Yugoslavia - Agricultural prices and subsidies case study (Vol. 3 of 3)

AGP- 8 DISCUSSION DRAFT Vol. 3 This internal working paper is prepared for Staff Use Only. The views expressed are not necessarily those of the World Bank. 1AT CnT YUGOSLAVIA: AGRICULTURAL PRICES AND SUBSIDIES CASE STUDY (Volume III: Supplementary Report) AGREP DIVISION WORKING PAPER NO. 8 Prepared by: ULG Consultants Limited in association with Economic Consultants Limited March 1977 C 0 NTENTS Page I. INTRODUCTION 1 II. ECONOMIC EFFICIENCY/COMPARATIVE ADVANTAGE 2 AND RELATED ISSUES A. GENERAL 2 B. COMPARISON OF DOMESTIC RESOURCE COST MEASURES (DRC) 3 C. NET ECONOMIC BENEFIT 9 D. EFFECT ON THE BALASSA COEFFICIENTS OF CHANGES IN YIELDS AND WORLD PRICES 19 E. SUPPLY RESPONSE 22 F. REGIONAL AND PRODUCTION SECTOR DIFFERENCES 24 G. LARGE AND NEGATIVE EPCs. 26 III. CONSUMER SUBSIDY EQUIVALENTS: REVISED CALCULATIONS 27 A. INTRODUCTION 27 B. BEEF 27 C. PIG MEAT 30 IV. COMPARATIVE ASSESSMENT OF PERFORMANCE MEASURES 33 A. GENERAL 33 B. EXAMPLE 35 C. INTERPRETATION 37 D. EMPIRICAL PRACTICABILITY 44 E. USE OF THE MEASURES IN EXPLAINING POLICY 48 F. USE OF THE MEASURES IN PROJECT APPRAISAL 49 G. USE OF THE MEASURES IN PRICING POLICY 52 H. USE OF THE MEASURES BY SECTOR AND ECONOMIC MISSIONS 54 ANNEX I - TABLES 1-24. ANNEX II - MARGINAL PRODUCT OF LABOUR ANNEX III - GRAPHS ANNEX IV - REFERENCES ANNEX V - METHOD OF CALCULATION OF MEASURES Page 1 I. INTRODUCTION. At a meeting in Warwick on April 14th 1977 between C.Bruce. Chief Economics & Resources Division. Agriculture and Rural Development Department of the World Bank and G.M.H.Cox of Economic Consultants Limited, it was agreed that some additional work on the Yugoslavia Agricultural Prices and Subsidies Case Study was desirable. The subjects to be covered in a supplementary report were discussed and agreed with the details being set out in three letters, two from the World Bank to ULG Consultants dated 22nd and 29th April and one from ULG Consultants to the World Bank dated April 28th. This report covers the additional work together with a revision of the original consumer subsidy equivalents as new data has become available.since March 1977. The report is divided into three main sections covering:- a) Economic Efficiency/Comparative Advantage including calculation of DRC coefficient according to equation 4 of the Supplementary Guidelines and the Net Economic Benefit per unit of output. b) The revised Consumer Subsidy Equivalents. c) A comparative assessment of the performance measures, their meaning use and practicability. The annexes include the calculations associated with (a) above together with a more detailed explanation of the estimates of the marginal product of labour used in the study. Page 2 II. ECONOMIC EFFICIENCY/COMPARATIVE ADVANTAGE AND RELATED ISSUES A. GENERAL The supplementary guidelines for consultants 1/propose three ways of expressing economic efficiency/comparative advantage, namely equations 4, 5 and 6. Equation 5 is the DRC coefficient already calculated in Annex II of the original report, while the calculation of equations 4 and 6 is shown in Annex I of this report. To enable comparisons to be made between wheat, maize, beef and pig meat produced by the social and individual sectors .equation 6 is expressed in all cases as the net economic benefit per ton. This required a limited amount of additional calculation as the original DRCs in equation 5 had not been derived in the same way. The individual sector ratios were calculated on a per farm basis while those from the social sector were either per ha or per 100 Kg Liveweight. 1/ Case Studies of Agricultural Prices and Subsidies Supplementary Guidelines for Consultants. March 10, 1977. Page 3 B. COMPARISON OF DOMESTIC RESOURCE COST MEASURES(DRC). 1. Agricultural Sector The DRC coefficients calculated according to equations 4 and 5 are compared in Table 1 and the following points should be noted:- a) In most instances the DRC coefficient from equation 4 is less than that calculated from equation 5, because the numerator in domestic prices is less than that in border prices. The denominator is the same in both equations. The numerator in domestic prices is generally less than that in border prices because:- i) with domestic cereal prices generally lower than border prices the opportunity cost of land in domestic prices is also less than in border prices because cereal prices are closely linked to the opportunity cost of land. This occurs because the opportunity cost of land in wheat production is taken as return 1/ to land in maize production and vice versa, while the opportunity cost of land in forage production is the average of the returns from wheat and maize. ii) farm machinery is cheaper in domestic than border prices which reduces the average capital investment and hence the annual cost of capital in equation 4 compared to equation 5. iii) in eight years out of ten studied the conversion factor used to convert labour and working capital costs from domestic to border prices is greater than one. Higher border than domestic prices for machinery and cereals are a major 2/ influence on the value of the conversion factor used. - 1/ Value of production - (material inputs + cost of labour and capital). 2/ See Annex II page 15 Yugoslavia Agricultural Prices and Subsidies Case Study for details of the calculation of the conversion factor. Page 4 Table 1 COMPARISON OF MEASURES OF ECONOMIC EFFICIENCY/COMPARATIVE ADVANTAGE WHEAT MAIZE BEEF PIG MEAT 1/ .ontiq 4 5 6 4 5 6 4 5 6 4 5 6 1. Social Sector FRY .366 0.53 1.16 (69) 0.93 0.89 49 - - (6408) 0.66 0.70 777 '967 0.11 0.48 179 1.60 1.56 (138) - - (5578) 0.61 0.63 1152 )68 - - 330 3.14 2.75 (258) * - (7913) 0.48 0.48 2037 1969 0.39 1.38 (76) 1.20 0.85 46 - - (5230) 0.35 (D.34J g19 970 * 2.09 (306) 0.70 3 OCfe 54 38.33 43.83 (4960) 0.59 0.62 1380 971 0.38 (6-4 279 2.19 1.67 (229.) 1.54 1.45 (1698) 1.35 1.34 (616) )72 0.23 0.73 284 0.88 1.25 (184) 1.51 1.83 (4207) - - (4476) 1973 0.06 0.56 11 0.58 1.60 (870) - - (18200) 0.86 1.09 (429) .74 0.39 0.59 841 1.10 1.51 (692) - - (21095) 1.69 1.87 (2439) .75 0.77 0.88 165 0.93 1.09 (99) - - (23855) 3.96 3.94 3940). Individual Sector Serbia Proper 966 0.64 0.92 37 0.83 1.07 (36) 6.40 7.48 (6022) 1.10 1.16 (489) .367 0.48 (0.64 160 1.12 1.43 (.15u) 3.00 3.42 (4674) 0.86 0.90 420 '968 0.87 0.94 20 0.90 1.05 (16) - (6570) 0.80 0.82 803 369 2.40 2.87 (330) 0.51 .45. 203 38.19 39.14 (6075) 0.75 ,0.76 ' (1252) 1970 1.04 1.45 (194) 0.40 0.73 138 5.93 7.43 (6709) 1.27 1.33 (1032) 971 1.93 1.24 (83) 0.96 0.84 65 1.94 1.66 (3419) 3.44 3.37 (3430) 1972 0.71 0.96 4 0.49 I.M__L261 1.85 2.51 (8492) - - (8719) '973 0.31 0.84 (323 0.23 1.17 (227) 13.08 30.79 (25135) 3.53 4.61 (7692) 974 0.83 1.37 (415) 0.52 0.77 69) - - (27084) 2.89 3.24 (7218) Individual Sector SAP Voyvodina ?66 0.66 1.05 (48) 0.54 0.96 29 - - (11112) 1.24 1.31 (837) .967 0.52 0.74 251 0.85 1.31 (171) 4.68 5.96 (8196) 0.82 0.86 606 968 0.81 1.04 (29) 0.75 0.96 18 - - (7560) 0.75 0.76 1125 369 1.09 1.51 (285) 0.65 0,2p 42.65 47.D4 (8021) 0.68 (J.6B 1194 1970 1.13 1.79 (640) 0.34 (.59) & 6.47 8.66 (8920) 0.98 1.03 (130) 371 1.06 0.74 282 0.88 1.31 (195) 2.24 2.24 (5827) 2.36 2.32 (2736) 72 0.61 0.84 0.43 1.17 (187) 1.26 1.82 (7023) - - (7921) 73 0.17 . ) (2253) 0.28 1.80 (1475) 2.17 5.58 (24643) 2.09 2.74 (6719) '74 0.58 0.82 531 0.53 1.21 (387) - (25519) 1.67 1.87 (4765) Page 5 Table 1 (Contd) WHEAT 1/ MAIZE 1/ BEEF 1/ PIG MEAT 1/ ,guation 4 5 6 4 5 6 4 5 6 4 5 6 . Individual Sector SAP Kosovo '966 0.54 0.70 138 1.14 1.30 (147) 6.86 7.43 (4779) 1.52 1.59 (1355) .967 0.50 0.64 192 1.29 1.41 (200) 2.09 2.23 (2842) 1.62 1.69 (1575) '968 0.68 0.62 143 1.31 1.38 (138) 7.68 7.76 (4165) 1.11 1.13 (446) 369 0.93 0.91 25 1.33 1.06 (25) 6.47 6.40 (4241) 0.80 0.80 984 1970 0.74 0.95 26 0.95 1.D4 (26) 22.17 24.14 (6110) 1.17 1.23 (764) 971 1.07 0.55 183 1.71 1.50 (200) 1.35 1.28 (1406) 5.48 5.38 (4010) 1972 0.57 0.61 344 0.84 1.47 (350) 1.37 1.57 (3402) - - (8903) 973 0.20 0.59 779 0.38 1.55 (624) 17.48 27.71 (14842) 4.21 5.52 (8135) *974 0.67 1.11 (117) 0.41 0.92 132 7.01 8.84 (13776) 3.37 3.78 (7761) 'ote: Some of the equation 5 coefficients differ slightly from those in Annex 11 volume 2 of the Yugoslavia Agricultural Prices and Subsidies Case Study as some small computational errors have been corrected. I/ Dinars per metric ton. Page 6 In most years these offset the effect of the moderate levels of duty on the other inputs eg fertiliser and fuel, which tend to make domestic prices higher than border prices. However only in two years out of ten did the conversion factor exceed 110. b) Equations 4 and 5 show less divergence for pig meat production than for the other three commodities because:- i) no land cost is included ii) little investment in farm machinery is required iii) the conversion factor is less than 110 in eight years out of ten so giving relatively small differences in the size of the numerator when expressed in domestic prices as opposed to border prices. c) Whereas the DRC coefficients calculated using equation 5 can be used to compare the economic efficiency across a whole range of commodities and to recommend which commodities should be expanded or contracted to maximise economic efficiency 1/ this cannot be done with DRC coefficients calculated from equation 4. In both equations 4 and 5 the opportunity cost of land is measured in terms of the return from the next best alternative but this together with other components of the numerator in equation 4 is expressed in domestic not border prices. As a result the DRC coefficient from equation 4 indicates whether or not that sector should be expanded or contracted to maximise returns to the producers only given that no change in policy affecting the alternative used to measure the opportunity cost of land is possible. 1/ Those sectors with a DRC coefficient (1 would be expanded, provided also that the DRC reflects the marginal production of the industry as well as the average, and those with a DRC coefficient-l would be contracted. The DRCs shown in Table 1 are averages. However, the DRC itself does not indicate by how much the sector should be expanded or contracted. Page 7 For example, if wheat has a DRC coefficient using equation 4 of 2.0, then wheat production should be contracted to maximise producer returns , but only if no change is made to policies affecting maize, since the return to land from maize production is the opportunity cost of land for wheat production. 2. Non Agricultural Sector The DRC coefficients from branches of the manufacturing and mining sector, calculated using the two alternative formulae, equations 4 and 5, are presented in Table 2.-1/ The estimates based on equation 4 are considerably higher than those on the other formula, by about 10% in 1970 and 11% in 1975. This is to be expected, since (on the available evidence) domestic 2/ prices in Yugoslavia have on the whole been above border prices.- The ranking of the four industries, in terms of size of DRC, is the same as on the equation 5 basis. DRCs based on equation 5 are the simpler to interpret, since (at least in theory) a DRC less than one on this basis shows that the industry concerned makes a "profit" for the economy, in terms of foreign exchange gained or saved; while a DRC above one indicates a "loss", by the industry taken as a whole. The equivalent dividing line between "profit" and "loss" is less easily identified on the equation 4 basis, because the numerator and (enominator of the coefficient are expressed in different price terms.!/ 1/ The opportunity has been taken here to correct a minor inconsistency of approach between the calculation for 1970 and 1975. However there is little difference between the equation 5 estimates in table 2 and those in the earlier report, and the comments there on pages 55-56 still apply. 2/ In contrast to the agricultural sector where border prices have more often been higher than domestic prices. 3/ Some caution is required in drawing policy conclusions from the estimates given here, even those based on equation 5, since they represent industry averages, not the marginal position. Page 8 Table 2 DOMESTIC RESOURCE COST COEFFICIENTS, MANUFACTURING INDUSTM 1970 1975 Iron and steel: equation 4 basis 0.92 1.27 equation 5 basis 0.83 1.13 Electrical equation 4 basis 0.97 1.07 equipment: equation 5 basis 0.88 0.96 Textiles and equation 4 basis 1.25 1.19 Clothing: equation 5 basis 1.13 1.07 Food and drink equation 4 basis 1.01 3.43 manufacturing: equation 5 basis 0.92 3.07 Page 9 C. NET ECONOMIC BENEFIT. 1. Agricultural Products Estimates of net economic benefit per ton of production using equation 6, are shown in table 3 both in current and constant 1970 dinars for wheat, maize, beef and pig meat production from both the social and individual sectors. Equation 6 is equation 5 expressed as an absolute number per unit of output rather than as a ratio and so gives a meaningful value when the value added in border prices is very small or even negative (see table 1). Equation 6 therefore indicates which sectors are not profitable and have been over eRAnd0d- i.e. those with negative net economic benefits,or profitable and should be expanded i.e. those with positive net economic benefits. Because of the recent high rates of inflation the current dinar figures are difficult to interpret and the following comments refer to the 1970 constant dinar figures. Beef production from both the social and individual sectors has consistently shown a negative benefit per ton, which in real terms has increased significantly since 1972. Beef production in both .sectors can be considered to have been over expanded. As mentioned in the original report there is evidence that this sector is also financially unsound. The use of high cost cereal diets is considered to be a major factor in the social sector, while in the individual sector this plus poor technical performance are likely to have contributed to the consistent negative net economic benefit. The use of some land with a high opportunity cost i.e. which could grow maize and/or wheat must also be a contributory factor. This is evident in Serbia Proper and SAP Voyvodina where cereal yields are higher than in SAP Kosovo and the negative benefit per ton of beef is also greater. Page 10 Table 3 NET ECONOMIC BENEFIT PER METRIC TON WHEAT MAIZE BEEF PIG MEAT (a) (b) (a) (b) (a) (b) (a) (b) 1. Social Sector FRY 1966 (69) (81) 49 58 (6408) (7539) 777 914 1967 179 218 (138) (168) (5578) (6802) 1152 1405 1968 330 418 (258) (327) (7913) (10016) 2037 2578 1969 (76) (87) 46 53 (5230) (6001) 3819 4390 1970 (306) (306) 154 154 (4960) (4960) 1380 1380 1971 279 221 (220) (175) (1698) (1348) (616) (489) 1972 284 182 (184) (118) (4207) (2697) (4476) (2869) 1973 1104 566 (870) (446) (18200) (9333) (429) (220) 1974 841 377 (692) (310) (21095) (9460) (2439) (1094) 1975 165 65 (99) (39) (23855) (9466) (3940) (1563) 2. Individual Sector Serbia Proper 1966 37 44 (36) (42) (6022) (7085) (489) (575) 1967 160 195 (150) (183) (4674) (5700) 420 512 1968 20 25 (16) (20) (6570) (8316) 803 1016 1969 (330) (379) 203 233 (6075) (6983) 1252 1439 1970 (194) (194) 130 138 (6709) (6709) (1032) (1032) 1971 (83) (66) 65 52 (3419) (2713) (3430) (2722) 1972 34 22 (26) (17) (8492) (5444) (8719) (5589) 1973 323 166 (227) (116) (25135) (12890) (7692) (3945) 1974 (415) (186) 319 143 (27084) (12145) (7218) (3237) 3.