1This paper analyses the water supply and sanitation in the city of Chennai. Section 1 briefly describes the physical and supply situation, while section 2 deals with financing aspects, tariffs and cost management. Then the paper concentrates on the analysis of contracting out, organisational and institutional innovations that constitute the particularity of Chennai in the Indian urban water scenario. It concludes by putting Chennai in perspective within the Indian urban water management models. This paper is based on various field interviews with the Chennai Metro water Board, with the Tamil Nadu Slum Clearance Board, NGOs, consulting companies, and service companies.

The physical and supply situation

The Chennai Metropolitan Water Supply and Sewerage Board

2The water supply in Chennai is ensured by the Water Supply and Sewerage Board, constituted on 1.8.1978 by an Act. Compared to other Indian cities, the situation is characterized by a structural shortage of resources. The share of ground water is relatively high for the Board water itself, further complemented with a large amount of extra-Board supply (36 % of the supply is not routed through the Board). It serves a population of 4.6 million, living in 174 km² and 3.10 lacs “consumers”, despite the fact that it charges municipal taxes from 4.87 “assessees”. Eighty per cent of the water originates from the ground, which is the particularity Chennai faces in India, due to the scarcity of local surface sources. The treatment capacity is 620 mld2, and the water actually produced is about 480 mld in good monsoon years. Indeed, overall, there is a water resource shortage in the Chennai area. In 1999-2000, due to monsoon failure, the piped water supply had to be reduced from 440 to 250 mld. The official supply figure is 110 lpcd for the connected users. But in 1999-2000, during the months of drought, the Board could maintain piped supply for only about 70 % of the population, at the rate of about 55-60 lpcd. The Board’s water only accounts for 64 % of the sources. Another 4 % comes from private tankers, 10 % from community tanks and wells, 21 % from in-house wells. Another estimate, for normal months, is that the Metrowater system provides 44 % of the consumption. Solid waste management is mostly taken care of by the Chennai Municipal Corporation. Sewerage and sanitation is relatively poor, despite the fact that the Board gets a good amount of its financing through an afferent tax. In that scarcity context, the board has initiated recycling and reuse experiments that have to be acknowledged and duly recognized. The waste water treatment capacity is of 268 mld, with a network of 119 pumping stations, fed by 260,838 sewer connections. Up to now, treatment is of 60 % of the Board-water (a substantially lesser percentage if one takes into account the non-Board supplies), divided in 5 % re-used by the industry, 10 % re-used by the electrical industry, and 45 % put back to rivers, as shown by interviews in the sewerage wing of the Board. The area suffers from the lack of perennial rivers, from a rainfall pattern of a short duration, and from storage capacities that are anyway limited, given the proximity of the seashore that alters the thickness of the aquifer (only a few meters). The Board has engaged leak detection and rectification works, rainwater harvesting (with NGO’s, voluntary groups and consumers’ associations), control and regulation of ground water extraction (Ground Water Regulation Act, 1987) and re-use of secondary treated sewage for industrial purposes. But ultimately, it is very surprising that an integrated vision of the resource within a cycle is still not generalized, and that no techno-economic comparisons are made between the long-term cost of recycling water and the long-term cost of such high capital-intensive projects, such as the Krishna river canal.

The Water Supply

380 % of the water is from ground-water, which is the particularity Chennai faces in India, due to scarcity of local surface sources. The Board enjoys the monopoly of extraction at Manali Hills, both the first ground source and the industrial district of the Chennai area. There is a seasonal water resource shortage in the Chennai area. But the treatment capacity is 620 mld3, and the water actually produced is about 480 mld in good monsoon years. Overall, the most important sources are at Manali Hills, but the Board also buys 20 mld from farmers, under a tripartite agreement between the Water Board, the Electricity Boards and the farmers. Over recent years, plans have been addressed to distribute additional water from the Krishna river project (the Krishna water has been received since 1996, and has been enhanced the supply by 200 mld). The next stage is the works for draw of 530 mld of raw water from the Chembarambakkam tank treatment and clear water transmission mains from the Chembarambakkam treatment plant to the city, to be completed by 2004.

