in 2003. China’s electricity consumption in 2003 was 1903.16 billion kilowatt hours (KWH), and total oil available for consumption was 1903.22 billion KWH. Increasing power demand as the country continues its modernization drive has put immense pressure on power grids in some areas, especially in the relatively developed coastal regions like Shanghai and Guangdong[32]. Increased industrial output, lower prices, and de-mand for high power-consuming appliances such as air-condition-ers are now causing power shortages in 16 provinces. The situation has become so serious that eastern China will have electrical power shortages year round, instead of just in the summer. To cope

with the problem of power supply, China launched a west-to-east power transmission project in 2000, making it one of China’s major strategies in energy development and an important step for devel-oping the western regions.

Steenhof[36]presented an analysis of the effect of changes in the industrial sector on electricity demand, an important economic sector contributing to these above patterns as it consumes nearly 70% of the electricity generated in China. He found that both increased industrial activity and fuel shifts have helped increase industrial sector electricity demand between 1998 and 2002 by using decomposition analysis. Edvardsen and Førsund [11] and Jamasb and Pollitt[22]analyzed the benchmarking of the electric-ity industry in Europe and Northern Europe at the plant level. In summary, by 2020 projections indicate that China will be respon-sible for approximately 16.1% of world energy consumption. There-fore, a potential energy crisis has become a great challenge to the economic development of China.

An economy’s macroeconomic policies generally have two objectives: the creation of wealth and good living conditions for citizens. Gross domestic product (GDP) is commonly used to assess an economy’s wealth, but it does not constitute a measure of wealth without dealing with environmental issues adequately. Although energy saving is mutually beneficial to China and the rest of the world, people may worry that a drastic reduction in energy will hamper economic growth. Therefore, given the limited avail-ability of economically viable alternative energy sources, reducing total domestic energy use without sacrificing economic growth is an important issue for economies all over the world[8]. This con-cept is also called ‘green GDP.’ Green GDP is derived from GDP through a deduction of negative environmental and social impacts. The future will certainly involve conflicts between environmen-tal protection and economic growth. Therefore, energy-saving tar-gets are very important for all economies. Efficient savings not only are feasible under China’s current technology, but they also will not reduce the maximum potential economic output. Energy effi-ciency improvement is the key to sustainable energy management. Therefore, the main interest of this study is to address the issues related to the analysis of the targets of energy saving and the po-tential application and strengths of DEA for regions in China. Dif-ferent from the traditional DEA model which emphasized efficiency, this thesis creates an input-saving index. It is main con-tribution of this paper too. This study can provide additional sug-gestions for the energy policies of China’s economy.

From the perspective of China’s development and political fac-tors, its provinces, autonomous regions, and municipalities are usually divided into three major areas: the east, central, and west. There is a distinct economic disparity between the coastal and in-land areas. Regional economic disparities are due to greater access to world markets, better infrastructure, a higher-educated labor force, and the government’s preferential policies on foreign invest-ment for the east area[41].

To the best of our knowledge, the existing literature of efficient targets of energy-saving ratios does not simultaneously incorpo-rate various types of energy. Hu and Wang[18]also indicated that China could improve its energy efficiency in various regions with-out reducing its potential economic growth. Hu et al.[19]find to-tal-factor water efficiency of regions in China by the DEA. Hu and Kao[16]used the DEA approach to construct environmental-en-ergy efficiency indicators for APEC economies. Färe et al.[12]used DEA to construct an environmental performance index focusing on pollution. In their study, energy is just one part of the inputs that are taken into account. The DEA approach was originally intended for use in microeconomic environments to measure the perfor-mance of schools, hospitals, and the like, and it is also ideally suited to macroeconomic performance analysis. Each region’s tar-get amounts of coal consumption, gasoline oil consumption, and

electricity consumption in each year can be found by comparing their actual consumption to the total-factor efficiency frontier of that year – that is, the efficiency frontier in each year represents the feasible and best performance of China in that year. Therefore, the imposition of an arbitrary saving target with a developed econ-omy’s standard is avoided herein. The efficient saving ratios for each region in each year are obtained by dividing the target con-sumption by the actual concon-sumption of energy.

This paper is organized as follows: Following this section, Sec-tion 2 introduces the data envelopment analysis to compute the energy-saving targets. This section also describes the data sources. Section3presents empirical results. Finally, Section4concludes this paper.

2. Method and data sources

2.1. Methodology of data envelopment analysis (DEA)

DEA is known as a mathematical programming method for assessing the comparative efficiencies of a decision-making unit (DMU). In our study a region is counted as a DMU. DEA is a non-parametric method using linear programming to construct a non-parametric piecewise frontier over the data for an efficiency measurement. DEA does not need to specify either the production functional form or weights on different inputs and outputs. Comprehensive reviews of the development of an efficiency measurement can be found in Lovell[29]. There are K inputs and M outputs for each of these N DMUs.

The envelopment of the ith DMU can be derived from the following linear programming problem:

Minh;k h

s:t:  yiþ Yk P 0;

hxi Xk P 0; k P 0; ð1Þ

where h is a scalar representing the efficiency score for the ith DMU; k is an N  1 vector of constants; yiis an M  1 output vector

of DMU i; Y is an M  N output matrix constituted by all output vec-tors of these N DMUs; xiis a K  1 input vector of DMU i; and X is a

K  N input matrix constituted by all input vectors of these N DMUs. The efficiency score will satisfy 0 6 h 6 1, with a value of 1 indicat-ing a point on the frontier and hence a technically efficient DMU[7]. The above procedure constructs a piecewise linear approximation to the frontier by minimizing the quantities of the K inputs required to meet the output levels of the ith DMU. The weight k serves to form a convex combination of observed inputs and outputs. It is an input-orientated measurement of efficiency.