-Individual Sector SAP Voyvodina 1966 (48) (56) 29 34 (11112) (13073) (837) (985) 1967 251 306 (171) (209) (8196) (9995) 606 739 1968 (29) (37) 18 23 (7560) (9570) 1125 1424 1969 (285) (328) 167 192 (8021) (9220) 1794 2062 1970 (640) (640) 312 312 (8920) (8920) (130) (130) 1971 282 224 (195) (155) (5827) (4625) (2736) (2171) 1972 315 202 (187) (120) (7023) (4502) (7921) (5078) 1973 2253 1155 (1475) (756) (24643) (12637) (6719) (3446) 974 531 238 (387) (174) (25519) (11444) (4765) (2137) Page 11 Table 3 (Contd) WHEAT MAIZE BEEF PIG MEAT (a) (b (a) (b) (a) (b) (a) _b) 4. Individual Sector SAP Kosovo 1966 138 162 (147) (173) (4779) (5622) (1355) (1594) 1967 192 234 (200) (244) (2842) (3466) (1575) (1921) 1968 143 181 (138) (175) (4165) (5272) (446) (565) 1969 25 29 (25) (29) (4241) (4875) 984 1131 1970 26 26 (26) (26) (6110) (6110) (764) (764) 1971 183 145 (200) (159) (1406) (1116) (4010) (3183) 1972 344 221 (350) (224) (3402) (2181) (8903) (5707) 1973 779 399 (624) (320) (14842) (7611) (8135) (4172) 1974 (117) (52) 132 59 (13776) (6178) (7761) (3480) (a) * Current New Dinars (b) * 1970 New Dinars obtained using the index of producers' prices of agricultural products. Page 12 Since 1971 pig meat production has been over expanded in both sectors as evidenced by the negative economic benefit per ton. The higher "loss" per ton from the individual sector reflects poorer technical performance. Prior to 1970 the sector was generally under expanded. The change was due to a decline in the value added at border prices relative to the opportunity cost of labour and capital as the cost of cereal inpyts increased by 2.47 while the value of pig meat rose by 1.79/between 1971 and '1975.' Excluding 1969 and 1970 wheat production in both sectors has been under expanded while maize has been over expanded..2 While the net economic benefit per ton indicates which sectors are or are not economically profitable and so should be expanded or contracted it gives an idea of the importance to the econouy of those sectors and where policy changes would have most impact. To do this the economic cost of the wheat, maize, beef and pig meat produced by both sectors has been calculated. These figures are shown in table 4 and have been obtained by multiplying the per ton figures from table 3 by the production of the relevant commodity. Although no survey data are available for the individual sector of republics other than Serbia, the Agricultural Economics Institute considers the figures for Serbia Proper to approximate to those that would be produced from a survey of the individual sector throughout the Federation. The per ton figures for Serbia Proper have therefore been multiplied by the production of the relevant commodity from the individual sector in all Republics. This enables a direct comparison to be made with the / Social Sector / The sum of the net economic benefits per ha from maize and wheat in any one year is zero, because the opportunity cost of land is measured in terms of the value of the production from the other cereals (see Tables Appendix 1.) Per ton negative or positive the maize figures are less than those for wheat because the yield per ha is higher. Page 13 Table 4 TOTAL ECONOMIC BENEFIT * 1. Social Sector FRY ------------- WHEA T---------- ------------MAI Z E------------- Production Net Economic Total Production Net Economic Total '000 tons Benefit/ton m.dinars '000 tons Benefit/ton m.dinars 1/ 2/ 2/ I/ 2/ 2/ 1966 1403 (81) (114) 1173 58 68 1967 1524 218 332 1270 (168) (213) 1968 1647 418 688 1280 (327) (419) 1969 1804 (87) (157) 1259 53 67 1970 1422 (306) (435) 1045 154 161 1971 1975 221 436 1324 (175) (232) 1972 1819 182 331 1347 (118) (159) 1973 1768 586 1036 1364 (446) (608) 1974 2423 377 913 1328 (310) (412) 1975 1739 65 113 1586 (39) (62) ----BE E ----- - ------PIG MEAT ---------- 1966 75 Y (7539) (565) 124 (914) (113) 1967 84 (6802) (571) 133 1405 187 1968 89 (10016) (891) 149 2578 384 1969 60 (6001) (360) 116 4390 509 1970 60 (4960) (298) 116 1380 160 1971 66 (1348) (89) 131 (489) (64) 1972 64 (2697) 4173) 125 (2869) (359) 1973 67 (9333) (625) 132 (220) (29) 1974 83 (9460) (785) 140 (1094) (153) 1975 81 (9466) (767) 142 (1563) (222) * Referred to as Economic oCost" with opposite sign in Section IV. 1/ Source: Table 11-10 Annex I. Yugoslavia Agricultural Prices and - Subsidies Case Study. 2/ 1970 constant prices. 3/ Source: Table 11-7 & 11-12 Annex I as above. Also assumes total production eplit between sectors in the same percentage as marketed production. See also Table 11-13. Assumes published figures for production are in tons DW. Page 14 Table 4 Contd 2. Individual Sector FRY ------------ WHEAT ----------- ------------ MAAZI 2 E *--------- Production Net Economic Total Production Net Economic Total '000 tons Benefit/ton m.dinars '000 tons Benefit/ton m.dinars 1/ 2/ 2/ If 2/ 2/ 1966 3197 44 141 6807 (42) (286) 1967 3296 195 643 5930 (183) (1085) 1968 2713 25 68 5530 (20) (111) 1969 3076 (379) (1166) 6562 233 1529 1970 2368 (194) (459) 5888 138 813 1971 3629 (66) (240) 6119 52 318 1972 3024 22 67 6583 (17) (112) 1973 2982 166 495 6889 (116) (799) 1974 3859 (186) (718) 6703 143 959 ---- BEE F --- - ------PIG NEA T----------- 1966 152 3/ (7085) (1077) 163 3/ (575) (94) 1967 172 (5700) (980) 176 512 90 1968 203 (8316) (1688) 174 1016 177 1969 215, (6983) (1501) 171 1439 246 1970 185 (6709) (1241) 223 (1032) (230) 1971 197 (2713) (534) 253 (2722) (689) 1972 202 (5444) (1100) 215 (5589) (1202) 1973 205 (12890) (2642) 176 (3945) (694) 1974 226 (12145) (2745) 253 (3237) (819) Page 15 Social Sector where the per ton figures in table 3 come from a survey which covers five out of six Republics. The figures calculated in table 4:- (a) Re-emphasize the relative importance of the individual sector compared to the social sector. Agricultural policy changes must affect small farmers if they are going significantly to reduce economic costs. (b) Highlight the recent relatively large cost to the economy of the beef and pig meat sectors. However the cost of beef from the individual sector was three to four times that fromthe social sector. (c) Show that the over expansion of social sector maize production has .had a considerably lower economic cost than that associated with beef production, while an expansion of wheat production from the social sector would be economically beneficial. In the individual sector there have been fluctuations between over and under expansion of whoat and maize which perhaps indicate that little change in the levels of production is economically desirable. 2. Non Agricultural Sector Estimates of net economic benefit, using equation 6 are presented in table 6 for the four manufacturing industries. The initial estimates represented the total net economic benefit in absolute terms of each industry. Table 5 shows these estimates both in current price terms and adjusted to a constant (1970) price / Net Benefit Equation 6 per unit of output x production referred to as economic cost with opposite sign in Section IV. Page 16 Table 5 NET ECONOMIC BENEFIT, MANUFACTURING INDUSTRIES 1970 1975 (a) Total value for industry, current prices (mn. dinars) Iron and steel 326 - 990 Electrical equipment 333 376 Textiles and clothing - 649 - 1043 Food and drink manufacturing 434 - 10.217 (b) Total value, 1970 prices (mn. dinars) Iron and steel 326 - 394 Electrical equipment 333 150 Textiles and clothing - 649 - 416 Food and drink manufacturing 434 4071 (c) As percentage of total output Iron and steel 4.8 - 2.2 Electrical equipment 3.5 1.2 Textiles and clothing * 3.9 - 1.7 Food and drink manufacturing 1.6 - 11.9 (d) As percentage of resource costs Iron and stee) 20.6 - 11.7 Electrical equipment 14.0 4.4 Textiles and clothing 11.9 - 6.1 Food and drink manufacturing 8.9 - 67.5 Page 17 basis using the retail prices index. It was neither meaningful nor possible to estimate net economic benefit per unit of output, since one was dealing with whole industries each with a multiplicity of products. Therefore, to remove the effect of differing industry sizes, net economic benefit was also calculated as a percentage of (1) total output measured in domestic prices; and (ii) domestic resource costs, in border prices. The resulting estimates are also given in table 5. The figures in table 5 correspond closely to those in table 2. By definition, where the estimate based on equation 5 in table 2 is less than one, the corresponding figure in table 5 is positive; and where it is above one, the figure in table 5 is negative. In absolute terms, the four industries taken together appear to have provided a modest net economic benefit in 1970 and at least in theory could have been expanded to the benefit of the economy, but to have made a large economic *loss' in 1975 when contraction would have had an economic benefit. In 1975 only the electronic equipment industry produced a positive result. Table 5 shows an especially heavy *loss" by the food industry in 1975. in both absolute and percentage terms; however, as mentioned in Volume 1 the precise figure must be regarded as dubious, although there probably was an appreciable "loss* in that particular year. / A warning was given in the original report against undue trust in the accuracy of the DRC estimates for non-agricultural industries; it applies equally to the estimates of net economic benefit presented here, which should therefore be interpreted rather cautiously. Page 18 The ranking of the industries in terms of total net economic benefit differs from that based on the percentage measures. For example, in 1970 the food industry was first and the smaller steel industry third, in terms of absolute value; whereas the positions are reversed, on the percentage basis. The ranking based on the percentages is the same as that based on the DRCs in Table 2.1/ / In the case of the percentages related to domestic resource costs (part (d) of table-B), this follows by definition . Page 19 D. EFFECT ON THE BALASSA COEFFICIENTS OF CHANGES IN YIELDS AND WORLD PRICES. Yields and world prices are both incorporated in the calculation of the EPC 11 and DRC coefficient. To examine the relative importance of these two items in the determination of these coefficients the effects of a ! 30% change in both the yield per ha 9 and the world price on the EPC and DRC were calculated 3/ using the 1975 Social Sector wheat data as an example. The results are summarised below:- EPC DRC (Equation 5) Yield per ha - 3D% 0.65 2.38 Actual 0.80 0.88 + 30% 0.83 0.55 Difference - 30% to + 30% 0.18 1.83 World price - 30% 0.49 2.35 Actual 0.80 0.88 + 30% 2.12 0.54 Difference - 30% to + 30% 1.63 1.81 Changes in yield per ha have much less effect on the EPC than changes in world wheat prices because the yield is included in both the denominator and numerator of the EPC equation, whereas the world wheat price is only included in the denominatorj t I) r. 1/ ESC is ignored here because of the difficulties involved in its calculation. These are referred to on page 13 of Annex II of Yugoslavia Prices and Subsidies Case Study. 2/ Equivalent to the highest and lowest wheat yields recorded between 1966-75. 3/ It has been assumed that the material inputs would remain unchanged in both cases. This is realistic as the materials used in cereal growing e.g. seed and fertiliser are applied per ha and machinery inputs are little affected by yield. Neither are influenced by the world grain prices. Page 20 The same percentage change in the yield and world price have a similar effect on the DRC coefficient as both are included in the denominator of the equation and neither is included in the numerator. A similar comparison for livestock products is not possible because there is no suitable yield data and because material input usage is generally linked to output. (Unlike cereals where climatic conditions have a major effect on Yugoslavia's cereal yields). Higher yields of meat per animal require additional feed, so to increase the yield by 30% with the same material inputs would be unrealistic. The effect of world prices on both the EPC and DRC coefficient is very important as the following example for beef production from the Individual Sector in SAP Kosovo in 1972 illustrates: EPC DRC (equation 5) World Price - 20% 4.07 11.24 Actual 0.57 1.57 + 20% 0.31 0.85 Difference - 20% to + 20% 3.76 10.39 In the above case the effect of world price changes is exaggerated because value added in border prices is only 23% of output compared with 48% in.the wheat example. Graphs at Appendix III show the relationship between the EPC and DRC coefficient,world prices,and in the case of cereals yields per ha. Graphs are included for the Social Sector and the Individual Sector in Serbia Proper. Results from this area are generally considered to approximate to the average for the Individual Sector throughout the Federation. / This assumes that there is no change in the yield of the alternative crop. Page 21 The graphs also show that for each comodity the EPC and DRC coefficient generally follow the same trend. This is because both coefficients have the same denominator.. Page 22 E. SUPPLY RESPONSE Production trends for the main agricultural products were briefly discussed on pages 14-18 of Volume I of the previous report. Comparison of production data with the Balassa protection coefficients.:- suggests the following. 