Jul 27, 2001 - the so-called Availability-based Tariff (ABT)2 order by changing the method of computing capacity charges and energy charges. The ABT ...
238 India Infrastructure Report 2002
9
FRAMEWORK FOR THE ENERGY SECTOR
9.1 REGULATORY OBJECTIVES, PROCESSES, AND OUTCOMES: FIRST POWER TARIFF NORMS Ajay Pandey The first review of bulk power tariffs by an independent regulator, the Central Electricity Regulatory Commisssion (CERC), in December 2000 was an important event for the power sector, as serious anomalies had crept in the bulk power tariff regulations, over a period of time. In this section, we trace the process followed by the CERC and analyse its final conclusions. While the process followed by the CERC seemed open and the CERC seemed inclined to build a regulatory rationale, stonewalling by the regulated entities and low quality of participation by purchasers of bulk power are also evident. Finally, the ‘review order’ is full of inconsistencies and seems more of a balancing act than a move to a sound regulatory framework for determination of tariffs. The Central Electricity Regulatory Commission (CERC) of India was set up in July 1998 as an independent regulator by the Electricity Regulatory Commissions Act (ERC Act), 1998 with the powers and functions to regulate (a) the tariff of generating companies owned or controlled by the central government, (b) the tariff of generating companies selling power to more than one state, and (c) the tariff of transmission utilities engaged in inter-state transmission, according to section 13 of the Act. The commission assumed jurisdiction with effect from 15 May 1999 consequent to notification of deletion of Section 43A(2) of the Electricity (Supply) Act, 1948 in respect of the entities covered under the ERC Act. The CERC, upon assuming tariff jurisdiction, went through the process of fixing ‘tariff norms’ and came out with its first tariff norms
for generation and transmission on 21 December 2000 albeit after some jurisdictional challenges.1 The CERC’s order dealt with norms related to the rate of return (cost of capital), rate base, depreciation, operation and maintenance (O&M) expenses, as well as incentives and operating norms for fixation of tariff for bulk power generation and transmission under the CERC’s jurisdiction. Given the lack of history of independent regulation in India and its importance in the evolution of the power sector in future, we believe that institutionalization of the regulator is itself a key issue of governance within the sector. The CERC had earlier (in January 2000) come out with the so-called Availability-based Tariff (ABT)2 order by changing the method of computing capacity charges and energy charges. The ABT norms also entailed the creation of charges related to unscheduled interchange (UI) of 1
National Thermal Power Corporation (NTPC) and Power Grid Corporation of India Limited (PGCIL) argued that the CERC’s jurisdiction is constrained in its powers to operate within national power plans as reflected in notifications of the Government of India. The NTPC also brought forth the ‘level playing field’ argument. As the CERC did not have jurisdiction on independent power producers (IPPs), who continued to be guided by special GOI notifications, the NTPC claimed that fairness would demand that the CERC have no jurisdiction over it. 2 For a review of the same, see Sidharth Sinha, ‘Regulation of Tariffs and Interconnections: Case Studies’, and Puneet Chitkara, Rajiv Shekhar, and Prem K. Kalra, ‘Missing Interconnections in the Power Systems’, in 3iNetwork (2001).
Framework for the Energy Sector power. The objective of the ABT was to promote merit order dispatch and maintain grid-discipline. After the removal of certain glitches, it is now being implemented. The ABT norms are certainly important for grid system performance, and also for developing the market for trading of power consistent with short run marginal costs of generation. Yet, the fixation of tariff norms by CERC in December 2000 has even greater financial and economic implications on the entities under the CERC’s jurisdiction, and also on the evolution of tariff regulations in the power sector within and outside the CERC jurisdiction. Our focus is on the process followed and the context of the order in meeting the objectives of the same. The then existing tariff norms had the following infirmities: • Tariff norms were biased financially in favour of short-gestation projects irrespective of their overall economic efficiency, as the return on equity begins after commercial production starts and is equal for all. • The multi-project expanding entities could earn indeterminately high returns on equity, as the equity was assumed constant in a liability-side approach3 used to determine the rate base. • The economic returns (or return on equity) were sensitive to the choice of liability (debt) maturity taken for funding the projects, with higher maturity debt yielding higher returns in multi-project expanding entity context.4 This seemed an inadvertent fallout of the tariff notifications as the internal resources generated by an entity5 are 3
Determination of allowable profits to the regulated entity under rate-of-return regulations requires information on the capital invested by the entity (called rate base) and estimation of appropriate rate of return (also called cost of capital, at times). Usually, the rate base is based on the assets deployed as they appear on the balance sheet. Current power tariff notifications, including for IPPs, are based on the liability side of the balance sheet. Normally both the approaches are equivalent if the regulator keeps track of the liabilities corresponding to the assets deployed. However, the tariff notifications assume equity invested in the project to be constant and debt repayment and servicing linked with the actual terms of debt. These lead to divergence between actual assets deployed and rate base on which tariffs are determined with the passage of time. 4 This happens because of divergence between the assets deployed and the rate base (see fn. 3). The excess return by an entity due to this divergence is possible as the equity invested can be doublecounted in two or more projects. 5 The internal resources generated by an entity are calculated by adding depreciation to net profits and by deducting dividends, actual debt payments, and increase in working capital. It is considered as equity and can be applied to another project as such. Ideally, internal resources calculations should be based on debt amortized over the life of the project, irrespective of actual payment. Actual repayment can be used as an approximation for other purposes, but in the context of regulated return entities, this approximation
calculated on the applied as equity to refinance debt incentive to take
239
basis of actual repayment of debt and to new projects. The lack of flexibility in proportion to net assets creates the long-term, bullet repayment debt.6
• The returns on equity were not sensitive to the leverage deployed, even though financial risks would vary with the leverage. • While the returns on equity are not sensitive to the leverage deployed, the tariffs of specific projects/plants and assets are sensitive to the leverage imputed7 despite debt raised for projects is with recourse to the entire balance sheet rather than on project finance basis. This results in cross-subsidization across beneficiaries/consumers. • Incentives and disincentives to the beneficiaries for neither early payments (1 per cent rebate on tariff ) nor late payments (1.5 per cent per month on the amount billed) were linked with their effect on economic returns to the entity. These anomalies affect different aspects of the sector and even different stakeholders differently. The first infirmity listed above is comparatively unfair to hydro-power generating utility, the second, third, and fourth to the State Electricity Boards (SEBs) and final consumers, the fifth to some SEBs and final consumers, and the sixth has an indeterminate effect. Late payments in fact can affect both the upstream and downstream players negatively, to the gain of the exchequer, and is dependent more on the financial condition of downstream players, that is, the SEBs. Besides these anomalies, the useful economic life of the assets affecting allowable depreciation, operating norms, allowable O&M expenses, incentives, and the use of market value of assets in the rate base to reflect long run marginal costs (LRMC)8, were some of the other issues expected to be taken up by the CERC in its first review and fixation of tariff norms. distorts incentive to take long-term bullet debt, if the debt servicing is allowed for such debt. 6 Loans/debt for very long-term on bullet payment basis are not easily available though. 7 The tariff for each plant or project is based on equity imputed in that project. Each project is assumed to have a different proportion of debt and equity, based on internal resources generated/equity capital raised and applied across projects. This affects tariff as equity attracts higher returns as well as tax liability. 8 Long run marginal cost is the opportunity cost of supplying one unit of any good or service assuming that there is no capacity available. In such a case, costs will include the cost of adding one unit of capacity as well, and may make it different from short run marginal costs, when there is excess capacity. In this context, LRMC of electricity would be the same for all electricity produced. Historical cost, however, is based on the costs related to investment made in the past and would be different across plants and projects.
240 India Infrastructure Report 2002
ISSUES
IN THE
CONSULTATION PAPER
While setting out on a review and fixation of tariff norms, the CERC came out with a consultation paper, which identified various issues and also spelt out the preliminary position it took on these issues. The executive summary of the CERC’s consultation paper on tariffs (September 1999) identified the following issues: Unbundling of tariffs and incentives for environment management: In the consultation paper, the CERC took the view that unbundling of tariff for various services provides discrete price signals to the users and promotes economically efficient usage and decision making by them. The unbundling could be for (a) distinct organizational domain engaged in the provision of a particular service, (b) nature of different services such as capacity made available vs. actual usage, and (c) time of use of service based on demand consideration to reflect the cost of congestion. In the case of organizational domain unbundling,9 the concern was with the difference in timing of the creation of sub-capacities within an aggregate capacity in generation on the one hand, and physical or technical demarcation of transmission assets on the other. In its concern for environment management, the CERC also mentioned exploring the possibility of cross-subsidizing generation through renewable sources by imposing a levy on fossil fuel based generators, and disallowing costs incurred by generators on meeting environmental norms without consequential benefits. Choice of tariff setting methodology and tariff base: In the consultation paper, the CERC took a fairly open position with respect to the two tariff setting methodologies followed world-wide (a) Cost-plus or Rate of Return approach (in reality, incentives and disincentives are used commonly qualifying it as middle-of-the-road, incentive regulations), and (b) Price Cap approach (in reality, inflation adjusted price cap regulations), though it viewed the differences between the two approaches as more of degree than of substance. On the issue of tariff base10 (set of allowable costs), the CERC clearly took a stance of favouring the existing practice of defining the tariff base in terms of historical costs and not on LRMC. This is also a key difference between the two tariff setting methodologies. The use of price cap regulation would have entailed moving 9 Organizational domain unbundling refers to the delineation of different parts or units of the organization when they produce the same services. For example, two units in a power generating station could be thought of as two distinct tariff entities. This is particularly meaningful, if there are different set of consumers. 10 Tariff base consists of allowable profits based on rate base (see fn. 3), and other allowable operating costs.
over to LRMC based price cap, in case the long run marginal costs were substantially different from historical costs. The CERC’s argument for favouring historical costs as tariff base was on the grounds that this would avoid the swings in marginal costs due to lumpiness in investments in the face of demand–supply mismatch! Specification of allowable costs: Under this heading, the consultation paper addressed two sets of issues: (a) operational costs and norms (on which work had been done by the Central Electricity Authority in 1997), and (b) the other costs—rate of depreciation, taxes passed through in tariffs, risk weighted returns, and recovery of costs ex post. While in the case of the former, the CERC indicated continuation of the existing practices, it indicated its stance to review the existing practices on the latter set of issues. The rate of depreciation allowed to power utilities had been delinked from the actual life of the assets, ostensibly to allow sufficient cash generation to pay back debt raised, while in the view of the CERC, the depreciation should be linked with the actual physical life and investors/ utilities should repay the debt from their returns rather than burdening the consumers for the same. It also felt that tax as pass-through item in tariffs increases the burden on the consumer. And returns, if defined on pre-tax basis, would promote tax planning and may also reduce overall tax burden on consumers. In its opinion the current tariff structure did not deal with differences in risks within the sector while determining returns to the utilities/investors. The CERC also took a position in the consultation paper on cost escalations and interest incurred thereon in projects. The cost escalations in its view should be admissible only under exceptional circumstances and interest due to such cost escalations should not be pass-through. This would provide incentives for accurate cost projection, and adherence to the same. Market structure and scope of regulations: The CERC recognized the effectiveness of markets and trends in other countries to develop them through deregulation and lowering of entry barriers. As a regulator, it stressed the need to create the conditions stimulating competition, and to move towards the development of markets, wherever feasible without creating regulatory uncertainty. Seized of the endemic receivables problems, it sought to signal its intent and role in mitigating the same. The consultation paper by the CERC dealt with these issues openly. Unfortunately it also clearly indicated a misplaced preference for historical costs as tariff base, and therefore, for rate-of-return as the tariff setting method. At this stage, however, the anomalies in the existing tariff structure, as brought out earlier, were not delineated sharply.
Framework for the Energy Sector
CONSULTANTS’ REPORTS ON RATE OF RETURN (COST OF CAPITAL), DEPRECIATION AND O&M COSTS The CERC invited comments from all stakeholders on the issues raised in the consultation paper and also appointed independent consultants to prepare a position/discussion paper on some of the issues. CRISIL Advisory Services (CAS) was asked to submit a report on the appropriate cost of capital (rate of return) for the entities under the CERC’s jurisdiction and ICRA Advisory Services (IAS) was appointed to study and suggest depreciation norms. The terms of reference to the former included the choice between cost of capital approach and cost of equity approach (existing), the level at which the risks and returns were to be determined, and the approach to estimating the cost of capital, frequency of review of cost of capital, the effect of ownership (government or private) on the cost of capital, the treatment of foreign exchange risks, and effect on noncore businesses of regulated entities on the cost of capital. Also included in the latter was the appropriate depreciation method for depreciation, the assessment of asset life, appropriateness of accelerated depreciation, and appropriate method of determination of the asset base. Similarly, Development Consultants Ltd (DCL) and Water and Power Consultancy Services (India) Ltd (WAPCOS) were appointed to benchmark best practices for O&M costs and their escalation characteristics for suggesting norms for thermal and hydro-power generation respectively. These reports and discussion papers as well stakeholders’ comments were used later as inputs in fixing tariff norms by the CERC.
CAS Discussion Report Cost of Capital: The discussion report prepared by CAS on appropriate cost of capital (rate of return) made recommendations entailing changes from the existing practice. The recommendations and arguments are summarized below: • While the cost of capital approach11 is preferable, the existing methodology based on the cost of equity be 11 In the cost of capital approach, the appropriate allowable returns do not take into account actual deployment of equity and debt in rate base. The returns are allowed based on assumed normative capital structure (proportion of debt and equity), and interest on debt is not allowed as part of tariff base to prevent double counting. In implementing the cost of capital approach, the cost of (returns on) debt and cost of (returns on) equity are estimated to arrive at overall cost of capital by weighing the cost of debt and equity with respect to normative capital structure. In the cost of equity approach, the actual equity in the rate base is ascertained or assumed, the latter being the case in the present
241
continued as it is difficult to predict the normative cost of debt and will lead to higher than warranted returns to equity as the existing cost of debt is very low for the central sector utilities. • The cost of equity (or return on equity) may not be differentiated on business risks, vintage of assets, ownership, or mode of financing, but should reflect different level of leverage (financial risks) assumed by different entities till normative capital structure12 based overall cost of capital (distinct from cost of equity) is followed. • While foreign exchange denominated debt may be protected as pass-through in tariffs, the returns on foreign exchange denominated equity need to be lowered13 by an amount reflecting risk premium on foreign exchange risk, even though for a finite period such risk may be passed through in the tariffs. • The cost of equity may be determined using the capital asset pricing model14 (CAPM), with listed power sector and refining companies being used as proxies for estimating asset betas.15 The risk-free rate16 may be estimated using 3–4 months’ average yield-to-maturity (YTM) on government securities of residual maturity above eight years, and market risk premium17 be estimated tariff regulations, and allowed profits are based on the cost of (returns to) equity so determined. The interest on debt in the cost of equity approach is part of the tariff base, as the debt is not separately accounted for in the rate base. 12 Proportion of debt and equity in the rate base, considered optimum or desirable (see fn. 11). 13 The assessment being that the allowed 16 per cent returns factor in the foreign exchange risk. If the risk is to be borne by the domestic consumers, the returns allowed should be lowered by an amount appropriate for hedging this risk. 14 An equilibrium asset-pricing model used extensively in the financial markets as well as by the regulators of private entities to estimate the required rate of return on equity by the market. According to this model, the expected return on equity of a company depends upon the risk-free rate (see fn. 16), the risk premium (see fn. 17), and the beta of the equity (see fn. 15). 15 The beta of equity indicates the extent to which the returns on the equity are sensitive to returns in the market as a whole (or the economy). The asset betas similarly indicate the sensitivity of asset returns’ to the market as whole. The asset betas are different from equity betas, as the equity beta is a linear increasing function of leverage (amount of debt) for a given asset beta. The asset betas are estimated to remove the effect of leverage from the observed equity betas and to find appropriate equity beta for a given leverage in mind. 16 Risk-free rate is the rate of return on the default-free (central government) bonds in the market. 17 Risk premium in the context of CAPM means the excess over risk-free rate expected by the market on a very highly diversified portfolio. The intuition behind the risk-premium is that investors would demand some returns over and above the risk-free rate on even a highly diversified portfolio, as they still remain exposed to risks.
242 India Infrastructure Report 2002 using excess returns earned on the Bombay Stock Exchange (BSE) Sensex with systematic investment plan over last 20 years. Using these proxies, the CAS report suggested 20– 1 per cent return on equity based on debt–equity ratio of 1:1. The estimates of the cost of equity based on CAPM be reviewed every five years, even though the cost of equity may be changed every year to adjust for changing leverage, in a manner that such reviews do not leave financial ratios worse than what would qualify for investment grade debt. • The risks in unregulated businesses taken up by entities should be ignored for cost of equity or capital estimations. In an early disclaimer in the report, CAS pointed out that though the study envisaged an assessment of the impact on the entities’ financial performance of the past five years of the application of the cost of capital approach, the data for it was not made available. Rate Base: Besides these recommendations and in the process of study, CAS looked into a closely related item, that is rate base itself, on which the cost of capital or equity needs to be applied for tariff determination. Probably guided by the need to evolve a conceptually sound framework for the other recommendations made by them and without critiquing the existing methodology explicitly, they suggested the following as regards the methodology for rate base determination: • The rate base for applying cost of equity should be based on ‘adjusted net worth’ (that is, net assets – longterm debt + statutory assets + capital works-in-progress18) and the adjusted net worth across projects/plants should add up to the balance sheet of the entity related to regulated business. The need for plant/project-wise balance sheet was highlighted to remove distortions during the implementation phase. • The implications of constant equity assumption in the liability side approach to tariff fixation were pointed out on p. 39 of the report: ‘The investor should return returns from capital that is deployed in the business. If post-tax cashflows are reinvested into the business, the investors would continue to earn returns on such funds.’ The cashflows referred to above are depreciation recovered as tariff and available for application as equity in some other projects and earning twice (or, multiple times) over.19 18
Capital works-in-progress is a balance sheet item on the asset side, representing the investments made in fixed assets, which are yet to be commissioned. 19 The divergence referred to in fns. 3 and 4 arises from the fact that for depreciable assets, the asset values and hence appropriate rate base keep going down by the amount of depreciation with the
• The returns on capital works-in–progress should be allowed and be capitalized till the project becomes operational, as the method of estimating the cost of equity or capital implies estimating the returns that investors would expect without any lag or gestation period.20 • The rate base of a tariff object (plant/project) should be allocated the same leverage (on the basis of an overall leverage), as opposed to varying leverage implied in existing tariffs.21 • Though receivables (if beyond the control of the entity) should theoretically be included in rate base, the potential moral hazard and cross-subsidization warrants recovery of returns through delayed payment charge from specific debtors. Between the two issues, rate of return and rate base, the CAS report dealt with the existing anomalies pointed out earlier. The suggested capitalization of the return on equity till completion removes the bias against long gestation period projects even though it removes incentives for early completion of any project, irrespective of technology. A slight modification of the same idea could also have created incentives for early completion without creating such an obvious bias. The asset based approach suggested for determination of tariffs neutralizes the excess returns on equity possible at present because of the constant equity assumption as well as by taking in longest possible maturity debt. The recommendations would also have corrected the anomaly of variations in tariffs based on different imputed leverages across plants and projects, leading to crosssubsidization. On the other hand, it would have linked the return on equity with the financial risks assumed by the respective entities. These recommendations occupied an important place in the discussions later and in the eventual CERC order, notwithstanding non-implementation of most of these recommendations as brought out later.
IAS Discussion Report on Depreciation Norms The ICRA Advisory Services’ report on depreciation norms focused on the depreciation method, rate and asset base used in determining allowable depreciation. The report argued for continued use of the ‘straight line method’ while suggesting that the rates prescribed are not linked with the fair life of assets—the fair life of assets as notified under passage of time. Unless the debt is repaid exactly by the same amount, the liability side approach of determining the rate base by keeping equity constant inflates the rate base. The reverse can also happen if the debt is paid at a faster rate, compared to depreciation. 20 This would have removed the bias against a long gestation period, as the investors would get higher returns later through the capitalization of interim returns in the rate base. 21 See fn. 7 for an explanation.
Framework for the Energy Sector the ES Act, 1948 being longer than the rates revised in 1994. Conceding that the notified lives of assets in India might be actually lower due to operating conditions, it pointed out that the internationally prescribed lives are in fact higher than the ones notified in India. On account of intergenerational equity (that is, consumers of the services of the same asset be charged the same depreciation) and incentive problems (early recovery of depreciation reduces the interest of operator in the asset22), it recommended a lowering of depreciation rates. As far as the asset base on which the depreciation was to be determined, the report suggested a gradual move towards Optimized Depreciation Replacement Cost (ODRC) asset-base. The term optimized refers to adjustments made on account of over-design, over-capacity, and redundancy of assets, whose replacement costs are being used for deciding the asset base. The implementation of ODRC methodology required having a detailed asset register and periodic reassessment of replacement costs. The report suggested that ODRC methodology will avoid tariff shocks associated with historical cost based approach experienced as and when assets are replaced. The implications of these recommendations are straightforward. The depreciation rates prevalent were higher than warranted, if only the asset life is brought into consideration. However, if the replacement cost are rising in a sector facing supply-constraints, the depreciation and therefore tariffs related to some of the old assets might be lower than warranted. Even though the report suggested lower depreciation rates (as the fair lives are higher), it also suggested a gradual movement to replacement cost of assets (which would have brought tariffs closer to LRMC based tariffs). The two recommendations of the report have exactly opposite effects on the existing tariff,23 even though independently both were conceptually sound. The report, however, did not relate these two issues explicitly or even implicitly, as on p. 16 of the report, the IAS argued: ‘We agree that supply of energy is inadequate to meet demand and that the market would probably clear at a higher price. However, investment flows would be driven by magnitude of the return on the capital rather than source of return of capital. The issue of intergenerational equity is important.’ The above observation was based on a non-existent asset side tariff determination approach, which was independently suggested in the CAS study. To quote the IAS study: ‘The depreciation profile can be however rendered irrelevant, i.e., constant net present value of total revenues, 22
Given the anomaly of return on equity assumed to be constant, this was/is not true as pointed out earlier. 23 Assuming that replacement costs have been rising continuously, which has not been the case of late.
243
if the return on capital is based on the net asset base rather than on a fixed base (equity subscribed) as at present.’ Clearly, the IAS understood that lowering of depreciation affects the economic returns (return on equity) negatively due to the anomaly existing despite claims to the contrary in the previous paragraph. On the ODRC recommendation, the IAS study was surprisingly silent on its financial impact on the entity. The excess depreciation due to rising replacement costs will go to replace assets, but what would be the incentive for the entity in operating such an asset? On the other hand, if any equity were attributed in the assets so created, that would amount to windfall gains for the shareholders/entity. Some of the issues discussed here were important in resolving the anomalies pointed out earlier. In both the CAS and IAS studies, the recommendations followed from the framework that these agencies had used. A smaller subset of core recommendations consistent with each other would have been useful to evolve the bases for tariff norms. However, both the studies to some extent, and the CERC later on, treated some of the recommendations independent of others leaving most of the anomalies much as they were.
DCL and WAPCOS Reports on O&M Costs The report prepared by DCL on O&M costs related to thermal generation was based on analysis of data of actual O&M costs of 11 domestic stations. The report found that the current norm of O&M costs at 2.5 per cent of capital costs was insufficient for steam plants (coal-based) and suggested increasing it to 2.8 per cent. The report pointed out that though in the case of 9 steam plants in the US, for which they had data, actual O&M costs were only 1.6 per cent of capital costs, and the work culture and other differences may make the US or international benchmarks inappropriate.24 The report could not suggest any O&M norms linked with capital costs in the case of gas or dieselbased plants. In the case of steam plants, the study found a relationship between O&M costs and escalated capital costs (project cost escalated based on actual escalation trends). In the case of gas and diesel based plants, the maintenance schedules are not annual and therefore cause swings in O&M costs. Further critiquing some aspect of current practice of linking O&M costs with capital costs, the report suggested linking O&M costs with the cost of generation, the arguments being that: (a) higher generation costs are driven by higher fuel consumption, a reflection of higher availability due to efficient operation and 24 The O&M costs, if any, are expected to be lower in India, given far lower wage norms. Very poor work norms and overmanning allowed to go on uncorrected have gone far as to affect investment decisions! [Ed.]