Eq.(1)is known as the constant returns to scale (CRS) DEA mod-el[3]. This model finds the overall technical efficiency (OTE) of each DMU. The variable returns to scale (VRS) DEA model[1] fur-ther decomposes the overall technical efficiency into pure techni-cal efficiency (PTE) and stechni-cale efficiency (SE): OTE = PTE  SE. In order to pursue overall technical efficiency with energy, this study adopts the CRS DEA model. Furthermore, both output-oriented and input-oriented CRS DEA models generate exactly the same effi-ciency scores, target inputs, and target outputs. However, the re-sults of a VRS DEA model can be drastically changed by shifting from an output orientation to an input orientation.

The DEA approach was originally intended for use in microeco-nomic environments to measure the performance of schools, hos-pitals, and the like, and it is also ideally suited to macroeconomic performance analysis. However, to the best of our knowledge, the existing literature of efficient targets of energy-saving ratios does not simultaneously incorporate various types of energy. For exam-ple, Hu[15]used three air emissions as inputs to compute the effi-cient air pollution abatement ratios in China. Hu and Lee [17]

found the target waste abatement of three wastes for 27 regions in China through DEA. Hu et al.[19]found total-factor water effi-ciency of regions in China by DEA.

2.2. Regional performance evaluation

We take the economic production function that is constructed by data envelopment analysis to analyze regional efficiencies in China. Three major types of energy (electricity, coal, and gasoline oil) are considered in conjunction with the inputs of labor and cap-ital stock (that are normally used in economic efficiency and pro-ductivity analysis) as the total inputs in order to produce economic output (GDP). The target inputs and outputs for a DMU to be efficient can be computed by the DEA approach. The target saving ratios of the regions are then calculated from dividing target consumption by actual consumption.

Labor and capital are two major inputs in production. When measuring a nation’s overall output, gross domestic product (GDP) is commonly used. For example, Färe et al. [13]analyzed the productivity growth of OECD countries, by considering labor and capital as inputs and GDP as an output. Chang and Luh [2] adopted similar inputs and outputs to analyze the productivity growth of ten Asian economies.

The change in income and energy is a two-way relation: First, increasing income deteriorates the environmental condition di-rectly, because waste is generally a by-product of the energy con-sumption and is costly to dispose. Conversely, the growth of income is accompanied by the public increasing its demand for better environmental quality through driving forces such as con-trol measures, technological progress, and the structural change of consumption. GDP and energy should be both taken into account in order to correct a nation’s output.

The following analytical process considers coal, gasoline oil, and electricity as inputs in order to find the target input levels by the DEA approach.

2.3. Data sources

The data of regional labor employment are established from the China Statistical Yearbook. Data of GDP output in each region are collected respectively as stated previously. Real capital stocks in 1996 prices are constructed based on Li’s method[28].1Monetary inputs and outputs such as GDP and capital stock are deflated to 1996 values. From the China Energy Statistical Yearbook, we establish the three types of energy dataset for 27 regions in China (24 prov-inces and three municipalities) during 2000–2003. Note that Chon-gqing became a municipality out of Sichuan in 1997 and in this study its outputs and inputs are included in Sichuan.

From the perspective of China’s development and political fac-tors, its provinces, autonomous regions, and municipalities are usually divided into three major areas: the east area (abbreviated as ‘E’), the central area (abbreviated as ‘C’), and the west area (abbreviated as ‘W’). There is an apparent economic disparity be-tween the coastal and inland areas. Regional economic disparities are due to greater access to world markets, better infrastructure, a higher-educated labor force, and the government’s preferential policies on foreign investment for the east area[41].

There are five inputs and one output in the DEA model to calcu-late the energy-saving targets. The five inputs are capital stock,

1

The capital stock data are not available in the China Statistical Yearbook. In this study, the authors calculate every regional capital stock in a specific year according to this formula: capital stock in the previous year + capital formation in the current year – capital depreciation in the current year. All the nominal values are deflated in 1995 prices before summations and deductions. This thesis finds the initial capital stock (capital stock data in 1995) from the research of Li[28].

number of employed persons, and electricity consumption, coal consumption, and gasoline oil consumption. The only one output is the GDP of a specific region. These include aggregated input and output proxies. Three inputs of energy are treated as the cost of production, and they are China’s three types of most important energy. To avoid the double counting problem, these three energy inputs are for final consumption. The values of monetary inputs and outputs such as GDP and capital are in 1996 prices.

Tables 1 and 2show summary statistics of these inputs and out-puts ordered by year and area, respectively. The east area has the highest GDP, capital stock, electricity consumption, and gasoline oil consumption. The central area has the largest number of employed people and the highest coal consumption. As shown in Table 3, all inputs have positive correlation coefficients with the output – that is, all inputs satisfy the isotonicity property with the output.

This paper uses the software Deap 2.1, kindly provided by Coelli [6], for computing the target inputs and outputs of each region in each year.