1. Despite fluctuations between individual years caused by the climatic conditions, cereal production from both sectors has increased over the ten years studied, although both theLPAnd EPCs have generally been below one. This modest expansion in proTuctonTas n-ebn aieved Irough a policy of high producer prices, although with domestic prices having recently been increased to near world price levels. the farmrs' financial returns from est rg_n_bth sectors have increased sharply. Higher yields per ha particularly in the individual sector through the uptake of modern technology have also helped maintain the relative profitability of cereal production. 2. Allowing for fluctuations in production beef production from the social sector appears to be stagnant or declining in spite of moderate levels of protection. This reflects the poor financial performance of this enterprise. The Institute's survey shows that on average beef production from the social sector was unprofitable six years out of the last ten. However, beef production from the individual sector seems to be increasing in response to the price protection. This apparent contradiction can probably be explained by the individual farmers placing a lower value on their labour than the wages paid by the social sector and so apparently having a higher level of profitability. / The measures of economic efficiency have already been related to production from the social and individual sectors in Sections II 8 & C. Page 23 3. During the period 1966-1969/70 both the NPC and EPC for pig meat production were less than one and production from the social sector declined while that from the individual sector stagnated. Since 1970/71 the EPC, from both sectors has increased and the higher effective protection has encouraged additional production and resulted in improved profitability. Page 24 F. REGIONAL AND PRODUCTION SECTOR DIFFERENCES The data available from the Institute of Agricultural Economics did not make possible direct comparisons of the protection and incentive coefficients between different agroclimatic zones. Although data from the social sector was available on a Republic basis for 5 out of the 6 Republics in the Federation, the Republic boundaries do not necessarily coincide with agroclimatic zones. Also a Republic may include more than one agroclimatic zone as for example Serbia does, where there is a marked difference in agricultural production potential between SAP Voyvodina and SAP Kosovo. In addition there is a tendency for social sector farms to be situated on the most productive land so reducing to some extent the effect of any natural regional variation. Production methods are the same in the social sector throughout the Federation. Units are large and production involves the use of modern technology and is highly mechanised. Data from the individual sector was only available from the three areas of Serbia. These have the following characteristics:- i) SAP Voyvodina: an area of high quality arable land from which high crop yields are obtained and which is relatively sparsely populated. ii) SAP Kosovo: a mountainous area with poor soils and limited crop production potential, but densely populated. iii)Serbia Proper: an area whose production potential lies somewhere between the above two extremes. In all three areas individual sector farms are small with less than 10 ha of arable land; they are based on family labour, but the use of modern technology and mechanisation is increasing. Page 25 A comparison of the coefficients from the individual sector in SAP Voyvodina and SAP Kosovo gives an idea of the likely range of results that would be obtained from similar studies throughout the Federation. As mentioned earlier the results from Serbia Proper are considered to reflect the average that would be obtained from all Republics, if data were available. A comparison of the coefficients calculated in this and the previous report for the individual sector in Serbia Proper and the social sector can be taken to represent the protection and economic efficiency of the two sectors throughout the Federation. Page 26 G. LARGE AND NEGATIVE EPCs. In the calculation of the coefficients for beef production the value added in border prices was frequently very small or negative. When value added in border prices is small relative to that in domestic prices the EPC is very large and subject to very considerable fluctuations as is any index with a small base. While a negative value added is a perfectly valid concept it is not possible to calculate an EPC comparable to those calculated with positive value added. A possible approach to these two problems would be to express the EPC in the following way: "EPC" - VAd * VAb instead of EPC * VAd The following example compares the two methods. Year VAd VAb VAd - VAb VAd TF_ VAd 1 100 90 0.10 1.11 2 110 120 (0.09) 0.92 3 110 150 (0.36) 0.73 4 100 120 (0.10) 0.83 5 90 60 0.33 1.50 6 80 20 0.75 4.00 7 8o 10 0.88 8.00 8 80 5 0.94 16.00 9 80 (10) 1.13 - 10 100 (20) 1.20 In the case of DRC coefficient where small or negative value added in border prices causes the same problem the use of equation 6 enables a meaningful figure to be obtained. (see Section II.C.). Page 27 III.CONSUMER SUBSIDY EQUIVALENTS: REVISED CALCULATIONS A. INTRODUCTION. It was mentioned in the Report of March 1977 (page 73) that further information was awaited on the "compensation payments* made to the meat packing industry. This has been received, and the compensation payments were first introduced in 1975. Accordingly none applied in the period 1972-74. contrary to the assumption made in calculating the "subsidy equivalents" presented in the Report. The estimates of consumer subsidy equivalents and of Government subsidy costs for beef and pig meat have now been recalculated, and are shown in Tables 6-9 (the producer subsidy equivalents are unaffected). B. BEEF. 1. Consumer Subsidies The measures of beef consumer subsidies in Table 6 show a considerable change from the previous estimates. The figures for 1972 and 1973 are much reduced, though still positive; while the "negative subsidy" in 1974. due to the protection of Yugoslav prices from the fall in the world market, is now estimated to have been much more substantial. 2. Government cost The estimates of financial transfers between beef consumers and the governmental sector and of total government costs, shown in Table 7 have also been much altered. It is now assumed that there were no direct payments by the Government sector to keep down consumer prices in 1972-74. As aconsequence, it is no longer estimated that total government expenditure on beef subsidies was significant in 1972 and 1973; while the estimate for 1974 is considerably reduced, although still appreciable owing to the cost of supporting producer prices. Page 28 Table 6 Beef: Consumer Subsidy Equivalents 1972 ,931974 1. Consumption* ('000 tons) 145 155 165 2. Price, ex-meat packer equivalent** (din.per kg) 26.36 35.81 32.20 3. Direct consumer payments (mn. din) - - - 4. Total consumer cost (mn. din) 3,822 5,551 5,313 5. Policy transfers to consumers (mn. din): Price protection*** 216 102 - 6B6 6. Proportional subsidy (%) 5.7 1.8 - 12.9 7. Subsidy per unit (din.per ton) 1,490 658 - 4,158 * Consumption of authorised purchases, less net exports; estimated. * Farm-gate equivalent price multiplied by 1.07. ** Difference between border price and producer price, multiplied by consumption. Page 29 Table 7 Beef: Subsidy Transfers between Sectors (million dinars) 1972 1973 1974 Producer Subsidy Value: - 237 - 63 986 from consumers - 216 - 102 686 from Government sector - 21 39 300 Consumer Subsidy Value: 216 102 - 686 from Producers 216 102 - 686 from Government sector - - - Government sector cost: - 21 39 300 to producers - 21 39 300 to consumers Page 30 C. PIG MEAT 1. Consumer Subsidies Table 8 presents estimates of consumer subsidies on pig-meat, which again differ considerably from the figures in the previous Report. Instead of receiving a substantial subsidy in 1972, pig meat consumers now appear to have suffered a net "loss" through relatively high domestic prices. Consumer subsidies in 1973 and 1974, when domestic prices were held below world market levels, are still reckoned to have been very high, but well below the previous estimates. In percentage terms, the consumer subsidy equivalent for pig meat was on average higher than for beef over the period 1972-74, but lower than for wheat and maize. 2. Government cost There was no governmental expenditure on pig meat consumer subsidies in 1972 or 1974, according to these estimates; some cost, however, was incurred in subsidising the 1973 import deficit. The previous estimates of overall expenditure on subsidising pig meat producers and consumers has been revised downwards sharply; it now appears that the total cost to the government sector was significant only in 1973. Table 9 shows our estimates. Total government sector expenditure on producer and consumer subsidies over the period 1972-74 was about the same for pig meat as for beef, but much less than for wheat; the government appears to have made some "profit" on maize. However with the introduction of compensation payments to the meat packing industry in 1975 the position is likely now to be considerably changed with respect to beef and pig meat. Page 31 Table 8 Pig-meat: Consumer Subsidy Equivalents 1972 1973 1974 1. Consumption* ('000 tons) 199 200 254 2. Price, ex-meat packer equivalent* (din. per kg) 14.04 21.35 20.36 3. Direct consumer payments (mn.din) - - - 4. Total consumer cost (an.din) 2.794 4.270 5,171 5. Policy transfers to consumers (mn. din): Price protection*** - 247 1.832 1,003 6. Proportional subsidy (%) - 8.8 42.9 19.4 7. Subsidy per unit (din. per ton) - 1,241 9.160 3,949 * Authorised purchases plus imports, less exports (for 1974. average of 1974 and 1975 exports taken). ** Farm-gate equivalent price multiplied by 1.07. ** Difference between border price and producer price, multiplied by consumption. Page 32 Table 9 Pig-meat: Suhsidy transfers between Sectors (million dinars) 1972 1973 1974 Producer Subsidy Value: 308 - 1,593 - 973 from Consumers 247 - 1,613 -1,003 from Government sector 61 20 30 Consumer Subsidy Value: - 247 1,832 1,003 from Producers - 247 1,613 1.003 from Government sector * 219 - Government sector cost: 61 239 30 to Producers 61 20 30 to Consumers - 219 - Page 33 IV. COMPARATIVE ASSESSMENT OF PERFORMANCE MEASURES A. GENERAL This section relates to the calculation, interpretation, and utilisation of the various measures employed in the study. Since each measure has its own correct interpretation, one such measure can only be said to be superior to another if both relate to the same .question. Similarly, each question relating to economic performance or policy evaluation strictly requires its own measure. It is argued below that the measures used in the study are complementary in that they emphasize different aspects of perform- ance, and moreover, that from the basic information assembled, a variety of other 'derived' measures can be calculated to answer specific additional policy questions. Four separate sets of measures are discussed, with a simple numerical example to aid interpretation: a) the 'original' Balassa coefficients (1) b) the Josling producer and consumer impact effects (6) c) the additional measures requested in the 'supplementary guidelines' (2), and d) the 'derived' measures, which follow suggestions made by Josling (4,5) in the context of policy evaluation. The Balassa measures are, by design, indicators of efficiency in resource use. They are all ratios where the values, if properly measured 'at the margin', will approximate to unity under competitive internal markets, free trade, and absence of government intervention. Deviations from unity in such ratios can be taken either as evidence of unwanted distortions in the economic system 1/ numbers in parentheses refer to references listed in Annex IV. Page 34 or as a conscious attempt by governments to manage the markets concerned to achieve particular ends. The measures themselves cannot distinguish between these two cases, but they can still be used to monitor the distortions in a way that indicates to governments the incentives which are being generated by the price policies that they are pursuing. 1/ The Josling measures, derived specifically for the purpose of evaluating the cost-effectiveness of different methods of income transfer associated with farm policies, concentrate on the impact on producers and consumers of a product of a set of price and other methods of market support. Tht.transfer measures as such do not MBare economic cost, but they allow for the direct assessment of 'benefits' in the sense of the achievement of policy objectives. The supplementary guidelines were intended to clarify the calculation of domestic resource cost. The measure represented by equation 5 in that paper, which valued domestic resources at their opportunity cost in foreign exchange terms has been included as one of the Balassa ratios: the comparison bAtween-domes tie.