FOR POLICY ANALYSIS MATRIX (PAM) IN INDONESIAN AGRICULTURE. Carl H. Gotsch Scott R. Pearson Sjaiful Bhari

Transcription

1 COMPUTER TUTORIAL FOR POLICY ANALYSIS MATRIX (PAM) IN INDONESIAN AGRICULTURE Carl H. Gotsch Scott R. Pearson Sjaiful Bhari June 2003

2 TABLE OF CONTENTS Introduction: Organization and Use of the Computer Tutorial Uses of the Computer Tutorial Organization of Tutorial Chapter 3. Private Profitability Analysis: The PAM s First Row Single Commodity Private Price Budgets Sensitivity Analysis Appendix 3.1. Estimating Capital Recovery Costs Estimating Capital Recovery Costs Modifications to the Spreadsheet Sensitivity Analysis Chapter 4. Farm Budgets at Social Prices: The PAM s Second Row Adding Social Prices Constructing Farm Budgets at Social Prices Appendix 4.1. Determining Export and Import Parity Prices Preparing an Import Parity Price Table Determining the Export Parity Price of Corn Linking Tables in the Spreadsheet Sensitivity Analysis Summary Appendix 4.2. Nontradable Good Prices Analysis of Non-Tradable Services Chapter 5: Policy and Market Failures Single Commodity PAM for Rice Farming Systems PAM Multi-Period PAM Sensitivity Analysis Appendix 5.1. Computing Summary Ratios The Ratio Table Chapter 6. Benefit-Cost Analysis Introduction With Project at Private and Social Prices Without Project Budgets at Private and Social Prices With and Without Project PAMs Investment Costs Computing a Discounted Benefit-Cost Ratio The Internal Rate of Return Sensitivity Analysis Chapter 7: Market Failures and Environmental Externalities Unsustainable Versus Sustainable Production Practices

3 Sensitivity Analysis

4 Introduction: Organization and Use of the Computer Tutorial Uses of the Computer Tutorial. Experience with PAM workshops has demonstrated that hands-on tutorials play an important role in developing PAM skills. Actually carrying out the operations helps to clarify concepts that may remain vague after reading or hearing them in lectures. Step-by-step repetition at a leisurely pace reinforces the central points in a way that no other medium can. The tutorial exercises in this manual can also serve as templates for actual research activities. Although the size of the problems have been kept small to minimize the typing required, the models have sufficient detail to permit their use as examples for real research activities. Having a quantitative outline reduces substantially the amount of time required to set up a research project. (Each successive project can be saved in a library for use in subsequent exercises.) Organization of Tutorial The tutorial has been organized so that it is integrated closely with the book Policy Analysis Matrix (PAM) in Indonesian Agriculture. Chapter 3 of the computer tutorial, for example, is based on Chapter 3 of the book; Chapter 4 is based on Chapter 4, etc. However, the tutorial chapters are standalone chapters. In addition to the computations, they provide a brief discussion of the theory behind the calculations as well as some commentary on the interpretation of the quantitative results. Footnotes and comments point to more detailed explanations in the new PAM book and the original book by Monke and Pearson. The manual is available in both HTML and pdf formats. Typically, the first step after becoming familiar with PAM concepts, either by reading or through lectures, is to download the computer manual in the pdf (Adobe Acrobat Reader) format and print out a hard copy. In the absence of a dual monitor or two-computer setup, it is difficult to work through the lessons in the tutorial by going back and forth between the on-line version and an on-line Excel document in which the calculations are being done. Experience suggests that a better approach is to use the hard copy of the tutorial as a guide to carrying out the Excel computations. 6

5 Chapter 3. Private Profitability Analysis: The PAM s First Row. This chapter covers the material in Chapter 3 of the new PAM book. Tables in the computer tutorial are numbered accordingly. For the most part, each chapter in the tutorial has been entered into a separate worksheet whose name shows up as tabs along the status bar at the bottom of the Excel page. This procedure helps keep track of the various components of the PAM computations and facilitates sensitivity analysis. The tabs also provide a visual reminder of the logic of the calculation sequence. The entire workbook is called PAMTutorial. Single Commodity Private Price Budgets The commodity budget used in this initial example is based on rice. Production and price data are used to calculate the returns to high yielding paddy in Indonesia s wet season. The area in which the rice is grown is assumed to have good water control. The physical components of the budget are laid out in the Input-Output table shown in Table Table 3.1. Physical Input-Output (Good Water Control) I-O Quantities HY Paddy Tradables Fertilizer (kg/ha) Urea 240 SP KCl 20 ZA 150 Chemicals Liquid pesticide (liters/ha) 3 Granulated pesticide (kgs/ha) 15 Seed (kg/ha) 35 Fuel (liters/ha) 65 Factors Labor (hr/ha) Seedbed Prep 100 Crop Care 600 Harvesting 200 Threshing 150 Capital Working Capital (Rp/ha) 2,000,000 Tractor Services (hr/ha) 20 Thresher (hr/ha) 35 Land (ha) 1 Output (kg/ha) 6,000 A highly disaggregated input-output table is useful in two ways: (1) it shows that care was taken in doing the field work, and (2) it facilitates sensitivity analyses that will be done in the final stages of the study. Subsequent tables provide data on private prices and compute the commodity budget at private prices. 7

