World Bank Reprint Series: Number 148 Mohan Munasinghe An Integrated Framework for ~nergyPricing in Developing Countries An Integrated Framework for Energy Pricing in Developing Countries Mohan Munasinghe* lNTRODUCTION In recent years, decisionmakers in an increasing number of countries have realized that energy sector investment planning and pricing should be carried out on an integrated basis, e.g., within the framework of a national energy master plan that determines energy policy, ranging from short-run supply-demand management to long- run planning. However, in practice investment planning and pricing are still carried out on an ad hoc and at best partial or subsector basis. Thus, electricity and oil subsector planning have traditionally been carried out independent of each other as well as independent of other energy subsectors. As long as energy was cheap, such partial ap- proaches and the resulting economic losses were acceptable, but lately, with rising energy costs, changes in relative fuel prices, and substitu- * Energy Department, the World Bank. The Energy Journal. Vol. 1 , No. 3 Copyright Q 1980 by the International Association of Energy Economists. All rights reserved. The opinions expressed in this paper are the authors, and do not necessarily represent the views of the World Bank or its affiliated organizations. I am grateful to Joy Dunker- ley and Gunter Schramm for helpful ideas. Jim Baldwin and Eric Daffern also provided useful comments on an eariier draft of this paper. 2 / The Energy Journal tion possibilities, the advantages of an integrated energy policy have become evident. In this paper the importance of coordinated energy planning and pricing will be emphasized, with particular reference to the interrela- tionships among the pricing policies adopted in various energy sub- sectors such as electric power, petroleum, natural gas, cod, and tradi- tional fuels (e.g., firewood, crop residues, and dung). Nonconventional sources can also be fitted into this framework. We will focus on the LDC context where, generally, higher levels of market distortion, shortages of foreign exchange and resources for development, larger numbers of poor households whose basic needs must be met, greater reliance on traditional fuels, and relative paucity of energy data add to the already complicated problems faced by energy planners in the de- veloped countries. We will also touch on the chief investment issues to the extent that they strongly influence pricing policy. Before developing an integrated framework for power and energy pricing, it is necessary to briefly discuss what is meant by national energy planning. The broad underlying rationale is to make the best use of energy resources to promote socioeconomic development and improve citizens' welfare and quality of life. Therefore energy planning is an essential part of overall national economic planning, and should be carried out and implemented in close coordination with the latter. However, in energy planning, the principal emphasis is on the com- prehensive and disaggregate analysis of the energy sector, with due regard for the main interactions with the rest of the economy. In a strictly technical sense, the energy planner's role might be confined to seeking the least-cost method of meeting future energy requirements. However, energy planning also includes a variety of other objectives, including reducing dependence on foreign sources, supplying basic energy needs of the poor, reducing the trade and foreign exchange defi- cit, priority development of special regions or sectors of the economy, raising sufficient revenues to fiance energy sector development (at least partially), ensuring continuity of supply and price stability, pre- serving the environment, and so on. In general, energy planning requires analysis at the following three hierarchical levels in relation to fundamental national objectives: (1) links between the energy sector and the rest of the economy; (2) inter- actions between different subsectors within the energy sector; and (3) activities in each individual energy subsector. The steps involved in the planning procedure usually include energy supply and demand analyses and forecasting, energy balancing, policy formulation, and impact analysis, to meet short-, medium-, and long-range goals. Initially, these activities may be carried out at a relatively simple level; later, as An Integrated Framework for Energy Pricing / 3 niques, including computer modeling, may be implemented. The insti- tutional structure should also be rationalized by setting up a central energy authority (CEA),or ministry of energy, with its principal focus on energy planning and policymaking. The execution of policy, and day-to-dayoperations, would remain the responsibility of the electricity utilities or petroleum corporations that already exist in practically all countries. SCOPE AND OBJECTIVES OF PRICING . To put pricing in proper context, we note that it is only one of the policy tools available for optimal supply-demand planning and management; others include physical controls, technical methods (in- cluding research and development), and education and propaganda. Since these tools are interrelated, their use should be well coordinated. Physical controls are most effective in the short run when there are unforeseen shortages of energy. All methods of physically limiting consumption are included in this category, such as load shedding and rotating power cuts in the electricity subsector, as well as reducing the supply of gasoline or banning the use of motor cars during some pe- riods. Technical means on the supply side include the least cost or cheapest means of producing a given form of energy, the best mix of fuels, and research and development of substitute fuels such as wood- alcohol for gasoline; on the demand side, they include introducing higher efficiency energy conversion devices, such as better stoves for woodfuel. Education and propaganda on the supply side include efforts to make citizens aware of external diseconomies such as pollution, and supportiveof re-afforestation schemes to preserve the environment; on the demand side, they include public education for energy conser- vation. Pricing is a very important tool, especially in the long run. As dis- cussed below, the pricing and investment decisions should be closely related. However, energy supply systems-e.g., electricity generation, transmission, and distribution; oil and gas wells and pipelines; coal mines; and forests-usually require large capital investments with long lead times and lifetimes. Therefore, once the investments decision is made, usually on the basis of the conventional least-cost method of meeting demand by subsector, with due regard for interfuel substitu- tion possibilities, there is a lock-in effect with respect to supply. Thus prices should be related to the long-run planning horizon. On the de- mand side also, energy conversion devices (e.g.. motor cars, gas stoves, , . . .. , q . , a . , 4 / TheEnergy Journal income levels and have relatively long lifetimes, thus limiting con- sumers' ability to respond in the short run to changes in relative fuel prices. The objectives of energy pricing are closely related to the goals of energy planning, but they are more specific. First, the economic growth objective requires that pricing policy should promote economically effi- cient allocation of resources, both within the energy sector and between it and the rest of the economy. In general terms, this implies that future energy use would be at optimal levels, with the price (or the consumer's willingness to pay) for the marginal unit of energy used reflecting the incremental resource cost of supply to the national economy. Relative fuel prices should also.influence the pattern of consumption in the di- rection of the optimal or least-cost mix of energy sourEes required to meet future demand. Distortions and constraints in the economy neces- sitate the use of shadow prices and economic second-bestadjustments, as described in the next section. Second, the social objective recognizes every citizen's basic right to be supplied with certain minimum energy needs. Given the exis- tence of significant numbers of poor consumers and also wide dispari- ties of income, this implies subsidized prices, at least for low-income consumers. Third, the government would be concerned with financial objectives relating to the viability and autonomy of the energy sector. This would usually be effected by pricing policies that permit institutions (typi- cally, government-owned) in the different energy subsectors to earn a fair rate of return on assets and to self-finance an acceptable portion of the investments required to develop future energy resources. Fourth, energy conservation is also an objective of pricing policy. While prevention of unnecessary waste is an important goal, other reasons often underlie the desire to conserve certain fuels. These