Economics and Statistics

	Table of Contents
p. 46 of 74 BACK | NEXT I. Introduction II. Background III. The Economic Rationale of A Project IV. Macroeconomic and Sectoral Context V. An Integrated Approach To Economic Analysis VI. Identification and Quantification of Costs and Benefits VII. Valuation of Economic Costs and Benefits VIII. Large Projects, Linkages, and National Affordability IX. Least-Cost and Cost-Effective Analysis X. Investment Criteria: Economic Viability XI. Discount Rate XII. Uncertainty: Sensitivity and Risk Analysis XIII. Sustainability of Project Effects XIV. Distribution of Project Effects XV. Projects and Policies XVI. Appendices Appendix 1: Key Questions For The Economic Analysis of Projects Appendix 2: Project Economic Rationale: Market and Nonmarket Failures Appendix 3: The Project Framework Appendix 4: Identification and Measurement of Consumer Surplus Appendix 5: Treatment of Working Capital Appendix 6: Depletion Premium Appendix 7: The Use of Constant Prices In The Economic Analysis of Projects Appendix 8: General Methodology For Building Up Project Statements Appendix 9: Economic Evaluation of Project Output and Input Appendix 10: Economic Price of Traded Goods and Services Appendix 11: Valuation of Nontraded Outputs and Inputs Appendix 12: Shadow Wage Rate and The Shadow Water Rate Factor Appendix 13: The Economic Price of Land Appendix 14: Treatment of Resettlement Components of Projects Appendix 15: Calculating Economic Prices At The Domestic Market Price Or World Market Price Levels Appendix 16: Estimating The Shadow Exchange Rate Factor and The Standard (Or Average) Conversion Factor Appendix 17: Example of An Economic Rate of Return: An Irrigation Rehabilitation Project Appendix 18: Effect On Net Foreign Exchange and Budget Flows: An Example Appendix 19: Least-Cost Analysis and Choosing Between Alternatives Appendix 20: Estimating The Economic Opportunity Cost of Capital Appendix 21: The Treatment of Uncertainty In The Economic Analysis of Projects: Sensitivity and Risk Analysis Appendix 22: User Charges, Cost Recovery, and Demand Management: An Example For Piped Water Appendix 23: Financial Returns To Project Participants: An Illustration Appendix 24: Economic Evaluation of Environmental Impacts Appendix 25: Distribution of Project Effects Appendix 26: Impact On Poverty Reduction Appendix 27: Difference Between Economic and Financial Prices Appendix 28: Use of Economic Prices In Measuring Effective Protection Appendix 29: Exchange Rate Issues In Project Analysis XVII. Others

Guidelines for the Economic Analysis of Projects : XVI. Appendices

---

Appendix 4 : Identification and Measurement of Consumer Surplus

1. Market competition is rarely perfect and large projects affect prices. The extent to which a project affects its output price depends on the price elasticity of demand for project output. If demand is less than fully elastic, the project faces a downward sloping demand curve, and buyers paying the market price for project output are paying less than what they and others would be willing to pay. The difference between the market price and what they and others would be willing to pay for project output is called consumer surplus. An increase in consumer surplus represents an increase in economic welfare.

2. The gross benefits of a large project, therefore, are made up of sales revenue and consumer surplus. Sales revenue equals the quantity of project output sold, multiplied by the actual price paid. Consumer surplus equals the sum across all consumers of the differences between the price actually paid for the project output and what buyers would be willing to pay for it. It is only when demand is perfectly elastic, such as for most internationally traded outputs, that no consumer surplus is created by an increase in project supply. In that case, the market price for project output does not fall and no consumer surplus is created. A project sometimes may create consumer surplus without reducing the price of its output. If supply is rationed at a price below what buyers would be willing to pay, an increase in supply capacity at the same price produces incremental consumer surplus.

3. Estimating consumer surplus requires knowledge of the demand curve for project output; but demand curves are difficult to estimate. Demand is often suppressed by nonprice factors, such as supply constraints. Econometric analysis using historical data is not always reliable, and the results are often controversial because estimates of the quantity demanded without the price change are counterfactual. The use of before and after price change data, instead of with and without price change data, may fail to account for shifts in the demand curve over time, and therefore mixes price consumption points on different demand curves.

4. Segments of demand curves are often built for the purpose of estimating consumer surplus. This approach is applicable where prices charged for an output are full cost recovery prices, that is, they are not depressed by subsidies to the supplier or by regulation. A segment can be drawn from two price consumption points. Alternatively, what is required is one price consumption point and an estimate of price elasticity of demand at that point.

5. The first step in measuring consumer surplus is to forecast the quantity demanded without the project (Q_WO). Over time, the demand for project output will be affected by population and income growth, as well as prices. If the price elasticity of demand, quantity demanded without the project, and the output price with and without the project (P_W and P_WO) are known, then the quantity demanded with the project (Q_W) can be estimated as part of the formula for an arc elasticity:

e = [(Q_W - Q_WO ) / ((Q_W + Q_WO ) / 2)] * [(( P_W + P_WO ) / 2) / (P_W - P_WO ))]

where e is an arc estimate of the price elasticity of demand.

