Procedures for Budget Preparation

A budget can be formulated wherever a market transaction occurs. This approach provides a maximum disaggregation of the marketing system and allows detailed comparisons between costs and margins for each stage of the marketing process. But if the number of budgets is large and at least some of the margins are small, a simplified analysis can consider fewer budgets. The PAM format described here uses only three levels of postfarm budgets: a farm-to-processing activity that includes transportation, bulking, and storage of the farm product; a processing activity that includes processing costs and possibly storage of both the processed and farm products; and a processing-to-market activity that includes transportation and delivery costs to the wholesale point.

Postfarm budget data utilize numeraires different from those of the farm production activity. Often, the numeraires differ among marketing activities. A typical rice system, for example, would have farm-to-processor costs denominated per metric ton of paddy rice, whereas processing and processor-to-market costs would be measured per metric ton of milled rice. Conversion ratios-number of kilograms of milled rice per kilogram of paddy rice (that is, milling outturn ratios)-are necessary to convert each marketing budget to a common numeraire before the total costs of the rice marketing system can be calculated. The conversion ratios can affect substantially the importance of particular activities in total system costs, and become prime candidates for sensitivity analyses.

Farm-to-Processor and Processor-to-Market Budgets

The postfarm activities other than processing are involved primarily in transportation and storage. The technologies of these activities are usually easy to describe, and relatively few inputs are involved. Labor needs include unskilled manual labor (handlers) and skilled labor (drivers, managers of warehouses, and merchants). Fixed input requirements are limited to warehouses, trucks, and machinery for loading and unloading. Intermediate input costs include working capital (to represent the opportunity cost of storage), fuel, maintenance and repairs on transportation equipment, and sacks or other handling materials. A final cost element, losses for commodities while in storage, can vary widely, depending on the commodity characteristics and storage technology. Unless the storage agent keeps records of warehouse throughput, estimates of losses can only be approximated.

The costs of transportation activities depend importantly on the analytical treatment of capacity utilization and transport distance. Because fixed costs are spread over annual throughput, some estimates of annual activity (number of trips per year, average number of tons stored per year) are necessary for the calculation of fixed costs per unit of product. Distance is determined by the designation of location in the representative budget. But knowledge of the rate structure of transport costs is useful when the PAM results are generalized to locations other than those specified in the representative budget. Transport costs generally vary less than proportionally with distance, reflecting economies of size and one-time costs of loading.

Time is also a critical parameter, particularly for the estimation of storage costs. The principal components of storage costs-losses and the opportunity cost of financial capital invested in the stored commodity-are directly related to the duration of storage. If the commodity is stored for a one-month period, the warehouse operator will seek to recover the purchase price of the commodity plus a premium to cover losses during the month plus one month's interest equivalent to the rate of return to investment. The interest payment represents forgone earnings if the commodity was purchased with the storage agent's own financial capital; if the money was borrowed, at least part of the rate of return payment represents a financial obligation to pay interest. In both instances, payment for the return to storage is part of the capital cost.

Budgets that include storage costs must therefore match costs with the marketing margin in a temporal sense. This task usually requires a detailed understanding of market behavior. If available data describe annual average margins, for example, the analyst might be tempted to estimate costs of storage on a six-month basis. But, as Figure 10.1 shows, this calculation could be incorrect. The diagram traces commodity price behavior over time for rural producing areas, urban consuming areas, and imports. The world price shows no seasonality, being fixed at P_cif. In a global perspective, production of most commodities occurs at all times of the year and world prices show little seasonal behavior. Domestic postfarm costs for transportation and handling are represented by the difference between urban and rural market prices. Costs of storage are reflected by the rate of increase in the rural market price.

The key insight from the diagram is that domestic storage is not undertaken during all times of the year. In rural areas, prices decline from October through December (the harvest period) and the commodity is put into storage. Over the January through May period, stocks are depleted and the commodity is taken out of storage to satisfy rural demands. But during the months of May through October, domestic stocks are eliminated and rural market prices no longer increase. Rural prices now reflect the cost of imports plus the cost of transport from the urban import point to the rural market. Storage is thus a six-month activity rather than a yearlong one, and the representative budget should be measured accordingly. An average storage period would be three months, January through March, and the return to storage would be represented by the difference between the March and January prices.

The recognition of intraseasonal price variation imposes a discipline on the output pricing for all the activities of the commodity system. In the current example, the farm-gate price is the January price (adjusted for transport costs between the farm gate and the rural market). The January price represents the purchase price of the commodity for the farmtoprocessor activity. This item appears in the commodity-in-process category of the latter budget. The March price is the sales price for the farm-to-processor activity and thus appears in the output category of this budget.

If the costs of transport and handling are small or if the postfarm activities are unimportant to the policy issues for analysis, development of detailed activity budgets will be unnecessary. In this circumstance, the analyst can assume that costs are equal to observed marketing margins or can develop implicit margins based on price quotations from transport companies and rough estimates of storage costs. These private

costs are allocated among labor, capital, land, and tradable-input categories. Each cost item is then revalued in social prices. Because these activities reflect services that are not available on world markets, the social price of each activity will necessarily equal the sum of the social costs.

Processing Budgets

Processing costs are usually the most prominent component of post-farm costs. Detailed price and quantity data for inputs and outputs of processing thus will generally be needed. The analyst can visit representative firms to find a comprehensive list of inputs. Labor, fuel and lubricants, electricity, repair services, and packaging materials are variable inputs common to most agricultural processing activities; chemicals and additives can be prominent for some products as well. For most processing technologies (household processing is the principal exception), fixed costs are a large share of processing costs. Processing machinery, buildings and storage facilities, and equipment to handle and transport both raw materials and the processed commodity often dominate other categories of input costs. Hence, particular attention must be given to careful description of fixed input requirements.

Since the annualized costs of fixed inputs have to be converted into cost per unit of output, detailed knowledge of annual throughput or capacity utilization is needed. Expected annual throughput is the preferred measure by which to determine annual cost, because this value represents the level of demand that motivates the long-run decision of the firm to operate. The expected value of throughput may differ from its observed value in a particular year: firms might be observed in a start-up stage, where construction of facilities is based on expected future increases in demand; infrastructural constraints, such as availability and reliability of power supply, storage capacity, or availability of transportation equipment, sometimes cause operating levels of firms to be temporarily low; and business cycles cause aggregate demand to fluctuate from year to year, particularly if the product has an income-elastic demand. Visits to a large number of firms and time-series data can be helpful in the formation of an estimate of expected annual throughput. When capacity utilization rates vary widely and chronically, multiple processing budgets can be constructed to represent alternative scenarios.

The valuation of outputs and the selection of outturn ratios are often interdependent, because the quality of output and price are inversely related to the outturn ratios. With rice or wheat, for example, process

ing yields can vary substantially within a single processing technology. Outturn ratios can be higher for parboiled brown rice than for white rice and higher for whole wheat than for white flour. But the prices of the alternative outputs differ as well. If they did not, processors would increase profits by producing the product with the highest outturn ratio. In addition, outturn ratios influence the quantity and quality of by-products, such as the bran collected from cereal processing. These by-products often are of significant value relative to processing costs. Rice, for example, is frequently milled for a nominal or zero charge in exchange for the bran; cotton ginning can be done similarly, with cotton-seeds the relevant by-product.