Integrated Water and CGE Model of the Impacts of Water Policy on the Beijing's Economy and Outp

Chinese Journal of Population, Resources and Environment Vol.8 No.2 June 2010Integrated Water and CGE Model of the Impacts of Water Policyon the Beijing's Economy and OutputXia Jun', Deng Qun12, Sun Yangbo1.21. Key Lab of Water Cycle and Related Surface Process, Institute of Geographical Sciences and Natural Resources,Chinese Academy of Sciences, Bejjing 100101, China;2. Graduate University of Chinese Academy of Sciences, Bejing 100101, ChinaAbstract: The article used general equilibrium model to analyzeis becoming less viable due to resource constraints and thethe change of gross domestic product and industry output afectedincreasing marginal costs of engineering solutions. Alterna-by water resources policies in Bejing City by using GEMPACKtive management options, such as demand-side manage-soft tool. The article researches on rules of water supply and de-ment, have to be considered (Ashton and Seetal, 2002).mand, evaluating water resources, building water resources inputBeijing Municipality has adopted several changes on waterand output table, establishing water computable general equilib-price since the 1990s (See Table 1). Now according torium model and stimulating water policy. The stimulation givesChina Water Law, the local government will increase watera scenario that increases water price by 10%. The result showsprice and waste water fee after 2008. So it is critical to ana-the fllowing aspects. First, water resources policy infets gross lyze the impact of water policy to Beijing's economics anddomestic product and industrty output in dferent ways. There areoutput. Water price and waste water fee are two major fac-different behaviors in dfferent industries as to the water policy.tors in water resources economics analysis and give a greatAgriculture industry has the same tendency as water price changeimpact to national economics. Most researchers have usedand it has more sensitive to water quantity than to water price. For part equilibrium model for analysis relationship betweenbasic energy industries such as oil and chemistry and gas, theywater price and gross domestic product (GDP) growth andshow diversity tendency. As to some high water consumer indus-industries output. But in some cases it would give an unrea-try such as paper and textile et., water resource economic policysonable result according to some experts research (Horridgecan infect them greatly and can promote them to accomplish moreMark, 1993). For this reason, scientists are using generalequilibrium model to analyze the change of GDP and indus-industries show the same tendency as to water policy. Second,try output affected by water resources policies. This articlegovernment should pay more attention to water resource policy byTable 1 Changes of water price in Beiing (unit: yuan RMB/m)Items/year2001 2002 2003 2004 2005the output and supply and consumer price change more than expectI. life and industry3.30 4.00 4.7 5.40 6.00as to water poliey in a free market economic in water industry. So1. tap water (include water conser-as to a government policy maker, one should be more carefully and2.13 2.13 2.64 2.63 2.60rancy fee)prepare suitable forecast and plan to water policy and its negative2. waster water0.80 1.00 1.20 1.40 1.50impact.3. integrated water resources fee0.40 0.91 0.87 1.37 1.90Key words: water nlicy analysis, CGE model, BejingII. water construction supply price 0.71 0.77 0.77 0.77 0.81input- output table general equilibriumII.. water resource fee to self-well 1.30 1.90 2.5 3.10 3.70IV. surface water of irigation0.15 0.22 0.30 0.35 0.401 Introduction1. water engineering0.12 0.17 0.23 0.27 0.302. water resources fee0.03 0.05 0.07 0.08 0.100.10 0.15 0.20 0.25 0.30limiting factor to development (Aronson et al, 2006). InVI. industry reuse water0.20 0.38 0.36 0.43 0.50the past, rising water demand was addressed through asVII. agriculture reuse water0.04 0.06 0.08 0.09 0.10supply-side mechanisms (Smakhtin et al, 2001), but thisSource: satistice data and report from Bureau of Water Afairs of Bejing.Received 4 March 2009; accepted 8 June 2009中国煤化工Corresponding author: Deng Qun (dq@ waterpub.com.cn)MHCNMHG61Chinese Journal of Population, Resources and Environment Vol.8 No.2 June 2010describes how we linked information about water price andtionship of basic valucs to production costs and to purchas-water use with a computable general equilibrium (CGE)ers' prices; market-clearing