Economic and social analysis of projects and of price policy : the Morocco fourth agricultural credit project

EcoDnomic and Social Analysis of Projects and of Price Policy: Thie Morocco Fourth Agricultural Credit Project SWP-369 Woirld Bank Staff Working Paper No. 369 Janiuary 1980 Prepiared by: Kevin M. Cleaver Europe, Middle East and North Africa Projects Department Copyright ( 1980 The World Bank 181 B H Street, N.W. _____- n, D.C. 20433, U.S.A. '+' * + and interpretations in this document are those of the author not be attributed to the World Bank, to its affiliated 4 H D os s, or to any individual acting in their behalf. 1440 *f* HD1440.M8 C53 . 118 Economic and social analysis of projects and of price policy 'zI C'A971991   The views and interpretations in this document are those of the author and should not be attributed to the World Bank, to its affiliated organizations, or to any individual acting on their behalf. WORLD BANK Staff Working Paper No. 369 January 1980 ECONOMIC AND SOCIAL ANALYSIS OF PROJECTS AND OF PRICE POLICY: THE MOROCCO FOU1RTH AGRICULTURAL CREDIT PROJECT The purpose of this paper is to (a) apply the economic and social cost benefit methodology of Little-Mirrless and Squire-van der Tak to the Fourth Agricultural Credit Project in Morocco, and (b) discuss the operational implications @f the results including price policy. The paper concludes that investments in mechanization, irrigation equipment, wells, draft animals and small implements generate a higher rate of return to Morocco's economy than to farmers due to price distortions biased against agriculture. The negative effective protection for most agricultural products is causing a level of investment in agriculture lower than is economically desirable, except for livestock where milk subsidies are causing over- investment. The social rate of return to the Project is lower than the economic rate of return, and for some investments, the social analysis suggests policies exactly opposite to those suggested by the economic analysis. Some agricultural investments which are economically viable are socially marginal. The price policy which was found by the economic analysis to excessively discourage such investments, is found in the social analysis to be desirable, or not to be discouraging enough. Prepared by: Kevin M. Cleaver Europe, Middle East and North Africa Projects Department Copyright @ 1980 The World Bank 1818 H Street, N.W. Washington, D.C. 20433 U.S.A.   ACKNOWLEDGEMENTS Country parameters used in the working paper are generally modifications of those estimated by Mr. Gordon Hughes and reported on in a forthcoming World Bank Economic Report on Morocco. Modifications were made to the Hughes' parameters to make them consistent with the Squire-van der Tak presentation and methodology. The work by Mr. Hughes, and comments by Messrs. 1. Hume, C. Merat, B. Merghoub, A. Ray, and P. Scandizzo have been instrumental in the completion of this work.   TABLE OF CONTENTS Pase No. I. BACKGROUND AND SUMMARY OF THE PROJECT .... ............ 1 Introduction ......................................... 1 The Project .......................................... 1 Farm Models .......................................... 4 II. EMPLOYMENT AND FINANCIAL ANALYSIS .................... 6 Employment ........................................... 6 Financial Rates of Return ............................ 7 III. ECONOMIC ANALYSIS .................................... 7 Shadow Prices ........................................ 7 Economic Rates of Return ............................. 12 Conclusions and Relevance to Agricultural Price Policy 13 IV. SOCIAL RATE OF RETURN ANALYSIS ....................... 13 Basic Concepts ....................................... 13 Distribution Weights ................................. 16 Value of Public Income and Investment Relative to Consumption ........................................ 19 Critical Consumption Level ........................... 19 The Net Social Cost/Benefit of Incremental Household Consumption ........................................ 20 Social Cost of Input Use ............................. 22 Social Cost of Commodity Use ......................... 23 The Accounting Rate of Interest ...................... 23 Social Rate of Return Analysis ....................... 24 V. SENSITIVITY TESTING .................................. 26  Table of Contents (Continued) TABLES Accountir. Ratios 1. Accounting Ratios for Converting Market Prices to Border Prices; Output 2. Accounting Ratios for Converting Market Prices to Border Prices; Inputs Social and Economic Rate of Return and Sensitivity Analysis of Farm Models 3. Well Drilling and Purchase of Pump 4. Draft Animals and Agricultural Implements 5. Stable Construction and Purchase of Cross Bred In-Calf Heifers 6. Purchase of Livestock for Milk Production 7. Purchase of a Tractor 8. Purchase of a Tractor and a Combine Harvestor Technical Coefficients, Income Statements, Financial Analysis of Farm Models 9. Well Drilling and Purchase of Pump 10. Draft Animals and Agricultural Implements 11. Stable Construction and Purchase of Cross Bred In-Calf Heifers 12. Purchase of Livestock for Milk Production 13. Purchase of a Tractor and a Combine Harvestor  I. BACKGROUND AND SUMMARY OF THE PROJECT Introduction 1.01 The economic and social analysis of the M1orocco Fourth Agricultural Credit Project was carried out to serve two purposes including (a) appraisal of the Project, and (b) analysis of the impact of Moroccan price policy on the agricultural sector for iLnclusion in an agricultural sector memorandum. Where previous Bank working papers on the social analysis of projects have oriented their work to testing the methodology, the present exercise was intended to be operational. Preliminary results were summarized in both the Project appraisal report and in the President's Report sent to the Bank's Board of Directors. Completed results have been included in an Agricultural Sector Memorandum. Nevertheless, the parameters estimated in this paper have not yet been agreed to by the Moroccan Government, and therefore certain results, particularly those dependent on income disribution weights, must be considered as prelimi- nary. The Project 1.02 Objectives of Project. The Project will finance farm investments and small scale agro-industry through Morocco's Agricultural Credit Bank (Caisse Nationale de Credit Agricole, CNCA) during the period September 1979 to August 1982. The objectives of the Project are to (i) improve the standard of living of about 475,000 of Morocco's 1,928,000 farm families, (ii) contribute to growth of agriculturaL production and to a decline in Moroccan dependence on imported food, (iii) create about 115,000 manyears of rural employment, (iv) generate value added from agro-industrial production, and (v) expand CNCA's capability as an agricultural credit institution. A summary of Project Cost estimates is as follows where Project Cost is defined as total investment cost financed by borrowers, CNCA credit, and Government subsidies: Project Cost US$ Million Estimates: Local Foreign Total Agricultural Machinery 33.0 49.5 82.5 Wells, Pumps, Irrigation Equipment 43.1 18.5 61.6 Transport and Other Equipment 29.3 29.3 58.6 Cattle and Sheep 165.4 45.0 210.4 Stables, Sheep Folds, Storage 79.5 53.0 132.5 Plantations 20.5 13.8 34.3 Land Improvement 15.2 6.5 21.7 Draft Animals 64.6 0.0 64.6 Agro-Industry 34.3 34.4 68.7 Other Farm Investment 1.3 0.0 1.3 Monitoring and Evaluation 0.5 0.5 1.0 Total 486.7 250.5 737.2 Of which Price Contingencies 45.8 23.6 69.4  1.03 Project Execution. CNCA would be the borrower and executing agency for Project lending. Farmers borrowing from CNCA would execute the invest- ments. CNCA lends up to 70% of a farm investment cost subject to its appraisal criteria. The most important criteria include the number of hectares of farm land required for a borrower to obtain a loan for a tract(- (60 hectares) or combine harvester (120 hectares), and the number of hectares of forage required to obtain a loan for pure-bred livestock. CNCA also undertakes an assessment of financial viability of proposed investments and the debt service burden on the borrower before approving a loan. The Fourth Agricultural Credit Project is likely to be economically viable for two reasons. Firstly, faced with a positive real interest rate (8.5% compared to expected price inflation of 7 to 8% per annum), and the certainty that repayment of credit is necessary (annual repayment rates are about 90%), borrowers will demand credit, almost exclusively for financially viable investments, and will avoid investments which are excessively risky. This results from the borrower's knowledge that if his investment does not generate the resources required for him to repay the loan, he may default on the loan, and CNCA's collection procedures are such that he is certain to lose his property. CNCA's lending norms and appraisal criteria help borrowers in making these decisions. Secondly, financially viable Lnvestments with several exceptions are likely to be economically viable in Morocco because price distortions facing agriculture tend to be mutually offsetting. Fdr example, effective taxes on labor use, fertilizers, mechani- zation, and pesticides offset effective subsidies of milk and some other agricultural commodities. This can be seen by comparing financial and economic rates of return to the investments represented in the farm models to be presented below (para. 3.15). Nevertheless, price distortions facing Moroccan agriculture, which are described in the following, are potential impediments to an economically desirable allocation of resources. Changes in Government price policy are therefore recommended (para. 4.16). 1.04 Production Generated by Project. The economic evaluation of the agricultural investments to be financed by the Project is based on the analysis of farm models representative of CNCA clients (tables 9-13 attached). The farm models are drawn from information provided by CNCA's monitoring and evaluation system. CNCA collects monthly data on farm inputs and outputs for a sample of 200 farms including 175 CNCA clients, and 25 farms which do not use institutional credit. This data permitted construction of farm budgets representative of average changes in the use of farm inputs, crop and live- stock production in response to credit financed farm investment. These farm models, when expanded to represent the results of the entire Project, provide the incremental crop production projections below. The farm models show that the Project would increase crop production in Morocco through changed cropping patterns and increased crop yields resulting from Project financed wells, pumps, mechanized farming, draft animals, plantations, agricultural implements, land levelling, destoning, and other land improvements. Development of meat and milk production would result from Project investment in pure bred and cross bred cows and sheep replacing lower yielding locally bred animals, development of forage crops and animal feed, improved animal shelters, comple- mentary Government livestock services, and increased marketing and processing facilities for milk and meat. All Project production would go to satisfying increased demand for food, except for some fruit production which may be  -3- exported. The alternative to the Project would be a comparable increase in imports of wheat, vegetables, industrial crops (vegetable oils, sugar), milk and meat. As a Incremental % of Increase in Annual Production Annual Moroccan Permitted Consumption between By Project 1977 and 1985 (tons) (%) Wheat 107,000 6.5 Barley -100,000 /1 - Forage 163,000 /1 16.0 Vegetables 215,000 9.5 Industrial Crops 1,283,000 13.0 Pulses 1,000 0.7 Fruit 500,000 70.0 /2 Milk 300,000 42.0 /3 Meat 55,000 18.0 /3 /1 The reduction in barley production is compensated by the increase in forage production. Forage production is measured as '000 feed units, and is comparable to 163,000 tons of barley. /2 Includes Project fruit production as percentage of projected increase in Moroccan consumption through 1990 excluding exports. /3 Measured as a percentage of incremental Moroccan consumption from 1977 to 1990, plus imports in 1977. Full Project production would occur by 1990. 