4Let us detail the tripartite agreement with the TNEB (Tamil Nadu Electricity Board) and farmers. Farmers are paid Rs.26 per hour of running by the Board, and have to run 18 hours a day (producing in each borewell 400 kl/day). The electricity is directly paid by the Water Board to the Tamil Nadu Electricity Board, at a rate of Rs.3 per kWh, as compared to Rs.2 for domestic usage and 75 paise for small domestic usage). But, if this scheme provides easy money to these farmers, as is often felt in local debates, this also highlights the fact that there is an implicit rural/urban competition for water in the Chennai area. The MIDS (1995) states that “water for urban/industrial uses can be obtained by restricting the water used in agriculture”. The statement is to be seen not so much in terms of direct competition, but rather in terms of a better use of the resource, since “irrigation consumes 70-80 % of the water resources of the State, even small efficiency improvements will release substantial quantities for non-agricultural uses”.

5The official supply figure is 110 lpcd for the connected users. But the area faces severe droughts, and in the year 1999-2000, during the months of drought, the Board could maintain piped supply for only about 70 % of the population, at the rate of about 55-60 lpcd. For the remaining 30 %, the water supply had to be temporarily provided through 5 500 stationary tanks and also through mobile supply by about 3,800 lorry trips per day (35 mld). During the May 2001 drought, interviews confirmed that the level had dipped to 35 lpcd, with one-third of the piped streets being actually empty, due to low pressure, and the tank supply by the Board went up to 50 mld.

6Groundwater was extensively used, through a total of 5000 pumps, especially for the poor and slum dwellers. Beyond the official system, experts quote the figure of an additional one-third for local, unofficial, groundwater extraction. The CMWSSB figures are, however, to be complemented with the private sources of water. Indeed, the Board’s water only accounts for 64 % of the sources. Another 4 % comes from private tankers, 10 % from community tanks and wells, 21 % from in-house wells (WSP 1999). Another estimate, for normal months, is that the Metrowater system provides 44 % of the consumption (Murugappa 1997, after a survey on 10,000 households). But this figure must rather stand for the domestic part of the all-users segment.

7It is especially worth comparing with sample surveys conducted by the company Economic Perspectives. The survey covers 4,500 users, domestic, non-domestic, on public taps and hand pumps, or on mobile lorry supply. The preliminary results of the survey show that 85 % of the domestic consumers4 have their own source (71 % of them have a bore well, and 28 % an open well). Five per cent of the domestic users use the Metrowater for drinking only, 3 % for drinking and cooking, 70 % for drinking, cooking and kitchen washing, 19 % for these usages plus laundering, 2 % for all purposes. Seven per cent of the domestic users consume mineral water, 26 % use the water tankers of the Board. As far as non-domestic users are concerned, 82 % have the metro-water connection, but 75 % have anyway their own source (91 % bore well. 8 % open well), and 60 % are actually pipe supplied, of whom only 76 % acknowledge a sufficient quantity. People on public taps/tankers5 are segregated as such: 56 % taps or pumps, 44 % on tanks. Forty-four per cent of the total category use the metro-water for all usages, and 56 % use it for drinking and cooking only; 63 % of these people feel the supply is insufficient. The usage of the water supplied by the Board is, then, a differentiated usage.

Sewerage and drainage

8The waste water treatment capacity is 268 mld, with a network of 119 pumping stations, fed by 260,838 sewer connections (2,106 km of mains). Since 1995, 180 crores have been spent in: construction of 60 new pumping stations; improvements to 56 existing pumping stations; and improvements and expansion of 6 sewage treatment plants. Plans to treat an additional quantity of 264 mld (Rs. 615 crores in treatment plants and Rs. 640 crores in strengthening the system accordingly) are proposed for the next 10 years. Up to now, treatment is of 60 % of the Board-water (thus a substantially lesser percentage if one takes into account the non-Board supplies), divided in 5 % re-used by the industry, 10 % re-used by the electrical industry, and 45 % put back to rivers, as shown by interviews with the sewerage wing of the Board show.

9In practice, the survey carried out by Economic Perspectives gives a clear idea of the situation: while 100 % of their ‘domestic users’ have toilets and 99 % a sewer connection, for people connected through public taps and pumps, only 44 % have an access to toilets. Of these 44 %, only 44 % are connected to sewers, 18 % to storm water drains, 33 % to septic tanks, and 5 % let it out into canals.