244 India Infrastructure Report 2002 maintenance, and (b) data shows relationship between O&M costs and cost of generation across domestic and US plants. This would also automatically take care of the need for deciding the basis for escalation of O&M cost with the passage of time. The report by WAPCOS analysed 29 medium to large hydro-power stations operated by SEBs and other licensees in India for O&M costs. The study found that the average of these costs was about 1.47 per cent of escalated capital costs, which was close to the currently allowed 1.5 per cent of capital costs in the first year. It suggested retaining the current norms and pointed that the central utility’s (National Hydro-Power Corporation Ltd, NHPC) costs were higher due to overstaffing. It however, recommended that the increased O&M costs be allowed to the extent additional expenses are incurred on security of certain stations by the NHPC. The WAPCOS report also suggested that escalation in O&M costs be linked more with the Consumer Price Index than the Wholesale Price Index. The WAPCOS report also dealt with the appropriate availability norms for different types of stations, the need for review of design energy on completion, and the need for determining the price of reactive power and looked into norms on auxiliary consumption and transformation losses based on actual data.
CERC’S ORDER
ON
TARIFF NORMS
Besides the reports from the consultants on specific issues, the CERC also asked for views of various stakeholders on the issues identified in the consultation paper as well as on the consultants’ reports. Based on these inputs, the CERC came out with a comprehensive review of tariff norms. The key decisions taken and the rationale offered in the order document are given below.
General Issues Tariff Entity: Continuation of station-wise generation tariffs and line-wise aggregated regional tariff was approved on account of its widespread acceptance and the need to avoid micro-management by the regulator. Periodicity of Tariff Review: The CERC set the date for the next review of tariffs after three years, without committing itself to any periodicity thereafter. Changes During Tariff Period and Retrospective Adjustments: On account of capital costs, the CERC ordered that any approved capital expenditure incurred during the tariff period shall have to wait till the next tariff revision unless it is more than 20 per cent of the approved capital costs. The argument behind the ruling is that these revisions create ‘tariff shocks’ for the beneficiaries and that there
should be as much certainty as feasible of tariffs within the tariff period. The fuel price adjustments were considered inevitable by the CERC while taxation, O&M costs, and foreign exchange variations were dealt with separately.
Return on Investment and Rate Base These two issues occupied the maximum space in the CERC order document. As already pointed out, there are several anomalies in tariff norms related to these two issues. The CAS report had already explicitly or implicitly unearthed these. The CERC order affected no change in the tariff norms on both the issues, despite agreeing in principle on several aspects of the CAS report! On the rate of return issue, the CAS report suggested linking return on equity with leverage and eventually moving over to cost of capital approach without differentiation across entities with respect to business risks. The cost of equity estimated by CAS using CAPM indicated higher than 16 per cent return—an expected return of about 20–1 per cent per annum. On the rate base, it suggested moving to the asset side approach for determining the asset base and suggested inclusion of capital works-in-progress to remove bias against certain type of projects. Rate of Return: The CERC retained 16 per cent return on equity for the next tariff period despite agreeing with the approach of the CAS study. The reasons for retaining the existing rate are cited as: • Use of a limited sample of 30 stocks of the BSE Sensex as a proxy for the determination of risks as well as market risk-premium; • Need for better assessment of risks associated with the power sector as the proxies used by CAS (listed power companies and oil-refining companies) were challenged by the central sector utilities as providing an inadequate comparison and downward bias in the estimation of risks; and • Non-completion of impact assessment study by CAS due to unavailability of data and imperfect risk assessment may make any change in return unfair to either central sector utilities or their beneficiaries/end consumers. Besides these, the upward impact of raising return on equity on the beneficiaries also seems to have played on the CERC’s mind. To quote the CERC, ‘There had been a great deal of resistance from the beneficiaries to this increase in ROE (return on equity) to 16 per cent which prior to 1991 was only 10 per cent’. The issue here is whether 16 per cent is an appropriate return or not. The fact that it was 10 per cent for public sector entities in an era when private sector participation, commercialization, as well as independent regulations were non-existent should not have guided the CERC’s decision.
Framework for the Energy Sector On the issue of return on equity being linked to leverage, the CERC order shows confusion as well as probably regulatory trade-off. On this issue, at one instance it says: ‘We have not been convinced on the practicability and advisability of the adjustment for individual company’s debt/equity ratio, though the principles (outlined by CAS in its report) are theoretically well founded.’ Later in the report, it says: ‘We do not recognise in principle the sector specific risk and adjustment for debt/equity.’ Finally in the findings, it says: ‘The model suggested by CAS lacks stability in as much as it is required to be linked to the actual debt equity of each company periodically. These companies do not have any manoeuverability to influence debt/equity mix.’ Further, ‘A constant revision of risk premium based on the behaviour of the equity and debt market would result in regulatory uncertainties which would be a great disincentive for investments in the sector, though the adjustment is contemplated every five years.’ It is unclear as to whether removal of uncertainty of all kinds in the tariffs over the tariff period is desirable. Just as it agreed that the fuel cost adjustment is inevitable, any unhedged exposure to exogenous risk if not passed through will either result in windfall or loss to the regulated entity. Such exposures if not passed through warrant an adjustment in returns to compensate for risks taken. Rate Base: On the rate base issue, the CERC order while retaining the current liability side approach, in which equity is assumed constant, came out with even more clumsy reasoning. The reasons highlighted were: • The balance sheets of tariff entities (station-wise/ line-wise) are not available; • The balance sheets, even if made, would reflect the actual position whereas tariffs are on normative basis; • The time lag in availability of balance sheets would delay the finalization of tariffs; • The need for an independent auditor verified and approved cost-based trail of asset, debt, and equity related to the tariff entity. These reasons are valid only if the CAS recommendation is taken literally. The essence of the recommendations only required approved cost based net assets (which is nothing but approved capital cost less allowed depreciation), pro-rata debt applicable to net assets and prorata statutory investments. It is clear that the CERC noted the anomaly caused by the present regime as it says: ‘In administrative pricing in all regulated industries return is allowed on net assets only. It should be so under Sec 43(2) of ES Act 1948 as well as normal interpretation read with schedule VI. Hence it would be only appropriate to provide for interest and return on fixed net assets.’ Further, ‘In this regard we find that in case of IPPs the
245
equity is kept constant and return is continued to be provided till the life of the plant. Though no justification on record is available for the treatment adopted by the government as at present, a legitimate assumption that depreciation is used to repay loan is appropriate though this is not borne out by any notification. After the loan is repaid, the balance of net fixed asset is represented completely by equity. There is no justification of providing return on equity not represented by assets . . . This may be viewed as unwarranted and unfair but this is based on the notifications dated 30 March 1992 and 16 December 1997.’ In the end, the reasons behind the CERC’s reluctance turn out to be not to upset the existing state of things due to jurisdictional anomalies (the IPPs facing the same regime would be unaffected by the changes, whereas central sector utilities would be affected negatively) and the government policy, despite obvious conflict with the regulatory objective of providing only fair return to the regulated entity. This is evident from, ‘From the study of policy of Government of India with regard to IPPs it is evident that there was a conscious decision to offer incentives to investors so that they can continue to sustain their plants and operate the services. In case they discontinue they would lose the return on the hypothetical capital which may not be the actual capital deployed. This policy appears to have found its way into the pricing for public sector utilities . . . We consider this as a deliberate policy of the Government though not formally communicated under section 38 of ERC Act.’ Finally, the CERC conceded that, ‘Public sector undertakings so long as they stand committed to the power sector and do not diversify to other sectors without prior approval and proper justification should be entitled to this incentive of a return on the equity employed based on liability side approach rather than strictly administered pricing approach based on the asset values.’ In a related issue of allowing returns on capital-worksin-progress, the CERC decided against it, as it was not in line with accepted accounting principles and procedures. Further, ‘This is the universally accepted accounting practice (not capitalizing opportunity costs of equity) and the departure from this that too only for the purpose of calculation of tariff may not be correct.’ Despite placing emphasis on the importance of regulatory rationale and conceding anomalies, on the two issues of rate of return and rate base, the CERC order is at best against its objective of providing fair return to the regulated entity. At its worst, it is inconsistent. In the case of rate base, the justification of maintaining the status quo is incentive for the central sector utilities at par with IPPs, while in the case of rate of return it is to protect beneficiaries.
246 India Infrastructure Report 2002 It seems that in the opinion of the CERC, two or multiple anomalies are worth preserving in case they cancel out each other’s effect. Even on jurisdiction, the rationale offered is weak as would be evident later from some other aspects of the order. Besides incorrectness of using accounting principles and practices as a criterion to accept or reject a proposal on an essentially economic matter, the CERC has been inconsistent on this issue as well. While allowing increased asset life for tariff determination (thereby lowering depreciation rates), it allowed the regulated entities to follow the depreciation as notified under section 43A(2) of ES Act 1948 for their books of account.
Depreciation Norms Despite not explicitly stating the effect of depreciation on economic returns in the relevant part of the order the CERC was aware of the same, as can be noted from the observations made by the CERC in its order on the discussion related to rate base, ‘We would like to sustain the underlying incentive feature behind the existing policy (of allowing returns on constant equity and current methodology of determining rate base) and would not like to upset the same in view of need for promotion of investments in this sector. All the same the acceleration of depreciation, needs proper justification though augmentation of cash in flow has an equal and opposite cash out flow on beneficiaries. We shall take note of the same with depreciation.’ On depreciation, the CERC accepted the IAS recommendation of use of straight-line method and linking depreciation rates with useful asset life. This would mean lower depreciation rates than were allowed and notified by the government. It also advised the central government to revise the depreciation rates under the ES Act, 1948 as otherwise the IPPs and other players outside CERC jurisdiction would be unfairly benefited. The anomaly in the situation is evident as the CERC used the same argument to maintain the rate base methodology, while in case of depreciation it decided to take on the jurisdictional anomalies albeit tangentially. The double impact of anomalies related to fictitious/hypothetical asset base and accelerated depreciation on beneficiaries must have played on its mind. The irony in all this is that the regulator instead of laying out principles for tariff determination is seen as performing a balancing act between the regulated entity and beneficiaries, and that too in an oblique manner. As could be expected from its inclinations of not accepting LRMC as the basis for depreciation/tariffs in the consultation paper itself, the replacement cost linked methodology suggested by the IAS was not accepted. It would, however, have been better for the CERC not to interpret depreciation (as outlined in the IAS report)
narrowly by saying that, ‘We are not convinced about the ODRC method since it has already been concluded that primarily depreciation is not a process for collecting money for replacement of asset but is a process for repayment of the capital instalments.’ While this is correct, the idea behind ODRC in the IAS report was to have smoothening of tariffs and more efficient signalling in the context of increasing replacement costs. In fact later, the CERC itself allowed a development surcharge to protect the cash flows of utilities for investments. For the investments made by collection of development surcharge, there would be no return on these funds used as equity in power projects. The same effect could have been achieved in case the ODRC methodology was used along with change in rate base to the asset side approach.
O&M Costs In the case of O&M costs for thermal generation, the CERC did not agree to the DCL report of linking O&M costs with generation costs as it will allow different O&M charges to the plants based on their location (depending on whether they are located at pit-head or at load centre), and as it did not find favour with any party. It also rejected other findings on account of small sample analysis carried out by DCL. The CERC decided to move to O&M charges based on actual O&M charges incurred in the last 5 years (after removing abnormal expenses), as it found that there were measurement problems in the current methodology of using current capital costs in the case of assets of old vintage. The current capital cost of an asset/ plant is defined as the capital cost, in the year of fixation of tariff, for a similar project. This invariably would bring in subjectivity and that is the main reason cited by the CERC for moving away from current capital costs. The CERC also modified the escalation norm of 10 per cent per annum in O&M costs to an escalation formula linked with CPI (Consumer Price Index) and the non-employee cost related components of WPI (Wholesale Price Index), as worked out in an internal paper (discussion paper dated 2 June 2000) by the CERC. Having decided on moving away from capital cost based O&M cost norms, the CERC prescribed actual O&M costs based norms for hydro-generation plants as well. In the case of transmission, the CERC modified it somewhat as the increase in O&M costs at regional level (unlike plant level O&M costs in generation) would be also due to on-going network expansion and the actual data would tend to understate the O&M costs, if the network continues to expand during the tariff period. The CERC accepted 30:70 as the ratio for allocating O&M costs between sub-station and lines, as recommended by the expert committee appointed by CERC. During the tariff period, the allowable costs
Framework for the Energy Sector would be linked with the number of bays and line-kilometres and would be escalated on the same basis as in generation. The decision of the CERC to move away from normative O&M costs (capital cost based regime is normative; whether it is liberal or constrictive is another issue) to actual experience based (except for abnormal expenses) O&M cost recovery leaves no incentive for the regulated utilities to improve operating efficiencies. This is particularly damaging in the light of O&M cost analysis carried out by WAPCOS in its report. The 29 hydro-generation plants within India on an average have lower O&M costs that the sole regulated entity (NHPC). Similarly in the DCL report, the operating costs seen elsewhere are low compared to the regulated entities. All the regulated entities felt that the existing norms were too tight and the actual costs incurred are higher, as is reflected in the order document. The move to an extreme actual based O&M norms in such a case is unfair to their beneficiaries and therefore, end consumers. The reason of measurement problem associated with current capital costs is too flimsy a reason. In the case of NHPC, the O&M costs were worked out by finding the appropriate capital cost linked O&M costs in the first year and were subsequently escalated as would be the case for any new project under the previous regime with respect to O&M costs. The same approach, instead of being critiqued by the CERC, could have been chosen to remove at least some of the problem associated with measurement of capital costs. Alternatively, any normative rather than actual costs based approach would have left the incentive for management of O&M costs with the entity on one hand and reduced the burden to collect and analyse information related to actual O&M expenses on the CERC on the other, consistent with its desire not to microregulate as expressed elsewhere.
Operating Norms and Variable Cost in Thermal Generation On this issue, the CERC could do nothing and was in a sense forced to maintain status quo except for PLF/ availability. In order to review the existing operating norms as notified by the government, the CERC appointed a committee headed by a chief engineer of the Central Electricity Authority (CEA) with representatives of affected utilities. The findings of the committee turned out be that of its chairman, and all others did not accept the same. In the absence of data, the CERC maintained status quo and ordered submission of data on heat rate, fuel consumption, secondary fuel oil consumption, auxiliary energy consumption, availability, PLF etc. at quarterly intervals, based on which some norms can be arrived at. The stonewalling and lack of data forced the issue, as it were, on the CERC.
247
Foreign Exchange Variation In line with the CAS recommendation, the CERC agreed with protection against foreign exchange variation. The CERC ordered that the treatment of capitalizing the variation25 followed by the National Thermal Power Corporation (NTPC) is consistent with accounting standard and hence be followed by the NHPC, which was billing it as it was incurred. In addition to being inconsistent with accounting standards, the CERC was concerned that practice followed by NHPC results in tariff shocks. What the CERC failed to see in this instance is that given the liability side approach followed, the NTPC practice and, therefore, its order biases it in favour of regulated entities by increasing the return on equity, even though it creates pressure on their cash flows in the year in which variation takes place.
Treatment of Corporate Tax The CERC accepted the current practice followed by the NTPC to pass through the actual tax despite the fluctuations/unpredictability it may cause to billing, as grossing up the tariff amounts to double taxation. It also accepted NTPC’s proposal to have an escrow account for each beneficiary in which monthly billing based on estimated tax liability will be deposited and the account will be drawn for tax payments based on the auditor’s certificate. The refunds, if any, shall also be deposited in the same account. This would reduce lumpiness in billings. The treatment in case of IPPs is based on grossed up billing and results in double taxation.
Incentives The CERC tightened the norms for incentives, as well as lowered in some cases incentive rates for the utilities. The base PLF for the NTPC was raised from 68.49 per cent to 77 per cent and for the Power Grid Corporation of India Limited (PGCIL) base availability was raised from 95 per cent to 98 per cent. Similarly, for Neyvali Lignite 25 When the foreign exchange denominated debt is raised for constructing any fixed asset, such as generation plant or transmission lines, the change in foreign exchange rate can increase the interest as well as principal repayment in rupees. Since the power sector players are allowed to recover this variation from beneficiaries, they can either raise a bill for increased rupee liability as and when it arises or can increase the rate base to the extent such increase has taken place. Both the approaches are equivalent in case the rate base is determined appropriately, except that in case of the former the impact on tariff is immediate and lumpy, whereas in the latter case it is levelled. The approaches are not equivalent if there are distortions in the determination of rate base of the kind pointed out in fns. 2, 3 and 18. The CERC, as is obvious, went in favour of the NTPC’s practice without realizing the full implications and without taking care that these anomalies do not come into play at least on this issue.
248 India Infrastructure Report 2002 Corporation (NLC) the base PLF was raised from 70.2 per cent to 72 per cent. The CERC allowed 50 per cent of fixed costs as incentive above base PLF subject to a maximum of 21.5 paise/kWh. The PGCIL was allowed 1 per cent extra return on equity (ROE) for every 0.5 per cent higher availability, capped at 4 per cent extra return on equity at availability of 99.75 per cent. The incentive rates for IPPs outside the CERC’s jurisdiction will now be relatively much higher with the revision, as their rates remain unaffected by the CERC order.
Development Surcharge The impact of reduction in depreciation rates by the CERC and unchanged return on equity would have resulted in lower cash flow availability to utilities for investments. This made all of them oppose the recommendations in the CAS report arguing for higher ROE (as was done by PGCIL and NTPC) or pray for a surcharge (NHPC). After looking into the legal feasibility of having such a surcharge, it allowed a 5 per cent surcharge to generation utilities and 10 per cent to the transmission utility to be billed on fixed charges in respect of operations at the regional level (and not within a state). While stating that it is not its intention to provide for all funds needed for capacity expansion, the CERC ended up allowing the surcharge, drawing analogy from similar provision in the tariff of licensees. The net impact of the development surcharge was exactly the opposite of change in the depreciation rate. The only saving grace was that the surcharge could be only applied for investments as equity without any corresponding returns. Though the design of the CERC (to have only one-third of total equity drawn from the surcharge reserves in any project) was intelligent to prevent loss of interest in the projects financed through surcharge, it did not consider the possible cross-subsidization built in any such arrangement, as such projects would be cheaper in terms of tariffs to their beneficiaries.
ON PROCESS
AND
OUTCOME
The CERC as an independent regulator undertook the exercise of reviewing bulk power tariffs for the first time after running into jurisdictional challenge in the aftermath of its ABT order. The process followed by the CERC was very open as is evident from the order document and the availability of all the material in the public domain. It also took expert opinion wherever required, as is evident for the appointment of consultants and committees. The regulated entities, on the other hand, seem to have taken a rather confrontational posture along this process on
various issues including recommendations of consultants and committees in addition to their reluctance to provide data and information. It could have been more appropriate if their suggestions and responses were more facilitative to evolve a framework for the underlying issues rather than challenging the methodologies and findings of others. This was most exemplified in the issues related to variable costs and operating norms of thermal generation. The problems of game-theoretic behaviour, particularly induced by information asymmetry between regulator and the regulated entities, are however well known, and the CERC as a regulator has to evolve ways to counter these problems. What was surprising, however, in the process of review was the low level of involvement and arguments put forth by the beneficiaries, that is, by the SEBs. Most of the anomalies affect them negatively and hence one would have expected them to raise issues accordingly. This was singularly missing in their responses to the CERC. In terms of outcome in the form of order, the CERC at times supported sound arguments (evident most in its comment on the CAS recommendations) but desisted from implementing them on grounds ranging from government policy and consistency with IPPs (asset based approach to rate base determination), accounting practices and standards (capitalization of equity component of capitalworks-in-progress for tariff purposes), measurement problems (current capital cost based O&M cost norms), interest of beneficiaries (in changing return on equity), investment requirement of regulated entities (development surcharge), practicability, advisability, acceptability, etc. Most of these are not valid arguments in so far as its own regulatory objective related to tariff determination is to provide fair return to the regulated entities on one hand and be fair to and across consumers/beneficiaries. In the process, not only has it failed to rectify anomalies pointed out earlier and identified in the course of the review, but also some of its decisions can worsen the impact (foreign exchange variation treatment). Another regulatory objective which was compromised was improvement of operating efficiency by allowing O&M cost recovery on the basis of actual O&M costs in the previous tariff period. Worse still, the arguments articulated in its order were inconsistent on accounting standard (depreciation vs. capitalization of equity), consistency with IPPs (asset based approach vs. incentives and depreciation), suggesting changes to government on IPP tariffs (depreciation vs. incentive), and fairness to the beneficiaries (depreciation vs. development surcharge). In the end, despite openness in the process the order looks more like a balancing act and we will have to wait at least till the next tariff review for a sound framework for bulk power tariff regulations to emerge.
Framework for the Energy Sector
249
9.2 INTER-STATE TRANSMISSION OF ELECTRICITY: LESSONS FROM THE NORTHERN REGIONAL GRID COLLAPSE Puneet Chitkara • Rajiv Shekhar • Prem K. Kalra For a modern economy, the electric power system is an axiomatic need. The woes of the power sector are generally seen as arising in generation or distribution. However, the collapse26 of the northern grid on 2 January 2001 is ominous. The transmission sector as well has severe problems. The PGCIL owns the high voltage transmission network in India. Broadly, the functions of PGCIL are as follows: • It transfers bulk power from the Central sector generating stations to the beneficiary states. • It transfers interstate power in a region both on short-term basis and on long-term basis. This is usually inter-SEB transfer and in the future is likely to be transfers from an IPP in the jurisdiction of one state to a distribution utility/SEB of another state. • It transfers inter-regional power from an excess region to a deficient region, usually through HVDC (high voltage direct current) lines. • It provides reactive support for maintaining a normal voltage profile in the high voltage network. In effect, PGCIL is both: (i) the system operator, operating what is commonly known as the Regional Load Despatch Centres (RLDCs) and (ii) the transmission asset owner (TO). The RLDCs functions cover all the operational activities required to keep the system balanced and within safe operating limits. The TO’s functions relate to the maintenance and long-term development and investment in the transmission system. The CEA’s investigation of the 2 January collapse revealed several disturbing facets of the transmission sector. Consequently, important issues from an organizational, and operational framework that have a bearing on the efficiency of the transmission sector will be discussed in this section. More specifically, this section will focus on the effectiveness of PGCIL in providing a trouble-free and vibrant integrated grid system. As on 1 January 2001, the Northern Regional Power System (hereinafter, the Region) had an installed capacity of 27,042 MW. The Region has a hydro-thermal mix of 31:69. Hydro and thermal capacities account for about 96 per cent of the total installed capacity of 27,042 MW. The Region has major thermal power stations located at the pit-heads of coal mines at Singrauli, Rihand, Obra, and Anpara. Major hydroelectric power stations are located in 26
The usual reasons for a grid collapse are given in Box 9.2.1.
the Himalayas such as Bhakra, Dehar, Pong, Chamera, Baira-Siul, Salal, Uri, and the hydro-power stations on Yamuna River in Uttar Pradesh (UP). The large coal-using pit-head thermal power stations are located in the extreme south-eastern part of the Region. Therefore, there is a large flow of power from the south-eastern part to the central and western parts of the grid throughout the year. During the winter months, when the water flows dwindle to their annual minimum value, many hydro stations are shut down during the night (off peak) hours. To handle the bulk transmission of power, a point-to-point high voltage direct current link, viz. the ± 500 kV HVDC Rihand–Dadri bipole with a capacity of 1500 MW has been established which operates in parallel with an extensive 400 kV AC transmission system and the underlying 220 kV network.
THE COLLAPSE Following widespread rains on 1 January 2001, the demand in the Northern Region had reduced. A few thermal units in Punjab and Haryana had been closed due to low demand. After 10 p.m. of 1 January 2001, the frequency started rising and backing down had to be resorted to at a number of thermal power stations. The salient events that led to the collapse of the Northern Region Grid (NRG) are brought out below27: • Pole-2 of the Rihand–Dadri HVDC bipole line, a crucial link in the NRG, had been non-functional for a while, resulting in increased loading of parallel AC transmission system in the east–west corridor of the NRG. Because of the prevailing high frequency in the grid,28 Singrauli and Rihand were asked to back down generation by 220 MW and 100 MW respectively. • Due to flashovers,29 pole 1 of the HVDC link was operated at 67 per cent of its rated capacity, and the 400 kV Obra–Panki line of UPPCL (Uttar Pradesh Power Corporation Limited) tripped, resulting in loss of 600 MW of transmission capacity on the east–west corridor. The NRLDC (Northern Regional Load Despatch Centre) asked Singrauli and Rihand STPS (Super Thermal Power Station) to back down by 80 MW each. 27 28
CEA report on NRG disturbance. See www.powermin.nic.in. Because of inclement weather and rains, there was a substantial reduction in the agriculture load. Agriculture load in the NRG constitutes about 35 per cent of total power demand. 29 An electric discharge around or over the surface of an insulator.
