Report No. 4285-IND Indonesia Selected Issues of Energy Pricing Policies (In Three Volumes) Volume IlI: Methodological and Statistical Appendices August 1, 1983 Programs Department East Asia and Pacific Regional Office FOR OFFICIAL USE ONLY Document of the oirld Bank This document has a restricted dstnbution and may be used by recipents only in the performance of their official duties. hts contents may not otherwise be disclosed without World Bank authorization CURRENCY EQUIVALENTS Before November 15, 1978 US$1.00 = Rp 415 Annual Averages 1979-82 1979 US$1.00 Rp 623 1980 US$1.00 Rp 627 1981 US$1.00 Rp 632 1982 US$1.00 = Rp 661 After March 30, 1983 US$1.00 = Rp 970 FISCAL YEAR Government - April 1 to March 31 Bank Indonesia - April 1 to March 31 State Banks - January 1 to December 31 FOR OFFICIAL USE ONLY INDONESIA SELECTED ISSUES OF ENERGY PRICING POLICIES This report is based on the findings of two missions to Indonesia; one in November 1981 and the second in February 1982. The missions consisted of the following: William Branson (Consultant, Princeton University/NBER); Dennis Framholzer (Consultant, Stanford University); Noriko Iwase (Indonesia Division, East Asia and Pacific Country Programs Department); Lawrence Lau (Consultant, Stanford University); Dan Morrow (Indonesia Division, East Asia and Pacific Country Programs Department); Mark Pitt (Consultant, University of Minnesota); and Armeane M. Choksi (Chief of Mission; Country Strategy and Trade Policy Division, Country Policy Department). The mission wishes to acknowledge the assistance of the officials of the Ministry of Mines and Energy, the Ministry of Industry, the Ministry of Transport and the Central Bureau of Statistics. Without their help and advice, this report would not have been possible. This report was discussed with the Government of Indonesia in June 1983. This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. i INDONESIA SELECTED ISSUES OF ENERGY PRICING POLICIES VOLU4E III: METHODOLOGICAL ANID STATISTICAL APPENDICES Table of Contents Page No. APPENI)IX 1: TABLES ....................................................... APPENDIX II: THE DE4AND FOR ENERGY BY INDONESIAN INDUSTRY: THIE MOUDEL AND ITS ESTIMATION ........................ 26 APPENDIX IIl: TIIE DEMAND FOR ENERGY BY IIOUSETIOLDS ........................ 34 Expenditure Equations for Household Fuels ....... .......................... 34 Parameter Estimates ............................................ ... 34 Test Statistics ........................................................... 35 Elasticities ............................................................. 35 APPENDIX IV: THE ESTIMATION OF COMPENSATING VARIATION ............... 39 The Data . ............................................................ 41 The Econometric Mlodel ..................................................... 42 APPENDIX V: A MACROECONOMIC MODEL ........................................ 49 Introduction .......................................... 49 The Demand Side .......................................... 49 The Supply Side .......................................... 50 Explicit Solution for Price and Output .................................... 52 Statistical Appendix Table No. Appendix I I.1 Abbreviations, Conversion Factors and Energy Equivalents ....... 1 I.2 Energy Equivalents of Fuels .................................... 2 I.3 General Energy Equivalents ..................................... 3 I.4 Three-Digit ISIC Codes ...... .............................. 4 1.5 Fuel Elasticities: Sector: Food Processing ................... 5 1.6 Fuel Elasticities: Sector: Other Food Products ............. .. 5 I.7 Fuel Elasticities: Sector: Beverages Elasticities ............ 6 1.8 Fuel Elasticities: Sector: Tobacco Elasticities .............. 6 I.9 Fuel Elasticities: Sector: Spinning and Weaving .............. 7 ii Appendix I (continued) I.10 Fuel Elasticities: Sector: Textiles .......................... 7 1.11 Fuel Elasticities: Sector: Wearing Apparel .... ............... 8 1.12 Fuel Elasticities: Sector: Leather and Leather Substitutes 8 I.13 Fuel Elasticities: Sector: Leather Footwear .... .............. 9 1.14 Fuel Elasticities: Sector: Wood and Wood Products .... ........ 9 I.15 Fuel Elasticities: Sector: Wood Furniture .................... 10 I.16 Fuel Elasticities: Sector: Paper and Paper Products .... .... 10 1.17 Fuel Elasticities: Sector: Printing and Publishing .... ..... 1l I.18 Fuel Elasticities: Sector: Basic Chemical .................... 11 I.19 Fuel Elasticities: Sector: Other Chemical Products .... ....... 12 1.20 Fuel Elasticities: Sector: Rubber ..................... 12 1.21 Fuel Elasticities: Sector: Plastic Wares .. 13 1.22 Fuel Elasticities: Sector: Ceramic and Porcelain .... ......... 13 I.23 Fuel Elasticities: Sector: Glass and Glass Products .... ...... 14 1.24 Fuel Elasticities: Sector: Cement and Cement Products ........ 14 1.25 Fuel Elasticities: Sector: Structural Clay Products .... ...... 15 1.26 Fuel Elasticities: Sector: Other Nonmetallic M-etal Products .... 15 I.27 Fuel Elasticities: Sector: Fabricated Metal Products .... ..... 16 1.28 Fuel Elasticities: Sector: 'Mlachinery ......................... 16 1.29 F)uel Elasticities: Sector: Electrical Machinery .... .......... 17 1.30 Fuel Elasticities: Sector: Transport Equipment .... ........... 17 T.31 Fuel Elasticities: Sector: Measuring & Optical Equipment ..... 18 I.32 Price Elasticities for Aggregate Inputs - Sector 31 - Subsectors ..19 I.33 Elasticity of Dernand for Variable Factors - ISIC Sector 31 - Subsector ..19 I.34 Elasticity of Total Cost XJRT Energy Price - Sector 31 - Sulbsectors ..19 1.35 Pr-ice Elasticity for Aggregate Inputs - Sector 32 - Subsectors ..20 I.36 Elasticity of Demand for Variable Factors - Sector 32 - Subsectors ....................... 20 I.37 Elasticity of Total Cost WRT Energy Price - Sector 32 - Subsectors ....................... 20 I.38 Price Elasticities for Aggregate Inputs - Sector 33 - Subsectors ....................... 21 I.39 Elasticity of Demand for Variable Factors - Sector 33 - Subsectors ....................... 21 I.40 Elasticity of Total Cost WRT Energy Price - Sector 33 - Subsectors .... ........ ........... 21 I.41 Elasticities for Aggregate Inputs - Sector 34 - Subsectors .... ........ ........... 22 I.42 Elasticity of Demand for Variable Factors - Sector 34 - Subsectors ....................... 22 I.43 Elasticity of Total Cost WRT Energy Price - Sector 34 - Subsectors .... ........ ........... 22 I.44 Price Elasticities for Aggregate Inputs - Sector 35 - Subsectors ....................... 23 iii Appendix I (conti nuetd) 1.45 ElastLcity of Doen.iand for V~iriable Factors - Sector 35 - Subsectors .......................... 23 1.46 Elasticity of lTotal Cost WRT Energy Price - Sector 35 - Subsectors ........................ 23 I.47 Price Elasticities for Aggregate ITputs - Sector 36 - Subsectors ....................... 24 1.48 Elasticity of Demand for Variable Factors - Sector 36 - Subsectors ....................... 24 1.49 Elasticity of Total Cost WRT Energy Price - Sector 36 - Subsectors ....................... 24 1.50 Price Elasticities for iggregate Inputs - Sector 38 - Subsectors ....................... 25 I.51 Elasticity of Demand for Variable Factors - Sector 38 - Subsectors ....................... 25 1.52 Elasticity of Total Cost WRT Energy Price - Sector 38 - Subsectors ......................................... 25 Appendix III III.1 Parameter Estimiates of the Fuel Expenditure Equations ... ....... 36 III.2 Test Statistics for the Fuel Expenditure Equations .............. 37 Appendix IV IV.1 Number of Households with Zero Expenditures in Each Budget Category ..43 IV.2 Geographical Distribution of Sample Households . .44 IV.3 Average Budget Shares ..45 IV.4 Equations ......... 48 Appendix V V.1 Gross Output and Inputs in Mlanufacturing, 1979 .54 V.2 Shares of Value Added .55 APPENDIX I TABLES 1 - Appen.dix I Talsl tI. AIaaPeiltTATTCZ, rCNuflFlCTn!J rAIWTAq AUn ENERGY EQUIVALENTS Abbreviations and Conversion Factors bbl U. S. barrel (1 bbl - 42 U.S. galloons - 0.159 kl) Btu British thermal unit; the amount of heat required to raise the temperature of one pound of water one degree Fahrenheit 30E Barrel of oil equivalent cal calorie cd calendar day cu ft cubic feet (1 cu ft = 0.02832 cu meter) cu meter cubic meter 1 m3 = 35.31 cu ft) ft foot (1 ft - 0.3048 m) gal U. S. Gallon (1 gal 3.785 1 - 0.003785 m3) Gwh gigawatthour (1 Gwh thermal - 123 TCE) in. inch (1 in. - 0.0254 m) kcal kilocalorie (1 kcal = 1,000 cal) kce kilogram of coal equivalent kg kilogram (1 kg - 1,000 gr = 2.205 lb) kl kiloliter (1 kl - 1,000 L - 6.29 bbl) km kilometer (1 km = 1,000 m = 0.621 miles) kw kilowatt (1 kw = 1,000 watt) kWh kilowatt-hour (1 kWh thermal - 3,413 Btu - 0.000123 TCE) 1 liter (1 liter - 0.2642 U.S. gallons - 0.001 m3) lb pound (1 lb - 0.4536 kilograms) m 1 meter (1 m - 3.28 ft) mi. mile (1 mile - 1.609 kilometers) mt metric ton (1 mt - 1,000 kg - 2.205 lb) MW megawatt (1 MW - 1,000 kw - 1,000,000 watt) MWe megawatt electric output MWth megawatt thermal output SCF one cubic foot of natural gas, measured at 60 0 F and 1 atm.. SD stream day or operating day for a production process ST short ton (1 ST - 907 kg) tonne metric ton (1 tonne - 1,000 kg - 2,205 lbs) TCE tonne of coal equivalent TOE tonne of oil equivalent Multiples: M - 103, thousand MM - 106, million G - 109, billion T - 1012, trillion - 2 - Appendix I Table I.2: ENERGY EQUIVALENTS OF FUELS Quantities Approximately Equivalent to 1 TCE Fuels in U.S. Units in Metric Units Liquid Crude Oil 4.79 bbls 0.76 kl All Refined Products 4.96 bbls 0.79 kl Kerosene 4.91 bbls 0.78 kl Ethanol 8.08 bbls 1.28 kl Methanol 11.87 bbls 1.89 kl Natural Gas Liquids 6.01 bbls 0.96 kl Syncrude 4.96 bbls 0.79 kl Liquefied Natural Gas (LNG) 8.08 bbls 1.28 kl Liquefied Petroleum Gas (LPG) 6.67 bbls 1.06 kl Gaseous Natural Gas 27,780 SCF 787 cu meter City Gas 27,780 SCF 787 cu meter Solid Coal (Bukit Asam - Air Laya) 1.23 ST 1.12 tonne Coal Briquets 1.23 ST 1.12 tonne Charcoal 1.08 ST 0.98 tonne Wood and Agricultural Wastes 2.21 ST 2.00 tonne Tobit 1.3; CGNERAL ENERGY EQUIVALENTS T. L-IvcCL: .Ioltipilkatlon Factors Millon Kll Tlrr kru= \ tece TCE TCe sue TUE Joule bo-le Joule Cal l Cal 8tu MMH 8t1 I. Kt I l-3 IU-9 4.79xlU3 0.b04.103 0.0Zxl 0.U029.l0 0.b29.10- 7006 7.0 P 27.7b.103 27. 