Chapter 3 - Economics of energy in agriculture and rural development projects

* by F.J. Hitzhusen Professor The Ohio State University

Key economic efficiency issues
Non-economic efficiency criteria
Case studies applying foregoing concepts
Towards some synthesis

Economics is generally defined as the study of how individuals and society choose, with or without the use of money, to employ scarce productive resources to produce goods and services over time and distribute them for consumption now and in the future among various competing ends of people and groups in society. Economics is a useful applied social science discipline for identification and analysis of the economic efficiency and income distribution trade-offs embodied in alternative solutions to important societal problems. This chapter focuses specifically on the choices related to the energy aspects in agriculture and rural development projects.

Prior to 1800, the world population was controlled primarily by famine and pestilence. Man was generally dependent on draft animals and wood for tillage, transportation and energy. Since 1800, the world population and energy use have increased dramatically. The discovery of fossil fuels and modern technology have greatly facilitated economic development. Increased energy efficiency, ease of attainment and a disregard for its finiteness made fossil fuels cheap energy sources that rapidly replaced previous biomass sources. What now seems evident is that the fossil fuel era will only be a brief period in the history of mankind - a "blip" on the scale of time.

The traditional fossil fuel energy sources are non-renewable and exhaustible, or stock resources that do not increase in physical quantity over time* Some, such as coal, are not significantly affected by natural deterioration. Others, such as oil and gas in cases of seepage and blow off, can be significantly affected by natural deterioration. However, concepts such as exhaustible and inexhaustible have meaning only in an economic context. Long before a given resource is physically used up or even appreciably diminished, it may be exhausted in the sense that further utilization is discontinued (due to its relative price or cost) in spite of continuing human wants. Alternatively, a resource may be Inexhaustible in the sense that utilization continues indefinitely, even though it is relatively limited in physical quantity compared with other resources.

In addition to supply limitations, there are some fundamental questions being raised relative to appropriate pricing of finite resources. It is argued Chat a combination of political expediency and the private market's inability to price external effects and non-renewable resources has created artificially cheap energy sources and minimal incentives for conservation, at least in some countries, leading to the substitution of high energy, capital intensive structures for labour. The failure of the "free" market to reflect full social costs from such externalities as oil spills, air pollution and congestion from automobiles, balance of payment deficits, and the potential disruptive economic and national security costs of an oil embargo are frequently cited reasons for higher gasoline and other energy taxes. Higher taxes on energy, particularly gasoline, may sometimes be a necessary condition, but rarely a sufficient condition for optimal energy use.

Boulding (1968) argues that the "spaceship" earth requires some revised economic principles from conventional economics, i.e. in the closed economy, throughput (production and consumption) is not a measure of success but rather something to be minimized. Economics, like biology, should evolve towards a greater consideration of the environment and understanding of the first and second laws of thermodynamics. The first law suggests that waste disposal is an integral part of production and consumption processes in energy as well as other areas. The second law supports the increased use of flow energy resources (e.g. biomass) and the development of more entropy-efficient technologies.

Key economic efficiency issues

This discussion assumes that the reader has knowledge of some basic financial accounting concepts such as fixed and variable costs, discounting, cash flow, debt service, etc. Elementary microeconomic concepts (see Appendix I Glossary), such as obtaining an optimal input when the marginal value product of each input is equal to its price, are not discussed. However, the non-economist reader might want to consult a basic microeconomics text such as Leftwich (1966)) for this and other microeconomic concepts, or may find it helpful to at least read the article by Haron (1983) where optimal input mix is discussed regarding agroforestry projects. Finally, some basic knowledge of elementary project evaluation methodology (such as found in Gittinger, 1982) is assumed, including financial vs. economic analysis, alternative decision models and the importance of the with/without project criterion.

The focus of this section is on those economic efficiency concepts most critical for analyzing the energy aspects of agricultural and rural development projects. Concepts discussed include accounting stance, discount rate, traded goods and foreign exchange, technological externalities, under- or unemployed factors, secondary benefits and costs, and risk and uncertainty.¹

¹ This discussion draws on a more generic development of several of these concepts in McCullough, Gowen, Hitzhusen and Feinstein (1987).

Accounting Stance

The concept of accounting stance is critical to project evaluation. It refers to the time (e.g. number of years) and space (e.g. territory) over which a project is to be evaluated. Social (or economic in Gittinger's framework) measures of project benefits and costs generally require a "long and wide view" to incorporate all of the relevant impacts and impacted groups. Alternatively, a financial analysis may refer to a single firm and a one-year time period.