conditions for commodities andmodel by GEMPACK soft package.primary factors; and numerous other macroeconomic vari-ables and price indices.2 Water CGE model description2.1.2 Model equation and percentage changeConventional neoclassical assumptions drive all privateThis section describes the standard parts of our CGEmodel and the extensions added to the model of water equa-agents' behavior in the model. Producers minimize costwhile consumers maximize utility, resulting in the corre-tion and function.sponding demand and supply equations of the model. The2.1 Computable general equilibrium modelagents are assumed to be price takers, with producers op-The model used here is similar to the ORANI-G modelof pure profits. In general, the static model with its overallof the Australian economy, which is fully presented andLeontief production structure allows for limited subtitutionexplained by Horridge (2002). The model consists of thou-sands of equations that could not be repeated here. Thein consumption. It has constant elasticity of substitutionreader may visit www.monash.edu.au/policy/oranig.htm for(CES) sub-structures for the choice between labor, capitala summary of all the country models that have been built inand land, the choice between the different labor types inthe ORANI-G style.the model, and the choice between imported and domestic2.1.1 Structure of CGE modelinputs into the production process (see Fig. 1). HouseholdWe present a diagrammatic overview of the structure ofdemand is modeled as a linear expenditure system that dif-the model in Fig. 1. The model has a theoretical structureferentiates between necessities and luxury goods, whilethat is typical of most static CGE models, and consists ofhouseholds' choices between imported and domestic goodsequations describing producers' demands for produced in-are modeled using the CES structure. Fig. 1 shows thatputs and primary factors; producers' supplies of commodi-commodity composites and a primary factor composite areties; demands for inputs for capital formation; householdcombined using a Leontief production function. Conse-demands; export demands; government demands; the rela-quently, they are all demanded in direct proportion to totalproduction ("activity level!"). Each commodity composite isOutputa CES fiunction of a domestic good and the imported equiv-alent. The primary factor composite is a CES aggregate ofland, capital and composite labour. Composite labour is aLeontietindustries share this common production struc- ture, input(Goodit up ----o3proportions and behavioral parameters may vary betweenindustries (Horridge, 2002). The elasticity used for the CESCESESfunctions in the model are three kinds: export demand elas-ticity equal 4; Arminton intermediates of industries equal 2;and CES between capital, labour and land between 0.1 and1 for different industries.(Land )( Labour)[ CapitalAt the heart of a CGE model are equations describingKEYindustry input demands and output prices. Although detailsCEr Functionarvary, most CGE models assume a“nested" arrangement ofFomy-CES sub-production functions like that of Fig. 1. Abstract-Inputs oring from details, the production function for one sector mayOutputs(Labour Labour. up _o---___our}_be represented as(type D (type2ltype yFig. 1 Structure of water CGE modelMin Z.PX+[P中国煤化工'THCNMHG62Chinese Journal of Population, Resources and Environment Vol.8 No.2 June 2010s.t: X,/A, = min{X;/ A,,i= .,.;total of 47x52 sectors. The oficial table has only one sectorfor agriculture and irrigation is mainly rural water use. Theweights used for the splits rural and industry water used arebased on the output value in the table. As to the SAM, weAnd Xy=CES[ x])14",s=I,D]can derive from statistic data and yearbook published bywhere Z is output and X1, X2, X3, ... are various inputslocal municipality. The basic structure of the SAM table is .with prices PI, P2, P3, ... Usually the production function shown in Table 2. We calculate the details data by GEM-F has the property that if all inputs are doubled, output alsoPACK file in mkoranig.tab according to some guessed cCo-doubles.It is assumed that the producer chooses input pro-efficients. For saving words we just give the concepts andportions to minimize the cost of producing given output Z. framework of this table. Some data are ilustrated in waterIn GEMPACK the input demand equations are representedextension sector below.in percent change form as2.2 Water extension to CGE modelxy=(ay +a)+x,We explain the balance function and equation of waternd )"0)-a")-a,"=x;+a,-oσ;[p,;"+a{'")-p%]resource in a CGE model and the data which are used inWe can also illustrate other percent change function likewater IO-table.follows.2.2.1 Water supply and demand equationFor pricep,=Z. H'"[D"+a"]Water cycle formed the process of the whole water re-source change (OV), which precipitation (P) falls and be-p"=Z H'"[p"+a{0]comes rainfall (R) and underground water (Ug) and waternet use (Qc), by following functionFor consumer demandP=R+En+Ug+Qc+OVx)-a{)=&,0)+Zima[p"'+a%]For life consume of water resourceFor export and importSp?"X{"=[YH-YHT-(nw:2Q!?".Q)+(°-fa(°"-1g"=σ!