6 To begin the computer tutorial, create a new workbook (file) and save it as PAMTutorial. Rename (right mouse button) the first worksheet tab to C3-P-Budget (no quotes). (This stands for Chapter 3, Private Budget.) Type in the labels and data for Table 3.1. at the top of the worksheet. (The label starts in Cell A1.) After constructing the I-O table, select the table and copy it below itself. Label the new table as Table 3.2. Private Prices. Make the necessary changes in the labels and units. Type over the cell values to enter data for prices instead of physical input-output coefficients. Copy Table 3.2 below itself, and identify the new table as Table 3.3. Private Prices Budget. Make the necessary changes in the labels (units). Table 3.2. Private Prices P-Prices Quantities HY Paddy Tradables Fertilizer (Rp/kg) Urea 1,100 SP-36 1,400 KCl 1,600 ZA 1,000 Chemicals Liquid insecticide (liters/ha) 30,000 Granulated insecticide (kg/ha) 8,000 Seed (Rp/kg) 2,500 Fuel (Rp/liter 1,500 Factors Labor (Rp/hr) Seedbed Prep 1,600 Crop Care 1,600 Harvesting 1,600 Threshing 1,600 Capital Working Capital (%) 5% Tractor Services (Rp/hr) 12,500 Thresher (Rp/hr) 1,500 Land (Rp/ha) 1,500,000 Output (Rp/kg) 1,205 Compute the cells in Table 3.3, the Private Prices Budget table, by multiplying the elements of the Prices table times the elements of the I-O table. To minimize typing, compute the first cell, either by typing in the formula or by creating it with the aid of the mouse, then drag the formula down across the other rows. (The first element might be something like =C4*C31; this becomes =C5*C32 in the second row, etc.) Because the tables have the same number of rows, the value for all elements of High Yield paddy up to and including Total Revenue, can be obtained by copying. (If you are unclear about how to copy in Excel, look up the topic under Excel Help.) The Private Prices Budget contains three additional rows, Total Costs, Profits, and Net Profits. To 8

7 compute Total Costs excluding Land, write a formula that sums the relevant cost elements, i.e., Urea through Thresher services. (e.g., =SUM(C56:75). To compute Profits (excluding Land), subtract Total Costs from Total Revenues. To compute Net Profits (including Land), subtract Land from Profits (excluding Land). Table 3.3. Private Prices Budget P-Budget Quantities HY Paddy Tradables Fertilizer (Rp/ha) Urea 264,000 SP ,000 KCl 32,000 ZA 150,000 Chemicals Liquid insecticide (Rp/ha) 75,000 Granulated insecticide (Rp/ha) 120,000 Seed (Rp/ha) 87,500 Fuel (Rp/ha) 97,500 Factors Labor (Rp/ha) Seedbed Prep 160,000 Crop Care 960,000 Harvesting 320,000 Threshing 240,000 Capital Working Capital (Rp/ha) 100,000 Tractor Services (Rp/ha) 250,000 Thresher (Rp/ha) 52,500 Land (Rp/ha) 1,500,000 Output Total Revenue (Rp/ha) 7,230,000 Total Cost (excluding land) (Rp/ha) 3,048,500 Profit (excluding land) (Rp/ha) 4,181,500 Net Profit (including land) (Rp/ha) 2,681,500 The distinction between profits that include or exclude returns to land is important. Whereas rental values can be observed and included in a private budget, the same is not true for social budgets. Land is unique because it is the only truly fixed factor in agriculture. In suburban locations, agriculture might not be the only use for land, and prices and rental values will be influenced by off-farm opportunities. But in most areas, the only alternative to agricultural use is no use at all (if forestry is included as an agricultural activity). In these cases, land acts as a residual claimant on the profits from farming. 1 Save the three-table file under the heading: PAMTutorial. With these calculations, we are now in a position to complete row 1 of the PAM. The figures, drawn directly from the numbers in Table 3.3, are as follows: 1 Further explanation can be found on pp in Monke-Pearson. 9

8 Table 3.4. PAM Rice (Good Water Control) Revenue Costs Profits Tradables Labor Capital Private 7,230, ,000 1,680, ,500 4,181,500 Social Divergence Sensitivity Analysis It is often helpful in farming systems analyses to do a sensitivity analysis of the most important parameters. Note that changes in the costs of inputs have much smaller effects on profits than changes in the prices of outputs. This is because each input makes up only a fraction of the cost, whereas the output price applies to the whole of revenues. Likewise, changes in productivity also apply to the whole of revenues. What happens if: 1) Fertilizer prices increase by 50% from their present values? 2) Fertilizer prices remain at their original values, but all labor costs double? The same percentage changes have a more pronounced effect when they are applied to outputs and revenue: 3) Paddy prices increase by 25%? 4) Yields increase by 25%. The results of the sensitivity analysis strongly suggest that the highest research priority is to obtain the best possible data on output prices and yields. The next highest priority would be the largest item in costs, say, labor. Only after the larger cost items have been determined should minor costs be investigated. The what-if feature of spreadsheets makes it possible to determine quickly exactly how much impact a particular price has on the overall results. Efforts to improve the database can be organized accordingly. Sensitivity analysis is normally done by changing one parameter at a time and observing what happens to other policy variables. However, it is often the case that different input and output polices are administered by different government agencies each of which has its own bureaucratic and political agenda. The total effect of all price and production policies determines farmer incentives to grow crops -- the farmer responds to changes in profitability, regardless of the source of these changes. Given the possibility that various policies could amplify or counteract one another, it is essential that different government agencies coordinate their efforts to ensure consistency. 10

