India - Maharashtra bagasse energy efficiency project

Mah sarhtra Bagasse Energ Efficiency Poject Report No. 120/91  JOINT UNDPIWORLD BANK ENERGY SECTOR. MANAGEMENT ASSISTANCE PROGRAMME (ESMAP) PURPOSE The Joint UNDP/World Bank Energy Sector Management Assistance Programme (ESMAP) was launched in 1983 to complement the Energy Assessment Progamme which had been established three years earlier. An international Comnmission was convened in 1990 to address the creation of ESMAP's role in the Nineties. It concluded that the Programme had a crucial part to play over the next decade in assisting the developing countries to better manage their energy sectors given that the supply of energy at reasonable prices is a critical determinant of the pace and magnitude of the growth process. The Commission's recommendations received broad endorsement at the November 1990 ESMAP Annual Meeting. Today, ESMAP is carrying out energy assessments, preinvestment and prefeasibility activities and is providing institutional and policy advice. The program aims to strengthen the impact of bilateral and multilateral resources and private sector investment through providing technical assistance to the energy sector of developing countries. The findings and recommendations emerging from ESMAP activities provide governments, donors, and potential investors with the information needed to identify economically and environmentally sound energy projects and to accelerate their preparation and implementation. ESMAP's operational activities are managed by two Divisions within the Industry and Energy Department at the World Bank and an ESMAP Secretariat. * The Programme's activities are governed by the ESMAP Consultative Group which consists of its co- sponsors, the UNDP and the World Bank, the governments which provide financial support and representatives of the recipients of its assistance. The Chairman of the Group is the World Bank's Vice President, Sector Policy and Research. He is assisted by a Secretariat headed by the Group's Executive Secretary who is also responsible for relations with the donors and securing funding for the Programmne's activities. The Secretariat also gives support and advice to a Technical Advisory Group of independent energy experts which meets periodically to review and scrutinize the Programme's strategic agenda, its work program and other issues related to ESMAP's functioning. * The ESMAP Strategy and Programs Division is responsible for advising on which countries should receive ESMAP assistance, preparing relevant ESMAP programs of technical assistance to these countries and supports the Secretariat on funding issues. It also carries out broadly based studies such as energy assessments. * The ESMAP Operations Division is responsible for the detailed design and implementation of tasks consisting mainly of sub-sectoral strategy formulation, preinvestment work, institutional studies, technical assistance and training within the framework of overall ESM AP country assistance programs. FUNDING The ESMAP represents a cooperative international effort supported by the World Bank, the United Nations Development Programme and other United Nations agencies, the European Community, Organization of American States (OAS), Latin American Energy Organization (OLADE), and a number of countries including Australia, Belgium, Canada, Denmark, Germany, Finland, France, Iceland, Ireland, Italy, Japan, the Netherlands, New Zealand, Norway, Portugal, Sweden, Switzerland, the United Kingdom and the United States. FURTHER INFORMATION For further information or copies of completed ESMAP reports, contact: Office of the Director OR The Executive Secretary Industry and Energy Department ESMAP Consultative Group The World Bank The World Bank 1818 H Street N.W. 1818 H Street, N.W. Washington, D.C. 20433 Washington, D.C. 20433 U.S.A. U.S.A.  CURRENCY EOUIVALENTS Currency Unit Rupees (Rs) Rs. 1.00 Paise 100 US$1.00 Rs. 17.00 Rs. 1,000,000 - US$58,824 MEASURES AND EOUIVALENTS 1 Kilometer (km) - 1,000 meters (m) - 0.6214 miles (mi) L Meter (m) 39.37 inches (in) 1 Cubic Meter (m3) - 1.31 cubic yard (cu yd) 35.35 cu.ft. 1 Thousand Cubic Meter (MCM) - 1,000 cubic meters 1 Barrel (Bbl) - 0.159 cubic meter 1 Normal Cubic Meter - 37.32 Standard Cubic Feet (SCF) of Natural Gas (Nm3) 1 Ton (t) - 1,000 kilograms (kg) 2,200 lbs 1 Metric Ton of Oil (390 API) - 7.60 barrels 1 Kilocalorie (kcal) 3.97 British Thermal Units (BTU) 1 Kilovolt (kV) - 1,000 volts (V) 1 Kilovolt ampere (kVA) 1,000 volt-amperes (VA) 1 Megawatt (MW) - 1,000 kilowatts (kW) -1 million watts 1 Kilowatt-hour (kWh) - 1,000 watt-hours 1 Megawatt-hour (MWh) - 1,000 kilowatt-hours 1 Gigawatt-hour (GWh) - 1,000,000 kilowatt-hours 1 Ton of Oil Equivalent (toe) 10 million kilocalories  ABBREVIATIONS AND ACRONYMS BICP - Bureau of Industrial Costs and Prices BSES - Bombay Suburban Electricity Supply CEA - Central Electricity Authority DCC - District Central Cooperative Bank DNES - Department of Non-conventional Energy Sources DSI - Deccan Sugar Institute EIRR - Economic Internal Rate of Return ESMAP - Energy Sector Management Assistance Program EUE Expected Unserved Energy FIRR - Finaacial Internal Rate of Return GOI Government of India COM - Government of Maharashtra IBRD - International Bank for Reconstruction and Development ICICI - Industrial Credit and Invetment Corporation of India Ltd. IDBI - Industrial Development Bank of India IFCI - Industrial Finance Corporation of India ISMA - Indian Sugar Mills Association LRAIC - Long Run Average Incremental Cost LRMC - Long Run Marginal Cost MECC - Marginal Expected Curtailment Cost MEDA - Maharashtra Energy Development Agency MSCB - Maharashtra State Cooperative Bank MSEB - Maharashcra State Electricity Board NCDC - National Cooperatives Development Corporation NFCSF - National Federation of Cooperative Sugar Factories Ltd. NPC - National Productivity Council NTPC - National Thermal Power Corporation PACS - Primary Agricultural Credit Societies PFC - Power Finance Corporation PURPA - Public Utilities Regulatory Policies Act (U.S.) RBI - Reserve Bank of India REC - Rural Electrification Corporation ROI - Return of Investment ROR - Rate of Return SCF - Standard Conversion Factor SDF - Sugar Development Fund SEB - State Electricity Board SOM - State of Maharashtra SRMC - Short Run Marginal Cost TCD - Tonnes Cane per Day TEC - Tata Electric Companies TNPL - Tamil Nadu Newsprint and Papers Ltd. UNDP - United Nations Development Program Fiscal Year April 1 - March 31  TABLE OF CONTENTS EXECUTIVE SUMMARY .. . 1. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . 1 Background . . . . . . . . . . . . . . . . . . . . . . . . Major Constraints . . . . . . . . . . . . . . . . . . . 2 Objectives . . . . . . . . . . . . . . . . . . . . . . . . 2 II. SECTORAL CONTEXT: THE POWER SECTOR . . . . . . . . . . . . . 4 Overview . . . . . . . . . . . . . . . . . . . . . . . . . 4 India's Commercial Energy Resources . . . . . . . . . . . 4 Organization of the Power Sector . . . . . . . . . . . . . 5 Tariffs ...... . . .. . . . .. . . .. . . . .. . 6 Electricity Supply and Demand in the Western Region . . . 6 Power Situation in Maharashtra . . . . . . . . . . . . . . 7 Private Power Generation .... . . . . . . . . . . . . . 9 III. SECTORAL CONTEXT: THE SUGAR SECTOR . . . . . . . . . . . . 11 Overview ...... . 11 The Demand for Sugarcane Products . . . . . . . . . . . . 11 The Supply of Sugarcane Products . . . . . . . . . . . . 12 The Conversion Industry .... . . . . . . . . . . . . . 13 Supply/Demand Prospects .... . . . . . . . . . . . . . 15 Institutional Aspects . . . . . . . . . . . . . . . . . . 15 Financial Aspects . . . I . . . . . . . . . . . . . . . . 16 Pricing and Taxation Policies .... . . . . . . . . . 16 The Financial Framework of the Sugar Sector in Maharashtra ..... . . . . . . . . . . . . . . 18 IV. BAGASSE UTILIZATION AND VALUATION IN MAHARASHTRA . . . . . . 26 Overview ...... . . . . .. . . . . .. . . . . .. . 26 Bagasse Production . . . . . . . . . . . . . . . . . . . . 26 Present Production . . . . . . . . . . . . . . . . . . 26 Estimating Future Availability . . . . . . . . . . . . 27 Bagasse Utilization . . . . . . . . . . . . . . . . . . . 30 Present Practices ..... . . . . . . . . . . . . . . 30 Potential Improvements . . . . . . . . . . . . . . . . 31 Optimal Bagasse Utilization . . . . . . . . . . . . . . 33 Valuation of Bagasse as a Fuel . . . . . . . . . . . . . . 33 Principles . . . . . . . . . . . . . . . . . . . . . . 34 Economic Efficiency . . . . . . . . . . . . . . . . . . 34 Marginal vs. Average Cost Pricing .... . . . . . . . 34 Cost-Plus Pricing ..... . . . . . . . . . . . . . . 35 Estimation of MSEB Avoided Energy Costs . . . . . . . . 35 Estimation of MSEB Avoided Capacity Costs . . . . . . . 37 Toward a Workable Power Purchase Tariff ... . . . . . 39 Valuation of Bagasse as a Feedstock . . . . . . . . . . . 41  V. THE PROPOSED PROJECT .............. ..... . 45 Project Objectives . . . . . . . . . . . . . . . . . . . . 45 Project Description . . . . . . . . . . . . . . . . . . . 45 Finarncing Plan . . . . . . . . . . . . . . . . . . . . . . 47 Sub-projec.ts . . . . . . . . . . . . . . . . . . . . . 48 Project Implementation . . . . . . . . . . . . . . . . . . 49 Project Operation . . . . . . . . . . . . . . . . . . . . 5s MSEB Bagasse Cogenerator Contract . . . . . . . . . . . 50 Grid Interconnection Standards . . . . . . . . . . . . 51 Project Justification . . . . . . . . . . . . . . . . . . 52 Financial Analysis . . . . . . . . . . . . . . . . . . 53 Economic Analysis . . . . . . . . . . . . . . . . . . . 59 Project Risks . . . . . . . . . . . . . . . . . . . . . 63 Environmental Impacts . . . . . . . . . . . . . . . . 65 Conclusions . . . . . . . . . . . . . . . . . . . . . . 66 ANNEXES 1 Power Sector . . . . . . . . . . . . . . . . . . . . . . . 67 2 Sugar Sector . . . . . . . . . . . . . . . . . . . . . . . 81 3 Maharashtra Sugar Mill Survey Results . . . . . . . . . . 89 4 Efficiency Standards and Standard Specifications for New or Expanding Sugar Mills . . . . . . . . . . . . . . 121 5 Case Studies of Bagasse Cogeneration Schemes . . . . . . . 130 6 Design of a Power Purchase Tariff in Maharashtra State.. ............ 195 7 Power Loss Reduction Credit . . . . . . . . . . . . . . . 228 8 Contract Model for MSEB Power Purchase . . . . . . . . . . 236 9 Sugar Mill to Utility Interconnection Standards . . . . . 260 Appendix 1 . . . . . . . . . . . . . . . . . . . . . . . 270 10 Off-Season Fuel Alternatives . . . . . . . . . . . . . . . 298 TABLES 2.1 Total Installed Capacity for Regional Power Systems . . . 7 3.1 Production and Use of Sugarcane, 1987/88 . . .13 3.2 Performance of Sugar Factories in the Main States, 1987/88 ..14 3.3 Key Financial Parameters . . . . . . . . . . . . . . . . 17 3.4 Financing Plan for New Mill . . . . . . . . . . . . . . . 20 3.5 Lending Terms for New Mills . . . . . . . . . . . . . . . 22 3.6 Lending Terms for Expansion and Modernization of Mills . . . . . . . . . . . . . . . . . . . . . . . . 22 4.1 Crushing Season Duration Analysis . . . . . . . . . . . . 29 4.2 Maharashtra Bagasse Utilization, 1987-88 . . .30 4.3 MSEB Avoided Energy Costs ..37 4.4 MSEB Avoided Capacity Costs . . . . . . . . . . . . . . . 39 4.5 Two Part, Three Period Tariff for Bagasse Cogenerated Power . . . . . . . . . . . . . . . . . . . 40 4.6 Two Part, Two Period Tariff for Bagasse Cogenerated Power. . . . . . . . . . . . . . . 40  4.7 Single Part Tariff for Bagasse Cogenerated Power . . . . . 40 4.8 Recommended MSEB Commercial Power Purchase Tariff for Bagasse Cogenerated Power . . . . . . . . . . . . . 41 4.9 Estimated Financial and Economic Bagasse Values . . . . . 43 5.1 Bagasse Cogeneration Investment Candidate Sugar Mills . . 46 5.2 Scope of the Project . . . . . . . . . . . . . . . . . . . 47 5.3 Project Financing . . . . . . . . . . . . . . . . . . . . 47 5.4 Sub-project Financing . . . . . . . . . . . . . . . . . . 48 5.5 Results of Case Study Financial Analyses . . . . . . . . . 56 5.6 Results of Case Study Economic Analyses . . . . . . . . 61 FIGURES 4.1 Maharashtra Grinding Season and Cane Production, 1978-88 . . . . . . . . . . . . . . . . . . . . . . . . 28 5.1 Financial Rate of Return vs. Mill Size . . . . . . . . . 55 5.2 Sensitivity of FIRR . . . . . . . . . . . . . . . . . . . 