62 Int. J. Environment and Sustainable Development, Vol. 2, No. 1, 2003
River water pollution in Indonesia: an input-output analysis
Budy P. Resosudarmo Economics Division, Research School of Pacific and Asian Studies, The Australian National University, Canberra, ACT 0200, Australia E-mail: [email protected]
Abstract: Data from global environmental monitoring activities have shown alarming environmental conditions in many developing countries. Environmental policies that could improve the environment significantly, while at the same time maintaining the growth of economic activities are needed. Using an input-output analysis, this paper researches such policies with a view to applying them to Indonesia’s river water pollution. Firstly, this paper reviews river water quality and current policies in Indonesia. Secondly, it develops future policies to control such pollution.
Keywords: Input-output analysis; environmental economics; development economics.
Reference to this paper should be made as follows: Resosudarmo, B.P. (2003) ‘River water pollution in Indonesia: an input-output analysis’, Int. J. Environment and Sustainable Development, Vol. 2, No. 1, pp.62-77.
Biographical notes: Budy P. Resosudarmo is a Research Fellow in the Economics Division, Research School of Pacific and Asian Studies at the Australian National University. In writing this paper, the author received useful suggestions from Dr. Raksaka Mahi and Dr. Ari Kuncoro. Santi Budi Handayani is the research assistant for this paper.
1 Introduction
Since the 1972 Stockholm Conference on the Environment, global and local monitoring programs to determine the status and trend of key environment issues have been established around the world. As a result, several publications this subject are available [1–3]. Data from these publications indicate that several environmental indicators in many developing countries are already above tolerable levels. Table 1 presents forest covers in some developing Asian countries. This table shows that countries, such as Indonesia, the Philippines, Malaysia, and Myanmar, experienced annual deforestation rates above 1% in the last 20 years. By 2000, Indonesia had lost approximately 20% of its 1980 forest cover, the Philippines 40%, Malaysia 30%, and Myanmar 25% [1,2]. These losses contribute significantly to the reduction of the world’s forest cover.
Copyright © 2003 Inderscience Enterprises Ltd.
River water pollution in Indonesia: an input-output analysis 63
Table 1 Forest cover and deforestation
Forest area Average change (thousand km2) (% per year) 1980a 1990a 1990b 2000b 1980-90 1990-2000 China 126,398 133,756 145,417 163,480 0.57 1.18 India 58,259 64,969 63,732 64,113 1.10 0.06 Indonesia 124,476 115,213 118,110 104,986 -0.77 -1.17 Malaysia 21,564 17,472 21,661 19,292 -2.08 -1.15 Thailand 18,123 13,277 15,886 14,762 -3.06 -0.73 Philippines 11,194 8,078 6,676 5,789 -3.21 -1.42 Myanmar 32,901 29,088 39,588 34,419 -1.22 -1.39 Cambodia 13,484 10,649 9,896 9,335 -2.33 -0.58 Vietnam 10,663 9,793 9,303 9,819 -0.85 0.54 a Source: WRI [1] b Source: FAO [2]
Table 2 exhibits the annual average of ambient air pollution in several large cities in the world. It can be seen that cities in developing economies have air pollution levels above the WHO air quality standards. It is suspected that air pollution in these cities causes a large number of human health problems, such as premature mortality, respiratory symptoms, and asthma attacks [4–6].
Table 2 Annual means of ambient air pollution (µg/m3) in several large cities in the world Country City City population Total suspended Sulphur Dioxide Nitrogen (thousands) particulates 1998 dioxide 1998 2000 1995
China Beijing 10,839 377 90 122 Guangzhu 3,893 295 57 136 Shanghai 12,887 246 53 73 Shengyang 44,828 374 99 73 India Calcutta 12,918 375 49 34 Delhi 11,695 415 24 41 Indonesia Jakarta 11,018 271 30 148 Japan Osaka 11,013 43 19 63 Tokyo 26,444 49 18 68 Malaysia Kuala Lumpur 1,378 85 24 N/A Thailand Bangkok 7,281 223 11 23 Philippines Manila 10,870 200 33 N/A Mexico Mexico City 18,131 279 74 130 USA Chicago 6,951 N/A 14 57 New York 16,640 N/A 26 79 Los Angeles 13,140 N/A 9 74 WHO standard < 90 < 50 < 50 Source: 2001 World Development Indicators
64 B.P. Resosudarmo
Table 3 presents the median annual levels of Biology Oxygen Demand (BOD), Chemical Oxygen Demand (COD) and Faecal Coliforms in several major rivers in Asia. These rivers are used as a source of drinking water and the water quality of many of them does not meet standards [7]. This is a suspected cause of a significant number of human health problems such as premature mortality and diarrhoea [8,9]. In Indonesia, diarrhoea is among the top five causes of mortality [10].