1.05 Financial, Economic and Social Analysis. The representative farm models permit financial and economic analysis of representative CNCA financed investments. Of course, actual financial and economic returns will vary between farms, with the average returns approximating those shown in the models. The models estimate the income received by farmers at the present time, project the changes in income likely to result from Project investments, quantify the benefits that will result from the Project, and demonstrate the financial incentive to farmers to invest under the Project. Financial rates of return are estimated as the value of incremental crop and livestock production resulting from Project investment; net of input costs, debt service on CNCA credit, and farmer equity investment. Market prices are used. Based on these financial models, an economic analysis of representative investments is under- taken. This is done by revaluing those outputs, inputs, and investments which are tradable on world markets at border prices (c.i.f. if importable, f.o.b. if exportable), adjusted for handling and transport costs within Morocco. Outputs and inputs which are not tradable on world markets (such as Moroccan labor, cement, transport, draft animals, etc.) are valued at the marginal  -4 - economic cost of their supply for Project purposes. 1/ Economic cost and benefit streams include the incremental net value of production less the cost of investment, valued in economic prices. Finally, a "social" analysis is undertaken based on the economic analysis, but allowing for Moroccan distri- butive ani growth objectives by attaching appropriate weights to costs and benefits. 2/ These weights are based on an analysis of Moroccan macro-economic policy as manifested in the productivity of public investments, tax and subsidy policy, foreign exchange policy, interest rates, and distribution policies benefitting the poor. The "social" analysis permits the systematic use of macro-economic constraints and policies in project analysis. 1.06 Model 1: Well Drilling and Pump (Table 9). This is a small 6 ha farm in a rainfed area. The farmer grows hard wheat, barley and maize in an area whose average rainfall is around 320 mm. Yields are low. The farmer also owns a cow (providing milk for family consumption) and 50 sheep fed either by grazing on collective land, or with straw in the winter and during dry periods. The investment (well and pump) allows the farmer to grow pota- toes on part of the land initially devoted to maize. Net farm income per capita increases from US$330 equivalent to US$430. For purposes of compari- son, the absolute rural poverty level in Morocco is estimated at US$200 per capita (1978), while per capita GNP is estimated at US$600, and is forecasted to increase to US$690 by 1982. About 40% of' CNCA's potential clientele now live below the absolute poverty level. Model 1 represents a client living above the absolute poverty level, but below average income in Morocco without the Project. About 55% of CNCA's actual clientele are in this category. The model results in an increase in family labor equal to 24 days per year, and a 50 day increase in the use of seasonal labor. Total CNCA financed investment in wells, pumps, and irrigation is expected to create about 17,000 manyears of employment. 1.07 Model 2: Two Draft Animals and Basic Agricultural Implements (Table 10). This is a small 8 ha farm in the grain-producing area of Morocco with no possibility of irrigation. One fourth of the surface is left fallow, and grain (hard wheat, barley and maize) is grown on the rest. The farmer rents a plow, harrow and draft animals for land preparation. The farmer also owns four cows (milk is sold at springtime when production exceeds family consump- tion) and 80 sheep (young male sheep are sold). Animals are fed on fallow and on collective land, with the resulting risk of feed limitation. The investment (two mules, a plow and a harrow) allows all land to be cultivated. The resulting availability of straw and stubble will help eliminate the 1/ This shadow pricing follows the approach outlined by: Lyn Squire and Herman van der Tak, Economic Analysis of Projects (Baltimore: Johns Hopkins Press, 1975),; and I.M.D. Little and J.A. Mirrless: Project Appraisal and Planning for Developing Countries (Basic Books Inc., New York, 1974). 2/ Shadow prices have been estimated by Mr. Gordon Hughes, but not yet agreed to by the Moroccan Government.  -5- feeding risk by creating feed reserves. The farmer may be a private borrower or a member of a newly-created Agrarian Reform Cooperative. Per capita farm income will increase from US$360 to US$375 equivalent. Employment on the representative farm will remain unchanged. 1.08 Model 3: Two In-Calf Heifers and a Stable (Table 11). The live- stock component of the Project, represented in farm models 3 and 4, forms a part of Government's strategy for livestock development. This strategy has been developed partly in response to the problem of overgrazing of pastureland by sheep and goats. Overgrazing is causing a decline in their milk and meat yields. Overgrazing has also caused destruction of vegetative cover, erosion, and desertification. The Government's strategy is to intensify livestock production. This will be done by concentrating herd development on farms able to grow adequate forage, introduction of high yielding livestock breeds based on imported varieties, improved technical and health services, creation of livestock cooperatives through which services will be provided and milk marketed, and creation of necessary infrastructure (milk collecting centers, processing plants, slaughterhouses, etc.). The Fourth Agricultural Credit Project would provide the medium-term credit for livestock development based on farm cultivation of forage and of industrial crops which provide residues used for animal feed. Farm model 3 represents a small 13 ha rainfed grain farm. Initially, grain (hard wheat, barley and maize) is grown and crop yields are low due to poor climatic conditions (425 mm rainfall ) and lack of fertilizers. Land is prepared by the farmer with his own draft animals and implements. Except for the two mules there is no livestock. The investment (two cross bred in-calf'heifers and construction of a stable for the two cows and their calves) will require the cultivation of one ha of vetch-oats. Enough feed will be available for the cross bred cattle. The farmer is either a private borrower or a member of a six-year old Agrarian Reform Cooperative. Both possibilities are tested in the cash flow analysis. Per capita farm income will increase from US$200 to US$285 equivalent, and employment will increase by 310 days per year. 1.09 Model 4: Four In-Calf Heifers and a Stable (Table 12). This is a medium 66 ha rainfed farm. Hard wheat, barley and broad beans are the main crops; sugar beet and berseem are grown on irrigated land. The farmer owns a tractor and rents a combine harvester; he initially owns two cows. The investment (four in-calf Frisian heifers and a stable) permits an increase in herd size. At Year 11 the herd would number 29 cows and heifers, and 18 calves. Male calves will be sold when one-year old, female calves will increase the herd and old cows will be sold after the sixth calving. Artifi- cial insemination will be provided free of charge by the livestock technicians of the Ministry of Agriculture. Twelve hectares will be irrigated;' half under sugar beet, half under berseem, with a second crop of maize forage on 3 ha. Forage produced by the farm covers the needs of the herd until Year 10. Per capita farm income would increase from US$1,550 to $2,285, employment would increase by 800 mandays. The total livestock component of the Fourth Agricul- tural Credit Project is expected to create about 40,000 manyears of employment. 1.10 Model 5: Tractor and Combine Harvestor (Table 13). CNCA loan criteria require a minimum of 60 hectares of land to obtain credit for a tractor and 120 hectares for a combine, thus limiting mechanization credit  - 6 - to farming units of a size for which animal or human cultivation and harvest would be uneconomic. The representative farm has 120 ha, growing grain and pulses. Without the investment, land preparation and transport are done by draft animals, and harvest is done by hand, necessitating a large number of seasonal workers (which are not always available). Land cancot be prepared for planting before the first rain with animal powered implements. Since the period of rainfall is short, some potential production is lost by the time animal powered land preparation is completed. The first investment is a tractor. Tractors can prepare the land before the rain, thus permitting increased use of scarce rain water by the plant. Tractors also permit better quality land preparation, can work for a longer period of time without stopping than can animal drawn implements, and since the land can be prepared at the proper moment, tractors permit fertilizer and high yielding seed to have a more beneficial impact on yields. Research in Morocco indicates an average 0.3 ton/ha increase in cereal yields as a result of early sowing permitted by tractors, compared to yields obtained when sowing is undertaken by animal powered implements. Use of tractors reduces the amount of land which must be used to produce feed for draft animals, and reduces the drudgery of agricul- tural labor (one hectare of land preparation by animal power requires the farmer to walk 60 kilometers). The farm model reflects these observations. A second analysis is undertaken of an investment in a tractor and a combine harvestor. The financial and economic benefit of a combine results from labor savings and from a reduction in crop losses at harvest compared to those incurred when harvest is manual. CNCA's relatively strict credit criteria for combines, together with a heavy (30%) indirect tax on imported combines which increases their price, also serve to insure that combines are acquired only by farms able to use them economically. Farmers will purchase combines only when the expected value of incremental crop harvested and the reduction in labor use permits amortization of the cost of the combine. Net farm per capita income would increase from US$1,900 to US$3,450 (equivalent). Family labor is fully utilized both with and without the investment. However, hired labor requirements would fall from 7,600 days to 2,500 days if a tractor alone is obtained. If both a tractor and a combine harvestor are obtained, labor requirements would fall to 1,400 days. The mechanization to be financed under the Fourth Agricultural Credit Project is expected to cause a 19,000 manyear reduction of employment. The economic benefits of mechanization are therefore offset by social costs. The weighting of the economic benefits and social costs is the issue. II. EMPLOYMENT AND FINANCIAL ANALYSIS Employment 2.01 Based on the above farm model analysis, the Fourth Agricultural Credit Project is expected to directly create the following number of manyears of employment (not including jobs saved by replacing depreciated plant and equipment):  -7- Manyears Employment Manyears of of Family Labor Hired Labor Livestock and Stables 18,000 22,000 Draft Animals 0 0 Wells, Pumps, Irrigation 5,300 11,900 Plantations 1,800 15,500 Mechanization -1,000 -18,000 Land Improvement 54,000 6,000 Total 78,000 37,400 This compares to the total rural labor force estimated at 3.1 million manyears in 1979 of which only 60% is fully employed. About 30,000 people enter the rural labor force each year. The total Project Cost (in 1979 prices) per job directly created would equal about US$5,800. Financial Rates of Return 2.02 CNCA loans to finance these investments and farm model financial rates of return are as follows based on the farm model cash flow analysis in tables 9 to 13. Financial Rate of Model Loan Amount Return (DH) ---- %--- 1 (well, pump) 7,140 37 2A (mules, small 3,675 9 equipment) B (mules, small 3,675 9 equipment) 3A (heifers, stable) 8,805 29 3B (heifers, stable) 8,805 31 4 (heifers, stable) 48,670 16 5 (tractor, combine) 271,770 18 5 (tractor) 82,140 20 5 (combine) 189,630 8 III. ECONOMIC ANALYSIS Shadow Prices 3.01 General. The shadow prices used in the economic analysis are obtained by breaking clown Project costs into:  - 8- (a) foreign exchange costs; (b) basic primary inputs (labor, land, Government revenue, profits, and certain categories of factor payments); and (c) other local (non-tradable) costs. The economic cost of a commodity tradable on world markets is it's c.i.f. or f.o.b. price adjusted for economic costs of transport and handling in Morocco. If non-tradable on world markets, the long run marginal cost of sup- plying the item is first estimated, and then decomposed into foreign exchange costs and primary inputs using Morocco's input-output table. Foreign exchange costs are valued at border prices, and primary inputs such as labor at their opportunity cost. Important causes for differences between market prices and economic prices are Moroccan indirect taxes and tariffs, the overvalued exchange rate, and the excess of wages for unskilled labor over their oppor- tunity cost. Specific "economic" prices are discussed in the following. 1/ Tables 1 and 2 give results expressed as ratios of economic prices to market prices for each commodity and agricultural input. It should be noted that the common unit of account (the "numeraire") is freely disposable foreign exchange in the hands of the Government 2/. 3.02 Milk. The average farmgate price of milk paid by processing plants is DH1.1 per liter. The most economic alternative to domestic supply of milk to processing plants is the reconstitution of non-fat dry milk powder and butter oil imported from the European Community or New Zealand. This import price is therefore used as the "economic" price which, when adjusted for domestic transport and handling can be compared to the farmgate price 3/. A potential problem with this formulation is that European exports of powdered skim milk are heavily subsidized, and this situation is likely to persist through the mid-1980's. However, the European community tends to be a price follower with New Zealand setting the price. The c.i.f. price of New Zealand non-fat dairy milk powder is about US$525 per ton; equaling DH3,350 per ton c.i.f. Casablanca. The c.i.f. price of milk fat delivered to Morocco equals DH6,930 per ton. Reconstituted milk requires 3.5% milk fat, 9.5% non-fat dry 1/ "Economic" price is the term used here when referring to prices used in the economic analysis. "Social" prices are used in the social analysis. 2/ The choice of the numeraire does not affect project analysis because the selection of projects depends on relative prices; the numeraire determines only the absolute price level. 3/ There is some evidence that there is a considerable quality difference between fresh milk and reconstituted non-fat dry milk in favor of the former. To the extent that this is true, the use of the world price of reconstituted non-fat dry milk as the "economic" price of local fresh milk incorrectly lowers the value of domestic milk production in the economic analysis.  -9 - milk powder and 87% water. Adding DH40/liter for the cost of reconstitution, and deducting the average cost of transporting domestic milk to processing plants (DH40 per 1,000 liters) yields an "economic" price at farmgate equal to DH0.59 per liter. This is lower than the domestic price of DH1.1 per liter. The "Caisse de Compensation" now pays a subsidy of DHO.28 per liter to milk processors which is in reality a producer subsidy. The subsidy is regressive since it benefits the largest farmers who are the commercial milk producers. Milk produced under the Project which is consumed on the farm is considered a non-traded good and is valued at the domestic price of DH1.l per liter muLti-- plied by a rural consumption conversion factor of 0.74 which converts market values of rural consumption of non-traded goods into border prices (Table 2). Despite the price subsidy on commercialized milk, detailed farm budget analysis (Models 3 and 4) indicates that the average CNCA client producing milk for the market will have an economically as well as financially viable operation, though the economic return will be less than the financial return (para. 3.15). This results partly from the labor intensive nature of livestock production, combined with a low opportunity cost of labor. 3.03 Meat. Morocco only began to import meat in 1975. High tariffs and quantitative controls constrain imports to a lower level than would be the case without such policies. The "economic" price of domestic meat is the import price since this represents the economic savings caused by local production. Beef imported into Morocco would cost' about US$1.50 per kg (DH6.45 per kg) when converted to carcass weight. The domestic price is comparable. In 1977, the average wholesale price for cattle at Casablanca was DH6.3 per kg live-- weight, corresponding to a farmgate price of DH6.1 per kg. For sheep there are two import prices which are considerably different. Firstly, Morocco imports live sheep; in 1977 importing 69,000 at an average c.i.f. price of DH6.45 per kg liveweight. However, imported frozen lamb carcasses would be considerably cheaper, equaling US$1.69 per kg boneless (DH4.4 when converted to liveweight equivalent) from New Zealand. The Moroccan consumer therefore pays a premium of about DH2.05 per kg for fresh meat over frozen boneless. The lower import price was used to value sheep meat in the economic analysis. 3.04 Cereals. The supply of cereals in Morocco does not satisfy demand, causing an increase in cereal imports. Most domestic production of cereals is consumed by farm producers or sold on local markets at free market prices. A public enterprise, the "Office National Interprofessionnel des Cereales et Legumineuses" (ONICL) imports soft wheat and buys cereal at support prices; reselling to retailers. The alternative to domestic cereal production destined to urban markets is to import, which would also save the cost of transporting wheat from the farm to the city. The economic value of local production is therefore the reduction in imports. The "economic" price for wheat destined to urban markets thus equals the c.i.f. import price -asablanca, minus average cost of transporting cereals produced in Morocco to the city. When adjusted for projected relative world price changes by 1985, the results are "economic" prices (a) for soft wheat 23% cheaper than the farmgate price paid by ONICL, (b) for hard wheat 14% less than the farmgate price paid by ONICL,  - 10 - (c) for barley 2% more than the farmgate price and (d) for maize 1% less than the farmgate price. For cereals consumed on the farm or produced in deficit production areas, the alternative is import plus transport and handling to these areas, resulting in an "economic" price above that for cereals destined to urban areas. Farmgate prices are also considerably highec for producers in deficit areas and for farmers not selling through ONICL. For production not destined to urban markets, "economic" prices are estimated to be 39% less than the rural market producers' price for soft wheat, 21% less for hard wheat, and 10% less for barley. 3.05 Pulses. About 37% of Moroccan production of pulses is exported, primarily to EEC countries. The increasing potential for dry broad beans and chick-peas as a filler in manufactured foods is a possible future market for Moroccan exports. ONICL buys and sells a small quantity of pulses. Most production has been, and would continue to be, traded on local markets. F.o.b. prices were used to value production of pulses at farmgate in "economic" prices. Due to a discriminatory exchange rate and taxes, farm- gate prices average 8% less than the f.o.b. price adjusted for handling and transport. 3.06 Sugar Beet. Morocco is a large importer of sugar. Sugar beet prices are fixed by Government, and since all production goes to processors, the fixed price holds. An increase in the real price of sugar from an aver- age of US$0.179 per kg in 1977, to US$0.290 in 1982 (in 1977 prices) is projected. To this is added transport and handling costs to obtain an "economic" price of DHI,505 per ton of sugar. When converted to sugar beet on the basis of recoverable sugar content and adjusted for transport costs and the value of by-products, the "economic" price is computed at DH135/ton of sugar beet compared to a domestic price of DH116/ton. Sugar beet producers are thus implicitly taxed by Government price policy. 3.07 Fruits and vegetables. Fruit and vegetable production is either consumed on farm by producers, sold directly to wholesalers or retailers, sold to canning or other processing enterprises, or sold to the public office responsible for exporting (Office de Commercialisation et d'Exportation, OCE). For the economic analysis, fruit and vegetables are treated as exportables and hence valued at f.o.b. prices. 