°[p;")-π-p!"]rmx,.V?Q!?"Q ](1- mps)While computationally tractable, the linear equationwhere YH and YHT represent tax of citizens' income andsystem is only a local approximation to the nonlinear modeltotal tax; rw; and rw2 are the cofficients for price changes,specification. However, the efects of larger changes can be V'Q}2) and v2Q2) represent average price and waste wateraccurately computed by stepwise integration techniques-fee; Q: is water quantity of life using, which is decided byin practice this mmeans that the linear equation system is per-water use ration estimated by econometric function; mps isturbed by a sequence of tiny shocks, all the while updatingsave rate. For industry water usingcofficients like the x; above.之w; =之(w, +w。+W,)2.1.3 Database of IO-table and social account matrixwhere Wy, Wwj and Wy represent surface water and under-(SAM) flowsground water and reuse water respectively. For irigationA standard CGE database consists mainly of an IO-tableZAW= Z(W.+W)of flows, showing the expenditures by each industry andfinal demander on a range of commodities and on primarywhere Ws and Wu are surface water and underground waterresources such as capital, labour, and land. Most of theusing in irrigation, in the situation we do not concern aboutdata for the CGE model are drawn from the official 2002the reuse water in agriculture. Now we move the whole wa-IO table of Bejjing City, published by Statistics Bureau ofter demandBeijing (SBB, 2003). The table divides households intomax Zuw, = 2UlW,+ 2UAW,two groups as urban and rural area, and distinguishes 42economic sectors. We further disaggregate the technologyand service industry and water supply intensive sectors to as.t.rw,+SAW,中国煤化工W, +W,)MHCNMHG63Chinese Journal of Population, Resources and Environment Vol.8 No.2 June 2010Table2 Structure of social accounting matrix1 Production activi- 2 Factors of produc- 3 Household current 4 Corporation cur- 5 Government cur- 6 Foreigners currenttioaccountrent accountConsumer expen-Govermment expen-1 Production activi-Inter industryditure on domesticditure on domestie Exports of goodstiesTransactionaoodsomesand servicesgoods ;2 Factors of produc- Gross value add intionproductionWages and gross3 Household currentoperating surplusPayments of inter-Government trans- Transfer by foreign-est and diviend toof unincorporationfer to household ers to householdhousehold4 Corporation cur-Gross operation sur- Transfer by house-plus of corporation hold to corporationfer to corporation ers to corporationIndex taxes on5 Govermmenteur-Govemment oper Index taxes on con-mosrasumption income Company taxesexports Direct taxestion surplusta:on foreigners6 Foreigners current Import of interme-Consumer expendi- Property income to Government expen- Import componentdiate goodsture on importsforeignersditure on importsof exportsHousehold saving7 Household capitalconsumption of fixcapitalCorporation saving8 Corporation capi-tal accountGovernment saving9 Governmentcapital account! Foreigners capi-Balance of exten11 TotalTotal cost of pro-Total factor costCurrent outlays by Current outlays by Current outlays by Current outlays byductioncorporationgovemment2.2.2 Water price and water valuefirst, irigation water quantity X, (=1, 2.... 