Table 3 Water quality in several rivers in the world in 1991-1993
BOD COD Faecal Coliforms (mg/l) (mg/l) (no. per 100ml) China Yangtze River (Changjiang) 1 n/a 490 Yellow River (Huanghe) 2 n/a 3,500 Pearl River (Zhujiang) 1 n/a 260 India Bhima River (Takali) 5 24 0 Godavri River (Dhalegaon) 4 20 0 Sabarmati River (Ahmedabad) 66 168 1,000,000 Wainganga River (Ashti) 5 20 0 Mahi River (Sevalia) 1 9 200 Indonesia Banjir Kanal River 9 20 1,000,000 Citarum River 12 37 600,000 Sunter River (Jakarta) 21 36 1,000,000 Surabaya River 12 26 7,100 Japan Sagami River (Samukawa) 1 1 490 Shinano River (Zuiun Bridge) 2 n/a 320 Tone River (Tone-Ozeki) 1 3 490 Yodo River (Hirakata Bridge) 2 n/a n/a Malaysia Klang River 4 38 460,000 Sekudai River 1 21 50,000 Korea Han River 1 n/a 14 River quality standards for drinking water <10 < 20 0 Source: UNEP-GEMS/Water < http://www.cciw.ca/gems/intro.html>
Better environmental policies are certainly needed in developing countries. However, it is important to analyse the impact these policies would have on the national economy, and particularly how they would affect the industrial sectors. Developing countries are seeking policies that could improve the environment significantly, while at the same time maintaining economic activities. This paper uses an input-output analysis to research such policy solutions in the case of river water pollution in Indonesia. Specifically, this paper will seek two categories of industrial sectors on which river water policies should be imposed as soon as possible. The first category is the heavy polluter group [11]. Industrial sectors in this group should be the target of regulations requiring industries to reduce their river water pollution.
River water pollution in Indonesia: an input-output analysis 65
The second category is the potential polluter group [12]. Industrial sectors in this group should not be allowed to grow excessively. Two criteria determine inclusion in this group: Firstly, a unit increase in the output of industrial sectors in this group will result in significant pollution to surrounding rivers. These sectors themselves might not be heavy polluters. However, various sectors that provide material inputs to them are. Secondly, sectors in this group are not the key sectors in the economy. The method to achieve the aforementioned specific goal of this paper has been available since 1970 when Leontief [13] expanded an input-output table to include pollution generation and abatement. Since then, many other suitable methods have become available. For example, in 1994 Duchin and Lange [14] developed a dynamic input-output model for the case of freshwater consumption, in 1996 Resosudarmo and Thorbecke [15] constructed a Social and Environmental Accounting Matrix, and in 1999 Garbaccio et al. [16] built up a computable general equilibrium model, both for the case of air pollution. However, all these studies utilise sophisticated methods and require very extensive data that are difficult to obtain in developing countries. The technique implemented in this paper is a simplification of the Leontief environmental input-output model and uses data that are most likely to be available in many developing countries. The Indonesian government and other developing countries could easily implement the technique utilised in this paper every year. Hence, this paper, besides being of use to the Indonesian government, is also important for other developing countries.
2 River water pollution in Indonesia
Before the economic crisis occurred, the Indonesian economy grew relatively fast. In the beginning of the 1990s, the economy grew at an annual average rate of approximately 7.5%, which is higher than the annual average growth rate of most other Asian countries, such as Korea, Taiwan, India, The Philippines, and Japan [17]. The main reason for this high economic growth was that its industrial sector (manufacturing sector) [18] developed rapidly. It grew much faster than the agricultural and services sectors, which is also true of other fast growing Asian countries. However, the industrial sector in Indonesia grew faster than that of most other fast growing Asian countries [17,19]. This high growth of the Indonesian economy caused environmental problems. One of the more important environmental problems is river water pollution. Table 3 indicates that pollution levels of Indonesian rivers, which are used as a source of drinking water, exceed the quality standard. Table 3 also shows that, in general, these levels are higher than those of China, Japan and Korea. Figures 1 and 2 exhibit river water pollution levels in several additional major rivers in Indonesia. Ciliwung and Bengawan Solo Rivers are in Java, Musi River is in Sumatra, and Mahakam River is in Kalimantan [20]. The graphs in Figures 1 and 2 indicate that industrial wastewater discharges, measured in BOD and COD, into rivers in the outer Java islands increased during the 1990s, while industrial wastewater discharges into rivers in Java decreased, albeit only by a very small amount.