3. Empirical analysis

3.1. Regional saving ratios for the three major types of energy Three major types of energy consumption (coal, gasoline oil, and electricity) are considered in conjunction with the inputs of la-bor and capital stock (that are normally used in economic effi-ciency and productivity analysis) as the total inputs in order to produce economic output (GDP). The target inputs and outputs for a DMU to be efficient can be computed by the DEA approach. The efficiency frontier can shift from year to year. DEA calculates the year-specific frontier with regional output and input (cross-sectional) data for each year. The target inputs of a DMU for a certain year are found by comparing its actual inputs to the effi-ciency frontier in that year. By this method, each region’s target amounts of coal consumption, gasoline oil consumption, and

electricity consumption in each year can be found by comparing their actual consumption to the total-factor efficiency frontier of that year - that is, the efficiency frontier in each year represents the feasible and best performance of China in that year. Therefore, an imposition of an arbitrary saving target with a developed econ-omy’s standard is avoided herein. Hu and Wang[18], Hu and Kao [16], and Honma and Hu[14]constructed a total-factor energy-saving ratio index to compute how far away a region’s energy input is from the efficient level. The higher the saving ratio is, the lower the total-factor efficiency will be.

Target Input Saving Ratio_kði;tÞ¼ 1

 Target Inputkði;tÞ=Actual Inputkði;tÞ;

ð2Þ

where it is in the ith region and the tth year for kth input. As Eq. (2) shows, the saving ratio represents how far away a region’s three major types of energy are from the efficient levels. The efficient tar-gets of energy-saving ratios for each region in each year are then obtained by dividing the target energy consumption with the actual energy consumption. The actual value is always larger than or equal to the target value such that the saving ratio will always be between zero and unity.

After the DEA computation,Tables 4–6present the regional tar-get saving ratios of coal consumption, gasoline oil, and electricity consumption during 2000–2003. Moreover, the average regional target saving ratios of coal consumption, gasoline oil, and electric-ity consumption during 2000–2003 are depicted in Figs. 1–3, showing the trends of regional target saving ratios.

3.2. Saving ratios for electricity consumption

The east area has one region with electricity consumption tar-get saving ratios always higher than 20% throughout the research period: Liaoning (06). The central area has two regions with elec-tricity consumption target saving ratios always higher than 30%:

Table 2

Summary statistics of inputs and outputs by area.

Area of China

East Central West Inputs

Capital stock Mean 19,616 8885 6989 (100 million RMB) Std. Dev. 9734 3297 5817 Number of employed Mean 2404 2430 2070 (10,000 persons) Std. Dev. 1395 1418 1806 Volume of electricity consumption Mean 820 471 396 (100 million KWH) Std. Dev. 442 194 245 Volume of coal consumption Mean 6580 7145 3870 (10,000 tons) Std. Dev. 3973 4155 2392 Volume of gasoline oil consumption Mean 180 126 98 (10,000 tons) Std. Dev. 77 71 63 Outputs

Gross domestic product Mean 4002 2196 1385 (100 million RMB) Std. Dev. 2070 910 1258 Notes:

1. The monetary values are in 1996 prices.

2. Data source: China Energy Statistical Yearbook, 2004–2005.

Table 3

The correlation coefficients between inputs and the output.

Real capital stock Labor Electricity Coal Gasoline oil Real GDP 0.81 0.68 0.93 0.47 0.81

Table 1

Summary statistics of inputs and outputs by year.

2000 2001 2002 2003 Inputs

Capital stock Mean 11,647 12,366 13,105 13,943 (100 million RMB) Std. Dev. 8607 8979 9376 9853 Number of employed persons Mean 2305 2291 2334 2374 (10,000 persons) Std. Dev. 1535 1536 1522 1540 Volume of electricity consumption Mean 501 565 611 699 (100 million KWH) Std. Dev. 296 366 377 448

Volume of coal consumption Mean 5396 5624 6166 7077 (10,000 tons) Std.

Dev.

3311 3487 4077 4588

Volume of gasoline oil consumption Mean 125 136 144 158 (10,000 tons) Std. Dev. 64 76 81 93 Outputs

Gross domestic product Mean 2623 2687 2728 2848 (100 million RMB) Std.

Dev.

1833 1889 1935 2055

Notes:

1. The monetary values are in 1996 prices.

Shanxi (04) and Inner Mongolia (05), while Shanxi (04) has target saving ratios higher than 50%. The west area has four regions with electricity consumption target saving ratios always higher than 30%: Guizhou (22), Shaanxi (24), Gansu (25), and Qinghai (26), where Guizhou (22) and Qinghai (26) have target saving ratios higher than 60%.

Table 6andFig. 3show the 2000–2003 average electricity con-sumption saving ratios in each area. The west area always has higher target saving ratios than the others. With respect to electric-ity consumption, the east, central, and west areas are the most, medium, and least efficient, respectively.

During the 2000–2003 period, the average electricity consump-tion target saving ratio in the three areas was stable. However, the average electricity consumption saving ratio of the west area stayed around 40% during the 2000–2003 period, showing no sig-nificant improvement.

3.3. Saving ratios for coal consumption

The east area has two regions with coal consumption saving ra-tios always higher than 30% throughout the research period: Tian-jin (02) and Liaoning (06). The central area has five regions with coal consumption target saving ratios always higher than 30%: Shanxi (04), Inner Mongolia (05), Jilin (07), Heilongjiang (08), and Hubei (17), especially Shanxi (04) and Inner Mongolia (05) with target saving ratios higher than 80%. The west area has six regions with coal consumption target saving ratios always higher than 30%: Guizhou (22), Yunnan (23), Shaanxi (24), Gansu (25), Qinghai

(26), and Xinjiang (27), especially Guizhou (22) and Gansu (25) with target saving ratios higher than 60%.