9 Appendix 3.1. Estimating Capital Recovery Costs Including the opportunity costs of fixed capital in the PAM analysis is somewhat awkward in a budgeting framework where the focus is on annual variable costs and not on fixed costs. However, over the usual lifetime of policies being analyzed, farmers make decisions about investment items whose costs are fixed. Failure to include annualized fixed input costs in some form would lead to distortions, not only in decisions about durable capital goods, but also in the selection of crops and technologies. As Monke and Pearson note in the text (pp ), one simplified, but incomplete, way to find the annual cost of a fixed input would be to divide its initial cost by the life of the input. This same method can be used to apportion the annual cost between different commodities, i.e., each crop could be debited in proportion to the time the fixed inputs were used in its cultivation. However, this approach ignores the opportunity cost of the capital tied up in the fixed input. The farmer could have banked the money rather than investing in a fixed production asset. The true cost of the capital, therefore, is the annual cost plus the interest on the embodied capital. This fixed input charge is then apportioned to the various commodities serviced by the investment item. Estimating Capital Recovery Costs Estimating capital recovery costs requires several steps. The first is to gather the information on the cost of recovering the capital from an investment. This includes the initial cost of the investment (in private and social terms), an estimate of the useful life of the machinery, its salvage value, and the total number of "horsepower" hours it is expected to provide. The initial cost and useful life of the machine represent the two most important parameters of the capital cost recovery calculations. After its useful life has expired, the machine may still have a salvage value as scrap and a source of parts. The salvage value is received several years in the future and, therefore, must be discounted using data on the private and social interest rates found in the existing Prices tables. It is then deducted from the initial cost to derive today's net cost. Perhaps the most complicated derivation is the recovery ratio, which involves adjusting the interest rate by the life expectancy of the investment. As defined by Monke and Pearson, this is the share of the net cost that must be recovered each year "to repay the cost of the fixed input at the end of its useful life." Once the recovery ratio is determined, the actual monetary sum is calculated. This figure, the annual recovery cost, is then prorated to an hourly basis. In the current chapter, these steps will be followed to incorporate the capital recovery costs of one particular investment: an irrigation pump. In reality, the farmer may own other implements that would require such accounting in the budgeting process such as the tractor or a thresher. Modifications to the Spreadsheet Creating the Capital Recovery Cost Table. Retrieve workbook called PAMTutorial. Insert a worksheet for calculating the Capital Recovery Cost and call it Appendix 3. Create the headings shown in Table Write the assumptions, shown in bold in Table 3.1.1, into the table: Initial Cost, Useful 11

10 Life, Salvage Value, and Total Pump Hours.2 Fill in the private Interest Rate cell by writing a formula that references the interest rate in the first column of the Private Prices table. Follow the same procedure for the Social Prices table. These steps integrate the spreadsheet so that subsequent sensitivity analysis on the interest rate will require the alteration of only one cell. Derive the Present Value of Salvage Value using the formula shown below where i = the interest rate and n = the number of time periods (in this case, years of useful life.) Salvage Value n ( 1+ i) The spreadsheet implementation: Cell address of salvage value/ (1 + cell for interest rate ) ^ cell address for useful life (^ is the character for exponentiation.) Calculate Net Cost as the Initial Cost minus Present Value of Salvage Value. Calculate the Recovery Ratio using the formula: n (1 + i) i n (1 + i) 1 The spreadsheet implementation of this formula: ((1 + i) ^ useful life * i)/ (((1 + i)^ useful life) -1) Calculate the Annual Recovery Cost as the product of the Recovery Ratio and Net Cost. Calculate Recovery per Pump-hour as the Annual Recovery Cost divided by Total Pump Hours. It will be multiplied by the hours if the irrigation pump was part of the Input-Output table. Appendix Table Annual Capital Recovery Costs Water Pump Private Prices Social Prices Initial Cost (Rp/15 hp unit) 1,120, ,000 Useful Life (years) Salvage Value (Rupiahs) 112,000 80,000 2 Total hours is annual machine capacity in hours. Using the capacity of the machine as the denominator seems preferable to the practice of computing percentage of use on the basis of total hours actually used. The latter depends upon the choice of a cropping pattern and thus is a function of the entire cropping system. By using capacity, the denominator becomes exogenous. The assumption that there is no surplus machine capacity in the sector also seems more consistent with the long run concerns of the PAM analysis than using total actual hours. 12