(or.private) oZportuniyg_,W and world-price (or social) value added, as expressed in equation 4 of that paper is regarded as an additional measure in the discussion below. Similarly, the difference between resource cost and value added, both measured at world prices, rather than their ratio, is also considered as an additional measure. The one tells of the private costs of achieving social benefits; the other of the net social benefit (or cost) from a particular sector per unit of output. The derived measures make use of the concepts developed by Josling which relate costs to policy objectives in a direct way. They also extend these concepts to include aggregate measures which reflect the differences in importance among sectors, and the specific / In the following discussion, EPC and ESC are taken as equivalent measures. Page 35 evaluation of incentives which make use of the difference between private and social costs and benefits. Section C touches on the question of the use of the various measures, and in particular stresses the difference between partial and general welfare indicators and between small changes in policy or in output relative to major shifts in economic variables. The subsequent sections discuss practicability of estimation, and the use of the measures in explaining policy implications, project appraisal and pricing issues. B. EXAMPLE A simplified example is used to illustrate these points. The following values are assumed for two prooucts which are substitutes in production: Variable A B Domestic price PD 12.0 25.0 World price P 0 (30.0j) Domestic-price value VAD added 4.5 q0.0 World-price value 3.0 added 3.0 15.0 Domestic-price R resource cost D 2.0 120.0 World-price resource R cost 3.5 12.5 Domestic-price input 5 cost D 7.5 15.0 World-price. input 0 cost 7.0 15.0 Domestic output 5 level D 150 50 Domestic consumption D level D 200 30 From these assumptions, the following measures can be calculated (for details, see Annex IV);- Page 36 Balassa efficiency measures Nominal Protection Coefficient NPC 1.20 0.83 Effective Protection Coefficient EPC 1.50 0.67 Domestic Resource Cost (eq.5) DRC(5) 1.17 0.83 Josling transfer measures Producer subsidy equivalent PSE 0.13 -0.20 Producer subsidy value PSV 225 -250 Consumer subsidy equivalent CSE -0.17 0.20 Consumer subsidy value _U.V -400 150 'Tax' burden TB -175 -100 Supplementary measures Domestic Resource Cost (Eq.4) DRC(4) 0.67 1.33 Net Benefit (eq.6) 8(6) -0.50 2.50 5' Derived measures ) Economic Cost EC - 75 125 Cost of producer transfer EC/PSV 0.33 n.a. Cost of consumer transfer EC/CSV n.a 0.83 Intra-sector transfer cost EC/ (-PSV-CSV) 0.12 0.31 Inter-sector transfer cost EC/ -(-PSV+CSV) 0.43 1.25 Budget cost of producer transfer TB/PSV -0.78 n.a. Budget cost of consumer transfer TB/CSV n.a. -0.66 Private incentive index Pl 0.55 -1.00 Social incentive index SI 0.22 -0.25 1/ Equation numbers refer to the 'supplementary guidelines' paper (2) n.a.Not applicable. Page 37 C. INTERPRETATION. 1. Balassa Coefficients The protection coefficients have a straightforward interpretation: domestic prices, relative to border prices, are 20 percent higher for product A and 17 percent lower for product B, and protected value-added is 50 percent higher and 33 percent lower for these products respectively. Resources have presumably been attracted from B-production to A-production relative to free trade, though if effective rates in other sectors were high it is possible that both activities could be at a disadvantage in attracting resources. The Domestic Resource Cost (DRC) coefficients indicate that resources in A-production are not 'paying their way': the resources have an opportunity cost, at world values, which is 17 percent greater than value added at world prices. In B-production, value added is 17 percent greater than resource cost. The implication is that a switch of resources from A to B would increase total value added from the same resources. 2. Josling measures The transfer measures represent another way of looking at the same situation. Support of product A represents a 13 percent subsidy equivalent to producers - i.e. a direct-subsidy of this amount would compensate producers for the loss of price support. The value of such a subsidy is 225. This measure differs from the protection coefficients in that it takes account of input cost differences (and hence is less than NPC) but has as a denominator the actual value of output rather than value added (and hence is less than EPC)I/ The consumer subsidy equivalent for A is comparable to the NPC, since no consumer programmes are included in the example: the CSV indicates that the price policy has the same effect as a tax of 400 on consumption. For product B, where no input policies are included (ID 0 1W), the PSE.and CSE reflect I/ In the simplified example here, inputs have been treated in the same way in both the EPC and the PSE calculations, but in the PSE estimates in the previous report only the specific agricultural policies affecting inputs were taken into account. Page 38 the value of the NPC; in this case producers suffer a loss equivalent to a tax of 250 and consumers gain by 150 in the form of an implicit subsidy. The 'tax burden' is the sum of the subsidy values, on the assumption that the net gain to producers and combined is at the expense of the exchequer. 3. Additional measures The other two basic measures used relate to the alternative formulation of DRC, using local rather than world prices to calculate opportunity cost and to the net benefit, which restates DRC at world prices as the difference between rather than the ratio of resource cost to value added. DRC(4), in the example, shows that resource cost is low in A and high in B production relative to value added. This is a reflection of the fact that A production is supported relative to B production: it is not an indication that A production should be increased, but merely that as a result of the policies already pursued A production would be more profitable to the sector concerned even if inputs and output were valued at world prices. DRC(5) would seem therefore to be superior to DRC(4) in policy evaluation. The net benefit calculation. B(6), is an indication of the 'per unit' social profit (or loss) of production. In the example, B production shows a larger net loss of resource cost over value added than A production when both are calculated at world prices. 4. Derived measures The derived measures fall into three headings: a) Aggregate Measures. Calculated by applying the quantities involved to the difference between costs and output value. Such a measure is illustrated by the Economi.c Cost (EC), which shows a loss of 75 on A- production andL1125 on B-production. These aggregate measures C,~ ' Page 39 capture the relative importance of the various sectors in the economy: on this evidence, a change in the policies relating to 8 might be considered to be more beneficial to the economy than those relating to product A. Aggregate transfers have already been discussed; though technically the subsidy equivalents were derived from the subsidy values, rather than vice versa, the value figures can be thought of as aggregate measures adding the sector-size dimension to the-indicative interpretation of the ratio and percentage figures. Conceptually all the elements of the Balassa ratios can be separated and aggregated: as an example, the difference between domestically- valued value added and resource cost, multiplied by domestic output, gives the aggregate 'profit' generated in the sector concerned. b) Cost/objective ratios Calculated by specifying the possible aim of the policy and relating this to economic, financial, or transfer cost. Assume in the example given that the objective of A-policy is to support producer incomes (for an imported product) and that B-policy is intended to lower consumer cost (by restricting exports) either to counter inflationary tendencies or to stimulate the domestic processing of home production. The measure indicates that for every 100 units transferred to producers there is a net cost of 33 units in lost value of output under A-policy: B-policy is even more expensive; for every 100 units cost saving to consumers there is, on average, a loss of 83 units to the economy as a whole. Suppose, as another example, that the objective of the policy were to transfer income to A-producers from A-consumers, and to B-consumers from A-producers. The intra- sector transfer costs, per unit, are 0.12 and 0.31 respectively. 1/ The economic cost is taken from B(6); an alternative, and more complete, calculation would estimate the welfare loss in the form of the traditional misallocation triangles. Page 40 If the objective were to raise tax revenue from these two sectors, the relative economic cost per unit raised is 0.43 in the case of A-policy, and 1.25 in the case of B-policy. The budget costs are also apparent from the example: in both cases, there is a gain to taxpayers of 0.78 and 0.66 units per unit transferred to A-producers and B-consumers. respectively. Expressing objectives and costs in this way allows for comparisons among policies: a straight transfer avoiding the price mechanism is likely to show up as having least economic cost, unless the government can alter its external terms of trade to its advantage. c) Incentive measures. A third type of derived measure makes use of the elements of the resource cost and value added calculations to examine incentives. Two examples are presented. The 'private incentive *index' (PI) is the proportionate difference between value added at domestic prices and resource cost at domestic prices. For product A, the 'profit' is 55 percent of net producer returns; for product 8 a 100 percent loss is indicated, i.e. : domestically valued opportunity cost is twice the domestic value added. The 'social incentive index' relates private value added to social resource cost. For product A, a 22 percent 'profit' is indicated, showing that private value added over- states social resource cost; for product B, the reverse is true. DRC(4), as discussed above, could also be thought of as an incentive index, in that it relates private resource cost to world-price value added. The advantage of such measures is that they give an indication of the extent to which private profitability can be manipulated to reflect social resource cost and private cost to accord with social value. This may allow for a broader view of policy alternatives through price manipulation than is implied by the 'pure' social cost/benefit measures such as DRC(5) and B(6). Page 41 5. Summary All these measures suffer from three problems. First, they are not well adapted to the evaluation of stabilisation policies; world price variability, in particular, will destabilise the measures. The aggregate variables, such as EC, PSV. CSV. can be summed over a period of years to indicate whether a particular policy has been primarily instrumental in offsetting such external instability. The answer lies in the sensitive interpretation of the measures as much as in the derivation of new indices per se. Secondly, all the indicators are open to the comment that small and large policy changes are being dealt with in the same analysis. For example, the NPC is a useful measure of the marginal protection costs of a tariff, since it indicates the difference between the marginal cost of protected domestic production (if this is approximately the domestic price) and the opportunity cost represented by the border price. But as the level of protection is reduced so is this distortion cost. The benefit from removing a tariff will depend on domestic supply and demand response, inter alia; hence not only does the marginal protection cost overstate the average cost per unit of protected production, but the ranking of policies by NPC may not reflect their actual relative costs. The values of DRC(5) will similarly show at the margin whether resources are being used efficiently. But one would assume that 'on average' resource costs are less than value added. In other words, the 'optimum' value of DRC(5) if total resource costs and value added are compared should be less than unity.!/ This has implications for 8(6). If marginal resource costs and net output valuations are used then this should be zero: a positive 'net benefit' is in fact a cost, an indication of under expansion in that sector. If averages are used then a positive value is a benefit to the economy but further benefit could be obtained 1/ The resource costs and value added calculated for the agricultural products were averages rather than marginal. Page 42 through expanding the sector. The original intention of the Balassa coefficients, and the two supplementary measures, seems to be as 'marginal' indicators.!/ As such they can be interpreted as giving a guide to 'small' policy and resource use changes; they may be less powerful in evaluating overall policy positions. By contrast the transfer measures are basically 'average' indicators in that they assess the impact and the cost of the policy as a whole. Indeed to get 'marginal' policy costs would require estimates of elasticity values - a procedure which would incidentally yield more complete estimates of the welfare cost of programmes than is inherent in the (RCW - YAW) calculation. Thirdly, there is a need to make a clear distinction between first and second-best welfare assumptions which shows up most clearly in relation to the question of opportunity cost. The alternative use of resources will depend on policies in other sectors. In the example of this paper, private resource cost in A depended on the level of protection of B-production. Consequently an evaluation of Policy A assuming policy B remains unchanged, will give quite a different answer from an evaluation of a general change in policy affecting both A and 8 sectors. Though these three points may be obvious, they do have a bearing on the use of the results. A policy maker will presumably wish to know the impact of his actions both in terms of resource allocation and income transfers. Sometimes he may be interested in small policy changes, at others he may wish for an overall evaluation. He may examine policies individually, or review a related set of measures such as the structure of relative agricultural prices. He may wish to know of the returns from investment in a particular enterprise, a decision which might be unrelated to the objectives of existing price support programmes 1/ See footnote at bottom of preceding page. Page 43 though nevertheless influenced by their impact. There are no universal indices which will answer all these 4uestions: the set of measures used in the study seem to give a useful starting point from which one can devise performance indices custom-made for any particular purpose. Page 44 D. EMPIRICAL PRACTICABILITY. 