in- clude the desire for greater independence from foreign sources (e.g., oil imports) and the necessity of reducing the consumption of woodfuel because of deforestation and erosion problems. Fifth, we recognize a number of additional objectives, such as the need for price stability, to prevent shocks to consumers from large price fluctuations, and the need for simplicity in energy pricing structures, to avoid confusing the public and to simplify metering and billing. Finally, there are other specific objectives, such as promoting re- gional development (e.g., rural electrification) or specific sectors (e.g., export-orientedindustries),and other sociopolitical, legal, and environ- mental constraints. In summary, therefore, price is most effective as a long-run policy tool. From the viewpoint of economic efficiency, the price indicates to An Integrated Framework for Energy Pricing / 5 suppliers the consumers' willingness to pay and the use-value of en- ergy; to the consumers, it signals the present and future opportunity costs of supply that draws on various energy sources. We conclude this section with a brief review of the pervasive role that most governments play in the pricing of commercial energy resources, and the relative neglect of issues relating to traditional forms of energy. Governments exercise direct influence, usually through the ownership of energy sources or price controls. Indirect influences occur through such means as taxes, import duties, subsidies, market quotas, taxes on energy-using equipment, and government-guided investments in en- ergy resources. In practically all developing countries, the electric utility is govern- ment owned. In oil and gas production, refining, and distribution, as well as in coal mining, both public and private organizations operate, often side by side. However, irrespective of the form of ownership, all governments exercise some form of wholesale or retail price control, usually at several levels, including during production, during refining, after transport or transmission, and so on. Income and excise taxes are also levied from both public and private energy sector companies. Generally, certain fuels in specific uses tend to be subsidized, al- though leakages and abuses of subsidies by nontargeted consumer groups also occur. Thus kerosene for lighting and cooking, rural elec- tricity for lighting and agricultural pumping, and diesel fuel for trans- portation commonly qualify for subsidies. Cross-subsidies exist be- tween different fuels, user groups, and geographic regions; therefore high-priced gasoline may finance the subsidy on kerosene, industrial electricity users may subsidize household consumers, and a uniform national pricing policy usually implies subsidization of energy users in remote areas by those living in urban centers. The principal problem associated with subsidies is that the energy producer may not be able to raise sufficient revenues to finance investment to meet expanding de- mand, or even to maintain existing facilities, and thus shortages even- tually result. Furthermore, cross-subsidies give consumers the wrong price signals, with consequent misallocation of investments. Import and export duties, excise taxes, and sales taxes are levied, often by several levels of government, from federal to municipal, at various stages in the production, processing, distribution, and retailing chain. In many developing countries, the combined levies are several hundred percent of the originalproduct price for some items, and nega- tive or close to zero for others. Several less obvious methods, such as property taxes, water rights and user charges, and franchise fees are also used to influence energy use. Energy prices are also affected by the wide range of royalty charges, profit sharing schemes, and exploration 6 / The Energy Journal agreements that are made for the development of oil and gas resources between governments and multinational companies. Other policy instruments are often used to reinforce pricing policies, such as quotas on imported or scarce forms of energy, coupled with high prices. Conservation regulations may affect depletion rates for oil and gas, while the availability of hydropower from some nlultipurpose dams may be subordinate to the use of water for imgation or river navigation. Many special policies involving tax holidays and conces- sion, import subsidies, export bonuses, government loans or grants, high taxes on large automobiles, etc., are also used to affectenergy use. The traditional fuels subsector has been relatively neglected because transactions involving these forms of energy are usually of a noncom- mercial nature. However, there is growing acceptance of the coordi- nated use of indirect methods such as displacement of fuelwood used in cooking by subsidizing kerosene and LPG, increasing the supply of fuelwood by re-afforestation programs and effective distribution of charcoal, enforcing stiffer penalties for illegal felling of trees, and proper watershed management. ECONOMIC FRAMEWORK Becausethe objectives mentionedabove are often not mutually consistent, a realistic integrated energy pricing structure must be flex- ible enough to permit tradeoffs among them. To allow this flexibility, the formulation of energy pricing policy must be carried out in two stages. In the first stage, a set of prices that strictly meets the eco- nomic efficiency objective is determined, based on a consistent and rigorous framework. The second stage consists of adjusting these effi- cient prices (established in the first step), to meet all the other objec- tives. The latter procedure is more ad hoc, with the extent of the ad- justments being determined by the relative importance attached to the different objectives. In the rest of this section, we discuss the impor- tance of shadow pricing and develop the economic framework that permits the efficient pricing of energy. The second stage adjustments due to noneconomic factors are discussed in the next section. Shadowpricing theory has been developed mainly for use in the cost- benefit analysis of projects. ' However, since investment decisions in the energy sector are closely related to the pricing of energy outputs, for consistency the same shadow pricing framework should be used in 1. For general use of shadow prices in developing countries, see Squire and van der Tak (1975);and Little and Mirrless (1974).For more specific application to the energy sprtnr see Munasin~hell979b). rhan. 9. An Integrated Framework for Energy Pricing / 7 both instances. Shadow prices are used instead of market prices (or private financial costs), to represent the true economic opportunity costs of resources. In the idealized world of perfect competition, the interaction of atomistic profit-maximizing producers and atomistic utility-maximiz- ing consumers yields market prices that reflect the correct economic opportunity costs, and scarce resources including energy will be effi- ciently allocated. However, in the real world, distortions may result from monopoly practices, external economies and diseconomies (which are not internalized in the private market), interventions in the market process through taxes, import duties, and subsidies, etc., and these distortions cause market prices for goods and services to diverge sub- stantially from their shadow prices or true economic opportunity costs. Therefore, shadow prices must be used in investment and output pric- ing decisions to ensure the economically efficient use of resources. Moreover, if there are large income disparities, we will see later that even these "efficient" shadow prices must be further adjusted, espe- cially to achieve socially equitable energy pricing policies for serving poor households. It is important to realize that lack of data, time, and manpower re- sources, particularly in the LDC context, will generally preclude the analysis of a full economy-wide model when energy-related decisions are made.2Instead, the partial approach shown in Figure 1 may be used, where key linkages and resource flows between the energy sector and the rest of the economy, as well as interactions among different energy subsectors, are selectively identified and analyzed, using appro- priate shadow prices such as the opportunity cost of capital, shadow wage rate, and marginal opportunity cost for different fuels. In prac- tice, surprisingly valuable results may be obtained from relatively simple models and assumptions. To clarify the basic concepts involved in optimal energy pricing we first analyze a relatively simple model. Next the effects of more com- plex features are examined, including short-run versus long-run dy- namic considerations, capital indivisibilities, joint output cost alloca- tion, quality of supply, and price feedback effects on demand. The process of establishing the efficient economic price in a given energy 2. This holistic approach or general equilibrium analysis is conceptually