Table 5andFig. 2describe the 2000–2003 average coal con-sumption saving ratios in each area. The coal concon-sumption saving ratio of the east area is the lowest, of which the central area is the highest. With respect to coal consumption, the east, central, and west areas are the most, medium, and least efficient, respec-tively. Among the three major types of energy, the coal consump-tion target saving ratios are generally the highest, implying that coal consumption may be the most critical task for saving energy in China.

3.4. Saving ratios for gasoline oil consumption

The east area has two regions with gasoline oil consumption target saving ratios always higher than 30% throughout the re-search period: Beijing (01) and Tianjin (02). The central area has two regions with gasoline oil consumption target saving ratios al-ways higher than 30%: Shanxi (04) and Hubei (17). The west area has five regions with gasoline oil consumption target saving ratios always higher than 30%: Guizhou (22), Shaanxi (24), Gansu (25), Qinghai (26), and Xinjiang (27), with Gansu (25) having target sav-ing ratios higher than 60%.

Table 6andFig. 3show the 2000–2003 average gasoline oil con-sumption target saving ratios in each area. The east, central, and west areas have the lowest, medium, and highest gasoline oil con-sumption saving ratios, respectively. With respect to gasoline oil consumption, the east, central, and west areas are the most, med-ium, and least efficient, respectively.

Table 4

2000–2003 actual consumption and target saving ratios of electricity for regions in China.

ID Region Area 2000 2001 2002 2003 Actual consumption Saving ratio Actual consumption Saving ratio Actual consumption Saving ratio Actual consumption Saving ratio 01 Beijing E 384.48 23.45 398.30 17.06 436.00 6.75 461.24 3.43 02 Tianjin E 236.55 12.83 250.47 9.81 281.00 8.86 313.00 3.52 03 Hebei E 809.33 0.00 869.55 23.29 965.08 16.49 1098.99 16.45 04 Shanxi C 506.09 61.25 557.08 62.41 628.83 56.43 731.77 55.23 05 Inner Mongolia C 256.07 36.79 279.68 34.70 320.44 33.27 406.62 29.51 06 Liaoning E 796.53 31.22 809.42 30.10 859.20 28.00 886.88 22.04 07 Jilin C 300.57 27.94 323.36 23.74 344.54 21.52 359.40 17.07 08 Heilongjiang C 397.24 12.98 468.13 11.11 463.02 6.27 503.63 0.00 09 Shanghai E 559.42 0.00 592.99 0.00 645.71 0.00 745.97 0.00 10 Jiangsu E 971.82 0.00 1078.44 3.50 1244.60 4.02 1505.13 3.75 11 Zhejiang E 742.89 13.50 855.29 12.19 1015.84 7.41 1240.35 4.91 12 Anhui C 338.92 0.00 359.62 0.00 389.94 0.00 445.44 2.57 13 Fujian E 403.02 0.00 439.98 0.00 497.86 0.00 585.35 0.00 14 Jiangxi C 209.39 1.65 222.29 0.00 246.56 0.00 299.53 0.00 15 Shandong E 1000.49 0.00 1560.20 0.00 1230.02 0.00 1395.72 0.00 16 Hennan C 717.62 12.85 808.41 3.77 927.56 20.61 1054.64 24.99 17 Hubei C 503.02 12.60 526.03 8.53 567.43 7.82 629.20 5.51 18 Hunan C 406.20 0.00 439.68 0.00 476.00 0.00 546.95 0.00 19 Guangdong E 1334.58 0.00 1458.43 0.00 1687.83 0.00 2031.29 0.00 20 Guangxi E 322.02 27.25 322.02 21.84 356.95 12.35 414.93 12.15 21 Sichuan W 769.87 25.22 866.82 26.30 954.27 22.71 1052.98 19.47 22 Guizhou W 334.76 63.85 449.05 72.51 491.67 66.31 551.07 63.82 23 Yunnan W 317.25 30.66 347.07 32.90 393.46 36.75 409.79 25.03 24 Shaanxi W 314.39 45.69 344.69 42.84 373.86 41.76 421.92 37.23 25 Gansu W 295.34 57.77 306.09 55.98 342.86 57.15 398.33 51.17 26 Qinghai W 115.96 74.25 111.81 67.75 132.67 66.29 158.51 67.87 27 Xinjiang W 182.98 21.38 197.62 20.84 212.24 19.42 234.62 10.57 Area average E 687.38 9.84 785.01 10.71 838.19 7.63 970.80 6.02 C 403.90 18.45 442.70 16.03 484.92 16.21 553.02 14.99 W 332.94 45.55 374.74 45.59 414.43 44.34 461.03 39.31 Notes:

1. Actual consumption is in 100 million KWH. 2. Saving ratios are in percentage terms.

The average gasoline oil consumption target saving ratios of the east and west areas are stable throughout the research period, with the central area decreasing. However, the average target saving ra-tio of the west area is always above 40% during the 2000–2003 periods, showing no significant improvement at all.

3.5. Energy-efficient benchmarks

FromTables 4–6, the five regions in China are found to always have zero target saving ratios of the three major types of energy, implying that their three major types of energy are efficient during the research period. One of these regions is located in the central area of Hunan (18), while the others are located in the east area: Shanghai (09), Fujian (13), Shandong (15), and Guangdong (19). It shows that the above five regions are the benchmark for the three major types of energy-saving ratios.

