Marginal cost pricing for coal fired electricity in coastal cities of China:The case of Mawan Electr

Jo-wrnal of Enrironmental Sciences Vol 15. No. 3, pp 401-413. 2003Aticle ID: I001-0742(2003)03-0401-12 CLC number: X32 Document code: AMarginal cost pricing for coal fired electricity in coastal cities ofChina: The case of Mawan Electricity Plant in Shenzhen CityChinaZHANG Shi- qiu, dUan Yan-xin(Center for Environmental Sciences, Peking University. Beijing 100871, China. E-mail: zhangshu ces. pku. edu.Ahstract: By developing a GDMOD model to eatimate the environmental extemalities associated with electricity generation. this project providesa detailed analysis of the damages and costs caused by different pollutants at varying distances from the Mawan Electricity Plant in ShenzhenChina, The major findings of this study cantal damages caused by electricity production are large and aremainly imposed on regions far away from the electricity plant: (2)air pollution is the most significant contributor to the total damages, andand particulate matter are the three major pollutants with highest damages:(3)the dansed per unit of particulate, NOxad S0, emissions are much higher than pollution treatment and prevention costs. The research results of this project showed that China needsto have a more effective levy system on SO2, and a more manageable electricity tariff mechanism to intemalize the environmental extemalitiesThe results have almo implications for pollution control strategies, compensation schemes as well as emission trading arrangementsKeywords: marginal cost pricing: environmental damages; environmental policy1 Issues and problemsChina is currently in the process of transition from a planned to a market economy. In a market economy, the allocationof scarce resources between competing uses is a problem that is solved through market pricing of resources. A precondition forthe optimal functioning of this allocation process is that the market prices reflect all costs involved in production( Pearce1993).The market mechanism cannot secure an optimal macroeconomic allocation ifreflected in the market price because they are passed on to third parties not involved as consumers or producers of a product( inthe instance of extemal or social costs)China has experienced remarkable economic growth in recent years, China's struggle to meet its enormous energy needsi a key element of the nations development strategy and is perhaps the single most important variable shaping the future ofChinas environment. Given the govemment's plans to continue depending heavily on coal to meet the increasing demands forcoals use in the power sector will be declining. As a consequence, it is expected that China's aipollution will increase substantiallyChina has made progressive steps towards the liberalisation of electricity prices, which has created a pricing structure thatmore accurately reflects the production costs( although not the extemalised environmental costs). However, furtherrat onalise energy prices and reform product markets are still needed. The major problems related to electricity price in Chinaan be summarized as followweo" irst. the current pnce system is the result of a number of incremental ad hoe changes instead of being the product of adesigned long-term price reform program. The electricity pricing system remains too complex and cumbersome in spite ofthe major reforms initiated in 1993. Pricing policies vary by region, supplier, and customer. Furthermore, municipalprovincial. and county governments frequently impose added fees to raise funds for electricity development, and costSecond, there is an internal distortion of electricity pricing. The price for commercial industries, service sectors, andhotels is set too low. Preferential prices for industry and agriculture are also too low. Furthermore, the currentefl ct differentiation in peak -load useFoundation item: The Environmental Economy Program of South East Asia(EEPSEA)and Trans century Training Program Foundation for theTake nts in China by Ministry of Education中国煤化工CNMHGZHANG Shi-qiu et alVol, 15Third, there is a problem with inefficient management in the electricity sector. The moles of the yarious administrativedepartments( agencies)are not clearly defined which is a major obstacle to coordinating efforts between differentdepartments. As a result, there is not a very strong scientific basis behind the calculation and classifications of electricityprices or in the determination of rational profit rateselectricity prices do not reflect the full social cost, such as environmental damages produced by electricitygeneration. This tends to discourage firms from adopting efficient pollution prevention and treatment measuresIn sum, the current price system is still very much a product of central planning. Such a price distortion sendsmisleading signals