important. For example, the efficient shadow price of a given resource may be represented by the change in value of aggregate national consumption or output, due to a small change in the availability of that resource. A more detailed discussion of general versus partial equilibrium in relation to energy sector analysis is given in Munasinghe (1979b).For a discussion of economy-wide energy models, see, for example, National Academy of C~;onmc1 1 Q7Q\ 8 / The Energy Journal Energy subsector ~nteractions Figure 1. Partial equilibrium framework for energy pricing. subsector may be conveniently analyzed in two steps (seeAppendix A for details). First, the marginal opportunity cost (MOC) or shadow price of supply must be determined. Second, this value has to be fur- ther adjusted to compensate for demand-side effects arising from dis- tortions in the prices of other goods, including other energy substi- tutes. From a practical viewpoint, an optimal pricing procedure that begins with MOC is easier to implement, because supply costs are generally well defined (from technological-economic considerations), whereas data on the demand curve are relatively poor. Suppose that the marginal opportunity cost of supply in a given energy subsector is the curve MOC(Q)shown in Figure 2. For a typical nontraded item like electricity, MOC that is generally upward sloping is calculated by first shadow pricing the inputs to the power sector and then estimating both the level and structure of marginal supply costs (MSC)based on a long-run system expansion pr~gram.~For tradable items likecrude oil and for fuels that are substitutes for tradables at the 3. The same model will be modified in the next section to establish socially equi- table subsidized prices for low-incomeconsumers. 4. For a detailed discussion of the procedures used in the electric power subsector, see Munasinghe (1979a).In this subsectorMSCis alsocalled the long-runmarginal cost ". 8 7 S", An Integrated Framework for Energy Pricing / 9 Figure 2. Efficient pricing with shadow prices. margin, the international or border prices of the tradables (i.e., c.i.f. price of imports or f.0.b. price of exports, with adjustments for internal transport and handling costs) are appropriate indicators of MOC. "or most developing countries, such import or export MOC curves will generally be flat or perfectly elastic. Other fuels such as coal and nat- ural gas could be treated either way, depending on whether they are tradables or nontraded. TheMOC of nonrenewable, nontraded energy sources will generally include a "user cost" or economic rent compo- nent, in addition to the marginal costs of production. The economic values of traditional fuels are the most difficult to determine,because in many cases there is no established market. However, as discussed later, they may be valued indirectly on the basis of the savings they allow on alternative fuels such as kerosene, the opportunity costs of labor for gathering firewood,and/or the external costs of deforestation and erosion. Thus, for a nontraded formofenergy,MOC is the opportunity cost of inputs used to produce it plus a user cost where relevant, while for a tradable fuel or a substitute, MOC represents the marginal foreign exchange cost of imports or the marginal export earnings foregone. In each case, MOC measures the shadow-priced economic value of alter- native output foregone because of increased consumption of a given form of energy. After identifying the correct supply curve, we next examine demand-side effects, especially second best corrections that 5 . We note that the use of border prices does not require the assumption of free trade, but implies that the numeraire or unit of value for shadow pricing is essentially uncommitted foreign exchange (but converted into local currency at the official ex- change rate). For details. see Squire and van der Tak (1975). 6. A nontraded item is generally characterized by a domestic supply price that lies shnvo tho f n h nriro nf pvnnrts hllt holnw tho r i f nriro nf imnnrts 10 / The Energy Jourrtal capture interactions between different energy subsectors. This second step is just as important as the first one, and therefore it will be exam- ined in some detail. In Figure 2, the market-priced demand curve for the form of energy under consideration is given by the curve PD(Q), which is the con- sumers' willingness to pay. Consider a small increment of consumption A Q at the market price level p. The traditional optimal pricing ap- proach attempts to compare the incremental benefit of consumption due to AQ, that is, the area between the demand curve and x-axis, with the corresponding supply cost, that is, the area between the supply curve and x-axis. However, since MOC is shadow priced, PD must also be transformed into a shadow-priced curve to make the comparison valid. This is done by taking the increment of expenditure p.AQ and asking what is the shadow-priced marginal cost of resources used up elsewhere in the economy if the amount p. AQ (in market prices) is devoted to alternative consumption (andlor invest- ment). Suppose that the shadow cost of this alternative pattern of expendi- ture is b(p.AQ),where b is called a conversion factor. Then the trans- formed P D curve, which represents the shadow costs of alternative consumption foregone, is given by b.PD(Q);in Figure 2, it is assumed that b < 1.Thus at the pricep , incremental benefits EGJL exceed in- cremental costs EFKL. The optimal consumption level is Qopt, where the MOC and b.PD curves cross, or equivalently where a new pseudo- supply curve MOC/b and the market demand curve P D intersect. The optimal or efficient selling price to be charged to consumers (because they react only along the market demand curve PD, rather than the shadow-pricedcurve b.PD) will be pe = MOC/b at the actual market clearing point B. At this level of consumption, the shadow costs and benefits of marginal consumption are equal, that is, MOC = b.PD. Since b depends on user specific consumption patterns, different values of the efficient pricep, may be derived for various consumer categories, all based on the same value of MOC. We clarify the foregoing by con- sidering several specific practical examples. First, suppose that all the expenditure (pAQ)is used to purchase a substitute fuel; that is, assume complete substitution. Then the con- version factor b is the relative distortion or ratio of the shadow price to market price of this other fuel. Therefore pe=MOC/b represents a specific second-best adjustment to the MOC of the first fuel, to com- pensate for the distortion in the price of the substitute fuel. Next, con- 7. The general theorem of the second-best shows, for example, that if the price of a given fuel is not set at its MOC, then the efficient price of a close substitute also must diverge from its own MOC. For a detailed discussion of the theory of the second-best in economics, see Winch (1971). An Integrated Framework for Energy Pricing / 11 sider a less specific case in which the amount (pA Q) is used to buy an average basket of goods. If the consumer is residential, b would be the ratio of the shadow price to the market price of the household's market basket (here, b is also called the consumption conversion factor). The most general case would be when the consumer was unspecified, or detailed information on consumer categories was unavailable, so that b would be the ratio of the officialexchange rate (OER)to the shadow ex- change rate (SER),which is also called the standard conversion factor (SCF).' This represents a global second-best correction for the divergence between market and shadow prices averaged throughout the economy.l o EXTENSIONS OF THE BASIC MODEL The analysis so far has been static. However, in many instances the situation with regard to the availability of a given energy source, interfuel substitution possibilities, and so on tends to vary over time, thus leading to disequilibrium in certain fuel markets, anddivergenceof the short-runprice from the long-runoptimal price. This aspect is illus- trated below by means of an example that shows how the optimal depletion rate and time path for MOC of a domestic nonrenewable resource will be affected by varying demand conditions, especially tra- dability, extent of reserves, and substitution possibilities. 8. For example, MOCEL could represent the long-run marginal cost of rural elec- tricity (for lighting),and the substitute fuel could be imported kerosene. Suppose that the (subsidized)domestic market price of kerosene is set at one-half its import (border) price for sociopolitical reasons. Then b = 2, and the efficient selling price of electricity p, = MOCEL/~(ignoringdifferences in the quality of the two fuels, and capital costs of conversion equipment such as light bulbs, kerosene lamps, and partial substitution effects; a more refined analysis of substitution possibilities would have to incorporate these additional considerations). It would be misleading, however, to then attempt to justify the subsidized kerosene price on the basis of comparison with the newly calcu- lated low price of electricity. Such circular reasoning is far more likely to occur when pricing policies in different energy subsectors are uncoordinated, rather than in an inte- grated energy pricing framework. We note that all these energy sector subsidies must be carefully targeted to avoid leakages and abuses, as discussed in the next section. 