58 5.3 Economic Rate of Return vs. Mill Size . . . . . . . . . . 60 5.4 Sensitivity of EIRR . . . . . . . . . . . . . . . . . . . 62 MAPS IBRD 22204: Electrification Scheme IBRD 22205: Major Sugar Mills  EXECUTIVE SUMMARY Background 1. Bagasse-based cogeneration for power export to the grid has long been of interest Zo the Government of India arid the private sector. In the recently completed ESMAP review of the country's nonconventional energy program being managed by the Department of Nonconventional Energy Sources (DNES), it is one of the major options identified as economically promising and worthy of more detailed study. Consequently, the Government, through DNES, requested further assistance from the World Bank to analyze the subject in greater detail and, if warranted by the findings, develop a project proposal suitable for financing consideration by international and/ot domestic lending agencies. 2. The technology for bagasse cogeneration for export power is well-known and commercially practiced in several countries, notably Australia, Cuba, Dominican Republic, Mauritius and the United States (Hawaii). The gestation period for such projects is less than that for conventional thermal or hydro power plants. In India, given adequate market opportunities for power sales, bagasse cogeneration projects appear as attractive private investment options, capable of adding to the utility's supply capability while reducing reliance on scarce public funds. For many sugar mills, bagasse cogeneration projects could result in added income from electricity or surplus bagasse sales and lower production costs in downstream mill activities. Major Constraints 3. Despite these potentially significant benefits to the economy and the proven success in other countries, a bagasse cogeneration project selling surplus power to the grid is still to be realized in India. There are several reasons for this. Firstly, its detailed mechanics and viability are still relatively unknown to both millers and electricity boards. Secondly, mill owners mistakenly assume tL.at cogeneration for export power cannot coexist with bagasse-based pulp and paper production. Finally, and perhaps the biggest impediment of all, is the lack of clear GOI and State government policies regarding the sale of privately generated power to the grid. In turn, this reflects the State Electricity Boards' (SEBs) 'inderstandable uncertainty about the reliability and value of cogenerated power and the impact of such schemes on their overall operations and system planning. There is lack of a concrete basis or framework for negotiations which would lead to a commitment by SEBs to purchase power from sugar mills at mutually acceptable rates.  - ii - Objectives 4. The main purpose of the present activity is to provide such a framework by addressing the above issues in adequate detail in a project- oriented study. Maharashtra State was chosen for the proposed project based on various considerations. It has the largest number of sugar mills of over 2,000 TCD (tonnes cane crushed per day) capacity. Currently, many of the smaller mills have applied for or have been granted permits to expand capacity. The State appears to lead all other sugar producing states in mill efficiency and profitability. On the utility side, Maharashtra is one of the more advanced and enlightened states in terms of examining grid power exchange between the SEB and private producers. Although its e[ectricity supply/demand situation is relatively bright compared to other states, peak demand restrictions are still in effect and energy demand is forecasted to grow at nearly 7% annum through 2000. The cane crushing season in Maharashtra is between November and April, coinciding with peak power demand from the grid and a period of lowered hydro generation. 5. Despite the focus on a single state, the methodology, analysis and implementation plan developed in the proposed project would clearly be applicable as well to other sugar-producing states, with minor modifications to account for differing utility system configurations. Private Power Generation 6. Maharashtra is home to two of India's four private utilities. Tata Electric Companies (TEC), a private generation and bulk supply utility, is based in Bombay. Bombay Suburban Electricity 3upply (BSES), a private distribution utility, provides supply to final consumers in the Bombay metro area. India wide, private utilities at present make only a marginal contribution to public electricity supply, though private captive generation is extensive (equivalent to about 15% of public supply). 7. In recent years, notwithstanding impressive achievements in commissioning generating plants, it has been increasingly recognized that the public sector alone will not be able to satisfy the ever increasing investments nekeds for power generation, transmission and distribution. This is both due to the size of the capital requirements (the preliminary estimate for the VIII Plan 1990-95 is of the order of US$60 billion), the limited capability of public sector utilities to generate surpluses available for investment, and the constraints on the availability of external financing. As a result, the possibility of mobilizing private investments for the sector is under review both at the national and the State levels. To that end, it will be necessary to remove a number of impediments which are essentially financial, technical and legal in nature. Working groups set up both by the Central Government and the  - iii - State Government of Maharashtra have recognized the potential benefits of private sector involvement in power supplies and have recommended some measures to accelerate the process. The Sugar Sector 8. Sugar plays an important role in India's agriculture, as it accounts for about 15% of India's agricultural output in value from only 2% of cultivated areas. There are an estimated 35 million planters, the majority of them small, and sugar mills employ an estimated 300,000 workers, of which about half are permanent and the rest seasonal. Being cash cropping, and largely targeted to modern agro industries, the sugar industry is often perceived as an instrument for rural development. Sugar mills not only create direct employment, but also invest in infrastructure (particularly roads) and promote rural development, for instance by sponsoring health facilities and learning institutions. 9. India has recently become the largest producer of sugar in the world. Growth of output has been uneven, largely because of changing Government policies, and climatic conditions. Over the past five years, however, Government policies have been increasingly supportive, so that it is anticipated that the historical pattern of growth and recession periods will be arrested. 10. Refined (centrifugal) sugar is obviously the most important product. Consumption of sugar has increased from 4.5 million tonnes in 1977/78 to 9.4 million tonnes in 1987/88, equivalent to 7.7% growth per annum. About 30-35% of the sugar is consumed in bulk form, and 60-65% directly by households. Overall per capital consumption at 11.8 kg at present is low by international standards (where it is of the order of 30-50 kg) even when taking into account the quantity of sugar consumed in its non-refined traditional forms. Bagasse Utilization 11. Important by-products of sugar manufacturing are molasses and bagasse. Bagasse is the fibrous residue of sugarcane, which accounts for about 30% of the cane in weight. In Maharashtra, bagasse today is utilized in two principal end-uses: (a) As a fuel in existing low pressure (-21 kg/cm2) sugar mill cogeneration systems designed to meet internal steam and electricity loads; and (b) As a pulping feedstock in six small scale (20-30 TPD) paper factories attached to sugar mills.  - iv - 12. Fuel. Internal steam and electricity needs consume some 94% of total bagasse production. Almost all the mills employ low pressure (21 kg/cm2) and temperature (320-340'C) water-tube boilers and turbo- alterators. These are sufficient to meet the energy needs of the plant (process steam, mechanical shaft power and about 15-20 kWh electricity per tonne of cane processed) but do not allow for any exportable energy surplus. Thus in their standard design configuration, the mills are basically bagasse incinerators designed to consume all available cane crush waste by the end of the crushing season. 13. Pulp. Nine Maharashtra mills have established small attached paper mill operations which can utilize depithed bagasse as a pulp for low quality writing papers. The profitability of these plants is marginal and three have closed down, yet the investments represent sunk costs to the owners who would like to assure a continuing bagasse supply. Efficiency Improvements 14. With appropriate investments, an existing sugar mill should normally be able to export about 40 kWh of electricity per tonne of cane processed. In a maximum elecitricity configuration, the cogeneration of electrical energy incidental to the process steam flow is augmented by electrical energy obtained from the steam generated by burning all the excess bagasse. 15. Optionally, the production of surplus bagasse may be optimized while still permitting 20 kWh/tonne of cane to be generated beyond the mill's internal needs. In this maximum bagasse configuration, cogeneration of elec.rical energy is incidental only to the process steam flow. All the es.ess bagasse produced is used in secondary processiig operations or sold on the market. Optimal Bagasse Utilization 16. Investment in cogeneration for export power would not result in a conflict with alternative bagasse use such as in pulp and paper. This is because almost 70% of the electric power obtained from a modern sugar mill cogeneration system is a by-product of efficient process steam raising. Only 30% of the power is generated from the combustion of the excess bagasse created through bagasse drying and efficient steam cycles. 17. Decisions as to which efficiency upgrade, bagasse or electricity maximizing, is the most appropriate investment depend on two factors: (a) The risk averseness of the investors: a sugar mill could lock in to a fairly stable contract electricity sales price, whereas the future demand for bagasse in other uses is less well known; and (b) The relative financial and economic values of the excess bagasse used as fuel and as a paper feedstock.  - v - Valuation of Bagasse Cogenerated Electricity 18. All sugar mills in Maharashtra are connected to the state grid, from which they draw start-up power. The value of electricity on MSEB's grid supply essentially determines the profitability of bagasse use as a fuel in sugar mills contemplating electricity export. But, a complication is that the price of electricity sales to sugar mills as occasional demanders of power may vary considerably from the price for MSEB purchases of electricity from sugar mills. This is because power tariffs reflect compromises between various and s-metimes conflicting economic, financial, social and administrative objectives. Economic Efficiency 19. Economic efficiency is promoted by designing a rate schedule which encourages the development of efficient cogeneration projects-- those which can supply power at a lower resour-te cost than the utility. Economic and legal interpretations of the efficiency criterion have consistently held that buy-back rates should be based on avoided cost principles, i.e., the cost which the utility would have incurred had it been required to generate the power through its own (or alternative) means. Marginal vs. Average Cost Pricing 20. The average (accounting) cost approact is concerned with the recovery of sunk costs, and implies that future economic resources will be as cheap or as expensive as in the past. There is ample evidence in Maharashtra that the cost of fuel resources is becomitg more expensive over time, so the state utility's historic costs of generation would not be the best guide to future costs. Examples are the steady decline in domesLic corl calorific value, and the plan to have coal for the Parli VI-VII extensions come from mines in Korba and the Ib Valley some 1,350 km away. 21. The avoided cost concept is the mirror image of marginal cost pricing. Marginal cost prices output on the