The distinction between financial and economic¹ analysis made by Gittinger is important, but the complementarity of these analytical approaches is equally relevant. Financial analysis provides information on the profitability of a given enterprise (e.g. combustion of crop residues for energy) to individual entrepreneurs or investors and thus gives an indication of the incentive structure and potential adoption rate. Economic or social cost benefit analysis attempts to determine profitability from a societal standpoint, taking into consideration externalities or environmental costs (e.g. reduction in soil cover, resulting in soil erosion impacts downstream from project), shadow pricing of under- or unemployed factors (e.g. underemployed labour utilized in residue combustion project), currency evaluation, etc. The appropriateness of these analytical alternatives depends on the question one is asking.

¹ Some authors refer to Gittinger's economic conceptualization as social cost-benefit analysis. A few authors have used "social appraisal of projects" terminology to refer to weighting of net benefits for income distribution considerations, particularly in the economic development literature. For purposes of this chapter, financial and economic analysis will refer to private and social concepts of economic efficiency analysis, respectively. Any attempts Co include income distribution impacts will be considered income distribution analysis.

Shadow pricing is a term used to describe the process whereby economists fully monetize, where possible, society's valuation of a benefit or cost. Thus, it is a critical concept to the notion of accounting stance. An economic or social cost-benefit analysis usually includes some shadow-valued benefits or costs along with market prices for goods whose social and private opportunity costs are equal. Although shadow values admittedly are difficult to estimate, attempts to quantify these values are usually better than ignoring them. Energy-related examples in rural areas include fuelwood, charcoal, agricultural residues and various byproducts such as tree fodder. Specific estimates of shadow values for some of these products are developed in the two case studies later in this chapter.

Discount Rate

In assessing the energy aspects of agricultural and rural development projects, problems may arise with the choice of discount rate. Two major types of discount rates, based upon a private (financial) or social (economic) sector valuation, are generally used in project analyses. A private discount rate reflects either a current or constant rate (i.e. including or excluding inflation, respectively) based upon projections from the financial market (private opportunity costs). However, for economic analysis (social) purposes, it has been argued that the private sector rate is generally too high relative to the time value (social time preference) society places on money. As a case in point, future costs or uncertainty associated with nuclear waste disposal may be either heavily discounted or overlooked, particularly under a private opportunity cost of capital comparison with an energy tree farm characterized by a several-year delay in its benefit stream. For example, if the private opportunity cost of capital is 10 per cent, a one million dollar clean-up cost of a nuclear waste site 30 years in the future has a discounted present value or cost of only $8,519!

Burkhead and Minor (1971) argue that the private opportunity cost of capital is usually higher than the social time preference due to:

(1) risk and uncertainty; (2) income taxes; (3) technological externalities (diseconomies) that have not been internalized; and (4) concern for future generations. Because a single social discount rate is difficult to estimate, most analysts use sensitivity analysis of several rates or use a rule of thumb for shadow pricing the market rate of interest. Examples of suggested rules of thumb include:

1. rate on interest bearing government securities of 15 years or more maturity;

2. long-term rate on high grade corporate bonds; and

3. long-term rate on government bonds with adjustment for private investment displaced.

Others have suggested that the only alternative is a politically-determined estimate of social time preference. The concern in this case is with making sure that the political process is relatively open and free.

Traded Goods and Foreign Exchange Factors

In developing countries, many agricultural and rural development project inputs (including energy) and outputs are bought from (imports) or sold to (exports) foreign markets. While floating exchange rates are increasingly the rule rather than the exception, a number of national exchange rates exists for converting between currencies,¹ and often under- or overvalue the domestic currency -more often overvalue. Because of this price distortion, the proportion of capital and other inputs and outputs requiring foreign exchange (as well as the foreign exchange saved from import substitution activities) needs to be shadow priced. The foreign-exchange shadow value is:

Foreign-Excange Shadow Value = (Total Domestic Price) x (% Total Cost as Imports) x (Shadow Excange Factor)

¹ A middle position is represented by a number of West African countries that fix their common currency (CFA) relative to the French France, which in turn "floats".

Suppose, for example, in a rural agro-forestry project, about 70 per cent of the investments (e.g. equipment and fertilizer) are imported. If the prevailing exchange race converting domestic to foreign currency overprices domestic currency by 15 per cent, then the real purchasing power of each unit of domestic currency needs to be multiplied by a shadow exchange factor of 1.15 to reflect actual social costs. Thus, If 750 million pesos are spent in year 1, of which 70 per cent is imported, the foreign exchange values of the project's economic analysis is 604 million pesos/(750)(0.7C)(1.15)/.

A common oversight in many economic analyses is forgetting to shadow price foreign donor or development bank loans. Provided the country had to use foreign exchange to pay back these loans, such loans could be shadow valued. An estimate for the shadow exchange rate can often be readily obtained by looking at regional exchange rate Cables in periodic publications such as International Monetary Fund (IMF) Financial Statistics or Far East Economic Review where both official and market rates are presented. If published data do not exist for the market or shadow exchange rate, one might estimate the shadow exchange rate from observation of "black market" transactions of currencies.