43), the 43is temporally added line in the table. We get the data X3)function. We estimate the water resource value based on tra-15.45 hundred million cubic meters and x1,=0 (=2....ditional economics theory, which is water value equal water43), from published materials; second, rural life water usingprice (P) multiwater quantity (W). There are two ways toquantity X43is 153 million m', where ruralconsumcalculate as followsis the original line of IO-table.For industry water quantity, we get data from BeijingV=rSw,Pwater resource bulletin. For each sector water quantity cal-culate through equation as followsOV=WP2.2.3 Water IO-table and database2X25..or-X25.23In original input- output table issued by SBB, there is nowhere Q is total industry water quantity equal 5.591 billionwaste water industry and whole water industry. We updatem'; Xrs; is output of the original table forj industry; Xxs,rorthe water supply sector and add waste water sector to theis subtotal output in line 25; X2s.23 is the output of originalIO-table. For the water supply sector, we first get rural wa-table for electrify and hot supply sector.ter supply, then we calculate urban water supply. So as to .Thus, we can get total output in water supply industrywaste water. Follows are the whole details process.14 477 million yuan RMB according to the quantity andFor rural water supply, irrigation was calculated inwater price, which is equal to input for water supply in-technology and service industry. It includes two parts as:dustry based the p( 中国煤化工divide totalTHCNMHG64Chinese Journal of Population, Resources and Environment Vol.8 No.2 June 2010input data into different industries according to equation asTable 3 Input cofficicnt of water supply industryfollowsRural water input Urban water inputName :LineP.2s=X.x2/Xxror.25coefficient (%)Agricuture0.0000 000 63where P;2s is input coffcient; X,25 is original input data inCoalMineProc1.635 564 242water supply sector; Xror25 is subtotal of original input dataCrudeOilGasof the same line of water sector.Metelmine0NonMetelmine0.017 690 844FoodToba0.000 523 64According to yearbook data issued by Bureau of Envi-CottonText0.008 041 662ronment Protection of Beijing in 2003 year, we know thatClothesShoes0.115751 389the whole waste water was 1.367 billion m3 and the dispos-WoodFurmitu0.143 309 441al rate was about 53%. In such situation, we also get wholePaperProd100.143 441133 .life waste water of 780 million m', which included ruralPetrolRef10.107 219 869life waste water 43 million m' and urban industry wasteChemDly121.502 000 953578 million m'. We adopt the same concept to calculate theNMtelProdu130.220 705 278output and input data of waste water sector.MetlSmelt140.147 390 018Finally, we have got new water IO-table which is 47*52IronProc0.347 118 407ComMachine1.068 251685sectors table.RoadMach0.468 857 875HhldElee0.684 376 4873 Model simulation resultsElecCommsEqp 190.098 158 454Meters200.254 990 262It is tendency to increase water price due to lack of wa-OtherManu20.013 171084.ter resource in Beijing. So we choose water prices increasedWasteScrap22ElecSteam2:16.812 25762by 10% and 50% for simulation. The whole stimulationGasSupply240.177 601 618program is shown in Fig. 2.WaterSupply1.458.955 615237When we increase water price for 10%, we get result asConstruction0.174 318388follows.RoadTrans270.465 421 056Post280.028 786 1353.1 Industry output changeComputSrve290.289 405 373From stimulation result (see Fig. 3) we conclude that in-Trade300.095 496 772Hotel30.068 573407 .dustry output shows a decrease tendency when water priceFinance .3.975 708 913prime file: io2002.xs change to: raw2002 harRealEstate330.116 970 088stepl: get result file of readldq.tab, readldq cmfLeasing340.950 884 957Tourism350.003 289 799step2 file: read1dq.diagnosefile:result file: scalio.tab, scalio.cmfdialreadldqharResearch3(0.039 105 038 .TechSrvc3"0.576 323 611step3 file: scalio.har,result file:diagnose file;OtherSrve1.334 744 201read2.tab, read2.cmfEducation390.037 206295SocWelfare4(0.000 104 165diagnose file:ArtsFilmTV410.057 723 044step4file: read2 .harmkorng.tab,diaread2.harPublicAdmin4:10.480.000 780 455mkormg.cmt _Wastewater0.002 598 796Deprec24.3213.691 036 64step5 file: oranig.har,oranig.sti,oranigsr.cmfWagesVA218.386.976 060353NetTaxesVA37.756.048 078 868SurplusVA4-0.097.569 416 347finish: oranigsr.slI~2, summary.harProfitVA524.575 929 48TotalFig. 2 Whole stimulation of water CGE model中国煤化工MYHCNMHG5Chinese Journal of Population, Resources and Environment Vol.8 No.2 June 2010信6导重与员三2合营曾豆省艺导营-0.5首是 三玉3夏皇31-industry outpu-1.52-Crude Oil Gas2.5 JFig. 3 Stimulation result one for industry output by 10% increase water priceincrease, (i) agriculture sector show medium size aboutincrease 10%, produce cost will increase and decrease in0.08%; (i) while energy industries such as oil and gas anddifferent ways, which causes two aspects, one is to promotecoal mine and electric power minus more than averageincrease cost and the other is to substitute local commoditylevel, especially of oil and nature gas industry of nega-with import and outside