9. Note that, with the foreign exchange numeraire, conversion of domestic price values into shadow-price equivalents by application of the SCF to the former is con- ceptually the inverse of the traditional practice of multiplying foreign currency costs by the SER (instead of the OER) to convert to the domestic price equivalent. 10. For example, suppose the border price of imported diesel is 4 pesos per liter (i.e., US$0.20 per liter, converted at the OER of 20 pesos per US$).Let the appropriate SER that reflects the average level of import duties and export subsidies be 25 pesos per US$. Therefore SCF = OER/SER = 0.8, and the appropriate strictly efficient selling price of diesel is p, = 4/03 = 5 pesos per liter. 12 / The Energy Journal Suppose that the present-day marginal supply cost (MSC)(including extraction costs, and additional transport and environmental costs, etc., where appropriate) of a domestic energy source such as coal lies below the thermal equivalency price of an internationally traded fuel (e.g., petroleum or high-quality coal),as indicated by points A andB in Figure 3.l The international energy price that acts as the benchmark is 1 assumed to rise steadily in real terms, along the path BE. Let us first examine two polar extremes based on simple, intuitively appealing arguments. First, if the reserves arepractically infinite and the use of this fuel at the margin will not affect exports or substitution for imports of traded fuels, then the MOC of the domestic energy source in the long run would continue to be based on the marginal supply cost, that is, along the path AC, which is upward sloping to allow for increases in real factor costs or extraction costs. On the other hand, suppose there is a ready export market for the indigenous resource, or substitution possi- bilities with respect to imported fuels. In this case the marginal use of this resource will reduce export earnings or increase the import bill for Figure 3. Dynamic price paths for domestic exhaustible energy. 11. Thermal equivalents are defined as the unit quantities of two substitutable fuels that provide the same useful energy output in a given use (i.e.,including the efficiency of conversion). We note that the choice between the energy forms would depend on the quality of the final heat output, capital and handling costs of conversion, and so on, but . . . . . .. A , f 7 1 An Integrated Framework for Energy Pricing / 13 the international fuels in the short run, because the reserves are small or output capacity is limited. Then, the marginal opportunity cost would tend to follow the path AD and rise quickly toward parity with the international energy price. The actual situation is likely to fallbetween these two extremes, thus yielding alternative price paths such as AFE, or AGHE. Here, the initial use of the resource has no marginal impact on exports or import substitution, but there is gradual depletion of finite domestic reserves over time, and eventual transition to higher-priced fuels in the future. For a given volume of reserves, the rate of depletion of the domestic energy source will be greater, and the time to depletion will be shorter if its price is maintained low (i.e.,on the path AGHE) for as long as pos- sible rather than when the price rises steadily (i.e., along path AFE). The macroeconomic consequences of the path AGHE are also more un- desirable because of the sudden price increase at the point of transition, when the domestic resource is exhausted. In practice the price path may well be determined by noneconomic factors. For example, the price of newly discovered gas or coal may have to be kept low for some years to capture the domestic market and displace the use of imported liquid fuels (whichcontinue to be subsidized for political reasons). In general, the desire to keep energy prices low as long as possible must be balanced against the need to avoid a large price shock in the future. More rigorous dynamic models, which maximize the net economic benefits of energy consumption over a long period, have been developed to determine the optimal price path and depletion rate; however, these models depend on factors such as the socialdiscount rate, the size of re- serves, the growth of demand, and the cost and time lag needed to develop a backstop technology (which could replace the international energy price as the upper bound on price). Uncertainties in future supply and demand- such as the possibility of discovering new energy resources or technologies-add to the complexities of dynamic analy- sis. The classical argument developed by Hotelling (1931) indicates that the rate of increase in the optimal rent (ordifference between price and marginal extraction cost) for the resource should equal the rate of return on capital.(7).l This implies that the optimal path of MOC would 3 be IJE in Figure 3, defined at any time t by 12. The preceding discussion is more useful for oil-importing or energy deficit LDCs. In the case of major oil exporters, the ability to influence the world market price and to determine the rate of resource depletion provides much greater flexibility. The huge for- eign exchange surpluses and limited capacity to absorb investment imply decreased at- tractiveness of marginal export earnings coupled with the need to conserve oil resources. There is also greater ability to subsidize domestic oil consumption to meet basic needs and to accelerate economic development by increasing investment and expanding nonoil gross domestic product. See, for example, Samii (1979),pp. 16-26. .,, , . , ., . . .. 1 , 0 8 1 , 8 I. , 14 / TheEnergy Journal whereJL is the rent at the time of depletion T. Thus MOC consists of l4 the current marginal costs of extraction, transport, environmental degradation, and soon (MSC),plus the appropriately discounted "user cost" or foregone surplus benefits of future consumption (JL).As T approaches infinity, IJ would tend toward AC, which is the infinite reserve case, while as T falls to zero, IJ would approximate AD more closely, corresponding to the case of very small reserves and rapid transition to the expensive fuel. We now consider another type of dynamic effect due to the growth of demand from year 0 to year 1, which leads to an outward shift in the market demand curve fromPD, to PD, as shownin Figure 4. Assuming that the correct market clearing pricep, was prevailing in year 0, excess demand equal to GK will occur in year 1. Ideally, the supply should be increased to Q ,and the new optimum market clearing price established atp ,.However, the available information concerning the demand curve PD, may be incomplete, making it difficult to locate the point L. Figure 4. Dynamic effects due to demand curve shifts. 14. However, Heal (1976) shows that under certain conditions the optimal rent will fall to zero at the time of depletion. Other recent papers explore a variety of conditions, rr Uowt..,;rL 1 1 0 7 Q b mrr 9 A 7 . 7 K A anrl narrierm1 1 0 7 Q \nn 9r\r\.QC7 An Integrated Framework for Energy Pricing / 15 Fortunately, the technical-economic relationships underlying the production function or known international prices usually permit the marginal opportunity cost curve to be determined more accurately. Therefore, as a first step, the supply may be increased to an inter- mediate level Q', at the pricep'. Observation of the excess demand MN indicates that both the supply and, if necessary, also the marginal cost price should be further increased. Conversely, if we overshoot L and end up in a situation of excess supply. then it may be necessary to wait until the growth of demand catches up with the oversupply. In this iterative manner, it is possible to move along the MOC curve toward the optimum market clearing point. As we approach it, note that the optimum is also shifting with demand growth, and therefore we may never hit this moving target. However, the basic guideline of pegging the price to the marginal opportunity cost of supply and expanding output until the market clears is still valid. Next, we examine the practical complications raised by price feed- back effects. Typically, a long-range demand forecast is made assum- ing some given future evolution of prices, a least-cost investment program is determined to meet this demand, and optimal prices are computed on the basis of the latter. However, if the estimated optimal price that is to be imposed on consumers is significantly different from the original assumption regarding