Technological Externalities

Sometimes the implementation and/or operation of a given project can result in a change in the output of another industry, individual or economic enterprise due to a change in the usefulness of a given resource. By definition, these effects can be either positive or negative, are "external" to the project boundary and, therefore, do not directly affect the project but have important real effects on the economy. These external effects would represent a physical interdependence between the given project and a government, industry or individual affected which is not fully priced or compensated.

Energy-related examples include oil spills, acid rain damage from coal combustion, soil erosion impacts on hydroelectric plants and deforestation from indiscriminant fuelwood cutting. In the latter two examples, the foregone hydroelectric benefits due to soil sediment and the cost of establishing a sustainable forest plantation for an equivalent fuelwood supply provide monetary estimates of these externalities (see Hitzhusen and Macgregor, 1937).

If physical interdependencies do exist, the individual or economic enterprise that is experiencing the external effects of the project should be compensated in some way. If full compensation has been made in the form of taxes on the project, or adjustments in project prices or quantities in amounts that will reflect the economic loss (or gain) to the individual or other economic enterprise, the technological externality has been "internalized" or compensation has already taken place. If compensation has not been made, an uncompensated technological externality exists and needs to be included in the analysis.

An example of a negative technological externality (or external diseconomy) would be a large energy project that disposes contaminated wastes in a local river at various intervals during the day. Downstream, the disposed wastes cause (a) decreased fish catch for one enterprise - illustrating a physical interdependence, i.e. change in the output of the fishing industry; and (b) due to health concerns, restricts the hours of safe bathing, i.e. change in the utility of a resource to users, the resource being the water in the stream. Both the fishing industry and bathing are activities falling outside the project boundary, i.e. they are external to the project. Since economic analysis attempts to examine the full societal impacts of a project, these external effects need to be considered so that the fishing industry and bathers can be compensated or adjustments can be made for their losses.

The next issue to consider once a technological externality has been identified and monetized is determining who should pay. If the property rights or entitlements are such that the fishing industry and the bathers own the resource (the stream), then clearly the project should compensate (e.g. pay) the fishing industry and bathers for their losses. Alternatively, if property rights allow the project managers or owners to control and own the stream, it is not clear how or if compensation might take place. Although it might appear unfair, it may be cheaper in financial terms if the fishing industry and bathers or government regulator pay the given project to seek an alternative waste disposal method to restore the fish catch and the health rating for the stream.

Because technological externalities generally affect natural resources, a common situation exists where none of the entities involved in the technological externality actually own the resource leading to the physical interdependence due to poorly- or commonly-defined resource property rights or entitlements. Suppose in the example above no one "owns" the property rights to the stream. Once again, the method to determine "who pays" is not clear. In such a case, the fishing industry and bathers might be forced to pay the given project owners to find alternative methods of waste disposal. In addition, government regulation, litigation and taxes are other possibilities of dealing with common property rights' problems and internalizing costs to the polluter.

Under- or Unemployed Factors: Labour, Capital and Land

The prevailing labour wage rate and private market prices for other factors of production such as land and capital are always used in financial analyses. However, if such factors are being underutilized by the economy due to the existence of factors such as a minimum wage rate, excess plant capacity and zoning constraints, then an economic analysis should use a shadow price value for labour, capital and land.

The shadow pricing of labour in many renewable energy projects, e.g. women's labour for collection of fuelwood, is usually appropriate. Some analysts have assumed the opportunity cost or marginal value product of this labour to be zero and have assigned a zero shadow price. This is rarely appropriate, since women's time is in fact very often a constraint to development activities. Others have argued that labour in renewable energy production has some positive value (either seasonally or in non-market activities) which can be estimated by multi-plying the wage rate per day when labour is scarce times the number of days of scarce labour or by estimating the value of the non-market activities. In addition, adjustments may be required to handle anticipated increases or decreases in the level of unemployment (e.g. seasonally) during the life of the energy-related project.

A final point on labour shadow pricing is that it may have significant implications for the distribution of income from alternative agricultural or rural development projects or levels of project labour and capital intensity. Labour shadow prices which are significantly lower than prevailing wage rates on labour-intensive projects (such as many renewable energy projects) can substantially increase the economic rate of return of these projects when compared with more capital-intensive projects. The difference is further amplified if a significant proportion of the capital is imported and local currencies are overvalued, requiring shadow pricing of foreign exchange. With a high proportion of the labour input below average skill and income, and an increase of labour-intensive projects in the investment portfolio, a major redistribution of income could result from the shadow pricing of labour.