goods. Such impact will cause neg-tive 2.24%; (il) some high water use industry textile andative to demand of capital and labour, thus decrease primerpaper and metal industries give largest decrease and showfactor price. The details are: () agriculture price increasesmedium size at about 0.08%; (i) energy industry shows a0.5%; (iv) other industry also show average decrease. Fornegative level at about 0.09%; (i1) consumer will save wa-instance, waste water sector decrease 0.38% for water us-ter and factory will adopt new equipments and water-savinging decrease and correspond to whole output decrease. Thetechnology. The market equilibrium mechanism will causestimulation also gives data of labour decrease 0.1 1%, whilesome negative impact to water supply industry, though theGDP decrease 0.11%. That is to say, when water price in-water price will decrease 1 .95% for equal function but thecrease 1% and the GDP will decrease about 0.011%.whole consume price will show an increase tendency.3.2 Produce price change3.3 Consume quantity changeAs stimulation shows (see Fig. 4) when water priceFrom simulation (see Fig. 5) we can find out that local1.5-0.5-0.5-点91营重岛昌2-1.5-produce pnce-2.ig.4 Stimulation result one for produce price by 10% incre中国煤化工TYHCNMHG66Chinese Journal of Population, Resources and Environment Vol.8 No.2 June 20100.40.2°TTTT号-0.2--0.4 --0.8-consume quantity-1.2--1.4--1.6Jig. 5 Stimulation result one for consumer quantity by 10% increase water pricemarket consumer quantity will decrease as (i) agriculture impacts to economics.consume reflex no sensitive at 0.02% changes; (i) energyindustries also show small change, for example, oil indus-Acknowledgements: Project supported by the Natural Sciencetry decrease 0.15%; (ii) high water consumption industryFoundation of China (Grant no. 40730632/40671035), the Specialachieve negative tendency at about 0. 12%; (iv) water sup-Fund of Ministry of Science and Technology, China (Grant no.ply industry shows that when water price increase 1% the2006DFA21890), the Key Project of Intermational Cooperation in CASwater using quantity will decrease 0.141%. This fact can(Grant no. GJHZ06)compare to some experts’calculate results, such as Japa-nese national elasticity of water use is 0.1255%, Beijing'sReferenceselasticity of water use is 0.164 finished by Mr. Wang Yingof Bejjing University.Aronson J, Blignaut J, Miton J, et al, 2006. Natural capital: the limit-ing factor. Ecological Engineering, 28:1-54 ConclusionsAshton P J, Haasbroek B, 2002. Water demand management and so-cial adaptive capacity: a South African case study. In: TurtonA RIn this article we presented a CGE model, which was ex-and Henwood R, eds., Hydropolitics in the Developing World: Atended by integrating physical water flows and tax functionsSouthem African Perspective. African Water Isues Research Unitinto the model. Thus we can advise water policy makers on(AWIRU) and International Water Management Institute (WMI),the use of market instruments to reduce water demand in24some of the most water-intensive sectors. The CGE modelCai X, McKinney D C, Rosegrant M W, 2003. Sustainability analysiswas needed to assess the economy-wide impacts of a waterfor irigation water management in the Aral Sea region. Agricul-demand reduction policy, using water charges. We furthertural Systerms, 76: 1043- 1066include excess burdens and equity considerations into themodel.of trade reform and lead to a“win-win”outcome? Jourmal of En-If the water tax aims only to cut water use, then wevironmental Management and Economics, 45: 708 -723should levy a tax on some high water using industry, sinceStatistics Yearbook of Beijing. Bureau of Statistics of Bejjing 2000-that will save much more water compared to other indus-2005tries. However, wider economic and social criteria may beVeck G A, Bill M R, 2000. Estimation of the residential price elastic-of concern. For example, agriculture shows less sensitive toity of demand for water by means of a contingent valuation ap-water price than water quantity, which means that technol-proach. Water Research Commission, Report no: 790/1/00ogy of water-saving shows more important in agriculture.Young M, Proctor W, Qureshi E, Wittwer G, 2006. Without water:Since increase water price will harm the economy, in termsof GDP as well as the consumption levels of poor house-Water for a Healthy Country Flagship report, CSIRO and Monashholds, water policy should not be short-term for its differentUniversity. Melbou中国煤化工THCNMHG37