|
|
4. Farm Investment Analysis
Once market prices have been determined for those items that enter the cost and benefit streams, this information must be arranged in "pattern" accounts to begin the assessment of the effects a proposed project will have on the farmers, public and private enterprises, and government agencies that will participate in project implementation. These accounts are central to the financial analysis of agricultural projects; they are always based on market prices.
Although we will touch on the essential elements of financial analysis, much more could be taken up here. Just how elaborate the financial analysis must be for a particular project will depend on the complexity of the project. Most agricultural projects will call for a financial projection based on at least one pattern farm plan that is assumed for participating farmers. This pattern (or model) farm plan projects resource use and income flows for a group of similar farms participating in the project. The financial projections for the private and public firms or project entities may be quite summary in nature for a simply organized project, but a project in which several different firms and project entities are concerned or one that poses special financial problems may involve a much more complex analysis. The major accounts needed will be outlined in this chapter and in chapters 5 and 6, and pattern formats suggested. This will enable the project analyst to proceed with the confidence that he is preparing an acceptable financial analysis. For more complex projects, especially those projects involving more complex public or private firms with specialized cost, revenue, or financing situations, the project analyst will have to move beyond what is discussed here. Agricultural project analysts may want to turn for more technical help to financial analysts or accountants, just as they would turn to agriculturalists or livestock technicians for their particular expertise. Many financial analysts and accountants, however, will not be familiar with methodologies of agricultural project analysis or with the particular analytical needs of the financial aspects of these projects, so even financial analysts and accountants may find that the following discussion may help them respond to a request for assistance.
Objectives of Financial Analysis
Six major objectives for financial analysis occur in analyzing agricultural projects.
Assessment of financial impact
The most important objective of financial analysis is to assess the financial effects the project will have on farmers, public and private firms, government operating agencies, and any others who may be participating in it. This assessment is based on an analysis of each participant's current financial status and on a projection of his future financial performance as the project is implemented. Detailed financial projections are needed for this analysis.
Judgment of efficient resource use
The overall return of the project and the repayment of loans extended to individual enterprises are important indicators of the efficiency of resource use. For management especially, overall return is important because managers must work within the market price framework they face. Farm investment analysis and financial ratio analysis provide the tools for this review. Project analysts and others concerned with decisions on policies for national economic growth and development will have to look beyond the financial analysis-at market prices-and form a judgment about the effects of the project on real resources for the economy as a whole. In chapter 7 we take up this issue in the discussion of how to determine economic values.
Assessment of incentives
The financial analysis is of critical importance in assessing the incentives for farmers, managers, and owners (including governments) who
will participate in the project. Will farm families have an incremental income large enough to compensate them for the additional effort and risk they will incur? Will private sector firms earn a sufficient return on their equity investment and borrowed resources to justify making the investment the project requires? For semipublic enterprises, will the return be sufficient for the enterprises to maintain a self-financing capability and to meet the financial objectives set out by the society?
Provision of a sound financing plan
A principal objective of the financial analysis is to work out a plan that projects the financial situation and sources of funds of the various project participants and of the project itself. The financial plan provides a basis for determining the amount and timing of investment by farmers and for setting repayment terms and conditions for the credit extended to support the investment. It provides the same basis for an assessment of the investment plans and debt repayment capacities of public and private firms participating in the project. Finally, for the project as a whole, the financial plan is the basis for determining the amount and timing of outside financing-whether from the national treasury or from international sources-and for establishing how rapidly the borrowed resources should be repaid. The estimated effect of inflation on both revenues and costs should be taken into account in making this assessment.
Coordination of financial contributions
The financial plan allows the coordination of the financial contributions of the various project participants. The coordination is made on the basis of an overall financial projection for the project as a whole. It addresses itself to such questions as whether the availability of resources from the treasury or international agency is matched with farmers' investment capacities and available funds for investment and operating expenses as well as with the timing of expenditures for project investments such as feeder roads and irrigation structures and for working capital needed for stocks in processing industries and the like.
Assessment of financial management competence
On the basis of a projection of the pattern financial accounts, especially for the larger firms and project entities, the analyst can form a judgment about the complexity of the financial management the project will require and about the capability of those who will manage the project's implementation. And from this assessment, the analyst can then judge what changes in organization and management may be necessary if the project is to proceed on schedule and what specialized training may be advisable.
Source: Schaefer-Kehnert (1980).a. Also called sources-and-uses-of-funds analysis.b. Benefit-cost analysis of on-farm investments.Preparing the Farm Investment Analysis
The starting point for both the financial and the economic analysis of an agricultural project is generally a group of investment analyses of pattern or model farms, based on budgets for individual pattern farms. These pattern farm budgets compare the situation with the project to that anticipated without the project for the duration of the project. They enable the analyst to form a sound judgment about the likely benefit to farmers of participating in an agricultural project and about the incentives for farmers to do so.
Farm investment analysis is the topic of this chapter. This analysis is similar to, and sometimes confused with, farm management analyses done by agricultural economists, which may be distinguished as farm income analysis and funds flow analysis. The differences are summarized in table 4-1.
Farm income analysis is generally used to evaluate the performance of a farm in a particular year. Its objective is to help improve the management of the farm. Current prices are used, and a depreciation allowance is included to account for that portion of longer-term capital investment used up in the year being considered. Noncash items such as home-consumed production and payments in kind are included. Off-farm income and expenditure are excluded because the analysis is intended to evaluate the performance only of the farm itself. The analysis provides an estimate of the return to capital invested and to the farmer's labor, and this may then be compared with the return to alternative cropping patterns or to off-farm opportunities.