66 B.P. Resosudarmo
Figure 1 Industrial liquid waste measured in BOD in several rivers
9,000 8,000 7,000 6,000 5,000
Ton 4,000 3,000 2,000 1,000 0 90/91 91/92 92/93 93/94 94/95 95/96
Musi Ciliwung Mahakam Bengawan Solo
Source: Bapedal, 1998
Figure 2 Industrial liquid waste measured in COD in several rivers
30,000
25,000
20,000
15,000 Ton 10,000
5,000
0 90/91 91/92 92/93 93/94 94/95 95/96
Musi Ciliwung Mahakam Bengawan Solo
Source: Bapedal, 1998
Since the end of the 1980s, Indonesia has developed a national clean river water program known as Program Kali Bersih or Prokasih. It mainly aims to reduce the pollutants discharged into the major rivers in Indonesia, and to improve the water quality of those rivers to meet or surpass the Indonesian standard. This program also attempts to strengthen human and institutional abilities in the management of river water quality and of riverbanks. Regional governments are responsible for implementing Prokasih in their regions. Prokasih-related activities vary from one region to another. Nationally, the Environmental Impact Assessment Agency (Badan Pengendalian Dampak Lingkungan or Bapedal) provides general guidance and coordinates all regional activities related to Prokasih.
River water pollution in Indonesia: an input-output analysis 67
2.1 Prokasih from 1988 to 1994
The implementation of Prokasih during this period can be divided into four stages: 1 Preparation stage: in this stage, Bapedal and regional governments conducted the necessary preparations to start implementing Prokasih. It was most important to develop Prokasih institutionally within the regional government agencies. Other preparations included procuring water-monitoring equipment, training, and developing an Indonesian river water quality standard. 2 Mobilisation stage: the main activity of this stage was to inform society, particularly industries located along riverbanks, about the program. Industries were the primary target, since Prokasih, not only during this period but also up until now, aims to reduce water pollution from industries. 3 Increased Participant stage: firstly, during this stage, it was expected that more industries should participate in Prokasih. Secondly, more regional governments implemented the program, meaning more rivers were included in the Prokasih program. 4 Improvement stage: the main goal of this stage was to start improving the water quality of rivers under the Prokasih program. In terms of the number of participants, Prokasih was a relative success. Firstly, the number of industries participating in the program increased from 381 factories in 1989/1990 to 1395 factories in 1994/1995. Secondly, the number of rivers included in the Prokasih program increased from 18 rivers in 1989/1990 to 31 rivers in 1994/1995 [21–25].
2.2 Prokasih after 1994
In 1994, Prokasih was extended into Proper Prokasih. Proper Prokasih is a form of environmental certification administered by the Bapedal that annually ranks factories and awards five different scores (black, red, blue, green, and gold) according to their success in reducing waste discharges, mainly wastewater. The results are published in newspapers and other media so that they are publicly known. The five different scores are as follows: • gold is for factories that potentially pollute a river, but are able to treat their waste so that the amount discharged is less than 10% of their total waste • green is for factories that are able to treat their waste so that less than 50% of their waste is discharged into the river • blue is for factories that are able to comply with the Bapedal’s wastewater discharge standard • red is for factories conducting wastewater treatment activities, but which are not able to comply with the Bapedal’s wastewater discharge standard • black is for factories without any wastewater treatment facilities. All their waste is discharged directly into the river.
68 B.P. Resosudarmo
The main idea of this program is to increase public pressure on industries along the rivers, so that they will reduce their waste discharges according to Bapedal’s targets. In June 1995, 187 factories were monitored under the Proper Prokasih program. In March 1997, 173 factories continued to be rated in the program. The 14 factories that dropped out of the program did so either because they closed down or were considered insignificant polluters. Table 4 shows the result of this program. It can be seen that more factories monitored under this program received green and blue labels in March 1997 than in June 1995. Proper Prokasih, hence, was considered very successful [26–28].
Table 4 Results from the Proper Prokasih Program
Jun-95 Mar-97
Factories Factories
Gold 0 0% 0 0% Green 4 2% 7 4%
Blue 59 32% 81 47% Red 118 63% 82 47% Black 6 3% 3 2% Total 187 100% 173 100%
Since March 1997, Bapedal has not published any results of their Proper Prokasih program and it is unclear whether or not they are still implementing it. However, now that Indonesia has adopted a decentralisation policy, many regional governments are showing an interest in controlling river water pollution. Among these regional governments are those of Jakarta, Surabaya and Semarang. The specific goal of this paper, which is to identify two types of industrial sectors on which river water policies should be imposed as soon as possible, will certainly help these regional governments in developing policies to control river water pollution.