250 India Infrastructure Report 2002 Box 9.2.1 Why Does the Grid Collapse? The distinguishing feature of electricity transport is that electricity cannot be stored: the supply of electricity should be matched by its demand. If supply tends to exceed demand, the generating stations are asked to back down. Conversely, when demand tends to exceed supply, load has to be shed. Three specific reasons can cause a grid to trip: • Overdrawal of power by states above the day-to-day directive of the RLDCs. When several states turn errant the frequency dips. If frequency falls below 47.5 Hz. (the desirable frequency is 50 Hz.), the grid trips. Cascading effects are possible, if the grid operator does not quickly isolate portions of the grid. • Refusal of generating stations to back down. RLDCs determine demand and advise plants to generate power accordingly. If generation is more than demand the grid frequency rises. If it rises above a critical value the grid trips. High frequency is experienced mainly during the monsoon months when the agricultural and air-conditioning and fans’ load crash due to widespread rains and fall in temperature. Load ‘collapse’ to the extent of 4000 MW due to rains, in the Northern Regional Grid, is common during the monsoon season. During the winter rains, similar load ‘collapse’ is observed. • Faulty equipment/poor maintenance leading to non-functional/under-performing transmission lines can also affect grid frequency. Grid frequency increases in generation-surplus regions and decreases in generation-deficit regions.
• Tripping of the 400 kV Panki–Muradnagar and Panki– Kanpur30 and 400 kV Unnao–Agra lines further reduced the transmission capacity below 600 MW. However, Singrauli and Rihand were, surprisingly, again asked to back down by a mere 260 MW, which meant that the remaining lines were loaded beyond their safe limits. • The NRLDC, realizing that their backing down instructions were inadequate, asked Singrauli and Anpara thermal stations to further back down their generation. However, the records at NRLDC and Anpara did not tally with respect to the time when the instructions were given and quantum of back down asked for and complied with. In fact, the entries in NRLDC’s record were overwritten! • Subsequently, tripping of the Kanpur–Agra, Kanpur– Ballabgarh, and Lucknow–Moradabad lines segregated the eastern and western parts of the NRG (Northern Regional Grid). • The western part of the regional grid was (before the separation) importing more than 2000 MW of power from the eastern part. Separation from the eastern part therefore resulted in a huge power deficit, and the frequency went down leading to the collapse of the western part of the grid. On the other hand, with excess generation, the
frequency of the Singrauli–Rihand complex rose above 52.5 Hz, ultimately culminating in the collapse of the eastern part of the grid. This led to a near-total collapse of the regional grid and a loss of generation of about 15,500 MW. Can the grid collapse be ignored as a rare isolated incident? Probably not, as the major grid accidents in year 2000 bear out (see Box 9.2.2). Important governance issues that have a bearing on the efficacy of the Inter-state Transmission System (ISTS) emerge from the NRG collapse: • Questionable reliability of the transmission system to withstand grid disturbances of the scale of 1–2 January, again. • The inability of the NRLDC operators to correctly assess the extent of required generation back down and to act swiftly to contain grid disturbances. • Possible lack of maintenance and monitoring of crucial equipments connected to the grid. • Inadequate acquisition and communication of real time data from grid equipments to RLDC.
Box 9.2.2 Other Grid Accidents in 2000 January: Punjab, Haryana, Delhi, Uttar Pradesh, and Rajasthan reel under a power blackout. April: West Bengal copes with acute power shortage as the eastern grid almost collapses due to low levels of generation and high levels of consumption. May: The southern grid faces disturbance; power cuts and blackouts unsettle Tamil Nadu and Karnataka. November: Maharashtra, Madhya Pradesh, and Gujarat are forced to cut down power consumption as the western grid falls short of the demand by a whopping 4500 MW. Low voltage problem looms over the entire region. 30
Due to failure of line breakers.
Framework for the Energy Sector • Quality of ancillary services for maintaining power quality, considering the fact that even off-peak conditions are characterized by high frequency and high voltages. • Reluctance of thermal power stations to back down generation.
RELIABLE TRANSMISSION Reliability has a special meaning in transmission. To be reliable, a transmission grid must have both adequacy and security. Adequacy implies that sufficient generation and transmission resources are available to meet the projected power needs at all times. A grid is secure when it can remain intact even after planned and unplanned outages or other equipment failures occur. Is the ISTS reliable? • The planning criterion of ISTS, as stated in the Indian Electricity Grid Code (IEGC), requires that the transmission system be designed in consonance with the N-2 reliability criterion, which requires that the system should be capable of operating normally31 even if any two lines in the system are not functional. However, events leading to the NRG collapse suggest the contrary. The grid was operating under normal steady state even with the outage of pole 2 of the Rihand–Dadri HVDC link. However, insulator failure on the 440 kV Obra–Panki line overloaded the parallel AC lines, triggering the sequence of events that ultimately led to the NRG collapse. • During the NRG restoration process, Dadri could not be supplied start up power from the Bhakra Beas Management Board (BBMB) system because of inadequate reactors on the Panipat–Dadri line and malfunctioning of circuit breakers at the Panipat sub-station. Inadequacy, especially in terms of peak power generation, is not a rare situation and has contributed to grid ‘accidents’, as shown in Box 9.2.2. The NRG collapse, illustrated by the above-stated examples, brought into sharp focus the reliability of ISTS. What are the possible remedies to increase the reliability of the ISTS? Increasing reliability entails: • Transmission capacity expansion to meet the N-2 reliability criterion. • Providing the means to suppress instability in the ISTS that may occur due to equipment/line failures and outages.
Transmission Increasing transmission capacity by building new lines is one way out of the problem. According to the IEGC, the 31 That is, without resorting to load shedding or rescheduling of generation.
251
CEA would develop a perspective long-term (10–15 years) transmission plan for the ISTS. These plans would be continuously updated to take care of the revisions in load projections and generation scenarios. The PGCIL, in turn, would carry out the annual planning process for identification of major inter-state and inter-regional 32 transmission systems and system strengthening schemes consistent with the perspective plans of the CEA. Construction of new lines takes time. It is obvious that there is a strong need to examine the possible options for increasing the transmission capability on present sites, and by making full use of existing transmission systems through upgradations. When feasible, upgrades are an attractive alternative, because the costs and lead times are less than those for constructing new transmission lines. Thermal limitations are the most common constraints that limit the capability of a transmission line, cable, or transformer to carry power. Thermal limits are imposed because overheating leads to two possible problems: (i) the transmission line loses strength because of overheating which can reduce the expected life of the line, and (ii) the transmission line expands and sags in the centre of each span between the supporting towers. Historically, transmission lines had been given static or fixed ratings that limit the amount of power that can be transferred on the line. These ratings have been based on assumptions regarding weather and conductor conditions. Worst-case expected conditions, such as the highest expected ambient temperature, no air movement, and low conductor emissivity, are usually used. At very high wind speeds, however, mechanical stability of the lines would be affected. This method provides a constant (static) rating that can be used by the system operator to ensure that transmission line conductors do not sag below the specified limits, and do come in contact with trees or other objects, thus affecting reliability and safety. In reality, transmission line ratings are highly dependent on the amount of cooling by wind and heating by the sun, and change constantly with the weather. It is possible that at certain times there is significantly more transmission line capacity than what the standard static transmission capacity would imply. The options and actions available for increasing power transfers by overcoming thermal limitations are described below.33 32 Inter-regional transmission system diverts can be used to divert power from power excess regions such as the Eastern Region Grid (ERG) to deficient regions such as the Southern Region Grid (SRG). The converse is also true. For example, the NRG collapse could have been averted if the excess power from the eastern part of the NRG was transferred to the Western Region Grid (WRG). However, this possibility was precluded because the WRG was also operating at high frequency during the NRG crisis. 33 John Maken: http://www.eia.doe.gov/cneaf/pubs_html/feat_trans_ capacity/w_sale.html.
252 India Infrastructure Report 2002 Box 9.2.3 Dynamic Line-Rating System The San Diego Gas and Electric (SDG&E) designed and installed a real-time dynamic transmission-line-rating system on one key 230 kV transmission line that limits import capability into the SDG&E system. The method monitors overhead conductor tension, ambient temperature, and net solar radiation temperature rise. Data is passed to a ground station via spread spectrum radio and sent to SDG&E’s energy management system (EMS). Calculations are performed to determine line condition such as sag and dynamic thermal to constraint, as well as operating warnings (time to thermal overload under system conditions) by the EMS, with results displayed on operator screens. Signals are given to operators, who can reduce load or generation to keep the line within thermal and vertical clearance constraints. The project successfully demonstrated the feasibility and reliability of providing real-time transmission line ratings to the system operator. Real-time line ratings for the transmission lines monitored in this study had 40 to 80 per cent more power transfer capacity than the static transmission line ratings presently applied. Fifteen utilities in the United States have implemented SDG&E’s monitoring system. Source: http://pier.saic.com/PDF/SDGE05.pdf.
• Monitoring ambient temperature and line tension is possible by special sensors. Using these data, the actual line sag of the line at its mid-span and the actual limit on the current that the line can handle can be calculated (see Box 9.2.3). • The thermal limit of a transmission line is also based on the component that would be the first to overheat. Hence, a substantial increase in the overall thermal rating of the line can sometimes result from replacing an inexpensive element such as a disconnect switch or circuit breaker. • The most obvious, but also the most expensive, method for minimizing thermal effects is to replace existing lines with larger (higher cross-section) conductors or to add one or more lines, forming ‘bundled’ lines. Replacing lines with larger ones, or bundling them, usually requires (i) substantial reinforcement of the tower structures and, possibly, the concrete footings of the towers and (ii) enhancing sub-station equipment. • Replacing HVAC (high voltage alternating current) by HVDC lines.34
Suppressing Instability Instability35 of the kind that plagued the NRG on 1 January resulted from excess power in the network, with respect to the available lines. Clearly, high generation excess, witnessed during the NRG crisis, can be handled best by backing down generation. However, under less severe conditions,36 redirection of power flows can reduce 34
Nationality we have accepted HVDC transmission as an alternative to HVAC transmission due to advantage like fast controllability of power and the possibility of interconnecting systems operating at disparate frequencies. 35 This type of instability is referred to as transient instability. 36 Such as those experienced under off-peak conditions, for example during the late nights and early mornings.
loading on critical lines, thereby permitting larger power transfers. Power redirection can be achieved by using: • Phase-angle regulators (PAR).37 The use of PARs has increased in recent years; however, their installations are relatively costly. A 230 kV, 300 MVA (mega volt ampere) PAR with a phase angle capability of plus or minus 60 degrees is estimated at $30,000,000. • Flexible AC Transmission System (FACTS) devices. The FACTS concept uses new power-electronics switches and other devices to provide faster and finer controls of equipment to change the way the system power flows divide over the system under normal conditions or during contingencies. Transient stability can also be maintained by dual generator control systems, namely automatic voltage regulator (AVR) and governor control systems,38 which can maintain a fixed voltage from the generator regardless of demand levels. In India, high frequency is experienced mainly during the monsoon months when the agricultural use and the usage of fans and air-conditioners falls due to the cooling effect of widespread rains. Load falls to the extent of 4000 MW in the NRG due to rains are common during the monsoon season or during winter rains. Sustained frequency overshoot can damage the shaft of the turbine, resulting in permanent damage to the generator. Overloading the grid due to excess generation results in poor voltages and power quality. Most modern thermal power plants are equipped with governors to regulate generation in line with demand. The CEA report on the NRG disturbance clearly states the reluctance of thermal 37
PAR is also referred as power-angle regulator or phase shifter. The AVRs maintain a fixed voltage level, while governors regulate the mechanical power output of turbine. 38
Framework for the Energy Sector power stations in reducing generation at times of low demand and operating their units on a ‘free governing mode.’ Despite the fact that ‘overgeneration’ is harmful both for the generator and the grid, why does this practice continue unabated? The main reason for the errant behaviour of generators is related to the commercial angle. Today many SEBs have ‘incentives’ for employees, which are directly linked to PLF (plant load factor). The tariff structure for bulk power supply, applicable to the NTPC, bases the fixed charge on a minimum PLF of 68.5 per cent, which may be raised to 80 per cent if the proposed ABT is implemented. With no specific penalties currently in place for violating the directions of RLDCs, the generators have a ‘free’ run, resulting in an ‘insecure’ grid. The proposed ABT however, has certain penalties on generators if they violate grid discipline. The calculation of variable charge is based on interalia a normative level of specific oil consumption (SOC). When the RLDCs instruct the power stations to operate under reduced load conditions, more oil support is required. Thus the tariffs of the generators need to be revised to appropriately compensate for excessive oil consumption under such conditions, and the policy has to recognize this need. Judicious combinations of both capacity expansion and instability suppression methods have to be implemented to increase reliability of ISTS. Any one or combination of the methods finally selected would have to depend on a number of factors such as cost, operating conditions, etc.
RESPONDING SWIFTLY
TO
EMERGENCIES
Directions to the power stations to back down on 2 January 2001 by the NRLDC were often inadequate and their execution delayed. For example, at 0107 hours the transfer capability of the east–west corridor had reduced by around 600 MW. However, the NRLDC asked Singrauli and Rihand STPS to back down by only 80 MW each.39 Why did the NRLDC not anticipate (i) the reduction in transmission capacity of the grid and hence the extent of power imbalance as a function of time and (ii) the sequence of events that ultimately led to the collapse of the NRG? Problems in making an accurate assessment of the real time status of the grid stem from the fact that: • The path followed by the electric current in a complex intermeshed grid, such as ours, is difficult to track due 39
During the restoration of the NRG, the BBMB system collapsed due to lack of coordination control on extending power supply to load centres without ascertaining the adequacy of generation available in the system. Power from the BBMB should have first increased generation in Dadri TPS before supplying power to load centres.
253
to lack of on-line communication and display facilities and their functional coordination. Further, the complete understanding of the system operation for reactive power and real power support is not available. Consequently, predicting the redistribution of power in the grid when transmission lines trip is no mean task. • Cascading effect, that is the sequential tripping of lines, happens in milliseconds. The use of modern technology/tools is crucial for the successful operation of any modern grid. Computer software that simulates the transmission network is available. Such software can be used not only to determine the distribution of power in the grid, but also the extent of power imbalance, and hence of the magnitude of backing down/stepping up generation of electricity. In fact, on-line computerized systems are available to predict cascading failures and to suggest preventive control. Grid disturbance can be triggered by a variety of sources such as failure of insulators, circuit breakers, or transformers at any of the numerous buses in the grid. Consequently, it is vital that a database of cascading failures resulting from all possible ‘triggers’ be stored for use during emergencies. Experience and thorough training in the use of such software is essential to the ability to respond swiftly to emergencies.
MAINTENANCE AND MONITORING The upkeep of the central transmission system is the responsibility of the PGCIL. It is the PGCIL’s duty to ensure that various constituents of the grid such as transmission lines, transformers, and control equipment are functioning within safe limits by having a monitoring and maintenance schedule in place. Unfortunately, the CEA’s report on the NRG disturbance suggests otherwise: • Pole 2 of the Rihand–Dadri HVDC bipole, which is a crucial link in the NRG, had been non-operational since 11 December 2000, because of the failure of converter transformers! On 1 January 2001, even spare transformers were not available. In the year 2000 there have been as many as four failures of converter transformers of which the last two have occurred within a span of one week. As a result, no spare transformer was available to revive the ‘dead’ pole on the night of 1 January 2001. Incidentally, all the transformers that failed were made by BHEL (Bharat Heavy Electricals Limited). In fact, because of transformer problems, the HVDC link was expected to operate with only one pole. • Pole 1 of the HVDC Rihand–Dadri link faced flashovers four times and subsequently had to be operated at a reduced (by 33 per cent) capacity. In a similar manner,
254 India Infrastructure Report 2002 the 400 kV Obra–Panki line of UPPCL also tripped due to flashovers. Several questions arise: • Did PGCIL explore ways of procuring replacement transformers considering the strategic nature of the HVDC link? • Did PGCIL conduct a proper technical assessment about the residual life of the BHEL transformers? In fact, the CEA also reported40 that the BHEL transformers also performed unsatisfactorily during the eastern grid disturbances. Were the correct lessons learnt from that experience? • Improper maintenance procedures or overloading can lead to transformer failures. Did the PGCIL regularly monitor the performance of the BHEL transformers and take appropriate corrective actions? Or were they subject to gross misuse? • Flashovers could also have come about in all likelihood due to heavy pollution and dense fog on the night of 1 January 2001. Since the area was prone to pollution/fog, did the PGCIL ensure periodic cleaning of the HVDC and Obra–Panki lines? Apart from a plan for the regular maintenance of grid equipments, a ‘condition monitoring scheme’ should have been implemented.
Condition Monitoring Condition monitoring refers to the techniques and procedures used for determining the residual life of equipments. In power systems, they provide ‘warning signals’ for specific problems that may plague transformers and associated tap changer mechanisms, circuit breakers, and switches. While traditional and modern methods will both detect typical events, modern methods allow utilities to focus on precursors of specific failure. For example, utilities periodically carry out dissolved gas analysis (DGA) on substation transformer oil to monitor the level of insulation degradation. Through a highly developed form of analysis, the ability of DGA to prevent transformer failure depends on when and how frequently the test is performed and how long it takes for a given problem to cause a failure. Online dissolved gas level sensors can carry out such tests relatively more frequently, and that too without disturbing the operation of transformers. These sensors do not provide information on the individual gas concentrations, but provide an overall indication of key gases. The test is much simpler than DGA, the measurements being made much more 40
See CERC order on NRG disturbance.
rapidly, thereby making it possible to avoid the limitations associated with discrete testing. Modern condition-monitoring sensors include signalprocessing techniques to determine trends from data automatically gathered. Thus stresses on transformers and status of ancillary pumps or fans41 can be continuously monitored and their effects analysed. Modern conditionmonitoring systems provide a powerful, fully automated and invaluable diagnostic tool that can be used for decision making (see Box 9.2.4). For example, monitoring systems can draw conclusions about transformer status by analysing numerous parameters such as operational and atmospheric conditions. When two transformers are operated in parallel, the base station can detect abnormalities through the comparison of operational parameters from either unit. Therefore on-line condition monitoring can be justified on its ability to increase the service life of a transformer.
COMMUNICATION The RLDCs need accurate real-time data from remote grid equipments such as transmission lines, transformers, and circuit breakers to proactively manage the transmission network. Very little monitoring is done today. Power meters themselves are hardly monitored except at key load centres and switch yards. With modern technology, it is possible to use networking and Supervisory Control and Data Acquisition (SCADA)42 to obtain the overall status of power system operations within milliseconds. Computerized link-ups would also facilitate and record communication between the RLDCs and TPSs (thermal power stations). It may be recalled that an important factor in the NRG collapse was the communication gap between NRLDC and the Anpara TPS with respect to the timing and quantum of generation back down.43
POWER QUALITY Here, we look at two issues: (i) Redefining power quality due to the changing profile of power consumption, and 41 The capacity of a transformer typically depends on the operation
of ancillary pumps or fans. Hence knowing the status of these can avoid the de-rated operation of transformers. 42 SCADA is a commonly used industry term for computerbased systems allowing system operators to obtain real-time data related to the status of an electric power system and to monitor and control elements of an electric power system over a wide geographic area. The SCADA system can determine: (i) fault location, (ii) position and status of breakers, (iii) amperage on transmission lines, (iv) power quality, (v) substation security fires and door alarms, (vi) substation primary transformer status. 43 See point 4 in the section on NRG collapse.
Framework for the Energy Sector
255
Box 9.2.4 Intelligent Systems and Condition Monitoring Interpretation of data is of fundamental importance in achieving the full advantages of modern condition monitoring. It is desirable that monitoring systems themselves be self-sufficient, requiring minimal user involvement. This has been achieved through the development of intelligent software systems, such as expert systems, that identify failure conditions with software algorithms. The expert system is a specialized area of Artificial Intelligence (AI) in which computer software is used to tackle the problems normally executed by highly educated human experts. Expert systems work, and save money, due to the fact they can accumulate and organize the expertise needed to keep operations running, and make that knowledge immediately available even to less experienced workers when problems occur. Commercially available computerized diagnostic expert systems, when implemented, can reduce the time and effort required by equipment specialists by comparing data with rule sets generated by human equipment experts, incorporating many years of machinery diagnostics expertise. The expert programme evaluates the machine conditions for magnitude of the vibration, and frequency of the vibration, form of the vibration, amplitude/phase angle to determine the status of various components.
(ii) who provides ancillary services for improving power quality?
Redefining Power Quality Over the years, the standard method for defining power quality was the power factor. A low reactive power in the system meant higher power quality. However, in the last decade, the increasing pace of office and plant automation, emergence of computer education institutes, dotcom companies, call centres, and changes in customer life style have introduced a new power quality parameter, ‘HARMONICS’.44 The presence of harmonics tends to generate more heat than pure sinusoidal currents, and causes overheating of equipments like transformers and generators. The percentage of loads that generate harmonic flows in the distribution network, such as computers, UPS, sodium vapour lamps, tube lights, and air-conditioners, have grown by leaps and bounds. In fact, HVDC transmission systems may also generate harmonics. Hence, the use of only the power factor as a measure of power quality merits reconsideration (see Box 9.2.5). In fact, remedial measures should be taken to tackle the problem of harmonics. Software, which can determine harmonic flows in various parts of transmission, can also be used.
Ancillary Services Section 3.7 of the Planning Code of the IEGC states that ‘The actual program of implementation of transmission lines, reactors and capacitors will be determined by CTU in consultation with the concerned agencies. The completion of these works, in the required time frame, shall be ensured by CTU through the concerned agency.’ Thus 44 In simple terms, harmonies refer to the flow of current waveforms that are distorted. These distorted current waveforms result from the use of non-linear loads.
ancillary services can be provided either by the states themselves or by PGCIL, on a chargeable basis. In our opinion, however, since the RLDCs are better aware of system requirements, they are best placed to make such investments through the Central Transmission Utility (CTU). The state power utilities may also be aware of their own systems’ ancillary support requirements. However, the voltage profile at the interface between the PGCIL network and a particular state may be affected by the system parameters at neighbouring buses, which connect other states to the ISTS, and is best taken care of by PGCIL. It should also be pointed out that assessment of power quality should be done both by the state power utilities as well as the CTU. This is because under the current load despatch mechanism, the states are required to furnish their load requirements 24 hours in advance to the RLDCs, which could then plan for provision of ancillary services. In the aftermath of the 2 January 2001 NRG collapse, various utilities involved were asked by the Ministry of Power (MoP) to implement certain actions aimed at making the Indian power system more reliable. What could have been the reasons behind the PGCIL’s lackadaisical approach to the maintenance of transmission line/equipment? Why does the system adequacy and reliability aspect of the Indian power system need to be put under the microscope? It took a disaster of the magnitude of the grid failure to stir the various players into action. As it exists, there is unfortunately no mechanism, either legal or economic, in place to the provide adequate motivation to the PGCIL to study and implement the system’s requirements suo moto. In fact, the tariff structure for PGCIL is such (as shown in a later section) that non-funtional/reduced capacity operation of some transmission lines does not affect its profitability significantly. It is in the backdrop of this anomaly that we suggest alternative management mechanisms for transmission services.
256 India Infrastructure Report 2002 Box 9.2.5 Power Factor Versus Displacement Power Factor When active power is divided by apparent power in the presence of harmonics, the result is known as total power factor (PF). The component of power factor not contributed by harmonics is known as displacement power factor (DPF). PF and DPF are equal in completely linear circuits, but with non-linear loads their values would be different. If PF and DPF differ by a facto r of 10 per cent or more, the difference is probably caused by harmonics. The degree of difference between PF and DPF may also suggest a course of action, depending on the types of loads in the system: • When PF and DPF are essentially the same value, linear loads dominate the circuit. In this case, low power factor can be compensated for with kVAr correction capacitance. Even in systems with low levels of harmonics, caution should be exercised because kVAr capacitors applied improperly can lead to over voltages. • When PF is significantly lower than DPF, low power factor can be corrected by applying line reactors directly to the sources of harmonic current or by using kVAr capacitor networks with series inductors to limit harmonic current in the capacitors. There is a need to exercise caution in the use of kVAr correction capacitors and compensating filters, to avoid resonance problems at harmonic frequencies.
TRANSMISSION OWNERSHIP The aim of any deregulatory process is to provide cheap, good quality power for an extended duration. This requires more competition in generation and a highly reliable transmission system. Consequently, it is necessary to deregulate access to the transmission network by not only removing technical bottlenecks but also by appropriately pricing the access. The transition to a deregulated market is difficult. We shall try to address the following issues, while trying to understand the existing pricing mechanism and the alternatives: • Should the transmission company (both PGCIL and the intervening state network used for inter-state transfer of power) be forced to allow any amount of wheeling over its transmission network? Who pays for the losses? How should the losses be priced? • In case of an emergency, which loads and generators should be dropped first? • Should the priority order in which transactions take place and/or are cancelled in the case of an emergency depend on the size of the utility, the amount of energy traded, the number of utilities affected by the transaction, or the effect of the transaction on the level of security? • Who will have to pay for the upgraded transmission network? In the case of state networks, who pays for the stranded investments? • How should the performance of the transmission system be measured?