Water conservation policy

10The area suffers from the lack of perennial rivers, from a rainfall pattern of a short duration, and from storage capacities that are anyway limited, given the proximity of the seashore that alters the thickness of the aquifer (only a few meters), as confirmed by interviews. The Board has engaged in leak detection and rectification works, rainwater harvesting (with NGOs, voluntary groups and consumers’ associations (see Singh 2001), control and regulation of ground water extraction (based on the Ground Water Regulation Act, 1987 that applies for above 5 HP pumps for commercial purposes; in practice, there have been no new licenses since 1993, and no renewal since 1997), and the re-use of secondary treated sewage for industrial purposes. The latter deserves being illustrated. An oil refinery, a fertilizer plant and a power plant have set up treatment units, supplied with a total of 60 mld of secondary sewage effluent by the Board. Regarding rainwater harvesting, a survey conducted by Economic Perspectives shows that only 2 % of the domestic consumers actually harvest rain water, while 48 % declare that they are keen on doing so, while 17 % declare they have no space or money to do so. As far as resource conservation is concerned, it should be noted that the agricultural sector, enjoying power free of charge, sells water for commercial purposes to tankers. Interviewed Metro officers have confirmed that, during drought situations, they do not insist on applying the Ground Water Regulation Act. In any case, there are loopholes in the Act that reduce its effciency: penalties are too low to be a deterrent and farmers get free electricity, even though they use the water for commercial purposes. Attempts to reduce leakages have been undertaken in a pilot distribution area, through new taps and non-corrosive techniques. Interviewed officers declared it had brought down the leakages from 24 % to 6 %, but this measure still has to be generalized and sustained. More substantially, 1.7 lacs of connections have been recently renewed at a cost of 160 crores. According to the interviews, this very measure has been the principal one in allowing the supply on a daily basis during droughts (during earlier droughts, it was on an alternate basis). The economic rationale of this measure, fully supported by the Board, is that, despite not allowing the provision of more water, it allows the base of consumers to be increased: once the pressure in pipes has improved, new connections become feasible.

The financial position, tariffs and cost management

Financing of the CMWSSB

11The Board managed last year to cover its operational expenses with a surplus (total income of Rs. 207 crores in 1999-2000, for a total expenditure of Rs. 181.5 crores, and a net surplus of Rs. 25.5 crores). However, the capital expenditure still comes from or routed through the state government, raising the issue of the long-term sustainability of water supply and sanitation in Chennai. More precisely, if the state government has, since 1997, taken the step not to provide financial support, it goes on routing the World Bank loans that it ultimately guarantees. The Board is presently considering an indexation of tariffs on a RPI minus X formula. As of now, the Board is still close to the functioning of an administrative body, as far as revenues are concerned. Indeed, a noticeable part of the ‘operational’ revenue (27 %) comes from the collected tax, which is rather an administrative prerogative than correlated to any supply or sanitation. As far as the financial results are concerned, its importance should be highlighted all the more, since increase of the tax value explains 64 % of the total increase in ‘operational revenue’. Interviewed officers stated that in return for these taxes, “there is no necessity to provide service” and that “they are advances to implement schemes”. However, as far as capital expenses are concerned, in 1999-00, 165 crores were granted by the government, while total loans from the government amounted 656 crores, and loans from financial institutions only 129 crores (CMWSSB 2000). The overall efficiency in collecting taxes varies yearly between 80 and 90 %, while bills are collected with a level of only 80 %, as interviews show. To sum up, and as far as water supply is concerned, the Board gets more than half of its operational income for water from only 40 big consumers only. It can retain them only on the basis of the Water Ground Conservation Act, that is, as a depositary of the public authority. For other consumers, it has, as shown here, an administrative behaviour. Then, the statement made during interviews that “the Board has shown operational profits from 1996 and real profits from 1998” has to be reexamined in the light of the fact that the Board does not supply either piped water to all its users nor extensive sanitation, despite continuing to collect “taxes”. Indeed, in terms of organizational efficiency, the Twinning Arrangement report states that for the year 1999-2000, the collection ratio is only 81.5 %6. The evolution of the total arrears was from 161 debtor days in 1996 to 221 in 2000.