3 Input-output analysis
As mentioned in the introduction, one of the available methods in achieving the specific goal of this paper is the input-output analysis. This paper modifies the traditional Leontief Input-Output (I-O) table by adding the pollution quantity or abatement cost account at n+1 row where n is the number of production sectors in the economy (Figure 3). The pollution and abatement costs are treated as primary inputs (value added). In this case, it is a ‘bad’ value added.
River water pollution in Indonesia: an input-output analysis 69
Figure 3 I-O Table with a Pollution Account
Producer Consumer Sector Final Total Sector Consumption Production 1 2 ... n
1 x11 x12 ... x1n f1 X1
2 x21 x22 ... x2n f2 X2 ...... n xn1 xn2 ... xnn fn Xn
Pollution* P1 P2 … Pn
Value Added v1 v2 ... vn
Import m1 m2 ... mn
Total Input X1 X2 ... Xn Note: * For the I-O Table with an Abatement Cost account, this Pollution column is substituted with the Abatement Cost column (Cj)
Recall that an I-O table is a matrix presenting complete transactions of goods and services in the industrial sectors of an economy. Each row in the matrix shows how much of its output a certain sector (under the heading of Producer Sectors in Figure 3) sells to various other sectors (under the heading of Consumer Sectors in Figure 3) in the economy. On the other hand, each column in the matrix exhibits how much a certain sector consumes of goods and services produced by various other sectors in the economy. Besides transactions among industrial sectors, an I-O table also records two other important transactions. Firstly, it notes down the amount of goods and services consumed by households and government and how much is exported (column Final Consumption in Figure 3). Secondly, it presents the amount of goods and services imported by a sector and (row Import in Figure 3) and the amount of compensation to labour and capital owners (row Value-Added in Figure 3). Pj in Figure 3 states the amount of pollution (in kg) resulting from the production activity of sector j. Let us define pn+1,j as pollution intensity, which is the amount of pollution loaded by sector j per one unit output of sector j:
pn+1,j = Pj / Xj (1) where Xj is the total output of sector j. Cj (see the note in Figure 3) states how much is needed to clean the environment by sector j. Then cn+1,j is defined as the abatement cost per one unit output of sector j:
cn+1,j = Cj / Xj (2) Several indices measuring the relationships between production activities and pollution quantities, as well as between production activities and pollution abatement costs can be calculated. The indices are as follows:
70 B.P. Resosudarmo
• Index of pollution intensity The index of pollution intensity compares pollution intensities from all sectors in the economy. A sector with a high index of pollution intensity (higher than one) is considered a high polluter sector. The index for sector j is: n. p = n+1, j IP j n (3) ∑ p n+1,i i=1 • Pollution effect and index of pollution effect The pollution effect defines the quantity of pollution emitted as a result of an increase in one unit final demand of sector j.
n P = α (4) E j ∑ p n +1,i . ij i =1 P α where E j is the pollution effect caused by sector j and ij is an element of the usual Leontief inverse matrix [29, p.15, 30] P The index of pollution effect ( IE j ) is an index enabling us to compare the quantity of pollution emitted caused by an increase in the final demand of a sector and that is caused by an increase in the final demand of other sectors.
n.E P P = j IE j n (5) P ∑ E i i=1 • Index of abatement cost intensity The index of abatement cost intensity compares an abatement cost per unit output of a certain sector to the average abatement cost per unit output of all sectors in the economy. The index can be determined using the following equation:
n.c = n+1, j IC j n (6) ∑ c n+1,i i=1 Cleaning pollutants from a sector with a high index of abatement cost (higher than one) is relatively expensive. • Abatement cost effect and index of abatement cost effect The abatement cost effect is the additional cost needed to clean the emitted pollution caused by an increase in a unit of final demand j.
River water pollution in Indonesia: an input-output analysis 71
n C = α E j ∑ c n +1,i . ij (7) i =1
C where E j is the abatement cost effect [31]. The index of abatement cost effect is defined as follows:
n.E C C = j IE j n (8) C ∑ E i i=1 • Output multiplier and index of output multiplier The output multiplier measures the increase of the total output of the economy caused by a unit increase in the final demand of sector j. The formula to calculate the output multiplier of sector j is [29, p.103]:
n = α O j ∑ ij (9) i=1 The index of output multiplier compares an output multiplier of a certain sector to the average output multipliers of all sectors in the economy.
n n. α ∑ ij O = i=1 = j IO j n n n n (10) α α ∑∑ ij ∑∑ ij i ==11j i ==11j where IO j is the index of output multiplier of sector j. Sectors with IO j greater than or equal to one are called key sectors in the economy, or the key economic sectors. Using the indices just mentioned, policies to control water pollution can be developed based on the following strategies:
1 Clean the sectors that have a high IPj index and a low IC j index. These sectors are the heavy polluters, but their costs are relatively low. To determine the priority sector in this category, a new parameter could be developed:
IP ˆ = j M j (11) IC j ˆ Hence, sectors with high M j are the priority sectors to be cleaned up or key sectors of pollution abatement. Let us call this parameter the Parameter of Effective Pollution Abatement.