Funds flow analysis, also called sources-and-uses-of-funds analysis, is used to determine a farmer's liquidity in an analysis of his credit situation. Only cash items, including purchase and sale of capital goods, enter the analysis. Off-farm cash income and expenditure are included, but home-consumed production is not. The analysis shows the cash available to the family over a period of time.
Farm investment analysis, in contrast, is undertaken to determine the attractiveness of a proposed investment to farmers and to other participants, including the society as a whole. It projects the effect on farm income of a particular investment and estimates the return to the capital engaged. It follows the principles of discounted cash flow analysis (discussed in detail in chapters 9 and 10). The analysis is projected over the useful life of the investment. The initial investment is shown at the beginning of the projection, and a residual value at the end. In general, the analysis is cast in constant prices, although allowance may have to be made for inflation. Off-farm income is included. Even though we use the term "cash flow," noncash elements enter the projection, including home-consumed production and payments and receipts in kind. (The term was first applied to industrial investments, in which noncash elements are less common.) When doing farm investment analysis, some elements of funds flow analysis are often incorporated to enable the analyst to assess the farmer's liquidity and his credit use. Those who wish to pursue farm income analysis or funds flow analysis in relation to project analysis may refer to standard farm management texts such as Harsh, Connor, and Schwab (1981) and Kay (1981). Those who want more detail about the application of farm budgets to project analysis may consult Brown (1979).
Pattern farm investment analysis that includes farm budgets should be prepared for almost every agricultural project. Although agricultural project analyses that do not have farm budgets are used, it is increasingly accepted that farm budgets are an extremely desirable, if not essential, part of project analysis. The benefit stream of an agricultural project may be built up simply by multiplying the total area to be planted by the expected yield, essentially treating the whole area as one undifferentiated farm. If this is all that is done, it may hide crucial information about the effects of the project on individual farmers and obscure underlying unrealistic assumptions. Even when the project involves only a public sector undertaking, a farm budget is likely to be necessary to test the feasibility of the cropping pattern and the financial viability of the enterprise.
The purpose in preparing farm investment analyses for a project is not to take a sample of the farms in the project area. Rather, it is to select major farm types expected to participate and to look at the impact of the project on them. These farm investment analyses are usually projections for the life of the project, often twenty to twenty-five years, not for just a single year. The analyst will want to examine the cropping pattern and perhaps to diagram it; to determine the labor that will be required if farmers are to participate in the project and perhaps to prepare a month-by-month labor budget showing requirements and the availability of family labor; to look at production and inputs; and, finally, to prepare a farm budget in the detail needed for understanding and evaluating the effects of the project on the income of participating farmers. From these, the analyst can assess the financial effect of the proposed project on typical farms-both to judge incentives for participation and to determine whether national policies on minimum incomes for project participants are being met.
Farm investment analyses and farm budgets can of course be prepared for farms of any size. The problems of analyses for smaller farms are the focus here, since many, if not most, agricultural projects in developing countries will be directed toward smallholders whose families consume a large part of the food they produce.
Large commercial farms and plantations, however, whether publicly or privately held, are more like other business enterprises than they are like small, family-operated farms. Projected accounts for these large agricultural undertakings are probably more appropriately cast in formal financial statements such as those of the agricultural processing industries discussed in chapter 5.
In considering small farms, the analyst will be particularly concerned with the effect of the project on the total income of the farm family. The aspects of the small farm as a family unit and as a business firm must be clearly understood and appreciated. These will differ from society to society, and the project analyst should either know the society well enough to anticipate the farmers' response or be advised by others who do. One must assess the attitude of the family to proposed cropping patterns that involve more days of labor, to patterns that increase cash crop output and reduce food crop production below household requirements, to patterns that change the work responsibilities of men and women, and to patterns that require the family to run a considerable market risk. Farmers are price responsive, of course; the extensive research has amply confirmed this (Krishna 1967). But farmers live in a particular cultural and risk environment, and project analysts must take this environment into account when they project their pattern farm investment analyses.
Backed by this understanding of the particular cultural environment, the analyst will prepare the farm investment analyses as realistically as possible to determine what the family gains by participating in the project. The projection must be based on a specific package of technological innovation. The effectiveness of the proposed new technology on small farms must be realistically assessed, and the technological assumptions must be checked to ensure that they reflect on-farm conditions and not those of an experiment station. The analyst must form a judgment about how rapidly farmers will be willing to adopt new practices. The farm investment analysis should confirm that adoption of a new technology will really be financially worthwhile, for farmers can respond to financial incentives only when it is truly remunerative for them to do so. The analyst must determine how much credit will stimulate farmers to adopt new practices and must assess how risky a new technology is and how variable the farm income may be under the project. The analyst will want to test the effect of risk on family income by determining what happens if yields fall below expectations or if prices are lower than anticipated and by undertaking similar sensitivity tests. Through such tests a margin to allow for bad years can be built into the farm plan.
Although in agricultural projects the analyst generally looks at budgets for entire farms, partial budgeting techniques can be used for undertakings that involve only a relatively minor change in the farm organization. To do this, one looks at the marginal cost (including opportunity cost) of adding a production activity and compares it with the marginal increase in benefit that the new activity will bring. Partial budgets are an effective tool for helping to search out the best combination of production activities. Brown (1979) discusses their use in some detail. In most projects, however, we expect rather substantial changes over a prolonged period, and under these circumstances it is better to project whole farm budgets. Then the total effect of the project on family income can be better assessed.