Current Mechanism of Transmission Pricing The basic elements of transmission tariff are as follows: • The allowed interest on loan capital is as actually incurred. Depreciation is calculated as per the norms laid
down by the CERC from time to time. Operation and Maintenance charges are calculated at 1.5 per cent of the actual expenditure at the time of commissioning of the transmission line in the plains and at 2 per cent of the actual expenditure at the time of commissioning of the transmission system in the hilly areas. These charges escalate every year by an index, which has a 60 per cent weight to the Wholesale Price Index (WPI) and 40 per cent weight to the Consumer Price Index (CPI). • For the existing transmission system, the equity and loan component of the transmission systems commissioned is each notionally presumed to be 50 per cent each of the book value of the transmission system for the computation of tariff. A return of 16 per cent on equity is a component of the fixed charges. Working capital consists of O and M expenses for one month, maintenance spares at a normative rate of 1 per cent of the capital cost, receivables equivalent to two months’ average billing calculated on normative availability level. Tax is computed on the following income streams: The 16 per cent return on equity. The extra rupee liability on account of foreign exchange variation is allowed in computing the return on equity not exceeding 16 per cent in the currency of the subscribed capital. In addition to these fixed charges, PGCIL is entitled to incentives if the availability of the transmission system is certified to be above 95 per cent. Incentive is applicable at the rate of one per cent return on equity for each percentage point increase in availability. There are many disadvantages in this arrangement: • This is a traditional cost-plus kind of a tariff determination mechanism. Pure cost-plus regulatory regimes are best suited to environments with high cost uncertainty. Since the transmission business in India is a regulated monopoly and is under government control, these
Framework for the Energy Sector uncertainties are reduced considerably, though not eliminated. In such conditions, sliding scale mechanisms (combinations of cost plus and price cap mechanisms) should ideally be used. • Transmission charges are allocated among the beneficiary states (in a particular region) in proportion to the energy drawn by them from the regional grid. This does not give the users an idea about the extent of resource usage, for example losses, and hence does not provide signals for efficient resource utilization. Moreover, there are certain resources in a region, which are not utilized for transmitting power for some of the states in a region. Such resources, however, are pooled together for the calculation of fixed charges. • Calculation of availability for the transmission system implies that the available time of each line and SVC (Static Var Compensators) equipment is weighted by its rated capacity or Surge Impedence Loading (SIL), in case of AC transmission lines. Hence, even if a line is partially loaded due to some failure, its availability will still be 100 per cent. Since the availability is system based, some states may have to pay towards incentives, even if specific equipment at the interface with that state is not available. • The tariff mechanism provides neither any incentive for optimal capacity expansion in transmission or for capacity expansion in generation. • It does not provide incentives for optimal capital structuring. • There is no economic basis of load shedding that can be derived from the tariff. • Total regional losses are calculated at the end of the month and allocated among the beneficiary states in proportion to their energy drawls. Hence there is no incentive mechanism which forces the beneficiary states to choose between own generation and drawls from the grid. • In the case of inter-regional transactions, the transmission charges for inter-regional assets are shared on 50:50 basis by the two contiguous regions. Some states in a particular region may have to pay even if they do not benefit from such transactions! • According to the order of the CERC on petition no. 86/2000, dated 8 December 2000, the states (in case of wheeling of power through the state utility system) should ideally agree on wheeling charges. If the two parties are not able to reach an agreement, the calculation is based on the contract path method or the verifiable opportunity cost of the wheeling utility, whichever is higher. The contract path, however, may not be the actual path of power transmission. Though it is well recognized that tracing of flows may be quite involved, the opportunity
257
cost criteria may be enforced for such transactions. This is because wheeling charges should provide incentives to build new lines and expand capacity when necessary. Wheeling charges should reflect the system reliability costs. Power system reliability and security must be preserved
Proposed Mechanism All transmission transactions can be unbundled as long term and short term. These can further be categorized as firm and non-firm transactions. The non-firm transactions can be sub-classified as curtailable and as-available. The firm long-term transactions (FLTT) can be viewed as commitments between the central sector power stations and the beneficiary states. Intra-regional and inter-state transactions due to surplus within a state (due to own generation or excess central sector quota) for a month to less than a year can be treated as firm short-term transactions (FSTT). Curtailable non-firm short-term transactions (CNFSTT) could use unused transmission capacity when surplus generation in one state/region and deficit in another state/region are encountered during real time operation. As-available non-firm short-term transactions (AANFSTT) arise during real time operation and may be executed in the interest of the players and the security of the grid. Each of these transactions may require reactive support to some extent. Ideally, Optimal Power Flow (OPF) is the most effective congestion management mechanism for highly intermeshed systems. Marginal costs based on OPF are spatially and temporally different. The prices based on these marginal costs give a precise indication of scarce resource use, indicate requirements for capacity addition in both generation and transmission, and can also be adjusted to meet the revenue reconciliation requirements. However, calculation of such prices is very complex even in highly computerized systems. In India, we do not even have the bare communication network in place to implement even a Unified Load Dispatch Scheme. Moreover, when a large scale OPF is run, users find it difficult to understand the rationale behind the charges and their volatility. Hence we are forced to consider certain second-best45 solutions to the problem that build on the existing infrastructure with some small enhancements. Each of the unbundled transactions described above use the available resources to a different extent and hence need to be priced differently. In India, the buyers of bulk power usually have direct contracts for the sale of power with the suppliers 45 Because prices here shall definitely be different from marginal costs, and hence will reflect the market power of the wheeling utility to some extent.
258 India Infrastructure Report 2002 (either central sector generators, or other states, or now the Power Trading Corporation). The transmission company (PGCIL) merely acts as a carrier of power and is required to ensure secure operations through the RLDCs.
Firm Long-term Transactions Calculation of FLTT Commitments would essentially require identification of various states and their quota from the central sector power stations. The method of recovery of fixed charges for FLTTs could be as follows: Connection charges: These are the payments or charges for interconnecting an SEB or a generator to the PGCIL grid. Thus, charges for facilities which serve only one SEB are recovered directly from that SEB on a cost of service basis. Use of the system charge: This is proposed to be paid by all the users in proportion of their maximum demand, measured as an average of their highest demands (in MW) during the system peak demand periods. These system peak demand periods could be estimated on a daily, weekly, monthly, or seasonal basis. The payment of this charge entitles the user state to a certain transmission capacity. The state, if it so desires, may trade these transmission rights in real time. All this requires the RLDCs to implement the Unified Load Despatch Scheme. The regulator could go in for a comprehensive price cap on the use of the system charge or introduce a mechanism that involves price cap on individual elements of the transmission charge. The basic elements of the tariff could be as below: • Return on capital employed (ROCE): This would cover both long-term and short-term capital. Given the institutional structure, the transmission projects are faced with minor risk during operation phase. Even these risks could be covered through O&M charges. The returns associated with risks during construction and commercial risks could be line specific and negotiated. ROCE (rather than ROE (return on equity)) gives an incentive to the transmission owner to optimally choose capital structure and long-term and short-term capital (working capital) requirements. Base working capital requirements could be based on RPI-X (retail price index with X as a measure of efficiency improvement) kind of a price cap mechanism. • Depreciation: This could be calculated as per the norms laid down by the commission from time to time. • O&M charges: Currently, O&M charges are linked to the capital costs. It would be desirable if each of the five regions in India can be operated as separate companies and these costs can be benchmarked. This would require ring fencing of the five regions and the regulator would need to ensure the correctness of the data provided.
Alternatively, sliding scale mechanism could be adopted for sharing profits or cost over-runs. To implement this, an initial hearing must establish the target O&M expenses, say O*. If Ot be the actual O&M expenses at the prices that initially prevail in year t, then the sliding scale would adjust prices so that the actual O&M charges, would be given by:
Ota = Ot + h (O* − Ot ),
(9.2.1)
where h is constant between 0 and 1 and must be decided by the commission depending on the risk perception. If h equals 1, the utility earns the target O&M charges in each period. This is essentially a cost-plus regulation. If h is smaller and closer to zero, it becomes a fixed price contract. The latter contract may be used if economic and technological uncertainties are perceived to be less important by the regulator. The sliding scale mechanism provides incentives for cost minimization, and could be used not only for the O&M expenses, but also for a comprehensive use of the system charge. The variable charges would be based on reactive support required for the firm transactions. These services could be procured from ancillary service providers, who would be required to inform the RLDCs and the states about the capacities available. Section 3.7 of the IEGC requires that ‘The actual program of implementation of transmission lines, reactors and capacitors will be determined by CTU in consultation with the concerned agencies. The completion of these works, in the required time frame, shall be ensured by CTU through the concerned agency.’ This requirement lays the entire responsibility of planning and implementation of reactive support (among other requirements) on the CTU. Since all inter-state bulk power agreements are between the central sector generators and the state utilities, PGCIL essentially acts as a wheeling utility and hence must provide the desired reactive support.
Short Term Transactions Transmission owners submit price/quantity offers for the transmission of electric power. The transmission owners must submit offers that reflect the current system conditions, equipment limits, security constraints, etc. To decide the amount of power they are capable of transmitting and, thus, the price of their offers, transmission companies need to solve an economic load dispatch (ELD) problem (off-line) with all the system security constraints. ELD is executed iteratively until there is no system constraint. This will help in calculating the available transmission capacity (ATC). Since the willingness to pay of the SEBs also needs to be taken into account, a double-sided auction mechanism where bids are invited from the SEBs, is
Framework for the Energy Sector adopted. This would act as an effective congestion management mechanism as some of the SEBs would have to opt for load shedding. Finally, the RLDC, in accordance with the norms laid down by the REB, must only assign contracts that do not violate system security and reliability. All kinds of short-term transactions can be dealt with in this framework, each of which would be valued differently by the beneficiary states/entities. In order to avoid double counting, the ATC for short-term transactions may be reduced from the capacity on which fixed charges are calculated for firm long -term transactions.
Incentives for Capacity Expansion, System Security, and Reliability The philosophy underlying these incentives is that the power system reliability and security should remain unaffected. The above mechanism may provide incentives for PGCIL to suo moto start strengthening the system. Bulk transmission facilities must provide: (a) adequate transmission capacity, which includes the task that voltage, frequency, and thermal limits are maintained and system stability is ensured following faults, switching and other transient disturbances; (b) power quality: The ability of the power system to operate loads without disturbing or damaging them, a property mainly concerned with voltage quality at points of interface between the state and the PGCIL network. The ability of the loads to operate without disturbing or reducing the efficiency of the power system, a property mainly, but not exclusively, concerned with the quality of current waveform. Power quality issues include short-term events such as voltage sags or dips lasting a few seconds caused by faults on nearby feeders, large loads switching on etc., and subcycle transients caused by switching power factor correction capacitors, lightening strikes etc. Power system harmonic and flicker issues also fall into the category of power quality. • Voltage sags:46 In India, at the central level, NTPC and NHPC are involved in generation, and PGCIL transmits this power to the beneficiary states, which transmit and distribute electricity to the final consumers. The responsibility of power quality is hence dispersed. Who is responsible for the customer’s power quality? The generator? The PGCIL? The STU? The state distribution company? Alternatively, the customer himself? A fault at the distribution level will lead to interruption for some customers and voltage sag for others. A fault at the transmission level will only lead to voltage sag, but often for many customers. The number of voltage sags will further increase with the transport of power over large distances. The responsibility of transmission becomes important because generation is concentrated in 46 A reduction in the AC voltage, at the power frequency, for a duration ranging from half a cycle (root mean square) to a few seconds.
259
the eastern and central coal belts in India while the major load centres are in the west. • Harmonic flows: An important issue is the allocation of financial responsibilities to guarantee agreed levels of power quality. The main requirement in this respect is the availability of accurate and sufficiently discriminating measurements. The index needs to distinguish between ‘injected’ and ‘absorbed’ harmonic currents. Such discrimination requires measuring the displacement angle between each harmonic current and its corresponding bus voltage. System-wide harmonic state estimation (HSE) is capable of providing information on harmonic generation and penetration throughout the power network. As long as the transmission utility remains a ‘natural monopoly’, the incentive for investments aimed at an overall reduction of the sag frequency, harmonic content etc. will be low. Even though the beneficiary states are able to choose suppliers, their physical interconnection through PGCIL’s network will remain unchanged. There are indications that power quality in terms of these voltage sags and harmonic content will increase in the future. Still, some beneficiary states will demand a reduction in them. There could be many options: • Power quality guarantees: The beneficiary state receives compensation for each event exceeding a certain severity (in magnitude, duration, or frequency). The transmission service provider or an ancillary service provider may provide such an additional service. Alternatively, the regulatory commission may fix a basic compensation to all customers as a part of the connection fee. • Another option could be to provide high quality power to certain beneficiary states that agree to pay for the investments involved. System reliability may be measured in terms of system indices and load point indices. Deficient power system states may arise either due to insufficient generation or due to grid disturbances. Both require system re-dispatch and load shedding. Assuming that PGCIL maximizes its profits, we need to have incentives based on bus isolation values (such as expected load curtailed). Such an index would provide incentives for PGCIL to invest optimally in capacity expansion and provide reactive support at various buses to improve the power flow. The reason for load curtailment, however, may be required to be certified by the Member Secretary, REB (Regional Electricity Board). Incentives/disincentives payable (or to be paid to) the beneficiary states may be adjusted in the fixed charges payable by that state. The Central Commission, in consultation with the REB, may estimate a dead band for such indices, below which incentives may be paid to PGCIL. Such indices would capture any system strengthening effort by the transmission company by way
260 India Infrastructure Report 2002 of either putting up additional lines, FACT devices, or phase-shifters etc. However, analysis of these indices requires a curtailment schedule, if cost benefit analysis is to be made. The loads may be curtailed based on the nature of transactions as discussed earlier.
In addition to load point indices, total transmission losses can be used as a system index, since losses indicate the adequacy of capacity. The return on equity of PGCIL may be revised downwards if the losses are deemed excessive by the regulator.
Box 9.2.6 Issues for Transmission Management Related to ABT Implementation of Availability-based Tariff would also require system strengthening and hence an appropriate transmission pricing mechanism, as is shown by the following arguments: 1. ABT is based on separate charges for generation capacity and energy. The commission notes that this will enhance the incentive for trading of power. Trading of power, however, requires a reliable power system. Hence, appropriate transmission pricing, with incentives for improvement of system reliability, becomes imperative for the successful implementation of ABT. 2. Court cases and stays, on the UI charge, have been major hindrances in the way of implementation of ABT. Use of FACT devices, which is facilitated by the incentive mechanisms above, can be used effectively for power flow control and regulation of unscheduled interchange of power. 3. The methodology of scheduling under ABT requires that in case of any forced outage, or in case of any transmission bottleneck, RLDC will revise the schedules. The charges for unscheduled interchanges that take place before the revised schedules become effective will have to be borne by the players. This may be unfair to the beneficiary states. In the case of grid disturbances, PGCIL bears no responsibility for unscheduled flows. The mechanism suggested above would directly provide incentives for PGCIL to put the equipment in place and be adequately compensated for improved system reliability. 4. The CERC order on ABT states that it facilitates merit order dispatch, grid discipline, and trading in capacity and energy. However, all this requires improvement in transmission capacity. Hence, implementation of ABT requires a sound transmission tariff policy in place. 5. The role of verification and certification of grid indiscipline, which leads to the imposition of an unscheduled interchange charge, rests with the RLDC, a unit of the CTU, which itself is a commercially interested party. Though the central commission has noted in the ABT order that ‘regarding the claim for payment of UI charges by transmission utility, specific instances of dereliction on the part of the transmission utility can always be brought up’, no rational commercial organization could be expected to adversely affect its own profits by pointing out its own ‘dereliction’. Moreover, given that the entire data on system operation is available only at the RLDC, it may be difficult for anyone outside the RLDC to know the exact nature of the power system failure. Under the transmission incentive mechanism suggested above, load point indices at the bus level are recorded even by the state utilities, and hence there is no monopoly of information. 6. Due to a transmission bottleneck or outage, the schedules of generation and drawl are revised. It would be unfair to deny the capacity charges to the generator, and at the same time since the benefit of this generation capacity is not available to the beneficiary states, it should not be charged to them. Who pays for it then? 7. Moreover, is the grid frequency likely to improve by imposition of a frequency linked UI charge? A combination of tariff mechanism and load curtailment needs to be used for congestion management. The following would clarify the point made that any power system essentially operates under three conditions: (i) Steady state, that is where demand equals supply, and voltage, frequency, active power, reactive power, and thermal limits of the lines are all within limits; (ii) A condition in which voltage, frequency, and thermal limits are violated. Such a condition may be corrected by using auxiliary services, FACTS devices and through proper coordination of various protection schemes; (iii) Disturbed condition, under which islanding and other extreme measures may have to be resorted to. Since, the beneficiary states need to pay explicitly for reactive support (provision of which, as stated above and as per IEGC, is the responsibility of the CTU) under the proposed mechanism, condition (ii) should be first corrected using technical measures. The proposed ABT resorts to penalties the moment the frequency deviates from the target frequency. Financial penalties here may be resorted to under conditions (ii) and (iii) only after all technical support has been optimally used (this needs to be verified by the REB).
9.3 ORISSA POWER SECTOR REFORMS: GETTING BACK ON TRACK Sidharth Sinha The objective of the Orissa power sector reforms is to improve the availability and efficiency of the power supply
by restructuring and substantially privatizing the power sector in Orissa. The government’s ultimate objective is
Framework for the Energy Sector to provide an appropriate policy environment for growth of the power sector and withdraw from it as an operator of utilities. The sector is expected to function with privately managed utilities operating where feasible in a competitive environment under an appropriately regulated power market. Like other states, Orissa is also faced with the problems of power shortages, imbalance between tariffs and costs, and high transmission and distribution (T&D) losses. However, the proportion of agriculture consumption in Orissa, at 6 per cent, is particularly low compared to other states which report agricultural sales of 30–40 per cent of total consumption. Orissa is, therefore, spared the problem of large, unmetered, and heavily subsidized agriculture consumption that most other states face. In addition hydropower accounts for about 40 per cent of total consumption, currently at an average price of approximately 60 paise per unit. The key principles of the Orissa power sector reform programme are: (i) Restructuring of the former Orissa State Electricity Board (OSEB) by corporatization and commercialization: unbundling and structural separation of generation, transmission, and distribution into separate corporations; (ii) Privatization: through private sector participation in the hydro generation and grid corporation and privatization of thermal generation and distribution; (iii) Competition: competitive bidding for new generation; (iv) Separate regulation: development of a power sector regulatory commission; and (v) Tariffs: related electricity tariff reforms at bulk power, transmission, and retail levels. A number of critical elements of the power sector reform process in Orissa have been completed. • The Talcher Thermal Power Station of the OSEB has been transferred to the National Thermal Power Corporation (NTPC) (3 June 1995). • OSEB was corporatized into the Grid Corporation of Orissa (GRIDCO) and Orissa Hydro Power Corporation (OHPC) (1 April 1996). • The Orissa Electricity Regulatory Commission (OERC) was established in August 1996. It has issued four tariff orders to date. Separate bulk supply, transmission and distribution, and retail supply tariffs have been issued. The level of cross-subsidy in the retail tariff has been progressively reduced. Tariff levels have been increased at about 10 per cent in successive years. • The Orissa Power Generation Corporation (OPGC) was privatized in 1998 with transfer of 49 per cent stake to a private operator, AES, for Rs 603 crore.
261
• Four distribution companies were corporatized under GRIDCO and privatized with transfer of 51 per cent stake to private operators (1 April 1999). The Bombay Suburban Electric Supply Company (BSES) acquired three distribution companies (DISTCOs) in the first round and AES acquired the fourth one in the second round. GRIDCO received Rs 1.6 billion for the 51 per cent stake of the distribution companies. The most crucial element in the power sector reform process is the improvement in efficiency of the distribution function. Improvements are necessary in the quality and the maintenance of the distribution network so as to provide high quality service to the consumers. Equally important is the need for improvements in revenue generation through metering, billing, and collections. Under the State Electricity Board (SEB) regime, the system got locked in a vicious circle of low revenue generation leading to low quality, in turn leading to low revenue. There was rampant theft in the system through ‘hooking’ and bypassing or slowing of meters, often with the connivance or even active participation of the SEB employees. This resulted in the so called high T&D losses. The reported T&D losses were kept low by inflating the consumption of consumers with no meters or ‘defective’ meters. These consumers and SEB employees benefited from the system even though the system as a whole was constantly deteriorating. The entire system was propped up by subsidies from the government and cross-subsidy by industrial consumers. However, this arrangement is no longer sustainable because the government is running out of funds to pay subsidies and industrial consumers are finding the cross-subsidizing tariffs unsustainable in a competitive environment. Many have resorted to captive power. The reform process has two main components to improve the efficiency of the distribution system. The first is the privatization of the distribution companies. This is expected to lead to better incentives for efficiency and an overall commercial approach to management. The second component is tariff setting by an independent regulatory agency through a transparent process. Tariffs set by a regulated agency are expected to ensure that the distribution companies will cover their costs and earn a reasonable return at efficient levels of operation. The current problem with the Orissa power sector reforms is that the expectations of efficiency improvements have not been met and T&D losses continue to be high. However, tariffs based on low T&D loss benchmarks are inadequate for the distribution companies to cover their costs and earn a return. The regulators blame the DISTCOs for the lack of improvement in efficiency. The DISTCOs blame the regulators for setting unrealistic efficiency targets
262 India Infrastructure Report 2002 and tariffs which are grossly inadequate to cover costs. The government, in turn, blames them both. The net result of the stand-off is that GRIDCO is financially bankrupt and the DISTCOs are facing serious liquidity problems. The resulting losses imply that the DISTCOs are not able to pay GRIDCO for bulk supply, which in turn is not able to pay the generators resulting in a chain of payables and receivables. Losses have to be ultimately financed, but the regulator does not allow the interest on such loans to be passed through to tariffs. The regulator takes the stand that the losses would not have occurred in the first place if only GRIDCO and DISTCOs were able to achieve the efficiency targets. The DISTCOs in turn claim that given the SEB legacy that they have inherited, including the manpower, they are unable to accelerate the pace of improvement in efficiency. The power sector problems reached such a pass that AES, the private sector partner for CESCO (Central Electricity Supply Company), the distribution company that serves Bhubaneshwar and Cuttak, threatened to withdraw from the company. The OERC, fearing an emergency situation, has stripped AES of the management control of CESCO and vested it with two government nominated officials from the Indian Administrative Service. Here, we analyse the major reasons for the stalemate in the power sector reform process in Orissa. The only way to put the reforms back on track is for the OERC, GRIDCO, DISTCOs, and the government of Orissa to arrive at an integrated plan for the sector which provides for realistic multi-year tariff increases, T&D loss reductions, and subsidies during the transition period. The plan should also identify the duties and responsibility of each party in ensuring that the plan is successful. This section provides the basic elements of such a plan. We first analyse the OERC’s approach to setting tariffs, especially the treatment of T&D losses and reduction of cross-subsidies. Then, we outline the privatization processs of DISTCOs, a key feature of the reform process. Next, we assess the current financial position of the privatized DISTCOs and look at the steps taken by them to improve their financial and operating performance. Then, we assess GRIDCO’s financial performance and the attempts at financial restructuring. Next, we look at the manner in which the government has gained from the reform process
but has failed to discharge its obligations, and conclude with some suggestions for setting the reform process back on track.
TARIFF SETTING
AND
T&D LOSSES
So far the OERC has issued four tariff orders incorporating the retail tariff increase shown in Table 9.3.1. The first two increases were made by the erstwhile OSEB; the remaining four by the OERC. The tariff increases in Table 9.3.1 are average tariff increases. There has been an attempt to rebalance tariffs and reduce cross-subsidies so that tariff increases have been much higher for the domestic sector and lower for the industrial sector. In the course of the four tariff orders the total increase in domestic and commercial tariff has been approximately 50–60 per cent as against an increase of about 10 per cent for large industry. However, the domestic sector continues to be subsidized quite significantly by the industrial sector. According to the OERC, during 2000–01 the per unit subsidy received by domestic consumers was 182 paise as against a per unit subsidy payment of 172 paise by large industry. In terms of total subsidy, the domestic sector received a subsidy of about Rs 445 crores during 2000–01 mainly from the industrial sector. In fixing tariffs the major problem that the OERC has had to contend with is the determination of T&D losses (See Box 9.3.1). In its first tariff filing in 1996, GRIDCO projected T&D losses of 42 per cent for 1997–8 as against the current level of 47 per cent. However, the OERC did not accept GRIDCO’s estimate and instead appears to have based its decision regarding allowable T&D losses on the targets established by the World Bank during the initial processing of the loan. In early 1996, OSEB’s system losses were estimated at 43 per cent for the financial year (FY) 1996 and from that base GRIDCO’s system loss reduction targets were fixed. Based on this, the OERC decided to allow T&D losses of 35 per cent for 1997–8, consisting of 20 per cent technical losses and 15 per cent nontechnical losses. During the project implementation, with the availability of detailed information and data, the 1996 base figure for OSEB’s last year of operation was raised
Table 9.3.1 Retail Tariff Increases Base year 1994–5 (in per cent)
1995–6 (5.11.1995) 17.5
1996–7 (21.5.1996) 17.0
1997–8 (1.4.1997)
1998–9 (1.12.1998)
10.33
Source: An Overview of GRIDCO; 1 April 1996 to 31 March 2001, p. 114.