to producers and consumers which leads to lower investments in altemative energy, improved efficiency inelectricity production, the installation of environmental pollution prevention facilities, and the use of clean inputs( such ascleaner coal with lower sulphur and ash content)2 Definition of marginal opportunity cost( MOC)pricing for this study2.1 Definition of coal fired electricity priceIn this study, the electricity price is defined as the Airm-gate price of eke2.2 Components of the electricity price at firm gaemvironmental and user costs. Therefore, the electricity price( MOC at firm-gate)should include MPC, MUC and MECMPC(marginal production cost)is the cost of production for an incremental unit of electricity( capital investmen andMuc(marginal user cost)is the depletion cost of the coal that is used as fuel. In theory, either the domestic marginalopportunity cost( MoC)price of coal or world market price of coal can be used as the cost of coal, which means the MUC willMEC(marginal extemal cost )is the extermal costs relaled to the production and consumption of electricity. It consists ofMECI,(environmental)costs( damages)associated with electricity consumption. Since the electricity price for this study is the firmprice, the mEC used here inchudes only the extemal costs of production, MECIThus the electricity price at firm gate for this study can be described as:P= MOc MPC MEC MPC MECI(1)3 Methodology usedThe major focus of this study is the estimation of the environmental costs( MEC1)associated with coal-fired electricityimate the possible environmental changes caused by the dierent stresSors. The physical impacts and damages caused bythese changes are assessed and quantified, and finally monetary values are assigned to the damages. The steps involved in thisprocess of MECI estimation are illustrated in Fig.1funetionpollutant, 2rpollutant i,CYonetary tor, EFig. I Process of damage valuation for estimation of MECI4 Develop the gdmod model to estmate the meCi中国煤化工CNMHGN.3Marginal cost pricing for coal fired electricity in coastal cities of China: The case of-...403Since the production processes used in the electricity industry vary little, coefficients of pollution emission and thenironmental damage functions from other models can be used. This allows for the use of computer programs to calculate thedanages and determine the monetary value the damages. Based on the program EXMOD. created for the"New York StateEnvironmental Extemalities Cost Study", we developed a GDMOD, which is a computer model to estimate the environmentalo-is produced by the Mawan Electricity Plant in Shenzhen. The model was developed based on the characteristics of pollutionprduced by coal-fired electricity plantsThe analytical parameters for GDMOD are defined as: (1)the regional focus of this study is Guangdong Provinceady and all data was converted into 1992 values using a discount rate of 12%:(3)the model developed here only considersthe environmental damages produced by electricity generation and does not inelude the extemal costs associated with the4.2 Receptor cell definitionThe model uses spatial units called receptor cells. These are the basic units for calculation. The affected area is dividedinto 139 receptor cells that are categorised based on their distance from the facility site. The four geographic classificationsd are:(1)local community locations in Guangdong within 30 km of the facility site; (2)rest of the region: locations inGuangdong between 30 and 80 km from the facility site;(3)rest of province: locations in Guangdong more than 80 km fromthe facility site;(4)out of province: locations in provinces neighbouring GuangdongEach receptor cell is assumed to be internally undiferentiated vis-a-vis pollution levels, population density and the4.3 Air quality models used in GDMODAir quality models are used to calculate changes in ambient air quality and pollution deposition using stack emissionsdata. This study uses the same air quality models used in EXMOD, though they have been adjustparameters in Guangdong Province. The following major models are used: (1) ISC2LT model: Used for short rangemodelling, to calculate the air quality changes with 50 km of the power plant. This model can be used to calculate theincremental concentration of SO,, NOx, TSP and other pollutants; (2)SCREEN2 model: ft is used to calculate the short-tem average concentration of PM10 for a maximum of 24 hours and NO(an O, precursor)for a maximum of I hour: (3)SLIM2 model: this model is used to calculate annual average impacts at long range( greater than 50 km from the power plant);(4)OLM: It is used to calculate changes in ambient ozone concentrations resulting from power plant emissions of Nox4. 