78.10i 2 TLe IUl I IU-6 4.79 0.b64 0.029xil'12 0.029.109 0.029 7?109 7.0106 27.781 ib 27.79 3 MI I I W T3 A 109 1ut I 4.79xiU6 0.0084.1 UI O.02 e1018 0.029o1015 i.029.10b 7.Iu5 7..1ou12 27.2ie112 27.78.)U0 4. bU0 2u9 U.21 U..21IU ° I 0.143 0.61.oblo 0.61e1U7 0.0101bf2 1.4.IU9 1l.46.b0 5.010U0 5.0d 5. tOE 1.4bl 1.4C I.46X0106 6.99 1 4.28.1i10 4.20ol07 4.28Ix0 2 l0.23.10 10.23.010 4U.5941U6 40.59^^ W 6. Joola 34.5.1. 9 34.5010U12 34.5.101 - 1.64.1IU0 0.23.1 U0 I 10- 1012 0.239 .239:1U-3 9.471o104 9.478.1010 7. Kilo Jo.le 34.5ol0 6 34.5.10f9 34.5.10-15 1.64.107 0.23.10o I3D I 10-9 0.239Kl03 0.239 U.9478 9.478.1U07 o T.rra 34.5.bU3 34.5 34. 5.1 6 1.b4.1U2 , 0.230102 0l12 I09 I 0.239oI01 0.239oIo9 9.478.l00 947.8 9. a-l U.142.1U-b 0.142.10-9 0142.1U15 U.68AlOl9 0.1l10 9 4.18 4.1g.1013 4.10010-12 I I0-3 3.980O10 0 3.9bb110 l9 IU. K (;I 0.142.l0 J U.142.101b 0.142.1Lu12 U.680lu0b 0 1e10b6 4.18-103 4.18 4.18.10 9 I03 1 3.9O8 3.968.10-6 II. 3tu 0.63b61( 3 0.036.10-6 O.0Ub.10-12 0.17.10U6 0.025.1]b6 0. 1o104 1.055 O.I05.L010 0.25103 0.25.106 I IU-6 12. I1M aLo 36 0.036 0.o01bo0-b U.17 0.U25 U.1050o10 0.I10500 U. l05310-2 0.25109 0.25oIU6 lob e AC dopLtd by tl. U. S. B----o of M31-. *0 Oes.Dlo a eloeItlo 0rooILy ol 6.99 bbl/Lt. - 4- Appendix I Table I.4: THREE-DIGIT ISIC CODES 31 38 311 Food Processing 381 Fabricated Metal Products 312 Other Food Products 382 Machinery 313 Beverages 383 Electrical Machinery 314 Tobacco 384 Transport Equipment 385 Measuring and Optical Equipment 32 3211 Spinning and Weaving Other 321 Textiles Except 3211 322 Wearing Apparel 323 Leather and Leather Substitutes 324 Leather Footwear 33 331 Wood and Wood Products 332 Wood Furniture 34 341 Paper and Paper Products 342 Printing and Publishing 35 351 Basic Chemicals 352 Other Chemical Products 355 Rubber 356 Plastic Wares 36 361 Ceramic and Porcelain 362 Glass and Glass Products 363 Cement and Cement Products 364 Structural Clay Products 369 Other Nonmetallic Metal Products - 5 - Appendix I Table 1.5: FUEL ELASTICITIES SECTOR: FOOD PROCESSING NO. OF OBSERVATIONS: 1808 Price Quantity Electricity Gasoline Fuel Oil Diesel Kerosene Electricity -1.3388 -.0168 .3813 .6138 .3487 .1149 .3440 .2030 .1974 .2383 Gasoline .1088 -2.3020 -.1602 .8987 .9502 .1558 .4168 .2592 .2442 .2962 Fuel Oil .6944 .3694 -.8084 .3345 .7517 .0671 .1840 .1098 .1056 .1284 Diesel -.1922 3.4725 -.3936 -4.0597 .5228 .2468 .6739 .4130 .3604 .4745 Kerosene .1896 .9904 1.1603 .8318 -3.5536 .1554 .4256 .2561 .2457 .2747 Table I.6: FUEL ELASTICITIES SECTOR: OTHER FOOD PRODUCTS NO. OF OBSERVATIONS: 906 Price Quantity Electricity Gasoline Fuel Oil Diesel Kerosene Electricity -1.5627 -.1621 .3328 .6217 .2921 .1429 .4277 .2524 .2454 .2963 Gasoline .1542 -2.0959 -.0841 .8543 .8999 .1380 .3894 .2297 .2164 .2625 Fuel Oil .7542 .4911 -.6528 .4629 .8006 .0543 .1490 .0887 .0855 .1040 Diesel -.2692 3.9740 -.5024 -4.5894 .5587 .2858 .7803 .4782 .4173 .5495 Kerosene .1854 .9964 1.1886 .8358 -3.5928 .1574 .4310 .2595 .2489 .2782 Note: In each column of Tables 1.5 - Table 1.52, the first number represents the elasticity (or cross-price elasticity) and the second number, the standard error; e.g., in Table I.5 the own-price elasticity for electricity is -1.3388 and the standard error is 0.1149. -6 Appendix I Table I.7: FUEL ELASTICITIES SECTOR: BEVERAGES NO. OF OBSERVATIONS: 93 Price Quantity Electricity Gasoline Fuel Oil Diesel Kerosene Electricity -.9725 .2488 .5222 .6619 .4998 .0790 .2363 .1395 .1358 .1637 Gasoline .2413 -1.7932 .0441 .8208 .8584 .1142 .3057 .1901 .1790 .2172 Fuel Oil .6078 .1591 -1.0863 .1110 .6867 .0926 .2540 .1512 .1457 .1772 Diesel -.3448 4.6458 -.6190 -5.2498 .6289 .3361 .9177 .5625 .4909 .6463 Kerosene .1981 .9792 1.1450 .82425 -3.4779 .1518 .4151 .2498 .2397 .2680 Table I.8: FUEL ELASTICITIES SECTOR: TOBACCO NO. OF OBSERVATIONS: 318 Price Quantity Electricity Gasoline Fuel Oil Diesel Kerosene Electricity -.9132 .2939 .5497 .6991 .5288 .0739 .2211 .1305 .1269 .1532 Gasoline .3289 -1.5449 .1623 .8183 .8502 .0965 .2582 .1606 .1512 .1835 Fuel Oil .5822 .0899 -1.2103 .0149 .6726 .1057 .2901 .1727 .1664 .2024 Diesel -.3025 4.2455 -.5524 -4.8611 .5849 .3063 .8363 .5126 .4473 .5890 Kerosene .1524 1.0653 1.2590 .8845 -3.9747 .1772 .4851 .2919 .2801 .3131 7 Appendix I Table I.9: FUEL ELASTICITIES SECTOR: SPINNING & WEAVING NO. OF OBSERVATIONS: 2130 Price Quantity Electricity Gasoline Fuel Oil Diesel Kerosene Electricity -.7572 -.9473 1.3548 1.0340 1.3558 .0567 .1789 .1207 .1225 .1184 Gasoline .9801 .0402 -.5852 -.9708 -.2197 .1859 .5592 -.4018 -.4068 -.3962 Fuel Oil .3583 3.4918 -2.8157 .1745 -1.1719 .1407 .4251 .2690 .2865 .2969 Diesel .3903 7.0368 -1.9038 -5.3467 -.9354 .2679 .8172 .6123 .5317 .5774 Kerosene 1.2326 -.5563 1.1676 .7971 -3.3197 .1638 .4999 .3581 .3579 .3139 Table I.10: FUEL ELASTICITIES SECTOR: TEXTILES NO. OF OBSERVATIONS: 487 Price Quantity Electricity Gasoline Fuel Oil Diesel Kerosene Electricity -.9502 -1.4356 1.4377 1.0374 1.4142 .0708 .2233 .1507 .1527 .1478 Gasoline .9124 -.0332 -.5018 -.8499 -.1714 .1879 .5052 .3628 .3673 .3579 Fuel Oil .4122 3.1345 -2.4766 .2525 -.9173 .1222 .3693 .2337 .2489 .2580 Diesel .4227 7.8761 -2.1980 -5.9782 -1.0961 .3048 .9300 .6968 .6051 .6571 Kerosene 1.2395 -.5626 1.1942 .8008 -3.3391 .1650 .5036 .3807 .3605 .3162 Appendix I Table I.11: FUEL ELASTICITIES SECTOR: WEARING APPAREL NO. OF OBSERVATIONS: 144 Price Quantity Electricity Gasoline Fuel Oil Diesel Kerosene Electricity -.1768 -.2489 1.2448 1.0565 1.2324 .0368 .1161 .0783 .0794 .0768 Gasoline .8294 -.1247 -.3633 -.6572 -.0848 .1416 .4262 .3061 .3099 .3019 Fuel Oil .2821 4.3795 -3.5486 .0492 -1.7084 .1838 .5549 .3512 .3740 .3876 Diesel .4819 9.0301 -2.5109 -6.6847 -1.2610 .3458 1.0549 .7904 .6861 .7454 Kerosene 1.7435 -.9187 1.6765 1.0954 -4.5430 .2438 .7440 .5329 .5326 .4871 Table I.12: FUEL ELASTICITIES SECTOR: LEATHER AND LEATHER SUBSTITUTES NO. OF OBSERVATIONS: 83 Price Quantity Electricity Gasoline Fuel Oil Diesel Kerosene Electricity -1.0181 -1.6132 1.4703 1.0408 1.4452 .0754 .2396 .1617 .1638 .1586 Gasoline .8628 .0878 -.4263 -.7439 -.1253 .1531 .4608 .3308 .3349 .3263 Fuel Oil .4915 2.7515 -2.0765 .3589 -.6121 .1015 .3066 .1940 .2067 .2142 Diesel .4029 7.2328 -1.9668 -5.4807 -.9704 .2758 .8409 .6300 .5472 .5942 Kerosene 1.5228 -.7798 1.4648 .9622 -4.0496 .2109 .6435 .4609 .4807 .4040 9 Appendix I Table I.13: FUEL ELASTICITIES SECTOR: LEATHER FOOTWEAR NO. OF OBSERVATIONS: 65 Price Quantity Electricity Gasoline Fuel Oil Diesel Kerosene Electricity -.9824 -1.5185 1.48527 1.0388 1.4285 .0732 .2309 .1558 .1579 .1528 Gasoline .7939 -.1651 -.2740 -.5371 -.0247 .1268 .3615 .2740 .2774 .2703 Fuel Oil .3978 3.2187 -2.5594 .2323 -.9798 .1266 .3827 .2422 .2530 .2673 Diesel .4223 7.9629 -2.1940 -5.9695 -1.0939 .3043 .9284 .6956 .6041 .6560 Kerosene 1.4494 -.7293 1.3945 .9.90 -3.8769 .1995 .6088 .4361 .4359 .3823 Table I.14: FUEL ELASTICITIES SECTOR: WOOD AND WOOD PRODUCTS NO. OF OBSERVATIONS: 856 Price Quantity Electricity Gasoline Fuel Oil Diesel Kerosene Electricity -1.1154 -.0498 .5528 -.4407 .3802 .4438 .7397 .4582 .6627 .7486 Gasoline .1234 -1.0133 -.3456 .2220 .8890 .2528 .3288 .2303 .3082 .3217 Fuel Oil .7615 .6730 -.2735 .6361 .5628 .0914 .1238 .0831 .1134 .1199 Diesel -.0823 1.9458 -.2293 .3913 -2.9676 .7803 1.0834 .6683 .8016 1.0302 Kerosene -.5073 -.7666 .1614 .0438 .1172 .5081 .6349 .4498 .5934 .5604 -10- Appendix I Table I.15: FUEL ELASTICITIES SECTOR: WOOD FURNITURE NO. OF OBSERVATIONS: 113 Price Quantity Electricity Gasoline Fuel Oil Diesel Kerosene Electricity -.7392 .2959 .6182 .0868 .5258 .2374 .3956 .2451 .3544 .4004 Gasoline .1226 -1.0143 -.3471 .2214 .8894 .2532 .3293 .2306 .3087 .3222 Fuel Oil .6002 .4570 -.4565 .3973 .2789 .1479 .2003 .1344 .1834 .1939 Diesel -.1038 2.6064 -.3292 .8580 -4.0249 1.0531 1.4622 .9019 1.0819 1.3904 Kerosene -.6396 -.9619 .1912 .0451 .3787 .6313 .7689 .5589 .7373 .6963 Table I.16: FUEL ELASTICITIES SECTOR: PAPER AND PAPER PRODUCTS NO. OF OBSERVATIONS: 151 Price Quantity Electricity Gasoline Fuel Oil Diesel Kerosene Electricity -.7262 .3571 .1202 .0805 .0098 .2146 .5056 .4125 .4484 .4098 Gasoline -.1297 -2.3868 .7571 .4831 .8093 .2294 .4979 .4212 .4614 .4289 Fuel Oil .3482 1.0066 -.5073 .4504 .1921 .1425 .3043 .2278 .2755 .2651 Diesel -.1269 -.1021 1.0977 -1.7880 .4241 .5299 .9656 .7947 .8750 .9096 Kerosene .1073 .0926 .5828 .2474 -1.7076 .2468 .5827 .4567 .5180 .4417 -11- Appendix I Table I.17: FUEL ELASTICITIES SECTOR: PRINTING & PUBLISHING NO. OF OBSERVATIONS: 462 Price Quantity Electricity Gasoline Fuel Oil Diesel Kerosene Electricity -.3586 .6770 .5758 .5503 .5287 .0916 .21598 .1761 .1915 .1750 Gasoline -.0293 -2.0828 .7139 .4843 .7577 .1922 .4173 .3530 .3867 .3595 Fuel Oil -.1262 1.1313 -.8502 .0690 -.4243 .2718 .5812 .4351 .5262 .5063 Diesel -.3118 -.2761 1.4505 -2.2632 .4811 .7625 1.3896 1.1436 1.2592 1.3089 Kerosene .1086 .0941 .5804 .2476 -1.7001 .2449 .5782 .4532 .5140 .4383 Table I.18: FUEL ELASTICITIES SECTOR: BASIC CHEMICAL NO. OF OBSERVATIONS: 138 Price Quantity Electricity Gasoline Fuel Oil Diesel Kerosene Electricity -.7434 -1.3251 .4268 1.0441 .2044 .1925 .4592 .3463 .3976 .3988 Gasoline -.3500 -3.0726 .9244 1.6605 .2312 .2206 .4718 .3630 .4078 .4150 Fuel Oil .,5219 1.2893 -.8970 .5816 .6789 .0695 .1678 .1123 .1286 .1353 Diesel .4474 .7743 1.0831 -2.9131 .1048 .2839 .6722 .4379 .4761 .5601 Kerosene .3605 1.1556 1.2244 .6082 -4.1445 .2698 .6039 .4399 .5153 .4736 - 12 - Appendix I Table I.19: FUEL ELASTICITIES SECTOR: OTHER CHEMICAL PRODUCTS NO. OF OBSERVATIONS: 393 Price Quantity Electricity Gasoline Fuel Oil Diesel Kerosene Electricity -.6113 -.7185 .5092 .9400 .3522 .1348 .3214 .2424 .2783 .2792 Gasoline -.0754 -2.2691 .8077 1.3179 .3274 .1529 .3270 .2515 .2826 .2876 Fuel Oil .1686 1.5031 -1.5597 .2725 .4416 .1208 .2918 .1952 .2236 .2352 Diesel .4943 .9359 1.3530 -3.6941 .0316 .3835 .9079 .5915 .6430 .7565 Kerosene .3504 .9825 1.0372 .5473 -3.4356 .2145 .4800 .3497 .4096 .3765 Table I.20: FUEL ELASTICITIES SECTOR: RUBBER NO. OF OBSERVATIONS: 299 Price Quantity Electricity Gasoline Fuel Oil Diesel Kerosene Electricity -.6652 -.9216 .4716 .9603 .2934 .1529 .3648 .2751 .3158 .3168 Gasoline -.2959 -2.8943 .8966 1.5697 .2434 .2047 .4378 .3368 .3784 .3851 Fuel Oil .3315 1.3906 -1.2466 .4139 .5481 .0959 .2316 .1549 .1775 .1867 Diesel .4488 .7621 1.0581 -2.8134 .1205 .2721 .6442 .4197 .4562 .5368 Kerosene .3499 1.0566 1.1178 .5701 -3.7655 .2398 .5368 .3910 .4580 .4210 - 13 - Appendix I Table I.21: FUEL ELASTICITIES SECTOR: PLASTIC WARES NO. OF OBSERVATIONS: 344 Price Quantity Electricity Gasoline Fuel Oil Diesel Kerosene Electricity -.6635 -.9144 .4727 .9594 .2853 .1525 .3632 .2739 .3144 .3154 Gasoline -.3632 -3.1180 .9350 1.6849 .2288 .2247 .4806 .3697 .4154 .4228 Fuel Oil .4083 1.3471 -1.1040 .4814 .6003 .0850 .2053 .1373 .1573 .1655 Diesel .4490 .7983 1.1282 -3.0729 .0830 .3034 .7181 .4679 .5086 .5984 Kerosene .3719 1.2231 1.2968 .6371 -4.3801 .2889 .6465 .4709 .5516 .5070 Table 1.22: FUEL ELASTICITIES SECTOR: CERAMIC AND PORCELAIN NO. OF OBSERVATIONS: 22 Price Quantity Electricity Gasoline Fuel Oil Diesel Kerosene Electricity -1.7483 1.7002 * .8395 -.0312 -.6702 .2218 .5399 .3467 .3995 .6517 Gasoline .5438 -1.8670 -.5117 .5084 .7615 .3072 .5148 .4338 .4597 .6909 Fuel Oil .7058 -.2258 -.7668 .3789 .8277 .1515 .2763 .2110 .2105 .3455 Diesel .4998 1.,2304 -.1618 -.4021 .0808 .1752 .3883 .2951 .2080 .4088 Kerosene .5776 2.1901 -.0420 -.3447 -2.8399 .3661 .7187 .5266 .5000 .6926 - 14 - Appendix I Table I.23: FUEL ELASTICITIES SECTOR: GLASS AND GLASS PRODUCTS NO. OF OBSERVATIONS: 102 Price Quantity Electricity Gasoline Fuel Oil Diesel Kerosene Electricity -1.7702 1.7147 .8412 -.0424 -.6909 .2251 .5479 .3518 .4054 .6614 Gasoline .5593 -1.9786 -.5913 .5208 .7968 .3349 .5809 .4729 .5011 .7531 Fuel Oil .6661 -.4871 -.9190 .2614 .8171 .1876 .3420 .2612 .2605 .4277 Diesel .5277 1.2161 -.0957 -.3799 .1328 .1651 3.659 .2780 .1960 .3852 Kerosene .5949 2.3678 -.0941 -.4306 -3.0930 .4070 .7990 .5857 .5559 .7701 Table I.24: FUEL ELASTICITIES SECTOR: CEMENT AND CEMENT PRODUCTS NO. OF OBSERVATIONS: 365 Price Quantity Electricity Gasoline Fuel Oil Diesel Kerosene Electricity -1.4434 1.5276 .8348 .1339 -.3605 .1786 .4346 .2791 .3216 .5246 Gasoline .5366 -1.7339 -.4140 .5046 .7326 .2767 .4634 .3907 .4140 .6222 Fuel Oil .7469 -.0349 -.6497 .4726 .8495 .1272 .2319 .1771 .1766 .2900 Diesel .1620 2.1222 -1.6133 -.5749 -.9624 .4701 1.0419 .7917 .5581 1.0969 Kerosene .6822 2.9468 -.1878 -.6129 -3.7124 .5140 1.0091 .7397 .7020 .9728 - 15 - Appendix I Table I.25: FUEL ELASTICITIES SECTOR: STRUCTURAL CLAY PRODUCTS NO. OF OBSERVATIONS: 210 Price Quantity Electricity Gasoline Fuel Oil Diesel Kerosene Electricity -2.4921 2.3949 1.0285 -.3538 -1.3681 .3521 .8570 .5504 .6342 1.0345 Gasoline .5521 -1.9542 -.5599 .5147 .7814 .3237 .5421 .4570 .4843 .7279 Fuel Oil .8208 .2514 -.4680 .6210 .8954 .0926 .1689 .1290 .1286 .2112 Diesel .2478 1.5174 -.9019 -.5817 -.4804 .3045 .6748 .5128 .3614 .7104 Kerosene .5895 2.3378 -.0822 -.4104 -3.0313 .3969 .7791 .5711 .5420 .7509 Table 1.26: FUEL ELASTICITIES SECTOR: OTHER NONMETALLIC METAL PRODUCTS NO. OF OBSERVATIONS: 93 Price Quantity Electricity Gasoline Fuel Oil Diesel Kerosene Electricity -2.9965 3.0805 1.2958 -.5096 -1.8345 .4600 1.1194 .7189 .8284 1.3513 Gasoline .5426 -1.8539 -.5022 .5075 .7580 .3041 .5093 .4294 .4550 .6839 Fuel Oil .9242 .6245 -.2279 .8190 .9634 0.488 .0889 .0679 .0677 .1112 Diesel .1690 1.9247 -1.4209 -.5953 -.8380 .4211 .9331 .7091 .4998 .9824 Kerosene .6731 2.8973 -.1815 -.5990 -3.6613 .5049 .9912 .7266 .6896 .9554 - 16 - Appendix I Table 1.27: FUEL ELASTICITIES SECTOR: FABRICATED METAL PRODUCTS NO. OF OBSERVATIONS: 616 Price Quantity Electricity Gasoline Fuel Oil Diesel Kerosene Electricity -2.8214 .9196 .4613 .9761 .8649 .1569 .3265 .2038 .2675 .2866 Gasoline .2181 -1.3492 -.2028-.3002 1.3016 .1806 .3670 .2387 .3176 .3307 Fuel Oil 1.6326 .0739 -.7477 -.0637 .1763 .1152 .2399 .1420 .1965 .2085 Diesel 1.2988 .8543 -.0260 -1.4636 -1.3779 .1802 .3849 .2318 .5452 .3533 Kerosene .3023 .6443 1.0906 1.0295 -3.7509 .2420 .5011 .306 .4139 .4150 Table 1.28: FUEL ELASTICITIES SECTOR: MACHINERY NO. OF OBSERVATIONS: 616 Price Quantity Electricity Gasoline Fuel Oil Diesel Kerosene Electricity -1.8414 .9253 .6299 .9618 .8901 .1012 .2105 .1313 .1724 .1847 Gasoline .2109 -1.3721 -.2216 -.3218 1.3242 .1855 .3771 .2453 .3264 .3399 Fuel Oil 1.8491 -.1611 -.9310 -.3385 -.0291 .1486 .3094 .1832 .2534 .2689 Diesel 1.4924 .9669 -.0737 -1.5911 -1.6720 .2131 .4551 .2740 .6445 .4176 Kerosene .3024 .6464 1.0953 1.0338 -3.7686 .2434 .5041 .3079 .4164 .4174 - 17 - Appendix I Table I.29: FUEL ELASTICITIES SECTOR: ELECTRICAL MACHINERY NO. OF OBSERVATIONS: 180 Price Quantity Electricity Gasoline Fuel Oil Diesel Kerosene Electricity -2.7204 .9161 .4753 .9705 .8635 .1510 .3141 .1960 .2573 .2756 Gasoline .2506 -1.2593 -.1277 -.2153 1.2245 .1623 .3299 .2146 .2855 .2973 Fuel Oil 1.6392 .0656 -.7543 -.0733 .1689 .1164 .2422 .1434 .1984 .2105 Diesel 1.2913 .8502 -.0234 -1.4577 -1.3650 .1789 .3820 .2300 .5411 .3506 Kerosene .3204 .6313 1.0606 1.0019 -3.6345 .2328 .4821 .2944 .3982 .3992 Table I.30: FUEL ELASTICITIES SECTOR: TRANSPORT EQUIPMENT NO. OF OBSERVATIONS: 261 Price Quantity Electricity Gasoline Fuel Oil Diesel Kerosene Electricity -2.6009 .9133 .4930 .9652 .8632 .1440 .2995 .1869 .2454 .2629 Gasoline .2568 -1.2389 -.1104 -.1960 1.2093 .1584 .3220 .2094 .2787 .2902 Fuel Oil 1.6374 .0679 -.7525 -.0707 .1709 .1161 .2416 .1430 .1979 .2100 Diesel 1.5606 1.0084 -.0850 -1.6287 -1.7644 .2239 .4782 .2879 .6773 .4388 Kerosene .3037 .6615 1.1285 1.0646 -3.8913 .2532 .5244 .3203 .4332 .4343 - 18 - Appendix Table I.31: FUEL ELASTICITIES SECTOR: MEASURING & OPTICAL EQUIPMENT NO. OF OBSERVATIONS: 32 Price Quantity Electricity Gasoline Fuel Oil Diesel Kerosene Electricity -1.7670 .9287 .6452 .9637 .8949 .0971 .2020 .1261 .1655 .1773 Gasoline .2040 -1.3950 -.2404 -.3433 1.3479 .1906 .3875 .2520 .3354 .3492 Fuel Oil 2.2104 -.4305 -1.1282 -.6636 -.2570 .1953 .4065 .2407 .3329 .3533 Diesel 2.6301 1.6886 -.1759 -2.1029 -3.0395 .3818 .8154 .4910 1.1549 .7483 Kerosene .3441 .5805 .8890 .8467 -2.7665 .1673 .3464 .2116 .2862 .2869 - 19 - Appendix I Table I.32: PRICE ELASTICITIES FOR AGGREGATE INPUTS SECTOR 31 SUBSECTORS Other Food Food Processing Products Beverages Tobacco Elasticity (E,E) -.7053 -.6777 -.7207 -.6899 (.0244) (.0196) (.0349) (.0542) Elasticity (E,L) .7053 .6777 .7207 .6899 (.0244) (.0196 (.0349) (.0542) Elasticity (L,E) .1704 .2160 .1132 .0662 (.0059) (.0063) (.0055) (.0052) Elasticity (L,L) -.1704 -.2160 -.1132 -.0662 (.0059) (.0063) (.0055) (.0052) Table I.33: ELASTICITY OF DEMAND FOR VARIABLE FACTORS - ISIC SECTOR 31 SUBSECTOR Other Food Food Processing Products Beverages Tobacco Elasticity (E,K) .1927 .2329 .2111 .2041 (.0154) (.0158) (.0511) (.0320) Elasticity (E,Y) .3703 .3422 .2614 .5361 (.0.53) (.0174) (..0557) (.0282) Elasticity (L,K) .0712 .1290 .0488 -.0342 (.0107) (.0129) (.0489) (.0222) Elasticity (L,Y) .3618 .3350 .2500 .5194 (.0105) (.0150) (.0536) (.0152) Table I.34: ELASTICITY OF TOTAL COST WRT ENERGY PRICE - SECTOR 31 SUBSECTORS Other Food Food Processing Products Beverages Tobacco Elasticity (TO,E) .0184 .0296 .0136 .0028 (.0003) (.0006) (.0015) (.0003) Elasticity (TO,1) .0038 .0032 .0050 .0011 Elasticity (TO,2) .0019 .0044 .0030 .0008 Elasticity (TO,3) .0090 .0173 .0037 .0006 Elasticity (TO,4) .0013 .0010 .0002 .0001 Elasticity (TO,5) .0024 .0036 .0017 .0002 20 - Appendix I Table I.35: PRICE ELASTICITIES