Commercial aspects

12The commercial aspects, in the spirit of the Board, only pertain to their “consumers”, who amount to 3.10 lacs, despite the fact that they charge taxes from 4.87 “assessees”, actually behaving as a municipal administration. In a situation of recurrent drought, the population is concerned with the ability of the Board to face it. The Board indicates (CMWSSB 2000) that, since the 1980s, it has been able to decrease cost per capita and per month from 10 to 1.67 Rs, according to the value 2000. It mentions as a success that, during 1998-99, total of 1,88,620 complaints were received and redressed, whereas during the 1999-2000, the total number of complaints received and attended to was only 1,66,610 only. However, it can also be noticed that (i) investments had already been made by that time, taking into account the structural dimension of droughts, and that (ii) the private supply sector was by that time quite developed.

13Compared to Calcutta and Mumbai, various measures were taken in the last few years to ameliorate relations with consumers through improving billing and collection, providing information and through various committees involving other institutions, elected representatives and the civil society. Until 1997. billing and collection was a separate department. In 1998, this department was abolished and billing and collection was merged at the depot level. Therefore, it falls under the responsibility of the depot engineer who has to achieve specific targets. Apparently, this change led to drastic improvement in collection from 60 % to 120 % (including arrears). In addition, the work culture has changed and the relationship with the consumers has improved (source: interviews). At each area depot, where bill collection takes places, various leaflets are freely available and provide information related to the overall water and sewerage situation in Chennai7, as well as specific information on the distribution area. This is part of a well-thought policy to become closer to one’s consumer that led to the setting up of an information cell headed by an Information and Facilitation Officer (IFO). This officer is in charge of external communication (public relations with newspapers, meetings with NGOs), supervising the complaint cell, developing information leaflets and the citizen’s charter. In addition, there is an area IFO. The IFO works directly under the Executive Director who is in charge of all the other public relations activities for the Board. Among them are the various committees: (i) an independent committee of consumers held once every two months and constituted of 5 members; (ii) ward committees’ meetings held at the ward level every month and which are attended by the engineer of the ward; (iii) a joint council held every three months and headed by the mayor; (iv) an industrial service committee which meets every three months and is constituted of 3 members of the Metro Water Board and 3 members of the Manali Hills industrial association. This office has also contracted Economics Perspectives for three years to conduct consumer surveys on a regular basis for a cost of Rs. 2.4 millions. At the end of the 3 years, all the consumers of the Board would have been surveyed and their level of satisfaction assessed.

14None of these efforts exist in Mumbai or Calcutta and there is no doubt that the Metrowater Board has developed adequate tools to inform its “consumers” about the service rendered. However, this does not conceal the considerable supply problem faced by the Board and therefore cannot be a substitute in terms of satisfaction level.

Costs

15The Board reports the total (operational) cost of water in 1998-99 to be Rs. 12.48 per kl8. The Board has calculated that, at a subsidised flat rate of Rs. 50/– per month for 237,943 consumers, “assuming an average size of the family to be 5, and a supply at the rate of 70-100 lpcd, every household is subsidised to the extent of about Rs.850/– per year” (source: CMWSSB 2000). The Board also underlines that “the number of employees per thousand connections which was of the order of 70 during 1986 has drastically came down to 27 during 1998-99”. Notwithstanding financial charges, revenue expenditure can be split like this: power and chemicals: 20 %, O & M: 11 %, drought expenditures: 8 %, employees: 45 % hire of water lorries: 16 %9.

Tariffs

16The Act specifies that the Board shall levy a water tax and a sewerage tax, which “in the case of water tax shall not be more than twenty per cent and in the case of sewerage tax shall not be more than ten per cent of the assessed annual value of the premises”. The actual amount is much lower. The CMWSSB (1991) gives respectively a water tax by 1.5 % and a sewerage tax by 5.5 % of assessed annual value. These are deemed to be paid whether the users are actually supplied or not, or even whether the residents are actually connected in piped network (supply and sewerage) zones or not.

17But the Board also enjoys tariffs. Metered tariffs are, broadly: for domestic use: 2.50 Rs/kl for small consumers (less than 10 kl monthly), increasing up to Rs. 25 per kl beyond 25 kl. Commercial use is billed at 25 Rs/kl up to 400 kl, and 50 Rs beyond. Users of public fountains collectively pay Rs. 400/– per fountain per month including maintenance charges, while mobile water supply to slums is charged Rs. 4/– per kl. The Board also offers mobile water supply to its consumers, at Rs. 600 for every load up to 12,000 litres, or to private tankers, at Rs.30/ kl, at its filling points.