Secondary Multiplier Effects

If an agricultural or rural development project indirectly generates additional jobs, or increases productivity beyond the project boundary, the project may be said to have secondary or multiplier effects. For instance, construction of a hydroelectric dam may greatly expand local fishing and create a tourism industry. If, say, for each $1.00 net benefit put into the local economy from a project the economy increases by $1.20, the project's multiplier is 1.2. Because a project can have both positive and detrimental effects on the region, both secondary benefits and costs should be estimated so that a net benefit multiplier is used. Such multipliers for a proposal project are often taken from regional input/output (I/O) models for a similar project and region. Unfortunately, precision about future multipliers almost never exists due to "leakages", and the fact that I/O models rely on historic data while project analysis is concerned with future situations.

The use of secondary multipliers in project analysis is thus regarded by many economists as often misleading and theoretically unsound. For these reasons, analysts should generally refrain from using multipliers on project benefits and costs unless such effects are not also generated by the private investment alternative that may have been displaced. Usually, simply expanding project boundaries to include employment and other effects, or shadow pricing unemployed or underemployed factors, are preferable solutions to using multipliers on the net benefits. Alternatively, multipliers may be more appropriate in evaluating total income distribution and physical environmental effects (see McCullough et al, 1987).

The Effects Method of economic analysis may be superior to the IRR method of project analysis, with regard to such externalities, in certain conditions of data availability (see Chervel and Le Gall, 1978).

Risk and Uncertainty

Future outcomes of agricultural and rural development projects are not generally known with certainty, particularly the energy aspects of these projects. Future uncertainty or variability in outcomes results from factors such as changing tastes, new sources of supply, technological innovations, weather variation and political instability. It may be useful to distinguish between risk and uncertainty where the risk involves a probability function based on experience, i.e. a concept of expected value while uncertainty is not subject to objective probability determination. However, the concept of a continuum ranging from complete certainty to complete uncertainty may be more realistic.

Whatever the distinctions, uncertainty in outcomes can result in major misallocations of resources due to both the tendency of vested interests (private or public) to overstate returns and to the differences in variability of outcomes between competing projects. Anand and Nalebuff (1987) develop a detailed discussion of uncertainty as it relates to specific energy projects. For example, the future price of oil and whether that price reflects full social costs is, in turn, dependent on variations in the interest rate, new discoveries, changes in consumption patterns, cartel activity, disruptions in supply from war¹, etc. The following are suggested responses to uncertainty in energy projects:

1. there is a great need for more ex post evaluation to systematically identify sources and magnitudes of uncertainty related to energy by project type, geographic region, etc.;

2. beware of arbitrary rules of thumb such as changing the discount rate for dealing with energy-related uncertainty;

3. if one is dealing with a large number of similar government energy projects (e.g. tree farms), risk is spread out and it is sufficient to look at the expected value of outcomes. If one is looking at one large "unique" energy project (e.g. nuclear power plant), both expected value and variance of outcomes are important, but data and experience may be limited;

4. with limited historical data, it may be possible to use:

a. sensitivity analysis of low, medium and high values for prices, yields, etc.;

b. breakeven analysis to determine how much an uncertain outcome can vary before an energy project becomes economically infeasible.

¹ For example, Tyner and Wright (1978) have developed a methodology for shadow pricing for energy independence or national security by adding to the price of oil a premium for the risk of a disruption of oil imports

Non-economic efficiency criteria

Income Distribution

Analysts should also be concerned with the equity impacts (both intra- and inter-generational) of alternative energy strategies in agricultural and rural development projects. An ongoing debate in Brazil involves the use of sugarcane or cassava for the production of ethanol for mixing with gasoline. Sugarcane is grown primarily on large plantations in the South and Northeast, whereas cassava is grown on small farms in more economically-depressed areas throughout Brazil. Promotion of one crop over the other will have important income distribution impacts that need to be estimated and reported to decision-makers.

Economists use several alternative methods for handling income distribution impacts, including: (1) explicit weighting of net benefits by income class, group or region; (2) provision of alternative weighting functions and their distributional consequences to decision-makers; (3) estimation of non-weighted net benefits by income class, group or region; and (4) a constrained maximum or minimum targets approach which maximizes economic efficiency subject to an income constraint or vice versa.

An early example of providing alternative net benefit weighting functions and their distribution consequences was developed by McGuire and Garn (1969). The alternative functions are essentially estimates of the marginal utility of a dollar of net benefits to various income groups affected by a given projector programme. Value judgments are required to formulate the alternative functions but the decision on which function to choose is left to the decision-maker(s)). The estimation of non-weighted net benefits by income class (approach 3) eliminates the need for the analyst to make any value judgment on distribution weights.