9.3
1999–2000 (1.2.2000)
2000–01 (1.2.2001)
4.0
10.2
Framework for the Energy Sector
263
Box 9.3.1 Measurement of T&D Losses There are some definitional ambiguities about the T&D loss measure. The three relevant figures are input (in units), billing or sales (in units and Rs), and collection (in Rs). T&D loss can be defined as the difference between inputs and billing in units and expressed as a percentage of the input, that is: billing in units 1− input in units In setting tariffs, this is the approach used by the OERC to determine the input into the system given the sales forecast. Tariffs are then fixed so as to recover the cost of power input as well as other costs. Given that only about 50 per cent of LT consumers have functioning meters a significant proportion of the billed units are estimates and not actual supply. Thus the T&D loss measure is only an estimate. This measure of loss effectively assumes that collection is 100 per cent of billings. In the case of Orissa this is not true, especially for LT consumers. In order to account for this a different measure can be used. billing in units collection in Rupees 1− × billing in Rupees input in units The expression in brackets can be referred to as 'the effective percentage of input units to the system realized as cash at the average tariff realized.' This measure of T&D loss may be more relevant for measuring the efficiency of DISTCOs given the less than 100 per cent collection of billings.
to 52–53 per cent. The figure for GRIDCO’s first year of operation was raised to 50.5 per cent, and the 35 per cent figure was revised to 47 per cent. However, as shown in Table 9.3.2, in subsequent tariff orders the OERC persisted with the 35 per cent T&D loss benchmark even though the licensees were demanding higher loss provisions. In the tariff order for 1998–9 the OERC offered ‘to treat the amount of difference between the revenue requirement calculated on the basis of GRIDCO’s estimated T&D loss of 41 per cent and the revenue requirement calculated on the basis of the reasonable level of 35 per cent as has been determined by us as special category of capital’. This amount with the RBI rate of interest would be considered Table 9.3.2 T&D Losses: Proposed and Approved Year
1997–8 1998–9 1999–00 Transmission Distribution 2000–01 Transmission Distribution
T&D Losses Proposed
Approved
42 41
35 35 35 4 31
5.3 40(BSES) 47(CESCO) 4 38(BSES) 42.66(CESCO)
Source: OERC various Tariff Orders.
3.7 31.46
for inclusion in the revenue requirement for tariff purposes only when GRIDCO or its successor licensees produced evidence of having reduced T&D losses to the level of 35 per cent. Additionally, if GRIDCO, or its successor licensees, was able to reduce the T&D loss below 35 per cent, the OERC would consider increasing the return from the existing level by one percentage point for every percentage decrease in the T&D losses. There are proposals to expand this programme through franchisees. These proposals are still being developed. However, this incentive has not been of much use since, according to the licensees, they are far from achieving the 35 per cent target. Meanwhile they have accumulated substantial financial losses because of the difference between the allowed and actual losses. The licensees in their tariff applications have requested the OERC to permit special appropriation to cover previous losses, arising on account of their inability to achieve the T&D loss targets set by the OERC. As per the Sixth Schedule, past losses can be financed by way of special appropriation in any year to the extent permitted by the Commission. However, the OERC concluded, ‘We are of the opinion that only where surplus of income over expenditure is available the licensee can request and commission may decide the extent to which surplus can be appropriated towards past losses in calculation of clear profit’. According to the OERC, since there was no surplus, there was no question of appropriating any amount for past losses. The OERC has also not allowed interest on bonds issued to fund losses. According to the OERC, the losses were the result of inefficiencies, the cost of which could not be passed on to consumers.
264 India Infrastructure Report 2002
PRIVATIZATION
OF
DISTRIBUTION
The key element of the reform process is the privatization of distribution. The purpose of distribution privatization is to create the right incentives and motivation for improvements in efficiency. Privatization also insulates the organization from political interference and interventions and allows it to function in a commercial manner. Overall, privatization is expected to achieve improvement in work culture, something considered difficult under a government set-up. The initial plan was to privatize the distribution system gradually, area by area, by the end of the year 2000. The entire distribution system in Orissa is divided into 10 circles, which are further sub-divided into 43 divisions. The 10 circles were grouped into four zones, each of which was incorporated as a distribution company—CESCO, WESCO (Western Electricity Supply Company), NESCO (Northern Electricity Supply Company), and SOUTHCO (Southern Electricity Supply Company). The distribution company areas were selected so that they were reasonably balanced in terms of consumer mix and future growth potential.
Distribution Operation Agreement The main obstacle to quick privatization was the lack of detailed and accurate information about the assets and liabilities of each circle and the fact that the new regulatory commission was not yet in place. Given the uncertainties about the regulatory commission and the lack of detailed information, any long-term transaction with the private sector would require the Orissa government to give such guarantees and assurances to the private sector that there would be very little transfer of risk. It was, therefore, decided to initiate the distribution privatization process through a short-term distribution operations agreement (DOA) that terminated at a fixed date. By this time it was expected that the regulatory commission would be in place and would also have had an initial track record. The bulk of investments in system rehabilitation and reinforcement would be made by GRIDCO and the private company would be invited to help manage and execute GRIDCO’s capital investment programme. It would be expected to provide its own staff for key management posts, the rest being seconded by GRIDCO to work for and be managed and paid by the company. The subsequent long-term lease or sale contracts would shift the investment obligation from GRIDCO to the new companies. Staff would similarly be transferred from GRIDCO and become regular employees of the companies. After several rounds of technical offers, price bids, and negotiations, DOA was negotiated for a period of three
years starting in October 1996 with the BSES. However, the DOA had to be terminated prematurely, only 6 months later in April 1997, due to non-performance as per the provisions of the contract. According to GRIDCO, the main reason for terminating the contract was that the BSES had consistently failed to perform. Examples of nonperformance included failure to produce revenue information, to assemble basic records for billing purposes, and to produce maintenance plans. According to BSES, the DOA had certain major drawbacks. Employees working in the central zone were not on deputation to the BSES and hence were not accountable to it. They belonged to GRIDCO and hence did not feel responsible to BSES. Before entering into the contract BSES tried to get officers and staff on deputation. But the Government of Orissa’s Tribunal ruled that personnel would not be placed on deputation and that they would remain employees of the government alone. In terms of investment, about 50 per cent of the meters in the central zone needed to be replaced. However, BSES was not willing to invest in improving the technical infrastructure since it only had management control but no ownership rights. The main lesson from this experience was that privatization, to be meaningful in the Orissa circumstances, had to ensure that the private sector had operational control and the responsibility for financing and implementing system investments. This would be possible only if the private sector had a majority equity stake (at least 51 per cent) in the company so as to ensure proper management and operational control, as well as the authority to make critical decisions. These features, taken together, would mean that the distribution businesses would be at least joint ventures (if less than 100 per cent of the shares were divested), in which the requisite authority and control were in the hands of the private sector, as a part of a well structured shareholders agreement.
Privatization The privatization process involved the sale of 51 per cent of the equity in each of the distribution companies. Up to 10 per cent of the shares in each distribution company was made available out of GRIDCO’s retained stake for the benefit of employees of the particular distribution company through an appropriate scheme. The Request for Proposal (RFP) imposed a condition that no purchaser could take more than two DISTCOs. This was to enable some degree of ‘yardstick’ competition and also provide for diversification across buyers. The Government of Orissa issued a notification in November 1998 called the Orissa Electricity Reform (Transfer of Assets, Liabilities, Proceedings and Personnel of GRIDCO to Distribution Companies) Rules, 1998.
Framework for the Energy Sector
265
Table 9.3.3 Financial Performance of DISTCOs (2000–1) Rs Crores Billing Collection Power Purchase Cost Gross Margin Billings Collections Other expenditure as approved by the regulator
CESCO
NESCO
WESCO
SOUTHCO
Total
584 437 526
365 276 352
472 364 367
235 190 194
1656 1267 1439
58 (89) 180
13 (76) 113
105 (3) 127
41 (4) 102
217 (172) 522
Source: An Overview of GRIDCO (2001).
The notification contained the distribution of assets and liabilities among the four DISTCOs and GRIDCO. In this distribution, GRIDCO retained 66 per cent of the share capital, 76 per cent of the loans, 60 per cent of the fixed assets, and all the accumulated losses. The notification stated that the terms and conditions of the service applicable to the GRIDCO employees being transferred to the DISTCOs ‘shall not in any way be less favourable than or inferior to those applicable to them immediately before the Appointed Date’. The privatization process had to be co-ordinated with obtaining a tariff order from the OERC since this would be an important input in the financial bids. GRIDCO made a tariff application to the OERC on August 1998. GRIDCO had previously incorporated four wholly owned subsidiaries to take over the business of distribution and retail supply; and the subsidiaries had applied to the OERC for licences. Hence, the tariff application included a proposed bulk supply tariff for transactions between the transmission and bulk supply licensee and the distribution and retail supply licensees. The tariff application, a public document, was available to the bidders for their review. The tariff order was issued by the OERC in November 1998. The major feature of the tariff order was that the OERC continued with a 35 per cent T&D loss figure against GRIDCO’s proposed figure of 41 per cent. The response to the privatization process was quite poor. Statements of Qualification (SOQs) were received from 12
Indian and foreign companies/consortia. However, finally only three bids were received—from BSES Limited (BSES), the consortium of Tata Electric Companies and Viridian Group (Tata/Viridian) and the consortium of Singapore Power and Grasim Industries Limited (‘Singapore Power’). The bids submitted by BSES were the highest for each of WESCO, NESCO, and SOUTHCO. There was a rebid for CESCO which did not yield any satisfactory outcome. GRIDCO then took up negotiations with a consortium of AES and Jyoti Structures, and finalized the share transfer. AES, which also owned 49 per cent of OPGC and was negotiating a PPA (power purchase agreement) for a new IPP, obtained OERC approval for entering the distribution business, in addition to its generation interests. Frontier Economics, as a part of its evaluation of the state power sector reform, carried out an Investor Survey to assess the reasons for lack of participation by several investors who had evinced interest in the early stages of the privatization process. The main reasons cited were: • High level of losses and collection risk: Poor condition of system and difficulty of collection for private company without political support. • Problems of inherited staff: Difficulties of imposing own management and bringing in own employees. • Size of system could not support the estimate of fixed costs (for example, 3–4 expatriate employees in India). Investment is too small, and there is a lack of paying capacity among consumers.
Table 9.3.4 Details of Collection of DISTCOs All DISTCOs
LT (low tension) HT (high tension) EHT (extra high tension) Total Input (million units)
1998–9 (Rs Crores)
1999–00 (Rs Crores)
Growth (%)
2000–01 (Rs Crores)
Growth (%)
253 327 528 1108
302 326 470 1097
19 0 –11 –1
380 383 503 1266
26 17 7 15
10,056
9989
–1
10,857
9
266 India Infrastructure Report 2002 Table 9.3.5 Distribution Losses 2000–01 Input (MU) Billing (MU) Billing/Input (%) Billing (Rs crores) Collection (Rs crores) Collection/Billing (%)
10,856 6080 56 1656 1267 77
• Tariff uncertainty: The final tariff decision was based on giving all participants a little, resulting in tariffs that are not high enough. There was also a lack of guarantees that the tariff will reward investment and efficiency gains. It was never made clear what investment would be allowed to enter the asset base and be recovered. • Poor information from the government of Orissa and GRIDCO. • Returns too low for risk.
PERFORMANCE
OF
PRIVATIZED DISTCOS
At the end of almost two years of privatization, the DISTCOs are showing some improvements in billings and collections. However, they continue to show financial losses. This is mainly because of their inability to meet the T&D loss targets set by the OERC. Currently the losses are being financed by stretching their payables to GRIDCO, which creates further problems for GRIDCO. In the first year of operations—1999–2000—the DISTCOs suffered significant losses. For all the four DISTCOs together, the Gross Margin47 in terms of billings was a positive Rs 141 crores, but a negative Rs 222 crores in terms of collections. In other words, collections were insufficient to cover their power purchase bills by Rs 222 crores. As shown in Table 9.3.3 there was some improvement during 2000–1 but collections were still inadequate to cover even power purchase costs. Table 9.3.4 summarizes the collection performance of the privatized DISTCOs. During 1999–00 while LT collections increased by about 19 per cent, EHT collection was down by 11 per cent resulting in an overall marginal decline in collection of 1 per cent. During 2000–1 the growth in LT collection was much higher at 26 per cent. Thus there is a definite improvement in performance during 2000–1 and it needs to be seen whether it can be sustained during 2001–2. 47
Gross Margin is the difference between billing or collection and the power purchase cost.
1999–00 9887 5679 57 1493 1097 73
1998–9
1997–8
10,056 5431 54 1336 1108 83
9646 5571 58 1375 1112 81
The main reason for the continuing losses is that distribution losses continue to be significantly higher than the benchmark set by the OERC for tariffs. As shown in Table 9.3.5, the billing to input ratio is about 56 per cent, implying that almost 44 per cent of the input power is not billed. While some of this represents technical losses the bulk of it represents theft. In addition, the collection to billing is only about 77 per cent as against the 100 per cent collection assumed by the OERC. The problem of T&D loses is likely to be difficult to resolve because domestic consumers are predominantly rural. Almost two-third of domestic consumption is rural and the collection to billing ratio for rural consumers is only 40 per cent as compared to 63 per cent for urban consumers.
Progress on Metering While there has been improvement in metering, almost half the consumers have either defective meters or no meters at all. It is for this reason that billing numbers tend to be unreliable and greater reliance has to be placed on collection figures. There is a major problem of defective meters with almost 35 per cent of installed meters being defective. According to the distribution companies the pace of metering has been slow, partly because of World Bank procedures for procurement of meters. This is being taken care of and the pace of metering is likely to increase in the coming months.
Initiatives for Improving Efficiency The DISTCOs have taken several initiatives to improve collection and reduce losses. The most important of these has been changing the work culture to introduce a commercial orientation among the employees. This continues to be the most difficult challenge faced by the distribution companies. Various initiatives have been taken to incentivize improvement in performance, ranging from monetary incentives to motivation through empowerment. CESCO has taken some bold steps, including openmarket recruitment for professionals (about 50) into their
Framework for the Energy Sector ‘business cadre’, assigning accountability and powers to managers and workers, and taking strong disciplinary measures for non-compliance. These measures have met with resistance among employees. Not many of the existing employees have exercised the option to join the alternate cadre which provides higher remuneration but no government job security. In the case of BSES companies only four executives have been brought from outside. The attempt is to manage the process by ‘reforming’ the existing employees. Other steps to improve efficiency include: • Disconnections to force consumers to pay their bills. • Bijli Adalats and consumer camps to provide speedy redressal of complaints • Establishment of a vigilance wing to detect theft and malpractice • Administrative action against erring employees • Establishment of customer care centres
Village Committees By far the most significant and innovative activity is the formation of village committees. The village committees have been instrumental in rural metering, billing and collections. There had been instances of village committees being formed during the RIAP (Revenue Improvement Action Plan) implementation in the pre-privatization phase of reforms. After privatization, WESCO initiated the process with the help of faculty from the local Xavier Institute of Management (XIM). This has now spread to the other BSES companies and CESCO has also followed. Village committees address the inherent difficulties in serving widely dispersed rural loads. The village committee is a committee of consumers from the village with participation of field staff of the utility with the objective of facilitating a single point contact between the utility and the consumers of the village. In the first phase the village committee facilitates interaction, grievance redressal, bill distribution, metering, and cash collection. A pilot project was launched in August 1999 with 100 villages in WESCO. There was substantial improvement in collections (almost 100 per cent although on a low base) and addition of new consumers for the utility. After this success the scheme was extended to an additional 3000 villages in the NESCO and WESCO areas. Of these in 1000 villages the second phase of the initiative has been launched whereby greater accountability is proposed to be transferred to the committee by making them aware of the input at transformer level and actual billing being achieved from the village. All the consumers in the village are proposed to be metered in this phase. In the third phase it is expected that energy
267
accounting will be entirely transferred to the village committee, whereby they will perform like a co-operative, that is, they will be billed based on transformer readings and the entire responsibility for collections will be transferred to them. At that stage the utility may approach the OERC for a separate tariff for these consumers. NESCO and WESCO have contracted out the process of formation and initial management of the process to Xavier Institute of Management (XIM), Bhubaneshwar, with participation from the utility. CESCO has taken the approach of introducing the same process through its own staff to increase ownership and ensure sustainability. Village committees are operational in more than 100 villages in the CESCO area now. According to the BSES, the microprivatization initiative has provided significant benefits. For example, in WESCO during 2000–01 while collection increased by only about 9 per cent in the non-project areas it increased by about 70 per cent in the areas in which the microprivatization project has been implemented. There are proposals to expand this programme through franchisees. These proposals are still being developed.
GRIDCO’S FINANCIAL PERFORMANCE During the three year period, from the corporatization of the OSEB into GRIDCO (1996–7) to the privatization of the DISTCOs (April 1999), GRIDCO suffered a total loss of Rs 765 crores before interest and Rs 1019 crores after interest. The bulk of the losses before interest can be explained by GRIDCO’s inability to achieve the T&D loss target of 35 per cent set by the OERC. GRIDCO’s actual T&D losses during this period was about 45 per cent. If GRIDCO had been able to achieve the target of 35 per cent it would have resulted in additional revenue of approximately Rs 720 crores, with very little additional costs. The additional revenue would have almost wiped out the losses before interest. With no losses, the interest burden would also have been much lower since the bulk of the additional loans were taken to finance losses and were not allowed to be passed through to tariffs by the OERC. The main reason for the poor performance has been the lack of a system of incentives and disincentives for ensuring managerial accountability for energy sale and revenue collection. Apart from this, other important reasons for inadequate progress were the lack of a management information system and weak commercial and accounting systems in the erstwhile Orissa State Electricity Board. After unbundling and corporatization, a large amount of advisory and implementation support has been provided to GRIDCO. In order to address the weaknesses in the
268 India Infrastructure Report 2002 commercial system, GRIDCO was provided support for developing a codified set of commercial practices and for implementation of the same at the field level. Manuals for commercial practices like new consumer registration, billing, revenue collection, credit control, temporary connections, regularization, consumer complaint redressal, etc. were prepared and were put into implementation in most areas within GRIDCO. A detailed management information system was prepared to address the lack of availability of management information of good quality for the information and action of the top management. However, the operational systems and mechanisms have been difficult to change under its public sector status, and despite the large quantum of support provided by DFID’UK the improvements were minimal and the financial situation deteriorated rapidly. With the separation of the DISTCOs in 1999, GRIDCO has shown profits before interest of Rs 80 crores in 1999– 2000 and Rs 23 crores in 2000–1. However, as a result of the distribution losses prior to privatization of DISTCOs and the continuing failure of privatized DISTCOs to make timely payments for power purchases, GRIDCO’s financial situation continues to worsen. As on 1 April 2001 GRIDCO owed generators about Rs 1240 crores and total borrowings stood at Rs 2750 crores.
GRIDCO’s Financial Restructuring Plan (FRP) The government of Orissa submitted a financial restructuring plan (FRP) for GRIDCO in October 1999 to the Ministry of Power, GOI, soon after the process for privatizing the DISTCOs had been completed. The plan has been under discussion since then while the condition of GRIDCO has continued to worsen. Recently the plan was submitted to the OERC for approval. The FRP presented to the OERC by GRIDCO essentially involves conversion of loans from financial institutions and payables to generators into tax-free bonds: • Rs 600 crores dues to the Central generating stations to be securitized by issue of tax free bonds. • Rs 800 crores of REC and PFC loans to be discharged by issuing tax free bonds with state government guarantees. • Rs 560 crores dues to the state owned generating companies (OHPC and OPGC) would be partly managed through the issue of tax free bonds with state government guarantee (Rs 360 crores) and the balance would be paid out of soft loans receivable from the World Bank. • Rs 200 crores loan from the World Bank for managing deficits until the turnaround of GRIDCO. The OERC held hearings on 8 November 2000 during which various affected parties expressed the following views:
• OPGC stated that the issue of bonds by GRIDCO to OPGC was not an attractive proposition. It also maintained that the waiver of DPS (Delayed Payment Surcharge) and penal charges on non-payment of dues was not an option it was considering. • OHPC complained that despite being the cheapest power supplier, GRIDCO was honouring liabilities of other suppliers like NTPC and OPGC prior to paying the OHPC dues. Since there was neither any inflow of funds from GRIDCO nor any commitments of early settlement of dues, the FRP would put additional burden on the finances of the OHPC. • SOUTHCO disputed the loss reduction figures quoted by GRIDCO, and offered the following loss reduction targets: 3 per cent per annum for the first 3 years and 2 per cent per annum in the fourth and fifth years. (The FRP assumed a decrease in distribution loss percentage of 5 per annum per annum for the first 3 years and 2.5 per cent for the subsequent years, till the overall loss percentage reaches 20 per cent). WESCO and NESCO held similar views on the FRP. • CESCO held that the loss reduction programme envisaged by GRIDCO was too ambitious. It contended that if all the assumptions taken by GRIDCO were fulfilled, CESCO might be able to achieve a loss reduction of 2.5 per cent per annum. • The government of Orissa held the view that given the current financial situation of Orissa, it would be extremely difficult for it to infuse any equity into GRIDCO or to intervene financially. According to the OERC the financial insolvency of OERC was largely the result of inefficiencies in operations, especially its inability to meet the T&D loss targets. The Commission concluded by giving no commitment on future tariffs and rejected the proposal for a multi-year tariff. The OERC Order on the FRP clearly shows that the FRP has serious problems, and in its current form, is not sustainable. The biggest problem is with respect to T&D losses. Unless this problem is sorted out between the regulator and the DISTCOs, the FRP will not provide a long-term solution.
ROLE
OF
GOVERNMENT
IN THE
REFORM PROCESS
The Orissa government played a pioneering role in accepting and initiating the power sector reform process. However, it appears to have decided that with the initiation of the reform process it has no further financial responsibilities towards the sector. This is not possible given that the sector is basically non-viable, given the high level of T&D losses. The Orissa government has benefited financially from the reform process in the following ways:
Framework for the Energy Sector • At the time of corporatization of GRIDCO, the OSEB assets were revalued by Rs 1194 crores. In return the government adjusted subsidies and electricity charges payable to GRIDCO of Rs 340 crores. • In 1998 the state government sold 49 per cent stake in OPGC to AES for Rs 603 crores. Out of this amount Rs 120 crores was lent to GRIDCO, and OPGC was allowed to retain Rs 98 crores. The balance Rs 385 crores was available to the government. A precondition for the sale was that GRIDCO should liquidate their dues to the OPGC. To comply with this requirement, GRIDCO raised Rs 300 crore from the market at a coupon rate of 14.5 per cent and paid off OPGC’s dues. • There was an upvaluation of Rs 500 crore when the Hydel assets of the state government and OSEB were transferred to the OHPC. As a result of the upvaluation, the cost of hydel energy went up from 20.52 paise per unit to 38 paise per unit. One of the results of this upvaluation was that the government was able to raise some funds for the cash starved Upper Indravati hydel project. It also enabled the OHPC to consistently show profits. While the government has benefited financially from the reform process it has not fulfilled its obligations to the sector in several instances. • Government subsidy to the sector has stopped since the reform process was initiated. The privatized DISTCOs have submitted subsidy claims, approved by the OERC, for rural electricification amounting to Rs 23.23 crores. The government is yet to disburse this amount. This is particularly embarrassing since the government attaches considerable importance to rural electrification and is committed to providing capital subsidy for the same. • State government departments and Public Sector Undertakings (PSUs) owed the DISTCOs Rs 144.8 crores as on 31 March 2001 for power consumed after 1 April 1999. • The state government has defaulted in passing on to GRIDCO and the distribution companies their part of the World Bank loans of US$ 350 million. As on 30 April 2001, GRIDCO was owed Rs 68 crores. Several times in the past, the World Bank had expressed unhappiness at the state government’s withholding of the World Bank, funds which were meant to be invested in the transmission and distribution system and would result in improvement in the quality of the electric supply. When matters did not improve, the Bank suspended the flow of World Bank funds to the state. • The OERC has been functioning with only two members for a long period of time and the government has been unable to appoint a third member. • The Orissa Act provides for a mechanism whereby funding of all reasonable expenditures of the OERC should
269
not be subject to the discretion of the government. The OERC is required to submit a statement of estimated expenditure to the government for the ensuing year by December of the preceding year. This is then tabled in the state legislature. The expenditure is funded out of the consolidated fund of the state and does not require approval of the assembly by voting. This is a privilege extended only to institutions where the government feels it is crucial to avoid political influence, such as the state judiciary and the Public Services Commission. In turn, the licence charges stipulated by the OERC for the transmission and distribution licensees are passed on to the government. The government budget for the OERC has decreased from Rs 150 lakhs in 1998–9 to Rs 85 lakhs for 2001–2, and it has not been getting the amount of funding requested. The government receives Rs 2.5 crores as licence fees from the four distribution companies and GRIDCO at the rate of Rs 50 lakhs per licence. • The DISTCOs have persistently complained about the lack of support from the law and order machinery for enforcing disconnections and preventing theft. At the minimum, the state government needs to fulfill the following obligations: subsidy for rural electrification, clearing government departments and PSU receivables, passing through World Bank loan amounts, and giving full support to the OERC. In addition, the government is ultimately responsible for the long term viability and success of the sector and needs to take suitable steps to put the reform process back on track. This needs to go beyond the financial restructuring of GRIDCO.