18Conversely, non-metered tariffs are: Rs. 50/-month for an individual house or per flat; Rs. 150/– for partly commercial; Rs. 200/– month for non-intensive commercial water; Rs. 400/– month for intensive commercial water. From this, it is clear that there are cross-subsidies. Indeed, in terms of quantity of water consumed, the commercial consumers’ share is only about 16 % of the total water supplied, while they contribute about 46 % of the Board’s revenue. On the other hand, the domestic consumers contribute only 40 % of the Board’s income and continue to consume about 69 % of the total quantity of water supplied. More than that, only 40 consumers (all located in the Manali Hills area) account for 96 % of the amount collected from all commercial consumers (Rs. 7,400 lacs out of Rs. 7,708 collected in 1999- 2000, see Table below). This indicates a strong dependence of the Board on these consumers who have to be wooed every year so that they do not exit the municipal system. To the water supply rates, a 25 % surcharge is generally added for the sewerage of this water.

19Table 7.1. shows that the collected tax is, if not the greater part, a fair amount of the revenues of the Board (27 %). This amount, will be recalled, is collected whether the supply is actually given or not, whether the sewerage is actually ensured or not. This part of the revenue is thus closer to the functioning of an administrative body, rather than to that of a corporate entity. The increase in revenue is largely imputable to an administrative increase of the tax value (it explains 64 % of the total increase).

20As far as connection charges are concerned, the procedure has been simplified, as any civil engineer can approve the request for a connection. Costs for a new connection, as in many other Indian cities, are high: Rs. 2,000 for water and Rs. 3,500 for sewerage. In the case of low-income households, payments can be made in 3 installments. However, this figure is certainly too high to enable poorer households to connect to the system.

The set of institutions: coordination and new arrangements

The so-called “privatization”

21The CMWSSB has engaged in openings to private sector that are innovative in the Indian public sector. It has interestingly taken stock of its past experience and issued some general lessons that are extensively reproduced here. The CMWSSB (2000) states that under limited local availability of capability for large-scale operations, “writing of contract documents is an important area where more expertise is needed; quantification of risks and risk allocation play a major role in response/economics”. But it notices that “on construction and operation contracts, as the contractor gains basic operational experience during mandatory operation and maintenance period; he is in a better position to properly assess the risks”, and concludes that “areas requiring large manpower input with less complex operations offer greater scope for privatisation and hence are economical”. This indeed goes along with the Board’s policy to reduce its staff expenses.

22It also underlines limitations in the writing of contracts that are well-known by academics and practitioners, such as: the lack of the private sector’s legal responsibility related to accidents for environment or living beings, lack of legal provision for compensation for losses due to lesser output, or lack of a framework for compensation for losses due to improper quality and as regards obligations not met by the private operator in respect of labour-related issues.

Institutional share in responsibilities

23Let us examine the autonomy the Board enjoys. It will be noted first that it remains fully integrated in the governmental sphere since the Chairman of the Board is the Minister in-charge for water supply, and the Board includes the water supply Secretary of the government, as well as the finance Secretary. It is, however, rooted in the municipal corporation as well, since the Member-Secretary of the Chennai Metropolitan Development Authority and the Commissioner, Municipal Corporation of Chennai, are also part of it. To balance these, the Managing Director is appointed for a period “not less than three years as the Government may determine and the person appointed shall be eligible for reappointment” (source: CMWSSB Act, 1978).

24Compared to other Boards in the country, it has the possibility to take external loans. It must, however, seek the prior sanction of the government for schemes and works, if above 10 million rupees in the case of replacements, reconstruction, and above 75 lacs of rupees in the case of new works including extensions and additions to existing works. As far as these amounts are concerned the area of discretional power of the Board is therefore limited to execute externally-aided projects, but provided the government accorded prior sanction, and to adopt any maintenance work, irrespective of the cost. Even as far as the power to write-off on irrecoverable amount, the Act specifies that “the Board shall, before writing off such an amount exceeding twenty-five thousand rupees, obtain the sanction of the Government”.

25The Board, according to the Act, enjoys regulatory power concerning the manner in which taxes are to be levied, the charges for consumption of water, but with the approval of its Chairman, that is, ultimately, the state government minister in charge.