The information on which the project analyst will base his farm investment analysis will come from many sources. Project analysts will have to rely heavily on their professional colleagues to determine a sensible cropping pattern and livestock activity for a proposed project, the output that may be expected, the inputs that will be required, and the relevant prices for products. The project analyst will want to pay particular attention to the realism of the estimates provided by the agriculturalists and livestock specialists he consults. Unrealistic assumptions about yields, input levels, or rates of farmer acceptance and, hence, of buildup in project benefits will negate the best of project analyses.
The analyst will certainly want to visit the site of the proposed project and typical farms that will be included. Nothing substitutes for the firsthand knowledge that being there brings.
A crucial source of information in every agricultural project is the farmers themselves. Only through interviews with farmers can a project analyst reach a valid conclusion about the realism of his farm investment analysis. The project analyst will want to interview farmers about their present cropping patterns, labor requirements, use of inputs, and the market prices actually received and paid. He will want to gain a sense about the farmers' willingness to participate in the project were it to proceed. In a project area or on a similar site some farmers often are already using a proposed new technology. It is most important that these farmers be interviewed to tap their experience. The analyst will want to know the yields the farmers actually have realized with the new technology, the inputs they actually must employ, and their general comments on the new technology and cropping pattern proposed for the project. The analyst will want to assess the labor requirements farmers have found necessary to use the new technology.
Interviewing farmers is an art in itself, and only a few comments can be made here. The information farmers give will usually be contradictory, but out of a group of interviews the project analyst can gain a sense of feasible technological and financial relations. Farmers will have to be interviewed in the field, not in the office. The analyst will have to know the local measurements and not expect everything to be reported neatly on a unit basis. The analyst or one of his staff should probably conduct the interviews alone or with very few other people around. Great care should be taken to establish a good atmosphere in the interview so that farmers are not overawed by the analyst's presence; farmers should also know that the information they give will not be used for tax purposes. A formal questionnaire may be helpful, especially if much information is to be collected by assistants, but any questionnaire should be carefully pretested in the field before use. It may be better for the analyst or his assistants to fill in the questionnaire only after the interview is complete and the interviewer has left the farmer. In any event, before the interviews the analyst should have formed a clear idea about the information he needs, perhaps in the form of a list of questions, so that critical information will not be overlooked. Questions put to farmers should be as specific as possible. Most information gathered should relate to actual experience, perhaps in the last cropping season. Questions about hypothetical situations should be avoided to the extent possible. The very nature of seeking information about a proposed project will, however, of necessity involve many "what if" kinds of questions. Cropping patterns are usually based on the judgment of the agriculturalists and livestock specialists working with the project analyst. Their judgment, in turn, will be based on their familiarity with the agriculture of the area, on research results and the results of pilot projects-perhaps undertaken especially as part of the project preparation-and on their knowledge of the farmers who will participate in the project. In most instances, experienced technicians can propose realistic patterns close to the optimum, but sometimes linear programming may be used as a more formal methodology to optimize cropping patterns. Linear programming has been applied in preparing agricultural project analyses but is not regularly used either in national planning agencies or in international lending agencies. It is a complex methodology that requires more formal input-output data than does simple budgeting, and in practice it requires computers. There are serious methodological limitations to the use of linear programming for agricultural project analysis: problems of dealing with risk, farmers' cultural traditions, variability of soils within farms, water availability in different areas of a farm, and other farm-level variations. Even so, when preparing a project for an area where there is inadequate experience to rely on in forming subjectively determined cropping patterns or when dealing with very complex patterns, the project analyst may want to consult specialists in linear programming for assistance. In these cases, the project preparation takes on some of the character of a research effort. Because it is a well-known methodology widely used in farm management research, many agricultural colleges have staff familiar with linear programming.
In most agricultural projects, about half a dozen or so pattern farm investment analyses will suffice, but generalization about this is dangerous. The number of pattern farm analyses depends entirely on the complexity of the project. The analyst will want a pattern farm investment analysis for each major group of soil and water conditions in the project area and for each major difference in the size of holdings. Of course, each major cropping pattern or livestock activity will require a separate farm investment analysis. Remember that the objective is an indication of the effect of the project, not some kind of rigorously drawn, random sample. In practice, the number of farm budgets prepared for any given project analysis is a tradeoff between the complexity of the proposed project and the availability of staff to prepare the investment analyses.
Each farm investment analysis will be the result of careful consultation with technical specialists and interviews with farmers. Just as it is not possible to generalize about how many pattern farm budgets will be necessary, neither is it possible to generalize about how many interviews with farmers will be needed. Thorough preparation of a complex project may require twenty-five to fifty or even more interviews to provide the information for each farm analysis. But a simpler project that will use a better-known technology may require only half a dozen to a dozen interviews for each pattern farm budget. A group doing an appraisal of a proposed project would probably interview fewer farmers than was necessary in the initial project preparation. Each situation will have to be judged by itself in the light of how confident the analyst needs to be about the project analysis, how complex the project is, how well known the technology is, and how available are staff for project preparation.
Similar considerations apply in deciding the level of detail necessary in a farm investment analysis. Any farm investment analysis is intended to improve the decisionmaking for a project. It is, of necessity, an abstraction. This imprecision is forced on us by the very fact that we must predict future events, but it also arises from the question of just how much detail is necessary. In every farm investment analysis, the project analyst will reach a point at which further elaboration or further detail would make such a marginal contribution to the investment decision that it is not worth the time. Just when that point is reached will vary from project to project according to the circumstances of the project and the circumstances of the decisionmaking process.
It is easy to conceive of a set of pattern farm investment analyses that would be so enormously detailed and have so many different budgets that the process would quickly become bogged down in detail. Because of staff limitations and because of the approximate nature of the underlying data, it is better to hold both the number of pattern farms and the level of detail to the minimum that will serve to lay out clearly the major points about the project.