The alternative of constrained maximum or minimum targets usually involves establishing a minimum acceptable distribution of net benefits to a designated low income class or group which must be met by each project or programme in the choice set. For example, one might include fuelwood projects in a choice set for further evaluation only if at least one-half of each project's benefits accrue to the poorest one-third of the residents in the target areas. The task is to then pick the most efficient project or rank the projects (meeting the minimum distribution constraint) on the basis of their economic efficiency, e.g. benefit-cost ratio or internal rate of return.

There are advantages and disadvantages of these alternative approaches for handling income distribution in the analysis of various energy aspects of agricultural and rural development projects. Key consideration in selecting an approach include: (1) data availability on income by income class with and without the project; (2) existence of a representative government including expenditure decisions and personal income tax rates from which to derive equity weights; and (3) the ability to specify and maintain distribution constraints for high target efficiency of projects.

Energy Theory of Value

Substantial methodological disagreement exists between most economists and the majority of physical and biological scientists involved in analysis of energy systems. The controversy is based primarily on quite different methods of accounting for or assigning values to energy inputs and outputs. The physical and biological science approach assigns an energy value in Joules, BTUs or kilocalories to both direct and indirect energy inputs and outputs in comparing the energy efficiency of alternative systems. This methodology is predicated primarily on the concern for the finite nature of fossil fuels (upon which industrialized societies have become heavily dependent) and their rates of depletion. Ulf Renborg (1979), a Swedish agricultural economist, describes in detail the development of this methodological approach in the case of agricultural biomass. Advocates for this method of energy accounting essentially distrust the measure of relative value provided by prices formed in markets or even by legislative intervention in markets (political shadow pricing or administered prices) for that matter.

Economists may see some similarities between this "energy theory of value" and the historical labour theory of value and make the same kind of criticism as, for example. Professor Samuelson (1964) "that a simple labour theory of value will lead to incorrect and inefficient use of both labour and non-labour resources in even the most perfect socialist society. So long as any economic resource is limited in quantity, i.e., scarce rather than free (market economies will and) socialist planners must give it a price and charge a rent for its use".

Criticism of energy analysis based on a simple "energy theory of value" does not mean that it is without utility or that market or politically-determined economic measures of energy value are without fault. Energy analysis is useful in describing the flow of energy resources (particularly non-renewable) through a choice set of agricultural and rural development projects or of comparing the net energy balance of alternative projects. Market or even politically-determined prices for energy may understate such factors as technological externalities, national security and considerations for future generations which may require shadow pricing by the analyst¹. However, a simple energy theory of value does not explicitly consider these factors either. In fact, it does not contain any other consumer preference than saving finite energy. Webb and Pearce (1975) point out that this "introduces the idea that energy as a constraint on economic activity is more important than any other constraint". Thus, policies or options with low energy input may have high total resource costs.

¹ Bishop and Heberlein (1979) argue that the willingness to pay for property rights held by others may be considerably less than the minimum level of compensation that property rights holders find acceptable. Specifically, the efficiency and income distribution consequences of a given policy or programme regarding energy options are dependent on who holds the property rights of concern.

Environmental and Social Impact

The environmental impact statement (EIS) has been primarily associated with increased environmental pollution and awareness in many countries. The EIS is Intended as a supplement to the typical cost-benefit analysis for projects that have potential environmental impacts that cannot be easily monetized. Specific to energy conversion projects, Table 1 is a modified version of a list of potential environmental impacts developed at the East-West Centre of Hawaii.

Table 1. Some Environmental Impacts of Energy Conversion

		Health	Property	Social Quality of Life	Environmental Services
Exploration
	Oil/gas	Accidents	-	Invasion of wilderness	- -
Production
	Coal mining	Accidents, black lung	Loss of farmland, subsidence	Use of aboriginal- Defaced landscape lands	Acid draining
	Offshore oil	Accidents	-	Oil on beaches	Oil as a biocide
	Hydroelectric dam	Dam collapse	Loss of farmland	Displacement Loss of wild rivers of residents	Fish passage, wildlife breeding grounds
	Microhydro	-	-	-	-
Processing
	Oil refining	Air/disease	Air crops	Smells, visibility	Pollution of estuaries
	Shale processing	Air/disease	Water consumption	Waste piles	Water pollution
Conversion
	Coal power plant	Air/disease	Air/crops, building	Noise, visibility	Acid rain, CO2 particles/climate
	Fuelwood plantation				Particulates and respiratory impacts
	Biogas	-	-	Odor	-
Transportation
Oil tanker		Fire	Fire, collision	Oil on beaches	Oil as biocide
	Electrical transmission	Electrocution	Restriction on land use	Unsightly towers
Consumption
	Automobile	Air/disease	Air/crops	Suburbanization Noise, visibility	Paved environment, heat/climate

Source: Adapted from John P. Holdren, "Energy Resources." in William W. Murdoch (ed.). Environment (Sunderland, Mass.: Sinamer Associates, 1975).