New institutional arrangements: ‘privatization’, or outsourcing of operation

26The Chennai Metro Water Board has involved the private sector by entrusting to it the operation and maintenance of a water treatment plant (since 1997), water production wells and several sewage pumping stations (this started in 1982 and almost half of the 118 stations are operated by 24 different companies). Interviews with the Board have indicated that the policy is merely conducted at the pace of retirements, from a perspective of savings on human resources. However, this has undoubtedly generated capacity-building for local Indian entrepreneurs and, as the CMWSSB (2000) shows, led to savings.

27For instance, a 56 % savings is achieved on operation and maintenance of well fields; 18.5 % on water transport through private lorries; 65 % on maintenance of sewage pumping stations; 33 % on maintenance of sewerage treatment plants. Only the maintenance of water treatment plants has shown an excess cost by 14.4 %. This is done along with the retirement of employees of the Board, which had 8,300 employees in 1994, and 5,400 in 2001 (with an objective of 4,000 in the mid-term (source: interviews). In practice, the Board continues to paying the expenses, and the private partners’ expenditures are limited to a monthly Rs.2,000/– in pumping stations, for instance (source: CMWSSB 2000a).

28Maintenance, however, must be comprehensively improved. For instance, the Economic Perspective survey points out that 77 % of public taps and pumps are collectively maintained by their users, 13 % by self-users, and only 10 % by the Board.

29The CMWSSB has also engaged in a joint venture licensing agreement with a private company for producing and marketing packaged drinking water. The firm constructed the plant and production started in December 1999, with responsibility for quality remaining with the CMWSSB.

Delegation of powers and internal management

30Compared to the Corporation structures in Bombay or Calcutta, the Board has, in Chennai, initiated or achieved noticeable steps in improving the overall management of resources. Of considerable importance in the general context of Boards in India, the CMWSSB has for two years promoted preventive maintenance, which CMWSSB (2000) declares “has resulted in substantial cost savings as also increased efficiency and productivity”. In practice, the Board has changed many procedures, and formalized them through memoranda of understanding, signed with the area engineers. The aim of the principal aspect of the MoUs is to fix targets of 80 % of collection over the collectables (CMWSSB 2001). The “financial delegation cf power”, which is, in actual terms, not a proper delegation in the managerial sense, but rather an administrative delineation of responsibilities that still includes seeking administrative clearances by the hierarchy, was anyway modified in 1998. Now, executive engineers enjoy a Rs. 2 lacs delegation, and even assistant engineers Rs.5,000/ More important, the number of hierarchy levels to actually clear the so-called “delegated decision” has been reduced from 3 to 2 (the two steps of estimates and technical sanction have been merged: an officer now only has to directly seek “administrative sanction”). The internal delegation of power still remains an administrative and a board-like one, but it is among the highest in a Board structure in India.

31Internal management also goes with improvement of reporting. The intermediate report of the Twinning Arrangement (Générale des eaux) concludes that “from a compilation of the figures reported by the depots and area offices during the diagnostic review, it appears that 78 % of properties have access to distribution within the CMWSSB’s area. This compares with the 96 % reported in the annual report”.

32A study on institutional options is also being carried out by KPMG (through some DFID findings). The rationale of this study is to divide the Board into Strategic Business Units (SBUs). There is a debate within the Board regarding this report, especially concerning billing and collection. KPMG proposes billing and collection to be a separate SBU, while the Board policy (see above) has, on the contrary, integrated billing and collection with operation at the ward level.

Contracting the water lorries

33Supplying water through lorries, though more expensive, has still to be done due to scarcity of the resource that precludes using the piped system in a normal way. It is partly done through contracting annually with a 500 owners10 of 750 lorries, grouped in a tankers association, for a total of 5,500 trips a day. Interviews show that, compared to the total production and transportation cost of, respectively, Rs. 6 per kl and Rs. 2.5 per kl for surface and ground water (and a total of 14 Rs when including distribution and sewerage), the cost of water lorries from is Rs. 8 for production plus transportation to the filling point, and then an additional Rs. 19.5 per kl from the filling points to the final users. This is monitored under a cadre contract, where the trip is paid Rs. 176 for a maximum return journey of 16 km per trip11. The lorry owners’ association takes part in the vigilance committee held every Saturday and trains the drivers in order to avoid any problems with consumers.