cost of the "last" (costliest) unit produced. It is inherently a "forward looking" approach, consistent with least-cost generation expansion planning. Pricing cogenerated power at a utility's average cosit of generation is inefficient. Assuming that MSEB's marginal costs exceed average costs, the result is that a less than optimal amount of cogenerated power will be purchased by the utility. Resources are wasted because the utility misses the opportunity to buy a block of pover which could be supplied at a cost lower than the utility's. 22. Resource savings will be maximized only if the power is purchased from the cogenerators with the lowest costs, i.e., those making up the lower part of the cogenerated power supply curve. Setting price  - vi - equal to marginal (avoided) cost effectively excludes the higher cost cogenerators because they will not be willing to sell at a price lower than their cost. Cost-Plus Pricing 23. An alternative pricing formula that has been proposed for the purchase of cogenerated power in India is the cost-plus method. Under the strictures of the Indian Electricity Supply Act of 1948, this formula is interpreted as the individual cogenerator's cost of power production plus a mark-up to achieve a return on investment (ROI) of the Reserve Bank of India rate plus 2%. However, such a pricing arrangement is inefficient and inconsistent with least-cost planning principles. Applying this cost plus formula to a cogeneration unit would likely provide inadequate incentives for cogeneration project adoption by private proposers. The Reserve Bank of India Rate is essentially the risk-free ROI for large investors; a 2% margin over this rate is unlikely to compensate for the relative riskiness of a bagasse cogeneration investment. Equally undesirable is the reverse outcome in which high cost, uneconomic schemes become validated under such pricing arrangements. 24. From an administrative viewpoint, the transaction costs of developing individual contracts with numerous cogeneration proposers would be high. Individual differentiation would be necessary under a cost-plus framework since the cost of production varies from site to site depending on the technology and scale of operations. Finally, electricity is not the sole output of a cogenerator, but rather a joint product of a process which simultaneously produces steam (and excess bagasse). Standard utility measures of cost of generation are thus not applicable to cogenerators. Estimation of MSEB Avoided Costs 25. Energy payments to cogenerators should be based on the avoided energy costs of the utility. The appropriate cost in any time period is the system lambda, familiar to utility dispatchers as the measure of rank order merit. System lambda is a short run marginal cost (SRMC) composed of the marginal unit in service fuel costs (unit cost of fuel times the marginal heat rate) plus variable operations and maintenance (O&M) costs. These costs should be adjusted for loss factors including a correction for marginal generation unit internal (parasitic) consumption, and a credit (charge) for cogenerator line loss reduction (increase), if appropriate. 26. Capacity credits are derived from: (a) the costs imposed on the electric system due to marginal increases in peak aemand, and (b) the cogenerators' effective luad carrying capacity. Bagasse cogenerated power is "non-firm" in that it is only available approximately six months (November-May) during the year. Standard utility analysis would define a minimum availability factor (say, 75%) that a plant would need to attain  - vii - before it could be classified as "firm" power. The traditional view would thus assign full capacity credit to the firm plant, and zero capacity credit to be bagasse cogeneration plant with an estimated 47% availability factor. 27. The methodology employed in this report recognizes that absolute distinctions between tirm and non-firm are artificial in a world of uncertainty. Through a combination of plant failures and demand uncertainty there is a finite probability of demand being curtailed at any time of the day and year. With capacity subject to random failure, demand characteristics alone do not indicate when the system is at stress. Therefore, all time periods should be responsible for some portion of the capacity costs and corresponding credits to cogenerators. 28. A power purchase tariff following the above principles is given in Table 1. Table 1: RECOMMENDED MSEB COMMERCIAL POWER PURCHASE TARIFF FOR BAGASSE COGENERATED POWER Duration Tariff Period b/ (hours/day) (Ps./kWh) Energy and Capacity Off-Peak 16 49 Peak 8 129 a/ Peak and off-peak periods defined as per MSEB Low Voltage Tariff Review Inception Report. Source: Annex 6. This is 76 Paise/kWh on a time-averaged basis, which compares with the FY1988/89 MSEB average consumer tariff of 78 Paise/kWh. Valuation of Begesse as a Feedstock 29. Estimations of bagasse values were based on detailed studies by the Indian Bureau of Industrial Costs and Prices, Development Consultants Pty. Ltd., and SPB Projects and Consultancy Ltd., as well as the World Bank appraisal and post-evaluation of the Tamil Nadu Newprint and Papers (TNPL) venture. In addition, information was gathered on market prices for traded excess bagasse, and an expressed willingness to pay of sugar mill owners utilizing bagasse in secondary manufacturing operations. On this basis, value estimates of Rs. 240 (economic) and 200 (financial) per tonne of mill wet bagasse were obtained.  - viii - The Proposed Project Project Description 30. The proposed project is essentially an investment program to upgrade sugar mills in Maharashtra whereby each participating mill would be retrofitted to enable the production of electricity and bagasse over and above internal mill requirements. The basic investment in each mill would typically involve the: (a) addition of a high-pressure boiler of 43 kg/cm2; (b) addition of a top ing turbo alternator with inlet steam pressure of 43 kg/cm ; (c) installation of a bagasse drying system which uses waste heat from the boiler flue gases; and (d) modification of the mill's electrical system to enable interconnection with the utility grid and dispatch of surplus power. 31. In structuring the proposed project, a conservative approach is taken by imposing an economic rate of return cut-off of 12% in selecting candidate mills. This requirement limits eligible mills to those of capacity 3,500 TCD and above, and those presently below 3,500 TCD but which would be expanding to at least this capacity. These are 13 sugar mills in Maharashtra which presently meet these criteria. If all 13 mills eventually participate in the project and each chooses to adopt the electricity maximizing configuration, the total power exported to the grid could be up to 102 MW with a total investment requirement of some Rs. 1,800 million. The most probable scenario, however, is for four mills out of the 13 which presently have attached paper mills to opt for the bagasse-maximizing configuration. In this case, the total power export would drop to about 87 MW. This is still a substantial amount of additional power. The total investment requirement would be about Rs. 1,465 million or about US$86 million. Each mill would require between US$5-9 million. Financing Plan 32. The proposed financing plan for the overall project is structured as follows:  - ix - Table 2: PROJECT FINANCING (Mn USS) Category Local Foreign Total Foreign Loan -- 28 28 Domestic Loan 15 -- 15 Total Loans 15 28 43 Farmers' Equity 13 -- 13 SDf 30 30 Total Equity 43 -43 Total Financing 58 28 86 Source: Mission estimates. 33. The foreign loan would provide for the foreign exchange cost of the project estimated at US$30 million. The loan would be made to GOI who would onlend the proceeds to the financing institution in charge of project implementation, IFCI. IFCI will onlend the proceeds of the foreign loan as well as consortium funds (i.e., funds also to be provided by ICICI and IDBI in accordance with present financing practices) to the sugar mills. Project Justification 34. The f.nancial rate of return to the planters (as project sponsors and equity investors) ranges from 36 to 65% in current terms, or 24 to 51% in real terms. This high rate of by return is explained by the small share of the investment cost provided and the soft nature of Sugar Development Fund contributory financing. 35. As a leveraged investment, a bagasse cogeneration project's return to the mill owners varies considerably over a range of project operational assumptions. Sensitivity analysis identified the duration of the cane grinding season (grinding days) and the investmenz cost as the variables having the greatest influence on project worth, followed by electricity and bagasse values. However, even under the most adverse circumstances (20% negative change in all parameters) the 3.7% real rate of return on a 3,500 TCD electricity maximizing investment remains above the cost of capital. Financial returns from a sugar mill viewpoint are thus considered acceptable. 36. Indicative economic rate of return figures for the overall Project can be obtained by weighting the rates of return for each mill with their respective investment costs and project sales. Assuming the  - x - participation of all of the 13 eligible mills which results in the export of a total of about 87 MW to the grid, the EIRR of the Project would be about 15%. Project Risks 37. The risks of this project are industry developments which may adversely affect continued operation of sugar mills, the improved technology for bagasse cogeneration which is largely untested in the Indian context, and the financial risks entailed by sales of surplus power to a utility. Environmental Impacts 38. The Project will have positive impacts on the environment through a number of ways. Firstly it will result in cleaner emissions from sugar mill boilers due to the installation of bagasse dryers. Boiler flue gases carry substantial amounts of unburnt particles which are all spewed into the atmosphere during normal operation. Bagasse dryers will capture some 65% of these entrained particles through a scrubbing effect. Secondly, the bagasse to be produceA in 4 of the 13 mills will be used as paper feedstock. They have the effect of substituting for pulp from trees which would otherwise have been felled, either locally or elsewhere. Thirdly, the 87 MW of total power generated from the Project would have obviated or delayed the construction of at least an equivalent capacity fossil-fuel thermal plant which is a far more environmentally polluting source of electrical energy than a bagasse cogenerator. Conclusions 39. While the Project faces agricultural, technical, and financial risks, they have been recognized and considered manageable given the potential benefits of the project. In addition, the Project will introduce a breakthrough to the Indian power sector by establishing an institutional, financial and contractual framework for cogeneration in the country on the basis of financial and economic criteria. As a result, its benefits could eventually extend beyond Maharashtra and beyond the sugar sector - as a considerable potential for cogeneration exists in the country, which is presently largely untapped due to institutional/financial constraints. Under these circumstances, the risks are considered manageable, and should be taken, particularly given the potential long term impact of the Project on the economy in general and the power sector in particular.  I. INTRODUCTION Background 1.1 Bagasse-based cogeneration for power export to the grid has Long been of interest to the Government of India and the private sector. In the recently completed EStAP 1/ review of the country's nonconventional energy program being managed by the Department of Nonconventional Energy Sources (DNES), it is one of the major options identified as economically promising and worthy of more detailed study. Consequently, the Government, through DNES, requested further assistance from the World Bank by way of a State-specific follow-up activity which would analyze the subject in greater detail and, if warranted by the findings, develop a project proposal suitable for financing consideration by international and/or domestic lending agencies. 