On the contrary, Shanxi (04), Guizhou (22), Shaanxi (24) Gansu (25), and Qinghai (26) have high target saving ratios of all energy types, implying that these regions are most inefficient among Chi-na. Among them, Shaanxi (04) and Shanxi (24) produce a large amount of coal, and these regions have abundant petroleum. Guiz-hou (22) also has abundant water resources for generating electricity.

3.6. General comments on coal, gasoline oil, and electricity consumption savings

From Table 7, the 4-year average target saving ratios of electricity consumption for the east, central, and west areas are

respectively 8.55%, 16.42%, and 43.70%. The 4-year average target saving ratios of coal consumption for the east, central, and west areas are respectively 18.58%, 44.00%, and 59.80%. The 4-year aver-age target saving ratios of gasoline oil consumption for the east, central, and west areas are respectively 13.43%, 22.70%, and 45.04%.

Our empirical findings show that the east area has most of the efficient regions with respect to the three major types of energy, while the energy consumption and the regional economic growth are out of step in the west area. The east area also has the lowest average target saving ratios for the three major types of energy. This implies that the most-developed east area is using environ-mental goods more efficiently. At the same time, the west area consumed the highest grade of energy, but still could not provide better living standards. This means that the least-developed west area is using environmental goods inefficiently. A better environ-mental performance has been accompanied with economic achievement for the more-developed east area than for the central and west areas.

Comparing to those cases of gasoline oil and electricity, the average target saving ratios for coal consumption are relatively much higher in all three areas. This shows that coal reduction is China’s most urgent task.

According to our data, most of efficient energy use areas locate in China east. We think the reason is highly related the step of China’s economic growth. After an open-door policy in 1978, FDI was promoted in most industries and centralized in coastal regions. In order to attract foreign firms with high technology, Chi-na’s government put most resources to build up the more sufficient

Table 5

2000–2003 actual consumption and target saving ratios of coal for regions in China.

ID Region Area 2000 2001 2002 2003 Actual consumption Saving ratio Actual consumption Saving ratio Actual consumption Saving ratio Actual consumption Saving ratio 01 Beijing E 2720 18.25 2675 10.84 2531 6.75 2674 3.43 02 Tianjin E 2473 41.38 2635 41.66 2929 43.63 3205 42.22 03 Hebei E 12,115 0.00 12,641 61.83 13,739 61.90 14,851 55.91 04 Shanxi C 14,262 92.92 14,856 92.83 18,055 91.59 20,502 89.78 05 Inner Mongolia C 5908 85.75 6265 85.50 6864 84.67 9025 85.35 06 Liaoning E 9582 50.91 9084 58.72 9355 56.92 10,454 59.58 07 Jilin C 4213 74.62 4484 75.09 4664 71.10 5202 62.80 08 Heilongjiang C 5815 68.26 5537 63.82 5543 57.53 6490 0.00 09 Shanghai E 4496 0.00 4610 0.00 4737 0.00 5018 0.00 10 Jiangsu E 8770 0.00 8963 34.42 9663 13.28 10,849 17.88 11 Zhejiang E 5385 31.67 5527 29.66 6595 18.83 7267 14.43 12 Anhui C 5909 0.00 6366 0.00 6679 0.00 7489 38.48 13 Fujian E 2160 0.00 2205 0.00 2711 0.00 3272 0.00 14 Jiangxi C 2469 55.30 2584 0.00 2557 0.00 3089 0.00 15 Shandong E 8698 0.00 11,098 0.00 12,938 0.00 15,166 0.00 16 Hennan C 8725 31.22 9325 18.89 10,333 30.93 11,420 24.73 17 Hubei C 6051 61.06 6096 50.93 6483 48.01 7238 42.92 18 Hunan C 3335 0.00 4100 0.00 4287 0.00 4984 0.00 19 Guangdong E 5890 0.00 6088 0.00 6649 0.00 7910 0.00 20 Guangxi E 2228 9.43 2228 8.45 2133 12.35 2621 13.28 21 Sichuan W 7804 60.46 7386 56.16 8515 22.71 9900 22.74 22 Guizhou W 5146 75.70 4946 77.22 5199 85.10 6794 87.70 23 Yunnan W 3062 43.99 3101 33.28 3556 39.44 4614 48.27 24 Shaanxi W 2766 66.91 3133 54.56 3451 52.13 3961 49.85 25 Gansu W 2480 71.14 2551 67.75 2798 67.76 3219 76.47 26 Qinghai W 522 70.88 642 74.05 620 53.05 675 48.67 27 Xinjiang W 2702 69.30 2734 69.26 2898 67.22 3184 62.53 Area average E 5865 13.78 6159 22.33 6725 19.42 7571 18.79 C 6298 52.13 6623 43.01 7273 42.65 8382 38.23 W 3497 65.48 3499 61.75 3862 55.35 4621 56.60 Notes:

1. Actual consumption is in 10,000 tons. 2. Saving ratios are in percentage terms.

infrastructure in the relatively developed provinces like Guang-dong, Shandong and Fujian. The local government purchased petrochemical manufacturing equipment, pipelines, substation integration and automation systems, and automatic control system of adding pressure station with the mixed coal gas to upgrade the infrastructure and then set up the national science and industrial park [5,45,46]. According to the survey of National Bureau of

Statistics of China[31], the leading industry of the relatively devel-oped provinces focus on the Information Technology (IT), con-sumer electronics, communications, semiconductor, medication, and biotechnology industry.