the evolution of prices, then the first-roundprice estimates must be fed back into the model to revise the demand forecast and repeat the calculation. In theory, this iterative procedure could be repeated until future demand, prices, and MOC estimates become mutually self-consistent. In practice, uncertainties in price elasticities of demand and other data may dictate a more pragmatic approach in which the MOC would be used to devise prices after only one iteration. The behavior of demand is then observed over some time period and the first-round prices are revised to move closer to the optimum, which may itself have shifted as described earlier. When MOC is based on marginal production costs, the effect of capi- tal indivisibilities or lumpiness of investments causes difficulties in many energy subsectors. Thus, owing to economies of scale, invest- ments for electric power systems, gas production and transport, oil re- fining, coal mining, reforestation, and so on tend to be large and long- lived. As shown in Figure 5, suppose that in year 0 the maximum si~pplycapacity is Q, while the optimum price and output combination (p,,,Qo) prevails, corresponding to demand curve Doand the short-run marginal cost curve SRMC(e.g., variable, operating, and maintenance costs). As demand grows from DOto D l over time and the limit of existing ~.nnoo;txr ;P vnoohnA t h n -&on -trot h n ;no-oonrl tn rr tn olnar thn 16 / The Energy Journal market-that is, "price rationing" occurs. When the demand curve has shifted to Dzand the price isp2,capacity is increased to Q. However, as soon as the capacity increment is completed and becomes a sunk cost, price should fall to the old trend of SRMC-for example, ps is the optimum price corresponding to demand D3.Generally, the large price fluctuations during this process will be disruptive and unacceptable to consumers. This practical problem may be avoided by adopting a long- run marginal cost (LRMC)approach, which provides the required price stability while retaining the basic principle of matching willingness to pay and incremental supply costs. Essentially, the future capital costs of a single project or an investment program are distributed over the stream of output expected during the lifetime of this plant.15 This average investment cost per unit of incremental output is added to variable costs (SRMC),to yield LRMC, as shown in Figure 5.'" Another method of allocating capacity costs, known as peak load pricing, is particularly relevant for electricity and also natural gas. The basic peak load pricing model shown in Figure 6 has two demand curves; for example,Dpk could represent the peak demand during the x daylight and evening hours of the day when electric loads are large, while Dop would indicate the off-peak demand during the remaining (24x)hours when loads are light. The marginal cost curve is simplified assuming a single type of plant with the fuel, operating, and main- tenance costs given by the constant a, and the incremental cost of capacity given by the constant b. The static diagram has been drawn to indicate that the pressure on capacity arises due to peak demand Dpk, while the off-peakdemandDop does not infringe on the capacity Q. The optimal pricing rule now has two parts corresponding to two distinct rating periods (i.e., differentiated by the time of day): peak period price ppk = a + b off-peak period price Pop = a 15. For example, capital costs could be annualized at the appropriate social discount rate and divided by the annual output, or an average incremental cost approach could be used; see Munasinghe (1979a). If continued demand growth is expected, consumers' initial willingness to pay a price equal to the annual equivalent LRMC is assumed to imply willingness to do so over the lifetime of the asset. 16. Exceptions to theLRMC rule may lead to efficiency gains in certain cases. If sub- stantial excess supply capacity exists, it could be appropriate to temporarily use SRMC (including both variable and user costs) as a basis for pricing to specific consumers. However, SRMC priced supplies must be decreased asLRMC priced demand grows, and the temporary users of low-priced supplies should not be permitted to become a perma- nent burden. e.e.. an interm~tibleload in electric Dower svstems. An Integrated Framework for Energy Pricing / 17 Unit price Figure 5. The effect of capital indivisibilities on price. The logic of this simple result is that peak period users, who are the cause of capacity additions, should bear full responsibility for the capacity costs as well as fuel, operating, and maintenance costs, while off-peak consumers pay only the latter costs.17Peak load pricing can also be applied in different seasons of the year. Related problems of allocating joint costs arise in other energy sub- sectors as well-an example is the allocation of capacity costs of natu- ral gas, or of refinery costs among different petroleum products. The former may be treated like the electricity case. For oil products, the light refinery cuts that are tradable, such as kerosene, .gasoline, and diesel, have benchmark international prices. However, other items like heavy residual oils may have to be treated like nontradables. Further- more, associated gas that may be flared at the refinery is often assumed to have a low MOC,although subsequent storage and handling for use as LPG will add to the costs. A more complicated approach would be to use a programming model of a refinery to solve the dual problem as a means of determining shadow prices of distillates. A more general aspect of the capacity constraint, which encompasses 17. The most recent peak load pricing models indicate that in an optimally planned system, marginal capacity costs should be allocated in proportion to marginal shortage costs during two or more different rating periods. We note that if the peak period is too narrowly defined, peak load pricing may shift the peak to another rating period; this would be an extreme case of price feedback effects, which were discussed earlier. 18 / The Energy Journal 1 'Do'? I 5 0 0 ave. kWh/h Figure 6. Peak load pricing model. peak load pricing, is that energy prices have to be structured. For ex- ample, the MOC shown in Figure 2 may vary by the type of consumer, geographic location, time and level of consumption, voltage level (for electricity),and so on. These values of MOC then have to be modified to reflect demand-side considerations (as discussed earlier). Therefore, the economically efficient prices in a given energy subsector may exhibit considerable structuring. The interrelated issues of supply and demand uncertainty, safety margins, and shortagecosts also raise complications. We first illustrate this issue using electricity as an example, and then generalize the results for the other subsectors. Thus the least-cost system expansion plan to meet an electricity demand forecast is generally determined assuming some (arbitrary)target level of system reliability-e.g., loss- of-load probability (LOLP),reserve margin, etc. Therefore, marginal costs depend on the target reliability level, when in fact economic theory suggests that reliability should also be treated as a variable to be optimized, and both price and capacity (or equivalently, reliability) levels should be optimized simultaneously. The optimum price is the marginal cost price as described earlier, while the optimum reliability level is achieved when the marginal cost of capacity additions (to im- prove the reserve margin) are equal to the expected value of economic cost savings to consumers due to electricity supply shortages averted by those capacity increments. These considerations lead to a more An Integrated Framework for Energy Pricing / 19 generalized approach to system expansion planning, as shown below. l8 Consider a simple expression for the net benefits (NB)of electricity consumption, which is to be maximized: where TB = total benefits of consumption if there were no outages; SC = supply costs (i.e., system costs);OC = outage costs (i.e., costs to consumers of supply shortages);D = demand; and R = reliability. In the traditional approach to system planning (i.e., least-cost system expansion planning),both D and R are exogenously fixed, and therefore NB is maximized when SC is minimized. However, if R is treated as a variable, is the necessary first-order maximization condition. Assuming aD/ aR =0, We have: Therefore, as described earlier, reliability should be increased by adding to capacity until the above condition is satisfied. An alternative way of expressing this result is that since TB is independent of R, NB is maximized when total costs, TC = (SC + OC), are minimized. The above criterion effectively subsumes the traditional system planning rule of minimizing only the system costs. l g We note that this approach may be generalized for application in other energy subsectors. Thus while sophisticated measures of relia- bility like LOLP do not exist outside the power subsector, the concept of minimizing total costs to society is still relevant. For example, in oil and gas investment planning, the costs of shortages due to gasoline queues, lack of furnace oil, or gas for domestic and industrial use may be traded off against the supply costs of increased storage capacity and greater delivery capability incurred by augmentingsurfacetransport or 18. For details see Munasinghe (1980a). 