4 Case creationWith this model, it is necessary to specify the facility location( including elevation), production specifications( type ofplant and production process), as well as the characteristics of the receptor cellsthe emissions and concentration amounts are determined based on information regarding the productionprocess, facilities and operation parameters as well as the pollution prevention measures for the Mawan Electricity Plant. Theelectricity generating equipment used at the Mawan Electricity Plant is a pulverised steam boiler with a capacity of 2 x 300MW. The plant uses high quality domestic and imported coal. Pollution prevention measures taken at the plant include theof high stack( 210M)emissions, low NOx emission boilers, electrostatic dust precipitators with 99% dust removal efficiencyand wastewater treatment facilitieFor the receptor cells, the data inputted include: (1)information for each receptor cell, such as place name, latitudelongitude,altitude, area, population, sex and age structure of the population;(2)environmental concentration monitoringdata in each receptor cell(collected prior to plant operation); and (3)meteorological data( joint frequency distributionwina speed and directionOnce the characteristics of the facility and receptor cells have been established, the dose-response functions must bespec fied, By modifying the parameters used for the EXMOD, appropriate functions for the GDMOD can be produced forFor detail, see Robert D. Rowe e al., The New Yor electricity externality atudy, Ocown publications hc. 1995中国煤化工CNMHGZHANG Shi-qiu et alregion being studied. The monetary value for physical damages must also specify before the model can be run4.5 The output of calculation resultsUsing the above mentioned data for the facility, surrounding regions, pollution characteristics and effects, the GDMODwill generate the following information:(1)value of damages based on regional proximity :(2)value of damages for eachmajor type of pollutant;(3)value of damages within environmental category( such as air, water, land, etc. ):(4)value ofdamages per unit of each major pollutant5 Estimation of marginal environmental cost5.1 Identifcation and classifcation of pollution damages produced by Mawan Electricity Plant: potential stressorsand impacts screeningCoal-fired electricity plants emitpollutants that when released into the environment impacts the health and wellng of humans and other forms of life. The major impacts can be summarised as:(1)impacts on human health;(2)impacts on human welfare;(3)impacts on environmental resources;(4)bal changeBased on data collected for this study and on results from previous research, the different stressors can be identifiedwhich was reviewed and categorized in terms of their potential impacts. Through this screening process, each stressor wasassigned one of four classifications. These classifications are used to select the applicable extemalities for the study. Thecriteria used in assigning categories were:(1)the impact ean be mitigated, which has or will be included in calculations ofpollution prevention cost(P, ):(2)the impact is relatively small, which will be deleted from further analysis:(3)there isnot enough scientific information available for a quantitative assessment, which will be qualitatively assessed and analysedseparately;(4)a quantitative assessment can be completed, which subject to be assigned full or partial economic valuation5.2 Estimation of dose-response function and mo5.2.1 Human health effects of air pollutants5.2.1.1 Human bealth ehects of PM10Many studies showed that PM10 is the major factor causing health problems, since they are small enough to enter into theairways of the lungs. Therefore, PM10 is more accurate to indicate the dose for damages. In the model we run, we use PM10instead of TSP. Abbey and others( Abbey, 1993)studied the conversion factor for TSP and PM10, and use a factor of 0.5-0.6. Brook(Brook, 1997)studied data collected over a 10-year period from 19 monitoring stations and also determined thecoefficient for PM10 and TSP to be 0.5-0.6. These studies are consistent with the research findings of the NewYork Study( Rowe, 1995). In this model we use 0. 55 as the conversion factor. ThereforeCPvo =0.55 x CTsPThe human health effects of PM10 include mortality and respiratory disease. Studies showed that the health problems andmpacts related to PM10 includes: chronic bronchitis( CB), respiratory hospital admissions(RHA),asthmaactivities days( RAD), acute respiratory symptoms( ARS), emergency room visits(ERV), and asthma for children. Fo(3)is used to calculate the dose-response of health effects of PM10△D.