11 Appendix Table Annual Capital Recovery Costs Water Pump Private Prices Social Prices Interest Rate (%/year) 28% 12% Present Value of Salvage (Rp) 9,487 25,758 Net Cost (Rp) 1,110, ,242 Recovery Ratio (%) Annual Recovery Cost (Rp) 339, ,029 Total Pump Hours 7,200 7,200 Recovery per Pump-Hour (Rp/hr) In many developing countries, manufactured equipment is often wholly or partially imported. It then must be treated like any other input. The tradable components must be valued at international prices, which are in turn converted to domestic prices using the equilibrium exchange rate. The nontradable factors used in its domestic manufacture or assembly must be valued at their shadow prices. Modifying the Input-Output, Prices, and Budget Tables. The structure of the Input-Output, Private Prices, Private Budget, Social Prices, and Social Budget tables are each modified in an identical fashion, by inserting a new row in the capital section. Data are then linked from the newly created Capital Recovery table. It is easiest to complete all structural and data changes on each table before progressing to the next. (Consistent with the original design of these tables, the units of measure depend on the table.) Go to the I-O table. Insert a row just above the Thresher row in the Capital section. Call it Irrigation Pump. Enter data for the number of pump hours per hectare. Go to the Private Prices table. Add a line for the water pump under the capital section in the same way that it was added in the Input-Output table. Enter the address of the Recovery rate per Pump-hour in the private prices column in Table Make the same change in the Private Budget table, i.e., insert a row above the Thresher row and call it Irrigation Pump. Copy the formulas that multiply the I-O table times the Prices table from the line above. Repeat the procedure for the Social Prices table and the Social Budget table. Sensitivity Analysis Note that assumptions about the interest rate play a significant role in determining capital recovery costs. What effect does a change in the social interest rate from 12% to 20% have on capital recovery costs? To 5%? 13

12 Chapter 4. Farm Budgets at Social Prices: The PAM s Second Row The preceding commodity budget was based on private prices, those that farmers face in the market place. As the PAM manual notes, in many instances, private prices do not reflect the true scarcity value of a good to the economy. Market failures and policy interventions may drive a wedge between the true opportunity cost, or social price of a good, and the observed market price. Because they are not directly observed, social prices must be estimated from other economic data. The process can be quite elaborate, depending on the extent to which a good is traded. To simplify the initial calculations, this chapter provides social prices for the tradables and nontradables needed to set up the basic PAM discussed in Chapter 2 of Monk and Pearson).The process of calculating those social prices and the sensitivity of those prices to economic policies are discussed in M-P s Chapter 6. Adding Social Prices The first step in adding social prices is to retrieve the workbook saved at the end of the last chapter (PAMTutorial). Insert a new worksheet and rename the tab Chapter 4. Select the worksheet under the Chapter 3 tab and copy it into the newly created Chapter 4 worksheet. This new worksheet will be devoted to computing the rice budget at social prices. (In the interest of clarity, rename Table 3.1. Physical Input- Output Data to Table 4.1. Physical Input-Output Data. As in all PAM analyses at this level, the basic underlying production data are the same for both the first and second rows of the PAM matrix. (Recall that these figures reflect high-yielding paddy grown on land assumed to have good water control.) Table 4.1. Physical Input-Output I-O Quantities HY Paddy Tradables Fertilizer (kg/ha) Urea 240 SP KCl 20 ZA 150 Chemicals Liquid pesticide (liters/ha) 3 Granulated pesticide (kg/ha) 15 Seed (kgs/ha) 35 Fuel (liters/ha) 65 Factors Labor (hr/ha) Seedbed Prep 100 Crop Care 600 Harvesting 200 Threshing 150 Capital Working Capital (Rp/ha) 2,000,000 Tractor Services (hr/ha) 20 Thresher (hr/ha) 35 Land (ha) 1 Output (kg/ha) 6,000 14

13 Rename Table 3.2. Private Prices to Table 4.2. Social Prices. For several reasons, this is the most difficult table to construct in the PAM analysis. Tradable commodities are hard to estimate because, not only does the export or import parity price involve getting international prices that may be difficult to find, but conversion to farmgate prices requires obtaining prices for such items as transportation, marketing, processing, etc. These may also be hard to obtain. Table 4.2. Social Prices S-Prices Quantities HY Paddy Tradables Fertilizer (Rp/kg) Urea 1,100 SP-36 1,400 KCl 1,600 ZA 1,000 Chemicals Liquid insecticide (liters/ha) 40,000 Granulated insecticide (kg/ha) 10,000 Seed (Rp/kg) 2,500 Fuel (Rp/liter 1,500 Factors Labor (Rp/hr) Seedbed Prep 1,600 Crop Care 1,600 Harvesting 1,600 Threshing 1,600 Capital Working Capital (%) 8% Tractor Services (Rp/hr) 12,500 Thresher (Rp/hr) 1,500 Land (Rp/ha) - Output (Rp/kg) 964 Estimating prices for fixed factors such as land, labor, and capital is no easier. There are no social prices for factors that can be observed directly, and the estimation procedure depends heavily on identifying institutional and market conditions that might produce distortions from observed private prices. (E.g., monopolies, monopsonies, and, in the case of land, the lack of markets that incorporate environmental degradation.) The details of computing the figures shown in Table 4.2 are developed in a series of appendices to Chapter 4. Appendix 4.1 deals with the estimation of export and import parity prices. Examples are given for rice imported from Bangkok and corn exported from Padang. Appendix 4.2 takes up the question of how analysts should go about decomposing goods and services that are a combination of traded and non-traded commodities. Note that the Land price is 0. In the absence of clearly specified cropping alternatives, imputing social opportunity costs to fixed factors within a single commodity budgeting framework is arbitrary. Consequently, the land price, and thus cost, equals 0 and all returns to land are included in the Profits residual, i.e., Profits and Net Profits are the same. Social profits thus measure the returns to land and 15