1.2 There would appear to be strong a priori justifications for bagasse-based cogeneration projects in India. The country suffers from chronic and widespread shortages in electricity supply which, among other detrimental effects to the economy, result in significant loss of industrial productivity. The prospects for the next decade do not appear to be much better. Efforts to upgrade the efficiency of existing thermal plants and expand generation capacity are being made but are hampered by resource constraints. Against this backdrop of power supply deficits, there is a vast potential for non-utility power generation, estimated for all sources in a study (Hagler & Bailly, 1986) to exceed 2,000 MW in terms of financially attractive investments. Cogeneration using bagasse and cane residues was cited in the study as being the least expensive of all non-utility power options with an estimated total potential of 350 MW in Haharashtra. 1.3 The technology for bagasse cogeneration for export power is well-known and commercially practiced in several countries, notably Australia, Cuba, Dominican Republic, Mauritius and the United States (Hawaii). The gestation period for such projects is less than that for conventional thermal or hydro power plants. In India, given adequate market opportunities for power sales, bagasse cogeneration projects appear as attractive private investment options, capable of adding to the utility's supply capability while reducing reliance on scarce public funds, For many sugar mills, bagasre cogeneration projects could result in added income from electricity or surplus bagasse sales and lower production costs in downstream mill activities. l/ Energy Sector Management Assistance Program, a joint World Bank/UNDP/Bilateral Aid program executed by the World Bank. The study was published in November 1988 and entitled "Opportunitie3 for Commercialization of Nonconventional Energy Systems".  -2- Major Constraints 1.4 Despite these potentially significant benefits to the economy and the proven success in other countries, a bagasse cogeneration project selling surplus power to the grid is still to be realized in India. There are several reasons for this. Firstly, although the subject has been discussed in recent energy and industry symposia and a few sugar mills have made preliminary feasibility studies, by and large its detailed mechanics and viability are itill relatively unknown to both millers and electricity boards. Secondly, millers are more familiar with pulp production as an alternative use for surplus bagasse, some having already invested in attached pulp and paper operations or are selling their surplus bagasse to such operations in r.eighboring mills. Many mistakenly assume that cogeneration for export power cannot coexist with bagasse-based pulp and paper production. Finally, and perhaps the biggest impediment of all, is the lack of clear GOI and State government policies regarding the sale of privately generated power to the grid. In turn, this reflects the State Electricity Boards' (SEBs) understandable uncertainty about the reliability and value of cogenerated power and the impact of such schemes on their overall operations and system planning. There is lack of a concrete basis or framework for negotiations which would lead to a commitment by SEBs to purchase power from sugar mills at mutually acceptable rates. Objectives 1.5 The main purpose of the present activity is to provide such a framework by addressing the above issues in adequate detail in a project- oriented study 2/. To make the evaluation manageable but at the same time develop a project that is significant in scope, it was decided to focus on one state and structure a state-wide scheme involving all sugar mills in the state satisfying certain technical and financial investment criteria. Maharashtra State was chosen for the proposed project based on various considerations. It has the largest number of sugar mills of over 2,000 TCD (tonnes cane crushed per day) capacity. Currently, many of the smaller mills have applied for or have been granted permits to expand capacity. The State appears to lead all other sugar producing states in mill efficiency and profitability. On the utility side, Maharashtra is one of the more advanced and enlightened states in terms of examining grid power exchange between the SEB and private producers. Although its 2/ This report is based on the findings of a mission which visited Delhi and Maharashtra State from March 20 to April 6, 1989. The mission members were Ernesto Terrado (Mission Leader), Charles Feinstein (Energy Economist), Marc Heitner (Consultant - Financial Analyst), Norland Suzor (Consultant - Sugar/Bagasse Specialist), and Bjarke Wiese (Consultant - Power Engineer).  - 3 - electricity supply/demand situation is relatively bright compared to other states, peak demand restrictions are still in effect and energy demand is forecasted to grow at nearly 7X per annum through 2000. The cane crushing season in Maharashtra is between November and April, coinciding with peak power demand from the grid and a period of lowered hydro generation. 1.6 Despite the focus on a single state, the methodology, analysis and implementation plan developed in the proposed project would clearly be applicable as well to other sugar-producing states, with minor modifications to account for differing utility system configurations. This is the broader objective of the present activity.  - 4 - II. SECTORAL CONTEXT: THE POWER SECTOR 3/ Overview 2.1 India has made impressive progress over the last few years in expanding its power supply, increasing its installed capacity from 32,000 MW and energy availability of 114,000 GWh in 1981/82, to 54,000 MW and 189,000 GWh in FY 1987/88. An additional 80,000 MW of capa -ty is planned to be installed by the year 2000. However, despite this progress, shortages equivalent to about 20% of maximum power demand and 10% of total energy demand persist and the quality of electricity supplies remains mostly unsatisfactory. The costs to the economy of shortages and poor quality of supplies are magnified by remaining inefficiencies in end-use of electricity - the result mostly of inefficient commercial incentives to many markets and subsidized power prices to low voltage consumers, particularly agriculture. 2.2 Key constraints are the divided responsibility (between GOI and the States) for power development, political interference in the operations of the State Electricity Boards (SEBs) and weaknesses in the financial structure of the sector. Under the Seventh Plan, GOI has mounted initiatives to address present inefficiencies. The development of the relatively efficient central utilities, particularly the National Thermal Power Corporation (NTPC), has been accelerated. Efforts are being made to bring financial discipline to the SEBs: the Power Finance Corporation (PFC) was formed to mobilize additional resources for SEBs willing to make needed institutional reforms. GOI is reviewing its fueL supply policy for the sector. In addition to considering several fuel import options, it has sanctioned more domestic natural gas for power generation. Finally, GOI is reviewing its policy on private sector involvement in power supply and aims to ease regulatory and financial disincentives to private sector participation. The principal challenge facing GOI under the Eighth Plan is to ensure that the utilities' institutional development and efficiency improvement keep pace with their physical expansion. India's Commercial Energy Resources 2.3 India's commercial energy resources comprise coal, oil, gas, hydroelectricity and nuclear energy. Reserves of thermal coal have been estimated at over 125 billion tonnes of which 60 billion tonnes are considered economically recoverable. However, coal quality generally is 3/ Much of the general power sector discussions in this Chapter are condensed versions of those in IBRD Staff Appraisal Report "Indla: Maharashtra Power Project", May 1989.  - 5 - poor (ash content up to 50%) and is getting worse. Recently, GOI has begun to give consideration to imports of coal. Proven and probable oil and gas reserves are estimated at 580 million tonnes oil equivalent (mtoe), sufficient for only 20 years' supply at present rates of consumption. With the recent completion of a large gas pipeline, gas is now becoming an important fuel for power generation. Overall, however, oil products continue to have limited use in thermal power generation, being confined primarily to stabilizing combustion in coal-fired stations and to fuelling captive generating plants. 2.4 India's hydroelectric potential is equivalent to about 100,000 MW but only 16,000 MW have been developed. Some 47,000 MW are under construction and a further 23,000 MW are at various stages of planning. Due to inadequate financial resources in states with the greatest hydro potential, lengthy disputes over water rights and environmental issues, and limited technical resources for the preparation of large hydro projects, the pace of India's, hydroelectric development has slowed over the past decade. Organization of the Power Sector 2.5 Responsibility for electricity supply is shared between GOI and the States. GOI controls CEA. NTPC9 NHPC, the Rural Electrification Corporation (REC), and (through CEA) the Regional Electricity Boards (REBs). The states control the SEBs and the day-to-day operations of the REBs. CEA was created in 1950 to develop a national power policy and to coordinate power utilities. CEA is part of the Department of Power within the Ministry of Energy. 2.6 NTPC, NHPC and REC are public corporations reporting to the Department of Power. NTPC and NHPC were formed in 1975 to construct and operate large power stations and transmission facilities and to sell bulk power to the SEBs. REC was formed in 1969 to coordinate rural electrification and provide financial and technical expertise for SEB schemes. REC finances more than 70% of rural electrification investments. 2.7 SEBs were instituted under the Electricity (Supply) Act, 1948, to promote power development and to regulate private licensees, such as the Tata Electric Companies. Although SEBs are supposed to be autonomous, in practice they are under the control of state governments in such matters as capital investment, tariffs, borrowings, and salary and personnel policies. As a first step towards integrating power supply nationally, SEBs have been grouped into five regional systems, each coordinated by an REB. Activities coordinated regionally include states' generation schedules and overhaul and maintenance programs, power transfers and concomitant tariffs.  - 6 - 2.8 Recently, GOI formed the Power Finance Corporation (PFC) the main functions of which are to mobilize additional resources for sector development and to pursue institutional reform of sector entities, particularly the SEBs. The Corporation's lending operations are focussed on completing priority rehabilitation and distribution projects being implemented by SEBs. Figure 1 in Annex 1 depicts the existing power sector organization. Tariffs 2.9 In general, SEBs' tariffs do not cover the long run marginal cost of supply (LRMC). Moreover, there are sharp differences in the structure of electricity tariffs with industrial tariffs at or above LRMC, and agricultural tariffs heavily subsidized. The average level of SEBs' tariffs is estimated at approximately 60-70% of long run marginal cost wnich represents an improvement from about 50% of LRMC in 1981 following real tariff increases which have averaged 5% per annum. In Maharashtra, the average level of retail tariffs is one of the highest in India and is estimated to be 74% of marginal cost. The level of SEBs' tariffs relative to LRMC will continue to improve as rates are increased to meet at least GOI's minimum rate of return target. However, the structure of tariffs remains unsatisfactory. Tariffs do not distinguish between the costs of supplying peak as opposed to off-peak energy, frequently are excessively complex and invariably heavily cross-subsidize low voltage consumers. Despite GOI accepting the principle in both its Sixth and Seventh Plans that energy prices should "reflect true costs", social and agricultural objectives have sharply limited progress towards this objective. Moreover, very little has so far been done to improve tariff structures through tapping the considerable potential for consumer load management. In Maharashtra, the World Bank financed Maharashtra Power Project includes a component which will help MSEB implement tariff reforms in the state. Most probably, these will include the introduction of peak and off-peak retail tariffs and the launching of additional load management initiatives. Notwithstanding these efforts, resistance to economic power pricing in India, at least for some categories of consumers, is such that progress in pricing reform with the SEBs is likely to be slow. 2.10 The tariffs of NTPC and NHPC have several shortcomings, notably that they are based on average generation costs and (similar to SEB tariffs) do not distinguish between supply costs in peak and off-peak periods. Recently, GOI has established a committee to recommend the principles on wnich both NTPC and NHPC should set their tariffs. It is widely expected that the committee will recommend a tariff structure more closely aligned to the structure of the Corporations' fixed and variable costs. This would be an important step towards ensuring that NTPC's and NHPC's stations are utilized as efficiently as possible, and towards more closely integrating operations of regional grids.  - 7 - Electricity Supply and Demand in the Western Region 2.11 The Western Region consists of the states of Maharashtra, Cujarat, and Madhya Pradesh and the Union Territories of Goa, Daman, Diu, Dadra and Wagar Haveli. The importance of the Region is reflected in its high share of installed generating capacity and electricity consumption shown in Table 2.1. Table 2.1: TOTAL INSTALLED CAPACITY FOR REGIONAL POWER SYSTEMS Region Total Installed Total Installed March 1987 March 1990 (Actual) (MW) (Estimated) (MW) Northern 13,365 18,490 Western 14,683 19,463 Southern 12,590 15,828 Eastern 7,766 9,666 North-Eastern 813 1,215 A&N, Lakshadeep Island 11 23 Total 49,288 64,865 Source: CEA. 2.12 The shortage of power in the Region is critical and results in severe supply restrictions in some of the states (Annex 1). In spite of the existing shortage of capacity, the constrained peak demand has grown at an average rate of 11.4% during the last seven years. Demand for peak capacity is expected to grow at an average annual rate of 8.6% until 2000. Planned rapid expansion of installed capacity notwithstanding, a capacity deficit of 3,050 MW to 3,500 MW by CEA norms is anticipated during the years 1989-2000. In the same period it is expected that the load factor will increase, so the long term energy demand projection will be met. Power Situation in Maharashtra 2.13 The power scenario in Maharashtra is in general characterized by steady development of generating capacity and transmission and distribution facilities. The state has given high priority to development of the power sector as a basic condition for a continued economic growth. Compared to the neighboring states, Maharashtra is in a favorable situation, but there are still constraints which have to be resolved before the power situation is fully satisfactory.  - 8 - 2.14 Between 1981/82 and 1987/88, electricity consumption in Maharashtra increased at 8.81 per annum in terms of energy consumed and 1.7Z per annum in terms of the mazimum load met (Annex 1)e Principal consumption shares in Maharashtra in 1987/88 were: industrial 52%, agricultural 19%, domestic 16%, and commercial 7%. As in the case of national consumption, energy consumption in Maharashtra and the maximum load met have both been constrained by inadequate supplv capacity. The installed capacity amounts at present to 7,032 MW of which 6,955 KW is available due to plant derating. When an availability factor of approximately 74% (average for 1987-88) is included, the firm energy in the system is approximately 5,145 KW. The annual peak demand occurring on January 23, 1989, was measured at 5,255 MW. In order to keep the capacity deficit manageable, at present certain restrictions are imposed on industry. These are mostly of administrative nature, in the form of agreements with some major consumers that efforts should be made to reduce their peak loads by 10-15% during system peak hours. As opposed to the other States in the Western Region, physical load shedding in M4aharashtra has up to present only been used occasionally and mostly in case of emergencies. CEA estimates that present unconstrained demand in Maharashtra is about 14% higher than maximum load met. However, CEA projects that through 1999/2000, maximum load in Maharashtra will increase more slowly than in the past (6.9% per annum), while maximum generating capability will increase relatively more quickly. Together, CEA expects that by 1999/2000, these effects will close Maharashtra's power deficit to the equivalent of about 6% of the maximum load met. Increases in Maharashtra's energy generating capability are projected to keep pace with increases in energy demand. Demand increases in Maharashtra are expected to follow the national pattern, with strongest growth in domestic and agricultural consumption (both expected to average about 10.7% per annum) and declining shares of industrial and commercial consumption (Annex 1). 2.15 The Maharashtra State Electricity Board (MSEB) is the region's largest utility. MSEB generated or purchased 32,962 MWh in 1988 (25,670 MWh on MSEB's own power stations). MSEB's own generation consists of 19,842 KWh (77%) from thermal plants, 2,562 MWh (10%) from gas-fired plants and 3,266 MWh (13%) from hydro plants. The total sales were 26,500 KWh. The difference represents exchange of energy with other States and NTPC, purchase of energy from private utilities and system losses. Import/export in 1987/88 from other bodies was 1,403/2,158 MWh, with the highest load imported 490 MW (NTPC) and the highest exported 464 MW (see Annex 1). Surplus energy is exported during MSEB's off-peak hours to Karnataka, Goa and Gujarat (the line to Karnataka was disconnected in early 1989). The inter-state lines are capable of transport of substantial amounts of energy, but due to institutional barriers and inter-state tariff inefficiencies, utilization of the transfer capability is very limited, Transmission and distribution losses were 14.5% in the period which are believed to be predominantly of a technical nature.  - 9 - Private Power Generation 2.16 Maharashtra is home to two of India's four rivate utilities 4/. Tata Electric Companies (TEC), a private generati.l and bulk supply utility, is based in Bombay. Bombay Suburban Electricity Supply (BSES), a private distribution utility, provides supply to final consumers in the Bombay metro ares. TEC is seeking to install additional gas-fired capacity and to improve the utilization of an existing hydro station through its conversion to a pumped-storage scheme. BSES, an efficiently- run distribution utility, has proposed expanding its involvement in the sector through a thermal generation project. India wide, private utilities at present make only a marginal contribution to public electricity supply, though private captive generation is extensive (equivalent to about 15X of public supply). 2.17 In recent years, notwithstanding impressive achievements in commissioning generating plants, it has been increasingly recognized that the public sector alone will not be able to satisfy the ever increasing investments needs for power generation, transmission and distribution. This is both due to the size of the capital requirements (the preliminary estimate for the VIII Plan 1990-95 is of the order of US$60 billion), the limited capability of public sector utilities to generate surpluses available for investment, and the constraints on the availability of external financing. As a result, the possibility of mobilizing private investments for the sector is under review both at the national and the State levels. 2.18 To that end, it will be necessary to rejAove a r.umber of impediments which are essentially financial, technical and legal in nature. On the financial side, it is unclear whether a private entity selling to a public utility would be restricted to a Reserve Bank of India rate plus 2% return in nominal terms on equity as provided for in the Electricity (Supply) Act (1948); such a return will obviously not meet private sector return targets, particularly considering the risks involved in building power plants. Furthermore, other constraints exist on allowable depreciation rates, and the availability of long term financing. On the technical side, procurement, if restricted to the domestic market, is not anticipated to be cost effective for many items in view of the small number of domestic suppliers. On the legal side, various provisions in the present legal instruments 5/, particularly in relation to certification requirements (a private generator could be a 4/ The other private utilities operate in Ahmedabad and Calcutta. 5/ The main ones being the Indian Electricity Act, 1910, the Electricity (Supply) Act, 1948, the Industrial Policy Resolution, 1956, and the Indian Electricity Rules, 1956.  - 10 - licensee or a promoter) may have adverse implications under certain circumstances; furthermore, 'he administrative procedures for the certification of private power plants are expected to be long - for instance in Maharashtra, some 19 approvals, certificates, clearances and sanctions have to be satisfied just for private cogeneration 6/. 2.19 Nevertheless, working groups set up both by the Central Government and the State Government of Maharashtra have recognized the potential benefits of private sector involvement in power supplies and have recommended some measures to accelerate the process. Among the schemes considered, it would appear that collective-captive generation (in which a group of industries establish jointly a generation facility designed primarily to meet its own power needs) and cogeneration projects will be the easier to implement in the Indian context in view of their nature and legal status. Still, important clarifications will be necessary prior to the entry of the private sector into this area. It is expected that the Central Government will announce its new policy shortly, which should pave the way for new sources of financing for the sector. 2.20 In the specific case of Maharashtra, while the state is relatively better off than most, it remains that a capacity shortage of 1,200 MW is expected at the end of the VIIth Plan, and of 1,300 MW at the end of the VIIIth Plan 7/. Furthermore, it is widely felt that investments in the power sector have been unbalanced, particularly at the expense of distribution. As a result, the Government of Maharashtra (COM) is particularly supportive of a strategy in which the private sector becomes involved in generation and even in certain major transmission programs, while MSEB focusses more on distribution and loss reduction programs. GOM is actively seeking proposals from the private sector regarding projects to be constructed during the Eighth Plan. While seeking to mobilize additional resources for the sector, COM is conside.-ing innovative project implementation arrangements, such as build-operate-turnover (BOT) schemes to ease pressure on technical resources available in MSEB and GOM. 6/ Report of the Working Group on Private Sector Participation in Power Generation, February 1988, p. 88. This list is reproduced in Annex 1. 7/ Ibid., pp. 5-6.  - 11 - III. SECTORAL CONTKE: THE SUGAR SECTOR Overview 3.1 Sugar plays an important role in India's agriculture, as it accounts for about 15Z of India's agricultural output in value from only 2% of cultivated areas. There are an estimated 35 million planters, the majority of them small, and sugar mills employ an estimated 300,000 workers, of which about half are permanent and the rest seasonal. Being cash cropping, and largely targeted to modern agro industries, the sugar industry is often perceived as an instrument for rural development. Sugar mills not only create direct employment, but also invest in infrastructure (particularly roads) and promote rural development, for instance by sponsoring health facilities and learning institutions. 