=Rx△PM10× POP x NIn which, AD, is the annual incremental cases of disease due to pM10; R is the deAPM10 is the annual concentration change of PM10, POP is the affected population; N is the days of PM10 exceedingstandards in one yeThe dose-response coeficient estimation is therefore the key issue for estimating the health eifects. Given the similaritiesve decided to use the dose response functions developed for the New York Study(EXMOD)'. These functions were developebased on the most comprehensive review of related research conducted in the US during the 1990's (Table 1)Sinee some double counting may exist between the indicators shown in Table 2, we adjust the functions in the GDMOD asfollowsThe average days for staying in hospital for a RHA case are 9.5 days( China's health statistical year book, 1996)There have been a number of studie conducted by Chinese researchers( Wang, 1989; 1993; Zhang, 1994; Chu, 1993),anppears to be no significant difference from fnding of audies conducted in other countrie中国煤化工CNMHGMarginal cost pricing for coal fired electricity in coastal cities of China: The case of405ERV= ERV-RHARADRAD-(r9,5d·RHA)-(r·ERV)-r·A4],in which, r refers to the proportion of population over age 18; ARS-ipu=ARS -RADTable 1 Damage functions of health effects of PM10age functionUnH10.1×10-16.9x103425.4×10Mortality( <65) caMe/d330.23x10340.35x103,0x10256.1x103x10case/dI ug/m1.8×103253.3×1050ERVcase/d1g/m3.2x10256.5×10509.7x10·109x10-331.6x104x10-4P3325273RAD(≥180.8x10342.5x10-4Asthma for0.8x10251.6×10-324×103e: L refers to low value; C refers to central value: H refers high value; P refers to probability(%); source: A.D. Rowe, 1995;Schwarz, 19925.2.1.2 Human bealth effects of ozone(AA), minimum restricted activities days (MRAD), and acute respiratory symptoms( ARS). Formula(4)is used to calculatehe human bealth efects of ozoneRx△O3 x POP xIn which, AD, is the annual incremental cases of disease due to PMO: R is the dose-response coefficient case/(d pg/m')40, is the annual average of daily changes in high-hour ozone: POP is the affected populatiethe days of ozoneSince there is no study on human health effects by ozone available in China, we use the functions used in EXMODwhish are shown in Table 2. due to the possibility of double counting, the following adjustments were madeMRAD detd MRAD-AA: ARSJnnd ARS-MRADTable 2 Dome- response functions of human heat ehects o O384×10-63313.7x1063419.0x10-6case/d·ppm3375MRADNotes:Refers to low value C refers to central value; H refers high value; P refers to possibility(%)5.2.1.3 Human health effects of lead and mercuryLead and mercury are emitted into the air with soot produced during coal combustion. Exposure can occur throughbrea hing and ingestion. Epidemiological studies have found that PbB levels can lead to higher rates of hypertension, nonfatalear attacks, nonfatal strokes and risks of premature death for adult men. Efects of mercury exposure are quite complex anddiffic ult to quantify. As a result, we use the damage value per unit of pollutant used in the New York Electricity ExtemalityStud after adjusting for per capita GDP values, to directly calculate the damages( Table 3)Formula(5)and(6)are used to calculate the human health effects of lead and mercury中国煤化工CNMHG406ZHANG Shi-qiu et alVol 15in which, E, is the damages due to health effeets of lead; POP is the population exposed to lead; Cu is the incremental leadlevel produced by electricity plant: R, is the damage function of leadin which, Ea is the damages due to health effects of mercury: C a is the incremental mercury level produced by electricityplant: R, is the damage funr-lion al mercury5.2.1.4 Human health efTects ofTable 3 Damage functions for human health effects of lead and mercuryradiationRadiation can cause various healtheffects, which can be estimated by Health effects of lead $/person'pg/dI 0.531 33 1.614 348.60033Formula(7). The dose received by an Health effects of mercury 3/kgdividualbe used as an indicator of Notes: 1. L refers to low value; C refers to centralH refers high value; P refers toprobability(%): 2. the value here is subjected to adjust by GNP ratio: source: RoweE,=R…rPoP100,(7)19hc:19in which, E, is the value of damages caused by radiation; R is the value of damages per unit radiation; r is the radiationexposure/(person a)due to 1000 MW electricity plant: PoP is the affected populationIt is estimated in the EXMOd model that the incremental radiation produced by a 1000 MW coal fired electricity on loc33×10-3rem/( person·a·1000Mw)5.2.1.5 Homan health impacts of sAir toxics included here are the suspected carcinogenic air emissions such as As, Be, Cd, Cr, Ni and POMs( BaP)ing and ingestion. According to the USEPA's Integrated Risk InformationSystem(IRIS)(USEPA, 1992), the dose response functions of air toxics are:in which, AD, is the incremental case of cancer by pollutant I: AC is the Table 4fumctions o selected toricincremental concentration of pollutant I; RF, is the cancer risk factor for chemicalsinhalation for chemical i (which refers to I incremental concentration exposure Pollutants DuThe value we used of RF, is shown in Table 4( Chen, 1992)5.2.1.6 Valuation for buman bealth electsdcmThere are several