14 management when all commodities are priced at their efficiency prices. Constructing Farm Budgets at Social Prices. Rename Table 3.3. Private Prices Budget to Table 4.3. Social Prices Budget. The formulas that multiply the Input-Output table and the Prices table should have been copied automatically and the Social Prices budget should be ready for the computation of the PAM s middle row. Table 4.3. Social Prices Budget S-Budget Quantities HY Paddy Tradables Fertilizer (Rp/ha) Urea 264,000 SP ,000 KCl 32,000 ZA 150,000 Chemicals - Liquid insecticides (Rp/ha) 100,000 Granulated insecticides (Rp/ha) 150,000 Seed (Rp/ha) 87,500 Fuel (Rp/ha) 97,500 Factors Labor (Rp/ha) - Seedbed Prep 160,000 Crop Care 960,000 Harvesting 320,000 Threshing 240,000 Capital - Working Capital (Rp/ha) 160,000 Tractor Services (Rp/ha) 250,000 Thresher (Rp/ha) 52,500 Land (Rp/ha) - Output Total Revenue (Rp/ha) 5,784,000 Total Cost (excluding land) (Rp/ha) 3,163,500 Profit (excluding land) (Rp/ha) 2,620,500 Net Profit (including land) (Rp/ha) 16

15 Appendix 4.1. Determining Export and Import Parity Prices The social price of a tradable output or input at the wholesale market nearest to the farm gate equals the international or border price adjusted for exchange rates and domestic transportation, processing, and marketing costs. The resulting farm gate prices are called import and export parity prices or sometimes border price equivalents. The general concepts for developing export and import parity prices are shown in Table Appendix Table Determining Import and Export Parity Prices Step Import Parity Prices Export Parity Prices STEP DATA PROCESS DATA PROCESS International Prices F.o.b. price at point of export Given C.i.f. price at point of import Given Freight to point of import Given Freight to point of Given export Insurance Given Insurance Given C.i.f. at point of import F.o.b + Freight + Insurance. F.o.b. at point of export C.i.f - Freight - Insurance Currency Conversions Foreign exchange rate Given Foreign exchange rate Given Foreign exchange premium Given Foreign exchange pre mium Given Equilibrium exchange rate ER * (1 + ERP) Equilibrium exchange rate ER * (1 + ERP) C.i.f. in domestic currency EER * C.i.f at point F.o.b in domestic cur EER * F.o.b. at of import rency point of export Weight Conversions Weight conversion factor Given Weight conv. factor Given C.i.f. in dom. curr. and C.i.f. in dom. curr. / F.o.b. in dom. curr. and F.o.b. in dom. curr. weight Weight conversion factor weight / Weight con version factor Distribution between port & wholesale mar ket. Distribution between wholesale & farm gate Result Local transport & mkting costs to wholesale mkt, in social prices Value before processing Given C.i.f. in dom. curr. and weight + dis trib. costs. Local transport & Given mkting costs to wholesale mkt, in social prices Value before F.o.b. in dom. curr. processing and weight - distrib. costs. Processing conv. factor Given Processing conv. factor Given Import parity value at whole sale market Value before pro cessing * conver sion factor Export parity value at wholesale market Value before pro cessing * conver sion factor Transport, marketing, & storage costs to farm, in social prices Import parity value at farm gate Given Import parity value at wholesale mar ket +/- distr. costs to farm gate (Deduct if output; Add if input) Transport, marketing, & Given storage costs to farm, in social prices Export parity value at farm gate Export parity value at whole sale market - distr. costs to farm gate Preparing an Import Parity Price Table. Create a table to the right of Table 4.1 on the Chapter 4 worksheet and label it Appendix Table The data to be used, and the required intermediate calculations, are shown in the table below. 17

16 Formulas for Determining the Import Parity Price of Rice The price of imported rice from Bangkok will serve as a starting point for deriving the import parity price for paddy in Indonesia. 1. To calculate the C.i.f. price for rice at the Indonesian port: C.i.f. at point of import = F.o.b. price at point of export + freight costs + insurance costs 2. Calculate the Equilibrium exchange rate: Exchange rate * (1 + exchange rate premium) 3. Convert international prices in dollars ($) to local currency (Rupiah). C.i.f. in domestic currency = C.i.f. at point of import * equilibrium exchange rate 4. Convert the unit of measure from tons, the usual international price unit, to kilograms. C.i.f. (Rp/kg) = C.i.f. (Rp/ton)/ Add distribution costs between the port and the wholesale market to the weight-adjusted c.i.f.price 6. Multiply "before processing" cost by the processing conversion factor. Adjust for the cost of milling net of the value of rice bran. 7. Adjust for the cost of distributing the commodity from the wholesale market to the farm gate. (Because paddy is an output, the costs of distribution between the wholesale market and farm gate are deducted from the import parity value at the wholesale market.) Table Adjustment of International Prices to Farmgate Level F.o.b. Thailand ($/ton) Freight & Insurance ($/ton) C.i.f. Indonesia ($/ton) Exchange rate (Rp/$) 9,000 Exchange rate premium (%) 0% Equilibrium exchange rate (Rp/$) 9,000 C.i.f. Indonesia in domestic currency (Rp/ton) 1,530,000 Weight conversion factor (kg/ton) 1000 C.i.f. Indonesia in dom. curr. (Rp/kg) Transportation and handling costs to wholesale market (Rp/kg) 133 Value before processing (Rp/kg) Processing conversion factor (%) 0.64 Cost of rice milling net of the value of rice bran 50 Import parity value (Rp/kg) Distribution costs to farm (Rp/kg) 50 18