1. Salassa Coefficients - Agricultural Products As mentioned in the original report it was possible to estimate incentive coefficients and DRC; for the agricultural products of the social sector without exorbitant difficulty. This was because:- a) A detailed survey of costs and returns for the major agricultural products was available for a 10 year period. This survey covered 12-13 percent of the land cultivated by the social sector. b) It was possible to break down:the major items not inter- nationally traded into their traded and non traded components e.g. young cattle purchased for fattening. c) Other inputs not internationally traded e.g. electricity which would have been more difficult to break down, were an insignificant proportion of total inputs. d) It was possible to estimate directly the border prices of the selected products and to compare them with the domestic prices. However for the individual sector no survey of costs and returns was available. The value added had to be calculated from farm income survey data. As in most instatices inputs were not allocated to specific outputs a considerable amount of cost allocation was required. This was very time consuming and of necessity some of the allocations were arbitrary. An alternative approach would have been to have prepared some farm nodels, but this would have been equally time consuming and Page 45 it would have been difficult to incorporate the changes in input use and increasing farm mechanisation that have occurred over the ten years studied. The ESC has the advantage over the EPC of allowing for the effects of differential taxation and measures other than those reflected in producer and input prices. However not only was it difficult to obtain full information on subsidies and concessions, partly because they were often selective or ad hoc, but quantifying the effects on the value added of the types of measures used in Yugoslavia eg transportation rebates and retention quotas is very difficult and so was only partially attempted in this study.! Therefore where survey data of costs and returns is available and with world price data for the major agricultural outputs and inputs 1/ easily located, the calculation of the NPC, EPC an(LDRC is straight forward. On farms producing several products there is always the problem when using surveys of costs and returns as to what is the appropriate allocation between products of the so called 'fixed costs.' However, in the majority of LDCs there is little if any reliable data on costs and returns from specific products. It is therefore very much more difficult and time consuming to calculate the coefficients, and the results will inevitably be less accurate, so reducing to some extent their usefulness. 2. Balassa Coefficients - Manufacturing industries In two respects, special difficulties arise in calculating the Balassa coefficients for non-agricultural products. The first is that data on inputs into individual products are not generally available. Time series such as the annual statistics of input costs for agricultural products in Yugoslavia are 1/ There is the problem of ensuring that the quality of the domestic production is similar to that for which world price data is available. 2/ The only adjustment made in calculating the ESC from the EPC was for the subsidy on credit. Page 46 particularly rare. Usually it will be necessary to fall back on the national input-output tables, where they exist. In these, even the most detailed breakdown will be by industry rather than by product. This is no disadvantage in principle, where the object is to make inter-sector or inter-industry comparisons; or where (as in the present study) the main interest is in the agricultural products and the purpose of calculating Balassa coefficients in the non- agricultural sector is to provide some comparative yardsticks. However, making estimates for a whole industry is inevitably more complicated in practice than doing so for a single product. The second problem arises where the import duty and the export subsidy. If any, do not provide an adequate index of nominal protection (owing to quantitative restrictions or other factors including incompleteness of information about incentives). In such a case, the ideal solution is to determine the border price directly, as was done for the agricultural products in this study, and divide it into the comparable domestic price. For manufactured products, however, differences of specification and quality mean that it is usually difficult to make a trustworthy comparison of prices, using available data. The problem is of course multiplied when one is dealing with a whole industry rather than a single product. If time and staff are available, inquiries can be made of,traders and manufacturers to determine comparable c.i.f./f.o.b. and domestic prices for a sample of specific products. But in practice the analyst may have to fall back on using import duties, and any known export subsidy or tax rates, to estimate nominal protection. In a country like Yugoslavia, where a considerable proportion of trade in industrial goods is officially liberalised, this approach may be defended, if the object is Page 47 primarily to establish some approximate measures for comparison with agricultural products. Nevertheless, it will be appreciated that the coefficients for manufacturing industries are more uncertain and may well be less accurate. 3. Balassa Coefficients - Summary The difficulties mentioned above do not necessarily invalidate the Balassa coefficients. They do provide information on effective protection and comparative advantage for which other measures are not a complete substitute. Their usefulness, in relation to other measures, is discussed above. The question which the Bank and governments must answer is whether the information given by the coefficients is useful enough to justify perhaps several man-months in their calculation, bearing in mind also the difficulty of getting accurate results. 4. Josling Measures The methodology of calculating Producer Subsidy Equivalents is comparatively easy both to grasp and to put into practice. Moreover, all the data required are in any case needed for the estimation of EPCs, except for total output figures;Y while EPCs require a great deal of further information on input use and the protection of inputs. Consumer Subsidy Equivalents are also easy to calculate, although the data needed are largely different from those required for the Balassa EPCs and ESCs, but they provide a different kind of information. "Government sector cost" or "tax burden" figures are very easily calculated once PSVs and CSVs have been estimated, and they constitute a useful kind of information which is not given by the Balassa coefficients. I/ These are sometimes needed for calculating EPCs. If not they are not required either for the PSEs as opposed to the PSVs. Page 48 E. USE OF THE MEASURES IN EXPLAINING POLICY The Josling measures are probably somewhat easier to define and explain to officials and other non-economists than are Balassa's EPCs and ESCs. They do not involve the concept of "value added", neither do the complications over "border prices" arise except in the case of the price of the output. Moreover, the elements of the Josling subsidy equivalents lend themselves to simple tabulations, which are factually informative and make the method of calculation clear (see the tables in Section IV of the report of March 1977). However, the concepts and meaning of the Balassa coefficients should be explicable to the educated layman, although the methodology of calculation is more difficult to explain. As discussed above each of the Balassa and Josling measures gives a different and useful kind of information. Therefore, relative difficulty of explanation would not seem a sufficient reason for preferring one set of measures to the other. Page 49 F. USE OF THE MEASURES IN PROJECT APPRAISAL 1. Domestic Resource Cost Coefficient The DRC (equation S basis) and the net economic benefit measure may be useful in the preliminary "sifting" of projects. Projects may be ranked in order of net economic benefit of the products they will generate, or reverse order of DRCs, and a "cut-off" may be applied so as to reduce a long list to a more manageable number. Also a project for products with a DRC exceeding one, or a negative net economic benefit, is prima fade not worth pursuing. However, a DRC less than one is not necessarily an indication that output of the product should be expanded, when the coeffiqlents are national or regional averager-4as-4 the present study); such a prima facie conclusion can only be drawn, when the DRC refers to the marginal product. When it comes to the final selection of projects, of course, even the marginal DRC is not a sufficient criterion, and an examination of each individual scheme cannot be avoided. The DRC can also be calculated for individual projects. Balassa shows 1/ that the DRC, net economic benefit and other project selection criteria all give the same "yes or no" answer as to whether a project is economically worthwhile; but the ranking of projects can differ on the alternative criteria. Balassa uses a simple one-period model to demonstrate his point. In practice, of course, the costs and returns of nearly all projects occur over a number of years. The selection criteria, including 1/ Methodology of the Western Africa Study, page 42-48 and Appendix 5. 2/ This fact is demonstrated empirically in the present report, where the DRCs and net economic benefit measures of the products under study are compared. See Table 1. Page 50 the DRC, can be adapted to this case by expressing them in present value terms, and the statement in the preceding paragraph remains true. However it seems that the DRC would need to be redefined, in respect of the treatment of capital costs. Normally in project appraisal capital expenditures, including replacement expenditures, are counted in full as a once and for all cost in the year in which they occur; whereas in the DRC as previously defined depreciation and the opportunity cost of capital are used instead. The numerator of the DRC can be redefined as the shadow cost of labour and land, discounted to present value. Capital expenditure, converted to border prices, would be subtracted in full in calculating world market value added for the year when it occurs; the annual value added figures would be discounted to present value to arrive at the denominator of the DRC. The calculation of the DRC is therefore very similar to that of "Net Present Value", except that the elements are arranged differently in the final expression. The advantages and disadvantages of the different project selection criteria have been much discussed in the literature of economics, and it is not part of the present study to review the whole question yet again. As regards the DRC, it gives the same "yes or no" answer as the other criteria; the NPV, which is often recommended as the preferred criterion, has the advantage of showing the absolute economic benefit (or loss) from a project, whereas the DRC (or rather its inverse) is an indicator of the "rate of profit" of the project. It will be appreciated that the discussion in this report has been in terms of "economic efficiency" rather than "social accounting" However, the DRC can be expressed in terms of social accounting prices just as well as the other selection criteria, and the comments above would still apply. Page 52 G. USE OF THE MEASURES IN PRICING POLICY The Balassa incentive coefficients are designed to be directly relevant to pricing policy, and are useful for this purpose, if used with care. Section IV earlier discusses at length the meaning and relevance of these coefficients and the additional derived measures. Some points relating to pricing decisions are discussed below. An NPC below or above one indicates a "distortion" compared with the national free market situation. However, the EPC is a more complete indicator as to whether pricing policies give the producer a greater or smaller incentive than he would get in a free market.i/ From a strict economic efficiency point of view, it might be presumed that where the EPC exceeds one the product's price should be reduced, and vice versa. Such a conclusion, however, would be very hazardous: the DRC may sometimes suggest the opposite, for example that there would be an economic gain in increasing output by raising incentives, even when the EPC already exceeds one. The DRC itself must be used with caution, for reasons explained elsewhere in this report. The EPCs and, preferably, the ESCs may be useful in comparing different products or sectors, so as to assess the degree of discrimination between them resulting from government policies. For example, the fact that ESCs in Sector A generally exceed those in other sectors, so that A is already receiving "preferential treatment", may militate against a general price increase for A which had been contemplated for economic efficiency or other reasons. The indices of economic efficiency and discrimination are of course not the only measures relevant to the pricing decision; the authorities will wish to take account of personal income distribution in different sectors and expected world market trends, inter alia. 1/ On the assumption that the effective foreign exchange rate would be the same in the free market situation. Page 51 2. Other Balassa measures The protection and subsidy coefficients are of course less complete and direct measures of economic benefit than is the DRC. An EPC exceeding one indicates that there is a greater incentive to produce than there would be in a free market situation so that economically inefficient production may be encouraged. When the EPC is less than one, the opposite conclusions may be drawn, and there may be a presumption that output should be expanded. However, it would be dangerous to use the EPC as a guide to project selection since (as demonstrated by some of the estimates in the present study) it may give an indication quite contrary to the DRC, which is a true economic benefit measure. 