3.2 India has recently become the largest producer of sugar in the world. Growth of output has been uneven, largely because of changing Government policies, and climatic conditions. Over the past five years, however, Government policies have been increasingly supportive, so that it is anticipated that the historical pattern of growth and recession periods will be arrested. This chapter reviews the sugar demand and supply situation and the sector's institutions. It concludes with an evaluation of the financial framework governing the operations of sugar mills, with a special emphasis on Maharashtra, the State where the proposed Project would be located. The Demand for Sugarcane Products 3.3 Refined (centrifugal) sugar is obviously the most important product. Consumption of sugar has increased from 4.5 million tonnes 8/ in 1977/78 9/ to 9.4 million tonnes in 1987/88, equivalent to 7.7% growth per annum. About 30-35% of the sugar is consumed in bulk form, and 60- 65% directly by households. Overall per capita consumption at 11.8 kg at present is low by international standards (where it is of the order of 8/ Unless otherwise noted, all statistics are from the February 1989 issue of Cooperative Sugar, a publication of the National Federation of Cooperative Sugar Factories Ltd., New Delhi. 9/ The sugar year corresponds to the milling season, i.e. October to May. The prevailing financial year in India is April 1-March 31. Sugar mills often followed a fiscal year in line with the crushing season, i.e. October 1-September 30. They are now switching to a fiscal year starting April 1.  - 12 - 30-50 kg) even when taking into account the quantity of sugar consumed in the form of gur and khandsari. 3.4 Gur and khandsari are the traditional forms of sugar in India. They are produced through an open pan process at atmospheric pressure, through traditional methods. The extraction process is inefficient so that a portion of the sugar in the cane is not recovered. In recent years, consumption of gur and khandsari has been on the order of 8-9 million tons, without a discernible trend. On the other hands per capita consumption has declined by nearly 30% since 1977/78. There has therefore been a shift in sugar consumption towards centrifugal sugar, which is taken as a clear indication of an improvement in the living standards of the population. The balance of the demand for sugarcane is for seeding and chewing purposes. 3.5 Important by-products of sugar manufacturing are molasses and bagasse. Molasses is used for the production of alcohol. India's alcohol production has thus increased from 430 million liters in 1977/78 to 619 million liters in 1986/87, which exceeds the domestic requirements (largely industry). As a result, India exports both alcohol and molasses. Some mills have diversified downstream to produce acetic acid, an important raw material for the chemical industry. 3.6 Bagasse is the fibrous residue of sugarcane, which accounts for about 30X of the cane in weight. In Maharashtra, bagasse today is utilized in two principal end-uses: (a) As a fuel in existing low pressure sugar mill cogeneration systems designed to meet internal steam and electricity loads; and (b) As a pulping feedstock in small scale paper factories attached to sugar mills. A detailed review of the status and prospects for the use of bagasse in Maharashtra is the subject of Chapter IV. The Supply of Sugarcane Products 3.7 Prrduction of sugarcane has increased from 176 million tonnes in 1977/78 to 197 million tonnes in 1987/88 (Annex 2). The main producing States are Uttar Pradesh, Maharashtra and Tamil Nadu. There are wide differences in yields and end-use between the States however as follows:  - 13 - Table 3.1: PRODUCTION AND USE OF SUGARCANE, 1987/88 Production Yields Utilization Cane Sugar Sugar Gur & Khand. (mn MT) (1) (MT/ha) (%) (W) (%) Andhra Pradesh 10.3 5 68 9.2 64 28 Karnataka 14.4 7 86 10.1 43 48 Maharashtra 25.0 13 86 10.8 90 0 Tamil Nadu 21.8 11 105 9.2 39 51 Uttar Pradesh 93.4 47 52 8.9 32 54 India 196.7 100 60 9.7 47 41 Source: NFCSF. As indicated above, Maharashtra has by .X,r the highest ratio of cane conversion to sugar, and with only 13X of the sugarcane, accounts for 30% of the centrifugal sugar production; the State also has a high yield of sugarcane, and the highest sugar content in the cane. In line with demand trends, the national percentage of sugarcane converted to gur and khandsari is declining (it was 50X in 1977/78). 3.8 Production of sugar has increased from 6.5 million tonnes in 1977/78 to 9.1 million tonnes in 1987/88 i.e. at an average rate of 3.5Z per annum. As production did not suffice to meet the demand, the balance of the requirements was imported. Net imports 10/ have been declining in recent years, from 1.6 million tons in 1985/86 to 40,000 tons in 1987/88. The Conversion Industry 3.9 During the 1987/88 season, India had 357 factories in operation. Their technical performance can be sunmmarized as follows: 10/ India is both an importer and exporter of sugar at the margin.  - 14 - Table 3.2: PERFORMANCE OF SUGAR FACTORIES IN THE MAIN STATES 1987/88 No. of Cane Crushed Average Milling Capacity Factories (000' MT) TCD Days Use (%) Andhra Pradesh 31 5,934 ,507 127 93 Karnataka 25 7,357 1,975 149 94 Maharashtra 90 25,828 1,839 156 104 Tamil Nadu 25 6,505 1,859 183 113 Uttar Pradesh 104 29,967 1 665 173 138 India 357 93,933 1,719 153 112 Source: NFCSf. The above table calls for a number of comments: (i) the size of the average factory in India is particularly small by international standards 11/; (ii) the apparently high capacity use is a function of the number of milling days during the year - in India, it is based on 130 milling days; and (iii) the number of milling days fluctuates widely from one year to the next 12/. 3.10 The Government is cognizant of the necessity to increase plant capacities and announced in January 1987 that new units will only be licensed if their capacity exceeds 2,500 TCD and the minimum spatial distance from the nearest mill exceeds 40 km; furthermore, expansion would be allowed up to 3,500 TCD subject to the availability of 11/ In other countries, it is estimated as follows: Argentina - 4,788 Australia - 8,268 Brazil - 8,296 Cuba - 4,037 Philippines - 4,798 South Africa - 6,371 Thailand - 5,833 Source: ISMA. 12/ Over 1978/79-1987/88, the lowest number of milling days was 86 (1979/80) and the highest 173 (1981/82).  - 15 - sugarcane. Expansions up to 5,000 TCD would be allowed only on the basis of increased productivity, as opposed to acreage increases. 13/ Supply/Demand Prospects 3.11 The VIlIth Five-Year Plan is presently under preparation. According to preliminary estimates, the per capita consumption of refined sugar would increase to 15 kg in 1994/95 resulting in an overall demand that year of 15.5 million tonnes, i.e. 60Z above the 1989/90 target of 9.8 million tonnes. This production target is anticipated to be met through a 25Z only increase in the area under cultivation, so that the balance would be made up through a combination of higher yields (from 60 tonnes/ha at present to 65 tonnes/ha); and a higher share of sugarcane being processed in sugar mills (from 40% at present to 601), at the expense of gur and khandsari. The installed capacity of sugar mills would be correspondingly increased from 9 million MT/year at present to 14.1 million MT in 1994/95, with the greater portion of the increase coming from expansions, and the balance from new sugar mills. Institutional Aspects 3.12 National responsibility for the sector is divided among several ministries. Sugarcane is under the Ministry of Agriculture. Regarding sugar, the main policy making body is the Sugar Directorate in the Department of Food under the Ministry of Food and Civil Supplies. The Directorate, headed by a Joint Secretary, plays a key role in the determination of pricing policies, and the articulation of incentives for new investments 14/. In addition, the Directorate administers the Sugar Development Fund, an important source of financing for the sector. Licensing of sugar mills comes under the purview of the Ministry of Industry which acts upon recommendations of the Sugar Directorate. Molasses and alcohol are under the care of the Department of Chemicals 13/ A relaxation of this policy was announced in early April 1989, whereby grassroots mills of up to l,t750 TCD capacity, at a minimum spatial distance of 25 km from existing mills, would be authorized under certain circumstances provided that they expand to 2,500 TCD within five years of initial production. This relaxation is not without certain risks, since it might trigger a competition for sugarcane from the established mills in a given area. 14/ The minimum price of sugar cane to be paid by the sugar mills to the farmers is determined annually in each growing area on the basis of yields and other factors by the Commission for Agricultural Costs and Prices.  - 16 - and Fertilizers. Problems of common interest are referred to a committee of secretaries from the different ministries. 3.13 State governments are also involved in the formulation of policies in the sugar sector, for instance by fixing certain prices, by licensing power generator sets, by providing equity financing to new mills, and by guaranteeing loans made to sugar mills. In Maharashtra, the Sugar Directorate under the State's Ministry of Cooperatives ensures that sugar cooperatives operate in accordance with the State's Cooperatives Act. 3.14 The cooperative sector accounts for 55% of the sugar mills in operation and 58% of sugar production at the national level. Sugar cooperatives, owned largely by small planters, are organized in State Federations, which, in turn, are coordinated by the National Federation of Cooperative Sugar Factories (NFCSF); NFCSF is very active in representing the interest of its members at the national level, in the provision of technical assistance to sugar cooperatives, and in the publication of periodicals concerned with the sugar sector. In addition, the cooperative movement has set up a research and training school for sugar technologists, the Deccan Sugar Institute (DSI) (now known as Vasantdada Sugar Institute) near Pune (Maharashtra). NFCSF is financed by way of contributions from its members. 3.15 Sugar mills in the private and joint sectors are grouped in the Indian Sugar Mills Association (ISMA), which aims at furthering the interest of its members at the Goverament level. ISMA also produces well-documented publications on policy matters. 3.16 The lead bank in the provision of long term financing to the sector is the Industrial Finance Corporation of India (IFCI); as such it carries out project appraisals on behalf of a consortium of development banks. IFCI has also been designated as nodal agency for administering the Government's Sugar Development Fund and as such supervises the use of the funds, and the repayment of loans. IFCI has hosted seminars on sugar industry matters, and publishes occasionally on sector issues. Financial Aspects Pricing and Taxation Policies 3.17 India's sugar economy has been largely isolated from the international markets. That is the result of a deliberate Government policy which has encouraged the development of the sector, essentially to meet the domestic demand for sugar. As a result, while the international sugar industry is facing considerable difficulties, the financial policies adopted in India have enabled the industry to expand, without  - 17 - being affected by low-priced imports or the necessity to export at a loss. 3.18 The sugar sector is heavily regulated from a financial standpoint as the Central Government determines: (i) a minimum purchase price of sugarcane from the farmers; (ii) a selling price for a portion of the sugar sold at a relatively low price - this sugar is referred to as levy sugar; (iii) the proportion of free and levy sugar; and (iv) taxes on free and levy sugar, and on sugarcane including the Sugar Development Fund 15/. In addition, by determining the monthly releases of free and levy sugar, as well as sugar import levels, the Government influences free sugar prices in the short term; by licensing new sugar mills or their expansion plans, the Government implicitly influences the free sugar prices in the long term. Regarding molasses and alcohol, these products are in surplus in India, and fetch relatively high prices on the international markets; the domestic prices of these products are regulated. Lastly, the percentage of sugar a mill is allowed to sell on the free markets is used as an incentive for investments in new mills, and their expansion 16/. In addition to the pricing and taxation regime determined by Central Government, state governments usually impose: (i) 15/ To the extent that SDF is financed by way of a cess (established through the Sugar Cess Act, 1982), it constitutes a tax on the mills. 