The check list type information in Table 1 is probably easier to develop for assessment of physical environmental impacts than for social impact assessment. The latter deals with impacts on such things as social infrastructure or institutions, behaviour norms, values, familial and friendship patterns, feelings of alienation, local autonomy and cultural artifacts.

It is obvious that the foregoing environmental and social impacts lack a single common denominator (similar to a monetary unit in economic analysis) for measurement. One approach has been to develop indices to scale the relative importance or uniqueness of various attributes of the natural environment and loss of community by the human population impacted by various types of projects. These and other measures of environmental and social factors become dimensions (some would say of a "social welfare function") or separate accounts in addition to financial and economic efficiency and the distribution of financial rewards to be compared by the appropriate decision-making group(s).

The important point is that there is no common denominator for the analyst to aggregate economic efficiency, income distribution, environmental and social impacts and they thus become multiple criteria for the evaluation of the energy aspects of agricultural and rural development projects.

Case studies applying foregoing concepts

The following case studies show how various energy aspects of agricultural and rural development projects have been addressed using financial, economic, income distribution and physical environmental analyis. Developing a sound analysis of alternative policy choices helps government authorities make more informed decisions. The two case studies are briefly presented to illustrate the analytical concepts in this chapter and to show policy implications from the result of the analysis. More detailed development of the analysis and discussions of each of these cases are available in the original references cited in the bibliography.

Valdesia Watershed and Hydroelectric Reservoir

The Dominican Republic constructed the Valdesia Dam in the 1970s to decrease their dependence on imported oil. By developing Valdesia and one other site in the 1970s, the government reduced the share of electricity generated at oil-burning facilities from over 90 per cent to less than 80 per cent by 1981. Sedimentation has been a major concern at the reservoir, however, and threatens to shorten the life of the generating capacity at the dam by 5 to 10 years, which results in increased importation of oil and the use of scarce foreign exchange.

The Valdesia watershed is quite rugged in many areas. Precipitation sometimes exceeds 150 cm/year. By the late 1970s, farmers had converted much of the watershed to rangeland and cropland on many steep slopes. Severe erosion has resulted on these hillsides, greatly contributing to the sediment buildup in the reservoir.

A watershed management plan for the Valdesia project was suggested to encourage soil conservation techniques. The project area was divided into four slope classes: Class A, 3 to 20 per cent slope; Class B, 21 to 35 per cent slope; Class C, 36 to 50 per cent slope; and Class D, over 50 per cent slope. These slope classes may be reasonable approximations of farmer income distribution classes, with slope Class A being the highest. The more severe slope classes suggested more restrictive land use policies, such as agro-forestry only. Class A slopes could meet erosion tolerances with mulching and contour farming. It was found that 57 per cent of the project area would not be affected by project implementation while 11 per cent would have to undergo a change in land use. In the remaining 32 per cent, soil conservation goals would be met by mulching, contour farming and range renovation.

Reductions in soil loss from project implementation and the resulting decreases in reservoir sedimentation were estimated, utilizing the Universal Soil Loss Equation and a sediment delivery ratio formula with and without the project.

The results of the financial analysis (excluding the off-site benefits of reduced reservoir sedimentation) are summarized in Table 2. Farmers in slope Class A are the only individuals on site who would benefit from the project implementation, even though farmers as an entire group show net benefits.

Table 2 Financial Analysis of Valdesia Watershed Management Project

Slope Class	Net Present Value
A	DR$ 12,500,000
B	- 550,000
C	-1,925,000
D	-1,100,000
TOTAL	DR$ 8,925,000

Note: Does not include benefits of reduced reservoir sedimentation.

When benefits due to reduced reservoir sedimentation are included in the analysis, substantially higher net benefits of soil conservation are obtained (see Table 3). These benefits were estimated by determining how many more years the hydro plant could operate with reduced sediment loads, reducing the necessity for oil imports to replace the lost hydropower. If project guidelines are followed on slope Classes A and B, net benefits to the project will increase more than DR$ 9,350,000, which represents a substantial saving in foreign exchange. One policy alternative would suggest that a portion of these increased benefits could be used to fully compensate farmers in slope Classes B, C and D (likely to be lower income farmers) for losses incurred from project implementation.