34Information on the non-members of the association has been collected through an interview with one of the earliest independent and leading lorry owners. They purchase their water from farmers, around 30 km outside the city (despite the Ground Water Regulation Act), and pay 50 Rs per load to the farmers, to then sell it for between Rs. 480 or 500 for large customers (foreign consulates, universities, hospitals, hotels), and up to 550 for private residential consumers. There are around 700 to 800 lorries in the city and 150 companies. Out of these 150, 10 are large companies and the remaining ones own only one or two trucks. Apparently, the increase in private water tankers has led to increased competition among existing companies but at the same time, due to considerable water scarcity, there is until now place for all these companies.

The poor: water supply, sanitation and health

35Water in slums is mostly provided with standposts and public fountains or through water tankers contracted by the Board (according to Economic Perspectives, the distribution is 56 % taps or pumps and 44 % tanks). Wells are still used by low-income households. Until 1997, there were water charges to poor households, but now water is provided free. Connection charges for low-income groups can be paid in 3 installments but, as mentioned earlier, this remains too substantial a cost to enable slum dwellers to connect. Regarding sanitation, there is a policy to provide sewer connections in slums. So far, 30 000 connections have been provided through some MLA funds and with the help of the Corporation, as it is a government supported project.

36In the slums developed by TNSCB, the Slum Clearance Board itself developed the infrastructure. In one visited settlement, the quality of the infrastructure was not satisfactory: standposts were provided without taps, the reason being that water is provided on an intermittent basis. Therefore, in the future, if water is provided on a 24 hours basis, these stand posts will be contributing to the wastage of water. As far as economic affordability is concerned, rehabilitated families have to pay Rs. 150 rent per month for 20 years to be granted the final ownership of the flats, for a population the Board estimates as having a Rs.1,000 to 1,400 monthly family incomes. The Slum Board estimates that 70 % of these rehabilitated people have a connection to sewerage and have an average water supply of 25-30 lpcd.

37Another example illustrates the lack of sustainable solutions found by these institutions to provide water to poor settlements. One specific scheme developed by the Metro Water Board is to supply water to a fishermen’s settlement where a piped water supply is not within reach. This settlement, called Nochikuppam, is at the end of Marina Beach and the water available is brackish and is unfit for drinking purposes. To supply water to this community, the Metro Water Board engaged a private contractor to build, operate and maintain a desalination plant. The cost of the operation and maintenance of the plant is high, as is the total cost of production (around Rs. 35 per kl) However, the cost provided is not necessarily reliable as it was unclear as to whether it included the cost of electricity. In addition, the process that was followed to give this contract was not really clear. It is, however, a scheme that the Metro Water Board considers one of its best actions in favour of low-income households, despite the cost issue and the small number of people that benefit from this scheme. This scheme is ample proof that this kind of project is not sustainable to provide water to low-income households. On the contrary, the very high cost of production (Rs. 35 per kl) is a deterrent for any other projects of this type.

Beyond administration of water: Chennai as an innovative system towards management of water

38The general picture established for Chennai (scarcity, financial resources available from taxes and few big customers paying cross-subsidies, recourse to private partnerships) allows the city to now be characterized within the Indian picture. Let us first specify here that ‘technical centralization’ will refer to the type of technical model: centralized through a piped system, and ‘technical decentralization’, will refer schemes based on to local groundwater, or local surface water. Technical centralization ultimately refers to the internal technical model adopted by the body in charge. Conversely, as far as internal management is considered, the word delegation will be preferred, including for the internal delegation of an activity to a private contractor. In management, decentralization will then refer to the sharing of administrative responsibilities between the state and the municipality.

The general picture: different kinds of administrations and administrative contexts

39The study of water arrangements in urban India (refer to Ruet and Zerah 2001), shows that, independently from the physical situation, the nature of the administration in charge can vary from a mono-technical (centralized) agency with principally a mono-output oriented objective (water supply), strongly embedded in the state/municipal sphere, to a rather autonomous agency, evenly focusing on different issues and aiming at a managerial delegation, and making attempting for decentralized techniques. In other words, this ranges from a Calcutta-like situation, surprisingly both politically centralized and unclear in terms of administrative responsibility, to a Chennai-like solution, where institutional and organizational innovations are worth being exemplified.