different ways to estimate the economic costs of humanRespiratory 0.00024health effects. The Willingness to Pay( WTP),Cost of Illness( COI)andHuman Capital Approaches are three common methods. However, given the Source:USEPA 1992difficulties associated with data and information collection, for this study we use the technology transfer methodology toestimate the human effects in monetary termsUsually, the estimation based on WTP is much higher than on COI( Pearce, 1989: Rowe, 1995).According to researchconducted by the Chinese Academy of Social Sciences, Table 5 The estimation for He nhe(converted by GNP ratio)the monetary value assigned to human life in 1992 forLife valueChina was about 160000 RMB Yuan World Bank, ffectsPopulation (10000 RMI992)1997).In the affectedHPlant, the per capita GDP is higher than the national Mortalityaverage. We therefore adjust the averageMortalitysing GNP ratios and values calculated in othecountries to assign life values for the areas around the Mortality ChildrenMawan Plant( Table 5 and Table 6). Given the Probability weighl. sarguments for and against benefit transfer approachSources: calculated based on Cropper 1991: Fisher 1989: Miller 1989; Mooreell as the value of human life approach, we presentmh中国煤化工CNMHGN,3Marginal cost pricing for coal fired electricity in coastal cities of China: The case offir al results two ways--once with costs of deaths included in total damages and once with deaths left out of the costTable6 The cost estimation for illness( converted by GDPRMB Yuan, 1992UnitTypeCB(adult)RMB Yuan/person 1980 3309 5200 Cropper. 1991: Miller, 1989 WTPRHARMB Yuan/cas110022003300 Viscusi,199lAdjusted COl125 Krupnick, 19sRMB Yuan/HAD(≥18MB Yuan/ca16.5 Abbey, 1993WTP adjusted COIRMB Yuan/d5.3 8.7 Lochman, 1979WTP6.5 Abbey, 1993; Rowe, 1986 WIPRMB Yuan/d2.4 Abbey, 1993: Rowe, 1986F'robability weight,5. 2.2 Damage to crops by so, and acid depositionStudies on the dose-response functions for $O, and acid deposition for agricultural crops have been done in China(Cao1991: Zhang, 1997). Cao et al. focused on Guangdong and Guangxi Province, making their results most appropriate for useto calculate the damages to varous crops. The calculated results are shown in Table 7△V=△Q.·P=PR.·△CsQ,=∑RR·ΔCso;·VIn which, AV, is the losses to crop I: 4Q, is the production loss of crop i: P, is the market price of crop I:R,is thedanage functions for crop i: ACso, is the incremental value of So, at region j: Q, is the total production of crop i at regionj: V, is the total output5.2.3 Material damages by acidTable 7 Damage functions for agricultural loses caused by SO, and acid depodddonAn empirical study conducted byL Pmd平mmbm,m0naa%(pgm3)0.025330.029340.03333various materials, such as covering vegetables%(pm3)0.0380.048340.06333materials, marble, galvanized steel,Source: calculated bavd on Cao ef al. 1991steal. Formula (10) is used to estimatehe cost of materials damages caused by acid deposition∑R·△C.·HHin which, E is the material damages cost by SO, and acid deposition; R is the damage function; ACso, is the incrementalconcentration of so, at receptor cell k: HH, is the number of households at receptor cell kYang( Yang, 1997)developed the following dose-response functions that we used as the input for R in Formula( 10)For covering materials: R=5.61+2.84S0,+0.74 x 10LH'JR=14.53+23.81902+3.8×10[H’]For galvanized steel: R=0.43+4.47S0 2+0.95 x 10[H']R=3928+814502+212x10H‘In which, R refers to the speed of the corrosion(um/a); SO, refers to the concentration of the So,; [H'I refers to thencentration of [h]of rainfall mol/L)中国煤化工CNMHGZHANG Shi-qiu et alVol 155.2.4 Estimation of damages by waste water pollutionThe wastewaler discharged by the Mawan Electricity Plant includes mainly domestic wastewater; cleaning wastewaterewater from ash flushing, and thermal wastewater. The wastewater has been treated primarily beforeHowever, the heavy metals and toxics in cleaning wastewater and wastewater from ash ponds may have an impact on theenvironment and it is difficult to quantify these impacts. Large amounts of thermal water discharged may also have impacts onhe ocean ecological systemWe use the cost for water treatment (outsideof the plant)to estimate the environmental cost ofnnual discharge, Treatment cost, Total losswastewater pollution. Specifically, we useRMB Yuan/t RMB Yuan1000secondary treatment cost for urban wastewater to Domestic waste water 14.4217.3calculate the losses produced by the wastewaler Rinsed water213000.0Monitoring Institute(1996),the treatment costsare:1.2 RMB Yuan/t for domestic wastewater, 2RMB Yuan /t for rinsing water, I RMB Yuan/t for ash fushing water and 0 RMB Yuan for thermal water. Table 8 showsthe treatment costs of wastewater6 Environmental impacts valuation results for the Mawan Electricity Plant6. 