17 Table Adjustment of International Prices to Farmgate Level Import parity value at farm gate (Rp/kg) Determining the Export Parity Price of Corn The social export parity price is the border price of an exportable good adjusted for transport and handling costs and revalued by the EER. The calculations for the export parity price resemble those for the import parity price, but generally work in the opposite direction. Follow the steps outlined in Table to calculate Table 4.1.3, the social price of corn in Padang, assuming that corn is an exportable good. Create a new table under Appendix Table and label it Table Social Export Parity Price of Corn. It can be created most easily by copying Table and changing the labels and computations where necessary. Data and Assumptions for Appendix Table C.i.f. U.S. Gulf price for no. 2 yellow corn = $115/ton Costs of insurance and freight between the U.S. and Jakarta = $17.50/ton Official exchange rate: $1 = Rp 1644 Foreign exchange premium = 10% Transportation costs from port of Jakarta to wholesale market = Rp 7/kg. Handling costs from port to wholesale market = Rp 8/kg. Conversion of weights: 1000 kilograms = 1 ton Farm to wholesale distribution costs= Rp 10/kg A conversion factor is not necessary for processing corn because the commodity is sold on international markets in an unprocessed form. Deriving the Export Parity Price for Corn in Padang. Derive the intermediate values based on the assumptions given above and the steps described in Table Several of the cell formulas must be modified to account for the difference between an imported output and an exported output. Appendix Table Calculation of the Social Export Parity Price of Corn C.i.f. ($/ton) 115 Freight & Insurance ($/ton) 17.5 C.i.f. Indonesia ($/ton) Exchange rate (Rp/$) 1,644 Exchange rate premium (%) 10% Equilibrium exchange rate (Rp/$) C.i.f. in domestic currency (Rp/ton) 239,613 Weight conversion factor (kg/ton) 1000 C.i.f. in dom. curr. (Rp/kg) Transportation costs (Rp/kg) 5 Marketing costs (Rp/kg) 7 Value before processing (Rp/kg) Processing conversion factor (%) 1 19

18 Appendix Table Calculation of the Social Export Parity Price of Corn Import parity value (Rp/kg) Distribution costs to farm (Rp/kg) 10 Import parity value at farm gate (Rp/kg) Linking Tables in the Spreadsheet Link the results of the import parity price calculations for rice directly into the Social Prices table, overwriting the preliminary values entered in Chapter 4. To link the relevant cells, move to the Social Prices Table, delete the existing entry, and click on =. Then select the appropriate (final price) entry from the parity price calculation table. Click OK. Sensitivity Analysis The spreadsheet is now integrated so that sensitivity analysis on international prices and exchange rates can be reflected in the social budgets. How does the social profitability for the paddy system change when: 1) The exchange rate premium rises to 30%? 2) The international price of rice rises by 25%? The Import and Export Parity Prices tables assume an exchange rate premium of 0 percent, which means that the exchange rate is not overvalued. Although many developing countries experience overvalued exchange rates, it is often difficult to ascertain the exact amount of the premium. Hence, it is desirable to test the results of different EER assumptions. Determining an international price of rice is also difficult. Care must be taken to ensure that the variety and quality of the rice from the world market is the same as the domestic rice with which it is being compared. Hence the international price is a prime target for sensitivity analysis. Summary This appendix reviewed the process for calculating the social prices of tradable commodities. Data are required for international commodity prices, distribution costs between various stages of the marketing chain, exchange rates, weight conversions and processing factors. The steps involved in transforming these data into parity prices were outlined in Table Distinctions were drawn between the calculations for imported outputs, imported inputs, and exports of both outputs and inputs. For purposes of illustration, sample data were provided for only two of these categories, permitting the construction of tables for paddy (occasionally an imported output) and corn (occasionally an exported output). The import and export parity prices so derived are the social prices faced by farmers. To test the effects of changes in international prices and exchange rates on social profitability, these results were linked to the original Social Prices table constructed in Chapter 4. Although the computations are straightforward, data requirements are often formidable. For example, in identifying the f.o.b. and c.i.f. prices in international markets, it is usually difficult to ensure equivalence in specifications (e.g., quality) between the traded product and the domestically available product. Even small mistakes in establishing the comparability of products can swamp large errors in input-output coefficients. 20