GETTING BACK
ON
TRACK
At the end of over five years of the reform process in Orissa, the transmission and distribution entities are financially insolvent. The privatized distribution companies do not have adequate cash flows to even pay salaries. AES, the private partner in CESCO, has effectively withdrawn from management of the company following the OERC Order which vests management control of CESCO in a government nominee. The main source of the problems is the gap between the efficiency and cost parameters underlying tariffs and those achievable by the licensees, and the government’s inability to bridge the resulting cash deficits. Resolution of this problem is crucial to the viability of the state power sector. It is, therefore, a prerequisite for taking up proposals for the financial restructuring of GRIDCO. Stakeholders will be willing to participate in a financial restructuring exercise only if there is a reasonable assurance of financial viability.
270 India Infrastructure Report 2002 Not much seems to have been achieved in terms of efficiency improvements during the period 1996–9, that is, the period from corporatization and unbundling to the privatization of the distribution companies. While considerable work may have been done by the consultants in terms of developing systems and procedures and building databases it does not seem to have translated into improvements in efficiency. This is not surprising since changes in systems and procedures can achieve little if not accompanied by change in work culture and motivation. Post privatization, there is a definite attempt to change the work culture and employee motivation. Various initiatives have been taken by the distribution companies to provide incentives for improvement in performance ranging from monetary incentives to motivation through empowerment. These measures have met with resistance from vested interests among employees and other interest groups. In tariff setting the most critical decision of the OERC has been a benchmark T&D loss level of 35 per cent. This was reduced to 34 per cent in the last tariff order. The 35 per cent benchmark was based on World Bank estimates and targets. However, even after the World Bank had accepted that its earlier loss figures were grossly underestimated the commission has refused to alter its benchmark. The commission has not accepted the DISTCOs’ T&D loss figures because it is not backed by adequate data. However, adequate data is unlikely to be available, given the current low levels of metering. In view of the financial losses of the DISTCOs it is clear that the actual T&D losses must be certainly higher than the benchmark established by the OERC. This gap between actual T&D loss and that allowed by OERC in the tariffs were apparent at the time of privatization. In the filing for the tariff order passed just prior to privatization GRIDCO had claimed an actual T&D loss figure of 46.6 per cent and a projected figure of 41 per cent. The OERC based its tariff order on a T&D loss of 35 per cent. However, during the privatization process there does not appear to have been any discussions about the magnitude and financing of the inevitable cash deficits. Financing concerns were expressed only at the time of the CESCO re-bid. AES asked for and received working capital financing for taking over CESCO. Overall, it appears that the privatization process was pushed through in haste without fully thinking through the implications. On the other hand, if these questions were seriously pursued it is quite likely that the deal would not have been consummated in its existing form. The only way out of the current imbroglio is to carry out the planning exercise that was not done at the time of privatization. This will need to be a three-way exercise among the licensees, OERC, and the government of Orissa. This exercise rests on the following principles:
1. Multi-year tariffs to enable licensees to plan efficiency improvements and investments: The starting point of the exercise is the acceptance of the principle of multi-year tariffs. The FRP submitted by GRIDCO proposed multiyear tariffs. The World Bank has also suggested that the OERC move to a system of multi-year tariffs. However, in its Order on the FRP submitted by GRIDCO, the OERC has rejected the idea of a multi-year tariff. According to the Order, ‘We are not in a position to give clearance and commitment for future tariffs, as these will be dealt separately on a year-to-year basis in accordance with the OER Act.’ Multi-year tariffs are essential to enable the various players in the Orissa power sector to plan for the future with some degree of certainty. If necessary, the government could consider amending the OER Act to enable the OERC to issue multi-year tariffs. Multi-year tariffs will also involve pass-through of power purchase costs. So far the OERC has only allowed passthrough of fuel costs but not other costs. In a multi-year tariff these other costs which are beyond the control of the licensees will also have to be passed through. 2. Tariffs to be based on standard costs and efficiency parameters: The multi-year tariffs should be based on standard costs and efficiency parameters. The standards are not hypothetical but based on the actual technologies in use. These standard costs and efficiency parameters would serve as targets which the DISTCOs would be required to achieve in a time bound manner. These standards will, therefore, be set by the OERC after negotiations with the DISTCOs and using inputs from independent experts. Ultimately, the DISTCOs will have to agree to the achievement of these targets in a phased manner. This manner of tariff setting will be a departure from the current practice in which tariffs are set using a mixture of standard and actual costs and efficiency. For example, the critical 35 per cent T&D loss figure is neither a standard nor actual. The current tariff burdens paying consumers with the inefficiencies and thefts of the system. This is not only unjust but can also lead to paying consumers becoming non-paying consumers. Such tariffs will also encourage industrial consumers to exit from the system and resort to captive supply. The tariffs should also provide for a phased reduction and ultimate elimination of cross-subsidies. In case the government wants to provide subsidies to certain segments, these should be clearly targeted and paid for separately by the government. Subsidies to target segments could be achieved through life-line tariffs limited to consumers with a certain basic level of consumption. 3. Licencees will have to commit to a binding multiyear programme to achieve the standard costs and efficiency parameters: Licensees will have to provide a multi-year schedule of progress towards achieving the standards on
Framework for the Energy Sector which the tariffs are based. For this schedule to be meaningful the efficiency parameters should be measurable. This requires a level of metering, not necessarily 100 per cent, which will provide reasonably reliable T&D loss data. This schedule will have to be negotiated between the licensees and the government since the government will be paying for the transition. 4. Need for subsidy during the transition phase to bridge the gap between targets and actuals: Given the multi-year tariff schedule and the multi-year programme of efficiency improvements the cash deficits of the licensees can be worked out. The cash deficit represents the cost of the gap between standard and actual costs and efficiency parameters. This cash deficit will have to be financed by the government. Other alternatives such as the creation of regulatory assets are unlikely to be credible. The government could levy a special temporary tax to finance this subsidy. The final outcome of this exercise will be a tariff plan from the OERC, an efficiency improvement plan from the DISTCOs, and a subsidy commitment from the government which together ensures break-even for the licensees. This will constitute a binding agreement among the OERC, DISTCOs, and the government. An essential element of the plan would be a clear statement of the commitments by the licensees and the government in achieving efficiency targets and ultimately viability of the sector. The job of the OERC would be to enforce these commitments. The Andhra reform process appears to be following some elements of the process described above. The Andhra tariff orders are based not on actual costs but on some standards of efficient costs. For example, the tariff orders provide for an ‘efficiency gain’ of Rs 500 crores. It also provided for an initial government subsidy of Rs 1345
271
crores which was later increased to Rs 1626 crores after the government requested a decrease in tariffs following public and political agitation in response to the original tariff order. In the tariff application for 2001–2, the Andhra Pradesh Electricity Regulatory Commission (APERC) estimated a loss of Rs 1024 crores for the year 2000–1 after taking the government subsidy into account. The APERC proposed to carry forward only an amount of Rs 90 crores of this loss ‘which is on account of factors beyond the reasonable control of the Licensee’, through a special appropriation in the annual revenue requirement for FY 2001–2. The balance uncovered financial loss of Rs 934 crores is to be compensated by the government of Andhra Pradesh (AP) as subvention as per the provisions of Section 27 (1) of the Reform Act. Thus in the AP approach the tariffs are based on some (undefined) efficient costs and the deficit is subsidized by the government. This needs to be accepted openly and be accompanied by a definite programme of loss reduction and phasing out of subsidies. The AP regulator has recently commissioned a study to estimate technical losses in transmission and sub-transmission systems in AP. The Maharashtra Electricity Regulatory Commission (MERC) also appears to be preparing to move to the kind of arrangement proposed above. According to press reports, MERC has commissioned three studies on cost of subsidies in the state power sector, benchmarking performance factors in generation and distribution, and share of electricity cost and power consumption in the state agriculture sector. According to the press report ‘MERC is looking towards manifold usefulness of this study such as providing reasonable performance indicators, which could be imposed as targets to the utilities and facilitate the dialogue between the utilities, the regulator and the consumers.’
9.4 POWER SECTOR REFORMS AND PROPOSED ELECTRICITY BILL Ajay Pandey The need for power sector reforms in India came into focus immediately after the initialization of broader economic reforms starting in 1991. Initially the focus was on attracting private and foreign investments to add generation capacity, as it was thought that the government was not in a position to undertake investments in capacity additions due to budgetary constraints. Over a period of time, the focus shifted to distribution and tariff rationalization once the rapidly deteriorating financial condition of the State Electricity Boards (SEBs; the distributors) was seen as a constraint in attracting investments in the sector. Unbundling, competition, reduction
in thefts and T&D losses, restructuring of SEBs, reduction in subsidies and cross-subsidies, etc. have been some of the themes, that have been in focus since last few years. Meanwhile, private sector in the form of independent power producers (IPPs) has added or is expected to add very limited capacities, the receivables of SEBs to central sector generation and transmission utilities have reached staggering sums necessitating concessions, and the escrowable revenues of SEBs for private investors have become nonexistent. The cost of IPP power is seen as very high (as in the case of Enron) and restructuring of SEB (as in the case of Orissa) is seen as having not yielded
272 India Infrastructure Report 2002 expected results. In this backdrop and given that electricity is on the Concurrent List of the Constitution and therefore requires concerted effort and clear understanding between Central and State governments to move forward, the proposed Electricity Bill pending before the parliament for its approval assumes significance for the reforms in this sector. In order to provide statutory basis to the reforms in the sector, this bill was drafted and finalized last year. The initial draft was prepared by the National Council of Applied Economic Research (NCAER) and given to the Ministry of Power (MoP) last year. Since then, it has been modified by the Ministry of Power and the bill is in waiting for legislative clearance. The primary motivation for this paper is to critically examine the provisions of the proposed bill (which has been posted on MoP web-site on 27 July 2001) in the light of issues that have been identified and discussed in various forums. Another interesting issue related to the proposed bill is that the base draft of this bill was prepared by an independent institution on request. The instances of asking an independent agency or institution to draft a legislative bill are rare in India and a positive development, as an independent institution is less likely to protect the interest of bureaucracy and the political establishment. Since both the drafts are in the public domain, our motivation has also been to look into the key differences and approaches in the two drafts. This is important to the extent it sheds light on the willingness and ability of the bureaucracy and the political establishment to move forward on the sectoral reforms. Below, we first examine the proposed Electricity Bill, 2001 to assess its effectiveness. Then, we compare the proposed bill with the earlier draft prepared by NCAER to critically examine the nature of changes brought about by the MoP and plausible reasons for the same. Finally, we summarize the arguments for and against various provisions of the proposed bill and the assessment of its likely impact on the reforms in the sector.
competition facilitated by unbundling, private participation, and independent regulations. The need for the bill arose as the existing laws—the Electricity Act of 1910 and Electric (Supply) Act of 1948—were too outdated and had vertically integrated operations by state monopolies and licensed monopolies as the basis of electric supply. When the need for an independent regulator was identified with private sector participation, the Electricity Regulatory Commissions Act was enacted in 1998 to enable independent regulations in the sector. With the proposed bill, the new unified law will replace all the three existing laws. In particular, the new law is expected to facilitate the evolution of competition including private participation by reducing regulatory and policy uncertainties, open access to transmission networks for socially optimal generation, independent regulator and tariff setting, removing crosssubsidization for efficient resource allocation and for enabling competition across geographical territories, development of inter-state markets, fixation of rational organizational jurisdictions and operating terms and conditions, penalties for unauthorized use or theft of electricity, etc. While the proposed bill covers some ground on these issues, it leaves a lot more desired as is clear from the following analysis.
DRAFT ELECTRICITY BILL, 2001
In the opening part II of the bill dealing with National Electricity Policy and Plans, Section 3 envisages the evolution of development in the sector through national policies and plans. What would be the basis of implementation of such policies once the sector has considerable private investment is not clear? If the objective is to have competition and let the markets take over, then national plans and policies spelt out by the state bring in potential regulatory risks for the private sector participants. The policy reviews, which may take place from time to time, have been left unconstrained in the bill. The political considerations, as a result, may continue to prevail in any future reviews in addition to the uncertainty, associated with such reviews. Policy and regulatory uncertainty, so enabled by the bill, will act as
With the power sector facing these issues, the draft Electricity Bill, 200148, which has been reportedly cleared recently by the Union cabinet, assumes importance as it is expected to provide a statutory framework for the resolution of some of these issues. It is also expected to enable the ‘second generation’ reforms in the sector, which are critical for efficient operations of the sector through 48 As pointed out earlier, our analysis is based on the draft posted on 27 July 2001 on the Ministry of Power web-site. The bill may undergo substantive changes by the time this Report is published.
Objectives of the Bill: Too Timid The preamble of the proposed bill focuses on development of the industry rather than defining it in terms of consumers’ interests. After all, the development of industry in an infrastructure sector has to commit itself to certain performance standards or has to at least reflect legislative intent to that effect. Section 6 of the bill only mentions that the appropriate governments shall endeavor to supply electricity to all, without any time frame being indicated.
Policy Uncertainty: Enabled through National Electricity Policy and Plans
Framework for the Energy Sector deterrent to any private sector participation, if that is also an objective (though explicitly not in the scope of bill, yet is evident implicitly). If the policy needs a review due to ‘public interest’, it should have been at least constrained by explicitly stating that certain principles not be violated except through legislation. These should have covered the following minimum—open access, moving towards incentive regulations, commitment to competition except in case of transmission and wires business of distribution, protection of private property and consumers, and supporting development of market. The worst part of the bill in this regard is section 79(3) of the Bill (part X, regulatory commission), wherein Central regulator is envisaged to discharge its function under the guidance of the national policy. These are retrograde provisions, which leave everything amenable to reviews by the state and undermine regulatory independence. Section 3 also includes tariff related matters in the scope of National Policy and Plans, which makes the power of the executive arm of the state or the government absolute, should it decide to exercise the same.
Regulatory Independence: Severely Undermined Instead of strengthening the independence of regulators, the proposed bill takes considerable pains to undermine the same in various ways. The selection committee proposed for selecting regulators for the central regulatory body as envisaged under section 78(1) of the bill is full of bureaucrats. Regulatory capture by the producers is a phenomenon widely experienced even in countries where strong tradition of independent regulations exists. In the Indian context, where the resource constraints as well as lack of political will, widespread corruption, and indifferent public opinion are already stumbling blocks for effective independent functioning of the regulators, such a move will ensure that the regulators will not be able to gain the stature and independence from the state and allied political pressures. The bill should have at least provided for representation of consumer interest as well as the judiciary in the selection committee, even though that by itself may not ensure the independent functioning of regulator. The ability retained by the state to influence, through policy, plans including tariff-related issues has already been pointed out. Implicitly, section 62 in Part VII of the bill dealing with tariffs binds the regulator with existing cost-of-service tariff regulations, as the tariff reviews are prescribed at one year frequency and include transmission. This practically rules out nodal pricing for transmission. The bill also envisages regulation of trading margins, that is, the difference between the price at which electricity is bought and sold by a trader. This is probably motivated by concern that in the presence of cross-subsidies, an electricity buyer can easily overcome the burden of cross-
273
subsidies by buying it from a trader. However, since it has not been specifically pointed out in the bill, the provision to regulate the trading margin may be construed as the basis for regulating the return and margins of the trader even beyond the cross-subsidy element in the margin. Bringing the trader within the regulatory ambit without any qualifications may result in regulations of margins per se, which is antithetical to the trader’s role being that of an arbitrageur.49 The dispatches based on contracts similarly violate and preclude pricing based on merit order and short run marginal costs. This also implies that tariffs for generation are going to be determined by cost-plus basis. As pointed out earlier, the cost-of-service (rate-of-return) tariff regulations create severe incentive distortions and more progressive incentive and marginal cost based tariff regulations would be desirable in stages before market based competition and price discovery. By laying down specifics in the bill, any possible moves by independent regulators to evolve tariff regulations to support efficiency have been prematurely aborted. The regulators are also sought to be advised by advisory and joint committees at various levels under section 81, 83, and 87 of the bill, without making it explicit whether such advise is binding or not. Even the role of these committees is unclear, other than fear of too much independence on the part of regulators. In the worst case scenario as a consequence of these committees, the regulatory process rather than being guided by techno-economic principles may come to reflect political realities. These are specific provisions which seek to undermine regulatory independence as the bill in any case gives general powers to the appropriate governments to give directives.
Unbundling and Open Access: Enabled but Left to the States One of the positive features of the bill is enabling open access of transmission and distribution networks for the generators, consumers, and traders through section 42(2) in Part VI of the bill dealing with the supply of electricity. The open access is also facilitated by unbundling of vertically integrated monopolies and having provisions envisaging delicensed generators, transmission, and distribution licensees. There is however no time frame specified within which the state regulators would ensure open access for the consumers. Only the distribution licensees will get open access immediately to buy power from any generator within the region, provided the generator has capacity to contract. While section 42(2) of the bill enables open access, section 45(3) does not insist upon providing break49 Some one who tries to gain from price difference(s) in the market(s).
274 India Infrastructure Report 2002 up of distinct charges for wheeling in billing by licensees, a necessary pre-condition for the eventual move towards open access. Similarly, section 51(3) of the bill does not clearly spell out the need to maintain separate account for the wires business (as opposed to customer servicing business) by the distribution licensees.
Phased Reduction of Cross-subsidies and Direct Subsidization: Positive but Without Commitment on Time Frame The bill correctly links the open access issue with the phased reduction in cross-subsidies, as cross-subsidies are not possible without regulatory co-ordination, particularly for inter-state trades. It also envisages direct subsidization for target consumers by the respective governments and imposes restrictions/discipline on them through section 65 of the bill by making the tariffs decided by the regulator applicable if no subsidy is received in 90 days. Despite a positive undertone, the bill allows for open access with additional surcharge under section 42, if it has not been possible to remove cross-subsidies by then. This provision would ease pressure on the states and regulators to eliminate cross-subsidies.
Generation Delicensed but Misplaced Emphasis on Long-term Contracts Another positive step taken in the bill is delicensing of generation capacities and players as stated in section 7 of the bill. This however, is not going to be very effective in improving the generation efficiency and attracting investments, as it is unclear as to how the sale of power so generated will be accomplished. Section 7 and 10(2) of the bill indicate that it can be sold to any licensee, but the remaining parts of the bill make it abundantly clear that the sale can only take place through contracts with licensees and not through markets. Under section 40(c(i)) of the bill, the transmission licensee has to provide open access, but the bill is silent on from whom and on what basis the distribution licensees can buy power. The bill seem to suggest that the regulator, in tandem with the transmission utility, would determine the dispatch schedules. Under section 32(2), the state load dispatch centers have been made responsible to dispatch in accordance with the contracts entered into. Given the fact that the bill does not envisage any spot market for selling power, any generator capacity will come up after entering into long-term contract(s) with distribution licensee(s). While the dependence of generation capacity on consumption is desirable, the long-term contracts prevent competition by preventing new capacities to come up even when desirable.
Long term contracts need to be designed carefully, otherwise they can affect economic efficiency. For example, if the variable costs and fixed costs are combined then the generation has to be paid for even if the plant is not operational. Alternatively, if the costs are not segregated appropriately in the contract, then the plant may be scheduled or backed down despite being optimal from the point of merit order. Similarly, if the capacity (or obligation to pay fixed costs or capacity charges) is not tradeable, then collective capacities contracted by the licensees might be different from what is socially optimal. Too much emphasis on long-term contracts as opposed to facilitating development of markets is not only misplaced, but the bill may also become a constraint for any development in the future in that direction. This feature of the bill also directly affects the role of electricity traders, distinct licensees envisaged in the bill.
Traders’ Role is Minimal In an open access scenario, the electricity traders could have induced competition even if the distribution licensees remain monopolies50 over their respective geographical regions. Ostensibly, all these preconditions have been facilitated in the bill. However, section 53 of the bill, which stipulates that no trader will buy power meant to be sold to a distribution licensee (presumably resulting from existing arrangements as well as because of any long-term contract) without consent of licensee, dilutes the role of traders considerably.51 It is not clear as to what a trader can do in such a context in which everyone is bound by or is likely to protect itself with long-term contracts. Even if there is surplus power available and open access allowed, why should a licensee give its consent?52 Similarly, under section 79(1d), the regulator is empowered to regulate trading margins for inter-state trading of power. As such, the costof-service based tariff framework is not conducive for having electricity trading as total tariff (fixed costs+ variable costs+ profits) would be lower for any generator with lower historical costs of assets and would always give rise to arbitrage opportunity. Section 79(1d) on trading margins is a recognition of this difficulty and is an attempt to reconcile the same by keeping the provision to regulate the margins. However, regulating the margins is not going to be easy due to incentive problems.53 50 The bill provides for more than one licence in the same geographical area though. 51 The contrary view is that the Act leaves an option with the licensee to decide the optimal mix of energy purchases through long- and short-term contracts. 52 See fn. 51. 53 The contrary view is that since the retail tariffs are regulated by SERCs, competition between the distribution licensee and the traders will ensure that trading margins remain competitive.
Framework for the Energy Sector
Transmission: Requires Strengthening of Independent System Operations Envisaged The bill envisages independent system operations through a clear operational hierarchy of the National Load Dispatch Center (NLDC), Regional Load Dispatch Centers (RLDCs), and State Load Dispatch Centers (SLDCs). This is one more positive feature in the bill in the light of transmission issues in India and for facilitating open access as well as development of markets. However, there are some problem areas in the bill on this aspect as well. Sections 26(2) and 171(2c) of the bill state that the NLDC will have its function prescribed by central government. The central regulator, in order to regulate the sector, may find itself in conflict with the functions handed over to the NLDC by the central government. Given the sensitivity of the NLDC as well as its centrality in the transmission system operations, it would have been better if the management of system was intended on the basis of techno-economic principles such as (i) centralized control for system reliability (grid), (ii) open access, and (iii) maximizing social welfare. Only under exceptional circumstances (and these need to be defined, such as war, disaster etc.), the central government may direct the NLDC to prioritize dispatches otherwise. The hierarchy of load despatch centers (RLDCs and SLDCs) similarly is desirable, as envisaged in sections 26(1) and 29(3), but needs further strengthening. There are conflicts within sections 28(2) and 29(4&5) of the bill because of simultaneous control of the Central Commission and Regional Power Committee envisaged. The Central Commission should have been given over-riding power, otherwise real time efficient functioning of the system can be easily jeopardized. Similarly sections 29(1) and 29(6) of the bill envisage reference to the Central Commission in case any dispute between SLDCs and RLDCs arises.
Transmission Ownership: Lack of Restrictions Consistent with Lack of Commitment to Development of the Market A major omission in the bill is lack of ownership restriction on transmission utilities. It is well known that collusion between transmission and generation or distribution can result in abuse of market power. In case of private transmission licensee, it would have been appropriate not to let them have any ownership interest in generation and distribution or trading. The government ownership could have been the only exception allowed. This omission is, however, critical only if the markets are envisaged upstream or downstream in the sector. Lack of ownership restriction on transmission is otherwise consistent with cost-of-service tariff regulations. It is an issue only if eventually there are plans to move
275
towards competitive markets and price discovery through them. It is evident that all major provisions in the bill do not support evolution of markets in the sector. There are essentially three major problem areas in the bill. The first one relates to competition and markets. In the distribution and supply sub-sector, the bill does not envisage competition except when multiple licenses are issued. This by itself is, however, less constrictive compared to excessive reliance on long-term contracts and disabling emergence of electricity markets, even at the bulk level. The second problem area is the closely related issue of tariffs. Even without markets, traders’ role could have been more effective in case appropriate tariff regulations were enabled. By positively committing (indirectly) on cost-ofservice regulations, such a possibility has been eliminated and existing perverse incentive problems have been preserved. Lastly, the independence of regulators and other key players (load dispatch centers) has been undermined through various provisions.