72 B.P. Resosudarmo
2 Prevent a significant increase in, or reduce as much as possible, the outputs of P C sectors for which both IE j and IE j indices are high and where the sectors are not key sectors in the economy. Firstly, a new parameter should be developed to find these sectors:
ˆ = P ⋅ C E j IE j IE j (12)
This index can be called the Parameter of Effective Pollution Prevention. Sectors with high Êj are the sectors where an increase in their output causes high quantities of pollution to be emitted and the associated abatement costs are high. Secondly, one should check whether or not the sectors with high Êj indices are among the key economic sectors; i.e. whether or not their indices of output multiplier are greater than or equal to one. If not, then they are the sectors where an increase in their final demands should be avoided.
4 Data sources
This paper mainly utilises the Indonesia Input-Output (I-O) Table for 1995, which was published by the Central Bureau of Statistics. Industries in this table are classified into 66 sectors. The table is extended to include river water pollution sectors and their abatement costs. Information on coefficients of pollution intensity and abatement costs is available from the World Bank Industrial Pollution Projection System (IPPS) [32]. These coefficients were derived from the US condition. The coefficients of pollution intensity might underestimate the true pollution intensity in Indonesia, while the coefficients of abatement costs most likely overestimate the true abatement costs. Since no detailed information on pollution intensity and abatement costs in Indonesia is available yet, adopting the numbers from the World Bank IPPS is a reasonable option. Pollution intensity coefficients are stated in kg/millions of 1987 USD. This coefficient is then converted to kg/millions of 1995 rupiah. Based on data availability, there are two variables measuring river water pollution, Biological Oxygen Demand (BOD) and Total Suspended Solids (TSS). Note that the IPPS data is only for water pollution from manufacturing sectors. River water pollutants from primary and services sectors, as well as domestic sources, are not taken into account. Accordingly, the I-O analysis was applied only to manufacturing sectors. The IPPS data follows the four-digit International Standard for Industrial Classification (ISIC-4). Information from the IPPS then needs to be adjusted to conform to sectors in the Indonesia I-O Table.
5 Results and discussion
This section first presents the calculated indices of pollution intensity, pollution effect, abatement cost intensity, abatement cost effect, and output multiplier. Secondly, policies to control water pollution in Indonesia are developed, using parameters of effective pollution abatement and effective pollution prevention, and indices of output multiplier.
River water pollution in Indonesia: an input-output analysis 73
Table 5 shows manufacturing sectors that emit wastewater to surrounding rivers and their indices of pollution intensity, pollution effect, abatement cost intensity, abatement cost effect, and output multiplier.
Table 5 The Indonesian economy and river water pollution
No. IO Sector BOD TSS j p c p c ICj IPj IE j IE j ICj IPj IE j IE j 1 Meat Products 0.03 0.05 0.19 0.17 0.00 0.00 0.05 0.06 1.10 2 Food Products 4.60 4.93 9.13 7.35 0.05 0.08 0.28 0.22 1.12 3 Oils and Fats 0.40 0.29 0.71 0.86 0.02 0.02 0.15 0.16 1.13 4 Grain Mill Products 0.00 0.00 0.05 0.07 0.00 0.00 0.10 0.11 1.17 5 Bakery Products 0.00 0.00 0.56 0.21 0.00 0.00 0.08 0.06 1.05 6 Sugar Factories and Refineries 0.19 3.57 6.36 0.40 0.01 0.27 0.59 0.12 0.99 7 Other Food Products 0.02 0.01 0.57 0.27 0.00 0.00 0.12 0.10 1.10 8 Drinking Industries 2.38 3.07 6.80 4.18 0.16 0.29 0.76 0.48 1.10 9 Tobacco Manufacturers 0.01 0.00 0.82 0.87 0.00 0.00 0.21 0.20 0.89 10 Spinning, Weaving, & Finishing Textiles 0.12 0.16 0.46 0.52 0.01 0.01 0.09 0.11 0.87 11 Textile, Cloth, and Leather Products 0.38 0.17 0.71 1.34 0.03 0.02 0.14 0.27 1.10 12 Cane, Wood, Rattan Products 0.06 0.06 0.29 0.46 0.01 0.01 0.12 0.15 1.12 13 Pulp, Paper, and Paperboard 5.96 5.89 13.79 12.36 0.74 1.04 2.53 1.98 1.02 14 Fertilisers and Pesticides 0.12 0.08 0.43 0.67 0.86 0.77 1.46 1.76 0.81 15 Chemical Products 5.53 1.22 2.90 10.36 1.04 0.33 0.81 2.48 0.94 16 Petroleum Refineries 0.20 0.15 0.29 0.35 0.03 0.04 0.09 0.09 0.86 17 Rubber and Plastics Products 1.05 0.29 1.07 3.35 0.25 0.10 0.37 0.94 1.07 18 Nonmetallic Mineral Products 0.03 0.03 0.45 0.91 0.00 0.00 0.15 0.24 0.99 19 Cement, Lime, and Plaster 0.00 0.00 0.57 0.58 0.14 0.23 0.55 0.40 1.03 20 Iron and Steel 0.02 0.02 0.27 0.65 9.97 17.26 33.43 20.67 0.92 21 Nonferrous Metals 3.83 4.96 9.51 6.47 2.04 3.80 7.41 4.29 0.99 22 Metal Products 0.03 0.01 0.66 0.75 0.03 0.02 4.19 2.67 0.93 23 Machinery and Equipment 0.03 0.01 0.27 0.51 0.01 0.00 0.50 0.47 0.86 24 Shipbuilding and Repairing 0.01 0.00 0.15 0.26 0.00 0.00 0.90 0.64 0.83 25 Other Manufacturers 0.01 0.00 1.63 1.61 9.61 0.71 2.91 19.56 1.00 = candidates to become key sectors of pollution abatement