3.08 Seed. High yielding certified seed is obtained by farmers from 64 seed distribution outlets throughout Morocco which are overseen by a Public Company (SONACOS). The price of certified wheat seed is fixed by Government at about 18% less than the border price, and additional subsidies (35%) are available for cooperatives, and 25 to 30% to farmers participating in Govern- ment's "operation cereales". Other farmers can obtain cereal seed subsidies ranging from 10% to 30% depending on the seed variety. Higher yielding seed is most heavily subsidized. About 12% of cereal seeds used are high yielding varieties; the rest is local seed. The low use of high yielding seed results primarily from inadequate supply, not from inadequate demand. A reduction in subsidies which artificially stimulate demand and an increase in prices paid to seed producers to expand supply, would be a better policy. Sugar beet seed,  - 11 - which is distributed by sugar factories, is subsidized by about 20% compared to the border price. Local seeds are used for pulses and barley. Forage seeds are subsidized (30% for non-irrigated farms, 60% for irrigated farms). 3.09 Pesticides. Pesticides are imported by several private distributors. Import duties and other indirect taxes result in a farmgate price about 38% above the border price adjusted for transport and handling. 3.10 Farm Machinery Investments and Machine Services. Because of high import duties and indirect taxes on imported agricultural machinery, the border price of tractors adjusted for transport cost to the farmgate is 33% lower than the farmgate price. These taxes are offset for Agrarian Reform Cooperatives and Farmers Associations by Government subsidies of about 25% of the domestic price of tractors, and 20% for combine harvestors. Individual farmers do not receive these subsidies. For other agricultural implements, subsidies of 20 to 30% are available to individual farmers as well as to cooperatives and farmers associations. 3.11 Fertilizers. Fertilizers, except for phosphate, are imported and processed by a public enterprise: Fertima. Phosphate based fertilizers are processed and distributed by a Moroccan firm (Societe Cherifienne des Engrais, SCE). Farmgate prices exceed border prices by 30% for imported fertilizers due to indirect taxes. This is partly offset by a 10% subsidy paid to Fertima and SCE. In addition, a 20% subsidy can be obtained by large farmers owning more than 40 ha of land and willing to enter an agreement with Government in which the farmer agrees to follow a bi-annual rotation. Farmers' groups can obtain subsidies equal to 30% of domestic price. Small and medium farmers receive no subsidy. The "Caisse de Compensation" paid DH50 million in subsidies in 1977 to fertilizer importers and processors, and DH58.5 million to farmers. 3.12 Labor. The opportunity cost of labor is estimated at about 50% of the average wage rate, which is close to the value of subsistence consumption. An alternative calculation, based on the percentage of labor employed per month and on the average daily wage, yields an average opportunity cost of unskilled labor equal to 30% of the daily wage. The higher opportunity cost was used in the economic analysis. 3.13 Livestock. A subsidy equal to 20% of the purchase price of pure- bred cows is available to all farmers. In addition, 20% of the cost of stable construction will be subsidized by Government. 3.14 Non-traded Agricultural Inputs. Non-traded inputs are converted from market to "economic" prices using conversion factors which represent the average ratio of economic to market prices for groups of such goods.  - 12 - In order to calculate the "economic cost" of using non-traded inputs, these inputs must be decomposed into traded and primary components. This is fol- lowed by application of border prices to traded components and opportunity costs to primary components. Such a procedure was undertaken for the trans- portation, t :!tribution, construction, and service sectors using Morocco's Input-Output Tables the traded goods "economic" prices discussed above, and the opportunity costs of various primary components (labor,capital). 1/ The following are the resulting ratios of "economic" prices to market prices for major non-traded goods and services. construction 0.48 repair services 0.51 tractor services 0.54 transport 0.50 non-traded agricultural goods 0.50 Economic Rates of Return 3.15 Application of economic prices (shown in Tables 1 and 2) to the farm models (Tables 9-13) gives the following economic rates of return (with finan- cial rates of return again shown for comparison): Financial Economic Model Rate of Return /1 Rate of Return /2 1 Well and Pump 37 60 2a Draft Animals and Implements (farm) 9 25 2B Draft Animals and Implements (coop) 9 25 3A Cross Bred Cows and Stable (farm) 29 15 3B Cross Bred Cows and Stable (coop) 31 15 4 Pure Bred Cows and Stable 16 25 5 Tractor and Implements 20 60 5 Tractor and Combine 18 25 5 Combine 8 10 /1 To Farmer's equity contribution /2 Rounded. The average economic rate of return on project financed investments, weighted by the probable percentage of each investment category in total Project cost, is 25%, excluding the agro-industry component. Sensitivity analysis is under- taken in para. 5.01. 1/ These estimates were made by Gordon Hughes.  - 1 3 - Conclusions and Relevance to Agricultural Price Policy 3.16 The average rates of return reported above conceal large variations between farms depending on particular agro-economic conditions, farmer manage- ment ability, labor availability, rainfall, etc. Investment in mechanization, irrigation equipment, wells, draft animals and small implements generate a higher rate of return to Morocco's economy than to farmers due to price dis- tortions biased against agriculture. Therefore it can be concluded that effective protection for most agricultural products is low or negative. Declining quantities of wheat and oilseeds marketed through enterprises paying official prices indicate the inadequacy of the prices for these commodities compared to farm production costs. Imports have increased to replace this decline in marketed local production. For smaller livestock producers, milk price subsidies are greater than effective input taxes (paras. 3.08 to 3.13) thus causing financial rates of return on dairy investments to exceed the rate of return to the economy, although the economic rate of return is satisfactory (model 3). Milk price subsidies may therefore be causing over-investment in livestock by small and medium farmers. However, CNCA's strict lending criteria for livestock loans increase the likelihood that such investments undertaken by CNCA clients will be economically viable. For larger milk producers (model 4) milk subsidies are not high enough to offset effective input taxes (chemical treatment, farm machinery, seeds), and therefore the economic rate of return to large farmer milk production is higher than the financial rate of return. Despite the likely viability of CNCA financed agricultural investments, review of price policy by Government would be desirable. The possibility of replacing input subsidies with agricultural credit and increased producer prices for some agricultural commodities should be investigated. However, as Chapter IV of this paper suggests, the degree of price distortion should be measured with respect not only to "economic" prices as is common in Bank work, but to social prices, or prices consistent with Government social/economic objectives. Deviation of a price from its "economic" level may be desirable if this deviation can be shown to be con- sistent with Government social/economic policy objectives. Deviations which are not consistent with Government objectives are true distortions. IV. SOCIAL RATE OF RETURN ANALYSIS Basic Concepts 1/ 4.01 Social analysis expands on the preceding economic analysis by giving weight to the impact of a Project on economic growth and income distribution. If an objective of Government is economic growth more rapid than is permitted by fiscal and monetary policy, then a premium is given in social analysis to Project generated income that leads to investment rather than consumption. This procedure will cause Project analysis to favor projects that generate 1/ The methodology on which this social analysis is based is developed in Squire, Lyn, and van der Tak, Herman; "Economic Analysis of Projects".  - 14 - more investment, hence, helping to overcome the constraint on reaching desired investment and growth targets. The impact of a Project on income distribution is estimated by giving weights to the incremental consumption accruing to each income group benefitting from the Project, including labor. Weights reflect Government's income distribution objectives. When combinirg the two weighting systems, the analysis will favor projects that benefit the poor and that result in higher savings and future growth rather than current consumption. The following formula summarizes these concepts: d_ Net Social Benefit = Net economic benefit - Ba C. (1 - where: net economic benefit = the incremental production arising from the Project at "economic" prices as defined above in Chapter III; ACi = the consumption increment of the jth income group resulting from the Project; A = the factor converting the national consump- tion basket from domestic into "social" prices reflecting the real resource cost of such goods in terms of uncommitted foreign exchange in the hands of the Government (the numeraire); d = weight attached by Government to the cgnsumption of income group j; v = the marginal social value of foreign exchange in the hands of the Government relative to private consump- tion at the average consumption level. v serves to convert consumption increases caused by a Project to its value in terms of the numeraire. In this analysis, all non consumption expenditures are considered to be equivalent. 4.02 The paragraphs which will follow describe the assumptions made in the estimation of the social parameters. It is useful to first scrutinize the calculations made in Tables 3 through 8. Taking Table 3.1 as an example (well drilling and Purchase of Pump, Social Rate of Return Analysis), scrutiny can begin with column 5 (Incremental Farm Income from Project). I/ This represents the incremental cash flow shown as the bottom line of the Cash Flow Projection (,the cash flow projection for model 1 is made in Table 9.3). 1/ Identical tables with identical column headings are shown to each model; see Tables 3.1, 4.1, 5.1, 6.1, 7.1, and 8.1.  - 15 - The incremental cash flow projection was made from the model's investment, cropping pattern, crop yield and farm input analysis (Tables 9.1 and 9.2 for farm model 1), as well as credit receipts and debt service (Table 9.3). The rate of return calculated from the incremental project cash flow is the financial rate of return to the investment represented by the model. The "economic" cash flow shown in column 1 of Table 3.1 is obtained by applying economic prices to crop production, farm inputs, and investments. The eco- nomic cash flow is used to calculate the economic rate of return. The appli- cation of the social rate of return formula is made by first estimating incremental consumption (AC). Incremental consumption, shown in column 6 of Table 3.1, equals incremental cash flow minus incremental savings out of cash flow. Savings rates are estimated for various farm sizes in para. 4.11. The social cost (or benefit) of this inc remental consumption is then estimated by applying to it the parameter (-p - -). The parameter -A defined is estimated as -.7 (para. 4.09). The parameter (v) is estimated as 3.65 (paras. 4.07 and 4.08). The distribution weights (d) are estimated for each farm investment beneficiary family in paras. 4.04 to 4.06. The resulting value of (-p + -) for each farm model is shown in column 8 (of Table 3.1 for example). It is a ratio representing the percentage of incremental farm consumption which constitutes a social benefit (if positive) or cost (if negative). This ratio is multiplied by incremental consumption to obtain the social benefit or cost of consumption. It should be noted that private savings are not costed, and are therefore assumed to have the same social value as foreign exchange revenue accruing to Government, or as consumption by people living at the critical consumption level. The net social benefit stream in column 4 of Table 3.1 is calculated as the net economic benefit stream minus the social cost of incremental farm family consumption (in some cases this consumption is added as a social benefit) and a third adjustment reflecting the difference between the social and economic prices of farm outputs and inputs. These price differences, shown in footnote b of Table 3.1 and in Tables 1 and 2, are explained in paras. 4.12 to 4.14 below. They reflect primarily the difference between the social and economic cost of labor use. The social cost of labor adjusts for benefits and costs of incremental consumption by laborers as a result of the project and distribution weights attached to that consumption. Since labor is involved in the production of non-traded inputs, there will be a difference between the social and economic costs of using non-traded inputs. Column 2 of Table 3.1 shows the difference between the economic and social benefits of farm outputs, inputs, and investment. The social rate of return is calculated as the rate of return to the resulting net social benefit stream shown in column 4 of Table 3.1 (for model 1). 4.03 The following paragraphs describe the estimates of the social para- meters, discussing firstly the distribution weights (d), the value of public income and investment relative to private consumption (v), and the consumption conversion factor (p). When combined, these three parameters give a ratio representing the net social cost (benefit) of incremental household consump- tion (-, + -). Incremental consumption, to which this ratio is applied, is then estimated from incremental farm cash flow through an analysis of savings behavior. The difference between social and economic pricing of labor and commodities is then discussed to arrive at the net social benefit stream.  - 16 - Distribution Weights 4.04 Distribution weights are estimated from the following: dj = c n d C =(-) C. J where d. = distribution weight attached to the consumption of income group j; c = average per capita consumption in Morocco c = household consumption per capita at income level j; n = the elasticity of the income weight with respect to income. A value for n equal to 1 is assumed here indicating that a percentage change in income is associated by Government with an identical percentage change in the income weight. This assumption is subjective, though based on Govern- ment's stated objective to improve income distribution. Actions taken by Government to reach the objective have been limited. The tax system is only mildly progressive, and wage and price policy have no positive distributional impact. If the fiscal system in Morocco were able to distribute income more efficiently and in a manner consistent with Government's distributional objectives, there would be no need to use distribution weights in project analysis. In Morocco, the fiscal system is unable to do this, and therefore there is a strong case for using projects to assist Government in achieving distributional goals. Morocco's 1971 household budget survey is the source of data for estimation of distribution weights. Weights are derived from average expenditure in each occupational group.  - 17 - Table 1: DERIVATION OF DISTRIBUTION WEIGHTS BY OCCUPATION Average Household Distribution Expenditure per Capita Weights (d % of (DH per Capita, Occupation Households 1971 Prices) n = 1 n = 2 White Collar 5.99 2,240 .40 .16 Commercial 9.21 1,040 .87 .76 Blue Collar Workers in Industry and Services 22.35 1,040 .87 .76 Artisans 7.60 710 1.27 1.61 Farmers 35.20 705 1.28 1.64 Farm Workers 11.27 565 1.60 2.56 Other Employed 3.38 670 1.35 1.82 Inactive 5.00 1,055 .86 .74 Average 905 1.00 1.00 Source: Morocco, 1970/71 Household Budget Survey, Vol. 1. Table 2: DISTRIBUITION WEIGHTS FOR FARM SECTOR Per Capita Consump- Distribution Weights (d)/ Size of Farm % of Farm tion as % of Average (ha) Families Farm Consumption n = 1 n 2 0 23.4 35 3.66 13.40 0 - 5 56.6 55 2.32 5.38 Average - 100 1.28 1.64 5 - 10 11.4 170 .75 .56 10 - 20 5.9 310 .42 .18 > 20 2.7 555 .23 .05 Total 100.0 /1 Computed on the basis of an estimate of average expenditure per capita equal to that for the average farm family shown in Table 1 (DH 705 per capita) multiplied by the percentage of this average appropriate to each farm class. Hence, for 0 - 5 ha farms, average expenditure per capita is estimated in 1.970/71 prices as .55 x DH 705 = DH 390, which yields a distribution weight of 2.32 (DH 905/DH 390). Source: MARA - Division des Affaires Economiques, "Etudes de la Capacite de Financement du Developpement par l'Agriculteur", Novembre 1977.  - 18 - 4.05 For the Fourth Agricultural Credit Project, each farm model repre- sents average agro-economic conditions facing a wide variety of income groups. Therefore, for each farm model, a distribution weight was used which repre- sents the average for the group of farmers who would undertake the investment. The following data indicating the distribution of CNCA cretlJ t by farm size is relevant. Table 3 % of CMCA % of CNCA Distribution % of Moroccan Expected Disbursements /a Weight from Farm Size Farm Families Clientele (1979/80-1981/82) Table 2 (ha) (1982) (Col. 1) (Col. 2) (Col. 3) (Col. 4) (Col. 5) 0 23.4 0 0 3.66 0 - 5 56.5 40.0 13.0 2.32 5 - 10 11.4 48.5 50.0 0.75 > 10 8.6 11.5 37.0 0.21 Total (weighted- average) 100.0 100.0 100.0 (0.75) /b /a Excluding agro-industry. lb Weighted average distribution weight for CNCA clients, with weights equal to column 4. The average distribution weight for beneficiaries of CNCA loans, calculated as column 4 multiplied by column 5, equals 0.75. This is low because large farmers borrow considerably more from CNCA than do small farmers. This average distribution weight is used for the models representing investment in pumps, irrigation equipment, wells, crossbred livestock, and stable construc- tion (models 1 and 3). For the models representing mechanization and intro- duction of highly efficient livestock operations using imported pure bred livestock (farm models 4 and 5), a distribution weight representing the wealthiest 9% of Moroccan farm families is used (equal to 0.21). The distri- bution weight for farm laborers displaced by mechanization is 2.54. For investments in draft animals and small implements (model 2), the distribution weight applicable to the farm size category representing CNCA's smallest clients is used (d = 2.54). 4.06 As a Project beneficiary's relative consumption level increases due to the Project, the distribution weight attached to that consumption will decline. This decline is estimated to equal the percentage rate of increase in consumption by the investor relative to the average growth of Moroccan per capita consumption. Moroccan per capita consumption is predicted to grow at 2% p.a. Growth of per capita consumption is predicted at 5.1% p.a. for model 1, or by 3.1% p.a. relative to average Moroccan consumption. The applicable distribution weight thus declines at 3.1% p.a. Similar calculations were made for each farm model (Tables 3 to 8).  - 19 - Value of Public Income and Investment Relative to Consumption (v) 4.07 The parameter v is normally defined as the value of public income (measured in foreign exchange) in terms of the marginal value of consumption for the consumer at the average income level. The formula for deriving v is as follows: q - S q ~CRI - sq where q = rate of return in economic prices to the marginal public investment (estimated to equal 6% in para 4.15); s = savings rate out of this return (estimated at 18%); which is the aggregate marginal real savings rate out of domestic income between the two periods 1967/1970 to 1974/77 (Bank Economic Report, Table 2.4); CRI = consumption rate of interest (estimated at 3%) 1/; = average consumption conversion factor for the rural and urban sectors (estimated at 0.7). The formula results in a high value for v, equal to 3.65, indicating the extremely high premium attached by Government to uncommitted public revenue compared to consumption due to the severe shortage of public revenue, and suggesting a relatively high return on public expenditure. The plausibility of this estimate is checked in para 4.08 by relating it to the estimate of the critical consumption level. Critical Consumption Level 4.08 A first approximation of the critical consumption level is that which corresponds to an income at which the tax system allows deductions. Few households with incomes below DH 9,000 (US$2,250) pay tax in Morocco. This is too high a level of critical consumption however, since 60% of households have an expenditure level below this. The critical consumption level was therefore estimated to be identical to the absolute rural poverty level as measured by the Bank (of US$200 per capita in 1978 prices). About 45% of Moroccan rural families live at or below this level. Table 2 above provides the information 1/ The consumption rate of interest (CRI) is the rate at which the value of a unit of consumption declines over time. Its estimating formula is: CRI = ng - p where n = the elasticity of marginal utility with respect to consumption, assumed to equal 1; g = the growth rate of per capita consumption (estimated at 1% per annum from 1971 to 1977); p = the rate of pure time preference (assumed to equal 2%).  - 20 - needed to estimate the distribution weight for farmers living at this cri- tical consumption level. The weight equals 2.54 1/. The social value of an extra dirham going to consumption by people at the critical consumption level is equal to the social value of an incremental dirham going to public revenue (-B + d./v = 0 when d. equals the distribution weight at the critical consumption levJl). Since P l 0.7, and (d.) equals 2.54 for people living at the critical consumption level, a value fo; v of 3.63 is implied 2/. This is remarkably close to the estimate of v, equal to 3.65, obtained in para. 4.07. 