3. Josling Measures Unlike the Balassa coefficients, the Josling measures are not designed as indicators of economic efficiency; and they cannot be used as project selection criteria. However, they may be of use for assessing the consequences of a project in term of transfers to and from producers and consumers and of cost to the government sector. For example, if a project is to increase output of a product subject to a maintained price, it should be possible to estimate quickly from the Josling measures the addition to total farm-sector income, and the additional cost to the government (in this example, assuming the price of the product is unchanged, consumers would be unaffected, except as tax-payers). However, especially where there is no fixed price guarantee, the Josling measures presented here must be used with great caution, since they are overall and average figures, and may give a misleading impression of the effect of the marginal increase in production. 1! / Professor Josling has however developed other "cost-objective" measures which reflect price elasticities, and may be calculated in marginal terms. (see Journal of Agricultural Economics May 1969). Page 53 The Josling measures do not purport to make an explicit comment on the economic efficiency implications of government policies. However, the percentage PSEs can be used in assessing discrimination between products and sectors;." unlike the EPCs and ESCs, they do not take full account of the effect of administrative action on input prices, but they have the advantage of isolating policies aimed at the particular product, and require less data and time to calculate. In addition, the Josling measures are useful in considering the income transfer effects of government policies and the implications for the exchequer. For example, take the case of product B, where current policies result in a large net subsidy to producers, a heavy price penalty for consumers, and substantial net government expenditure in supporting B: these factors would weigh against a policy for raising prices, although by themselves they would not necessarily be conclusive. It should be noted, however, that the Josling measures described in this study do not quantify the consequences of a potential change in policy; they show the overall effects of recent or past policies. To conclude, it may be said that the Balassa coefficients, the Josling measures and the derived measures described in this section can all be relevant to the pricing decision. Each provides a particular kind of information, different from the others, so that one should not think in terms of "preferring" some measures and rejecting others on general grounds. Which measures will be useful in each individual case will depend on the authorities' policy objectives and criteria and, sometimes, on availability of data and practicability of calculation. I/ The Josling measures were conceived in relation to agricultural products, but there is no reason in principle why they should not be calculated for products of other sectors. 2/ Professor Josling has however developed other measures which can be used for this purpose, provided that price elasticities can be determined: see Journal of Agricultural Economics, May 1969. Page 54 H. USE OF THE MEASURES BY SECTOR AND ECONOMIC MISSIONS As these missions are Concerned with agricultural policy, pricing issues and investment priorities a discussion of the usefulness and practicability of the Balassa, Josling and the derived measures to their work Is covered by sections IV C-G of this report. At-- 1 Table I SOCIAL SECTOR FRY WHEAT PRODUCTION - CALCULATION OF THE DOMESTIC RESOURCE COST COEFFICIENT (DRC: calculated as per equation 4) ND/ ha 1966/67 1967/68 1968/69 1969/70 1970/71 1971/72 1972/73 1973/74 1974/75 1975/76 CAPITAL 1/ MA3chinery 2510 2800 2455 2600 2775 3035 3440 3390 4780 5645 Breeding Livestock - - - -- - - - - - - Buildings - - - Working 1670 1747 1633 1681 1781 2027 2328 2792 3704 5090 Total c. 1803 4547 4088 4281 4556 5062 5768 6182 8484 1j715 Opportunity cost 9 9% 376 409 368 385 410 456 519 556 764 966 LABOUR 1/ farm skilled 548 565 506 541 612 754 858 1041 1490 1899 Farm unskilled 68 70 62 67 76 93 106 129 184 23S Non farm skilled 58 58 53 56 71 85 97 141 210 Total 674 693 626 661 744 918 1049 1267 1815 2344 LAND 1/ (40) (935) (1287) (743) (1232) (550) (575) (1151) 1706 1151 Value Added pot 1010 167 (293) 303 (78) 824 993 672 4285 4461 Value Added Border Prices 1904' 1493 1287 785 878 2171 4397 11382 11089 5789 Domestic Resource Cost Coefficient DRC 0.53 0.11 - 0.39 - 0.38 0.23 0.06 0.39 0.77 11 Domestic Prices IEX Table 2 SOCIAL SECTOR FRY MAIZE PRODUCTION - CALCULATION OF THE DOMESTIC RESOURCE COST COEFFICIENT (DRC: calculated as per equation 4) ND/ha 1966/67 1967/68 1968/69 1969/70 1970/71 1971/72 1972/73 1973/74 1974/75 1975/76 -CAPITAL J/ Machinery 3370 3860 4075 4745 5720 4795 6525 6850 10495 10715 Breeding Livestock - - - - - - - - Buildings - - - - - - - - - - Working 2064 2181 2117 2285 2642 2750 3585 4065 5830 7404 Total 5434 6041 6192 7030 8362 7545 10110 10915 163-r, 1.11 Opportunity cost @ 9% 489 544 557 633 753 679 910 982 1469 1631 LABOUR 1/ Farm skilled 1080 1031 1006 1102 1311 1399 1857 2102 2888 3615 Farm unskilled 133 127 124 136 162 173 229 260 357 447 Non farm skilled 44 51 42 35 37 58 .75 67 95 118 Total 1257 1209 1172 1273 1510 1630 2161 2429 3340 4180 LAND 1/ 995 489 601 447 (139) 1779 1207 1473 5054 1325 Value Added 1/(Opportunity 2741 2242 2330 2353 2124 4088 4278 4884 9863 7136 Cost) Value Added Border Prices 2949 1397 743 1967 3016 1865 4836 8353 8997 76F1 DomEstic Resource Cost Coefficient DRC 0.93 1.60 3.14 1.20 0.70 2.19 0.88 0.58 1.10 0.93 !/ Dowestic Prices 11%14EX I lable 3 S3CIAL SECTOR FRY BEEF PRODUCTION - CALCULATION OF THE DOMESTIC RESOURCE COST COEFFICIENT (DRC: calculated as per equation 4) ND/ 100 kg LW 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 CAPITAL 1/ Machlnery 175 170 175 165 180 200 285 385. 435 455 Breeding Livestock 235 250 240 235 265 285 375 580 615 635 Buildings 78 198 265 234 242 237 439 431 657 739 1orking 428 427 427 425 499 617 795 1017 1101 1226 Total 916 1045 1107 1059 1186 1339 1894 2413 2808 3055 Opportunity cost @ 9% 82 94 100 95 107 121 170 217 253- 275 LABOUR I/ Farm skilled 116 115 114 111 123 141 186 252 298 325 Farm unskilled 14 14 14 14 15 17 23 31 37 40 Non farm skilled 4 7 5 3 6 6 10 11 17 21 Total 134 136 133 128 144 164 219 294 352 I6 LAND 1/ 14 C7) (10) (4) (21) 18 9 5 101 37 Value Added 1/(Opportunity 230 223 223 219 230 303 398 516 706 698 Cost) Value Added Border Prices (80) (54) (192) (54) 6 197 263 (96) (254) (523) Domestic Resource Cost Coefficient DRC - - - - 38.33 1.54 1.51 - - - 11 Domestic Prices A""YEX I lable 4 SOCIAL SECTOR FRY PIG MEAT PRODUCTION - CALCULATION OF THE DOMESTIC RESOURCE COST COEFFICIENT (DRC: calculated as per equation 4) ND/100 Kg LW 1966 1967 1968 1969 1970 1971 - 1972 1973 1974 1975 CAPITAL 1/ Machinery 50 55 55 60 /5 75 95 100 110 100 Breeding Livestock 40 40 40 40 45 50 60 80 95 105 Buildings 130 194 207 243 209 197 438 402 501 521 Working 367 379 361 375 427 SIS 566 794 952 1059 Total 587 668 663 718 756 837 1159 1376 1658 1785 Opportunity cost @ 9% 53 60 60 65 68 75 104 124 149 161 LABOUR 1/ Farm skilled 62 67 66 70 76 91 114 136 167 195 Farm unskilled 8 8 8 9 9 11 14 17 21 24 Non farm skilled 6 8 7 7 7 8 .11 13 17 16 Total 76 83 81 86 92 110 139 166 205 235 LAND I/ - - - - - - - - Value Added ]/(Opportunity 129 143 141 151 160 185 243 290 354 396 Cost) Value Added Border Prices 194 234 294 433 269 137 (78) 339 209 100 Domestic Resource Cost Coefficient DRC 0.66 0.61 0.48 0.35 0.59 1.35 - 0.86 1.69 3.96 IL Domestic Prices .X I SERBIA PROPER Table 5 INDIVIDUAL SECTOR WHEAT PRODUCTION - CALCULATION OF THE DOMESTIC RESOURCE COST COEFFICIENT (DRC: calculated as per equation 4) ND/Farm 1966/67 1967/68 1968/69 1969/70 1970/71 1971/72 .1972/73 1973/74 1974/75 CAPITAL 1/ Machinery 710 680 720 710 780 950 1110 1280 1690 Breeding Livestock - - - - - - - - - Buildings - - - - - - - - - Working 518 478 550 592 630 834 1055 1 120 1813 Total 1228 1158 1270 1302 1410 1784 2165 2560 3273 Opportunity cost 0 9% 111 104 114 117 127 161 195 230 295 LABOUR 1/ Farm 2/ 108 110 107 114 122 146 195 283 315 Non farm 35 33 38 38 43 59 76 85 108 Total 143 143 145 152 165 205 271 368 423 LAND I/ 535 285 335 545 483 1266 1153 876 1757 Value Added 1/(Opportunity 789 532 594 814 775 1632 1619 1474 2475 Cost) Value Added Border Prices 1236 1106 681 339 746 845 2272 4685 2982 Domestic Resource Cost Coefficient DRC 0.64 0.48 0.87 2.40 1.04 1.93 0.71 0.31 0.83 1/ Domestic Prices 2/ Opportunity Cost 70% market wage ANNFX 1 SERBIA PROPER INDIVIfUAL SECTOR MAIZE PRODUCTION - CALCULATION OF THE DOMESTIC RESOURCE COST COEFFICIENT (DRC: calculated as per equation 4) NO/Farm 1966/67 1967/68 1968/69 1969/70 1970/71 1971/72 1972/73 1973/74 1974/75 CAPITAL 1/ Machinery 910 880 940 950 1150 1380 1560 1910 2470 Breeding Livestock - - - - - - - - - Buildings - - - - - - - - - Working 695 718 727 775 868 1096 1406 17 2A Total 1605 1598 1667 1725 2018 2476 2966 3788 489ti Opportunity cost @ 9% 144 144 150 155 182 223 267 341 441 LABOUR 1/ Farm 2/ 379 387 376 392 424 512 684 989 1333 Non farm 45 44 49 51 64 84 106 126 159 Total 424 431 425 443 488 596 790 1115 1492 LAND 1/ 868 765 387 191 77 891 675 87 1722 Value Added 1/(Opportunity 1436 1340 962 789 747 1710 1732 1543 3655 Cost) ____ ____ ________ Value Added Border Prices 1727 1197 1070 1540 1856 1790 3499 6622 7001 Domestic Resource Cost Coefficient DRC 0.83 1.12 0.90 0.51 0.40 0.96 0.49 0.23 0.52 Domestic Prices 2/ Opportunity Cost 70% market wage .AN"r" 1 Table 7 SERBIA PROPER INDIVIDUAL SECTOR BEEF PRODUCTION - CALCULATION OF THE DOMESTIC RESOURCE COST COEFFICIENT (DRC: calculated as per equation 4) ND/Farm 1966 1967 . 1968' 1969 1970 1971 1972 - 1973 1974. CAPITAL 1/ M4achinery 125 145 160 185 275 320 365 555 152 Breeding Livestock - - - - - - - - - Buildings 670 680 660 620 770 840 830, 1140 1590 Working 760 784 814 858 1063 1467 1842 2437 3230 Total 1555 1609 1634 1663 2108 2627 3037 4132 4972 Opportunity cost 9 9% 140 145 147 150 190 236 273 372 447 LABOUR 1/ Farm 2/ 284 302 298 314 366 470 610 810 1089 Non farm 38 38 39 38 51 56 62 87 118 Total 322 340 337 352 417 526 672 897 1207 LAND 1/ 133 122 87 109 111 392 336 248 832 Value Added 1/(Opportunity 595 607 571 611 718 1154 1281 1517 2486 Cost) -- - _1 Value Added Border Prices 93 202 (71) 16 121 596 694 116 (366) Domestic Resource Cost Coefficient DRC 6.40 3.00 - 38.19 5.93 1.94 1.85 13.08 - l Domestic Prices 2/ Opportunity Cost 70% market wage ANN-1 I Table 8 SERBIA PROPER INDIVIDUAL SECTOR PIG PRODUCTION - CALCULATION OF THE DOMESTIC RESOURCE COST COEFFICIENT (DRC: calculated as per equation 4) ND/Farm 1966 1967 1968 - 1969 1970 1971 1972 1973 1974 CAPITAL Machinery 85 115 125 180 235 230 325 365 365 Breeding Livestock 105 80 90 120 140. 155 205 290 315 Buildings 300 370 430 570 570 590 1270 1640 1620 Working 1041 1093 1153 1404 1614 2088 2681 3461 4229 Total 1531 1658 1798 2274 2559 3063 4481 5756 6529 Opportunity cost 0 9% 138 149 162 '205 230 276 403 518 588 LABOUR If Farm 2/ 429 459 473 559 617 799 1040 1285 1729 Non farm 13 16 16 21 22 25 38 47 56 Total 442 475 489 580 639 824 1078 1332 1785 LAUD ! Value Added 1/(Opportunity 580 624 651 785 869 1100 1481 1850 2373 Cost) ____ ________ Value Added Border Prices 527 728 811 1043 686 320 (480) 524 821 Domestic Resource Cost Coefficient DRC 1.10 0.86 0.80 0.75 1.27 3.44 - 3.53 2.89 1 Domestic Prices 2/ Opportunity Cost 70% market wage NNFv ' Table t S.A.P. VOYVODINA INDIVIDUAL SECTOR WHEAT PRODUCTION - CALCULATION OF THE DOMESTIC RESOURCE COST COEFFICIENT (DRC: calculated as per equation 4) ND/ Farm 1966/67 1967/68 1968/69 1969/70 1970/71 1971/72 1972/73 1973/74 1974/75 CAPITAL 1/ Machinery 960 1030 1300 1640 1760 1830 2230 2440 3070 Breeding Livestock - - - - - - Buildings - - - - - Working 482 475 726 831 892 1039 1252 1231 1967 Total 1442 1505 2026 2471 2642 2869 3482 3671 5037 Opportunity cost 0 9% 130 135 182 222 238 258 313 330 453 LABOUR 1/ Farm 2/ 78 92 105 115 116 127 163 160 228 Non farm 53 53 77 116 114 131 173 192 214 Total 131 145 182 231 230 258 336 352 442 LAND 1/ 1066 1135 .1038 1420 1764 2915 2688 1584 4663 Value Added 1/(Opportunity 1327 1415 1402 1873 2232 3431 3337 2266 5558 Cost) Value Added Border Prices 2005 2720 1741 1721 1970 3228 5486 13106 9546 Domrestic Resource Cost Doefficient DRC 0.66 0.52 0.81 1.09 1.13 1.06 0.61 0.17 0.58 l_ Domestic Prices 2/ Opportunity Cost 70% market wage S.A.P. VOYVODINA Table 10 INDIVIDUAL SECTOR MAIZE PRODUCTION - CALCULATION OF THE DOMESTIC RESOURCE COST COEFFICIENT (DRC: calculated as per equation 4) ND/Farm_ 1966/67 1967/68 1968/69 1969/70 1970/71 1971/72 1972/73 1973/74 1974/75 CAPITAL I/ Machinery 2600 2560 2810 3230 .3620 4690 5430 6770 7570 Breeding Livestock - Buildings Working 1379 1560 1676 1692 1867 2624 3162 3513 4858 Total 3979 4120 4486 4922 5487 7314 8592 10283 12428 Opportunity cost @ 9% 358 371 404 443 494 658 773 925 1119 LAOUR 1/ Farm 2/ 750 806 793 798 836 1139 1387 1549 1968 Non farm 146 133 165 230 236 335 420 531 526 Total 896 939 958 1028 1072 1474 1807 2080 2494 LAND 1/ 2195 3497 1990 2210 1130 3909 2779 2892 7457 Value Added 1/(Opportunity 3449 4807 3352 3681 2696 6041 5359 5897 11070 Cost) I _ -- - Value Added Border Prices 6338 5679 4492 5705 7938 6898 12320 20953 20829 Domestic Resource Cost 0.54 0.85 0.75 0.65 0.34 0.88 0.43 0.28 0.53 Coefficient DRC - - 1 Dorestic Prices 2/ Opportunity Cost 70% market wage ANV^ I Table 11 S.A.P. VOYVODINA INDIVIDUAL SECTOR BEEF PRODUCTION - CALCULATION OF THE DOMESTIC RESOURCE COST COEFFICIENT (DRC: calculated as per equation 4) NO/Farm 1966 1967 1968. 