Water pollution abatement by Chinese industry : cost estimates and policy implications

_Ws tl63o POLICY RESEARCH WORKING PAPER 1630 Water Pollution Abatement Data on pollution abatement costs in Chinese industry by Chinese Industry suggest that the benefits of stricter discharge standards Cost Estimates and should be weighed carefully against the costs. China's Policy Implications current eegulatory system provides an economic Susmita Dasgupta incentive to abate by Mainul Huq charging a levy on pollution David Wheeler that exceeds the standard. Chonghua Zhang But changing to a full emissions charge system would greatly reduce total abatement costs. The World Bank Policy Research Department Environment, Infrastructure, and Agriculture Division August 1996  I POLICY RESEARCH WORKING PAPER 1630 Summary findings Using factory-level data provided by China's National * The current regulatory system provides an economic Environmental Protection Agency (NEPA) and the incentive to abate by charging a levy on pollution that Tianjin Environmental Protection Bureau, Dasgupta, exceeds the standard. But the results of this analysis Huq, Wheeler, and Zhang estimate the costs of water suggest that changing to a full emissions charge system pollution abatement for Chinese industry. Using their would greatly reduce overall abatement costs. For the econometric results, they analyze the cost-effectiveness of sample of 260 factories, the current overall abatement current pollution control policy in China - and rate could be attained under a charge system with conclude that: present-value savings of $344 million. At a cost * For each pollutant, marginal abatement costs exhibit equivalent to that of the current system, uniform great differences by sector, scale, and degree of pollution charges would produce much better abatement. Ratios of 20:1 in each dimension are not environmental quality. uncommon. Approach: To measure the costs of abatement, they * The benefits of stricter discharge standards should use joint abatement cost functions that relate total costs be weighed carefully against the costs. For a sample of to treatment volume and the simultaneous effect of 260 factories, a shift across the existing range of reductions in suspended solids, chemical oxygen standards entails a present-value difference in abatement demand, biological oxygen demand, and other costs of $330 million. pollutants. Tests of alternative functional forms suggests * Emissions charges as low as $1 a ton would be that a simple (constant elasticity) model fits the data as sufficient to induce 80 percent abatement of suspended well as a complex (translog) models does, permitting solids for cost-minimizing factories. Charges of $3 a ton, sophisticated policy experiments with relatively simple $15 a ton, and $30 a ton would be sufficient to induce calculations. 90 percent abatement of suspended solids, chemical oxygen demand, and biological oxygen demand, respectively. This paper - a product of the Environment, Infrastructure, and Agriculture Division, Policy Research Department - is part of a larger effort in the department to understand the economics of industrial pollution control in developing countries. The study was funded by the Bank's Research Support Budget under research project "The Economics of Industrial Pollution Control in Developing Countries (RPO 680-20)." Copies of the paper are available free from the World Bank, 1818 H Street NW, Washington, DC 20433. Please contact Susmita Dasgupta, room N10-035, telephone 202-473-2679, fax 202-522- 3230, Internet address sdasgupta@worldbank.org. August 1996. (23 pages) The Policy Research Working Paper Series disseminates the findings of uwork in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be used and cited accordingly. The findings, interpretations, and conclusions are the authors' own and should not be attributed to the World Bank, its Executive Board of Directors, or any of its member countries. Produced by the Policy Research Dissemination Center  WATER POLLUTION ABATEMENT BY CHINESE INDUSTRY: COST ESTIMATES AND POLICY IMPLICATIONS by Susmita Dasgupta* Mainul Huq David Wheeler Chonghua Zhang PRDEI Dasgupta, Huq and Zhang are Consultants, and Wheeler is Principal Economist in the Environment, Infrastructure and Agriculture Division of the World Bank's Policy Research Department. Our thanks to China's National Environmental Protection Agency and the Tianjin Environmental Protection Bureau for providing us with the data analyzed in this paper. Thanks also to Shakeb Afsah, Ken Chomitz and Raymond Hartman for useful comments and suggestions.   