4.09 The consumption conversion factor (B). The consumption conversion factor represents the ratio of the value of marginal consumption in social prices relative to its value at market prices. Based on expenditure patterns derived from Morocco's last household budget survey and relevant social prices, the consumption conversion factor equals 0.74 for rural households, 0.7 for the average Moroccan household. The Net Social Cost/Benefit of Incremental Household Consumption 4.10 The net social benefit (cost if negative) of incremental household consumption is estimated from the formulation introduced in para 4.01: d. b C. (- p + 1) I v Parameters are as defined in para. 4.01, and estimated above. )f equals .7, v equals 3.65, and the values for d. are as indicated for each farm model in para 4.05. These parameters result in the following initial consumption adjustment factors (- f + d,/v) for each farm model: J~~~~~~~~~~~~~ Model Initial d1 Initial (- A + d /v) 1 1 1 .75 -.5 2 2.54 0 3 .75 -.5 4 .21 - .64 5 .21 - .64 This indicates, for model 1 for example, that 50% of the increase in farm consumption is treated as a social cost. These values change as per capita consumption generated by a Project investment increases relative to the average increase in Moroccan consumption per capita (para. 4.06). The consumption adjustment factor is shown for each farm model in column 8 of Tables 3.1, 4.1, 5.1, 6.1, 7.1 and 8.1. It is multiplied by the value of incremental consump- tion for each model to determine the social cost of that consumption. 1/ Computed by intrapolation from Table 2 (51.7% - 45%)/(51.7%-11.7%) x [(3.66 - 2.32) + 2.32] = dist. weight at the 45% consumption level. 2/ - l=;-7 - =0v32.54 2/ - + s= 0; - .7 + = 0; v = 3.63. v v  - 21 - Ihe Estimate of Incremental Consumption Out of Income from Analysis of Savings Behavior L.11. In order to estimate incremental consumption, incremental savings is subtracted from the incremental cash flow estimated for each farm model (in ''ables 9.3, 10.3, 11.3, 12.3 and 13.4). There is no reliable data indicating savings behavior in the farm sector. Estimates were made by the Ministry of iLgriculture on the basis of the 1971 household budget survey and the 1973/1974 census of agriculture as follows: 1/ Farm Size Savings/Disposable Income 0 - 5 ha 0 5 - 10 16 10 - 20 42 20 - 50 61 > 50 41 Average 22 rhe average rural savings rate of 22% is slightly higher than that for the economy as a whole (18%). The data show that the smallest farmers, living close to subsistence have virtually no savings. The savings rate increases with farm size up to the 20 - 50 ha farm category, after which it declines. 2/ 1/ Morocco, Ministry of Agriculture, "Evaluation de la Capacite de Finance- ment du Developpement par l'Agriculteur", Nov. 1977; page 7. 2/ Knowledge of savings rates, in combination with other information given above, permitsthe calculation of the social cost of incremental consumption caused by an investment. Because of the interest of the farm mechani- zation model, it is used as an example (Table 7.1). The distribu- tion weight (d) for mechanization beneficiaries of .21 declines at 4.1% p.a. after the mechanization investment is made. This results from the increase in beneficiary consumption more rapid (by 4.1% p.a.) than the projected Moroccan average annual increase in per capita consumption (para. 4.06). This declining distribution weight is shown in column 7 of Table 7.1. Incremental consumption is then estimated in column 6 by subtracting savings from incremental income. Forty percent of incremental income is saved for this income group. The social cost of incremental consumption is the value of that increase multiplied by the consumption conversion factor (social cost = 6cB). The benefit (Ac.d./v) is obtained by multiplying the increased consumption (Ac1), by the dlstri- bution weight for beneficiaries of combines (.21), (declining at 4.1% p.a.), divided by the marginal value of public income to private consump- tion (v = 3.65). The result in the initial year is a social benefit of 5.7% of incremental consumption, minus a social cost of 70% yielding a net social cost of 64.3% of incremental consumption.  - 22 - Social Cost of Input Use 4.12 In addition to the social cost (benefit) of incremental consumption by farm family beneficiaries, there are additional social costs and benefits of increased input use which are not captured in the economic analysis. The differences between economic and social prices of commodity use, and farm output, are shown in Tables 1 and 2. The differences are due almost entirely to the added social cost (benefit) of incremental consumption out of added wage payments. 4.13 The social cost of using labor includes: (a) the cost of labor estimated for the economic analysis; equal to the opportunity cost at "economic" prices of the marginal production sacrified elsewhere in the economy by increasing employment in the Project (para. 3.12); and (b) the net social cost (benefit) of incremental consumption caused by added wage payments. These costs are represented in the following formula: d. j j v) where m = opportunity cost, measured in border prices, of using labor in income group j as estimated in para 3.12. All other parameters are as defined above. Ac; = change in consumption by labor resulting from the Project. d = distribution weights attached to the consumption of Project labor. 1/ Since rural labor lives approximately at the critical consumption leval (although many live at an even lower level), the expression ac (B - equals zero. The social cost of using rural labor is thus equal to the economic cost, which is valued at about 50% of its wage. The social cost of semi-skilled labor is higher than the opportunity cost because of the low distribution weight attached to additional consumption by this group (see 1/ In terms of the formula in para. 4.01, m* is included as an economic d. cost in the net economic benefit stream, and - Ac ( - ) for incre- mental labor is included along with the social coat or benefit of farm family incremental consumption.  - 23 - Table 2). In the case of mechanization, less labor is used as a result of investment. There is a benefit of this replacement of labor by farm machinery equal to the alternative production expected from the freed labor plus the expected value of reduced consumption by labor. The economic benefit of labor savings is estimated at 50% of its wage. The social value of reduced consump- tion caused by labor deplacing mechanization is estimated at zero since laborers live at about the rural absolute poverty level (or critical consump- tion level). The net result is that labor displacement in isolation has a benefit equal to the likely production of that displaced labor in its alterna- tive employment. This productivity is estimated to equal about 50% of its agricultural wage. The margin of error in this estimate is very large. Social Cost of Commodity Use 4.14 The "social" prices of traded commodities equal their border prices as described for "economic" prices, plus domestic handling revalued in social prices. As in the economic analysis, domestic handling is a non-traded good, and is therefore decomposed into its component costs. These consist of labor which is costed in social prices, traded goods costed at border prices, and other non-traded goods and services which are further decomposed into labor and traded inputs using the input-output table. Other non-traded goods are also treated in this way. Table 2 shows that social prices are generally higher than "economic" prices. This is caused by the fact that the economic cost (opportunity cost) of using semi-skilled labor is lower than the social cost due to the low distribution weight attached to incremental consumption by this group. Non-traded goods and services have a high input of semi-skilled labor in Morocco. Relevant ratios of economic and social prices to market prices for non-traded goods are as follows: Economic price Social price to market price to market price Repair services 0.51 0.58 Tractor services 0.54 0.60 Transport 0.50 0.57 Construction 0.48 0.54 Non-traded agricultural goods 0.50 0.80 Animal drawn implements 0.50 0.55 Tables 1 and 2 show relevant ratios of economic and social prices to market prices for traded goods. The Accounting Rate of Interest (ARI) 4.15 The accounting rate of interest against which the social rate of return should be compared is estimated at 6%. This compares to an opportunity cost of capital in Morocco, against which the economic rate of return is compared, equal to 8%. The ARI should equal the real rate of return on marginal projects in the public sector evaluated using social prices and accounting for the marginal projects' impact on growth and income distribution.  - 24 - Alternatively, if the Government is borrowing heavily from abroad, as is Morocco, and if such borrowing is based on a social evaluation of benefits and costs of borrowing, the ARI should equal the marginal real cost of borrowing. Estimated rates of return on Bank Projects in Morocco over the past 4 years have varied setween 10 and 30%. For industrial projects it has probably been lower (about 6-8%). The social rate of return to the marginal project is probably about 6%. Morocco pays an interest rate of about 11-12% for Euro- dollar loans. With the rate of inflation at 7-8%, the real rate of interest is also about 5%. These estimates suggest the use of 6% as the marginal social rate of return against which social rates of return on investments in Morocco should be compared. Social Rate of Return Analysis 4.16 Applying to the farm models the social prices and weights derived above and shown in Tables 1 and 2, results in the net social benefits streams shown in Tables 3 to 8. The following social rates of return result, with the economic and financial rates of return again shown for comparison: Economic Rate Social Rate Financial Rate Model (Investment and Income Level) of Return/ of Return- of Return/ Well and pump, (average farm income level) 60 50 37 Draft animals and implements; (for farm at average income level) 25 14 9 Cross bred cows and stable; (Average farm income level) 15 5 30 Pure bred cows and stable (for farm at upper 25% income level) 25 20 16 Tractor and Implements (upper 25% income level) 60 45 20 Tractor and Combine; (upper 25% income level) 25 20 18 Combine (upper 25% income level) 10 2 8 /1 Rounded. The social rate of return to the' Project, equal to the weighted average return on the several components is 15%. This is lower than the 25% economic rate of return. The social rate of return should be compared to the 6% accounting rate of interest. The social rate of return is lower than the economic rate of return for each farm model because of the extremely high premium put on public income compared to private consumption, and the fact that most of the  - 25 - benefits of the Project go to medium and large farmers having low distribution weights. Despite this, the overall Project and all of its components except combine harvestors have acceptable social and economic rates of return. 