FOR AGGREGATE INPUTS - SECTOR 32 SSUBSECTORS Leather Spinning Wearing & Leather Leather & Weaving Textiles Apparel Substitutes Footwear Elasticity (E,E) -.5436 -.5902 -.5768 -.3419 -.5576 (.0495) (.0396) (.0427) (.0831) (.0468) Elasticity (E,L) .5436 .5902 .5768 .3419 .5576 (.0495) (.0396) (.0427) (.0831) (.04687) Elasticity (L,E) .0602 .0841 .0754 .0216 .0658 (.0055) (.0056) (.0056) (.0052) (.0055) Elasticity (L,L) -.0602 -.0841 -.0754 -.0216 -.0658 (.0055) (.0056) (.0056) (.0052) (.0055) Table 1.36: ELASTICITY OF DEMAND FOR VARIABLE FACTORS - SECTOR 32 SUBSECTORS Leather Spinning Wearing & Leather Leather * Weaving Textiles Apparel Substitutes Footwear Elasticity (E,K) .3758 .2288 .2152 .5603 .4472 (.0201) (.0225) (.0280) (.0759) (.0359) Elasticity (E,Y) .8064 .6575 .6356 .8270 .6635 (.0363) .0508) (.0300) (.0419) (.0249) Elasticity (L,K) .2050 .0884 .0653 .2859 .2849 (.0084) (.0170) (.0230) (.0701) (.0356) Elasticity (L,Y) .6209 .5050 .4728 .5291 .4871 (.0283) (.0481) (.0200) (.0189) (.0138) Table I.37: ELASTICITY OF TOTAL COST WRT ENERGY PRICE - SECTOR 32 SUBSECTORS Leather Spinning Wearing & Leather Leather & Weaving Textiles Apparel Substitutes Footwear Elasticity (TC,E) .0145 .0193 .0147 .0029 .0187 (.0019) (.0024) (.0013) (.0010) (.0012) Elasticity (TC,l) .0089 .0102 .0113 .0013 .0094 Elasticity (TC,2) .0007 .0015 .0018 .0003 .0031 Elasticity (TC,3) .0031 .0057 .0014 .0011 .0051 Elasticity (TC,4) .0007 .0005 .0002 .0001 .0005 Elasticity (TC,5) .0010 .0014 .0002 .0001 .0006 - 21 - Appendix I Table I.38: PRICE ELASTICITIES FOR AGGREGATE INPUTS SECTOR 33 SUBSECTORS Wood and Wood Wood Products Furniture * Elasticity (E,E) -.4168 -.0742 (.0440) (.0873) Elasticity (E,L) .4168 .0742 (.0440) (.0873) Elasticity (L,E) .0775 .0064 (.0082) (.0075) Elasticity (L,L) -.0775 -.0064 (.0082) (.0075) Table 1.39: ELASTICITY OF DEMAND FOR VARIABLE FACTORS - SECTOR 33 SUBSECTOR Wood and Wood Wood Products Furniture Elasticity (E,K) .2660 .3217 (.0291) (.0544) Elasticity (E,Y) .4305 .5758 (.0234) (.0556) Elasticity (L,K) .1352 .0839 (.0181) (.0310) Elasticity (L,Y) .4234 .5630 (.0134) (.0405) Table 1.40: ELASTICITY OF TOTAL COST WRT ENERGY PRICE - SECTOR 33 SUBSECTORS Wood and Wood Wood Products Furniture Elasticity (TC,E) .0222 .0220 (.0005) (.0028) Elasticity (TC,1) .0015 .0079 Elasticity (TC,2) .0041 .0039 Elasticity (TC,3) .0159 .0101 Elasticity (TC,4) .0004 .0000 Elasticity (TC,5) .0002 .0001 - 22 - Appendix I Table 1.41: ELASTICITIES FOR AGGREGATE INPUTS - SECTOR 34 SUBSECTORS Paper and Printing Paper Products & Publishing Elasticity (E,E) -.4923 -.3724 (.0416) (.0656) Elasticity (E,L) .4923 .3724 (.0416) (.0656) Elasticity (L,E) .1107 .0491 (.0093) (.0086) Elasticity (L,L) -.1107 -.0491 (.0093) (.0086) Table I.42: ELASTICITY OF DEMAND FOR VARIABLE FACTORS - SECTOR 34 SUBSECTORS Paper & Printing Paper Products & Publishing Elasticity (E,K) .3543 .4313 (.0356) (.0396) Elasticity (E,Y) .3861 .4426 (.0360) (.0412) Elasticity (L,K) .1592 .1471 (.0291) (.0226) Elasticity (L,Y) .3288 .3591 (.0283) (.0201) Table I.43: ELASTICITY OF TOTAL COST WRT ENERGY PRICE - SECTOR 34 SUBSECTOR Paper & Printing Paper Products & Publishing Elasticity (TC,E) .0233 .0240 (.0010) (.0009) Elasticity (TC,1) .0041 .0142 Elasticity (TC,2) .0027 .0040 Elasticity (TC,3) .0125 .0038 Elasticity (TC,4) .0025 .0004 Elasticity (TC,5) .0016 .0016 - 23 - Appendix I Table I.44: PRICE ELASTICITIES FOR AGGREGATE INPUTS - SECTOR 35 SUBSECTORS Other Basic Chemical Plastic Chemicals Products Rubber Wares Elasticity (E,E) -.5747 -.5013 -.5679 -.5749 (.0331) (.0618) (.0288) (.0357) Elasticity (E,L) .5747 .5013 .5679 .5749 (.0331) (.0618) (..0288) (.0357) Elasticity (L,E) .1637 .0676 .1947 .1487 (.0094) (.0083) (.0099) (.0092) Elasticity (L,L) -.1637 -.0676 -.1947 -.1487 (.0094) (.0083) (.0099) (.0092) Table 1.45: ELASTICITY OF DEMAND FOR VARIABLE FACTORS - SECTOR 35 SUB SECTORS Other Basic Chemical Plastic Chemicals Products Rubber Wares Elasticity (E,K) .2296 .3228 .3350 .2337 (.0316) (.0316) (.0251) (.0238) Elasticity (E,Y) .4755 .5328 .4289 .5224 (.0343) (.0345) (.0243) (.0302) Elasticity (L,K) .0941 .0994 .2120 .0905 (.0293) (.0213) (.0212) (.0257) Elasticity (L,Y) .4078 .4212 .3674 .4509 (.0312) (.0212) (.0212) (.0257) Table 1.46: ELASTICITY OF TOTAL COST WRT ENERGY PRICE - SECTOR 35 SUB SECTORS Other Basic Chemical Plastic Chemicals Products Rubber Wares Elasticity (TO,E) .0377 .0134 .0141 .0289 (.0019) (.0008) (.0004) (.0010) Elasticity (TO,1) .0047 .0039 .0032 .0069 Elasticity (TO,2) .0030 .0030 .0014 .0022 Elasticity (TO,3) .0246 .0049 .0069 .0165 Elasticity (TO,4) .0038 .0003 .0016 .0023 Elasticity (TO,5) .0016 .0013 .0009 .0009 - 24 - Appendix I Table 1.47: PRICE ELASTICITIES FOR AGGREGATE INPUTS - SECTOR 36 SUBSECTORS Glass & Cement Other Non- Ceramic and Glass & Cement Structural Metallic Metal Porcelain Products Products Clay Products Products Elasticity (E,E) -.6835 -.6047 -.6425 -.7616 -.6587 (.0407 (.0300) (.0342) (.0896) (.0364) Elasticity (L) .6835 .6047 .6425 .7616 .6587 (.0407) (.0300) (.0342) (.0896) (.0364) Elasticity (E) .2431 .3333 .2913 .1030 .2727 (.0145) (.0165) (.0155) (.0121) (.0151) Elasticity (L,L) -.2431 -.3333 -.2913 -.1030 -.2727 (.0145) (.0165) (.0155) (.0121) (.0151) Table I.48: ELASTICITY OF DEMAND FOR VARIABLE FACTORS - SECTOR 36 SUBSECTORS Glass & Cement Other Non- Ceramic and Glass & Cement Structural Metallic Metal Porcelain Products Products Clay Products Products Elasticity (E,K) .2482 .2341 .1321 .1933 .1186 (.0551) (.0340) (.0263) (.0819) (.0590) Elasticity (E,Y) .5519 .4617 .5795 .7487 .6099 (.0664) (.0778) (.0521) (.0691) (.0485) Elasticity (L,K) .1838 .1797 .0740 .0746 .0585 (.0517) (.0286) (.0180) (.0680) (.0553) Elasticity (L,Y) .4039 (.3367) .4460 .4757 .4715 (.0612) (.0777) (.0450) (.0418) (.0420) Table 1.49: ELASTICITY OF TOTAL COST WRT ENERGY PRICE - SECTOR 36 SUBSECTORS Glass & Cament Other Non- Ceramic and Glass & Cement Structural Metallic Metal Porcelain Products Products Clay Products Products Elasticity (TO,E) .0892 .0833 .0799 .0502 .0653 (.0085) (.0052) (.0065) (.0157) (.0031) Elasticity (TO,1) .0156 .0160 .0271 .0022 .0005 Elasticity (TO,2) .0062 .0047 .0101 .0036 .0060 Elasticity (TO,3) .0275 .0204 .0386 .0320 .0541 Elasticity (TO,4) .0322 .0366 .0022 .0082 .0032 Elasticity (TO,5) .0077 .0056 .0019 .0041 .0017 - 25 - . Appendix I Table I.50: PRICE ELASTICITIES FOR AGGREGATE INPUTS - SECTOR 38 SUBSECTORS Fabricated Measuring Metal Electrical Transport & Optical Products Machinery Machinery Equipment Equipment Elasticity (E,E) -.5127 -.5279 -.5498 -.4939 -.5262 (.0518) (.0489) (.0446) (.0551) (.0493) Elasticity (E,L) .5127 .5279 .5498 .4939 .5262 (.0518) (.0489) (.0446) (.0551) (.0493) Elasticity (L,E) .0572 .0627 .0724 .0514 .0620 (.0058) (.0058) (.0059) (.0057) (.0058) Elasticity (L,L) -.0572 -.0627 -.0724 -.0514 -.0620 (.0058) (.0058) (.0059) (.0057) (.0058) Table I.51: ELASTICITY OF DEMAND FOR VARIABLE FACTORS - SECTOR 38 SUBSECTORS Fabricated Measuring Metal Electrical Transport & Optical Products Machinery Machinery Equipment Equipment Elasticity (E,K) .3043 .3765 .2311 .3781 .3695 (.0279) (.0368) (.0300) (.0984) (.0935) Elasticity (E,Y) .5478 .6587 .4546 .4616 .4323 (.0769) (.0748) (.0361) (.0359) (.0454) Elasticity (L,K) .1393 .2195 .0862 .2038 .2116 (.0163) (.0302) (.0228) (.0960) (.0911) Elasticity (L,Y) .4370 .5532 .3573 .3444 .3262 (.0727) (.0707) (.0270) (.0256) (.0382) Table 1.52: ELASTICITY OF TOTAL COST WRT ENERGY PRICE - SECTOR 38 SUBSECTORS Fabricated Measuring Metal Electrical Transport & Optical Products Machinery Machinery Equipment Equipment Elasticity (TO,E) .0112 .0206 .0106 .0085 .0129 (.0020) (.0023) (.0013) (.0016) (.0010) Elasticity (TO,1) .0033 .0106 .0031 .0028 .0071 Elasticity (TO,2) .0016 .0026 .0018 .0016 .0015 Elasticity (TO,3) .0049 .0057 .0043 .0035 .0019 Elasticity (TO,4) .0007 .0005 .0006 .0002 .0000 Elasticity (TO,5) - 26 Appendix II THE DEMAND FOR ENERGY BY INDONESIAN INDUSTRY: THE MODEL AND ITS ESTIMATION The theoretical model used in deriving energy demand relationships in this study parallels that of Fuss (1977) and Pindyck (1979). First, it is assumed that the production function is weakly separable in the major kinds of energy inputs. Thus, the cost-minimizing mix of these energy inputs is independent of the mix of other factors - capital, labor and materials. Furthermore, if the energy aggregate is homothetic in its components (electricity, gasoline, fuel oil, diesel and kerosene), cost-minimization becomes a two stage procedure: optimize the mix of fuels which make up the energy aggregate and then optimize the mix of the energy aggregate, labor, capital and materials. Finally, it is assumed that materials are weakly separable from the labor, capital and energy inputs. This assumption is required because the data required to construct a materials price index are not available. These assumptions on the structure of production can be summarized by the following production function: Q = F [{K, L, E (E1, E2, E3, E4, E5)I; Ml (1) where K, L, and M are capital labor and materials respectively, and E is the energy aggregate which is a homothetic function of the five fuels. If factor prices and output levels are exogenously determined and if the stock of capital is fixed in the short run, duality