= candidates to become key sectors of pollution prevention
74 B.P. Resosudarmo
Recall the definition of the index of pollution intensity. A sector, having an index of pollution intensity greater than one, is considered to be a high polluter sector. Meanwhile, having an index of abatement cost intensity less than one means that it is relatively cheap to clean pollutants from that sector. A sector in which one of these scenarios operates should then be a candidate to become a key sector of pollution abatement; i.e. cleaning pollutants from this sector should be prioritised. From Table 5, it can be seen that all sectors, except Rubber and Plastics Products sector for the case of BOD and Chemical Products and Other Manufactures sectors for the case of TSS, are candidates for key sectors of pollution abatement. Recall also the definition of the pollution effect index. A sector with an index of pollution effect greater than one means that a unit increase in the final demand in the sector causes an increase in river water pollution more than the average caused by other sectors. A sector with an index of abatement cost effect greater than one means that a unit increase in the final demand of this sector increases the total abatement cost needed more than the average induced by other sectors. In addition, a sector with an index of output multiplier greater than one is a key economic sector. One then can conclude that a sector with either an index of pollution effect or an index of abatement cost effect greater than one, and an index of output multiplier less than one, is considered to become a key sector of pollution prevention. From Table 5, it can be seen that, for the case of BOD, Sugar Factories and Refineries, Chemical Products, and Nonferrous Metals sectors are such candidates. While for the case of TSS, Fertilisers and Pesticides, Chemical Products, Iron and Steel, Nonferrous Metals, and Metal Products sectors are candidates. This paper then calculates the parameters of effective pollution abatement and of effective pollution prevention to find key sectors of pollution abatement and of pollution prevention. As mentioned before, there are two parameters measuring river water pollution emitted from a sector. Parameters of effective pollution abatement need to be calculated to discover key sectors of pollution abatement or priority sectors to be cleaned up in Indonesia. Since there are two variables (BOD and TSS) measuring river water pollution emitted from a sector, an arithmetic average of the effective pollution abatement parameter is applied, meaning
ˆ = 1 ⋅ ( ˆ + ˆ ) M i 2 M i,BOD M i,TSS (13)
Table 6 presents ten sectors in which the effective pollution abatement parameters are high. Sector number one in that list is the sector with the highest number of effective pollution abatement parameters, sector number two the second highest, etc. To determine key sectors of pollution prevention or priority sectors where an increase in output should be avoided, this paper first calculates sectoral parameters of effective pollution abatement. For the same reason as in the case of the effective pollution abatement parameter, an arithmetic average of the effective pollution prevention parameter is calculated, meaning
ˆ = 1 ⋅ ( ˆ + ˆ ) Ei 2 Ei,BOD Ei,TSS (14)
Secondly, this paper checks whether or not the sectors with high effective pollution prevention parameters are key economic sectors; i.e. whether or not their indices of
River water pollution in Indonesia: an input-output analysis 75
output multiplier are greater than or equal to one. Ten sectors with the highest parameters of pollution prevention and with indices of output multiplier less than one are shown in Table 6. The list is sorted from the sector with the highest number of pollution prevention parameters to the sector with the lowest.