PROPOSED BILL
AND THE
NCAER DRAFT
The proposed Electricity Bill is based on a draft written by the NCAER on the request of the MoP. The draft bills proposed by NCAER earlier as well as the cleared version of MoP are both available in the public domain. The NCAER, being an independent body, is likely to have been more unconstrained in envisioning the changes required and therefore, the comparison between the two sheds light on the changes brought about by the MoP because of the bureaucratic and political concerns. On the three major defects of the bill, the NCAER draft was indeed much better, as analysed forthwith.
NCAER Draft Facilitated Competition in Supply and Development of Bulk Electricity Markets In the NCAER draft, section 45 of the bill explicitly enabled development of bulk electricity spot market after 2 years and also envisaged evolution of market based tariffs under section 44. Despite envisaging licensed generation in section 3, notwithstanding exemptions for a pure generating company in section 4, it facilitated more competition by fixing a time frame for open access at 3 years for bulk consumers in section 19(2,3) and unbundling of supply from distribution in 5 years under section 27(2). It also explicitly mentions in section 34(1) that the transmission and supply of electricity shall be free from inter-state barriers and restrictions, thereby reducing the uncertainty from lack of coordinated steps taken by different states. Though some leeway for states was left in section 34(3), it was only with the consent of the President of India. Whether such provisions would have been possible
276 India Infrastructure Report 2002 legally given the constitutional status of the subject is suspect, but the provisions reflected the need of reforms in the sector. Similarly, the functions of transmission or load dispatch centers under sections 54(5), 56(3), and 60(3) were not constrained by contractual arrangement as in the bill cleared by the MoP. All these provisions are consistent with the development of markets as well as merit-order dispatch as objectives.
Market Based Tariffs were Envisaged Consistent with Competition and Development of Market The NCAER draft only envisaged tariff regulations for those sub-sectors which remain as natural monopolies, namely transmission and distribution. The other businesses were left for market based tariffs as competition was envisaged. Under section 40(2), it left generation, supply, and spot markets out of the purview of tariff regulations. Under circumstances spelt out in section 42(4), even if such regulations were required the tariffs envisaged were based on price cap regulation. On the subsidies by the government, it envisaged under section 43 that the subsidies are given directly to beneficiaries through a separate agency. This would have been useful in better targeting of the intended beneficiaries, and importantly, in reducing incentive problems related to T&D losses. The provisions in the draft directly follow from the importance given to competition and spot markets. Even if the same is not accepted, the provisions in the MoP draft on tariffs are restrictive and preclude the possibility of any improvement in tariff regulations, as they have moved to the other extreme.
Regulatory Independence and Independent System Operations In the third major area of weakness in the current MoP cleared bill, the regulatory independence has been undermined compared to the NCAER draft. The NCAER draft in the section 83 had more members in the selection committee for the regulators’ appointment, which were independent from the direct control of the MoP and were drawn from the judiciary, RBI, and UPSC etc. Though the general power of government to give directions was there in the draft, the regulators were not specifically constrained by the national policies and plans including tariff related matters, as is the case with the MoP cleared draft. The NLDC in the NCAER draft similarly was envisaged to function in an autonomous manner based on
the Grid Code and certain principles, as laid down by the Central regulator. This has been changed to give powers to the central government to give directions in its functioning. From both the drafts, it is clear that the current thinking of the MoP is distrustful of the role of markets and regulators to be able to steer the direction of change in the sector. The political compulsions related to tariff rationalization (subsidies and cross-subsidies), widespread thefts, and potential volatility in the spot markets are seemingly the cause. Difficulty to absorb these in the presence of an independent regulator makes up for the difficulty in accepting independent regulators.
POWER SECTOR ASSESSMENT
AND
ELECTRICITY BILL, 2001: AN
While the passage of the Electricity Bill, 2001 may pave the way for the unbundling of integrated SEBs and the setting up of independent regulators by the states, it may not bring about any major change in the sector beyond that. The open access and limited competition by the traders though envisaged in the bill, is likely to play a relatively minor role till appropriate tariff framework is in place and some leeway is made available within long-term contracts between generators and suppliers/distributors without their consent. Unfortunately, the bill is quite restrictive on the use and evolution of appropriate tariffs by the regulators. The bill does not positively remove restrictions on inter-state trading of electricity, does not enable excess supply of captive generators to meet peak demand, and does not create enough pressures for the removal of cross-subsidies and inefficiencies in the system. More damaging than all of these criticisms is of course the provisions which preclude the possibilities for the sector in future such as competition in supply, development of pool or spot markets. Any such move will require further legislative change, a time-consuming and costly process. The passage of the bill, however, can be expected to hasten the reforms and restructuring at state levels (even though the direction may be away from optimal), more inter-state transfer of electricity (once again, less than optimal), and possibly the emergence of new sets of problems and issues built in the new framework. These are expected in the areas of regulatory coordination, conflicts between state and the regulators, T&D losses, and conflicts between state and central transmission utilities, besides the political fallout of attempted tariff rationalizations.
Framework for the Energy Sector
277
9.5 OIL AND GAS CONTRACTS: LESSONS FROM MUKTA–PANNA OIL FIELDS T.V.S. Ramamohan Rao • Puneet Chitkara • Kaushal K. Saxena The oil and gas industry has the following segments: exploration of potential oil basins; extraction of crude oil from the oil fields which bear oil; storage, transportation, distribution, and marketing of crude oil; refining to end product stage; and transportation, storage, and marketing of end products. The following specific features make the oil and gas industry an important infrastructure activity. First, oil and gas products are universal intermediaries in many economic activities. Second, the strategic position of petroleum products in economic activity necessitate considering the oil bearing lands as a national resource. Third, substantial investments in exploration, refining, and distribution networks are necessary to derive any benefit from these products. Fourth, in the aftermath of OPEC the international prices and the availability of crude oil have become uncertain and such price shocks have had wide ranging effects on various other sectors of the economy. Fifth, the investments and time lags involved in exploration and extraction of crude oil are substantial. Downstream production, especially the refining related activity, is different from the upstream activity, viz. exploration and extraction. In particular, the production and cost structure of this activity is different from the upstream activities (the two cost functions are separable), that is the governance structures of these two activities can be different; there is much less uncertainty and time lag associated with the recovery of sunk costs; the technological and logistic requirements are different; and they have a different industry structure—in particular, a major share of refinery capacity is already in the hands of private sector firms like Reliance Petroleum. In essence, the organizational requirements of this segment of the oil and gas sector are independent of the details regarding the upstream exploration and extraction. Similarly, due to networking requirements, marketing and distribution (at both the upstream and downstream levels) were entrusted to the national oil companies (NOCs), viz. Indian Oil Corporation (IOC), Bharat Petroleum Company Limited (BPCL), and Hindustan Petroleum Company Limited (HPCL). Issues involved in pipeline and distribution networks are of a different nature from those of exploration and extraction. Das and Parikh (2000) and Barua and Madhavan (2000) spelt out a broad spectrum of the issues involved at these levels. Hence, this report will not go into the issues related to the downstream markets. The primary emphasis of this study is on the organizational arrangements in the upstream segment of
the oil and gas industry. The government owned enterprises controlled this segment of the petroleum industry until about 1991. Traditionally, the Oil and Natural Gas Commission (subsequently made a Corporation) (ONGC) and Oil India Limited (OIL) made large investments in exploration, development of the oil fields, and production of crude oil. In other words, the upstream activity was basically organized as a vertically integrated structure.
THE BACKGROUND
FOR
RESTRUCTURING
The R-Group (MoP 1996) noted that the extent of vertical integration at all levels of the oil and gas sector was excessive. There was some inefficiency in its operations due to this organizational overhang. In particular, commercial exploitation was restricted to only about 30 per cent of the known oil fields. The basic reasons for the failure were the capital intensity of operations and the need for large fixed investments; the difficulties associated with obtaining the necessary foreign exchange; the difficulties associated with and the indigenous non-availability of techniques to measure the extent of recoverable oil from a given oil field; and the necessity to switch to innovative new technology to maintain globally acceptable environmental standards. Also the pricing of crude oil adopted by the Directorate General of Hydrocarbons (DGH)—the regulator at the upstream stage—did not leave enough resources to sustain investments in further exploration Thus, the basic bottleneck in the oil and gas sector was the exploration and extraction stage. Symptomatic of the problems of the oil and gas sector was the excessive burning of gas at Bombay High. See, for instance, the figures quoted in Upadhyay and Raman (1995). Back in 1991 there was a necessity to raise $450 million to finance the requisite technology upgradation. Similar to the problems in the other infrastructure sectors the paucity of funds was a problem for the oil and gas sector as well. Note that in the traditional infrastructure activities, viz. communications, power, and transportation, the entire output and/or services demanded must be met out of the domestic supply. The oil and gas sector, on the other hand, allows imports from other oil producing countries. In fact, even as of today, a substantial portion of the requirements are imported. For all practical purposes, the emphasis on self-sufficiency in traditional infrastructure activities is an axiom. This is not the case in the context of the oil and
278 India Infrastructure Report 2002 gas sector. However, the same type of thinking permeated the deregulation exercise in this sector as well. Depending on international sources for the procurement of crude oil presented its brand of hazards. For example, as Upadhyay and Raman (1995: p. 55) pointed out, ‘if the oil price rises of the early 1970s and the arms-twisting tactics of the Organization of Petroleum Exporting Countries (OPEC) taught the world anything, it was this: no country could count on another for a major part of its oil supplies without running the risk of a massive depletion of its foreign exchange reserves’. Attempts to import crude oil at international prices did not have much chance of success because (i) oil prices were highly variable, (ii) the requisite quantities were not available, and (iii) foreign exchange requirements became a binding constraint. This neatly sums up the euphoria about self-reliance and deregulation in the oil and gas sector. By way of comment on the thinking of the government note the following two aspects. First, the concept of selfreliance underlying the organizational changes in the oil and gas sector must be kept in proper perspective. It basically consisted of a clamour to have enough crude oil and petroleum products produced indigenously so that the dependence on imports and uncertainties associated with it can be minimized. Self-reliance did not emphasize, the need to depend on or develop technologies indigenously, the desirability of allowing only domestic companies to participate in the sector, and the need to reduce or eliminate dependence on foreign capital and foreign exchange. Second, the anxiety to obtain adequate supplies undermined the necessity to keep the cost dimension in perspective and emphasize the economics of the operations. In particular, the policy makers implicitly showed a willingness to provide more subsidies (than what was involved when the public sector was in operation) and commit the nation to greater public debt. Even the economic burden associated with foreign capital and repatriation of profits by the MNCs was underestimated or ignored altogether. For all practical purposes the concept of self-sufficiency was one-dimensional. This is the backdrop against which the contracts for oil and gas should be interpreted. The Government of India (GOI) approached the World Bank (WB) for a bailout. In its turn the WB reiterated its penchant for private enterprise. They set up deregulation, privatization, and allowing foreign participation on an equal footing as their conditionalities for granting loans. The GOI accepted the WB conditionalities.
EMERGENCE STRUCTURE
OF A
NEW ORGANIZATIONAL
The government agreed to allow domestic private firms, either on their own or jointly in collaboration with foreign
ventures, to enter the upstream phase. The strategy emphasized two aspects. First, it is essential to augment oil and gas production beyond the levels that can be achieved if public enterprises alone were responsible. There is a necessity for a large number of firms at every stage of operation. Therefore, attracting and retaining private investment in the exploration and production of crude oil is imperative to maintain activities at a high level commensurate with domestic demand. Second, institutional arrangements to get the necessary technology and financing (especially the requisite foreign exchange) are in order. In a strictly market oriented economy the organizational structure to achieve these objectives will be along the following lines. A firm will identify an owner of oil bearing land; assess the market prices of crude oil, the demand for oil at these prices, and the costs of exploration and extraction; negotiate the details of royalty payments keeping the exposure to risk in perspective; choose a joint venture partner if appropriate; and conduct the sale of crude oil to refineries on the basis of contracting fundamentals. Identification of oil-bearing land and the initial investments for extraction are too risky for any private enterprise to take the initiative. Hence such a high level of integration is not likely to emerge in the future. The following organizational structures were conceptualized as possible options in the context of the upstream activities of the oil and gas sector: separate markets for exploration and extraction; a vertically integrated firm structure where exploration and extraction will be taken up by the same company; a vertical contractual relation (a long-term leasing agreement) wherein the extraction of crude oil is contracted to another firm after the ONGC (or, for that matter, another private firm) does the initial exploration. In particular, there is a significant risk involved at the exploration stage. Even if oil is struck in a specified field the available geological data and techniques of measurement are inadequate to clearly define the amount of crude oil that can be extracted from any one oil-bearing field. Hence, the assessment of the sale price of a discovery to another firm that will undertake extraction of crude oil becomes difficult. As a result, there are difficulties in the recovery of capital costs sunk into exploration. Hence, private initiative is unlikely to materialize at the exploration stage. The importance attached to self-sufficiency therefore suggested that the exploration phase may have to be kept separate and the government, or a public sector firm as its representative, will have to dip into its deep pockets and finance such projects through public debt. They can be justified so long as the returns at least cover the interest burden on such debt. This option was exercised.
Framework for the Energy Sector It appears more rational to seek some vertical integration. In particular, allowing the original developer to continue at the production stage would be efficient. The following observations reinforce this viewpoint. Teece (1980: p. 232) observed that ‘internal trading changes the incentives of the parties and enables . . . (them to attenuate) costly haggling . . . and other non-cooperative (disruptive) behavior’. In the specific context of oil and gas contracts, Libecap and Smith (1999) pointed out that the initial explorer will be more efficient in production because they have to recover the sunk costs. Any subcontract for production of crude oil will be inefficient if they do not bear the fixed costs. While vertical integration offers incentive advantages it is inefficient in terms of the requirements of financial and capital investments, and concentrating the risks of sunk cost recovery in a few hands. As noted above, it is difficult to attract private enterprises to undertake the exploration stage. It was argued that some divestiture would perhaps render the activities more competitive, reduce the financial and capital requirements, and disperse the risks more efficiently. In particular, the government, through the medium of a public sector enterprise like the ONGC, was considered necessary at the exploration stage. In sum, it was recognized that the efficient organizational structure can be either a vertically integrated structure or one in which the two different activities are taken up by independent firms. Though this has been acknowledged, selling off a potential field to a private explorer and developer is not considered feasible for the following reasons. In recent times, the identification of potential sources as well as the percentage of recovery of oil have been subject to significant technological developments and changes over time. As a consequence, neither the developer nor the government is in a position to assess the potential of an oil field and fix an appropriate sale price. There is also a difficulty in defining the costs of restoring the land to nature after the oil resource is exhausted, let alone making the new owner responsible for such restoration. The domestic private companies, on their part, are unwilling to take the risk even at the extraction stage. Instead, they found the collaboration with a multinational company more lucrative because the compulsions on them to develop R&D and new technologies is reduced and they can claim higher rates of return on investment in the name of internationally competitive rates of return (recall that the ONGC was paid fairly low prices for the crude oil they were producing under the earlier leasing agreement, and cost based pricing would have persisted if domestic enterprises alone were involved). The government, on its part, felt that a vertical contract at the extraction stage would reduce its investment burden
279
and, bring in state of the art technologies of the MNCs. The leasehold arrangement, given to the player who bids the highest amount of payment per unit of output from the field, may be on the basis of the minimum expected production levels, eliminates the ex ante uncertainty. Thus, the basic organizational structure chosen for the upstream activities of the oil and gas sector was a vertical contract. There are two main dimensions of this governance structure. First, exploration of potential oil fields was left exclusively to the ONGC in the initial stages. However, it was acknowledged that a vertically integrated market structure is desirable. The private parties entering the extraction phase would be encouraged to integrate vertically backward into exploration as well. Second, the rights to lease and use of oil fields was given to the private players due to randomness associated with the identification of oil reserves, oil extraction, and technological changes that the firms are expected to bring about. It was a priori clear that fixed costs of exploration cannot be recovered upfront at the stage of contract writing. For, if this was possible the firm taking the lease for extraction would have found it advantageous to integrate backwards. Hence, the only major mechanism for recovering whatever portion of sunk costs that can be reimbursed is through a share in the sale of crude oil produced. The government felt compelled to retain some regulatory powers. The conventional aspects of regulation, viz. ensuring free entry and market competition, and setting an overall price cap were not of the essence. Instead, the emphasis was on ensuring that the output of crude oil is maximized, and the necessary technology and financing is forthcoming
NEW EXPLORATION LICENSING POLICY The basis of contracting was the New Exploration Licensing Policy (NELP). Two aspects have been spelt out in detail to achieve speedy implementation of the oil and gas projects. At the bidding stage, the private parties were assured that: geological data will be made available to them to evaluate the prospects of the blocks on offer for exploration and extraction; even the fields that have already been discovered by ONGC will be offered for development; attractive fiscal terms (taxes, cess, duties on imported capital goods) will be offered to ensure recovery of sunk costs (that is, capital investments made by firms at the stage of development and extraction of crude oil) and adequate (globally competitive) returns to the investors; and the increase in the number of companies will be primarily to focus on the adoption of new and innovative technologies. The companies were also assured that at the stage of contract execution they need not pay any signature bonus (at the time of contract writing), delivery bonus, or
280 India Infrastructure Report 2002 Box 9.5.1 Functions of the DGH As a monitoring and regulatory agency, the DGH has the following functions: • Provide technical advice on exploration and exploitation of hydrocarbons. Recently the DGH has undertaken geoscientific work in less explored virgin fields to upgrade information so that such areas can be opened up for utilization in the future • Review the exploration activities of the concerned companies. The R&D institute of ONGC recently sponsored a study to improve the recovery ratio of the Panna fields • Assess and periodically revise the estimates of hydrocarbon reserves. Three surveys, one each for magneto telluric, seismic, and aeromagnetic projects, have been conducted • Advise the government on the blocks to be offered for exploration and extraction of crude oil • Standardize and monitor environmental norms • Maintain adequate database for operational purposes. Along with the Norwegian Petroleum Directorate they are in the process of acquiring satellite gravity data for several potential blocks • Be responsible for oil field safety and training. From the above description it appears that the MoP wanted to utilize DGH as the coordinator of all technical data and technical advice. However, as it happened in the case of Mukta–Panna, the contracting companies would like to utilize their own in-house expertise. In a completely deregulated set-up the firms may also utilize vertical contracting with private companies who specialize in providing such technical services. Once this market reaches a mature state it would be unnecessary for the DGH to be involved in this phase of the operations since it is not really a legitimate function of the regulator (see also Box 9.5.5). Source: Ministry of Petroleum (1996), p. 31.
production bonus; will have the freedom to market oil and gas from the new blocks in the domestic market; will be offered a price that will be comparable to what they can get on international markets (in practice, the standard used was ‘Brent crude minus ten cents’); will be exempted from duty for capital goods imported to undertake production; and will be allowed a seven year tax holiday from the date of commencement of production. The Comptroller and Auditor General (CAG) brought out two essential arguments during the initial implementation phase of NELP. First, prior to deregulation ONGC was offered unremunerative prices for crude oil. More than anything else this was the major reason why adequate resources could not be generated to augment production. Second, since vertical integration is feasible, ONGC, and NOCs in general, should be given an equal opportunity in the bidding process. In essence, the point is that though constraints of various kinds do exist they do not justify unequal advantage to foreign participants and/or denying a level playing field to NOCs. Consequent upon this line of thinking, NELP 2 was modified so that national oil companies will be allowed to compete in the bidding for petroleum exploration and development licensces, and NOCs will be provided a level playing field alongside private parties with respect to price and fiscal policies. Stated in this manner, NELP was a specification of the enabling clauses. It does not describe the obligations placed on the private firm who eventually wins the bid. In particular,
details regarding the choice of appropriate contract parameters and their quantitative magnitudes, the institutional and organizational structures, governance and control of operations, and grievance and redressal mechanisms were left open and to be addressed on a caseby-case basis. Similarly, the necessity for coordination with the downstream level and OCC (Oil Coordination Committee) was acknowledged. In November 1997, the Administered Pricing Mechanism (APM) for oil products was notionally dismantled in favour of international pricing. Clearly, the NELP accepted the participation of MNCs due to the unavailability of the state of the art technology, and paucity of finances (in particular, foreign exchange) for capital investments with domestic companies (public or private). The role and functions of DGH during the transition to complete deregulation are summarized in Box 9.5.1.
MODEL PRODUCTION SHARING CONTRACT The presentation of the NELP in the previous section clearly suggests that it is a statement of enabling clauses for vertical contracting and that the regulator wants to maintain some control to ensure that the fixed cost of exploration is recovered to the extent possible without disrupting the production volume (that, in any case, is of prime concern). It also suggest that the details of a contract will be negotiated on a field-by-field basis.
Framework for the Energy Sector The model production sharing contract (MPSC) was drawn up to provide the broad guidelines for working out these details. The highlights of this approach will be presented in this section. In most contracts, at the initial stages of deregulation, the ONGC has already made certain investments in exploration and partially developing the oil fields. However, there are some further development costs, viz. capital investment in drilling machinery to extract oil, on-site storage, and other operating expenses, in the conduct of petroleum operations. The model contract stipulated that the contracting parties would be solely responsible for these costs. Financing becomes one of the crucial aspects. It consists of a rupee component as well as the foreign exchange requirement to import capital equipment, obtaining technical expertise, and so on. Clearly the normal domestic banking and financial institutions will be utilized to finance the Rupee requirements. With respect to the foreign exchange component the companies have been entrusted with the responsibility of freely obtaining, through normal banking channels, funds necessary to carry out petroleum operations; making payments outside the country for the purchases and services; repatriate in foreign currency the proceeds of a sale; and receive, retain, and use abroad the proceeds of export sales of petroleum, if any. For all practical purposes, this arrangement replaced the earlier operation of the Oil Industry Development Board (OIDB). It was originally established with a view to provide financial assistance for the development of oil and gas produced (see, for example, Ministry of Petroleum 1996: p. 40). The primary advantage in having an institution like the OIDB was in the pooling of risks. For, if each of the firms approach different financial institutions they will be subject to market fluctuations and risks. However, the resources available with the OIDB were very inadequate. Further, they could not deal with foreign exchange requirements efficiently. Technology is another major aspect. The MPSC stipulates that it should be specified clearly at the stage of contract writing especially because it has fundamental cost implications. However, since the technological changes in extraction have been quite dynamic the best that could be done, as in the Mukta–Panna case, is to stipulate that the contractor should utilize modern oil fields technologies and petroleum operation practices keeping in perspective the prevention of environmental damage and provide the regulator all the relevant information, data, and samples to facilitate monitoring. Before commencing any petroleum operations there should be transparency and agreement regarding the points, over space and time, at which measurements (geo-physical measurements etc.) will be undertaken and the methods
281
of measuring petroleum production. Similarly, during the operation of the contract, the regulator needs periodic (monthly, quarterly) information relating to capital assets, property, and inventories acquired by the contractor, regarding production and other petroleum operations the quality characteristics and quantity of crude oil produced and saved, and the quantity of crude oil pumped to field storage and reinjected, and data pertaining to the remaining potential and development needs. An accurate assessment of cost is an equally important requirement before defining the revenue sharing agreement. In general, the production of crude oil necessitates the following costs: exploration costs until the date of first commencement of commercial production; exploration costs incurred in any year after production begins; development costs prior to the start of production; development costs incurred on a recurring basis; and payments committed to the owner of the land. Clearly, in the initial stages of commercial exploitation the output and revenues will not be sufficient to cover all the costs. In that case, the priority of commitment is defined as payments to the government, current production cost, exploration cost, and/or development cost. However, the contractor will be entitled to a fixed fraction (negotiated at the time of contract writing) of his expenses as cost petroleum. The remainder will be carried over to the next time period. The production sharing contract (PSC) was operationalized along the following lines. The contractor will prepare a monthly statement of prices in the spot markets and those charged by the main oil producing countries, the quantity of crude oil sold to third parties, and the value of inventory stocks. The net cash income of the contractor is defined as cost petroleum + profit petroleum + incidental income from petroleum operations – royalty payments – current production costs. Profit petroleum, that is, the share of the profit accruing to the contractor, is itself defined as a function of an investment multiple, where investment multiple = accumulated net cash income/ accumulated investment made by the company. The basic logic for this PSC was the necessity, for the contracting company, to cover costs and obtain a minimum rate of return on investment. To facilitate implementation of the contract it was stipulated that an operating committee (with adequate representation of all contracting parties) would be constituted to look after the day-to-day operations. Similarly, a management committee would monitor the annual works programmes and budgets for exploration operations, the progress of annual works, especially with respect to costs, additions to property and assets, and the appointment of auditors and approval of audited accounts.