EXECUTIVE SUMMARY In this paper, we use factory-level data provided by China's National Environmental Protection Agency (NEPA) and the Tianjin Environmental Protection Bureau to estimate water pollution abatement costs for Chinese industry. We utilize joint abatement cost functions which relate total costs to treatment volume and the simultaneous effect of reductions in Suspended Solids, Chemical Oxygen Demand, Biological Oxygen Demand and other pollutants. Tests of alternative functional forms suggest that a very sirnple (constant elasticity) model fits the data as well as a complex (translog) model, permitting sophisticated policy experiments vith relatively simple calculations. Using our econometric results, we analyze the cost-effectiveness of current pollution control policy in China. Our basic conclusions are as follows: * The benefits of stricter discharge standards should be weighed carefully against the costs. For our samnple of 260 factories, a shift across the existing range of standards entails a present-value difference of $330 million in abatement costs. - Emissions charges as low as $1.00/ton would be sufficient to induce 80% abatement of suspended solids for cost-minimizing factories. Charges of $3, $15 and $30 per ton would be sufficient to induce 90% abatement of TSS, COD and BOD. - The current regulatory system provides an economic incentive to abate by charging a levy on pollution in excess of the standard. However, our results suggest that changing to a full emissions charge system would greatly reduce overall abatement costs. For the 260 factories in our sample, the current overall abatement rate could be attained under a charge system at a reduced annual cost whose present value is $344 million. At a cost equivalent to that of the current system, uniform pollution charges could produce much higher environmental quality.  I  1. INTRODUCTION Although the potential benefits of industrial pollution control are clear in many developing countries, policy makers continue to worry about the costs. It has been difficult to address this concern explicitly, because little empirical evidence has been available. In addition, information about abatement costs would be extremely useful for the design of cost-effective regulation. In this paper, we use a new plant-level database to produce such infornation for China. We estimate a joint abatement cost function for major water pollutants, and use the results to evaluate the economic efficiency of current regulation. We also use the econometric results to simulate the impact of an emissions charge system. The paper is organized as follows. Section 2 provides an introduction to industrial pollution control issues in China. Section 3 uses the new dataset to develop measures of abatement and compliance for our sample of factories. In Sections 4 and 5, we specify and estimate an econometric model of abatement costs which is appropriate for simultaneous control of several pollutants. Section 6 discusses the policy implications of our results, while Section 7 summarizes the paper. 2. INDUSTRL4L POLLUTION CONTROL IN CHINA Industrial air and water pollution in China have been major concerns for the past two decades. A recent assessment by the Chinese Research Academy of Environmenital Sciences (CRAES) has identified industrial pollution as the source of approximately 70% of China's total environmental pollution. Current estimates of human health damage from urban air pollution are very high for some areas.' Such high levels of damage are primarily due to the rapid growth of pollution-intensive industries, not to lack of effort by China's environmental regulators. Indeed, the pollution control program of China's National Environmental Protection Agency (NEPA) and the provincial Environmental Protection Bureaus (EPBs) is probably the most extensive in the developing world. According to CRAES (1994), pollution abatement in the past decade has been sufficient to maintain at least constant levels of industrial waste water discharge and flue dust emission from coal combustion. Total estimated emissions of suspended particulates have dropped from 13.5 million tons to 5.8 million tons. At the same time, industrial output has approximately quadrupled. The pollution intensity of output in certain key emissions categories has dropped sharply since 1985, at least in factories which are regulated by the environmental agencies.' ' See CRAES (1994), X. Xu (1994), and Z. Xu (1995). 2 One cautionary note is warranted here: regulatory coverage is bv no means universal. It seems to be particularly sparse for Township and Village Enterprises. the fastest-growing ownership class. I  This record is impressive, but several factors suggest that the next decade will pose major challenges for NEPA and the EPBs. Continued rapid decline in pollution intensity will be necessary just to stay even with the pace of industrial growth. Moreover, recent findings on pollution-related health damage suggest that considerable improvement in ambient quality would be appropriate. Faced with the simultaneous need to reduce pollution and increase industrial output and employment, the Chinese government has become very interested in cost-effective regulation. The current pollution control system is under scrutiny, because its peculiar mix of regulations and economic incentives