1 Damages based on geographe divisionTable 9 summarizes the annual average total costs, and the annual average extemality cost per k Wh of power generationalculated for each environmental externality group, broken down imo 4 sub-regions, The total environmental externalitiesproduced annually, if mortality valuation is included, are USD 3. 8 million to USD 6.7 million. The results also show thatpollution produced by the Mawan Electricity Plant imposed significant impacts on remote areas(rest-of-province and oul-of-e). These remote areas account for 78.4% to 84. 1% of the total environmental damages. Such a situation is perhapsdue to the use of high stacks for emissions, which will increase pollution dispersion towards more remote regionsenvironmental cost for generating one kWh of electricity ranges from USD 1.028 x 10" to 1. 832 x 10(1992 pricesonverted to 1997 prices, the cost would be USD 1. 8117 x 10"/Wh to USD 3. 2286 x 10"/Wh, which is 0.015 RMBYuan/kWh to 0.027 RMB Yuan/kWh. Currently, the electricity price at firm gate is 0.52 RMB Yuan/kWh. Therefore, theosts associated with environmental damages range from 2.9% to 5.2% of the current price of electricity6.2 Damages produced by each major pollutantBy breaking down the extemal costs in terms of pollutant type, the environmental costs imposed by each major pollutebe compared and the key pollutants and environmental impacts can be identified. The annual average extemalities and thepresent value of total extemalities of major pollutants are presented in Table 10, which one will find that air pollutants are themajor sources of environmental extemalities. They account for 86.0% to 91.9% of the total demages, while sO,, NOx, andparticulates( PM10), the three major pollutants, alone account for 80.5%to 90.0% of the total damages. SO, is the largestontributor among the air pollutants, with the dameges accounting for 66.6% to 70.3% of the damages caused by airpollution and accounts for 56.4% to 64.0% of total damages. The second largest contributor is NOx, with the damagesaccounting for 22.2% to 22. 7% of the damages caused by air pollution and 19.5% to 20.4% of total damages; the thirdlargest contributor is PMIO, which accounts for 5.2% to 5.3% of the damages by air pollution and 4.6% to 4.8% of thetal damages6.3 Damages per physeal unit of emissionsTable 11 summarises the total emissions and damages per unit emissions of major pollutants. Although the particularemissions is not much in comparison with $o,, it has the highest damages per unit at USD 592/ton to USD 1040/ ton; whileSO, damages per unit emission is from USD 106. 24/ton to USD 173, 50/ton. The treatment cost for SO, in CIestimated to be less than USD 100 per ton. This difference between damage and treatment costs for $O2 should provide furtherevidence in support of increasing levels of $O2 levies or charges to encourage polluters to reduce sO, emissions. Further effortscontrol pollution is cost elThe Environmental Impact Assesment Report for the Mawan Plant states that the thermal water will not produce significat imp中国煤化工CNMHGN.3Marginal cost pricing for coal fired electricity in coastal cities of China: The case of409Table Annual average total externalities and per kwh generated by geographic divisionsAnnual average total extemalitieskWh generatedExternalityDamages( USD 1000)Damages(USD milEs/kWh)Low, 209High, 80% Low, 20%High. 80%31.51448.2340.0090.013Water287.946575.893Land/waste52.20058600.014Local subtotal327.618659607993.373test-of- region369.900462.8320.0740.10l0.1260.0000.0000,00000000,000Land/ waNle0.0000.0000.0000.000Other0.0000.0000.00272.57999004628320.u0695,4330.000m00050.0000.4611914.204000031793750.5200.0000000,000Other0.000out-o- province subtotal19142042563.2373179.375Total extemalities3782.4555288.177Low and high totalsof Central Limit Theorem; * 1 mill USD 0.001Table 10 Annual average and present vahe o total externalities of major pollutantsAnmual average0)Externality groupLow, 20%Hi,80%Low,20%High, 80%Greenhouse gaScO,N/ALead48,90031.000Mercury330.0008642.52210843,919248.000312,0002022.0242543.8370.017phur dioxide3898.75025572.68331787767228.293476.969688.956A r subtotal4651.3905856.89327700,65837924.281Water863.8392347,7147043.lToxics in ash0.0000.000Water subtotal2879463∞863.839end/waste25.00036221.77039.463Land/waste subtotal406.60236985155279.483681.1333015518743045.32555533.359Note: Due to the argument an中国煤化工CNMHGZHANG Shi-qiu et alVol. 156.4 SummaryTable l1 Annual damages per physical unit emissiondamages and damages per physical unit ofAir pollutant Annualmissions for each major pollutant, Thepresent value of damages for SO2 is clearly Particulates 299.00 tonston592,0000001040.000the most significant one, with about USD Nitrogen oxides 4710.00 tonss/tonl60.000226,000282.00025. 6 million in total damages. NOx ranks Sulphur dioxide 22500.00 toms $/ton 106. 