4.17 Combine harvestors, although economically viable, often displace considerable labor and are obtained by the wealthiest farmers. The social impact of such investmenc when it displaces labor, given the parameters described above, is undesirable. This result is modified if investors in combine harvestors save more of the incremental income derived from the investment. 1/ If for example, beneficiaries of investments in combines save 100% of incremental income, the social rate of return would approximate the economic rate of return; 10%. In addition, in some regions of Morocco, labor has become scarce during harvest periods and use of combines is the only feasible method for harvesting. If a combine is purchased because adequate labor is not available at harvest, then the social rate of return approaches the economic rate of return. Furthermore, if without the combine output would decline because of increasing shortages of seasonal labor at harvest time, then both the economic and social rates of return would tend to be higher than shown here. This analysis, along with the sensitivity tests reported in the following, indicates only that investments in combine harvestors will, in some cases, be undesirable. A more detailed analysis of labor markets, savings behavior, and opportunity costs of labor is required to obtain more precise answers to the question of combine viability. The results do suggest that existing import duties and other indirect taxes on combines should be main- tained in order to discourage over-mechanization. Subsidies presently avail- able to cooperatives and farmers groups for combine investments should be re-examined. CNCA lending criteria for combines should be tightened in labor surplus regions. Availability of harvest labor should become one criterion evaluated by CNCA in providing medium term loans for combine harvestors. It is very significant that this policy conclusion is exactly opposite to that resulting from the economic analysis alone in para. 3.15. For combine har- vestors, the financial rate of return (8%) is lower than the economic rate of return (10%) indicating that price policy discriminates against combines, inducing a level of investment which is lower than economically desirable. However, the social analysis gives a return of only 2%. Since the financial return is higher than this social rate of return, it can be concluded that price policy should discriminate even more against combines, not less as is suggested by the economic analysis. 4.18 Livestock investment. The inferences derived from the social rate of return analysis of livestock investment are different from those derived from the economic rate of return analysis. The large livestock investor is considerably more productive than the average (or small) livestock investor (ERR 25% for the former,, 15% for the latter). However, high milk subsidies paid to producers, and low taxes on the kinds of inputs used in traditional livestock production, causes the financial rate of return to small and medium livestock investors (30%) to be much higher than the economic rate of return. For large efficient livestock investors, the financial rate of return (16%) is lower than the economic rate of return due to high effective taxes on the 1/ It was assumed in the analysis that 40% of incremental income is saved.  - 26 - modern inputs used in such production, which more than offset the benefits of milk subsidies. The conclusion of economic analysis is that livestock invest- ments financed by CNCA are generally economically viable. However, milk producer price subsidies are too high, as are effective taxes on the inputs used in modern livestock production. This causes over-inve-stment by small and medium farmers, and under-investment by large efficient farmers. The economic policy conclusion is that milk price subsidies and effective taxes on modern inputs should be reduced. The social rate of return to livestock investments by the average CNCA client is 5%, while it is 20% to the large efficient farmer. For the medium farmer, this low social rate of return is caused by the relatively low distribution weight (0.75), a low savings rate (20% of incremental income), the milk subsidy, and lack of economic efficiency. As in the case of the economic analysis, the policy conclusion is that there is an incentive to over-invest in livestock by medium farmers, and that milk price subsidies should therefore be reduced. For smaller investors, the.distribu- tion weight increases, and at the same level of efficiency the social rate of return on their livestock investments increases. From the perspective of the social rate of return analysis, credit should be allocated to the small livestock investors in priority. For the large efficient livestock investor, the social rate of return is above the financial rate of return, but below the economic rate of return. 1/ The policy conclusion remains that effective taxes on modern inputs should be reduced, and that credit should be allocated to the large efficient investors in priority. With more credit going to the smallest investors, and to the large efficient investors, less would be allocated to the medium investors. 4.19 Other farm investments have remarkably high social returns according to these models. This does not mean that such investments will be productive on all or most farms. Only about 20% of Moroccan farmers have borrowed money from CNCA and the models represent only CNCA clients. Farmers who borrow medium term from CNCA have productive investment possibilities, and are con- fident that credit financed investment will permit debt service on the loan plus a reasonable return. This results from the stiff penalty imposed on loan defaults, including loss of collateral, often meaning loss of land. The aver- age farmer will not risk such a loss to finance investment which is marginal. To the extent that economic and social prices approach market prices, and to the extent that investments are undertaken equally by various income groups, such a system will assure an acceptable average social and economic rate of return on credit financed investments. The major price distortions face the livestock subsector, and the major inequity in investment distribution' occurs with mechanization, hence explaining the lower average social rates of teturn for these investments. V. SENSITIVITY TESTING 5.01 Tables 3.2 to 8.2 show the results of sensitivity tests under alter- native assumptions about the elasticity of the income weight with respect to 1/ The distribution weight for the large investor is 0.21, but the large livestock investor saves 60% of his incremental income compared to 20% by the smaller farmer.  - 27 - income (n), and the value of foreign exchange in the hands of the Government relative to private consumption (v). Five assumptions are tested, from the most extreme of n=0, v=1.43 indicating no distribution objective; to n=2, v=3.65 indicating a pronounced distribution objective combined with the expected value of public income relative to private consumption. The n=0, v=1.43 case is identical to the economic rate of return. Social Rates of Return - % n=O,v=1.43 (equals economic rate of return) n=O,v=3.65 n=l,v=3.65 n=2,v=1.43 n=2,v=3.65 Well & pump 62 59 50 50 50 Draft animals & implements 24 21 14 24 20 Cross bred cows & stables 14 6 5 9 3 Pure bred cows & stable 28 27 24 24 24 Tractor & implements 59 53 48 48 48 Tractor & combine 28 24 21 20 20 Combine 10 6 2 2 2 Weighted average 25 19 16 19 16 /1 These rates of return have not been rounded off, as they are in paras. 3.15 and 4.16 to facilitate comparison. The margin of error is high and therefore these rates should be rounded to the nearest 5%. Although the overall Project social rate of return is not sensitive to alter- native possible social parameters, conclusions regarding livestock and combine harvestor investments are sensitive. The less the weight given to distribu- tional objectives (n approaches 0), the more acceptable are these investments on average. The greater the weight given to distributional objectives, the less acceptable are these components. For livestock, the greater the weight given to public revenue versus private consumption, the less acceptable becomes such investment. This result emphasizes the importance in this analysis of accurate quantification of Government objectives if social analysis is to provide results useful for policy making. The results pre- sented in this working paper are clear enough to discuss with Government, and suggest the direction in which agricultural credit lending criteria might be modified. Development of additional farm models representing the different agro-economic circumstances under which livestock and combine investments are made, would provide information which could be used to narrow policy choices. However, this narrowing will require increasingly accurate social parameters.  - 28 - TABLE 1: ACCOUNTING RATIOS FOR ODNVERTING MARKET PRICES TO BORDER PRICES; OUTPUT Ratio of Economic Price to Social Price to Market Price Market Price Soft Wheat (Large Farmers) 0.77 0.77 Soft Wheat (Small Farmers) 0.71 0.72 Hard Wheat (Surplus Area) 0.86 0.86 Hard Wheat (Deficit Area) 0.79 0.79 Barley (Surplus Area) 1.02 1.02 Barley (Deficit Area) 0.90 0.91 Maize 0.99 0.99 Milk Sold for Processing 0.54 0.54 Milk Consumed on Farm 0.80 0.84 Sugar Beet 1.51 1.48 Cattle, for Slaughter 0.94 0.93 Urban Beef, Retail 0.81 0.82 Imported Breeding Cows 1.16 1.16 Sheep (Live Imports) 0.84 0.84 Sheep (Meat Imports) 0.55 0.55 Pulses 1.08 1.07 Exportable Agricultural Produce 1.07 1.08  -29 - TABLE 2: AiCCOUNTING RATIOS FOR CONVERTING MARKET PRICES TO BORDER PRICES; INPUTS Ratio of Economic Price to Social Price to Market Price Market Price Tractors 0.61 0.62 Miscellaneous Agricultural Machinery 0.67 0.68 Animal Drawn Implements 0.50 0.55 Vehicles 0.65 0.57 Pesticides 0.62 0.64 Urea 1.06 1.08 Phosphatic Fertilizers 1.17 1.18 Construction 0.48 0.54 Repair Services 0.51 0.58 Tractor Services 0.54 0.60 Improved Seeds 1.18 1.19 Transport 0.50 0.57 Taxes 0.00 0.00 Skilled Labor 0.65 0.65 Semi-skilled Labor 0.17 0.51 Unskilled Rural Labor 0.51 0.51 Profits 0.20 0.20 Non-traded Agricultural Goods 0.50 0.80 Rural Rents 0.55 0.60 Extra Income for the Average Rural Household /1 0.00 0.22 Standard Conversion Factor for Traded Goods 0.60 0.60 Non-traded Conversion Factor 0.47 0.53 Rural Consumption 0.71 0.74 /1 Extra consumption for the average rural household was estimated to have a net social cost equal to 25% of its value (para. 36). Assuming that about 15% of extra household income is saved, and rural savings have no resource cost (i.e. are as valuable as uncommitted public income), then the net social cost of extra income at the average rural income level is 22% of that income.  