Table 3 Benefits of Reduced Reservoir Sedimentation

Slope Classes where	Dam	Net Present Value of
Project Guidelines are followed	Lifetime (years)	Benefits (DR$)
None	19	0
A,	20	1,755,000
A & B	25+	9.350.000+
D	25	9,350,000

Subri Forestry Project

In 1971, the Ghana Forestry Department embarked on an intensive reforestation programme with the aim of converting 100,000 hectares of degraded forest reserves to plantations of fast-growing species. Large areas were clear-cut and the ground cover burned, wasting an enormous amount of resources. The Government of Ghana then proposed a charcoal production and utilization project. The Subri Forest Reserve is one of the areas designated for this programme.

The objective of the Subri Forestry Project is to establish either: (1) an industrial plantation capable of supplying pulpwood for the proposed pulp and paper mill; or (2) a fuelwood plantation. A benefit-cost analysis will determine which of these objectives or scenarios is more financially and economically viable.

Several parameters and conditions are common for both project scenarios. The total project area is 2,500 hectares. Major physical operations of the project are based upon a 5-year cycle, each involving 500 hectares of land. All input and output prices are assumed to be affected equally by inflation over the 20-year life of the project. The rate of discount is taken as 10 per cent.

Scenario 1, the pulpwood scenario, involves the following major operations: (1) land survey, planning and demarcation; (2) underbrush clearing; (3) logging operations; (4) sawmill operations; (5) wood utilization; (6) charcoal-making operations; (7) nursery operations;(8) planting and maintaining seedlings; (9) harvesting the 500 hectares planted each year; (10) road construction and maintenance;(11) workshop operations; (12) building maintenance; (13) administrative costs; and (14) management costs. Operations 1, 2 and 7 are performed in years 1-5. Planting of the Gmelina spp. extends year 2 through year 5 and then maintenance of the coppice is required through year 16. Workshop operations include maintenance activities for tools and equipment.

The capital costs include purchase of heavy, medium and light equipment and buildings. The heavy equipment is used for construction, land clearing, harvesting and production, and has an estimated life of 10 years. The medium equipment includes tractors and small vehicles for clearing, logging and administration, and has a projected life of 5 years. The light equipment (handsaws, tools, waterpumps, etc.) has a technical life of 3 years. The buildings include staff housing, offices and storage facilities.

The benefits of the project include utilizing the waste products during land clearing in the first 5 years. Charcoal, sawn timber and logs are all sold from the waste products, i.e. markets and prices exist for these products. In year 6, harvesting of Gmelina spp. begins and continues through year 20.

The financial analysis results for scenario 1 in Table 4 show that the summation of the NPV is $ 6,625,000 and the IRR is 4.3 per cent.

For economic analysis of scenario 1, several adjustments have been made. First of all, the domestic currency (the Ghana Cedi) was widely believed to be overvalued at the official US$ 1 = Ghana $ 2.75 exchange rate. A shadow exchange rate was estimated to be US$ 1 = Ghana $ 4.125. This latter rate was then applied to all imported inputs.

Shadow wage rates were calculated to reflect under- or unemployment according to labour groups. Project workers were categorized as: (a) professionals; (b) skilled workers; (c) semiskilled workers; or (d) unskilled workers. Since professionals and skilled workers are scarce in Ghana, it was assumed that their salaries reflected the marginal value of their services and no adjustments were necessary. In estimating the shadow wage rate for semi-skilled and unskilled workers, the unemployment rate and the seasonal pattern of agricultural employment were both taken into consideration.

Table 4

Summary of Financial and Economic Analysis for Pulpwood and Fuelwood Scenarios: Subri Forest Reserve

Decision Criteria	Pulpwood Scenario	Fuelwood Scenario
Financial Analysis
a. Summation of NPV	6,626,000	25,680,000
b. IRR	4.3%	30.4%
Economic Analyis with labour shadow pricing
a. Summation of NPV	9,559,787	40,290,000
b. IRR	20.2.%.	40.1%
Economic Analysis with Shadow Exchange Rate
a. Summation of NPV	- 10,611,000	15,120,000
b. IRR	1.9%	21.4%
Economic Analysis with labour shadow pricing and Shadow Exchange Rate
a. Summation of NPV	267,441	30,600,000
b. IRR	10.2%	30.6%

The results of the economic analysis for scenario 1 are mixed. Unfavourable results (IRR = 1.9 per cent) are obtained when only the shadow exchange rate adjustments are made. If adjustments with the shadow wage rate are made (and not with the shadow exchange rate), the IRR equals 20.2 per cent. If both the shadow exchange rate and the shadow wage rate are applied, the results are marginally favourable (IRR = 10.2 per cent).

The framework for the fuelwood scenario (scenario 2) is similar to the pulpwood scenario. One major difference is that Cassia siamea is planted and harvested for scenario 2. In general, operating costs and capital costs are higher for scenario 2. However, the benefits for scenario 2 are also higher, especially in years 2 through 5.