implies that cost- minimization given the production function (1) can be uniquely represented by a short-run cost-function of the form. C = G [g {P P(P P, PE25 ,PE P ), K,Q}; M] (2) L E E1 E2 E3 E4 E5 where PE is an aggregate price index for energy. In Fuss (1977) and Pindyck (1979), the price of energy PE' which is also unit energy cost to the optimizing firm, is represented by an arbitrary unit cost function such as the translog. Estimation of this cost function provides estimates of the elasticities of substitution among alternative fuels as well as their own and cross-price demand elasticities. In addition, estimates of the parameters of the translog aggregate price index can be used to calculate PE, an estimate of the price index, up to an arbitrary scaling factor. In the second-stage, the cost of 6utput is represented by a nonhomothetic translog cost function, and PE can be used as an instrumental variable for the price of energy. Estimation of this cost function provides - 27 - Appendix II us with estimates of elasticities of substitution and demand elasticities for capital, labor and energy. The econometric model described above must be altered when the data used in estimation are at the level of the firm, as they are in this study. Estimation of the production structure represented by (1) and (2) in most studies does not permit a positive probability of observing zero levels of input use. In addition standard approaches of estimating systems of share equations derived from cost functions would result in biased and inconsistent estimates because the random disturbances have nonzero means and are correlated with the exogenous variables. Moreover, dropping those firms which do not use at least one of the inputs would reduce the sample size severely and still result in biased estimates. The model developed here assumes that the homothetic energy aggregator functions are randomly distributed over the population of firms, that is pE = E (P El *-'pE53 U1, -.,. u5) (3) where ui represents a random component which enters the cost function in such a way that optimal cost shares consist of both a deterministic and a random component. The cost function is non-stochastic to each firm since it knows the value of u = (u1, ..., U5) and, thus, chooses the optimal energy fuel mix. However, from the investigators point of view the random component vector, u, and hence the observed shares, are random drawings from the population of firms. Among close substitutes, such as alternative energy inputs, cost- minimization may often result in corner solutions, that is, zero levels of use for one or more fuels. Clearly, at least one input will be used. Calling this input j, it is clear that if input i is also used by the firm then the marginal rate of transformation in production between inputs i and j will equal their price ratio at the cost minimizing solution. If input is i not used, the marginal rate of transformation in production may not equal the price ratio. Furthermore, changes in the prices of inputs i or j may not affect the level of its use if it is already unused. The relationship between optimal input shares and prices depends on whether any of the shares are at a corner. Optimal share levels may be written as Si = Si {p, d (p, u), u}, i = 1, 5 (4) where p is a vector of inputs prices and di is a vector of dummy variables whose jth element has the value 1 if input j is consumed and zero otherwise. - 28 - Appendix II Note that at most all but one elment of d can be zero. In this case, consumption of that one fuel is at the share upper-bound of unity. The elements of the vecor d are, of course, also functions of the vectors p and u. Reduced form equations for optimal inputs levels are Si = Si (P, u), i = 1, 5 (5) It is these reduced-form equations which are estimated in the first-stage of the analysis. Note that because the estimated equations are reduced forms, the coefficients estimated are not those of the underlying cost function. As a result, the usual symmetry restrictions of cost functions are not applicable in estimating these reduced forms. The reduced form will still be characterized by zero homogeneity in prices and adding-up. In estimating the model (5), we assume that the vector u is additive and has a joint normal distribution with zero means and covariance matrix L . In addition, we estimate first-order approximations to the reduced form share equations of the form i ai + i Yij log P. + ui, i = 1, 5 (6) The limited dependent variable model of Tobin (1958) (tobit) provides a likely candidate for estimating equations such as (6) since it permits a positive probability of observing zero input levels. However, because there are a significant number of unit shares, Nelson and Rosett's (1975) two-limit probit extension of Tobin's model is the appropriate estimator. Fully efficient estimation would require a multivariate extension of the two-limit probit model which could also take account of the adding-up restriction. The lack of such a computationally tractable multivariate estimator requires us to estimate the share equations singly. Although the resulting parameter estimates are consistent, there is no guarantee that predicted shares will add-up. The stochastic model underlying two-limit probit regression is given by the following relationship: Nt =X t + ut if L1 SXt a + ut < L2 = L2 if Xt + ut > L2 - 29 - Appendix II (7) L, if X+t t - 1, 2, ..., N Where N is the number of observations, Yt is the dependent variable, Xt is a vector of independent variables, a is a vector of unknown coefficients, L1 and L2 are known upper and lower truncation values and ut is an independently distributel error term assumed to be normally distributed with zero mean and variance a . The expected value of the dependent variable Yt is nonlinear in Xt and with L= 0 and L2 = 1 is given by E(Y) = azF(z) = ow {l-F(w)} + az + a {f(z)-f(w)} (8) where z = -X6/a, w = (1-Xa)/a and F( ) and f( ) represent the normal cumulative distribution and unit normal density functions respectively (subscripts have been omitted for simplicity). In Fuss (1977) and Pindyck (1979), the price index for energy is just the unit translog cost function. InE 0 +i ilog Pi + 1/2 i j .ij log Pi log P. (9) and an estimate of the price index PEup to a scalar multiple is obtained by substituting the parameter estimates of the associated share equations into (9). In our case, we^cannot identify the parameters of the cost function (9), and thus approximate P as proportional to some known index PE . Deaton and Muellbauer (1980) foung that Stone's (1953) index log PE= i Si log P (10) provided a reasonably close approximation to a translog price index. This is the form of the price index used in the aggregate model as an instrumental variable for the price of energy. The shares Si used in calculating P are the expected shares E(s) of the reduced form equations (6) calculated as in (8), normalized so that - 30 - Appendix II E (S . In the second-stage, in which the demand for aggregate inputs is modelled, the cost function given by (2) is represented by a nonhomothetic translog second-order approximation of the form Log C = a + £ ai log P. + Zak log Fk + 1/2 i E Y log P log P. + 1/2 yk m k log F log F + 1/2 Yki g F L (11) km km k in k ki k + 1/2 y log P lg i k ik i gFk where i, j = E, L; k, m Q, K, and Fk is the quantity of the kth fixed factor. From Shepards lemma (Diewert (1971)), the variable cost minimizing level of use of the ith variable factor Vi = ac/ao . Therefore, input demand functions in terms of cost shares are given by P.V. 11 a log C/3 log Pi = = S. C or (12) S. = a. + . 'i log P. + k Yki g k Since all aggregate inputs have positive levels of use for all firms, all marginal rates of transformation in production equal their relevant price ratios at the optimal shares. Therefore, the linear structural share equations (12) can be estimated by the usual Zellner efficient continuous dependent variable techniques. Since the two variable input shares must add to 1, only one of them needs to be estimated. In order to identify the parameters a , a and y , the cost function itself (11) must be estimated 0 k km - 31 - Appendix II along with one of the share equations (12). In order that the cost function and the share equations satisfy the properties of a neoclassical production structure, the following parameter restrictions are required: E a. = ~~~~~~~~~~~~~~(13) i ai = 1, i = E, L JYi i y = 0, i, j = E, L (14) i Yik = , i= E, L; k =Q,K (15) yij = Yi') i, j = E, L (16) Yik = Yki' i = E, L; = Q, K (17) Ykm = Ymk' k, m = Q, K (18) In summary, estimation of the complete model is accomplished with the following two-stage procedure: 1. Estimate the set of reduced form share equations (6) by two- limit probit regression, imposing zero homogeneity of prices in each equation. An estimate of an aggregate price index PE is obtained by using the normalized expected shares as exponential weights in (10). 2. Estimate the cost function (11) along with one share equation (12) by Zellner efficient techniques, replacing PE by its instrumental variable PE. - 32 - Appendix II From the first-stage estimates we can obtain estimates of the price elasticity of demand, which taking into account the tobit estimator, are calculated as Yi[F(w.)