240 139.670 73 500as second with a total present value of LeadUSD 8.6 million and particulates rank Mercury340,00lbsb15.4003980061.400693.00lbs/bthird with a total present value of USDeryllium 69 50 lbs0.0200.0402.0 millionChromium208.00lbsThe major conclusions that can be Nickel243.00lbs/lb0.000104.00l813.20032.00048.000section are as follows:(I) The regionsTable 1z Ranking of damages by major palsufer most of the damages, about 78. 4% Total damages in present valueto 84.1% of the total:(2)air pollutionin central value term(in central value termemssionis the most significant contributor to theRanklotal damages, with 86.0%to.9% of Sulfur dioxide25572. 683 22500 tons Particulates( PM10) 830 s/tonthe extemalities resulting from Nitrogen oxides710 tons Nitrogen oxides26 S/tontants:(3)So,,NOrParticulates( PMIO) 2022. 024 299 tons Sulfur oxides 139.7 $/ton1206. 692 1010 Ibs Mercury398sand particulate matter are the three major Mercurythey account for 80.5% to 90.0% of the total damages, while S0, accounts for 56.4% to 64.0%, NOx for 22.2% to22.7%, and particulates for 4.6% to 4.8% of the total environmental costs;(4)damages produced per unit of particulatesNOx, and so, are significant with the values of USD 830/t, USD 226/t, and USD 139.7/t respectively: (5)totalenvironmental cost of electricity generation accounts for 2.9% to 5.2% of the current electricity price at firm gate7 Policy implications of this studyIn most cities of China the concentrations of particulate matter and So, far exceed WHO guidelines of 70 pg/m and 50Hg/respectively( Sunman, 1998). The results of this study suggest the importance and cost-effectiveness of increased7.1 A more effective polution levy system on so, i needed to internalize the externalitiesThe analysis above shows that the costs of environmental damages is the main factor leading to price distortion andccounts for 2.9% to 5.2% of the current firm gate price of electricity. Taking into account the current production cost ofelectricity generation at the Mawan Plant, which is about 0. 31 RMB Yuan/kWh, if the environmental costs are intemalisedthen the cost ofcity generation should be increased from 0. 325 RMB Yuan/kWh to 0. 337 RMB Yuan/kWh, which is anincrease of 4.8 %to.7% of the current production costLevies or charges on pollutant emissions are necessary, especially for SO,. This study shows that $O, emissions produceenvironmental costs of USD 106. 24/t to USD 173. 50/t (or about 881. 8 RMB Yuan/t to 1440. 1 RMBSO, mitigation cost is about USD 100/t(830 RMB Yuan/t). China introduced a O, charge in 1992 on a trial basis atof only 200 RMB Yuan/ton emissions. This rate is well below the damage costs of $o,, covering only 13.9% to 22.7% ofthe damages caused. If the levy or charge on $0, were to reflect the full damage costs, they would help to encourageelectricity producers to better mitigate the pollution and would bring greater environmental gains for the whole of societyable Electricty tarmportant obstacle to the effective intemaliration of environmental externalities is the current electricity tariffs中国煤化工CNMHGNo. 3Marginal cost pricing for coal fired electricity in coastal cities of China: The case of"..slructure. More manageable electricity tariffs should be set up which would not only take into account the price at firm gateb it also demand-related issues such as user price, different end-uses and timing (e.g, peak versus off-peak use ).Asignificant problem related to electricity tariffs in Shenzhen( and Cuangdong Province)is that, the consumer price of electricityis already very high, with some users paying I RMB Yuan/kWh to 2 RMB Yuan/kWh, which is about 2-4 times the firmgate priee, At the same time, the producers complain they eam very their low profits(some even operate at a deficit).Theoblem is due in part to the complicated and irrational price system, and to the fact that most of the profits are collected bye distribution and transmission of electricity7. 3 Other implicationshis study showed that the environmental damages produced by the Mawan Electricity Plant have greater effectsr gions further from the plant. This finding has a number of important policy implications. When one includes all ofuangdong Province and surrounding regions into the damage valuation analysis, it may be necessary to re-examine whetherthe high stack is the best choice for mitigating pollution or whether other options might be better, Where the high stack strategyis used, a compensation scheme should be considered. If a compensation policy is needed, then the results in Table 10 and1 I provide a reference baseline rate forThe results of this study also have implications for emissions trading of $O2, SO2 emissions have long distance transportcharacteristics and may affect more than one air shed. With a permit trading scheme, the quota being treated should not onlytake into account the total emission, but also the geographic location