The project analyst will have to determine how best to present his information so that those who must review his work and make decisions about the project can work efficiently and yet have the information they need. The major entries, the level of detail, and the like will vary from project to project. Some of the elements given in separate tables in the illustrative examples in this chapter may be better combined in the tables that present a particular project. In many project reports, only a summary of background information, plus a detailed farm budget, will be needed. Other, more detailed tables can be included in annexes or in a separate volume of background information reproduced in limited quantities and circulated only to those most interested in the project. This kind of additional data can even be kept in a separate project information file that can be made available to anyone seriously interested. (In the Paraguay report used as an example in this chapter, the analyst presented his farm investment analysis in four annex tables and collected the supporting information in a separate project information file.)
What we will present here is a pattern format that includes the features most commonly of significance in agricultural project analysis. This pattern format uses a terminology generally accepted by both farm management specialists and accountants. In using the format, the analyst will have to determine for himself, for each project in his charge, exactly how much detail is necessary to support the analysis and exactly how this detail is best reported to facilitate decisions about the particular project.
Elements of Farm Investment Analysis
The principal elements of farm investment analysis are outlined in this subsection and are listed in table 4-2. A flow chart for preparing the analysis is given in figure 4-1. Not every element will be necessary in every analysis, and the means of presenting the elements will vary from project to project according to circumstances.
The most important elements of a farm investment analysis can be illustrated by an example adapted from the Paraguay Livestock and Agricultural Development Project. (Tables that illustrate particular elements of the analysis are noted in table 4-2.) The project is to increase agricultural production, productivity, and income on some 940 livestock farms and some 3,000 mixed farms, mostly small, through on-farm investments supported by credit, technical assistance, feeder roads, and market improvement. Most of the important aspects of preparing a farm investment analysis are touched on in this example, but it is clear that no single example can cover every possibility. Each project analyst will want to build on the illustrative tables that outline this one example for the purposes of his own project analysis. As in all accounts, the objectives of the farm investment account determine its content and format.
Accounting convention for farm investment analysis
Because farm investment analysis follows the principles of discounted cash flow, it is convenient to adopt an accounting convention that is congruent with those principles. [This convention has been called "time-adjusted" by Schaefer-Kehnert (1980), who has elaborated its use.] The discounting process used in discounted cash flow analysis implicitly assumes that every transaction falls at the end of the accounting period. It is desirable that the farm investment analysis match this assumption. This is simply accomplished if we consider the initial investment to take place at the end of year 1 of the project, regardless of whether it will actually take a full year or only a few weeks. Year 2, then, is the first accounting period in which increases in operating cost and incremental benefits occur. Thus, the dividing line between the end of the initial investment period and the beginning of the incremental production operations coincides exactly with the dividing line between years 1 and 2 of the project. (Some analysts accomplish much the same result by considering investment to fall in year 0, but this gives rise to problems when cash flows are aggregated.) Considering that preparing a farm plan, making a loan, constructing or purchasing investment items, and purchasing new inputs can take at least several months to a year, reserving year 1 for investment is not unrealistic. Doing so, however, is not dictated by real events but by the accounting convention.
If all transactions are considered to fall at the end of the accounting period, then we must allow for the availability of the needed operating
Figure 4-1. Flow Chart for Farm Investment Analysis
expenditure at the beginning of the cropping season. This is accomplished by incorporating in the analysis an entry for incremental working capital at the end of the preceding year. The amount of the working capital needed is related to the farming system being analyzed. If a single annual crop is produced, then nearly all the operating expenditure will be needed at the beginning of the crop year. But if two crops are to be produced in succession, only the operating expenditure for the first crop need be on hand at the beginning of the crop year, since there will be a harvest during the year that will provide proceeds to replace the input supplies needed before the second crop is harvested. Thus, only half the total annual operating expenditure need be on hand at the beginning of the year. The incremental working capital needed (either an increase or a decrease) at the beginning of the year, then, is entered at the end of the year preceding the year when it will be expended for production. A set of recommended adjustments in incremental operating expenditure to obtain incremental working capital is given in table 4-3. Introducing an incremental working capital stream reflects real resource use. When an investment is undertaken, short-term inputs such as seed, fertilizer, feed, and the like must be on hand. They are replaced from the proceeds of the harvest or livestock sales during the year and are again on hand for production in the following year. If operations are to expand the next year, then stocks of inputs for production must be increased, and this will be reflected in another incremental working capital entry. Since the incremental working capital is entered separately, it will easily be included in the total investment shown when the farm budget is prepared and will not be inadvertently overlooked. At the end of the project, the incremental working capital for each year is added together algebraically and taken out of the project as part of the residual value. Thus, including incremental working capital in the accounts does not result in double counting.
One practical outcome of this accounting convention is that operating expenditures and benefits in year 1 generally remain the same as they were without the project. In some cases new investment might require an increase in operating expenditure in year 1, even though production
Source: Schaefer-Kehnert (1980).would not be affected until year 2. In other instances, both operating expenditures and production might actually decrease-as might happen if new irrigation canals were to be constructed, disrupting farming operations. In laying out the farm investment analysis, however, it is usually considered that working capital is not freed as a result of the investment so that the layout of the farm budget can be simplified. If this were not done, both data for the case without the project and data for a "preproject" year or "year 1" would be needed to accommodate the decision rule that working capital be a proportion of the increase or decrease in the operating expenditure for the following year.
Showing working capital as a separate entry facilitates determining how much short-term credit may be needed by the farmer. A judgment may be made about whether the farmer will have savings from which to finance increased working capital or whether some proportion or all will have to be covered by extending the farmer a short-term loan, which then can be incorporated into the financing section of the farm budget.