Table 6 List of priority sectors
No. Pollution Abatement Sectors Pollution Prevention Sectors (1) (2) (3) 1 Sugar Factories and Refineries Iron and Steel 2 Meat Products Nonferrous Metals 3 Grain Mill Products Chemical Products 4 Spinning, Weaving, & Finishing Textiles Metal Products 5 Nonferrous Metals Fertilisers and Pesticides 6 Drinking Industries Sugar Factories and Refineries 7 Iron and Steel Tobacco Manufacturers 8 Cement, Lime, and Plaster Shipbuilding and Repairing 9 Food Products Non-metallic Mineral Products 10 Non-metallic Mineral Products Machinery and Equipment
5.1 Policy implications
The policy implications of input-output analysis in this paper can be summarised as follows: The ten sectors with high parameters of effective pollution abatement (second column in Table 6) should be given high incentives to clean up their pollution as much as possible because it is relatively inexpensive. One combination of incentives is providing pollution abatement subsidies and strict control of their wastewater discharge. Another incentive is applying a high progressive wastewater discharge tax to those sectors. Consumers should be given a high disincentive to increase their demands on outputs of the ten sectors with high parameters of effective pollution prevention and with indices of output multiplier less than one (third column in Table 6). Increasing demand on these sectors induces relatively large amounts of river water pollution that is relatively expensive to clean up. One disincentive is applying a relatively higher sales tax on the outputs of those sectors just mentioned [33]. The Indonesian government might want to adopt the aforementioned policies as an effective and efficient way to achieve higher quality river water in Indonesia.
6 Conclusion
This paper has shown, using the case of Indonesia, a procedure/method to develop an effective strategy in implementing policies to control river water pollution in developing countries. Using this method, a developing country will be able to identify the ten sectors in the country with high parameters of effective pollution abatement. Such sectors should
76 B.P. Resosudarmo
be given high incentives to clean up their pollution. A developing country will also able to determine the ten sectors with high parameters of effective pollution prevention and with indices of output multiplier less than one. Consumers should be given a high disincentive to increase their demands on outputs of these sectors. However, it is important to note several weaknesses of the method developed in this paper: Firstly, an I-O analysis assumes that all industries have a fixed proportion production function; i.e. there is a linear relationship between output and inputs and inputs are not substitutable. In many cases, this assumption is only true for a small change in output. Secondly, an I-O analysis assumes that the structure of an economy remains the same over time. This assumption can only be applied to a short-run analysis. Thirdly, although there are 66 different categories of sector in this paper, each one still covers a large range of industrial characteristics. A more disaggregated sector will produce a more detailed result. Fourthly, the use of a national I-O table, rather than a regional I-O table, might underestimate or overestimate the local impact of policies developed in this paper. One should also remember that river water pollution is only one of several important pollution problems in developing countries; others include urban air and ground water pollution. It is most likely that implementing policies mentioned in this paper would not solve pollution problems other than that of river water. Further study aiming to overcome the weaknesses of the method in this paper, but yet able to keep the method simple, is certainly needed.
References and Notes 1 WRI, World Resources Institute (1998) World Resources 1998-99, Washington, D.C., WRI. 2 FAO, Food and Agriculture Organisation of the United Nations (2001) Global Forest Resources Assessment 2000, Rome, FAO. 3 UNEP, United Nations Environment Programme (2002) Global Environmental Outlook-3, Rome, UNEP. 4 Ostro, B. (1994) ‘Estimating the health effects of air pollutants: a method with an application to Jakarta’, Policy Research Working Paper No. 1301, World Bank. 5 Ostro, B., Sanchez, J.M., Aranda, C. and Eskeland, G.S. (1996) ‘Air pollution and mortality: results from a study of Santiago, Chile’, Journal of Exposure Analysis and Environmental Epidemiology, Vol. 6, pp.97-114. 6 Chestnut, L.G., Ostro, B.D. and Vadakan, N.V. (1997) ‘Transferability of air pollution control health benefits estimates from the United States to developing countries: evidence from the Bangkok Studi’, American Journal of Agricultural Economics, Vol. 79, No. 1997, pp.1630-1635. 7 This is the Indonesian standard quality for river water, which is utilised as a source of drinking water. 8 Esrey, S.A. et al. (1990) Health Benefits from Improvement in Water Supply and Sanitation, WASH Technical Report No. 66. Arlington, VA., Water and Sanitation for Health Project, July. 9 Brockerhoff, M. and Derose, L.F. (1996) ‘The impact of preventive health care’, World Development, Vol. 24, No. 12, pp.1841-1857. 10 DepKes, Departemen Kesehatan (2000) Profil Kesehatan Indonesia 1999, Jakarta, Departement Kesehatan. 11 Later on in this paper, sectors in this group will be called Pollution Abatement Sectors. 12 Later on in this paper, sectors in this group will be called Pollution Prevention Sectors. 13 Leontief, W. (1970) ‘Environmental repercussions and the economic structure: an Input- Output Approach’, The Review of Economics and Statistics, Vol. 52, No. 3, pp.262-271.