19 Analysis of Non-Tradable Services Appendix 4.2. Nontradable Good Prices Previous chapters have ignored the issue of nontradable services such as machine rentals, transportation, processing, and handling. Collecting data for these nontradable services is one of the most challenging, and often frustrating, exercises in agricultural policy analysis. Because the information is difficult to collect and, once gathered, may only have a marginal impact on the results, analysts frequently resort to broad assumptions about the data, using sensitivity analysis to verify that these assumptions would not do violence to their conclusions. But as Monke and Pearson point out in their Chapter 10 ("Postfarm Budgets and Analysis"), market imperfections or policy divergences in nontradable goods and services should not be treated in a cavalier fashion. In theory, the costs of all nontradable inputs (goods and services) should be decomposed into their tradable inputs and domestic factor cost components. These costs, standardized on units such as hours or measures of volume or weight, then can be substituted into the appropriate cells of the Private and Social prices tables. The current chapter focuses on policy divergences in the tradable component of tractor services and illustrates how these divergences affect private and social budgets. Decomposing Tractor Costs In Chapter 2, tractor services were considered as domestic factors rather than tradable components of the farm budget. As such, the labor associated with tractor services was included in various farming operations; the capital cost associated with these services was included in the capital account. But the rental price of these machines masks a significant tradable component in the form of machine depreciation, fuel, and grease and oil. This exercise decomposes tractor services and assigns the tradable, labor, and capital components to their respective accounts in the private and social budgets. As with the linking of import and export parity prices to the Social Prices table (last chapter), the steps involved in identifying the tradable portion of nontradable services are not conceptually difficult. The process of modifying the spreadsheet, however, requires several calculations and careful adjustments to early tables. The first step is to identify the aspects of tractor services that are tradable. In this example, the tradable component consists of three parts -- the tractor itself, fuel, and grease and oil. The nontradable component consists of two parts -- the use of labor for repairs, maintenance, and management operations, and the working capital required to finance operational expenses. Second, one needs to determine a meaningful unit of tractor use (usually taken as the tractor service hour) and the quantity of each tradable and nontradable component used during that time. For example, how much tractor capital is used up during one hour of tractor use (depreciation)? How much fuel is used? Grease and oil? Labor? Capital? Third, appropriate private and social prices for these quantities must be found. For tradables, social prices are derived in a manner similar to that used to calculate import and export parity prices in the last chapter. In this particular example, tractors, fuel, and grease/oil are imported and thus the calculations follow the steps used to derive import parity prices. In general, private prices for both tradables and nontradables are those observed in the market place. In this example, the private prices for tradable goods have been further decomposed to highlight the effect of government interventions and the 21

20 assumptions concerning depreciation. For nontradables, private prices are observed in the market and thus are taken as given. In both cases, private prices are presented on a per tractor hour basis. Once the quantities and prices for the newly defined components of tractor services have been determined, they must be linked back into the initial budgetary calculations. The Input-Output table must be expanded to incorporate rows for tradable tractor services, tractor labor, and tractor capital. The original figures for hours of tractor services (previously lumped under the capital account of the I-O table) must be moved to the tradables portion of the table. There they serve as the base for calculating the hours of repair and maintenance labor (R & M) used per hectare and working capital required for tractor services per hectare. Next, similar rows are inserted in the Private Prices and Social Prices tables. Private and social price data must be linked from the table containing the various "tractor services" calculations. Once the appropriate lines and formulas are inserted, the Private Budget and Social Budget tables recalculate automatically. Sensitivity analysis can be undertaken to evaluate the importance of decomposing the tradable components of domestic services. Modifications to the Spreadsheet Step 1: Creating the Tractor Inputs Table. Retrieve PAMTutorial. Insert a new worksheet, rename it Nontradables. Create a data Table called Tractor Inputs. Enter the quantity of each of the tradable components (tractors, fuel, grease/oil, and nontradable components (R&M labor, working capital)) used per tractor hour. The R&M coefficient describes the amount of repair, maintenance, and administrative labor used by the vendor to deliver one hour of tractor services. The working capital coefficient describes the amount of working capital used by the vendor to deliver one hour of tractor services. Table Tractor Inputs Items # of Units per Tradable Components Tractor Hour Tractor (hours) 1 Fuel (liters) Grease and oil (liters) Nontradable Components R&M Labor (hours) Working Capital (Rupiahs) 500 Step 2: Creating the Tractor Prices Table Create the Tractor Prices table below the Tractor Inputs table. It resembles the Import Parity table in structure and in arithmetic logic (i.e., the calculations are very similar). The necessary additional labels and formulas to incorporate duties, subsidies, time conversions, and depreciation can then be added. (The labels in the social price portion of the Tractor table are identical to those in the private portion.) The costs of tradable tractor services are constructed along the same lines as the commodity import parity prices. F.o.b. prices obtained from exporting countries are the point of departure if local estimates of the c.i.f. prices in foreign currency are unavailable. The likelihood that the latter can be found is very high since local importers will know what their landed costs are. 22