6. The project will have an incremental and a nonincremental effect. Incremental project output is equal to (Q_W - Q_WO ). At the same time, the lower price of output with the project means that existing production is displaced. The total project output, Q_P , consists of the incremental output (Q_W - Q_WO ) plus the nonincremental output (Q_N) that displaces existing producers. Nonincremental output, Q_N, is, by definition, equal to (Q_N - (Q_W - Q_WO )).

7. By assuming a linear demand segment, consumer surplus (CS) from incremental output can be estimated as half the product of the difference between the price of project output without and with the project, and the quantity of incremental output (area d in Figure 1). ¹ Consumer surplus from nonincremental output is estimated by multiplying the difference between the price of project output with the project and the price of project output without the project by the quantity of nonincremental project output (area a in Figure 1). Together, this gives total consumer surplus from:

CS = ½ (P_WO - P_W )(Q_W - Q_WO ) + (P_WO - P_W )(QP - (Q_W - Q_WO ))

8. The following example shows the estimation of gross benefits, consisting of sales revenue and consumer surplus, for a rural electrification project that supplies electricity to households. In year 10, for example, the quantity of illumination demanded without the project QWO is forecast at 14,331 kwh. Without the project, households would continue using kerosene lanterns for illumination. The price of illumination (equivalent to that supplied by a 10 watt bulb) without the project (PWO ) is Rp189.6/kwh, depending on (i) the rate of kerosene consumption by a kerosene lantern (0.0076 l/hr); (ii) the price of kerosene (Rp235/l); (iii) the illumination produced by a kerosene lantern (0.01 kWh); and (iv) the cost of kerosene lantern operation other than for kerosene (Rp11/kwh). The price of electricity with the project (PW ) will be Rp140/kwh.

9. Market research in the project area shows that the price elasticity of demand for illumination is -0.8. Using the arc elasticity formula above, QW is calculated as 18,254 kwh. Consumer surplus on incremental output is Rp97,290 and consumer surplus on nonincremental output is Rp710,818. Adding total consumer surplus to sales revenue of Rp2,555,560, gives a total gross benefit in year 10 of Rp3,363,668.

10. Gross benefits, made up of consumer surplus and sales revenue, are estimated at the domestic price level. Project costs may be denominated at either the world price or the domestic price levels. To be able to compare gross benefits with project costs, tradable components valued at the border price level must be converted to the domestic price level using the shadow exchange rate factor. Alternatively, nontraded components of costs and the gross benefits must be converted to the border price level using the standard conversion factor (see Appendix 16).

11. In many projects in which consumer surplus is an important benefit source, the price of project output is set by the government, not the market. If project output is not rationed by price and demand exceeds supply, the price of project output would not reflect marginal willingness to pay. If project supply remains constant, as demand shifts over time and the quantity demanded without the project approaches the quantity demanded with the project, the magnitude of consumer surplus increases to reach a maximum, equal to the difference between the price of project output with the project and the price of project output without the project multiplied by project output (see Figure 2). Thus, in the case of many public utility projects, with markets exhibiting high levels of suppressed demand, this approach to measuring consumer surplus would be warranted.

12. In some cases, projects supply services to new markets and empirical price-quantity data are not yet available. Market surveys have been used to provide willingness to pay data in such circumstances. Surveys often take the form of a contingent valuation when a respondent is asked what he or she would do in a hypothetical situation.² For example, a consumer may be asked whether she would be willing to pay $x per cubic meter for a 24-hour piped water service. Each consumer would be asked this question for a range of tariffs. It is on the basis of a sample of households' willingness to pay bids that statistical demand curves can be estimated.³

13. Unlike private goods and services, public goods, such as rural roads, and social services, such as primary education, are not sold. With the output of projects producing such goods and services being supplied free of charge, consumer surplus accounts for all of project benefits and can be estimated using a willingness to pay approach.

____________________________________

¹ In principle, the demand curve segment should be treated as nonlinear. However, it is frequently convenient to treat the demand curve segment as linear, as in the text, especially where the CS from incremental outputs is much smaller than the CS from nonincremental output. A value of less than ½ can also be used in the text calculation to approximate the result for a nonlinear demand curve.

² Refer Water and Sanitation for Health Project, USAID, Guidelines for Conducting Willingness-to-Pay Studies For Improved Water Services in Developing Counties, Field Report No. 306, 1988.

³ The problem of uncovering the true preferences of community members cannot be underestimated. Surveys have shown many households felt that the government had a responsibility to provide them with clean water, and that understating its value would reduce the price they would be charged for it. However, most practitioners now generally agree that the greatest problem for contingent valuation arises not from strategic behavior, but from "hypothetical bias". Simply put, hypothetical bias occurs because people have little practical experience evaluating hypothetical offers and so are more likely to make mistakes when they respond to hypothetical offers than when they respond to cash offers. This does not suggest that the opinions of beneficiaries should not be solicited, but rather that care must be taken to design mechanisms that minimize potential sources of bias.

Back Appendix 3: The Project Framework	Next Appendix 5: Treatment of Working Capital