-F( i) - Y1 S[F(w1) 1 + Si - 1, i = 1,5 (19) Si Yij [F(w i)-F(z d] -i = 3 S] + Si, i,j = 1,5 (20) 3. Price elasticities of demand for the aggregate variable factors can be calculated from the parameters of the second stage estimates as: -Y.. + S .-1, i = E,L (21) yij ij S -+ Si, i,j = E,L (22) The elasticity of demand for variable factors with respect to change in the quantity of fixd factors can be calculated as: + k YklnF + i Y In P + yki/Si (23) i = E, L; k = Q, K; m = Q, K The elasticities of total cost with respect to the price of aggregate energy and each of the five fuels are calculated as: ( p vc TClE aE + YEJ log Pi + k YkE log Fk) - (24) Tand ~ TC and - 33 - Appendix II TlTC,i SifTC,ES i 1 1,5 (25) where VC/TC is the share of variable costs in total costs. Calculating standard errors for these elasticities is complicated because the elasticities are nonlinear functions of the estimated parameters. Approximate estimates of the standard errors are obtained by assuming that the shares and [F(w)-F(z)3, the probability of nonlimit levels of input use, are constant and equal to their means. - 34 - Appendix III THE DEMAND FOR ENERGY BY HOUSEHOLDS Expenditure Equations for Household Fuels Expenditure equations relating household expenditure of each combustible fuel to its price, the level of total household expenditure and a set of household characteristics are given by V = a + yij log p. + . log m + kaikhk' i K, F, C (1) j = K, W where p. is the price of fuel j, m is total household expenditure, hk is the kth household characteristic, a., y.., 5. and O. are parameters to be estimated and K, F, C and W refer to-kerosene, irewood, charcoal and wood fuel respectively. Furthermore, the parameters yj., 5i and 0ik vary as follows: O m yij Yij + yi. log (m/s) (2) S. = i. + i m log (m/s) (3) O m 0k ak0 + a0k log (mIs) (4) ik ik ik where s is household size. Parameter Estimates The expenditure equations for kerosene, charcoal and firewood were estimated by tobit (Tobin 1958) maximum likelihood techniques. The stochastic model underlying tobit is given by the following relationship: yt =Xt + utif Xta > O (Al) = 0 if X < O t = 1, 2 ... N - 35 - Appendix III where N is the number of observations, yt is the dependent variable, Xt is a vector of independent variables, 6 is a vector of unknown coefficients aed ut is an independently distributed error term with zero mean and variance a . Table III.1 presents the parameter estimates and their asymptotic standard errors. Test Statistics Table III.2 provides test statistics for the null hypothesis that each of the underlying varying parameters (Yi., i) 0 ik) of the expenditure equations are zero. The Wald test statistics indicate that all the price parameters Y. are significantly different from zero at the .05 level. Households arl sensitive to both own- and cross-prices in choosing levels of fuel expenditure for all three fuels. The location variables are significant in all cases as well, but one of the two demographic variables is not significantly different from zero in each of the equations. A likelihood ratio test strongly supports the contention that household demand response to exogenous variables varies with household m m m expenditure per capita. The set of interaction parameters (y.. , *'k 0)k are jointly different from zero at the .05 level in every expenditHre equation. Elasticities The formula used in calculating the own-and cross- price demand elasticties presented in Table 3 are [Y. + Y m log (m/s)] F(z) ii V- Vi (A2) 0 m I[Y .+ Y i log (m/s)] F(z) C = iJ _ J _ (A3) 2ij V. where F(zi) is the standard normal cumulative distribution evaluated at 1 z ~~~~~~~~~~z zi C ( Y+ log p. + a. log m + k hikhk). a jiki - 36 - Appendix III Table III.1; PARAMETER ESTIMATES OF THE FUEL EXPENDITURE EQUATIONS (asymptotic standard errors in parenthesis) Kerosene Charcoal Firewood 1. Constant -1872.0 -11936.3 4512.4 (1779.4) (2773.7) (4781.1) 2. Log household size (s) -716.58 -1246.6 1335.7 (307.64) (4589.6) (827.6) 3. Log members age > 10 years -474.86 512.77 -2150.4 (279.16) (427.24) (750.4) 4. Log total expenditure (m) 669.88 2818.7 945.82 (264.66) (409.5) (757.50) 5. Urban 18.082 -948.37 -1181.1 (210.38) (290.41) (526.6) 6. Java 163.66 -223.94 7057.2 (173.44) (263.10) (495.3) 7. Log price of kerosene -1186.5 -1583.7 -5511.3 (390.40) (634.8) (977.0) 8. Log price of wood fuels 83.875 -259.62 -320.88 (188.84) (295.49) (480.02) 9. Log size * log (m/s) 97.639 15.260 -62.751 (28.792) (44.566) (76.64) 10. Log members * log (m/s) 58.473 -55.599 267.71 (32.320) (48.082) (87.78) 11. Log expenditure * log (m/s) -52.432 -170.29 -184.35 (10.875) (17.60) (33.21) 12. Urban * log (m/s) -28.910 83.670 205.01 (24.143) (32.87) (60.55) 13. Java * log (m/s) 5.567 36.262 -803.67 (20.00) (29.553) (57.73) 14. Log price of kerosene * log (m/s) 135.47 161.012 663.77 (44.94) (971.037) (112.63) 15. Log price of wood fuels * log (m/s) -3.663 37.328 25.417 (21.80 (33.206) (55.632) 16. Sigma 465.13 399.22 1036.02 Limits 70 4737 2782 Nonlimits 5810 1143 3098 Observations 5880 5880 5880 - 37 - Appendix III Table III.2: TEST STATISTICS FOR THE FUEL EXPENDITURE EQUATIONS Expenditure Equation Coefficients Tested Kerosene Charcoal Firewood Price of Kerosene a/ 9.29 18.32 46.22 Price of Wood Fuel a/ 14.17 16.07 9.11 Household Expenditure 25.69 93.85 61.23 Household Size -a 11.60 13.88 2.94 Household Members a/ aged 5.18 1.88 15.71 11 years and above Rural _/ 267.22 142.38 299.67 Java_/ 215.72 30.47 218.13 All Interaction Parameters ˇ/ 94.45 203.59 389.81 a The test statistics are Wald tests. The critical values of the Chi- squared distribution with two degrees of freedom are 5.99 and 9.21 at the .05 and .01 levels respectively. b/ The test statistics are log-likelihood ratios. The critical Chi-squared values with 7 degrees of freedom are 14.07 and 18.48 at the .05 and .01 levels respectively. - 38 - Appendix III In the tobit model, F(zi) represents the probability of a household consuming a positive quantity of fuel i given zi1 Approximate standard errors for these price elasticities were calculated by treating Vi and F(zi) as fixed and equal to their expected values. Fuel demand elasticities with respect to total household expenditure are calculated as (S. + . log m = n. log (m/s) + . ym. log p. + k am E;. - i iki k ()(A4) im V. I The elasticity of the kerosene subsidy with respect to its price is calculated as 5KK ¢ _ i' ~~~~~~~~~~~~~~~(5) K,K sK where eK is the kerosene own-price elasticity given by equation (A2) above P is the'oimestic price per liter of kerosene and SK is the subsidy per liter of kerosene. - 39 - Appendix IV THE ESTIMATION OF COMPENSATING VARIATION In order to evaluate the distribution of welfare among different groups of households in an economy, it is first of all necessary to know each group's utility function. Groups of households may be distinguished by observable attributes, which include the size and composition of the household; age, sex, and education of the head of household; ethnicity; and other attributes that are relevant in the determination of demand patterns. In order to identify the utility function (up to a monotonic transformation) of each group, it is necessary to have observations of demands corresponding to different levels of prices. Without such variations in prices, the degrees of substitutability among different consumption commodities (or, what amount to the same thing, the curvatures of the indifference surfaces) can be determined only by a priori assumptions such as zero or unitary elasticities of substitution. This is the major drawback of most approaches based only on cross-section data. The advantage of the proposed approach, which combines both aggregate time series and individual cross-section data, lies in the fact that the degrees of substitutability are allowed to be determined empirically from the actual data. No a priori assumptions are necessary. Consequently, the utility function of each group can be identified to a much greater degree of reliability and correspondence with reality. Given the knowledge of each group's utility function (up to a monotonic transformation), it is possible to perform compensating variation calculations separately for each group of households. For example, let V(p/M,A) be the indirect utility function of a member of the group of households with a vector of attributes equal to A. Let P0 and M be the base period (before the implementation of the policy) vectors of prices and income respectively. Then, the base period utility level of the household is given by V (P0/Mo,A) - V0. Corresponding to the indirect utility function V (.), there is an expenditure function M (p, V, A) which gives the minimum expenditure required for the household to achieve utility level V at prices p. In the base period, assuming that the household maximizes utility, M (PO, VO, A) m0. Now suppose that as a result of the implementation of the policy or project, the price vector changes from Po to P1 and the income for a household with a vector of attributes equal to A changes from M0 to M1. Then, in order for this household tomaintain its utility at the same level VO as before, the minimum expenditure required is given by M (P1, VO, A). Then, the compensating variation for this