1969 1970 1971 .1972 1973 1974 CAPITAL 1/ Machfnery 220 230 180 230 320 300 355 51 500 Breeding Livestock - - - - - - - - Buildings 510 530 530 530 610 630 660 1030 1290 Working 676 752 714 815 1008 1401 1585 2254 3099 Total 1406 1512 1424 1575 1938 2331 2600 3799 4889 Opportunity cost 0 9 127 136 128 142 174 210 234 342 440 LABOUR 1/ Fari 2/ 235 284 272 296 346 470 579 826 1089 Non farm 41 46 40 48 54 60 70 105 103 Total 276 330 312 344 400 530 649 931 1192 LAND 1/ 258 292 169 239 254 470 383 355 B54 Value Added 1/(Opportunity 661 758 609 725 828 1210 1266 1628 2486 Cost) - - - - Value Added Border Prices (6) 162 (18) 17 128 539 1007 750 (77) Domestic Resource Cost Coefficient DRC - 4.68 - 42.65 6.47 2.24 1.26 2.17 - I_ Dowestic Prices 2/ Opportunity Cost 70% market wage AWt Table 12 S.A.P. VOYVODINA INDIVIDUAL SECTOR PIG PRODUCTION - CALCULATION OF THE DOMESTIC RESOURCE COST COEFFICIENT (ORC: calculated as per equation 4) ND/Pam 1966. 1967 1968 1969 ' 1970 1971 1972 1973 1974 CAPITAL / Machinery 270 275 245 340 575 590 705 765 805 Breeding Livestock 220 200 175 230 345 390 445 610 695 Bufldings 780 890 840 1080 1380 1490 2750 3440 3570 orking 2668 2675 2320 2750 4090 5439 5935 7389- 9780 Total 3938 4040 3580 4400 6390 7909 9835 12204 14850 Opportunity cost 9 9 354 364 322 396 575 712 885 1098. 1337 LABOUR 1/ Farm - 2/ 1072 1146 981 1148 1676 2175 2388 2834 4280 Non farm 34 40 31 40 54 63 82 99 124 Total 1106 1186 1012 1188 1730 2238 2470 2933 4404 LAN) 1/ - - - - - - - - - Value Added 1/(Opportunity 1460 1550 1334 1584 2305 2950 3355 4031 5741 Cost) Value Added Border Prices 1177 1893 1789 2340 2348 1248 (703) 1926 3442 Domestic Resource Cost Coefficient DRC 1.24 0.82 0.75 0.68 0.98 2.36 - 2.09 1.67 1/ Doriestic Prices 2/ Opportunity Cost 70% market wage 'VEX Table 13 S.A.P. KOSOVO INDIVIDUAL SECTOR WHEAT PRODUCTION - CALCULATION OF THE DOMESTIC RESOURCE COST COEFFICIENT (DRC: calculated as per equation 4) ND/Farm 1966/67 1967/68 1968/69 1969/70 1970/71 1971/72 .1972/73 1973/74 1974/75 CAPITAL I/ Machinery 410 430 450 510 640 850 1050 1250 1420 Breeding Livestock - - Buildings - dorkIng 480 428 527 584 717 991 1213 1408 1624 Total 890 858 977 1094 , 1357 1841 2263 2658 3044 Opportunity cost 0 9% 80 77 88 98 122 166 204 239 274 LABOUR 1/ Farm 2/ 105 109 104 113 119 158 186 214 258 Non farm 25 28 31 37 48 68 93 110 119 Total 130 137 135 150 167 226 279 324 377 LAND 1/ 247 370 232 292 623 785 888 372 1113 Value Added 1/(Opportunity 457 584 455 540 912 1177 1371 935 1764 Cost) Value Added Border Prices 844 1168 668 580 1225 1100 2417 4630 2634 Domestic Resource Cost Coefficient DRC 0.54 0.50 0.68 0.93 0.74 1.07 0.57 0.20 0.67 { Domestic Prices 2/ Opportunity Cost 70% market wage TEX S.A.P. KOSOVO Table 14 IhDIVIDUAL SECTOR MAIZE PRODUCTION - CALCULATION OF THE DOMESTIC RESOURCE COST COEFFICIENT (DRC: calculated as per equation 4) ND/ Farm 1966/67 1967/68 1968/69 1969/70 1970/71 1971/72 .1972/73 1973/74 1974/75 CAPITAL 1/ Machinery 440 430 490 540 650 820 990 1270 1480 Breeding Livestock - - - - - - - - - Buildings - - - - - - - - Working 564 540 620 693 739 968 1136 1425 1628 Total 1004 970 1110 1233 1389 1788 2126 2695 3108 Opportunity cost 0 9% 90 87 100 . 111 125 161 191 243 280 LABOUR 1/ Farm 2/ 389 372 398 427 424 533 620 -763 935 Non farm 26 27 33 40 49 66 89 111 122 Total 415 399 431 467 473 599 709 874 1057 LAND 1/ 537 822 429 594 826 875 673 235 199 Value Added 1/(Opportunity 1042 1308 960 1172 1424 1635 1573 1352 1536 Cost)_____ Value Added Border Prices 918 1015 735 879 1503 958 1883 3537 3763 Domestic Resource Cost Coefficient DRC 1.14 1.29 1.31 1.33 0.95 1.71 0.84 0.38 0.4 1/ Dormstic Prices 2/ Opportunity Cost 70% market wage .4NN-" ' S.A.P. KOSOVO Table 15 INDIVIDUAL SECTOR BEEF PRODUCTION - CALCULATION OF THE DOMESTIC RESOURCE COST COEFFICIENT (DRC: calculated as per equation 4) ND/Farm 1966 1967 .. 1968 1969 1970 1971 .1972 1973 1974 CAPITAL 1/ Machinery 10 10 15 15 25 20 45 s0 135 Breeding Livestock - - - - Buildings 490 490 380 470 560 610 660 960 1370 Working 379 260 437 605 657 939 1268 1582 2259 Total 879 760 832 1090 1242 1569 1973 2622 3764 Opportunity cost 9 9% 79 68 75 98 112 141 178 236 339 LABOUR 1/ Farm 2/ 139 178 176 225 216 317 425 529 851 Non farm 25 25 20 25 29 32 36 54 78 Total, 164 203 196 250 245 349 461 583 929 LAND 1/ 11 24 13 21 42 32 54 55 92 Value Added 1/(Opportunity 254 295 284 369 399 522 693 874 1360 Cost) _ Value Added Border Prices 37 141 37 57 18 388 507 50 194 Domestic Resource Cost Coefficient DRC 6.86 2.09 7.68 6.47 22.17 1.35 1.37 17.48 7. 01 1! Domestic Prices 2/ Opportunity Cost 70% market wage NNE S.A.P. KOSOVO Table 16 INDIVIDUAL SECTOR PIG PRODUCTION - CALCULATION OF THE DOMESTIC RESOURCE COST COEFFICIENT (ORC: calculated as per equation 4) ND/Farm 1966 1967 1968 1969 1970 1971 1972 1973 1974 CAPITAL If Machinery 40 45 60 80 95 90 115 130 160 Breeding Livestock 30 35 45 55 60 60 - 70 105 136 Buildings 110 150 200 250 230 230 450 590 690 Working 362 408 515 608 664 801 937 1236 1832 Total 542 638 820 993 1049 lIi 1577 2061 PR17 Opportunity cost 9 9% 49 57 74 89 94 106 141 185 254 LABOUR I/ Farm / 139 146 192 237 260 295 362 454 738 Ron frm 5 7 7 9 9 10 13 17 25 Total 144 153 199 246 269 305 375 471 763 LAID I - - - - - - - - Value Added 1/(Opportunity 193 210 273 335 363 411 516 656 1017 Cost) 13 20 23 356 411 516 65 1017 Value Added Border Prices 127 130 245 421 311 75 (177) 156 302 Domestic Resource Cost Coefficient DRC 1.52 1.62 1.11 0.80 1.17 5.48 - 4.21 3.37 1/ Domestic Prices 2/ Opportunity Cost 70% market wage 'ANt 1 Table 17 SOCIAL SECTOR FRY WHEAT PRODUCTION - CALCULATION OF THE NET ECONOMIC BENEFIT (Equation 6) ND 1966/67 1967/68 1968/69 1969/70 1970/71 1971/72 1972/73 1973/74 1974/75 1975/76 Value Added from Wheat per ha 1/ 1904 1493 1287 785 878 2171 4397 11382 11089 5789 Less opportunity cost of land labour & capital 1/ 2214 730 (12) 1099 1868 941 3227 6302 6485 4992 I.et Economic Benefit from Wheat per ha (310) 763 1299 (314) (990) 1230 1170 5080 4604 797 Yield of wheat tons per ha 4.47 4.35 3.95 3.97 3.13 4.46 4.20 4.57 5.45 4.30 Ret Economic Benefit per ton (69) 175 329 (79) (316) 276 219 1112 845 185 MAIZE PRODUCTION - CALCULATION OF THE NET ECONOMIC BENEFIT (Equation 6) ND 1966/67 1967/68 1968/69 1969/70 1970/71 1971/72 1972/73 1973/74 1974/75 1975/76 Value Added from Maize per ha 2/ 2949 1397 743 1967 3016 1865 4836 8353 8997 7651 Less opportunity cost of land labour & capital 2/ 2639 2160 2042 1653 2026 3095 6006 13433 13601 8448 liet Economic Benefit from I Maize per ha 310 (763) (1299) 314 990 (1230) (1170) (5080) (4604) (797) Yield of Maize tons per ha 6.36 5.64 5.03 6.59 6.21 5.65 6.47 5.80 6.63 7.15 Net Economic Benefit per ton 49 (135) (258) 48 159 (218) (181) (876) (694) (1l) 1/ Border prices fror i 5 Ane II Volume 2 Yugoslavia Agricultural Prices and Subsidies Case Study 2/ " " 10 " " " " " " " S ANL... I SOCIAL SECTOR FRY BEEF PRODUCTION - CALCULATION OF THE NET ECONOMIC BENEFIT (Equation 6) ND Table 18 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 Value added from Beef per 100 Kg LW 1/ (80) (54) (192) (54) 6 197 263 (96) (254) (523) Less opportunity cost of land, labour & capital 1/ 252 235 218 217 262 285 481 847 839 715 Net Economic Benefit from Beef per 100 Kg LW (332) (289) (410) (271) (256) (88) (218) (943) (1093) (1238) Conversion factor 3/ 19.3 19.3 19.3 19.3 19.3 19.3 19.3 19.3 19.3 19.3 Net Economic Benefit from Beef _ _ I per ton OW . (6408) (5578) (7913) (5230) (4941) (1698) (4207) (18200) (21095) (23893) PIG MEAT PRODUCTION - CALCULATION OF THE NET ECONOMIC BENEFIT (Equation 6) ND. 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 vlue Added from Pig meat per 100 Kg LW 2/ 194 234 294 433 269 137 (78) 339 209 100 Less opportunity cost of land, labour & capital 2/ 136 148 142 148 166 183 256 371 391 394 Net Economic Benefit from Pig meat per 100 Kg LW 58 86 152 285 103 (46) (334) (32) (182) (294) Conversion factor 4/ 13.4 13.4 13.4 13.4 13.4 13.4 13.4 13.4 13.4 13.4 Net Economic Benefit from Pig -- reat per ton DW 777 1152 2037 3819 1380 (616) (4476) (429) (2439) (3940) 1/ 3(rder prices from Table 13 Annex II Volume 2 Yugoslavia Agricultural Prices and Subsidies Case Study. 2/ " " 16 a " " 3/ x 19 3 (1.93 FAO conversion factor x 10) 4/ x 13.4 (1.34 FAO conversion factor x 10) ANN I Table 19 INDIVIDUAL SECTOR SERBIA PROPER WHEAT PRODUCTION CALCULATION OF THE NET ECONOMIC BENEFIT (Equation 6) ND 1966/67 1967/68 1968/69 1969/70 1970/71 1971/72 1972/73 1973/74 1974/75 Value Added from Wheat per farm 1/ 1236 1106 681 339 746 845 2272 4685 2982 Less Opportunity Cost of land, 1/ 1136 712 639 973 1083 1045 2187 3940 4096 labour & capital Net Economic Benefit from Wheat per farm 100 394 42 (634) (337) 1200) 85 745 (lll4) Hectares of Wheat per farm 1.04 1.00 1.02 0.98 0.87 0.89 0.90 0.83 0.81 Net Economic Benefit per ha of Wheat 96 394 41 (647) (387) (225) 94 898 (1375) Yield of wheat tons per ha 2.59 2.47 2.08 1.96 1.99 2.71 2.75 2.78 3.31 'let Economic Benefit per ton 37 160 20 (330) (194) (83) 34 323 (415) MAIZE PRODUCTION CALCULATION OF THE NET ECONOMI1C BENEFIT (Equation 6) No Value Added from Maize per farm 2/ 1727 1197 1070 1540 1856 1790 3499 6622 7001 Opportunity Cost of '.and, 2/ 1856 1714 1124 692 1355 1504 3617 7736 5364 labour & Capital [ot Economic Benefit from Maize (129) (517) (54) 848 501 286 (118) (1114) 1637 per farm I Hectares of Maize per fa-rm 1.35 1.31 1.32 1.31 1.29 1.28 1.26 1.24 1.19 :!,,t Economic Benefit per ha of Maize (96) (395) (41) 647 388 223 (94) (898) 1376 Yield of Maize tons per ha 2.67 2.64 2.60 3.18 2.82 3.42 3.62 3.96 4.32 let Economic Benefit per ton (36) (150) (16) 203 138 65 (26) (227) 1 319 1/ Border Prices from Table 18 Annex II Volume 2 Yugoslavia AgricultLral Prices and Subsidies Case Study 2/ 20 " Al - I Table 20 INDIVInUAL SECTOR SERBIA PROPER - BEEF PRODUCTION CALCULATION OF THE NET ECONOMIC BENEFIT (Equation 6) ND N.D. 1966 1967 1968 1969 1970 1971 1972 1973 1974 Value Added from Beef per farm 1/ 93 202 (71) 16 121 596 694 116 (366) Less Opportunity Cost of land, 696 691 600 626 900 989 1741 3572 3471 labour & capital 1I flet Economic Benefit from Beef per farm (603) (489) (671) (610) (779) (393) (1047) (3456) (3837) Kgs 0W Beef produced per farm 100.13 104.62 102.13 100.41 116.12 114.96 123.29 137.28 141.67 Net Economic Benefit per ton (6022) (4674) (6570) (6075) (6709) (3419) (8492) (25175)1(27084) - PIG HEAT PRODUCTION CALCULATION OF THE NET ECONOMIC BENEFIT (Equation 6) ND Value Added from Pig meat per farm 2/ 527 728 811 1043 -686 320 (480) 524 821 Less Opportunity Cost of land, 613 653 663 789 914 1080 1596 2417 2657 labour & capital 2/ Ukt Economic Benefit from Pig (86) 75 148 254 (228) (760) (2076) (1893) (1836) r;at per farm Kgs DW Pig Meat produced per farm 171.80 178.73 184.38 202.81 220.97 221.55 238.11 246.09 254.37 let Economic Benefit per ton (501) 420 803 1252 (1032) (3430) (8719) (7692) (7218) 1/ Border Prices from Table 23 Annex II Volume 2 Yugoslavia Agricultural Prices and Subsidies Case Study 2/ ,,s 0 o 26 It9 , 1 0 A. It * . ""EX - Table 21 INDIVIDUAL SECTOR SAP VOYVODINA WHEAT PRODUCTION CALCULATION OF THE NET ECONOMIC BENEFIT (Equation 6) ND 1966/67 1967/68 1968/69 1969/70 1970/71 1971/72 1972/73 1973/74 1974/75 Value Added from Wheat per farm 1/ 2005 2720 1741 1721 1970 3228 5486 13106 9546 Less Opportunity Cost of land., 1/ labour & capital 2104 2023 1813 892 3531 2389 4619 7036 7801 Net Economic Benefit from Wheat per farm (99) 697 (72) (871) (1561) 839 867 6070 1745 Hectares ofWheat per,farm 0.76 0.87 0.99 1.01 1.00 0.85 0.90 0.74 0.81 [let Economic Benefit per ha of Wheat (130) 801 (73) (862) (1561) 987 963 8203 2154 Yield of wheat tons per ha 2.77 3,19 2.66 3.03 2.44 3.50 3.06 3.64 4.06 Net Economic Benefit per ton (47) 251 (27) (285) (640) 282 315 2253 531 MAIZE PRODUCTION CALCULATION OF THE NET ECONOMIC BENEFIT (Equation 6) NO Value Added from Maize per farm 2/ 6338 5679 4492 5705 7938 6898 12320 20953 20829 Odoortunity Cost of Land, 2/ labour & Capital 6060 7418 4337 3980 4706 9049 14429 37768 25139 i-st Economic Benefit from Maize per farna 278 (1739) 155 1725 3232 21511 (21091 16815) (4110). lectares of Maize per farm 2.08 2.17 2.13 2.00 2.07 2.18 2.19 2.05 I,00 Htet Economic Benefit per ha of Maize 134 (801) 73 863 1561 (987) (963) (8202) (2155) Yield of F*jize tons per ha 4.66 4.69 4.34 5.17 5.01 15.05 5.14 5.56 5.57 Ilet Economic Benefit per ton 29 (171) 17 167 312 195) (187) (1475) ( 387) I/ Border Prices from Table 28 Annex II Volume 2 Yugoslavia Agricultural Prices and Subsidies Case Study 2/ . 