On the other hand, many foreign firms have advanced capacity such as Surface Mount Technology (SMT) lines. Moreover, they pay more attention on conform the environment certification such as

Table 6

2000–2003 actual consumption and target saving ratios of gasoline oil for regions in China.

ID Region Area 2000 2001 2002 2003 Actual consumption Saving ratio Actual consumption Saving ratio Actual consumption Saving ratio Actual consumption Saving ratio 01 Beijing E 106.60 33.52 138.69 44.35 152.00 37.95 165.22 39.56 02 Tianjin E 112.43 57.02 116.27 55.35 94.76 36.09 106.42 36.51 03 Hebei E 136.44 0.00 141.85 0.35 147.41 1.22 157.00 1.39 04 Shanxi C 88.84 45.93 88.77 44.64 89.23 41.83 89.27 36.80 05 Inner Mongolia C 64.81 38.00 72.10 40.75 79.35 36.64 83.12 29.51 06 Liaoning E 149.47 22.27 235.79 43.42 236.10 32.90 227.94 28.80 07 Jilin C 90.67 41.95 93.61 39.34 96.99 33.14 103.41 17.07 08 Heilongjiang C 244.04 63.54 269.84 63.95 258.57 56.95 310.17 0.00 09 Shanghai E 132.25 0.00 137.33 0.00 160.09 0.00 173.24 0.00 10 Jiangsu E 187.30 0.00 247.71 3.50 293.39 4.02 339.17 3.75 11 Zhejiang E 196.19 16.21 212.87 16.63 231.44 7.41 262.15 4.91 12 Anhui C 68.54 0.00 70.35 0.00 73.90 0.00 76.70 2.57 13 Fujian E 105.11 0.00 106.35 0.00 132.76 0.00 138.66 0.00 14 Jiangxi C 58.46 8.13 60.37 0.00 82.19 0.00 59.63 0.00 15 Shandong E 188.52 0.00 188.92 0.00 176.83 0.00 209.51 0.00 16 Hennan C 120.86 3.97 124.03 3.77 119.50 6.73 121.99 2.66 17 Hubei C 169.17 32.22 185.55 35.91 232.78 37.04 292.86 54.12 18 Hunan C 115.40 0.00 113.70 0.00 134.63 0.00 135.93 0.00 19 Guangdong E 301.16 0.00 324.82 0.00 344.58 0.00 375.04 0.00 20 Guangxi E 65.87 9.43 65.87 8.45 84.37 12.47 116.70 33.38 21 Sichuan W 209.31 28.27 222.10 30.41 236.88 22.71 247.53 24.65 22 Guizhou W 46.46 35.45 47.83 36.45 50.48 35.22 58.94 38.04 23 Yunnan W 90.79 32.71 111.47 46.61 97.60 29.28 106.10 32.80 24 Shaanxi W 103.51 56.98 78.16 42.84 95.00 41.76 105.43 43.28 25 Gansu W 98.41 68.82 103.96 69.39 97.37 64.04 97.82 63.29 26 Qinghai W 16.31 54.59 17.61 52.67 16.00 48.53 17.16 48.67 27 Xinjiang W 101.93 63.72 86.35 56.45 86.70 47.61 91.35 45.77 Area average E 152.85 12.59 174.22 15.64 186.70 12.01 206.46 13.48 C 113.42 25.97 119.81 25.37 129.68 23.59 141.45 15.86 W 95.25 48.65 95.35 47.83 97.15 41.31 103.48 42.36 Notes:

1. Actual consumption is in 10,000 tons. 2. Saving ratios are in percentage terms.

Electricity 0.00 10.00 20.00 30.00 40.00 50.00 2000 2001 2002 2003 year % E C W

Fig. 1. The average target electricity saving ratios in the three major areas of China.

Coal 0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00 2000 2001 2002 2003 year % E C W

Waste Electrical and Electronic Equipment (WEEE) and Restriction of Hazardous Substance (RoHS), and manufacturers have to ensure that the product cannot have the substance like lead or chromium. They bring the concept of the efficient energy use and higher man-ufacturing performance into coastal regions[37,42]. As a result, unbalanced regional development, the gap between advanced re-gions and lagged rere-gions was widening.

Furthermore, based on the Tenth Five year Economic and Social Develop Plan Codes in People Republic of China[35], the China’s economic policy has been trying to upgrade the industry in order to enhance the competitiveness of China’s local firms. China’s government promotes the knowledge economy like the software,

Integrated Circuit (IC) design, digital content and healthcare indus-try[39]. It also wants to change industrial structure toward the service industry such as retail, distribution and financing. They can raise GNP without more environmental goods input. These im-ply that the most-developed east area is using environmental goods more efficiently.

4. Conclusions and discussion

Electricity, coal, and gasoline oil are the three major types of en-ergy that are inputs of industrial production. Finding out the effi-cient targets of energy-saving ratios according to the feasible Chinese production frontier is hence an important academic and policy issue.

This paper computes the efficient energy-saving ratios of 27 re-gions in China during the period 2000–2003. Different from the traditional DEA model which emphasized efficiency, this thesis creates an input-saving index. The data envelopment analysis is used to construct China’s annual production frontiers in each year. A single output (real GDP) and five inputs (labor, real capital stock, coal consumption, gasoline oil consumption, and electricity con-sumption) are taken into the DEA model. The annual production frontier is constructed from 27 Chinese regions in each year. The efficiency scores and target values of three types of energy for each region in each year are hence obtained by comparing to the pro-duction frontier in that year.