19. The emphasis on outage costs requires greater effort to measure these costs; see M ~ ~ n a ~ i n randh ~ r r.ollorenn 11 '4791. and M ~ ~ n a s i n o h(1o9Rnhl 20 / The Energy Journal pipeline systems. Clearly, these additional considerations would modify the marginal costs of energy supply and thus affect optimal pricing policies. Finally, externalities, especially environmental considerations, have to be includedas faras possible in the determination of efficient energy prices. For example, if the building of a new hydroelectric dam results in the flooding of land that has recreational or agricultural value, or if urban transportation growth leads to congestion and air pollution, these costs should be reflected in MOC.20 While such externality costs may, in certain cases, be quite difficult to quantify, they may already be included (atleast partially) on the supply side, in terms of measures taken to avoidenvironmentaldegradation, for example, the cost of pol- lution control equipment at an oil refinery or coal-burning electricity plant, or the cost of landscaping strip-mined land. Estimation of environmental costs is most problematic in the case of noncommerical or traditional energy sources such as woodfuel, where marginal opportunity costs could be based (when appropriate) on the externality costs of deforestation, erosion,loss of watershed, and so on. Other measures of the economic value of traditional fuel would include the opportunity cost of labor required to collect woodfuel, or the cost savings from displaced substitute fuels such as kerosene and LPG. ADJUSTMENTS TO EFFICIENT PRICES TO MEET OTHER OBJECTIVES Once efficient energy prices have been determined, the second stage of pricing must be carried out to meet social, financial, political, and other constraints We note that efficient energy prices deviate from the prices calculated on the basis of financial costs, because shadow prices are used instead of the market prices. This is done to correct for distortions in the economy. Therefore, the constraints that forcefurther departures from efficientprices (inthe second stage of the pricing procedure) may also be consideredas distortions that impose their own shadow values on the calculation.2 1 20. The latter case could form the basis for raising fuel prices, vehicle licencing fees, road user charges, parking charges, and so on in a large metropolis, relative to rural areas. 21. For details, see Munasinghe (1979a). An Integrated Framework for Energy Pricing / 21 Subsidized Prices and Lifeline Rates Sociopolitical or equity arguments are often advanced in favor of sub- sidized prices or "lifeline" rates for energy, especially where the costs of energy consumption are high relative to the incomes of poor house- holds. Economic reasoning based on externality effects may also he used to support subsidies, for example, cheap kerosene to reduce ex- cessive firewood use and prevent deforestation, erosion, and so on. To prevent leakages and abuse of such subsidies, energy suppliers must act as discriminating n~onopolists.Targeting specific consumer classes (forexample, poor households) and limiting the cheap price only to a minimum block of consumption are easiest to achieve, in practice, for metered forms of energy like gas or electricity. Other means of dis- crimination, such as rationing, licensing, etc., may also be required.2 2 All these complex and interrelated issues require detailed analysis. The concept of a subsidized "social" block, or "lifeline" rate, for low- income consumers has another important economic rationale, based on the income redistribution argument. We clarify this point with the aid of Figure 7, which shows the respective demand curves for energy AB and GH of low (I,)and average (I2) income domestic users, the social tariff ps over the minimum consumption block 0 to Qmin, and the effi- cient price levelp,. All tariff levels are in domestic market prices. If the actual price p = pe, the average household will be consuming at the "optimal" level Q2,but the poor household will not be able to afford the service. If increased benefits accruing to the poor have a high social value, then, although in nominal domestic prices the pointA lies below p,, the consumer surplus portion ABF multiplied by an appropriate social weight w could be greater than the shadow price of supply (see the ap- pendix for details). The adoption of the block tariff shown in Figure 7, consisting of the lifelinerateps, followedby the full tariff pe, helps cap- ture the consumer surplus of the poor user but does not affect the optimum consumption pattern of the average consumer.23 In practice, the magnitude Qmin has to be carefully determined, to avoid subsidizing relatively well-offconsumers; it should be based on acceptable criteria for identifying "low-income" groups and reasonable estimates of their minimum consumption levels (e.g., sufficient to supply basic energy requirements for the household). The level of p, 22. For example, a minimum ration of cheap kerosene for households, or a special li- cense for trucks using subsidized diesel oil and a ban on diesel-driven passenger cars. 23. This ignores the income effect that is due to reduced expenditure by the average consumer for the first block of consumption, that is, up to Qmin. 22 / The Energy Journal Unit prlce I I I o 0, Q,,,i,, Q2 w a n t ~ t y Figure 7. Economic basis for the social or lifeline rate. relative to the efficientprice may be determined on the basis of the poor consumer's income level relative to some critical consumption level, as shown in the appendix. The financial requirements of the energy sector would also be considered in determining ps and Qmin. This approach may be reinforced by an appropriate supply policy (e.g., subsidized house connections for electricity and special supply points for kero- sene). Financial Viability The financial constraints most often encountered relate to meeting the revenue requirements of the sector, and are often embodied in cri- teria such as some target financial rate of return on assets, or an ac- ceptable rate of contribution toward the future investment program. In principle, for state-owned energy suppliers, the most efficient solution would be to set the price at the efficient level, and to rely on govern- ment to subsidize losses or tax surpluses exceeding sector financial needs. In practice, some measure of financial-autonomy and self-suf- - - - - An Integrated Framework for Energy Pricing / 23 that is placed on public funds, a pricing policy that results in failure to achieveminimum financialtargets for continued operation of the sector would rarely be acceptable. The converse and more typical case, where efficient pricing would result in financial surpluses well in excess of traditional revenue targets, may be politically unpopular, especially for an electric utility. Thereforein either case, changes in revenues have to be achieved by adjusting the efficient prices. It is intuitively clear that discriminating between the various con- sumer categories, so that the greatest divergence from the marginal opportunity cost-based price occurs for the consumer group with the lowest price elasticity of demand, and vice versa, will result in the smallest deviations from the "optimal" levels of consumption consis- tent with a strict efficiency pricing regime.14In many countries the ne- cessary data for the analysis of demand by consumer categories is rarely available, so rule-of-thumb methods of determining the appro- priate tariff structure have to be adopted. However, if the energy subsector exhibits increasing costs (i.e., if marginal costs are greater than average costs),the fiscal implications should be exploited to the full. Thus, for example, electricpower tariffs (especiallyin a developing country) constitute a practical means of raising public revenues in a manner that is generally consistent with the economic efficiency objec- tive, at least for the bulk of the consumers who are not subsidized; at the same time they help supply basic energy needs to low-income groups. Similar arguments may be made in the petroleum subsector, where high prices for gasoline, based on efficiency, externality, and conservation arguments, may be used to cross-subsidize the "poor man's" fuel-kerosene, or diesel used for trans~ortation.