16/ Other variables introduced to determine the incentive level (in the form of ratio of sales on the free market) are the investment cost, and the anticipated sugar yields. For instance, for new mills whose capital cost exceeds Rs. 70 million, the free sale quota for the first 55,000 tonnes of sugar produced during the year, and subject to other condition3 being met (essentially related to productivity targets) is as follows: Year High Recovery Area Low Recovery Area 1 60% 90% 2 78% 92% 3 78% 92% 4 78% 93% 5 78% 93% 6 - 93% 7 93% The normal quota for free sugar in 1988/89 is 55%, so that the incentive given to new mills is substantial - free sugar prices are 30-50% higher than those of levy sugar. Moreover, since the incentive is higher in low recovery areas, it reduces the natural advantage of high recovery areas - to that extent, this policy is at the expense of the efficient use of investment capital in the economy, but may be justified on developmental grounds.  - 18 - various taxes on sugar and cane; (ii) an electricity duty (on power generated to meet the internal requirements of the mills); and (iii) a minimum sugarcane price, which is usually higher than the one imposed by Central Government. The only product whose pricing is free appears therefore to be bagasse. 3.19 The evolution of important prices and taxes in recent years can be summarized as follows: Table 3.3: KEY FINANCIAL PARAMETERS Year Min. Price Cane Levy Sugar Price a/ Free Sugar Price Levy Sugar (Rs./MT) Index (Rs./kg) Index (Rs./kg) Indox (%) 1983/84 135 100 3.46 10o 4.56 100 65 1984/85 140 104 3.46 100 4.73 104 65 1985/86 165 122 3.91 113 5.86 129 55 1986/87 170 126 3.98 115 5.87 129 50 1987/88 185 137 4.22 122 6.13 134 50 1988/89 195 144 4.30 124 6.30 138 45 a/ The levy sugar price varies from State to State, and sometimes, within the State. Source: NFCSF; Mission estimates. 3.20 Nevertheless, the financial framework, which pre-dates India's independence, is extremely complex, and as a result of this complexity: vi) it is rather difficult to assess objectively the financial condition of the sector as a whole, and its contribution to the economy, so that each growing region and almost every mill has to be analyzed on its own merits; (ii) an investor in the sugar industry takes a "political" risk to the extent that favorable policies today may be changed adversely by future Governments; and (iii) it explains the importance to the industry of having representative groups which advocate their cause with the Government. As a result, sugar is a matter which attracts an unusual level of political interest, undeniably also because of the large rural population which derives its livelihood from the sector. The Financial Framework of the Sugar Sector in Maharashtra 3.21 Out of 90 factories in operation in Maharashtra during the 1987/88 season, 85 were in the cooperative sector. The cooperative sector began to develop in the State in the early 1950's and, with political support, has expanded rapidly, so that at present the private and joint sectors play a minor role in the State. 3.22 Financial Objective, The most important characteristic of the sugar mills in the cooperative sector is that they aim at maximizing the income of the planters, through a pricing system for sugarcane which  - 19 - ensures that the mills do not make any profits; this is achieved by basing the initial price of sugarcane on the Government's minimum price (para. 3.18), and its final price on the financial performance of the sugar mill 17/. While the cooperatives do not generate surpluses, with the consent of the planters, they retain a fraction of the funds due in the form of deposits (some of which are non-refundable) which are used in turn to finance productive investments, infrastructure items (particularly roads) or social amenities (health, colleges including medical and engineering institutions). This approach also has the advantage of enabling the mills to avoid income tax liabilities. 3.23 Prices and Taxation of Sugar in Maharashtra. The centrally- fixed minimum price of sugarcane is at present Rs.220/tonne, and the minimum fixed by the State is Rs.240/tonne. From the latter amount some Rs.34/tonne have to be deducted for State taxes 18/, with a corresponding reduction in the net price paid by the mill. As mentioned above, the ultimate price of sugarcane depends on the mills' financial performance. As an indication, in 1987/88, when taxes were essentially similar, the actual price paid was of the order of Rs.230-302/tonne. 3.24 Mills pay an electricity duty of 10 Paise/kWh (US40.6/kWh) which goes to the State. 3.25 The centrally-fixed price of levy sugar varies from State to State, and for big producers, within the States; thus for South Maharashtra it is Rs.4,060/tonne (the lowest in India) and for North Maharashtra, is at Rs.4,470/tonne (about the national average). Needless to say such a policy runs the risks of discriminating against efficient producers, and/or promoting the growth of cane in areas where other crops would otherwise be more economic. Taxes on levy sugar are additional to the selling prices, and are of Rs.520/tonne 19/. 17/ The final price is to be submitted to Maharashtra's Sugar Directorate for approval. 18/ Taxes on sugarcane are as follows (Rs./tonne): Chief Minister's Fund 5.00 Housing of Homeless 0.50 Small Scale Assistance 1.00 Drought Relief 2.00 Cane Development 8.00 Education Tax 1.00 Purchase Tax 16.60 Total Rs. 34.10 19/ R.s380/tonne of excise duty, and a Rs.140/tonne contribution to the Sugar Development Furd.  - 20 - 3.26 The price of free sugar varies during the year; in recent years, it has been of the order of Rs.6,OO0-7,OO/tonne, to which Rs.640/tonne of taxes have to be added 20/. 3.27 Contributions of Rs.7.50/tonne and Rs.3/tonne are levied on refined sugar for the Deccan Sugar Institute (para. 3.14) and the Maharashtra Federation of Sugar Cooperatives, respectively. 3.28 Financing of the Cooperative Sector. Different arrangements are in place for the financing of new mills and the expansion of the existing ones. Regarding new mills, they are typically financed 40% in the form of equity, and 60% in the form of long term debt. The equity portion is provided largely by the State Government (32.5% of total mill cost) and the balance by the planters (7.5%). While the State makes an equity investment in the new sugar mill, about 50% of the financing required for that purpose is obtained in the form of long term financing from the National Cooperatives Development Corporation (NCDC). Regarding the planter's equity contribution, a 2,000 TCD mill would normally have 6,000 planters. As the share price is Rs.2,000, planters are usually unable to provide such an amount in cash, so that only Rs.200 is required as a deposit while the balance is financed by way of a loan from the Reserve Bank of India (RBI). Regarding the loan portion, it is provided usually by a consortium of banks, with IFCI acting as the lead bank. As a result, the financing plan for a new mill can be summarized as follows: Table 3.4: FINANCING PLAN FOR NEW MILL (Rs. miliion) Equlty Government of Maharashtra Direct Contribution 26.0 Through NCDC Loan 26.0 Subtotal 52.0 Planters Directly 1.2 Through RtBi Loans 10.8 Total Equity 64.0 Loans 96.0 Total FInancIng 160.0 Source: IFCI; Mission estimates. 20/ Rs.500/tonne of excise duty, an a Rs.140/tonne contribution to the Sugar Development Fund.  - 21 - The striking feature of the financing plan is that through a cash investment of only Rs.1.2 million, planters are able to mobilize financing of the order of Rs.160 million. However, in the inital years of operation of new sugar mill, they have to repay the RBI loans (for shares acquisition), the long term loans, and purchase the Government of Maharashtra equity share in their mill 21/. In seven out of ninety cases, planters have been able to purchase the Government's equity in full, and in a number of others, they are close to attaining that goal. 3.29 Long term financing is normally obtained from the consortium of institutional development banks, i.e. the Industrial Credit and Investment Corporation of India Ltd. (ICICI), the Industrial Development Bank of India (IDBI), and the Industrial Financial Corporation of India (IFCI), all three well established financial institutions which play a major role in the financing of the public and private sectors throughout India. In accordance with the working arrangements between the three institutions, IFCI is the lead bank for the sector, so that it appraises and moni,ors almost all the ventures submitted for financing; its approach and methodology in evaluating projects are broadly similar to the World Bank's. The other two Banks rely on IFCI's appraisal reports to process projects. The Maharashtra State Cooperative Bank (MSCB) occasionally participates in the consortium 22/, and NCDC has recently been authorized to channel funds directly to sugar mills both for new ventures and for expansion projects. NCDC financial terms are anticipated to be better than those of the traditional financing institutions, which could create for the latter some problem in the long run. 3.30 Lending terms depend on the area (backward or other), and the size of the financing required as follows: 21/ For all practical purposes, this is achieved through deductions from payments made to the planters for sugarcane purchases. 22/ Precise data on the banks' exposure to the sugar sector in Maharashtra is estimated at Rs. 2.2 billion. At the national level, loans sanctioned to the sugar sector up to March 31, 1988 (including loans made for expansion) were as follows: Amount Sanctioned % of Portfolio (Rs. million) ICICI 1,442 2.0 IDBI 454 1.6 IFCI 2,419 4.0 MSCB (Maharashtra only) 503 14.0 Source: Annual Reports.  - 22 - Table 3.5: LENDING TERMS FOR NEW MILLS Backward Area Other Areas Interest Rate 12.5% 14% Amount (Rs, in) 20-50 no limit Grace Period (yrs) 2 2 Maturity (yrs) 6-7 6-7 Such a policy reduces the natural advantages of areas better suited for sugarcane production and processing through higher interest charges; this is done to achieve social and other goals. 3.31 Regarding expansion and modernization Rrojects, these are usually financed by way of equity from the farmers (15-20X%, semi-equity from the Sugar Development Fund (30-35%), and 50% by loans. The equity contribution of the planters is obtained through withholdings on payments f-or sugarcane. SDF loans are at 6% interest rate, up to 13 years, including a grace period of up to 8 years. Regarding the development bank loans, terms are broadly similar to those applicable to new mills as follows: Table 3.6: LENDING TEiRMS FOR EXPANSION AND MOOERNIZATION OF MILLS Backward Area Other Areas Interest Rate 11.5% 14% Amount (Rs. mn) up to 60 no limit Grace Period (yrs) 2 2 Maturity (yrs) 6-7 6-7 As the financing of expansion and modernization projects entails a lower level of risk, financing terms are more favorable, particularly if one takes into account that SDF provides long term financing at below-market rates. To the extent that SDF is financed by way of a tax on sugar and  - 23 - its financing terms are soft, its involvement can be assimilated to a form of equity 23/. 3.32 Working capital loans to Maharashtra sugar mills are extended essentially by the Maharashtra State Cooperative Bank (MSCB) and by Primary Agricultural Credit Societies (PACS) financed by District Central Cooperative Banks (DCCs) which are financed in turn by MSCB. MSCB has sanctioned some Rs.2.8 billion for that purpose which are for one year, and carry a 17.5% interest rate. The repayment of these loans is asaumed by the cooperatives through deductions from cane prices. 