Shanghai (09), Fujian (13), Shandong (15), Hunan (18), and Guangdong (19) are found to always have zero target saving ratios for the three types of energy, implying that they produce outputs efficiently with respect to China’s production frontier. On the con-trary, Shanxi (04), Guizhou (22), Shaanxi (24) Gansu (25), and Qinghai (26) have the highest target saving ratios of all energy types.

Generally speaking, the east area performs the best in China with respect to efficient for the three types of energy considered. The east, central, and west areas have the lowest, medium, and highest target saving ratios on the three types of energy.

We also discuss the difference of energy use from regional views. Due to China government’s preferential policies on foreign investment for the east area, it built up the better infrastructure in coastal regions, especially around science and industrial parks. Foreign firms have the more efficient energy use and higher man-ufacturing performance than local ones. The local firms in the most-developed east area learned how to use environmental goods more efficiently in the long term. China government furthers the upgrading of industries toward knowledge economy and the ser-vice industry. It will be helpful to reduce environmental goods using and add value at the same time.

The results will provide policy suggestions for regions in China in order for them to evaluate and identify their policies and pro-grams according to their income levels, and to improve an overall technical efficiency by adjusting their inputs of energy. For exam-ple, in the west areas, based on the experience of the east areas, build up the more sufficient infrastructure and higher manufactur-ing capacity will be useful to improve their efficient energy use.

The average target saving ratios for coal consumption are rela-tively much higher for all three areas, showing that coal saving is China’s most urgent task. China should immediately engage in improving production efficiency and reducing coal consumption as its top priorities. Therefore, China could improve these regions’ energy efficiency by the following policies: 1. Utilizing hi-technol-ogy to transform the traditional industry to reduce energy con-sumption. 2. Replacing the high-consuming energy industry with the low-consuming energy industry. 3. Developing regenerated en-ergy to increase the enen-ergy utilization ratio. 4. Improving the

pro-Gasoline Oil 0.00 10.00 20.00 30.00 40.00 50.00 2000 2001 2002 2003 year % E C W

Fig. 3. The average target gasoline oil saving ratios in the three major areas of China.

Table 7

2000–2003 average annual target saving ratios for regions in China. ID Region Area Electricity

saving ratio Coal saving ratio Gasoline oil saving ratio 01 Beijing E 12.67 9.82 38.84 02 Tianjin E 8.76 42.22 46.24 03 Hebei E 14.06 44.91 0.74 04 Shanxi C 58.83 91.78 42.30 05 Inner Mongolia C 33.57 85.32 36.22 06 Liaoning E 27.84 56.53 31.85 07 Jilin C 22.57 70.90 32.88 08 Heilongjiang C 7.59 47.40 46.11 09 Shanghai E 0.00 0.00 0.00 10 Jiangsu E 2.82 16.39 2.82 11 Zhejiang E 9.50 23.65 11.29 12 Anhui C 0.64 9.62 0.64 13 Fujian E 0.00 0.00 0.00 14 Jiangxi C 0.41 13.82 2.03 15 Shandong E 0.00 0.00 0.00 16 Hennan C 15.55 26.44 4.28 17 Hubei C 8.62 50.73 39.82 18 Hunan C 0.00 0.00 0.00 19 Guangdong E 0.00 0.00 0.00 20 Guangxi E 18.40 10.88 15.93 21 Sichuan W 23.43 40.52 26.51 22 Guizhou W 66.62 81.43 36.29 23 Yunnan W 31.33 41.25 35.35 24 Shaanxi W 41.88 55.86 46.21 25 Gansu W 55.52 70.78 66.39 26 Qinghai W 69.04 61.66 51.12 27 Xinjiang W 18.05 67.08 53.39 Area average E 8.55 18.58 13.43 C 16.42 44.00 22.70 W 43.70 59.80 45.04 Note: Saving ratios are in percentage terms.

portion of coal washing, and building large-scale hydropower sta-tions near coalmines to convert coal conveying into electricity con-veying. 5. Raising the energy price and levying the energy tax in order to improve energy savings.

Acknowledgements

The authors are indebted to Tzu-Pu Chang, Yu-Hsueh Hsu, and seminar participants at National Chiao Tung University for their helpful comments. Partial financial support from Taiwan’s National Science Council (NSC 95-2415-H-009-001) to the first two authors is gratefully acknowledged.

References

[1] Banker RD, Charnes A, Cooper WW. Some models for estimating technical and scale inefficiencies in data envelopment analysis. Manage Sci 1984;30:1078–92.

[2] Chang CC, Luh YH. Efficiency change and growth in productivity: the Asian growth experience. J Asian Econ 2000;10:551–70.

[3] Charnes A, Cooper WW, Rhodes E. Measuring the efficiency of decision making units. Eur J Oper Res 1978;2:429–44.

[4] Chen EKY. The total factor productivity debate: determinants of economic growth in East Asia. Asian Pac Econ Lit 1997;11:18–38.

[5] Chen JG, Huang QH, Zhong HW. The synthetic evaluation and analysis on regional industrialization. Econ Res 2006;6:4–15.

[6] Coelli TJ. A guide to Deap Version 2.1: a data envelopment analysis (computer) program. Armidale, Australia: University of New England; 1996.

[7] Coelli TJ, Rao DSP, Battese GE. An introduction to efficiency and productivity analysis. Boston: Kluwer Academic Publishers; 1998.

[8] de Nooij M, van der Kruk R, van Soest DP. International comparisons of domestic energy consumption. Energy Econ 2003;25:359–73.