bears little resemblance to a conventional emissions charge system. NEPA regulations specify effluent standards by sector, and a schedule of fees (the pollution levy) to be paid by any enterprise whose effluent discharge exceeds the mandated standard. With the approval of NEPA, local areas may raise both standards and fees above national levels. Levies are charged only on the 'worst case' pollutant from each source.3 This incentive system is supplemented by more traditional pollution control measures. Under the 'Three Simultaneous Steps' system, new enterprises are required to construct abatement facilities with capacity sufficient to meet the relevant effluent standards. Chinese regulators are debating whether this mixed-instrument regime should be changed to a conventional emissions charge system. Since transition costs are likely to be high, there is strong interest in estimating the potential net gain from such a change. Theoretically, the gain could be substantial. Commonly-cited simulation results from the OECD economies suggest that total costs in a standards-based system can be several times those in a charge-based system.4 For China, a well-informed judgment should be based on much better knowledge about actual abatement costs. 3. THE DATA Our data are drawn from two sources: (1) the China Monitoring Station in Beijing, which monitors the 3000 factories currently rated as China's largest potential polluters; (2) the Tianjin Environmental Protection Bureau, which monitors industrial facilities in the Tianjin urban region. The available data bases include information on production, emissions by pollutant, abatement, abatement costs, and pollution-related penalties such as pollution levies, fines and compensation paid for damage. Relevant variables for the present study are summarized in Table 3.1. For this exercise, NEPA has provided us with 1994 data for 200 factories scattered across China's urban/industrial areas. The Tianjin EPB has supplied data for 60 additional plants. The data base provides separate information for each emissions source. Many facilities have multiple sources, yielding a total sample size of 370 obsrvations. Although the data base is exceptionally complete, it records only three pollutants abated for each source. This is probably sufficient for most actual 3 For more extensive discussion of the pollution levy svstem. see Wang and Wheeler (1996). 4 See Wheeler (1992) 2  cases, but in theory it could cause some truncation problems. We have estimated total abatement costs for each treatment point by adding operations and maintenance expenditure to annualized services from abatement capital. 5 Data on treatment volume, influent concentration and effluent concentration are taken directly from the recorded measurements. The current state of compliance in China is highlighted by the range of variation in abatement activity recorded in the data base. As Table 3.2b indicates, end-of-pipe abatement ranges from 0 to 100%, with median abatment of standard water pollutants (BOD, COD, TSS) in the 70-80%/o range and first-quartile abatement around 50%. The willingness of many plants to report non- compliant discharges to NEPA indicates that the levy system is working as intended. Excess discharges are subjected to a fee under the levy system, but they are not illegal. China's abatement statistics compare very favorably with those of wealthier Southeast Asian economies such as Indonesia and Philippines, whose median abatement activity is closer to 50% (Hettige, et. al., 1995). Table 3.2a shows that water pollution would be much worse without existing control. Influent (pre-abatement wastestream) concentrations for most plants greatly exceed Chinese discharge standards (Table 3.3). Median influent concentrations for TSS, COD and BOD (567, 850 and 264 mg/l, respectively), are far above discharge standards for the lowest- quality water bodies in both Guangdong and Beijing. As Table 3.3b shows, median effluent concentrations are much closer to existing standards for medium-low quality water bodies, and 1 st-quartile concentrations are generally in the medium-high quality range. However, a large number of plants remain out of compliance. Tables 3.2 and 3.3 show that Chinese factory managers in polluting sectors must contend with great diversity in both process emissions and location-sensitive concentration standards. Their abatement decisions, and the overall level of abatement costs, will be significantly affected by the degree of pollution control necessary to bring their emissions into compliance with prevailing standards. 4. THE DETERMINANTS OF ABATEMENT COSTS Traditionally, abatement cost estimates have been based on plants' reported direct costs of installing and operating pollution control equipment. Coupled with information about the benefits of reducing pollution, such cost estimates can provide a basis for setting sensible regulatory standards. Until recently, the scarcity of appropriate plant-level data has prevented detailed empirical studies of average and marginal abatement costs by pollutant.6 Policy analyses have 5 We have used an interest rate of 10% for this exercise. 