household (relative to the base period) is given by CV = M (P15 VO, A) -Mo - 40 - Appendix IV However, since the income of the household has also undergone a change in the process, one may define a concept of net compensating variation as CV* = CV - (Ml - MO) = M (P1 VO, A) - MO - (M1 -MO) = M (P1 V0, A) - M1. The net compensating variation measures the additional expenditure (possibly negative for some groups) that is required to maintain a household with a vector of attributes equal to A at its base period level of utility, taking into account both price and income changes. It is, therefore, clear that if the net compensating variation for a group is positive, then the group is worse off than before. Since CV* depends on A in addition to P1, M1 and VO, it is group specific, reflecting the differential tastes and needs of each group. Thus, the group-specific compensating variations can be used collectively to assess the distribution of welfare gains and losses across different groups resulting from the implementation of the policy or project. This exercise also identifies those groups of households which are most significantly affected. In addition, it is a theorem in welfare economics that if the sum of the net compensating variations across all households is negative, then it is possible to find a redistribution scheme that every household is either better off or, at least, as well off as during the base period, as a result of the implementation of the policy. Thus, the distinguishing features of this model of aggregate consumer demand are: (a) the ability to accept directly full or partial information on the joint distribution of incomes and demographic characteristics; (b) the ability to identify uniquely the individual household consumer demand functions which depend on prices as well as on income of individual consumer groups distinguished by demographic characteristics; (c) the ability to recover the utility function of individual consumer groups and, hence, to perform compensating variation calculations; (d) full consistency between the microeconomic model of individual consumer behavior and the macroeconomic model of aggregate consumer behavior; and (e) the ability to combine consistently time-series aggregate consumption data and cross-sectional individual household consumption data. - 41 - Appendix IV Thus, the group-specific compensating variations can be used to assess the relative distribution of the burdens of any policy change across different groups. In this study, alternate policy changes -- defined by varying (increased) price levels of energy and time-paths for the elimination of subsidies -- will be developed in collaboration with the Government to determine their welfare implications on the households. This exercise will, therefore, identify those consumer groups which are most significantly affected by different policy regimes. It also provides information which may be useful for targeting conservation efforts towards specific consumer groups. However, the computation of group-specific welfare impacts is not the only application of such a model. Since unique group-specific demand functions can be derived from this model, it can also be used to project group-specific consumption patterns in response to change in prices, incomes, quantity constraints, and other Government policies. Moreover, one can also focus on the quantity of aggregate consumption. In this case, the model can be used to generate a projection of aggregate consumption, given the prices, incomes, and information on the joint distribution of household incomes and attributes. No microeconomic level simulation is required. In addition, the model can directly assimilate information of the changing demographic and income distributions which may indeed be highly relevant for medium- to long-term projections. The impact of different policies can be analyzed both in terms of their direct effects on consumption -- conservation efforts, rationing -- and in terms of their indirect effects through changes in prices and incomes. These policies may be introduced into the model at the regional, as well as the national, level. The methodolgy can also be extended in other directions. First, it can be adapted to accommodate prices which may vary across different groups of households because of transportation costs, progressivity of the income tax, and other factors. Thus, for example, any policy which has a differential impact on the prices of different regions can be evaluated with this methodology with respect to the distribution of the welfare gains and losses. Second, the methodology can be adapted to take into account the varying "basic needs" across groups of households. This can be done, for example, through the introduction of price-dependent indices of incomes and attributes in the aggregate demand function. Third, the methodology can be extended to include an analysis of household purchase and ownership of durable consumer goods, and the effect of such ownership on the consumption pattern. This can be done, for example, by treating the quantity and type of consumer durables owned as another dimension of the vector of attributes. This approach is currently being followed in a research project, "The Welfare Implications of Eliminating Energy Subsidies in Indonesia" (672-70), which is currently underway and addresses the issues discussed in Chapter 5 in much greater depth and detail. The Data The data used in this study are taken from the 1976 SUSENAS survey conducted by the Bureau of Statistics of the Government of Indonesia. The surveys are conducted in three subrounds -- January-April (Subround 1), May- August (Subround 2), and September-December (Subround 3). These data are supplemented with price data also obtained from the Bureau of Statistics. - 42 - Appendix IV In each subround approximately 17,000 households were included in the survey. However, many of the households show a zero level of expenditure for one or more the expenditure categories distinguished in our study -- namely, food, clothing, transportation, fuel, housing and miscellaneous. These households, with the exception of those which show a zero level of expenditure for transportation, were dropped from the sample because there does not seem to be a reasonable way to adjust the data. The number of zero observations for each category of expenditures for each province (in Subround 2) is presented in Table IV.1. A total of 7,342 households remain in the sample for further analysis. The other subrounds show similar tendencies. The households included for further anlaysis are distributed as follows: Number of Households Subround Urban Rural Total 1 4,776 2,565 7,341 2 4,865 2,233 7,098 3 4,720 3,305 8,025 In Table IV.2 the detailed geographical distribution is presented. In Table IV.3 the average budget shares devoted to each of the six expenditure categories are presented. It is clear that there is a sigrificant seasonal pattern especially with regard to food and clothing expenditures. Food expenditures dominate the average budget, constituting more than 70%. The remaining categories are approximately comparable in magnitude with the exception of transportation. Transportation accounts for such a small share of the average budget -- less than 0.3% -- that consideration shouid be given to aggregating it with another expenditure category in future analysis. The Econometric Model It is assumed that the preferences of each household, distinguished by province, urban or rural location, and size, as measured by the number of members of the household, can be represented by an indirect homogeneous transcendental logarithmic utility function of the form: 1 V =ca + Aailn6 P. 6 6 P. P. Table IV.1: NUMBER OF HOUSEHOLDS WITH ZERO EXPENDITURES IN EACH BUDGET CATEGORY Total Number Observations Province Food Clothing Transportation Fuel Housing Miscellaneous in Province Jakarta 0 2063 2674 19 145 2 3228 Jawa Barat 0 853 1622 3 355 19 1665 Jawa Tenga 0 1168 2094 3 684 16 2148 Yogjakarta 0 405 799 0 269 5 837 Jawa Timur 0 1068 2070 5 364 30 2158 D.I. Aceh 0 173 534 6 113 1 547 Sumatera Utara 1 379 800 45 149 10 815 Riau 0 140 302 4 86 1 319 Jambi 0 138 327 1 44 0 340 1 Sumatera Barat 0 225 474 3 113 2 479 J Sumatera Selatan 0 161 331 15 62 8 334 Bengkulu 0 47 166 0 30 2 178 Lampung 0 104 287 0 4 1 288 Kalimantan Barat 0 136 336 3 .120 1 340 Kalimantan Tengah 0 100 380 0 32 1 381 Kalimantan Selatan 0 108 416 5 39 2 420 Kalimantan Timur 0 118 251 2 33 0 252 Bali 0 79 198 0 91 0 199 Sulawesi Utara 0 117 273 2 62 2 295 Sulawesi Tengah 0 129 269 5 48 4 278 Sulawesi Selatan 0 291 651 2 110 10 659 Sulawesi Tenggara 0 112 234 4 30 7 234 Nusa Tenggara Barat 0 137 300 1 95 3 300 Nusa Tenggara Timur 0 173 369 3 110 4 376 m Maluku 0 78 240 2 55 0 246 :3 TOTAL 1 8,502 16,397 133 3,243 131 17,316 - 44 - Appendix IV Table IV.2: GEOGRAPHICAL DISTRIBUTION OF SAMPLE HOUSEHOLDS Subround Province (1) (2) (3) Jakarta 1,126 1,026 1,462 Jawa Barat 664 800 728 Jawa Tenga 714 713 564 Yog jakarta 303 374 308 Jawa Timur 953 882 944 D.I. Aceh 306 295 400 Sumatera Utara 363 401 416 Riau 134 120 150 Jambi 172 119 139 Sumatera Barat 200 216 279 Sumatera Selatan 153 143 192 Bengkulu 110 113 80 Lampung 182 115 172 Kalimantan Barat 123 135 166 Kalimantan