Mki IC A I. ilRAl OR A CIL311.7RAl I RI-I r 75`1 1fE Mliodel I - Will iIi,Ilinr and FDr.li.is.oiisvpd Social alteL ot Kretr At lolys ! i ces, of Net leaL o l Consumpiia lei,etil of Coui_m 3 Social Over W I w Net Economic Social Net i-rernial Intruriial Appi I'd 1o Net En-efic of -eneftr Of Social Fa I cnct ions Distrlbution - uemuial Ec-n-me Output And lvcineal tlme fit i' Cr,nu usuoiiu.i/c W-,ght/t Cus..ic. -p. Sear Benefits/a Inputs/h ConsuCmption/c Streanlg i'rojCtld (Ca) (d) (d3, )/ (I) k27) (l) (4) (5) G() (7) I 1) -4.495 - 30 0 -4,525 -3,390 0 75 - 411 2 2.100 -220 0 14880 h4n 0 73 - 3 2,755 -215 0 2.540 810 0 71 - Si 4 3,400 -215 0 3,185 I,o85 0 68 - *I 5 3 720 -215 -405 3,100 l,256 7hO 66 - 6 l3720 -215 -l.110 2,395 2,685 2,095 i!. - 7 3,720 -215 -905 2,600 2,I85 1,705 62 - 5S 8 3,720 -215 -920 2 585 2,185 I,70S 611 - 9 3,720 -215 -1,185 2,3210 2,B5 2,19S i1 - Y. 10 3,720 -215 -1,130 2,J75 2,6dS 2,095 57 /a Fron Table I /b CoMncted by applying the Ratios of Social Prices to Market Prices shown In Tables I and 2, to LIi: Well DrillinIg and Pomp MHdel presented I Tables - Relevant conversion factors are as follors (Note that conversion l s.cial prices yieldu hIighr re-s-re- c-sts lac coneers-on to economic prices). OUTPfIT l14PIJS Ra.I Otio f Rui_ _ EconomLi Price ru Soc"tal Price to ccorooic price C,, Social Price L Market Price Market Price lLIk,e,l rice MaCrket PI ic Dunn Wheat 0.86 0.86 Seed 050 0 80 Barley 1. 02 1.02 Fertilizers, MaJne 0 99 0 99 Pesticides 1.17 1 i8 Potatoes 1.07 1.08 Land Preparation 0.4 V. 0 Milk 0 80 0.84 HIred labor 0 51 0.51 Meat (Calves) 0 94 0.93 Ani,Lal Feed I 02 1 02 Sheep, Marketed 0G55 0 55 Irrigaton Loo.- 1 00 I 00 Sheep, (In farm Contingencies 0 60 0 60 Consum.ptIon 0.54 0 84 Taxes 0 00 0 00 Well Drilling 0.60 n 6i Pump 0 i67 0 5.h it Derived by altiplyiiig cultmn 6 by -ulu,Ii 8 /d Incremental In-coe is caklo from the crah flow table, lnrre,sental Net Ivenue, page /e Incremzental fan, conaisptiun is estimated by subtracting from increm.ental nvcise, the e.peCetd incremental -avi-k,s Sanings rates were estimated for various farm sines In paragraph 4 09. sice ihis modi I Is r-preveittin, loct la,s of all sizes In Morocco, the avera.ge srag f an-gs rate ni 22t of i-ciermnial i--.ase is I oed i e I,,..oc Ial consumption equals 78% of incresental icmve. lncremeiiMal -uns nption of hired labor is treoLtd ii the -aul-v wage Consuiition i assumed not to Increase until cianolatioc -sh fluo ren IC 1 o ---l i.iC 10,111 Is .tal r - than nero (L e. onCLI Incre-ental Il-eoe In the years liilloul g that is wirl, lici ., --....I. was nadi . tid t,c los of Income In year 1. /f TlILs weight vas stlmated Is paragraph 4 07, tiere d * 0 75. v J 65. P= 0 7 "d is cRs iiaied i. deIls it 3 L% p a. Which Ls the rate of Increase ol relative farm intcoe (S I4 - 2 .0 p a I /intal ot rotuns 1, 2, and 3 A,-.usl 1979  119115$ 135 FOIR11 I ALKIC lC.11Ai CR1 1)11ills 141 iodt I I - Ws 11 On I , lrtisj s Duv Iif S cial Rate of Rlt- S IS. ,s- tyivi A Assl y j.ss Year =O, v-1.43 eO, v-3.65 .-I, v3.65 ->2, v-I 43 ,i-2, -3.65 d l Co,s-ptitlon Adl-st-Ct Parameter (-8i+v) I 0 -.43 -.50 -. 11 -.S5 0 -.43 - So -.33 -.S 3 0 -.43 -.51 -.35 - 56 4 0 -.43 - Sl - 18 -.S7 5 0 -.43 -.5! -.40 -.58 6 0 -.43 -.53 -.41 -.59 7 0 -.43 -.53 -.41 -.59 8 0 -.43 -.54 -.45 -.60 9 0 -.043 -.54 -.40 -.61 10 0 -.43 -.54 -.47 -G.I1 d b/ Conon-ptioo Ad luat-nnt -AC(Jk-v) I 0 (I0 0 0 2 0 0 0 0 0 3 0 0 0 0 0 4 0 0 0 0 0 5 0 -335 -405 -310 -450 6 0 -900 -1,110 -660 -1,235 7 0 -735 -905 -735 -1.005 8 0 -735 -902 -765 -1,025 9 0 -945 -1,185 -1,010 -1,340 0 0 -900 -1.130 -985 -1,280 d c/ Wet Social HIvefit Stream (Net Eco..,it Ilsefit -ac(S-;) I -4,495 -4,495 -4.525 -4,525 -4,525 2 2,100 2,100 1,880 1,880 1,880 1 2,755 2.7SS 2,540 2,540 2,540 4 3,400 3,400 3,185 3,185 3,185 5 3,720 3,385 3,100 3,195 3,055 6 3,720 2,820 2,395 2,645 2,270 7 3,720 2.985 2,.slO 2,770 2,500 8 3,720 2.775 2,320 2,495 1,665 9 3,720 2,820 2,J75 2,520 2,225 10 3,720 2,820 2,375 2,520 2,225 Rat. ot Reto,rn 62 59 50 SI) Sit .1 When n=2, d iv ercnstd by squa-.n. th, val-e .f Id sA-so s-1 (i - s LsIarii ' tic cc -sit shtiAc Iii cslooo 7 In the precious cable). h/ Apl-is the culssm-ptioo adI-sir-st paral-e-r to lis ,-srssu,st- cvlsisri iosi 'is tts sreced-lrg tilo, colsase 6. Tiu co-osimptisl.s Edjua.sIL st tIlt E n-I, v- (5S ci .1is I 1ic.Ii 1. 1 1 I . f tilt sreo-diog table. It hbsldby cured ti. , s It tht c rc adjist-sLt o.nl.Iy s -. I sr scsrsisv ,st.il. .ir..i.i niy th, faror, net the far-t 's ,l red taboE rhL cos.t of rr issir-- - 11 I i,lict Isis by hired labor so. been rocisided In tl,s soial act 41 labor lis 11 . EIl. dil 1iis. l's ttttc . stuossmic .t- h,- -,cIii tols of labor use winch is resposscthl for tLI. di.IfersLsi1 i1stlss II.' ..i. 1.I s,s.1s IIsn- ac-os.sill tr lt-s Siven in foustote (b) Xn the I-rEctdiu4, hbl, *ad I- i.sl adi,dst--enr s,is is s.,s v.ls i..v. ii. ssslss-o 2 u1 the preced1og table. c/ Ths so l isecI irsir l tic is sllcncs is,. lii is, 1s,s ill, issI.s t * .ssijI sell rtsiid..III . . ...os c-vi of icisists siss ifstssss.s1,i I . .-isi .s 1 f.lI I sIt,.,,assr August 1279  MOROCCO FOURlTH ACRICULTURAI. CREDIT PROJECT Model 2 - Draft Animals and Agrictiltural Implements Social Rate of Analysis (0Dirhem) Excess of DerivatioI Of ConsumpLton Benefit/Cost of Colsi 3 Net Social Economic Social Net lnrre.ental Coi-sup t ion Net Benefit of Benefit Of Social Incremental Fam Distribution Adjus rmit.nt Economic Output Anid Incremental Benefit Farm Consumption/b Weight/c Weigght Year Benefits/a Input/a Consumption Stream Income/b ( C) (d) (d/v-B)/d (1) (2) (3) (4) (5) (6) (7) (8) I -1,410 -280 0 -1,690 -390 -390 2 54 0 2 350 -75 0 275 -350 -350 2.54 0 3 350 -75 0 275 -300 -300 2 54 0 4 350 -75 0 275 405 405 2 54 0 5 350 -75 0 275 405 405 2 54 a 6 350 -75 0 275 405 405 2.54 7 350 -75 0 275 405 405 2 54 a B 350 -75 0 275 405 405 2 54 0 9 350 -75 0 275 405 405 2.54 0 10 350 -75 0 275 405 405 2.54 0 11 350 -75 0 275 405 405 2.54 0 12 350 -75 0 275 405 405 2.54 a 13 350 -75 0 275 405 405 2. 54 0 14 350 -75 0 275 405 405 2 54 0 15 350 -75 0 275 405 405 2 54 0 /a Relevant ratios of social and economic prices to market prices, taken from tables I and 2 and appiLed to this model are as follows OUTPUT INPUTS Ratio Of Ratio Of Economic Price To Social Price To Economic Price To Social Price lo Market Price Market Price Mark,t Price Market Price Durum Wheat 0.86 0.86 Seeds 0.50 0 80 Barley 1.02 1.02 Land Preparation 0.54 0 60 Maize 0.99 0.99 Hlired Labor 0 51 0.51 Milk 0.80 0.84 Animal Feed 1.02 1.02 Sheep 0.55 0.55 Physical Contingencies 0 60 0 60 Taxes 0.00 0 00 Draft Mtules 0 47 0 53 Plow 0 Su 0.55 llarrow 0.50 0.55 /b Incremental incorie is taken from the cash flow projectioni for his model. Since the savings rate is zero, Incremental uonsumption equals incremental income. /c The distribution weight does not decline since Incremental consumption increases at about 1% p a. ewer tlhe first 4 years ol the Pru,ercc life, which is less than the expected increase in average per capita consumption in Morocco. /d The cousuu.iption adjustment equals zero since this tarm model represinis that at the crttical coLnsumption le.e August 1979  nuiwuCCu FOURTH AGRICULTURAL CREDIT PROJECT Model 2 - Draft Animals and Agricultural Implements Social Rate of Return Sensitivity Analysis Year n=O, v=1.43 n=O, v=3.65 n=l, v=3.65 n=2, v=1.43 n=2, v=3.65 d Consumption Adjustment Parameter (B-v) 1 0 -.43 0 3.81 1.07 2 0 -.43 0 3.81 1.07 3 0 -.43 0 3.81 1.07 4 - 15 0 -.43 0 3.81 1.07 d Consumption Adjustment - C(B-v) 1 0 165 0 -1,485 -415 2 0 150 0 -1,335 -375 3 0 130 0 -1,145 -320 4 - 15 0 -175 0 1,545 435 d Net Social Benefit Stream (Net Economic Benefit - C(B-v) 1 -1,410 -1,245 -1,690 -3,175 -2,105 2 350 500 275 -1,060 -100 3 350 480 275 -870 -45 4 - 15 350 175 275 1,820 710 Rate of Return 24 21 14 24 20 August 1979  MOROCCO F(UKlTl ACRICIJL'nlRAL CREDIIT VR0J IcT Model 3 - Stable Constr,uctiont and Purchase of 2 Crossbred In-Calf D.ifers for Milk Productijn Social Rate of Return Analysis Excess of Net Derivation of cL,Lsumpti.n :Ienefit/Cost I r C.lsu,-n 3 Social over Social Net Economic Benefit Of Neta Social Ilcremental jistribul ion ConSuqpLt inn Net Economic Benefit Of Incremental Ben fiL locre,wnnta I Cuonsninsjption WeighL Ad just mi I Year Benefits /a Inputs/Outputs Consumption Stream Income (8C)/b (d)/c (d/v-B) (1) (2) (3) (4) (5) (6) (7) (8) I -7,340 -150 0 -7,490 -3,34, 0 775 -.51 2 110 95 0 205 -620 0 .74 -.50 3 1,240 85 0 1,325 670 0 7!, -.50 4 1,990 75 0 2,065 1 555 0 .73 - .51) 5 1,240 85 0 1,325 840 0 .73 - .50 6 1,240 85 --800 525 2,960 1,605 .72 -.50 7 1,240 85 -860 465 2,t60 1,685 .71 -.51 8 1,241) 80 -380 940 960 750 .71 r>I 9 1,240 80 -860 460 2,16,0 1,685 .70 - ,1 10 1,520 80 -1,180 421) 2,960 2,310 .70 -.I! 11 1,520 80 -1,180 420 2,960 2,310 .7)) - 1 12 1,520 80 -1,180 420 2,960 2,310 .71) -. A 13 1,520 80 -1,180 420 2,960 2,310 .70 -.91 14 1,520 80 -1,180 420 ?,960 2,310 .70 51 15 1,520 80 -1,180 421) 2, ,0 2,310 .70 - . /a Relevant ratios of social and economic prices to markec prices are as follows OliTPUT [N1UTS Ratio )f Rati, o1 Eco,,onic Price S,-cial Ilrice Eco,,onic P rcc- S,,c i I Iri,e to to to L Market [rice Market Pri-e Market Irice Marknt irIc, Dur-,, VWLeat 0.8b 0.86 Se.d 0. 51 n. 80 Barley 1.02 1.0' Iahor 0.51 (,SI Maii, 0.99 0.99 A,, Ial Feedl 1.02 1 AK Milk - Sold U, prU. ess r 05'1 (54 M I eL ().47 ' 1 Milk.- Cons6med fa frat, 0.8() IJ. 84 (Cn I 0leS 1). 6') .0 MeaL. 0.9) ().93 Cc 1sslCLed 11.eifIr, . i i.t, 0.48 () 707. of milk produced is assumed con,sumed on the farm or traded locally (wi LIh ncighlbors or in thie village). 30. is assn,,ed sold lJ processors.- T)is corresponds roughly to the 20-30% of Moroccan milk production whii(J is sold to processors. /b Therc is a loss of income during the first two years of thie Project. This is ass-,med to be financed fromn, fann savings or credit from traditional sources (money lenders, family, friends). ilousehold censum,ption is assumed not atfected. 1owever, lousehold cn,n- sumptiont is assumed nut to increase until this negative cash flow f r the firsL Lwo years of tihe Iroject is rec..irstit,,ted (ie. Ot Lil1 after ctunulative incremental cash flow isgreater tian zero). /c Ilncome be fore debt service i,,creases at the average rate ol 2.77. ptr a, h..m, co,,q,ar,d to average Mo,roccan 1ner capi ta in-co,ve erowinL, at 27. Per ann,nm. August 1979  MOROCC( FOURTII ACR ICILTURAI. CREISiT PRO.J1I(:E( Model 3 - Stable Constructioni and Purcliase of 2 Cr,,ss-Bred In-C.l1L Ileifers for Milk Production Social Bate of Return Se-sitivity Analysis Year n=O, v=1.43 n=0, v=3.65 n-1, v=3.65 n=2, v=1.43 n-2, v=3.65 d Consumption Adjustment 'Factor 1 0 -.43 -0 -.31 -.55 2 0 -.43 -.'50 -.32 -.55 3 0 -.43 -.50 -.32 -.55 4 0 -.43 -.3( -.33 -.55 5 0 -.43 -0 -.33 -.55 6 0 -.43 -.j5 -.34 -.56 7 0 -.43 -.51 -.35 -.56 8 0 -.43 51 -.35 -.56 9 0 -.43 -.'1 -.36 -.57 10 0 -.43 -.1 -.36 -.57 d Consumption Adjustment - f-V) 1 0 0 0 0 0 2 0 0 0 0 0 3 0 0 0 0 0 4 0 0 0 0 0 5 0 0 0 0 0 6 0 -690 -800 -545 -900 7 0 -725 -860 -590 -945 8 0 -320 -3130 -260 -420 9 0 -725 -8150 -605 -960 10-15 0 -995 -1,1130 -830 -1,315 d Net Social Benefit Stream (Nel: Econoxmic Benefit -ac(,B-V) I -7,