1 is 4 30 11 81 so mm a . Is 1. a Table 22 INDIVIDUAL SECTOR SAP VOYVODINA - BEEF PRODUCTION CALCULATION OF THE NET ECONOMIC BENEFIT (Equation 6) !D 1966 1967 1968 1969 1970 1971 1972 1973 1974 Value Added from Beef per farm 1/ (6) 162 (18) 17 128 539 1007 750 (77) Less Opportunity Cost of land, labour & capital 1/ 920 965 687 780 1108 1208 1832 4187 3536 Net Economic Benefit from Beef per farm (926) (803) -(705) (763) (980) (669) (825) (3437) (3613) Kgs OW Beef produced per farm 83.33 97.97 93.25 95.13 109.87 114.81 117.47 139.47 141.58 Net Economic Benefit per ton (11112) (8196) (7560) (8021) (8920) (5827) (7023) (24643) (25519) - PIG HEAT PRODUCTION CALCULATION OF THE NET ECONOMIC BENEFIT (Equation 6) L.D Value Added from Pig meat per firm 2/ 1177 1893 1789 2340 2348 1248 (703) 1926 3442 Less Opportunity Cost of land, labour & capital 2/ 1537 1622 1360 1594 2426 2897 3619 5572 6438 Net Economic Benefit from Pig meat per farm (360) 271 429 746 (78) (1649) (4322) (3646) (2996) Kgs DW Pig Meat produced per farm 430.00 446.86 381.17 415.93 599.90 602.64 545.66 542.65 628.81 Het Ecenonic Benefit per ton (837) 606 1125 1794 (130) (2736) (7921) (6719) 47 1/ Border Prices from Table33 Annex*II Volume 2 Yugoslavia Agricultural Prices and Subsidies Case Study 2/ is I , ,, 36 ,, , is it is . Ii NEX . Table 23 INDIVIDUAL SECTOR SAP KOSOVO - WHEAT PRODUCTION CALCULATION OF THE NET ECONOMIC BENEFIT (Equation 6) ND 1966/67 1967/68 1968/69 1969/70 1970/71 1971/72 1972/73 1973/74 1974/75 Value Added from Wheat per farm 1/ 844 1168 668 580 1225 1100 2417 4630 2634 Less Opportunity Cost of land, 1/ labour & capital 587 742 414 527 1161 605 1478 2725 2923 lat Economic Benefit 'from Wheat per farm 257 426 254 53 64 495 939 1905 (289) Hectares of Wheat per farm 1.02 1.02 0.98 0.99 1.03 1.06 1.03 0.99 0.92 Net Economic Benefit per ha of Wheat 252 418 259 54 62 467 912 1924 (314) Yield of wheat tons per ha 1.83 .2.17 1.81 2.14 2.40 2.55 2.65 2.47 2.68 Economic Benefit per ton 138 193 143 25 26 183 344 779 (117) MAIZE PRODUCTION CALCULATION OF THE NET ECONCIUC BENEFIT (Equation 6) NO Value Added from Maize per farm 2/ 918 1015 735 879 1503 958 1883 3537 3763 ; ,oportunity Cost of Land, 2/ lajour & Capital 1191 1433 1013 936 1568 1434 2776 5240 3452 ':ot Economic Benefit from Maize per farm (273) (418) (278) (57) (65) (476) (893) (1703) 311 Hectares of Maize per farm 1.08 1.00 1.07 1.07 1.05 1.02 0.98 1.01 0.j95 i,et Economic Benefit per ha of Maize (253) (418) (260) (53) (62) (467) (911) 1686 327 YielJ of Maize tons per ha 1.72 2.09 1.88 2.16 2.39 2.33 2.60 2.70 2.48 at Ecoroic Benefit per ton (147) (200) (138) (25) (26) (200) (350) (624) 132 / Bordyr Prices from Table 38 Annex II Volume 2 Yugoslavia Agricultural Prices and Subsidies Case Study 2/ 40 " " " 5 5 INE) Table 24 INDIVIDUAL SECTOR SAP KOSOVO - BEEF PRODUCTION CALCULATION OF THE NET ECCNOMIC BENEFIT (Equation 6) NO 1966 1967 1968 1969 1970 1971 1972 1973 1974 Value Added from Beef per farm 1/ 37 141 37 57 18 388 507 so 194 Less Opportunity Cost of land, labour & capital 1/ 275 314 287 365 435 497 798 1385 1714 Net Economic Benefit from Beef per farm (238) (173) (250) (308) (417) (109) (291) (1335) (1520) Kgs DW Beef produced per farm 49.80 60.88 60.02 72.63 68.25 77.50 85.54 89.95 110.34 Net Economic Benefit per ton (4779) (2842) (4165) (4241) (6110) (1406) (3402) (14842) 1(13776) - PIG MEAT PRODUCTION CALCULATION OF THE NET ECONOMIC BENEFIT (Equation 6) ND Value Added from Pig meat per farm 2/ 127 130 245 421 311 75 (177) 156 302 Less Opportunity Cost of land, labour & capital 2/ 202 219 278 337 382 403 560 861 1141 ?let Economic Benefit from Pig nf;at per farm (75) (89) (33) 84 (71) (328) (737) (705) (839) Kgs DI Pig Meat produced per farm 55.37 56.52 74.03 85.33 92.99 81.79 82.78 86.66 108.11 flt Economic Benefit per ton (1355) (1575) (446) 984 (764) (4010) (8903) (8135) (7761) 1/ Border Prices from Table 43 Annex II Volume 2 Yugoslavia Agricultural Prices and Subsidies Case Study S, 46 , ,, 1 is . . a . ANNEX II Page 1 MARGINAL PRODUCT OF LABOUR 1. INTRODUCTION The estimation of the "shadow wages" used in calculating DRCs was summarised in the original Report.-1A more detailed explanation of the calculation of the marginal product of labour is given in this Annex. The marginal product of labour represents the opportunity cost to the economy of employing a given type of worker; it is the output foregone "at the margin" due to the shift of labour caused by the worker's engagement. Estimates of the marginal product of labour are presented in the World Bank internal paper. "Yugoslavia: Preliminary Estimates of Country Parameters for Economic Analysis of Projects", by Martin Schrenk and Shu-Chin Yang (January 10, 1975). The same general approach was followed in the present study, for two main reasons: (i) the basic principles of the earlier estimates were considered to be well-founded, and it was impracticable to attempt alternative approaches in a depth comparable to that of the Schrenk-Yang analysis; (ii) it was obviously desirable to maintain consistency, as far as possible,.with the parameters used by the Bank itself in project analysis. However, using recently available data, some revisions were made to the earlier World Bank calculations. II. ESTIMATION OF MARGINAL PRODUCT OF LABOUR A. Unskilled Workers By far the majQrtsUrce of unskilled labour is the rural population, in particular the individual sector of agriculture. This conclusion is justified at length in the Schrenk-Yang report. The reasons 1/ Yugoslavia, Agricultural Prices and Subsidies Case Study pages 43-45. March 1977. ANNEX II Page 2 include:- i) Incomes in the private agriculture sector are low (averaging 14,381 dinars per active household member in 1973, as against an average of 24,194 dinars in the nation's working population as a whole). ii) Agriculture still accounts for a disproportionately high share of the active population (38.5 percent in 1975, whereas agriculture's share of the social product was only 16 percent). iii) Agricultural workers constitute about one-half of the registered unemployment of unskilled and semi-skilled labour; and it is acknowledged that there is a great deal of under-employment in the private sector of agriculture. iv) There is a long-standing trend for workers to leave the land for other occupations (the total agricultural work-force has fallen from 5 4 million in 1953 to about 3.7 million in 1975). The social sector of agriculture is not likely to .be a significant source of unskilled labour because its employment is a small fraction of the total agricultural workforce (less than 7% in 1975); unskilled workers constitute a relatively small proportion of the employment in social sector agriculture (about 20 percent); and earnings in the sector are well up to urban levels. Of all the republics and provinces in Yugo!lavia, Kosovo probably has the lowest rural incomes (about 85 percent of the national average in 1973, in terms of income per active member of agricultural households). It also has a particularly rapid natural increase of population (2.8 percent a year, according to Schrenk and Yang, about double the overall national rate). Further, there is already a high density of rural population in Kosovo, and the average private ANNEX II Page 3 agricultural holding is smaller than in Yugoslavia as a whole. For these reasons, the average income per active member of private sector agricultural households in Kosovo was taken as an approximation to the marginal product of unskilled labour. This was the approach adopted by Schrenk and Yang. In the present study, the quinquennial household budget surveys were used as the source of income data. Income was defined as comprising earnings from employment off the farm, net income from sales of the farm's produce, remittances from household members working abroad and consumption in kind of farm produce (this definition is slightly narrower than that of the income figures quoted earlier). It approximates to the output foregone when a "representative group" of Kosovo rural workers moves to full-time employment. Remittances were included because family members working abroad are included in the household statistics. Table 1 summarises the relevant income figures. Table 1: Income of rural workers in Kosovo 1963 1968 1973 Income per household (dinars p.a.). 326.0* 8,464 27,327 Active members per household 2.D* 1.9* 2.3 Income per active member (dinars p.a.). 1,630 4,455 11,881 Income per active member (dinars per month) 136 371 990 *Estimated 1/ The latest is entitled "Survey on Receipts, Expenditure and Consumption of Households in 1973" (Part 1, in Statistical Bulletin 876, December 1974, was used). ANNEX II Page 4 The monthly income figures were also expressed in terms of a ratio to average net earnirgs of unskilled workers in (i) social sector agriculture, and (ii) the economy as a whole. Table 2 demonstrates the calculations. Table 2: Kosovo rural incomes as ratio to unskilled wages 1963 1968 1973 (a) Kosovo rural incomes, monthly 136 371 990 (b) Unskilled earnings, social sector agriculture (net) 177 604 1,358 (c) (a) as ratio to (b) 0.77 0.60 0.72 (d) Unskilled earnings, average for all employees (net) 208 615 1,371 (e) (a) as ratio to (d) 0.65 0.61 0.73 Since there is no good reason to suppose that the marginal product of labour fluctuated in relation to market "wages" in the way suggested by the ratios above, their averages - i.e. 0.70 for agricultural products and 0.66 for non-agricultural industries - were used for all years in the DRC calculations. In many developing countries the "urban migration response" exceeds one, i.e. more workers move from rural areas to the towns than the jobs available for them. Schrenk and Yang multiplied the marginal product of labour by 1.26 tp allow for such an excess response to urban employment. However, further examination of the relevant statistics shows that the difference between the non-agricultural work force and non- agricultural employment is less than was previously estimated, and it can be accounted for by private sector employment in the towns and by ANNEX II Page 5 registered unemployment. The percentage rate of unemployment is well within "Western' levels. Accordingly, since there was no more direct information on the urban migration response, it was thought reasonable to assume a factor of 1.0 - i.e. each unskilled job in a town would directly or indirectly draw away one active person from rural areas. The marginal product of labour in any given year, therefore, was calculated using the ratios derived from Table 2, as follows: i) in the case.of agricultural products, multiplying the average net "wage" of unskilled workers in social sector agriculture in that year by 0.70; ii) in the case of non-agricultural industries, multiplying the average unskilled waae in the econory as-a whole by 0.66. In fact, a,more internally consistent method would have been to establish the marginal product of labour using (ii) only. It was convenient to use a ratio of the kind above for estimating the shadow wage cost of agricultural products, but this could have been derived by dividing the marginal production from (ii) by the actual unskilled wage in social agriculture in the year concerned.- However, any difference between the -results on the two methods would be very slight. To calculate DRCs conforming with equation 5 of the "Supplementary Guidelines fnr Consultants". the marginal product of unskilled labour was converted to border prices. An-agricultural conversion factor was used, since the marginal output foregone by employing unskilled labour is primarily agricultural. The conversion factor is discussed in Annex II (pages 15-16) of the original Report. ANNEX II Page 6 B. Other Workers For workers other than the unskilled, the market wage was considered to represent the opportunity cost to the economy. The same assumption was adopted by Schrenk and Yang, on the grounds that unemployment rates of more qualified workers are comparatively low, and may be regarded as representing "frictional" unemployment; the total demand for such workers has grown steadily, and there is no evidence of significant under-employment. Moreover, earnings rates differ little between sectors. Average "wage" levels by branch of the economy are given in the Statistical Yearbooks of Yugoslavia, net of personal taxes and compulsory "contributions" to social services etc. The marginal product of labour in the more skilled categories is, on our assumptions, the gross wage. From the Yearbooks, the ratio of gross to net incomes for all workers can be calculated, and the same ratio was assumed to apply to the more skilled workers in each industry examined. (This calculation was required only for the manufacturing industries covered in this study - see Annex III, page 15, of the original Report for an example - as the wage data for agricultural products were already in gross terms). To convert the marginal product of more skilled labour into border prices, the SCF 1/ was used in the case of manufacturing industries and the agricultural conversion factor in the case of agricultural products. The rationale was that workers in manufacturing industry might be drawn from a variety of activities, whereas the alternative work for those in agriculture would most probably be in the same sector. If Described in Annex III, page 12, of the original Report, March 1977. ANNEX II Page 7 III."SHADOW WAGES" The "shadow wage" of unskilled labour used in the calculation of DRCs comprised not only the marginal Draduct of labour, but also, in the case of manufacturing industries, an "urbanisation cost". This represented the additional costs of housing, infrastructure, social services etc. caused by the movement of workers from rural areas to urban jobs. It did not apply, of course, In the calculation of D_Cs_ of agricultural proxts. The estimation of the urbanisation cost is discussed in Section III.C and Annex III, part 5, of the original Report. For workers other than the unskilled, the marginal product of labour was also the "shadow wage", since it was assumed that nearly all such workers taking urban jobs would already be living in towns. An example of the calculation of the "shadow wage" costs, for a particular manufacturing industry, is given on pages 14-15 of Annex III of the original Report. It will be noted that the "shadow wage" employed in this study is the "economic efficiency" price of labour, not the "social accounting" price. ANNEX III WHEAT SOCIAL SECTOR - EPC/DRC AND WORLD PRICES FIG. 1. - - / -.. - . - - .4 EPC/ -'0 World DRC Price (ND/kg) gf i i 41 le 7# 74 15 7* l . ,Al. , WHEAT SOCIAL SECTOR - DRC YIELDS AND TONS PER HA. FIG. 2. 1-0 DRC biYd/ha (tons 37 0 ANNEX I I I MAIZE SOCIAL SECTOR - EPC/DRC AND WORLD PRICES FIG. 3 EPC/ Worl d DRC 1Price (ND/kg) - o.~ -- .o.0 D R 41 -1 ( - Yd/ 04 // qu 7 ao 70 72* 73 7«i 71.lr4 ANNEX III BEEF SOCIAL SECTOR EPCfDRC AND WORLD PRICES FlG. 5 1.. EPC/ - --- -- -- World )RC . ....Pri ce - (ND/kg ) _01 1 l7 71 7/ 7; 7# ---- - >~ r1Íñ PIG MEAT SOCIAL SECTOR - EPC/DRC AND WORLD PRICES FIG. 6 -. -I. ii.~/ -Y.PC World MRC Price (ND/kg) ? 7.o . - - .. - -. - --- . I y ly H li le 11 73 14 ANNEX III INDIVIDUAL SECTOR SERBIA PROPER WHEAT EPC/DRC AND WORLD PRICES -* EPC1 ---r- \ /(ND/kg) 11~~ ~ 72 -i707 ••4 -- l-g 77 7X ( 70 7' 7~ I ~ Fl 8 ANNEX III INDIVIDUAL SECTOR SERBIA PROPER MAIZE EPC/DRC AND WORLD PRICES to - - - -- ---.~,Fig 9 . - .-- - EP/ fo . ~ - ~q World ~. Price -- (ND/kg) o-L- o .4 / -\/ -o . S C - /fé 77 73 4<7 7 INDIVIDUAL SECTOR SERBIA PROPER MAIZE DRC YIELDS AND TONS PER HA 04. DRC / • Yd/ha oSo o.4 gHL 1 I 7l' 7k 3 7' 7 L - ANNEX III INDIVIDUAL SECTOR SERBIA PROPER BEEF EPC/DRC AND WORLD PRICES Fig 11 EPC/ -- ol S- (ND/kg) 1*1 o 4- -I---------- - - I. - ---- - 7~. _~. - DRC 1-,Yd/ha - - - (tons) 7 - --7- I ___Prc