21 Table Tractor Prices Private Prices Social Prices Tractor Fuel Grease, Oil Total Tractor Fuel Grease, Oil Total (machine) (liters) (liters) (machine) (liters) (liters) F.o.b. ($/Unit) Freight and Insurance ($/Unit) C.i.f. ($/Unit) Official Exchange Rate (Rp/$) Exchange Rate Premium (%) 0% 0% 0% 10% 10% 10% Equilibrium Exchange Rate (Rp/$) 1,644 1,644 1,644 1,808 1,808 1,808 C.i.f. (Rp/Unit) 8,877, ,836 9,765, ,420 Domestic Duties (%) Domestic Subsidies (%) Import Parity at Border (Rp/tractor) 12,428, ,587 9,765, ,420 Expected Life (hours of life/tractor) 25, n.a. n.a. 25, n.a. n.a. Cost per hour transportation (Rp/hr) 497 n.a. n.a. 391 n.a. n.a. Transportation (Rp/unit) Farm Gate Value (Rp/unit) , Total Tractor Service Hour (Rp/hr) Calculate the Import value at border as the c.i.f. price plus the Rupiah value of domestic duties less the Rupiah value of domestic subsidies. Because duties and subsidies are proportions, the formula is: Prorate the total capital cost of the tractor over its useful life: Calculate the Total cost per tractor service hour for each of the tradable components as the product of its Farm gate value and the # of units per tractor service hour (found in the Tractor Inputs table). The social price formulas are identical to those for private prices. So too are the data, with the exception of the Domestic duties and Equilibrium exchange rate. The differences between private and social costs in tractor services originate from the same sources as the divergences in the agricultural sector. For example, many governments subsidize tractor services by permitting tractor dealers to import tractors and spare parts using foreign exchange obtained at an overvalued exchange rate. Gasoline and diesel fuel are also frequently subsidized. (No subsidies are shown in the current example.) Conversely, imports of machines are heavily taxed to encourage and protect domestic production, especially of smaller machines like two-wheeled tractors. Imports of oil are also significantly taxed instead of subsidized. For both machines and production inputs such as fuel, the degree of taxation is somewhat offset by granting foreign exchange allocations at a cost below the equilibrium exchange rate. The offsetting effects of an overvalued currency and import duties can be seen from the private and social price calculations in Table Step 3: Modifying the Input-Output, Prices, and Budget Tables 23

22 The structure of the Input-Output, Private Prices, Private Budget, Social Prices, and Social Budget tables are each modified in an identical fashion, by inserting a new row in the tradables and labor sections. (The capital section already includes tractor services). It is easiest to complete all structural changes on each table before progressing to the next. A sample drawn from the I-O table is shown in Table Consistent with the original design of these tables, the units of measure depend on the table, and in some cases, on the nature of the input. Apply the following steps to each of the five tables cited. Modify the tradables section by inserting a row entitled Tractor Services (hr/ha). Modify the labor section by inserting a row entitled Tractor R&M. In the I-O table only, move the hours from the Tractor services row currently in the capital section to the new tradables row for these same services. These coefficients describe the number of tractors hours used by each commodity. The Tractor Services row in the capital section and the Tractor R&M row in the labor section should be devoid of data. New data will be written in later. Repeat the first 2 steps for the Private Prices, Private Budget, Social Prices, and Social Budget tables. Adjust units according to the nature of the table (e.g., hr/ha for the I-O table, Rp/hr for the Prices tables, and Rp/ha for the Budget tables.) Note that Tractor Services Capital is now on a percent basis in the Prices tables (rather than Rp/hr). Table Physical Input-Output I-O Quantities HY Paddy Tradables Fertilizer (kg/ha) Urea 240 SP KCl 20 ZA 150 Chemicals Liquid pesticide (liters/ha) 3 Granulated pesticide (kg/ha) 15 Seed (kgs/ha) 35 Tractor Services (hr/ha) Fuel (liters/ha) 65 Factors Labor (hr/ha) Seedbed Prep 100 Crop Care 600 Harvesting 200 Tractor R & M Threshing 150 Capital Working Capital (Rp/ha) 2,000,000 Tractor Services (Rp/ha) 20 Thresher (hr/ha) 35 Land (ha) 1 Output (kg/ha) 6,000 24

23 Step 4: Linking the Tractor Tables to the Input-Output and Prices Tables The Tractor Inputs table contains information on the quantities of tradables and nontradables used per tractor service hour. The Tractor Prices table contains the private and social prices of the tradable components (the tractor, fuel, and grease/oil) per tractor service hour. These must be converted to the unit used throughout this analysis -- hectares -- and linked to the appropriate tables. The prices of labor and capital have not changed and are already included in the prices tables. In the labor section of the I-O table, calculate Tractor R&M labor per hectare by multiplying the number of hours of Tractor Services per hectare (in the tradables section of the I-O table) by R&M labor per tractor service hour (on the Tractor Inputs table). In the capital section of the I-O table, calculate Tractor Services capital per hectare by multiplying the hours of Tractor Services per hectare (in the tradables section of the I-O table) by Working Capital per tractor service hour (on the Tractor Inputs table). Format the new rows to be consistent with other data presented in the I-0 table. In the tradables section of the Private Prices table, calculate the private price of Tractor Services per hectare as total Cost per Tractor Service Hour, aggregated across Tractor Services, Fuel, and Grease/Oil (on the Tractor Prices table). In the labor section of the Private Prices table, copy the private price of Tractor R&M per hectare directly down from the wages figure in the preceding row. In the capital section of the Private Prices table, copy the private price of Tractor Capital per hectare directly down from the season interest rate in the preceding row. Format the new rows to be consistent with other data presented in the table. Calculate the social price of tradable Tractor Services, Tractor R&M labor, and Tractor Services capital in the same manner as that used for the private price, referencing data in the Social Prices section of the Tractor Services table. Update the Private and Social Budget tables by copying the formula in adjacent rows to those for the tradable component of tractor services and the labor component of tractor services. The capital component of tractor services should have adjusted automatically. Save the spreadsheet as PAMTutorial Sensitivity Analysis Integrating the results of the decomposition into the existing tables is the most time-consuming part of the exercise. However, such integration is highly desirable because it simplifies the sensitivity analysis of various types of policy proposals. Once the template has been properly implemented, it will be easy to see if significant changes in the international price of tractors has an impact on commodity PAMs. If the PAMs have been done correctly in the earlier exercises, changes in the Tractor 25