The accounting convention adopted here is not much different from that most commonly used by project analysts. The most important difference is the rule of reserving year 1 for investment only and assuming the investment to fall at the end of the year. It is more common to include investment in year 1 but to assume that it will occur at the beginning of the year, even though the discounting process assumes it falls at the end of the year. Production is then assumed to be increased in year 1, an assumption that leads to an overestimate of the rate of return on the capital used. It also leads to a considerable overestimate of the farmer's income in the early years of the project and, hence, to an underestimate of his need for both long-term and short-term credit. The other difference between the accounting convention adopted here and that most commonly used by project analysts is only a matter of completeness. It is easy inadvertently to omit or underestimate working capital unless such capital is included in the convention. This convention for working capital leaves the crop year intact and therefore facilitates the supporting technical projections.
Farm resource use
Once the agronomists, livestock technicians, and other technical specialists have determined the components of a proposed farming system for a pattern farm, the analyst may proceed to prepare the farm investment analysis.
LAND USE.
The first step is to determine what the land use on the farm will be. The land use for the Paraguay project is given in table 4-4. Note that the crop year is taken to extend from July of one calendar year to June of the next, since this arrangement makes the break in the year come during the Paraguayan winter season, when there are the fewest crops in the ground. The total farm area is 20.0 hectares, divided into cultivated area, pasture, forest, and a house plot. (Throughout the text of this chapter,
Source: Adapted from World Bank, "Staff Appraisal Report: Paraguay-Livestock and Agricultural Development Project," 2272-PP. (Washington, D.C., 1979; restricted circulation), annex 1, table 17.a. Operated by one family with six members and a work potential of 70 work days a month. A work day is the time (generally eight hours) devoted by one person during one day.b. Maize and manioc are intercropped.c. Double-cropped after cotton or soybeans. The area given is the area planted in the year shown.d. Does not include second crop area, in this case the area planted in sunflower.e. Cropping intensity is determined by dividing the total cultivated area by the total cropland.
the most common or generalized categories from the pattern tables will be shown in italic type. These items would be considered in any farm investment analysis. In all tables the generalized analytical framework applies, but in some the categories are all specific to the project analyzed.)
The land use, in accord with the accounting convention adopted, would remain unchanged in year 1, except for establishment of improved pasture that is part of the investment. In year 2, in which the proposed cropping pattern calls for sunflower to be introduced and to be double-cropped after either cotton or soybeans (depending on the year), both the total cultivated area and the total cropland are shown. The total cultivated area is the total area planted in crops, whereas the total cropland is the area available to cultivate. When the total cultivated area is divided by the total cropland, the cropping intensity is obtained. In year 2, for example, the cropping intensity is 1.4 (7.0 - 5.0 = 1.4). Many analysts prefer to report cropping intensity in percentages, so this would be reported as 140 percent. One check of the feasibility of the cropping pattern is the intensity. The analyst should be cautious about accepting a cropping pattern that has a very high cropping intensity or one that is markedly different from the pattern existing in the area. Farmers may well have good reasons for not driving up the intensity.
Many analysts also like to devise a cropping diagram, such as that given in figure 4-2, and this should be subdivided to indicate any existing farm plots. Such a diagram is usually drawn up only for the full-development situation. The diagram indicates the area to be devoted to each category of land use and each crop. In the case of our example, it extends over two years to show that the seasonal timing of the cotton-sunflower-soybean rotation occurs on one plot in one year and on another in the next. Checking vertically at any one time, we can be sure the cropping pattern does not call for more area than the farm has. Checking horizontally, we can determine when each crop must be planted and whether enough land will be available at the proper season. The left and right sides of the boxes showing the area to be planted in each crop are slanted to indicate the planting and harvest time necessary for each crop. Examining the cropping diagram can help determine if there will be adequate time between crops to prepare the land.
Figure 4-2. Land Use Calendar for Project Years 6-20, 20-Hectare Mixed Farm, Livestock and Agricultural Development Project, Paraguay
LABOR USE
A second aspect of the farm resource use is labor. To determine the labor the farm will require, we need to know the labor used to cultivate a hectare of each crop in each project year. It is desirable to be able to see this in two forms, by operation and by month. Table 4-5 shows the annual labor requirements for 1 hectare by crop and operation for the Paraguay project example. It includes the labor required not only for the various crops to be produced, but also for the pasture. The labor unit is a work day, the time (generally eight hours) devoted by one person during one day. The labor requirement for crops drops sharply between years 1 and 2 because of the introduction of draft animal power. In the case of pasture, the labor requirement for fencing and seeding is included in years 1 through 3. In the Paraguay model, other activities of pasture establishment are to be undertaken by a contractor, so there will be no call for labor from the farmer. If, however, labor for establishing some other kind of improvement were expected, such as farmers' digging their own tertiary canals in an irrigation project, then this should be included in the labor requirement.
Note: For the area in various crops, see table 4-4.Source: Adapted from A. O. Ballantyne, "Paraguay-Small Farmer Credit Component, Livestock and Agricultural Development Project," working papers on file (Washington, D.C.: World Bank, 1978; restricted circulation).a. Maize and manioc are intercropped. Hence, during the period when both are growing, the allocation of cultivation time between the two crops has an arbitrary element.
The total labor requirement per hectare for each crop is distributed by month in table 4-6. The monthly distribution is most important because we must determine not only the total annual labor requirement on the farm but also its timing to assess whether sufficient family labor will be available and, if there is not enough, how much hired labor will be needed. Although some farm management analysts break down the labor requirement by week or fortnight, for purposes of project analysis the monthly distribution is sufficient.