River water pollution in Indonesia: an input-output analysis 77
14 Duchin, F. and Lange, G.M. (1994) ‘Strategies for environmentally sound economic development’, in A.M Jansson, J. Hammer, C. Folke, and R. Costanza (Eds.), Investing in Natural Capital: The Ecological Economics Approach to Sustainability, Island Press, Washington, pp.250-265. 15 Resosudarmo, B.P. and Thorbecke, E. (1996) ‘The impact of environmental policies on household incomes for different socio-economic classes: the case of air pollutants in Indonesia’, Ecological Economics, Vol. 17, pp.83-94. 16 Garbaccio, F.G., Ho, M.S. and Jorgenson, D. (1999) ‘Controlling carbon emission in China’, Environment and Development Economics, Vol. 4, No. 4, pp.493-518. 17 Asia Pacific Economics Group (1998) Asia Pacific Profiles, The Australian National University, Canberra. 18 Please note that what are generally known as industrial sectors, in input-output tables, are defined as manufacturing sectors or manufacturing industries. 19 Heaton, G. and Resosudarmo, B.P. (2000) ‘Technology and environmental performance: leveraging growth and sustainability’, in D.P. Angel, T. Feridhanusetyawan and M. Rock (Eds.), Towards Clean Shared Growth in Asia: A Policy and Research Agenda, forthcoming. 20 The Indonesian rivers in Table 3 are all in Java. 21 PPSDML-UI, Pusat Penelitian Sumber Daya Manusia dan Lingkungan-Universitas Indonesia (1991) Seminar Report on Clean River and Urban Environmental Management, CRHRE – UI, Jakarta. 22 NRMP (1996) ‘Economic benefits of improved water quality in the Ciliwung River, Jakarta’, Natural Resource Management Project (NRMP) Working Paper No. 68, Jakarta. 23 Resosudarmo, I.A.P., Resosudarmo, B.P. and Isham, B. (1997) ‘The Indonesian clean river program (Prokasih) as perceived by the people residing along the rivers in Jakarta’, Indonesian Journal of Geography, Vol. 29, No. 74, pp.47-64. 24 Widayanto, Y., Kusrestuwardhani, Resosudarmo, B.P. and Resosudarmo, I.A.P. (1999) ‘Pengembangan Wilayah Dalam Hal Perbaikan Kualitas Air Sungai: Nilai Air Sungai Bersih Bagi Masyarakat Sekitar Sungai di DKI Jakarta (Improving river water quality as an aspect of regional development: the value of clean river water for people residing along the rivers in Jakarta)’, in Alkadri, Muhdi, Suhandojo, Maryadi (Eds.) Tiga Pilar Pengembangan Wilayah: Sumber Daya Alam, Sumber Daya Manusia dan Teknologi, BPP Teknologi, Jakarta. 25 The number is still relatively very small compared to the total number of rivers in Indonesia, or even to the number of rivers in Java. 26 Afsah, S., Laplante, B. and Makarim, N. (1996) ‘Program-based pollution control management: the Indonesia Prokasih Program’, Policy Research Working Paper No. 1602, PRDEI Division, the World Bank. 27 Wheeler, D. and Afsah, S. (1996) Going Public on Polluters in Indonesia: Bapedal’s Proper Prokasih Program, East Asian Executive Report, The World Bank. 28 Afsah, S. (1998) ‘Impact of financial crisis on industrial growth and environmental performance in Indonesia’, Briefing-note from USAEP to Bapedal. 29 Miller, R.E. and Blair, P.D. (1985) Input-Output Analysis: Foundation and Extensions, Prentice-Hall, New Jersey. P 30 The formula for E j is equivalent to the formula for Type I household income multiplier [29, p.106]. C 31 The formula for E j is also equivalent to the formula for Type I household income multiplier [29, p.106]. 32 This information is publicly available at