282 India Infrastructure Report 2002
THE MUKTA–PANNA CONTRACT In pursuance of the NELP and the MPSC, the government agreed to lease out (or provide the rights to use) of oil and gas fields that have already been developed by ONGC and OIL. In August 1992, MoP invited bids for 12 medium sized and 31 small sized oil fields. One of these was the Mukta–Panna (MP) oil fields off the Bombay High. They are located approximately 120 and 100 kms, respectively, west and northwest of Mumbai. Both were expected to bear oil and gas.
The ONGC incurred a cost of Rs 546.39 crores while developing the MP fields. In 1990 these fields were leased to ONGC for 20 years for development and extraction of crude oil. In 1992 they were already producing oil and selling it to the MoP. The lease was withdrawn to make way for the deregulation under the new dispensation. In its place, a 25 year production sharing contract (PSC) was signed on 22 December 1994 with Enron Oil and Gas India Limited (EOGIL) and Reliance Industries Limited (RIL) consortium. According to this agreement EOGIL,
Box 9.5.2 Ravva Fields Contract The Ravva fields contract was the other major agreement signed alongside the EOGIL–RIL contract for the MP fields. And yet it is somewhat different. The major aspects will now be highlighted. The Ravva fields are in the Krishna–Godavari (KG) basin off the Andhra Pradesh coastline. It is awarded to Videocon, Command (currently Cairns Energy India Ltd), and Marubeni consortium. The shares of the different participants are as follows. Command: 22.5 per cent Videocon: 25.0 per cent Marubeni: 12.5 per cent ONGC: 40.0 per cent ONGC spent Rs 351.05 crores in exploration and initial development of these fields. The consortium offered Rs 173.25 crores as signature bonus. This is roughly equal to the survey expenditure plus the expenditure on production facilities. Their present value was reimbursed since these assets were in use subsequently. If the same logic was applied in the context of the MP fields the signature bonus should have been of the order of Rs 223.29 crores. However, only Rs 12 crores was paid. The contractors took advantage of the invitation of the bids that mentioned that the production facilities that were being built up prior to the contract would be completed without specifying the commitment to costs. The MoP (in July 1994) argued that the international price of crude oil is currently low, and that the consortium must anyway pay royalty and cess on the oil produced. This does not leave enough return if past costs must be reimbursed. The need of the hour is to bring the fields to the production stream in a technically efficient manner. Overcoming the resource constraint, especially the foreign exchange component, was the prime concern. The reserve estimates of the Ravva fields was not disclosed. As of 1997 it produced 35 thousand tons of oil and 0.7 million cubic meters of gas per day. The incremental production bonus of the two fields were: Production (million barrels) Ravva (million $) MP
25 9.0 –
50 9.0 6.0
75 9.0 –
80 1.8 –
85 1.8 –
90 1.8 –
95 1.8 –
100 1.8 9.0
200 – 15.0
The differences were attributed to the perceived risk of recovery. The Ravva field contract explicitly provided for an abandonment fund to restore the site to nature at the end of the contract. There was no similar commitment in the MP case. The basic problems in both the contracts were: • The absence of firm information on production and costs to justify contract parameters at the time of contract writing • The absence of procedures for renegotiation as and when more definite information is available • Interpreting self-reliance in terms of making production possible without regard to cost commitments and the excessive dependence on MNCs and their profitability requirements. In a fully deregulated environment these risks will be taken by the promoter or shared with the contracting parties after periodic appraisal. The MPSC did not adopt this commonsense approach. Note: –: Data not available. Source: CAG Report on Ravva and MP Fields.
Framework for the Energy Sector
283
Box 9.5.3 Enron is Selling its Stake It appears that there is a change of corporate philosophy at Enron. Basically, the shift amounts to its perception that its exp ertise is in risk management, and a restructuring of its activities recommends pulling out of the MP fields. However, the PSC did not prohibit withdrawal, stipulate first right of refusal to ONGC or RIL, or allow any of the partners to withdraw without the permission of the government because continuity of production is desired. In October 2000, EOGIL sought the permission of the government to withdraw. EOGIL invited open bids for their 30 per cent stake. The current reports indicate that ONGC and RIL are actively considering bids. They appear to consider the fields as a means to augment their cash flows and assist in their exploration activities. Fundamentally, the rate of return of the MP fields do not seem to be as low as it was originally made out to be. The specific estimates by Mackenzie—a leading energy consultant—indicate that the remaining assets have a net present value of $545 million. This is much higher than the original estimate offered by the EOGIL–RIL consortium in 1994. There are fresh doubts about the validity of the original PSC. The information contained in these bids should be utilized by DGH to learn a lesson in evaluating the potential, and to impose adequate penalty before granting permission to Enron to sell its stake if it can be inferred that the company cheated initially. It would be premature to come to any conclusion without further quantitative evaluation. However, it is relevant to note that Enron is trying to recover the entire remaining value of the capital investments that it made. This is in contrast to its unwillingness to pay any part of the capital investments of ONGC at the time of the original contract. For, as already noted, the EOGIL–RIL consortium paid only Rs 12 crores as a signature bonus and did not agree to cover any of the major capital costs incurred by ONGC prior to the termination of their 20 year lease. Thus, ONGC would have lost twice in the contract for the Mukta–Panna oil fields.
RIL, and ONGC held shares of 30, 30, and 40 respectively in the post contract production operations. The model contract does not stipulate the contractor paying for the initial cost incurred by ONGC. Hence, the DGH accepted the consortium’s offer of Rs 12 crores as signature bonus, that is, an amount paid to the existing holder of the lease for the right to carry out exploration and drilling in a given contract area. The CAG pointed out subsequently that there is no evidence that this figure is based on the net present value (NPV) of the project, the profits that would have accrued to ONGC if it did not surrender the lease, the investments that ONGC already sunk into the project, and/or the experiences with respect to contracts for other oil fields. See the contract for the Ravva fields summarized in Box 9.5.2. The MoP took the position that it would not be possible to recover these sunk costs even if the lease for ONGC was continued. In other words, there was an argument about the inevitability of financing exploration and development costs from the budgetary resources of the government. By way of contrast, as noted recently, Enron’s policy is asymmetric in so far as they are trying to recover all their capital costs and possible profits that they forego while attempting to sell their stake in the MP oil fields (see Box 9.5.3). The highlights of the PSC are an agreement about the potential output from the fields, an estimate of the operating and capital expenditure of the firm, the price at which crude oil will be sold to the government, a cess based on
the volume of production, production bonus when the cumulative production level reaches a specified level, and a share of post tax profit (called ‘profit oil’ or ‘profit petroleum’). In 1990 the ONGC estimated the oil reserves of the MP fields to be 54 million tons (roughly 1 ton = 7.5 barrels of crude oil, 1 barrel = 158.91 litres). However, the prospectus to the bidders mentioned only 31.5 million tons. In the EOGIL–RIL bid, this was reduced to 20 million tons of recoverable oil. The PSC itself was based on an estimate of 14 million tons. This revision was apparently approved by the ONGC. However, the basis for downgrading was not spelt out explicitly. The estimate of capital expenditure (CAPEX) was Rs 173.36 crores. Any increase during the execution of the PSC was to be borne by all the parties to the contract. This will have an adverse effect on ONGC in terms of its recovery of sunk costs of exploration. The operating expenditures (OPEX) have not been included as a firm commitment and no disincentives were built in to protect against excessive spending. In practice, the operating cost for the first year worked out to Rs 370.52 per barrel. This was considered to be very high. The government agreed to purchase oil from the MP fields at the international price plus a premium of $4 per barrel (or, alternatively designated as ‘Brent crude minus ten cents’ rule). The reason given for this is the low sulphur content of the oil from the MP fields. The consortium was expected to pay cess and royalties for the entire period of the contract out of this price. The cess was Rs 990 per
284 India Infrastructure Report 2002 Box 9.5.4 Public Interest Litigation There have been allegations about manipulation in setting up the MP contract, the process of evaluating bids, and many other dimensions. The ministry took the position that the contract must be viewed from the perspective of national interest instead of the limited interest of the ONGC. The CAG, on the other hand, maintained that this ‘is an unacceptable generalization, especially in view of the manifest infirmities of the bidding and contract formalization process brought out by the auditor.’ This resulted in the Center for Public Interest Litigation (CPIL) and the National Alliance of People’s Movement (NAPM) filing a public interest litigation. The objections raised by the CAG can be summarized as follows. • The initial costs incurred by ONGC have not been recovered. Compensation offered in the form of signature and production bonus is inadequate. Given the paucity of resources available for investment purposes it will be difficult for the government to sustain further oil and gas exploration and development. • ONGC, that discovered the MP fields, was already producing oil and contributing 13.44 per cent post-tax internal rate of return. On the other hand, the EOGIL–RIL consortium claimed a similar return for itself. Hence, awarding the contract amounts to discrimination against ONGC. The ONGC probably suffered a loss of at least Rs 700 crores. • In 1991, ONGC estimated the potential oil reserve as almost four times the figure utilized to assess the bids. Even the figures mentioned in the tender were not followed up. The original estimates may be more nearly correct. On 27 June 1997 Yashwant Sinha himself said that ‘there should be reassessment of oil reserves in all the fields leased to private companies.’ The annual report of Reliance for 1998–9 noted that it has ‘upgraded its oil reserve estimate by 36 per cent and gas reserves by 210 per cent.’ The current production of over 27,000 barrels of crude oil and 3.0 million cubic meters of gas are well above the expected quantities. Further, the EOGIL–RIL estimate in their bid was higher than the final amount utilized for calculating the net present value. In essence the ministry did not adequately ascertain the reserves before finalizing the contract. • Even the price of Brent crude, that was utilized as the basis for compensation for the oil produced, was nowhere near the $24 per barrel figure used. This and other concessions, viz. the profit tax, cess, royalty, OPEX, and CAPEX, were not specifically evaluated. • ‘The price for crude oil produced by the NOCs is administered and stands, as in May 1996, at Rs 1741 per ton. Against this the joint venture would get the international price for crude, which is substantially higher at Rs 4545 per ton. Hence, whatever the potential oil estimate, there is a huge loss.’ • The ONGC and NOCs were not allowed to bid and did not have a level playing field. This is also a discrimination against ONGC. In its 1998–9 report the Estimates Committee of the Parliament (ECP) also expressed its reservations about the contract. The basic points of the report were: (i) the foreign exchange problem is not an adequate explanation since the temporary problem of 1991 did not persist for long; (ii) unattractive rate of return on international standards, promised to Enron in the MP context, does not justify further losses and subsidies to GOI; and (iii) foreign oil prospectors should invest in exploration activities and not just undertake production in fields that are already developed. On 19 October 2000 the Supreme Court dismissed the case in favour of the contract on the following grounds. • The allegations of bribes and misdeeds have not been substantiated. • ‘It is possible that out of (the) 34.4 million tons of oil estimated originally as being the reserve . . . the recoverable oil could be only 20 million tons or near about that quantity, as evaluated by Reliance and Enron.’ • ‘It will be extremely difficult for a court to decide whether a particular price agreed to be paid under the contract is fair and reasonable or not.’ • The contract provided for a share to the government in the event of an increase in the quantity of recoverable oil. • The NELP 1993 stipulated that the government ‘expects companies proposing to develop medium-sized fields to take one or more exploration blocks from amongst those on offer by the government of India’. This is adequate to ensure that the fixed costs of development are not exclusively left to the ONGC. In general, it can be concluded that the court ordering, as a governance mechanism, is weak. It can only check whether the original contract was along the lines of the model contract, and if there has been a breach of contract terms at the execution stage. It cannot objectively examine the validity of the contract terms themselves. The following issues are worth recording despite the fact that they could not stand legal scrutiny. One, there is a need for more accurate technical information about the quantum of crude oil. The estimates must be clearly known to the contracting parties and they should be verifiable. Second, there is a necessity to reconsider the different aspects involving the recovery of costs. Third, there is a need for contract clauses based on verifiable production data rather than fixed amounts specified as production and other bonuses.
Framework for the Energy Sector ton. The royalty, on the other hand, was Rs 481 per ton. These rates were in vogue in March 1992. However, on 23 February 1994—the date of signing the EOGIL–RIL contract—the government increased the royalty rate for related projects to Rs 528 per ton with retrospective effect from 1 April 1993. There was no explanation for the same rates not being applied to the MP contract. The production bonus was defined as $6 million when the joint venture records a cumulative output of 50 million barrels, an additional $9 million when output crosses 100 million barrels, and $15 million more when the cumulative output exceeds 200 million barrels. It is not altogether clear if even these figures are related to the recovery of past costs in any way. On the whole, it appears that the PSC may have been drawn up keeping the international competitive requirements. However, many variables have been fixed a priori or left to be interpreted by the governing board of the consortium. The contract has been controversial, and was contested in the courts (see Box 9.5.4).
LESSONS
AND
RECOMMENDATIONS
The EOGIL–RIL contract for MP fields vividly illustrates that the dependence on subsidies and government control has been perpetuated rather than eliminated. If selfsufficiency was the real essence, only domestic private enterprise should have been allowed and they would then necessarily have accepted whatever competitive return is possible rather than claim internationally comparable rates. The ground reality is that, given the sequence of events over the past decade and their irretrievable momentum, there is not much point in harping on self-reliance. It would be more meaningful to concentrate on the economic viability of the new contracts, especially with a view to eliminate the dependence on administered prices and government subsidies. Private sector companies have shown a willingness to make quick profits on oil fields that have already been discovered and developed. But they are not interested in making investments for exploration of new oil fields since the financial risks are high. The public sector undertakings, like the ONGC, continue to underwrite financial risks while discovering oil fields. The NELP stipulates that parties who will win competitive bids for extraction of oil from discovered fields must invest in exploration of new fields as well. The contracts did not operationalize this aspect. They have not been made accountable in any sense. This is one of the weakest aspects of the oil sector deregulation. The following observations are also pertinent. From the reaction of ONGC, both in terms of its assertion that its
285
interests are subverted and its willingness to bid for the 30 per cent of Enron stake, it appears that they are willing to act independently and take the initiative to maintain competitiveness. Further, even the DGH eventually accepted that the NOCs should be offered a level playing field. Subcontracting arrangements will be insufficient and exploration activity cannot be sustained on a long term basis if the fixed costs are not recovered. The role of signature bonus and cess in this recovery process should receive appropriate attention. Unlike the case of the other infrastructure sectors, where the necessity is for a wide network and/or the gestation periods are large, oil and gas projects at the upstream level are relatively isolated and do not have long gestation periods. Hence, especially in this sector, sustained subsidies and budgetary provisions for capital investments will be unwarranted. Accepting international prices for crude oil, irrespective of whether production is from MNCs or NOCs, leads to another related issue. Under the regime of APM, the downstream prices of petrol, diesel, etc. have been kept low resulting in an oil pool deficit and government subsidies at this end as well. However, the international prices for crude should indicate a corresponding change in the prices charged from the final user. This is possible since, at least notionally, the APM has been withdrawn. But these adjustments in oil pool accounts are not really complete as of today. A better coordination between the OCC and DGH is warranted. Though OPEX and CAPEX in the Indian context are relatively low, participation in production by international enterprises and the returns they expect may indicate that international prices of crude oil will not be adequate. Appropriate flexibility in crude oil pricing is warranted. Two further observations are in order. First, there is a possibility that a marginal field may not be able to cover the fixed costs at all prices of crude oil even at international prices. Clearly, exploiting such fields is not feasible irrespective of who is entrusted with the job. Such projects may have to be shelved until an opportune time arises. Note that this is what Canada has done with respect to its oil sands projects. Subsidies cannot sustain uneconomical projects on a long term basis. Second, in general then, any oil field that has the potential to produce crude oil will generate some profits. In such a case, either an upfront amount of profits or a share depending on equity participation should be worked out to cover the fixed costs. This may not leave ‘adequate’ returns to entice the MNCs. But the PSUs and NOCs, operating on a no-profit basis, can still do the job. Subsidizing private enterprise, operating either on their own or jointly with an MNC, for the sake of deregulation is self-defeating and unwarranted.
286 India Infrastructure Report 2002 In practice, the EOGIL–RIL contract accepted the resolution of risk ab initio by taking the minimum estimate of oil production. However, it is well known that dynamic changes in technology have been allowing greater recovery rates. The regulatory process also expects complete data on production costs to be supplied on an annual basis. Hence, an incomplete contract, that fixes contract parameters periodically after the risk is resolved, is efficient. However, this is subject to hold-up and costly recontracting. An incentive to reveal information and share benefits would have been superior. In either case there is a necessity to keep the option of periodic appraisal (say, once in three years) open. Rao (2001) expressed a similar view recently. This was, however, not attempted. The following two observations of the CAG are also pertinent. First, ‘As a matter of prudence, all the elements comprising the . . . government stake should have been carefully benchmarked and expected returns from these elements diligently evaluated before initiating the bidding process.’ Second, OPEX and CAPEX have been loosely stated. ‘Principles of computing cost escalation and control’ should be explicit in the PSC. In its absence, the management committee cannot supervise activities appropriately. In the foreseeable future, deregulation would be complete. Then, the government will have no role in the organization of this activity. In particular, the dichotomy between government enterprises operating at the exploration stage and private firms taking up extraction must be eliminated. Vertical contracting, if it still persists as an efficient organizational mechanism, will be due to such divestiture being a superior risk sharing arrangement between private parties. In general, the contracting parties will work out the details and procedures for monitoring and control. The MPSC, as it stands currently, is unnecessarily intrusive in terms of the nature of data that the contractor must provide and the monitoring and control that the regulators wish to exercise. It is evident from the presentation that the production sharing contract is only one of the dimensions of the problems confronting the oil and gas sector. More generally, organizational arrangements are necessary to obtain more investment in exploration and extraction, better technology, and more foreign exchange whenever it becomes essential. The second and third aspects will now be examined briefly. It is also necessary to examine the necessity for MNCs being assigned specific production rights if the only thing they are expected to bring in is foreign exchange. For, privatization and international participation in the capital markets certainly has better prospects of efficiently pooling risks across a variety of projects instead of the contracts dealing with each of the projects in isolation. This option
has not been evaluated appropriately. The OIDB and Infrastructure Development Finance Company (IDFC) have a pivotal role in this context. The real problem may be in obtaining the state of the art technology. The existence of risk in evaluation should have indicated greater caution. In particular, there is a necessity to develop more meaningful standards for evaluating the potential output from the oil fields, and it is necessary to monitor actual production levels to improve future estimation as well as revise the terms of the contract in case of discrepancies. Further, when events like the attempts by Enron to sell off its stake occur there is more information. It must be utilized appropriately. These are technical details. They cannot be left to the CAG or the Estimates Committee of the Parliament (ECP). As the R-Group report suggests, a national petroleum data archives may be established under the DGH. Cores, logs, seismic sections, geological maps, etc. should be treated as national assets. They should be accessible to all concerned parties on appropriate payment. All companies should surrender any data they collect to be kept in the archives. The need to obtain and utilize expertise on technologies has been underestimated. But this will be critical to make a proper assessment of the blocks on offer. Domestic private companies should be encouraged to negotiate technology transfer with the MNCs. This process, instead of a regulator getting involved, would be in the spirit of deregulation. It is possible to conceptualize further deregulation of this nature. In conclusion, the following aspects may be reiterated. First, technological upgradation at the exploration, production, and refining stages is warranted. It may be efficient to leave this to market coordination and the private contracting parties that have a stake in the venture. Second, financial arrangements, especially the participation of foreign capital, need not be project specific. Since deregulation is occurring more generally there should be proper coordination across sectors. Third, the CAG or the ECP cannot satisfactorily handle the evaluation of technical details. Expertise of technologists must be sought by the private contractors according to the requirements of the market. Fourth, contracting and monitoring details leave a lot to be desired. Major disasters occurred, not only in the oil and gas sector, but also in power and telecommunications. Fifth, deregulation and privatization are specialized tasks for them to be left to bureaucrats and politicians. There is a necessity to redress the balance. In the final analysis, when deregulation has run its course, the private parties, viz. the owners of different assets as well as the contractors, will make offers on the markets commensurate with the desires of the consumers.
Framework for the Energy Sector Suppose the following constraints operate: (i) domestic availability of petroleum products is a compulsion due to shortages on international markets; (ii) domestic capital is inadequate and indigenous enterprises are unequal to the tasks required by technology and capital; and (iii) foreign company participation, both in terms of providing risk capital and technology as well as managing day-to-day production operations becomes inevitable and they must be paid the rates of return they expect. Under such
287
conditions the domestic prices of petroleum products must bear the brunt of the constraints. Otherwise an overall fiscal burden on the government cannot be avoided. The limited deregulation attempted in the oil and gas sector has not given proper attention to the economic viability of the activity. The current attempts to balance many parameters in the hope of maintaining the social welfare of the current generation of consumers is inefficient from the long run perspective.
REFERENCES 3iNetwork (2001), India Infrastructure Report 2001: Issues in Regulation and Market Structure, Oxford University Press, New Delhi. Barua, S.K. and T. Madhavan (2001), ‘At the Cross-Roads: Pipeline Infrastructure for Oil Products’, in 3iNetwork (2001). Business Review (1999), ‘Annual Report of Reliance 1998–99’, available at www.ril.com/rpg99-htm. CERC (1999), Consultation Paper on Bulk Electricity Tariffs, 15 September 1999. Chitkara, Puneet, Rajiv Shekhar and Prem K. Kalra (2001), ‘Missing Interconnections in the Power Systems’, in 3iNetwork (2001). CMIE, 1995, India’s Energy Sector, July, Center for Monitoring Indian Economy, Mumbai. Comptroller and Auditor General (1996), Infirmities in the Bidding Process and the Production Sharing Contracts. Das, A. and J. Parikh (2000), ‘Coal, Oil, and Gas’, in K.S. Parikh (ed.), India Development Report 1999–2000, Oxford University Press, New Delhi. Ganguli, B. (1997), ‘A Spanner in the Works’, Business India, 19 May–1 June. Libecap, G.D. and R.L. Smith (1999), ‘The Self-Enforcing Provisions of Oil and Gas Unit Operating Agreements: Theory and Evidence’, Working Paper 7142, National Bureau of Economic Research, Cambridge, MA. Ministry of Petroleum (1996), Report of the Strategic Planning Group on Restructuring of Oil Industries (R-Group). Morris, Sebastian (1999), ‘The Political Economy of Electric Power in India’, in G. Raghuram et al. (ed.), Infrastructure Development and Financing: Towards a Public–Private Partnership, Macmillan India, New Delhi.
Morris, Sebastian (2001), ‘Power Sector Reforms and Regulation: The Road Ahead’, in 3iNetwork (2001). Pandey, Ajay (2001), ‘Regulatory Processes, Objectives and Outcomes: Case of first Tariff Review by CERC’, in 3iNetwork, India Infrastructure Report 2002, Oxford University Press, New Delhi. Pandey, Ajay, and Debashish Sanyal (2001), ‘Electricity Transmission Pricing Issues in India’, paper presented in Conference on Global Participation in Indian National Grid, Energy Management and Convergence, August 17–18, Mumbai. Rao, S.L. (2001), ‘Private Investment in Infrastructure’, Economic Times, 19 March. ICRA Advisory Services (2001), ‘Cost of Capital for Power Sector Utilities’, CERC. ICRA Advisory Services (2001), ‘Depreciation Norms’, CRISIL Advisory Services, CERC. ICRA Advisory Services (2001), ‘O&M Costs and Escalation Characteristics Cost for Thermal Power Generation’, Development Consultants Ltd., CERC. ICRA Advisory Services (2001), ‘Operating Norms and O&M Cost Norms of Hydro-Power Stations’, WAPCOS, CERC. Sinha, Sidharth (1999), ‘Appropriate Return to Equity in Private Power Projects: Dabhol Power Company’, in G. Raghuram et al. (ed.), Infrastructure Development and Financing: Towards a Public–Private Partnership, Macmillan India, New Delhi. ICRA Advisory Services (2001), Tariff Norms (order dated 21 December 2000), CERC. Teece, D.J. (1980), ‘Towards an Economic Theory of the Multiproduct Firm’, Journal of Economic Behavior and Organization, Vol. 3, No. 1, 237–47. Upadhyay, A. and A.T. Raman (1995), ‘Learning to Cope’, Business India, 13–26 February.