of emissions as well as the potential damages per unitAcknowledgements: The authors wish to acknowledge the research grant extended by EEPSEA: and thankful for the valuablecomments of Jeremy Warford, Stein Hansen, David Clover and Hermi francisco, as well as other members of the advisoryReferencesAlbey D E, Petersen F, Mills P K et al., 1993. Long term ambient concentrations of total suspended particulates, oone and sulfur dioxideand respiratory symptoms in a non-amoking population[J]. Archives of Environmental Health, 48(1): 33-46Asian Development Bank, Office of Environment, 1994. National response strategy for globel climate change: People's Republic of ChinalZICao H F, Shu J M, Liu Y J, 1991. Study on economic loses of crops in Guangdong and Guangri aree[JI. Research of Environmentallences,4(2):29-33.ChenB Y, 1992. Study on concentration of benmo(a)pyrene[ J]. Journal of Environment and Health, 9: 245-24Chestnut LG. Rowe R D. 1990. Economic valuation of changes in visibility: a state of the science assessment for NAPAP!R].Acidicdeposition: State of science and technology report 27( Brown Jr. G. M, Callaway. J. M. ed. ) National Acid Precipitation AssessmentProgram, Washington, D.C. OctoberChina State Statistical Bureau, 1994: 1995: 1996: 1997. Statistical yearbook of Chinal M]. Beijing: China Statistical Publishing HouseCopper M L. Freeman M A M, 1991. Environmental health effects[ M]. Measuring the demand for environmental quality(Braden J.BFisher A, Chestnut L G, Violette D M, 1989. The value of reducing risks of death: A note on mew evidence[ j]. Journal of Policy AnalysisEPA,1990.Methodology for assessing health risks associated with indirect exposure to combustor emissions[R]. Interim final. Prepared byJeffrey R B, Dann T F, 1997. The relationship among TSP, PM10, PM2. 5, and inorganic constituents of atmospheric particulate mailer atultiple Canadian Locations[ J), J Air& Wase Manage Assoc, 47(January): 2-19Kn pick A J, 1986. A preliminary benefits analysis of the control of photochemical oxidants R]. Report prepared for the U.S. EPAKrupnick A J. Cropper M L, 1989. Valuing chromic morbidity damages: medical costs, labor market effects, and individual valuations[RIinal report to U. S. EPA, Office Policy AnalysisKnpnick A J. Cropper M L, 1992. The effect of information on health risk valuation[ J]. Journal of Risk and Uncertainty,5:29-48Loehman E T, Berg S V. Ammo a d al., 1979, Distributional analyis o regel benefits and co o air quality control[J],Joumal of中国煤化工CNMHG412HANG Shiqiu et alvolEnvironmental and Economie Management, 6: 222--243iller T R, 1989, Willingness to pay comes of age: will the system survive?[JJ. North-westen University Law Review, 83: 876-907Pearce D, Warford ], 1993. World without end: economics, environment, and sustainable development[ M]. New York: Oxford UniversityPressPearce M. 1989. The benefits of environmentamonetary valuation[ Z].Paris: OECDRowe, Lang C M, Chestut L G et al., 1995. The New York electricity extemality study M). Oceana Publications Inc.ation for economic losses by TVIE pollution in Ba County of Chongqing. Chinas Environmental Economics in PracticeIM]. Beijing: China Environmental Sciences PressSunman H, Munasinghe M, Zhang S Q, 1998. Economics and environmental management for industry in China[R]. A report submitted toChina Council for Intemational Cooperation on Environment and Development( CCICED), BeijingUSEPA, 1986. Guideline on air quality models(Revised)[R]. Prepared by the Office of Air Quality Planning and Standards,ResearchEPA450278027RUSEPA. 1992. Integrated risk information syslem( IRIS)[R]. Office of Reaearch and Development, Office of Health and EnvironmentalViscusi WK, Magat W A. Huber J, 1991, Pricing environmental health risk: survey aseeMments of risk-risk and risk-dollar trade-offs forchronic bronchitis[ J]. Journal of Environmental Economics and Management, 21(1):32-51ang J Q, Lo D Y, 1989. Study on air pollution and residents'health in Chengdu City[ J]. Joumal of Environment and Health, 6(2):1-Wang R H, 1993. Discussion on the relations between bronchial asthma of children and the air quality, metrologic factor in Chongqing City[J]. Joumal of Environment and Health, 10(5):205-206Yang L, Tian Y, 1991. The strategy choice ford development and reform of the power industry in China[ Ml, Beijing: China Price PublishingYang Z M. Wang W X, Zhang W H, 1997. Study on estimation of economic los of materials by acid deposition[ J]. chongging EnvironmentalZhangC Q, Li Q, 1994. Analysis for air pollution and mortality of lung cancer in Chongging urban area[J]. Joumal of Emvironment andZhang H Q, 1995. Analysis for environmental benefits of thermal power development in China[j], Energy o China,3:47-49Zhang Y M, Wu L Y, 1997. Impacts on seeds quality and crops by acid nin(J]. Agricultural Environmental Protection, 16(1): 1-10中国煤化工CNMHG