~~ Other Considerations There are several additional economic, political, and social consider- ations that may be adequate justification for departing from a strict efficient pricing policy. The decision to provide commercialenergy like kerosene or electricity in a remote rural area (which often also entails subsidies because the beneficiaries are not able to pay the full price based on high unit costs), could be made on completely noneconomic grounds, e.g., for general sociopolitical reasons such as maintaining a viable regional industrial or agricultural base, stemming rural to urban 24. See Baumol and Bradford (1970). 25. However, a number of undesirableside effectsmay follow,such as the practice of mixing gasoline with kerosene and the substitution of diesel for gasoline. The income dist.ributioneffects may also be perverse, with the relatively wealthy diverting cheap .. . - . .. . . . . 24 / The Energy Journal migration, or alleviating local political discontent. Similarly, uniform nationwide energy prices are a political necessity in many countries, although this policy may, for example, imply subsidization of con- sumers in remote rural areas (whereenergy transport costs are high) by energy users in urban centers. However, the full economic benefits of such a course of action may be much greater than the apparent effi- ciency costs that arise from any divergence between actual and efficient price levels. Again this possibility is likely to be much more significant in a developing country than in a developed one, not only because of the high cost of energy relative to incomes in the former, but also because the available administrative or fiscal machinery to redistribute incomes (or to achieve regional or industrial development objectives by other means) is frequently ineffective. The conservation objective (to reduce dependence on imported energy, improve the trade balance, and so on) usually runs counter to subsidy arguments. Therefore, it may be necessary to restrict cheap energy to productive economic sectors that need to be strengthened, while in the case of the basic energy needs of households, the energy price could be sharply increased for consumption beyond appropriate minimum levels. In other cases, conservation and subsidized energy prices may be consistent. For example, cheap kerosene might be re- quired, especially in rural areas, to reduce excessive woodfuel consump- tion and thus prevent deforestation and erosion. It is particularly difficult to raise prices to anywhere near the effi- cient levels where low incomes and a tradition of subsidized energy have increased consumer resistance. In practice, price changes have to be gradual, in view of the costs that may be imposed on those who have already incurred expenditures on energy using equipment and made other decisions, while expecting little or no change in traditional energy pricing policies. At the same time, a steady price rise will prepare consumers for high future energy prices. The efficiency costs of a gradual price increase can be seen as an implicit shadow value placed on the social benefits that result from this policy. Finally, owing to the practical difficultiesof metering, price discrimi- nation, and billing, and the need to avoid confusing consumers, the pricing structure may have to be simplified. Thus, the number of cus- tomer categories, rating periods, consumption blocks, and so on, will have to be limited. Electricity and gas offer the greatest possibilities for structuring. The degree of sophistication of metering depends among other things, on the net benefits of metering and on problems of instal- lation and maintenance. In general, various forms of peak electricity pricing (i.e.,using maximum demand or time-of-day metering) would be particularly applicable to large-, medium-, and high-voltage indus- An Integrated Framework for Energy Pricing / 25 receiving a subsidized rate for electricity, a simple current limiting device may suffice, because the cost of even simple kwh metering may exceed the net benefits (which equal the savings in supply costs due to reduced consumption, less the decrease in consumption bene- fits). For electricity or gas, different charges for various consump- tion blocks may be effectively applied with conventional metering. However, for liquid fuels like kerosene, subsidized or discriminatory pricing would usually require schemes involving rationing and cou- pons, and could lead to leakage and abuses. SUMMARY AND CONCLUSIONS The 1970s have been characterized by increasing real costs of energy and fluctuations in relative fuel prices. This article has empha- sized the importance of comprehensive energy planning, especially an integrated approach to energy pricing, due to theconfusion arising from the often conflicting nature of national objectives, from the complexity of energy policy tools currently in use, including pricing, physical con- trols, technical methods, and public education, and fromthemany types of energy sources that may be used in a variety of applications. Energy policymakers in developingcountries face special difficulties, such as high levels of market distortion, shortages of foreign exchange and investment funds, large numbers of poor consumers whose basic energy needs must be met, and relatively greater usage of traditional fuels, in addition to the energy issues found in industrialized countries. Thus, an integrated pricing framework must begin with a clear state- ment of national objectives, and must provide a method for trading off among mutually contradictory goals. Important linkages between the energy sector and the rest of the economy, as well as interactions be- tween and activities within different energy subsectors, must be ana- lyzed using shadow prices, essentially within a partial equilibrium framework. For consistency, the shadow pricing methodology used for pricing energy sector outputs must be the same as the one used to make investment decisions. Special attention must also be paid to the hith- erto neglected area of traditional fuels. Energy pricing structures, disaggregated by energy subsector, are derived in two stages. First, the shadow-priced marginal opportunity cost (MOC)of a given form of energy is determined, based essentially on supply-sideconsiderations. For a tradable form of energy. an appro- priate measure of MOC would be the marginal cost of imports or export earnings foregone, with adjustments for local transport and handling costs. For nontraded fuels, MOC would be the marginal supply cost, nlus a user mst, mmnnnent. (in t.he rase nf nnnrenewnhle mnnnrresb 26 / The Energy Journal Next, demand-side effects including distortions in the prices of other goods, esp&ially substitute fuels, are used to derive from the MOC the strictly efficient energy price levelpe. In practice, this basic theoretical framework may be extended to cover dynamic effects relating to both supply and demand, price feedback effects, capital indivisibilities, problems of joint cost allocation, supply and demand uncertainty, shortage costs, and externalities. In the second stageof the pricing procedure, the efficient price be)is further adjusted to yield a realistic pricing structure that meets social- subsidy considerations, sector financial requirements, and other prac- tical constraints such as the need to change prices gradually, simplicity of price structure for metering and billing, and so on. Direct pricing policies are usually inapplicable in the traditional fuels subsector, due to the lack of well-developed markets for these forms of energy. Therefore, indirect methods-including augmentation of supply, the appropriate pricing of substitute fuels, improvements in the efficiencyof woodfuel energy conversion, and punitive measures for excessive use-must be used in close coordination. APPENDIX: MODEL FOR OP'TIMAL ENERGY PRICING USING SHADOW PRICES In this appendix, a general expression for the socially optimal price in the subsector for energy type A is developed based on shadow prices, to com- pensate for distortions in the economy. From the general equation, results for optimal energy pricing are derived, for cases that reflect the following: 1. a perfectly competitive economy (classicalresult) 2. efficient prices, including economic second best considerations 3. subsidized social prices or lifeline rates for poor consumers. The supply and demand for a form of energy A is shown in Figure A-1, where Sis the supply curve represented by the marginal cost of supply (evaluatedat domestic market prices),and D is the corresponding demand curve for a speci- fic consumer. Starting with the initial combination of price and consumption (p,Q),consider the impact of a small price reduction (dp),and the resultant in- crease in demand (dQ),on the net social benefits of energy A consumption. Before evaluating the net social benefit of this price change, let us define the shadow pricing framework.?' First, suppose we calculate the marginal cost of supply MC without shadow pricing, i.e., in market prices. Then up is defined as the energy A conversion factor (ACF),which transformsMCinto the corres- 26. For details, see Squire and van der Tak (1975). 