For financial analysis, the IRR of scenario 2 is 30.4 per cent. The IRR is high, primarily due to the sale of lumber, timber, charcoal and fuelwood during the first few years of the project.

The results of economic analysis for scenario 2 reflect the same adjustments made for scenario 1. When imports are adjusted by the shadow exchange rate, the IRR equals 21.4 per cent. If the shadow wage rate is applied, the IRR is 40.1 per cent. If both the shadow wage rate and the shadow exchange rate are applied, the IRR becomes 30.6 per cent.

For each variation of the financial and economic decision criteria, the fuelwood scenario is a more attractive option for the Subri Forest Reserve. It is likely that the pulpwood option will have greater foreign exchange earnings but these must be compared with foreign exchange savings from any reduction in oil imports (e.g. Kerosine) from increased fuelwood production. Attempts to estimate a shadow price for charcoal (increase it), due to the local market price not reflecting full cost (replanting, etc.) of a sustainable supply, would further increase the relative economic attractiveness of the fuelwood option.

Towards some synthesis

Economists are in general agreement regarding the desirability of relative prices or rents for inputs and outputs determining the allocation of scarce resources (including energy) among competing ends. This includes a discount rate or time price for allocation over time. Some disagreement exists among economists oh the extent to which these various prices or rents, and the resulting distribution of income and/or wealth, should be determined exclusively by the market. If the market is competitive, i.e. large numbers of buyers and sellers, mobility of resources, externalities or third party effects fully internalized (priced or compensated), etc., it is usually the preferred institutional mechanism for the efficient allocation of scarce resources. In addition, most economists would opt for a direct transfer mechanism (e.g. negative income tax) for any desired adjustments in the resulting distribution of income and/or wealth.

Where markets do not exhibit competitive charateristics, some form of intervention (usually by the government), such as regulations, public ownership, taxes and subsidies, frequently results. The degree and form of this intervention is the source of considerable disagreement among economists, as well as non-economists. Economists have generally preferred taxes and subsidies over regulations or public ownership as a more efficient form of intervention. However, an increasing body of literature in natural resource economics suggests that this choice may not be either most efficient or independent of income distribution without explicit consideration of the property rights involved¹. This is particularly true in situations involving non-renewable resources and pervasive environmental externalities with important implications for future generations, i.e. common ingredients of the continuing debate over alternative energy sources and futures.

¹ See footnote on page 63.

Several implications seem clear regarding the "economic" analysis of energy in agricultural and rural development projects. First, we should not limit our search for alternative energy sources to present use technologies. New sources as well as more entropy-efficient sources and end uses need to be explored in part, because the direct and indirect costs of extraction of traditional fossil fuels will continue to rise in the future. With increased pressure on both the raw materials producing and residual assimilating capacity of the natural environment, more attention needs to be given to quantifying and pricing technological externalities (or providing supplementary physical environmental impact statements) and accounting for depletion of non-renewable resources when comparing traditional fossil fuels to renewable bioenergy alternatives.

Shadow values or prices are critical to economic analysis of energy aspects of projects, but carrying out both financial and economic analysis is also important. Financial returns to equity capital and other measures of profitability are critical for determining whether or not a new energy feedstock, conversion and end use scenario will be adopted by entrepreneurs. For example, economic shadow values for national security and environmental externalities may show a sustainably managed fuelwood plantation as an economically viable enterprise. However, financial analysis may show adoption to be unprofitable unless taxes and subsidies are implemented.

The difficulty of combining income distribution and economic efficiency analysis and the continued concern over rural poverty in most regions of the world argues for separate but complementary analysis of income distribution consequences as well as social impacts in agricultural and rural development projects. In this context, more labour intensive (as opposed to imported capital input intensive) energy production and conversion technologies may be consistent with a targeted income distribution approach, particularly in developing countries. The economic feasibility of these more labour intensive options may also increase substantially if shadow values are estimated for under- or unemployed labour, foreign exchange savings and foregone urban migration costs, as well as for non-renewable energy inputs.

A final note is in order on "the energy theory of value". Use of energy accounting as an exclusive measure of value for optimizing alternative renewable or fossil fuel energy systems has at least as many flaws as the earlier-refuted labour theory of value. Careful social cost-benefit analysis or shadow pricing for the foregoing concerns of externalities, future generations, etc., seems to be a much more defensible optimization methodology. However, given the difficulty of both estimating and implementing shadow prices, some compromise may be possible. It would seem useful to measure energy inputs and outputs, particularly of critical liquid fossil fuels for alternative systems and to use this information to supplement social cost-benefit analysis. There may even be cases where the energy accounting can be a direct input in the formulation of shadow prices. Multi-criteria analysis of project effects and long-term impact may be a complementary approach.