6 For other recent work oii this issue. see Hartman. Wheeler and Singh (1995) and Mundle, et. al., (1994). 3  frequently developed abatement cost estimates from engineering models. However, failure to rely 7 on behavioral data has led to considerable estimation errors. While environmental economists and policymakers have focused almost exclusively on direct abatement costs, we recognize that these provide an incomplete measure of the cost of pollution reduction. Firms can adjust to the threat of higher pollution-related costs along many dimensions, including new process technology, pollution control equipment, improved efficiency, and allocation of more resources to legal representation or negotiation. At the plant level, all these options will register as changes in the scale and mix of inputs and, consequently, total production costs. Thus, pollution control will have an impact on conventionally-defined total factor productivity (TFP) which may be significantly different than directly-reported abatement costs.8 In this study, we focus on direct abatement costs because the available data do not permit estimation of a TFP-based cost function. 4.1 The Direct Abatement Cost Function Industrial facilities can abate pollution by scaling back polluting activities or by diverting resources to cleanup. In either case, pollution reduction will entail costs.9 Moreover, diminishing returns will apply: more resources will have to be devoted to cleaning up each additional unit of pollutant. Hence, the marginal abatement cost (MAC) function slopes upward from right to left as pollution falls. The position and slope of the MAC function are affected by factors such as the scale and sectoral composition of production; the average operating efficiency of the firm; the available process technologies; and the efficiency of waste treatment technologies. For any given level of pollution, more costly pollution control is associated with rightward movement of the MAC function. Conceptually, abatement cost functions are dual to abatement functions which relate inputs of capital, labor, energy and materials to pollution reduction. Abatement processes frequently reduce more than one air or water pollutant, so joint cost function estimation is appropriate. For example, Biological Oxygen Demand (BOD), Chemical Oxygen Demand (COD), and Suspended Solids (SS) can all be reduced by treatment in common facilities. These joint equipment 1 A useful illustration is provided by the trading price for SO2 emissions permits under the U.S. Clean Air Act (Hamilton, 1994). Using engineering models. the U.S. Goverrunent forecast a price around $600/ton before the trading system was instituted. In fact. permits have recentlv traded at prices around $150/ton. Recent plant-level econometric work by Hartman, Wheeler and Singh (1995) on S02 reduction costs in the U.S. has yielded estimates much closer to the latter figure. 8 In an econometric study of TFP impact for several U.S. industries. Gray and Shadbegian (1993) find that regulation imposes a TFP loss approximatelv three times higher than the reported direct cost of abatement. 9 There is currently an important debate, initiated by Porter (1993). on whether or not firms can be made more profitable by forcing them to undertake pollution control activities. In a verv simple static framework of analysis, Oates et al. (1993) have shown Porter's argument to be * Tong, and point out that the route from pollution abatement to higher profits is much more complex than mav have been tlhought. Ultimately, this is an empirical question. 4  requirements are associated with common use of skilled and unskilled labor, energy and materials. 4.2 Cost Function Specification For k pollutants, the environmental engineering literature suggests that an appropriate joint cost function for plant i should include the following variables: (4. 1) Ci = f (Wi, , Mj, X) where Cj: Total annual cost of abatement for the plant Wj: Total annual wastewater volume Ei,,/I,,: Vector of effluent/influent ratios for n pollutants, which can be interpreted either as concentration ratios or volume ratios (since waste water volume is constant across influent and effluent for each plant, it cancels out of the concentration ratio). Mj: Vector of input prices at location j Xi: Vector of relevant plant characteristics (sector, age, ownership, productive efficiency, etc.) For the kth pollutant, the marginal abatement cost function is given by: r(W, Ei , M,X,) (4.2) i i aEik atik We exclude the vector of input prices, since appropriate cross-regional price indices are not presently available to us. '0 We specify a second-order quadratic approximation to the general cost function (or translog function) as follows: InC = ao + a1 InW + a InW2 + fi In; + fiy InI In W (4.3) y~~j ln~L) + /Jij ~~ln~