40In Chennai, like in all major Indian cities, the tariff structure does not reflect the cost to produce water (neither average cost, nr marginal cost) and is submitted to political decisions for its revision. Most of the time, a change in tariff means a marginal increase in order to reflect the increase of the electricity cost. In no case is it linked to a strong political decision to revise the tariff regime or to link water charges to cost of production. In that context, no policy can be sustainable on the long run, and the subsidies to the poor are (i) not credible, and even (ii) adverse to the poor. Indeed, since the subsidies suppose the connection, they result not only in not benefiting the poor directly, but also in precluding them from being connected to the system. Any sound structural reform should hence target the relatively well-off sections of the users first. Target not in the usual sense of deriving cross-subsidies from them, but in developing a sustainable cost-integrated approach, namely an approach based on recovering the costs, but also on lowering them through re-use of water. This would be all the more feasible in Chennai, given the scarcity of water and the institutional arrangements already existing, and could serve as an example for the other cities. Then the issue of supplying the poor is less stringent (in the present situation, they anyway consume a lesser part of the water), and can be solved through direct subsidies on actual tariffs.

41For those familiar with the Indian administration, either under its ‘Department’ or ‘Board’ form, Chennai can well exemplify the levels managerial decentralization can reach within the administrative frame. There is no doubt that, relying on tax property collection and government support (or government-routed support) for its schemes, the CMWSSB is still mostly structured around procedures (technical and administrative) and not yet around services. As a matter of fact, providing full sanitation although a tax is collected is not looked up as compulsory. Similarly, dwelling on the water scarcity situation, the Board has de facto given up on satisfying and supplying a part of the population, or rather a segment of the consumption. But, whatever the pragmatic reasons of impossibility to face the demand through a centralized technical system, the Board has had the courage to promote some more decentralized resources. Internally, this did mean that the main office promoted a less centralized model over their technical and engineering officers. But the Board has also gone beyond that, through promoting an internal delegation of power, as well as through promoting a better service interface as far as the monitoring of complaints is concerned. What should be remembered is that, whatever the contextual situation in terms of resources and perhaps human resources at the upper echelons of the Board, and despite the fact that outsourcing has been mostly decided on a human resource policy basis, the institutional aspects matter. It can be argued that, compared to the Corporation form, the Board form, with a more clear-cut budget and allocation from resources to expenditures, allows of a better isolation and identification of the tasks, rights and duties of the organization. Similarly, the integration of the supply activity and planning and maintenance, brings some clarity in the establishment and management of control rights over the assets. And, last but not least, a sound definition of enforceable and reasonably sustainable (at least in the short run) tariffs ends up bringing this soundness in decision-making. This better definition and delineation of the very perimeter of the institution thus allows one to focus on the margins of this perimeter and to consider what the most suitable policies in terms of outsourcing are. It should be noted that, in the distribution aspects, outsourcing concerns the operation side, and not the supply activity (including commercial, etc.). Thus, the proper integration between control rights and financial rights remains, for a better incentive in to the decision-making. In other words, compared to Calcutta, the clear and joint allocation of control and cash flow rights contributes to the definition of sound property rights. Within that framework, the marginal optimization of solutions is envisaged, but keeping this integration intact.

Conclusion

42To conclude, as far as water supply is concerned, based on pragmatic experiences, the solution to promote is certainly one of an autonomous Board-like structure keeping both the cash-flow and control rights over the planning, building and supplying of the water. In other words, keeping a clear allocation of funds for the management of the activity, and then orienting the internal organization towards an integrated approach of this activity. To be sustainable, this supposes a proper tariff structure, which is detailed in the next subsection. From this ‘ideal’ allocation of control, joint strategies of outsourcing of the operation with local entrepreneurs’ capacity building can be thought of. Similarly, local strategies of integrated management of resources can be better achieved, since the Board would at the same time be the promoter or facilitator and the direct benefiter of this. The example of Chennai is worth being advertized, despite the fact that such measures occurred in a context of lowering the human resource and of the mere inability to ensure the full demand. But the second lesson is that, from this ‘ideal solution’, a pragmatic approach should also integrate the present situation. If it is possible to imagine that a Board could be constituted – even with some financial transfers to balance the loss for the corporation – from the hydraulic department in Bombay, it does not seem relevant to do so in Calcutta. Indeed, in the latter case, the state-led organizations perform better. Then, a concentration of control and cash flow rights still has to be reached, but it is difficult to imagine that as a first top this can be at the city level. This, all the more that in the case of Calcutta, the tariff issue is striking as the next subsection shows. Beyond organisational efficiency, the financial support of the State Government is still needed in Calcutta.