On a mixed farm, livestock will also require labor. This may be calculated by determining how much time will be needed per animal unit in the livestock herd. An animal unit is a measurement of feed demand by a particular class of animal. (This is discussed in more detail in the appendix to this chapter. The total animal units for each year are reported in table 4-11. They are given in that table for the livestock herd that could exist on 100 model farms to avoid the problem of divisibility that arises when the increase in large livestock on a small farm is projected. Thus, to obtain the animal units on one farm, the total reported in table 4-11 must be divided by 100. We will return to a discussion of this convention in the next subsection, when we discuss the herd composition.) In the Paraguay example, it is assumed that each animal unit will require five minutes of care a day and that the requirement will be the same each month throughout the year. The labor requirement is determined on the basis of the animal units at the beginning of the year. Using animal units rather
Note: Same as table 4-5. Source: Same as table 4-5.a. See note a, table 4-5.b. Labor requirements for sunflower apply to the year of planting. Thus, in year 2 the labor requirement is for the crop planted in May of year 2 and harvested in October of year 3.
than each individual class of animal as the basis for estimating the labor requirement considerably simplifies the computation and is not unrealistic. In the convention recommended here, for example, closing livestock figures do not include heifers two to three years old or steers sold during the year, in this case steers three to four years old. Opening livestock figures, however, omit calves. In reality, closing and opening figures tend to balance each other, since heifers will be transferred to the breeding herd throughout the year and surplus heifers and steers will be sold throughout the year, whereas calves will be born throughout the year. Attempting a more precise estimate would only lead to superficial precision because the error in estimating the daily requirement for labor considerably exceeds any gain in accuracy.
Having determined the labor requirement for each crop or animal unit by month, we proceed to calculate the labor requirement for the pattern farm. This is given for the Paraguay example in table 4-7. Here the labor required for each crop during each project year is given. The total by month and the amount to be provided by family labor and by hired labor are determined. In the Paraguay example, it is assumed that the family on the pattern farm will have available 70 work days of labor a month and that any labor requirement in excess of this amount will be supplied by hired labor. This is a very mechanistic assumption, of course. Not only will families vary widely in the labor they have available-even on farms of quite comparable size and cropping pattern-but families will also tend to work longer hours in busy seasons and rest in the off-season. For purposes of the farm investment analysis, however, this approximation is quite sufficient, given the wide margin of error in the estimates of the labor requirement in general. In the Paraguay example, table 4-7 shows that hired labor will be needed on the farm from year 4 onward. By year 7, in the peak month of March, about 44 percent of the total labor required for cotton and soybeans will have to be hired.
When a labor budget shows a need for hired labor, as this example does, the project analyst must consider carefully whether the labor will be available in the project area. Totaling the hired labor requirement for the project as a whole is one of the real advantages of including the labor budget in the farm investment analysis, since the analyst must then consider the realism of the proposed pattern in light of the added hired labor that can reasonably be expected to be available in the project area. Postulating 56 additional work days of hired labor in March is one thing; whether such additional labor would be available for an entire proposed project is another. It may be that a proposed cropping pattern will prove unrealistic in its requirements of additional hired labor, and a less labor-intensive cropping pattern must be proposed. Furthermore, if the project will call for substantial amounts of additional hired labor in relation to the supply available in the region, this may have implications about the sources from which the labor must be drawn and, hence, about the opportunity cost of the hired labor. In turn, this opportunity cost will have to be considered when making the estimates of the economic value of the labor (see chapter 7).
ha = Hectares; a.u.= animal units.Source: Calculated from tables 4-4 and 4-6.a. Assumes five minutes a day per animal unit, eight hours a day, and thirty days a month. The labor requirement is based on the animal units at the beginning of the year (as given in table 4-11) divided by 100 to give the labor requirement for a single farm. See text (the subsection "Farm production. Livestock") for a discussion of this convention.b. Assumes 70 work days per month of available family labor.c. Area and labor requirement for sunflower apply to the year of planting. Thus, the area assumed in the first five months of project year 5 is the 2.2 hectares planted in May of project year 4.d. Calculated on the basis of improved pasture established in previous years (as shown in table 4-4). Thus, in year 2 it is based on 3.5 hectares and in year 3 on 7.0 hectares.
Some analysts like to work out a labor use diagram such as the one shown in figure 4-3. This usually is done only for the full-development situation. The graphic presentation makes the problem of peak labor requirements readily apparent.
Once the total hired labor has been determined, it must be allocated among the various crops so that it may be included in the proper category of operating expenditure. This is done in table 4-8. In the Paraguay example the allocation is made in proportion to the total work days
Figure 4-3. Labor Use Diagram for Project Years 7-20, 20-Hectare Mixed Farm, Paraguay Project
Source: Table 4-7.
required for each cash crop in each month for which hired labor would be needed. In other circumstances, such a mechanistic allocation would be inappropriate. In many areas, certain crop operations are done by hired labor and not by family labor, even if family labor is available. Thus, in Southeast Asia, transplanting rice is done in many areas entirely by hired labor; the only family labor engaged is that of the farmer himself, who supervises the work. Both the amount of hired labor and its allocation among crops should be closely related to the expected cultural practices of people in the project area.
Farm production
Having determined the use of land and labor resources for the pattern farm, the analyst next assesses the projected farm production. The investment analysis of crop and pasture and livestock production is discussed in this subsection, and issues of valuation (both of farm production and incremental residual value on the farm) are addressed.