[9] Dorian JP. Minerals, energy, and economic development in China. Oxford: Clarendon Press; 1994.

[10] Drysdale P, Huang Y. Technological catch-up and economic growth in East Asia and the Pacific. Econ Rec 1997;73:201–11.

[11] Edvardsen DF, Førsund FR. International benchmarking of electricity distribution utilities. Resour Energy Econ 2003;25:353–71.

[12] Färe R, Grosskopf S, Hernandez-Sancho F. Environmental performance. An index number approach. Resour Energy Econ 2004;26:343–52.

[13] Färe R, Grosskopf S, Norris M, Zhang Z. Productivity growth, technical progress, and efficiency change in industrialized countries. Am Econ Rev 1994;84:66–83.

[14] Honma S, Hu JL. Total-factor energy efficiency of regions in Japan. Energy Policy; forthcoming.

[15] Hu JL. Efficient air pollution abatements for regions in China. Int J Sustain Dev World Ecol 2006;13:327–40.

[16] Hu JL, Kao CH. Efficient energy-saving targets for APEC economies. Energy Policy 2007;35:373–82.

[17] Hu JL, Lee YC. Efficient three waste abatements for regions in China. Int J Sustain Dev World Ecol; forthcoming.

[18] Hu JL, Wang SC. Total-factor energy efficiency of regions in China. Energy Policy 2006;34:3206–17.

[19] Hu JL, Wang SC, Yeh FY. Total-factor water efficiency of regions in China. Resour Policy 2006;31:217–30.

[20] International energy outlook; 2002. <http://www.allbusiness.com/utilities/ 186568-1.html>.

[21] Isakov I. Global energy issue at the G8 Agenda. Int Aff, Moscow 2006;52:21–30.

[22] Jamasb T, Pollitt M. International benchmarking and regulation: an application to European electricity distribution utilities. Energy Policy 2003;31:1609–22. [23] Kambara T. The energy situation in China. China Quart 1992;131:608–36. [24] Kim JI, Lau LJ. The sources of growth of the East Asian newly industrialized

economies. J Jpn Int Econ 1994;8:235–71.

[25] Kim JI, Lau LJ. The role of human capital in the economic growth of the East Asian newly industrialized countries. Asia-Pac Econ Rev 1995;1:3–22. [26] Krüger JJ, Cantner U, Hanusch H. Total factor productivity, the East Asian

miracle and the world production frontier. Weltwirtschaftliches Archiv 2000;136:111–36.

[27] Krugman P. The myth of Asia’s miracle. Foreign Aff 1994;73:62–78. [28] Li KW. China’s capital and productivity measurement using financial

resources. Economic Growth Center, Yale University; 2003. <http:// www.econ.yale.edu/~egcenter/research.htm>.

[29] Lovell CAK. Production frontiers and productive efficiency. In: Fried HO, Lovell CAK, Schmidt SS, editors. The measurement of productive efficiency: techniques and applications. New York: Oxford University Press; 1993. [30] Luft G. Fueling the dragon: China’s race into the oil market. The institute for

the analysis of global security 2003–2004; 2006. <http://www.iags.org/ china.htm>.

[31] National Bureau of Statistics of China. Provincial annual stat data; 2000–2005. <http://www.stats.gov.cn/> [in Chinese].

[32] National development and reform commission (NDRC). Xinhua News Agency; 2003 [June 23]. <http://www.china.org.cn/english/BAT/67733.htm>. [33] National Development and Reform Commission (NDRC). Xinhua News Agency;

2006 [October 4]. <http://china.org.cn/english/2006/Oct/182953.htm>. [34] Oster S. Energy: China plans to curb oil, coal use despite expected demand

growth. New York: Wall Street Journal; 2006. p. 17.

[35] State Council of the People’s Republic of China. the tenth five year economic and social develop plan codes in People Republic of China; 2001. <http:// big5.china.com.cn/ch-15/15p8/2.htm/>.

[36] Steenhof PA. Decomposition of electricity demand in China’s industrial sector. Energy Econ 2006;28:370–84.

[37] Tan ZA. Product cycle theory and telecommunications industry-foreign direct investment, government policy, and indigenous manufacturing in China. Telecommun Policy 2002;26:17–30.

[38] US–China Economic and Security Review Commission. 2004 Report to congress of the US–China economic and security review commission. Washington, DC; 2004.

[39] Wilson EJ, Segal A. Trends in China’s transition toward a knowledge economy. Asian Surv 2005;45:886–906.

[40] World Bank. The East Asian miracle: economic growth and public policy, Washington, DC; 1993.

[41] World Bank. China’s challenges: ensuring growth with equity; 1997. <http:// www.worldbank.org>.

[42] Xie W. Technological learning in China’s colour TV (CTV) industry. Technovation 2004;24(6):499–512.

[43] Young A. Lessons from the East Asian NICs: a contrarian view. Eur Econ Rev 1994;8:964–73.

[44] Young A. The tyranny of numbers: confronting the statistical realities of the East Asian growth experience. Quart J Econ 1995;110:641–80.

[45] Zhang GJ. Upgrade the competitiveness of China’s petroleum equipment manufacturing industry. China Oil Gas 2005;12(4):10–3.

[46] Zhao MS. Achievements and developing strategy of petroleum and petrochemical equipment manufacturing industry in China. China Petroleum Mach 2001;29(2):1–4.

[47] Zhijun X, Kuby M. Supply-side–demand-side optimization and cost-environment tradeoffs for China’s coal and electricity system. Energy Policy 1997;25:313–26.