27. The little triangle LFG may be neglected throughout this analysis because its :-IA- .An\ /9 ..rho- hnth An onrl All 0- rrnoll inrrommntc A n Integrated Framework for Energy Pricing / 27 Unit price Figure A-1. Supply and demand in energy subsector A. ponding real economic resource cost, i.e., with correct shadow pricing, the marginal opportunity cost is MOC = (apMC).Second, we assign a specific social weight Wc to each marginal unit of consumption (valued in market prices) of a given individual i in the economy. For example, if this user of energy A is poor, the corresponding social weight may be much larger than for a rich customer, to reflect society's emphasis on the increased consumption of low-incomegroups. Third, if the given individual's consumption of goods and services other than in the energy A subsector (valued in market prices) in- creases by one unit, then the shadow-priced marginal cost of economic re- sources used (or the shadow cost to the economy) is bi. As a result of the price reduction, the consumer is using dq units more of energy A, which has a market value of (p . dQ) (i.e., area IFGH).17However, the consumer's income has increased by the amountpQ - (p - dP) (Q +dQ), and assuming none of it is saved, this individual's consumption of other goods and services will increase by the amount (Q .dp -p dQ),also valued in market prices (i.e.,area BEFG minus area IFGH). Therefore, the consumer's total consumption- that is, energy A plus other goods-will increase by the net amount (Q dp) in market prices. This is the traditional increase in con- sumer surplus benefits. The shadow value of this increased consumption is Wc (Q dp) where Wc is the social weight appropriate to this consumer's incomelconsumption level. Next, consider the resource costs of these changes in consumption. The shadow cost of increasing the supply of energy A is (a . MC . dQ) (i.e., ap times area IJKH), and the resources used up to provi& the other additional goods consumed b, (Q dp - p dQ),where an is the conversion factor for . . 28 / The Energy Journal energy A, and bc is the conversion factor for other goods consumed by this consumer. Finally, the income change of producer of energy A (if any) must also be considered, but this effect may be ignored if we assume quite plausibly that the producer is the government. The total increase in net social benefits due to the energy A price decrease is given by: Therefore: where n = (p. dQ/Q . dp) is the elasticity of demand (magnitude).The neces- sary first-order condition for maximizing net social benefits, in the limit, is d(NB)/dp = 0.This yields the optimal price level: This expression may be reduced to a more familiar form, by making some simplifying assumptions. Case 1: Perfectly competitive economy where market prices and shadow prices are the same, and income transfer effects are ignored, i.e., no social weighting. Therefore. up = Wc = bc = 1,and equation (A.l)reduces to This is the classical marginal cost pricing result, where net social benefits are maximized when price is set equal to marginal cost at the market clearing point (p,, Qc) in Figure A-1. Case 2: Income transfer effects ignored, because the marginal social benefit of consumption is equal to the marginal social cost to the economy of providing this consumption. Therefore, Wc = bc, and equation (A.1) becomes This is the optimal efficient price that emphasizes the efficient allocation of re- sources and neglects income distributional considerations. As mentioned earlier, the marginal opportunity cost of energy A (MOC)may usually be evaluated in a straightforward manner (e.g., the international border price for a tradable fuel, or, in the case of a nontradable like electricity, hv nnnlvinu tho annrnnriato sharlnw nriros tn tho least m s t mix nf twhnirnllv An Integrated Framework for Energy Pricing / 29 determined inputs used in production). However, the conversion factor bc de- pends crucially on the type of consumer involved. For residential consumers of energy A, bc represents the consumption con- version factor (CCF),which reflects the resource cost or shadow value of one (market-priced)unit of the household's marginal consumption basket. If the CCF < 1, then p; > MOC. Anor her interesting case illustrates the application of equation (A.3) to cor- n ? ~for economic second-best consideration arising from energy substitution ! ponsit)ilities. As an extreme case, suppose all expenditures diverted from energy A consumption will be used to purchase alternative energy which is sut)sidized by the government (e.g., kerosene for lighting, or diesel for auto- generation). In this case, b, is the ratio of the marginal opportunity cost of ;il~.ernativeenergy to its market price. and may be written Thus. Irorn equation (A.3),'" p; = MOC (pdMOC,) (A.4) Since the alternative energy is priced below its border marginal cost, i.e., b, > 1, then p,f < MOC also. Therefore, the subsidization of substitute energy pricfhs will result in an optimal energy A price that is below its shadow supply c.ost. I f it is not possible to determine the consumption patterns of specific con- sumer groups, b, could be defined very broadly as the average conversion factor for all energy A users, e.g., the SCF, as discussed in the text. CUSP .'I:General Equation (A.l) is the optimal energy A price when shadow prices are used, which incorporate income distributional concerns. Consider the case of a group of very poor consumers for whom we may assume W, S b,(n - 1).Therefore, equation (A.l)may be written An even greater simplification is possible if it is assumed that n = 1; thus 28. The logic of this expressionmay be clarifiedby consideringthe case when the ac- tual p > pz Then the shadow cost of one unit of expenditure on energy A is MOC/p while if this sum was used to purchase alternative energy the shadow cost would be MOCdp, Sincep > p: MOUp > MOCdp,, and the country is better off if more energy A is used instead of the alternative energy.there fore,^ shouldbe reduced top: Similar reasoningcan be used to show that if p < p: then p should be increasedto the 30 / The Energy Journal For illustration, suppose that the income/consumption level of these poor con- sumers (c)is one-thirdthe critical income/consumption level ( F ), which is like a poverty line. Then a simple expression for the social weight is Therefore, p,* = MOC/3, which is the "lifeline" rate or subsidized tariff ap- propriate to this group of low-incomeconsumers. REFERENCES Baumol, W. J., and Bradford, D. F. "Optimal Departures from Marginal Cost Pricing." American Economic Review (June 1970): 265-283. Davison, R. "Optimal Depletion of an Exhaustible Resource with Research and Development towards an Alternative Technology." Review of Econo- mic Studies (March 1978):355-367. Hartwick, J. M. "Substitution Among Exhaustible Resources and Intergener- ational Equity." Review of Economic Studies (March 1978):347-354. Heal, G. "The Relationship Between Price and Extraction Cost for a Resource with a Backstop Technology." The Bell Journal of Economics 7 (Autumn 1976):371-378. Hotelling, H. "The Economics of Exhaustible Resources." Journal of Political Economy 39 (April 1931):137-175. Little, I. M. D., and Mirrless, J. A. Project Appraisal and Planning for Deve- loping Countries. New York: Basic Books, 1974. Munasinghe, M., and Gellerson, M. "Economic Criteria for Optimizing Power System Reliability Levels." The Bell Journal of Economics 10 (Spring 1979):353-365. Munasinghe, M. "Electric Power Pricing Policy." Staff Working Paper No. 340. Washington, D.C.: World Bank, June 1979a. Munasinghe, M. The Economics of Power System Reliability and Planning. Baltimore: Johns Hopkins. 1979b. Munasinghe, M. "The Costs Incurred by Residential Electricity Consumers Due to Power Failures," Journal of Consumer Research (March 1980a): 361-369. Munasinghe, M. "A New Approach to System Planning." IEEE Transactions on Power Apparatus and Systems, vol. PAS-79. May-June 1980b. National Academy of Sciences. Energy Modeling for an Uncertain Future. Washington D.C., 1978. Samii, M. Vajed. "Economic Growth and Optimal Rate of Oil Extraction." OPEC Review 3 (Autumn 1979): 16-26. Squire, L., and Van der Tak, H. Economic Analysis of Projects. Baltimore: Johns Hopkins, 1975. Winch, D. M. Analytical Welfare Economics. Harmondsworth, England: Penguin Books, 1971. THE WORLD BANK Headquarters: 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. European Office: 66, avenue d'Iena 75116 Paris, France Tokyo Ofice: Kokusai Building, 1 - 1 Marunouchi 3-chome Chiyoda-ku, Tokyo 100, Japan The full range of World Bank publications, both free and for sale, is described in the World Bank Catalog of Publications, and of the continuing research program of the World Bank, in World Bank Research Program: Abstracts of Current Studies. The most recent edition of each is available without charge from: PUBLICATIONS UNIT THE WORLD BANK 1 R 1 R U STPFl7 N W