Source: Calculated from table 4-7.a. Hired labor is allocated to cash crops-cotton, soybeans, and sunflower-in proportion to their total labor requirements in that month. In November of year 4, for example, 55 work days are required for cash crops, of which 3 work days are to be hired. Eighteen work days are required for cotton. To determine the hired labor for cotton, the proportion of total labor required for cash crops that is to be applied to cotton is multiplied by the total hired labor requirement for the month; this gives the hired labor to be applied to cotton, or 1 work day {[18 _ (18 + 29 + 8)] x 3 = i}.b. Does not equal the total in table 4-7 because of rounding.CROPS AND PASTURE.
For crops and pasture, the yield and carrying capacity are tabulated as illustrated in table 4-9. In the table, yields are shown only for crops and pasture actually in the cropping pattern in the year reported. Thus, no without-project yield is reported for soybeans, whereas for sunflower, which will be planted following cotton in year 2, yield is reported only beginning in year 3 since the first crop is not harvested until that time.
Multiplying the production per hectare by the number of hectares of each of the crops and of pasture in the land use pattern shown in table 4-4, we obtain the crop and pasture production illustrated in table 4-10. Again, since sunflower is first planted in year 2 but first harvested in year 3, the production from the first planting of 1.8 tons is shown in year 3. Similarly, sunflower planted each year produces in the following year.
Because all the feed for the livestock to be produced on the pattern farm in the Paraguay project is assumed to come from pasture, no deduction is made in table 4-10 for crops to be used for feed. Should the production pattern of a model farm call for the use of crops and crop by-products for feed, table 4-10 would then be adjusted to show that use. Total production of crops would be shown and expanded to include crop by-products if these were to be fed or if they have a sale value. From this total would be deducted the feed consumption taken from an estimate of feed requirement and production such as that illustrated in table 4-30. The result would be the net production available for sale or household consumption.
LIVESTOCK.
Herd composition, purchases, and sales are given in table 4-11. Projecting the herd (or flock) composition, purchases, and sales in a farm investment analysis that involves livestock introduces a computational process that can become quite complex.
Source: Same as table 4-4.
Source: Calculated from tables 4-4 and 4-9.a. Sunflower is harvested in the year following planting. Thus, the production in year 2 is zero, and in year 3 it is 1.8 tons, which is determined by multiplying the 2.0 hectares planted in year 2 by the yield of 0.9 tons per hectare harvested in year 3 (2.0 x 0.9 = 1.8).b. If there were a substantial livestock production activity on the farm that used crops for feed, the amounts would be estimated in a feed requirement and production table such as table 4-30 and tabulated here. The table would be expanded to include crop by-products used for feed. As indicated, the use of crops for feed would be deducted from the total production, and the result would be the net production available for sale or household consumption.
Herd projections are done to forecast use of future facilities, pasture, or feed by applying technical coefficients, such as those shown at the bottom of table 4-11, to trace the changes in the size and composition of the herd.
For poultry, stall feeding, or feedlot projections, it is usually assumed that enough young animals can be purchased in a given year to bring the numbers up to the level of feed availability or of proposed production facilities. Brown (1979, pp. 76-85) gives a methodology for broiler and egg production. For pigs, a projection is made applying the technical coefficients; since the gestation period of swine is short, the projection is simplified and uncomplicated.
Table 4-11. Herd Composition, Purchases, and Sales, 100 20-Hectare Mixed Farms, Paraguay Project (head)
Source: Same as table 4-4 (annex 1, table 18). See computations in table 4-27.a. In animal units. The carrying capacity at the beginning of the year is determined by multiplying the animal units per farm at the end of the previous year (given in table 4-10) by the 100 farms in the model. The carrying capacity per hectare at the end of the year is determined by dividing the animal units per farm by the 10.5 hectares of pasture on each farm; it is thus a weighted average of natural and improved pasture.b. Herd productivity is the sum of the off-take rate and the herd growth rate. Only the values for a stable herd are given.c. Represents a weighted average between the calving rate of breeding cows in the existing herd, which is 70 percent, and that of purchased in-calf heifers, which is nearly 100 percent.d. Note that in this project a minimum of one bull per farm is assumed, or a minimum of 100 bulls on 100 farms. Normally the number of bulls per 100 breeding females would be three or four for all years.The projection can become computationally quite complicated, however, for larger animals fed mainly pasture, such as sheep or cattle used for dairy or beef production. The project analyst often relies on the livestock technician for these projections and simply incorporates them into his farm investment analysis. But livestock technicians themselves may be unfamiliar with the details of how to make these computations-and especially with how to make them so they conform to the accounting convention adopted here for the farm investment analysis. For this reason, the computation for the herd projection in the Paraguay example is discussed in considerable detail in the appendix to this chapter (where definitions of the specialized livestock terms may also be found). This methodology can be adapted, with only minor variations, to projections for dairy animals. As in the treatment here, the details of the projection need not form part of the main body of most project reports; only a summary need be given (as in table 4-11 from the Paraguay example), with the details laid out in an annex or in the project file.
Projecting the herd composition on a small farm when larger animals are to be produced introduces the difficult problems of divisibility. As noted, herd projections for animals that are mainly grazed on pasture are based on the estimated feed availability. Technical coefficients, such as mortality and calving rates, are often directly influenced by the amount of available feed, but in cattle production changes do not happen immediately. For example, an increase in feed availability in one season will improve the calving rate and decrease calf losses only during the next season.
The projected coefficients are applied to the herd at the beginning of the project. The result begins to appear in the herd composition at the beginning of year 2. Often, the projected coefficients indicate that the herd's composition and its overall size will not change fast enough to utilize the increased feed available. As a solution, in-calf heifers can be purchased to increase the reproductive component of the herd, or feeder steers can be purchased for fattening until the herd can utilize the forage resources.