3.33 Financial Conditionality. Barring a few exceptions, the medium and long term loans from the development banks require a State Government guarantee 24/. This practice is largely attributed to the overwhelming influence of the Central and State Governments on the financial performance of sugar mills, for instance through the fixation of prices and tax levels, the licensing of new and the expansion of mills (which impact for instance on cane availability) and, particularly in Maharashtra's case, the support provided to the cooperatives, and involvement in day to day operations (for instance in the selection of key personnel), Furthermore, past experience has shown that presently supportive Government policies for the sector can be altered adversely by future Governments, hence the need for a protection from this risk at the Government level. The guarantee has been invoked in a number of instances, and the banks report that they were able to limit their losses through this mechanism, although not to the full extent (for instance, compensation for interest on payment in arrears has seldom been recognized); furthermore, recovery of amounts due required some time. 3.34 The banks have obtained additional undertakings from the sugar mills typically as follows: (i) the additional revenues on account of the excess of the incentive quota of free sugar (para. 3.18) over the prevailing quota has to be deposited in a special bank account earmarked for debt service; (ii) the sugarcane tax due to the State Government (para. 3.18) is to be treated as an interest free loan until bank loans have been repaid; (iii) a special provision in the purchase price of cane (Rs.10/tonne) is treated as a non-refundable deposit to meet debt service obligations; and (iv) a special provision in the price of sugar 23/ SDF does not maintain separate accounting records such as balance sheets and income statements, so that its financial standing could not be readily ascertained. Up to January 31, 1989, it had collected Rs.5.7 billion and disbursed Rs.1.4 billion (i.e. 24% of the proceeds). The authorities are now keen to accelerate disbursements from SDF. 24/ The State requires a 0.5% guarantee fee on the principal and interest balances outstanding.  - 24 - (Rs.200/tonne) is earmarked for debt service through a separate account established for that purpose. Lastly, IFCI often seeks a seat on the board of its borrowers to protect its interest. It would therefore appear that the development banks, with the support of the State Covernment, have taken a comprehensive set of measures to safeguard their loans to cooperatives. 3.35 Regarding working capital loans, these are not guaranteed by State Government but through pledges against receipts from sugar sales. The cooperative banks closely monitor sugar movement and sales receipts (by posting, for instance, their own representatives at the mills). Hence, it would seem that in this instance too, precautions have been taken to satisfy the lenders. 3.36 Financial Standing of Sugar Mills. The accounts of sugar mills in the cooperative sector are subject to an audit by an independent auditor appointed by the State. The accounts are published late, inter alia because of the obligation to seek State Government agreement on the final price paid to the planters for sugarcane. Sugar mills used to follow an October 1 - September 30 fiscal year reflecting the crushing season; however, in accordance with a recently approved regulation, the fiscal year to be adopted will be from now on April 1 - March 31. As a result of this change, it is anticipated that the next fiscal year will cover 18 months, i.e. October 1, 1988 - March 31, 1990. This will vastly complicate the monitoring of mills' financial performance for at least two years. 3.37 The financial condition of sugar mills in Maharashtra has not been the subject of recent systematic surveys or studies. Nevertheless, from discussions with development bank officials who are daily in contact with the industry, it would appear that there are wide differences between mills, some performing extremely well, while others are in financial difficulty so much so that at some point they reach the "sick" list 25/. It is estimated that about 40 mills out of 90 in the cooperative sector are in this category. The underlying factors seem to be: (i) Poor management - in the cooperative sector, managers are elected by the assembly of members. Candidates do not always have the required background, experience and qualifications to run a complex industrial facility such as a sugar mill. The State is aware of this issue, and now mill managers are selected through a panel, and it is envisaged that the appointment to senior positions in sugar mills will be subjected to similar procedures; (ii) Availability of sugarcane - for a number of reasons, cane may not be available in the quantities required 25/ "Sick" units being defined as those which have a negative net worth, or those whose accumulated losses have resulted in the erosior. of 50Z or more of their peak net worth during the most recent five year period. Source: IFCI - Sugar Development Fund Scheme (undated).  - 25 - to operate a mill efficiently. The reasons may be varied including drought, poor planting, competition from other mills, and inability of management to convince farmers to plant cane. The remedy here is better irrigation systems and the planting of high yield varieties, as well as better incentives to the farmers. During the 1987/88 season, 5 mills did not operate on this account; and (iii) Insufficient retention of deposits (through deductions from sugarcane payments) to enable mills to build up resources towards the financing of investments from internal resources. The Government is well aware of these problems and has appointed a committee 26/ to examine the problem and develop a rehabilitation program. The findings and recommendations of this committee are expected around June 1989. IFCI is also expected to carry out an in-depth review of the financial problems of the sugar industry whose conclusions are expected to be available in September 1989. 3.38 Financial Burden on the State. A persistent issue in Maharashtra is the financial burden cooperatives represent on State finances, particularly since the State does not obtain dividends on its equity investment in sugar mills, and payment of sugarcane tax is subordinated to debt service (para 3.34). 27/ While the improvements in the living standards in rural areas brought by sugar mills cannot be denied, it remains that approaches will have to be developed to wean cooperatives from State Government support in the form of equity investment or even as a guarantor of external loans. However, in the present social/political climate, this realistically represents a long term proposition. Meanwhile, the Central and State Governments have adopted measures designed to improve the financial performance of mills by requiring the implementation of larger mills and/or expansion projects. The proposed project supports these goals since it aims at improving the financial performance of sugar mills by taking maximum advantage of bagasse, the residue of cane whose full potential as an energy source has not been realized yet in India. 26/ The Shivajirao Patil Committee. 27/ The issue is potentially contentious, since cooperatives argue that they are providing services (education, health) which under normal circumstances should be provided by the State at no cost.  - 26 - IV. BAGASSE UTILIZATION AND VALUATION IN MAHARASHTRA Overview 4.1 In order to establish the current bagasse production, availability and utilization in the State of Maharashtra (SOM) sugar industry, a field survey was conducted 28/ in 37 of the 100 sugar mills located in SOM and listed in Annex 3. The surveyed sugar mills processed slightly more than 55% of the total quantity of cane ground by the SOM sugar industry during the 1987-88 season. In addition to the survey, a review and analysis of the past ten years published performance data was conducted. The composite picture of mill performance points to considerable scope for improved energy efficiency, with benefits potentially accruing to both the electricity and bagasse-based paper industries. Bagasse Production Present Production 4.2 Approximately 8 million tonnes of bagasse was produced by the SON sugar industry during the 1987-88 season. The cane milling process produces bagasse with a moisture content which generally averages 50% (wet basis). This bagasse consists of two major fractions, namely pith and fiber. The pith fraction varies from 45-50% of the bagasse. The fiber fraction is a proven feedstock for paper pulp, partially substi- tuting for wood or bamboo. 4.3 The parameter relating the weight of bagasse output to the weight of cane crushed is commonly referred to as "Bagasse X Cane" in the sugar industry. This percentage is mostly influenced by the fiber content of the cane which is referred to as "Fiber Z Cane." The fiber content of cane varies from mill to mill, from year to year and within seasons as a function of climatic conditions and soil characteristics, and hence Bagasse x Cane flucttuates from 28% to 36% at SOM mills. 4.4 For the purpose of evaluating the current availability of bagasse in SON, the statistical average of 13.9% has been taken as the fiber content of the cane. This ratio was applied to an empirically- derived linear regression relationship to establish that for most of the duration of the cane grinding season the average bagasse production from the milling process is of the order of 29.1% on cane. 28/ Consultants India, Ltd.  - 27 - Estimating Future Availability 4.5 Bagasse is produced and is available for use during the cane harvesting and grinding season which usually starts in October/November and terminates in April/May. The use of the bagasse may be extended into the off-season period only if the bagasse is adequately treated and properly stored to eliminate the risk of spontaneous combustion in storage. This combustion is caused by the progressive exothermic fermentation process which generally occurs in stored bagasse with a high moisture content. 4.6 The two principal factors affecting year-to-year bagasse availability, total cane supply and duration of the crushing season, have shown significant variability over the past decade (See Figure 4.1) 4.7 A simple analysis of Maharashtra-wide average crushing season data would obtain a season length of 151 days via arithmetic average methods. Such an analysis ignores the following complications: (a) The larger mills are generally "healthier" in terms of financial condition and access to adequate cane supplies. The large mills are often successful in outbidding smaller mills for cane supply from independent cane growers due to the formers' higher willingness to pay. Thus larger mills can be expected to have a longer crushing season. (b) A number of mills process cane at a higher than licensed equipment grinding rate. The result is to shorten the duration of the grinding season. Profit maximization under conditions of joint sugar, bagasse an, electricity production would provide mill owners incentives to lengthen the duration of grinding operations. 4.8 Cane supply is the fundamental variable affecting length of grinding season and hence potential capacity utilization of bagasse cogeneration equipment. An analysis of cane availability at the 37 surveyed Maharashtra mills is given in Annex 3. The figures confirm the above observations, and also suggest that substantial trading of cane occurs between Maharashtra mills. 4.9 The cane supply data was utilized to project the number of grinding days that would result at large (>2,000 TCD) SOM mills if available cane were crushed at licensed capacity (Table 4.1). The projected 184 grinding days was then normalized to an average cane growing year by comparing the survey year cane and sugar production to ten year historical averages. This index shows that 1987/88 was a representative year, being only 2% above the averages. The normalized figure of 180 days was used as the baseline indicator of bagasse availability at the large mills which are candidates for cogeneration investments.  India Bagasse Cogeneration Study Maharashtra Grinding Season and Cane Production, 1978-1988 Crushing Season (Days) Cane Production ('000 Tonnes) 300 3 250 30 200/2 6 150; 0 50. - 0 I I 78/79 79/80 80/81 81/82 82/83 83/84 84/85 86/86 86/87 87/88 Year ouc eason Trend r Cane Trend *Length of Season