This use of technical coefficients raises few problems of interpretation for larger farms or ranches with herds of 100 or more animals. For small farms, however, the technical coefficients lead to many "fractional animals." In the Paraguay example, for instance, at full development from years 7 through 20 the farm is expected to have eight breeding cows. The adult mortality is expected to be 2 percent, so do we report that 0.16 cows die each year (8 x 0.02 = 0.16)? Such nonsensical results have led project analysts to seek means to overcome the divisibility problem. Some have simply ignored technical coefficients, such as mortality, that result in very small, fractional animal figures. This omission, however, considerably distorts the pattern farm investment analysis. To avoid such distortion, other analysts have devised systems that carry fractional animals in the computation until the fractions add up to a whole animal, which is then reported. In the Paraguay example, for instance, the calving rate at full development is 75 percent. Thus, six calves are born, of which half may be expected to be female. Calf mortality is 5 percent, and this gives a figure of 2.8 to be carried over to the next year as heifers 1-2 years old [8 x 0.75 - 2 - (8 x 0.75 - 2 x 0.05) = 2.8]. Of these, 2 percent are reported and 0.8 is carried over to the following year, which then will show a figure of 3.6 (0.8 + 2.8 = 3.6). Of these, 3 are reported and 0.6 is carried forward, and so forth.
Mortality may sometimes be treated by incorporating a more formal probability assumption. Such systems become quite complex and in the end do not satisfactorily project an individual small herd. Another approach that is increasingly used-and that is adopted and recommended here-is to do the herd projection for a number of farms, say 100, that will contain or eliminate the divisibility problem. Purchases and sales are then valued, and only the values are entered in the farm investment analysis for a single pattern farm. In effect, this says that on the average a farm will have a certain level of purchases and sales. This, too, is not a fully satisfactory convention. Its results do not state, for example, how many animals are actually on the farm at any given time. It does have the virtue, however, of being simpler than other systems-even if it is still quite complex-and of generating somewhat less distortion.
Project designers may want to introduce an insurance scheme to protect project participants from, say, the loss of a bull. Then, in effect, the values in the farm investment analysis for a single farm for the purchase of bulls include an insurance premium that insures that the farmer will be reimbursed in the event of the death of a bull. Such insurance schemes are found in developing countries, but they give rise to possibilities of abuse and to difficult administrative problems, and they often are not very effective.
In the Paraguay example summarized in table 4-11, the herd composition is given for each major class of animals without the project and for each project year. (The table is drawn from the worksheet reproduced in table 4-27.) Note that the analyst assumed that each farm would have a bull, so that the number of bulls remains 100, many more than would be needed if the analysis were, indeed, for a single herd rather than for 100 small farms. Note, also, the purchase of draft animals at the end of the first year. It is assumed that each farm will purchase two work oxen. The total animal units for the herd are shown, and for convenience this figure is compared with the carrying capacity. As noted in the appendix to this chapter, the number of breeding cows has been rounded to an even multiple of the number of farms in the model so that each farm has five breeding cows without the project and increases its herd to eight breeding cows at full development. As a result, the total of animal units does not very closely approximate the carrying capacity. Since estimates of carrying capacity are quite approximate, however, overstocking of up to 10 percent would probably be acceptable.
Purchases of each class of animal are treated next in the investment analysis; these will form the basis for the investment and operating expenditure for the livestock aspect of the 20-hectare mixed farm. The sales give the basis for the inflow for the farm. The herd productivity, a measure of the efficiency of the herd, is also given; it relates the number of head sold plus the increase in herd size to the number of head carried at the beginning of the year. Only the figures for the stable herd without the project and at full development are given. The dynamics of herd growth tend to distort the measure during the period when the herd is increasing in size. (The details of the computation are given in table 4-28.) Finally, the technical coefficients for the herd are given. These are crucial parameters of the herd growth and are indicators of management effectiveness, animal health care, and feed availability.
When feed concentrates are important in the farm production pattern, it may be desirable to project the feed requirements the livestock activity will involve. (An illustrative example is included in tables 4-29 and 4-30 in the appendix to this chapter in connection with the discussion of herd projections.)
In some instances it may be desirable to report yield per animal if the valuation system is based, say, on kilograms. In the Paraguay example the prices are based on individual animals without regard to weight, so yield per animal is not needed and is therefore not illustrated.
VALUATION
To begin the valuation of the farm production, the farm-gate prices for items entering the farm investment analysis are listed as shown in table 4-12. (The symbol for Paraguayan guaranis is 0.) If a farm-gate price is used in only one table of the investment analysis, it may not be included in the farm-gate price table but may appear in the appropriate table. (Such is the case, for example, of the prices for land improvement, which are included in table 4-15, devoted to investment, and not in the table of farm-gate prices.) Some farm-gate prices were collected and projected by the project analyst on the basis of field observation. Other prices were collected in the field but forecast using the projections of the World Bank. Prices and their derivation were discussed in more detail in chapter 3.
The value of production for the farm is given in table 4-13. For crops, values are determined by multiplying the production in table 4-10 by the price per ton in table 4-12. For livestock, the value is obtained by multi-plying sales from table 4-11 by the price per animal given in table 4-12. The product is then divided by 100 to give the value for a single farm, in line with the convention recommended to avoid the divisibility problem.
INCREMENTAL RESIDUAL VALUE
In the last year of the farm investment analysis, the incremental residual (or terminal) value on the farm is
Table 4-12. Farm-Gate Prices, 20-Hectare Mixed Farm, Paraguay Project
(thousands of 0)