Tengah 254 201 281 Kalimantan Selatan 288 272 298 Kalimantan Timur 121 102 131 Bali 74 48 87 Sulawesi Utara 157 144 183 Sulawesi Tengah 139 92 132 Sulawesi Selatan 328 344 470 Sulawesi Tenggara 109 112 121 Nusa Tenggara Barat 111 102 172 Nusa Tenggara Timur 128 141 66 Maluku 129 88 82 TOTAL 7,341 7,098 8,025 - 45 - Appendix IV Table IV.3: AVERAGE BUDGET SHARES (Percent) Subround (1) ~~~(2) (3) Food 72.1 72.3 70.9 Clothing 7.4 8.3 13.6 Transportation 0.3 0.2 0.2 Fuel 4.3 4.5 4.2 Housing 6.5 6.7 7.0 Miscellaneous 9.3 8.0 4.1 TOTAL 100.0 100.0 100.0 - 46 - Appendix IV 6 P. + i1 Yi In ( M-) D 6 26 P. + *E Z i n M Dj i-1 j-l ln( ) Di 6 P. •iEl E, In ( M ') N where Pi is the price of the ith commodity, i=1, ...,6 (food, clothing, transportation, fuel, housing and miscellaneous), M is the total expenditure, Du is the dummy variable for urban location (urban = 1), D. is the provincial dummy variable, j=1,...,26 (province 13, however, is not represented in the sample), N is the number of members in the household, 6 6 6 ,El -1; 1,j -'= i' j; i jEl j = 0, i=l,...,6; El X. = 0; 6 6 i-l 7ij °' j; i-l 'i= °- P.X. 6 26 - -1 I = . + El ij ln P. + X. D j i uD + i N, i=1,...,6. N j=l i I u j= iu Note that there are linear restrictions on the parameters, both within each equation and across the equations. These restrictions are imposed in the estimation. The system of equations (3.2) is estimated with the individual household data for each round separately. The data are then pooled together, and a system of equations with subround effects is estimated from the pooled data. The subround effects take the form of additional dummy variables. - 47 - Appendix IV The system of demand equations have many parameters because of the presence of all the provincial dummy variables. The large number of parameters does not create any degree of freedom problem because of the large sample. However, it is still desirable to be able to simplify the model if possible. Thus a series of statistical hypotheses were tested with the data withl regard to the provincial dummy variables. The first hypothesis tested is that the parameters corresponding to all the provincial dummy variables are zeroes. This hypothesis is strongly rejected. The second hypothesis tested is that the parameters corresponding to the provincial dummy variables of all provinces within the same island are identical. This hypothesis is also strongly rejected with the exception of two provinces -- Nusatenggara Barat and Timur - whose parameters for the provincial dummy variables are not statistically significantly different. Further explorations of identical parameters within smaller groups on the same island have been made. However, the overall conclusion of these explorations is that the provinces are sufficiently different, even on the same island, that it is better to let each province have its own individual set of parameters corresponding to its dummy variable. The parameter e7timates are presented together with the associated t-ratios in Table IV.4.1 It is clear that the estimates are highly statistically significant by any ordinary standards. Thus, a great deal of confidence can be placed on the parameter values. / The parameters corresponding to the provincial dummy variables are ommitted. Table IV.4: EQUATIONS Y Food Clothing Transportation Fuel Housing Urlban Dummy 0.083 0.000 -0.002 -0.014 -0.337 (40.363) (0.137) (-6.708) (-22.716) (-24.047) Number of Members 0.001 -0.002 0.0003 -0.002 -0.002 (3.362) (-7.563) (8.336) (-21.68)) (-10.271) Pril(e of Food 0.016 -0.008 0.006 -0.013 -0.001 (8.695) (-6.152) (17.350) (-22.360) (-2.018) Price of Clothing -0.008 0.011 -0.000 -0.006 -0.001 (-6.152) (6.760) (-0.175) (-12.143) (-2.192) Pri'ze of Transportation 0.006 -0.000 -0.008 0.002 0.000 (17.350) (-0.175) (-18.816) (8.024) (0.579) Price of Fuel -0.013 -0.006 0.002 0.021 -0.002 (-22.104) (-12.143) (8.024) (41.247) (-8.242) Priee of Housing -0.001 -0.001 0.000 -0.002 0.001 (-2.018) -2.192) (0.579) (-8.242) (1.826) Prica of Miscellaneous 0.000 0.004 0.000 -0.002 0.003 Subrvund 1 0.000 0.007 0.000 0.001 0.003 (0.107) (3.528) (0.747) (1.291) (2.581) Subround 2 -0.006 0.060 -0.000 0.000 0.001 (-3.902) (32.600) (-0.978) (0.455) (0.797) "/ ,he parameter estimates of the "Miscellaneous" equation can be obtained by adding up the five (!quations and subtracting unity. - 49 - Appendix V A MACROECONOMIC MODEL Introduction In this section we set out the model for analyzing the effect of the energy price increase. The analytical point of the exercise is to see how the energy price increase is passed on to the price of output, both directly and through effects on wages. We also want to be able to see how the answers are influenced by the existence of differences in energy use across sectors and by differences in factor shares of output. Finally, the parameters of the model should be interpretable using the Indonesian data. The simplest model structure that meets our needs is the following. The agricultural sector supplies food at a constant price which will initially be assumed to be insensitive to the energy price. This assumption will be relaxed to see its effect. There is an industrial sector using labor L, energy E, and capital K in a Cobb-Douglas production function to produce output Q. Capital is fixed; energy and labor are variable inputs. Energy is supplied elastically by the government at a fixed price PE' Labor is supplied to the industrial sector elastically at a given real wage. The private sector consumes food, industrial output, and energy. Their prices make up the CPI, which is the real-wage deflator. The nominal demand for the output of the industrial sector is fixed by the money supply. Tnis makes the price elasticity of demand -1. We will lay out the model and its solution, and then see how the reaction of the price of output is sensitive to (a) whether the agriculture price is stabilized and (b) the high share of capital in manufacturing. The Demand Side Since the focus is on supply relationships, we strip demand down to one equation. Assume constant velocity in the modern and industrial sector, so that M = kPQQ. where Q is industrial output, and PQ is its price. Then changes in demoand are specified by PQ + Q =1, (1) where a ^ means percentage change: M dM/M. - 50 - Appendix V The Supply Side Initially, we take the price of agricultural output PA as given, and the energy price P1 as fixed by the Government. The real wage in terms of the CPI is also fixed by the assumption that labor can be drawn from the agricultural sector at that wage. This means that movements in the nominal wage follow the CPI, which is a weighted average of PA, PE' and P W = yAPA + YE 'P, + YQPQ' (2) where y. is the share of sector i in the CPI. The object is to see how the industrial sector's price and output, and demands for labor and energy input, react to an exogenous increase in P.. The production function is assurmed to be Cobb-Doug:las in the inputs labor L, energy E, and capital K. The results here are not particularly sensitive to the form of the production function, so it is best to keep things simple. Thus Q output is given by al a2 U Q = Q(L,E,K) = L E K (a, + a2 + 13 = 1). With capital fixed, the first-order conditions for variable L and 1 are given by al-l a2 (3 MPL = alL E K W/PQ MIPE =2L E B,/ EQ These can be totally differenitiated to obtain the solution for changes in L and E inputs as the prices vary. Remermber that W will be replaced by the real wage expresion (2.2). The total diEferentials are (a1-l)L + a2E = (yQ-1)1 Q+ YAPA +Y E; (IPL) (3) a1L + (a 2-)E -PQ + PE (,IPE) - 51 - Appendi7, V The solution for L and E can be obtained by substitution. In matrix form equation (3) can be written as F;' A = () Q E A The detQrninant of the coefficient martix A is 1- a - a 2= B. The solution for L, E can be written as L 1 a 2 -21 | QI YE YA| AQ (3") 1 1 | - A L~~~~~~ the separate solutions for L and E as all three prices vary are now L =(Y -1)(a2-1) + a2PQ - [a2 + YE(l-a2)]P y A(1 -a2)PA E = 1ff(l-alyQ - [l-al+ alAy aly * Q Q ~ ~ 1IF alYA AI 5 An increase in PQ raises both inputs and output Q. An increase in PE or PA reduces both inputs and the output. Thus we can draw the supply cruve in Figure 1. An increase in P raises output Q along the supply curve. An increase in PA or P shifts it up. The supply curve in Figure 1, along with the denand curve of equation (1) give equilibrium PQ and Q for given PA and PE' An increase in PE shifts the supply curve.up, raising P and reducing Q. With PA fixed the CPI will,go up by y PE + Y P . The central question is: What is the magnitude of the P reaction; Q Q Q - 52 - Appendix V Figure 1: INDUSTRY DEBMAD AND SUPPLY \~~~~~~~~E A PQQ = X Explicit Solution for Price and Output The change in Q resulting from changes in prices can be obtained by total differentiation of the production function for Q as,L and E,change, and substitution for L and F from (4) and (5). Thus, Q = aIL + a2E from the production function and substitution from (4) and (5) gives the explicit supply curve. Q -~ X [ai(l-y ) + X2APQ - (q +altF)PF- YA A (6) Q sQ The coeffecient of P is the positive supply elasticity of Q, denoted by s. Note tPat in the supSly-demand diagram of Figure 1, an increase in s = - [aj(1-y ) + a2] flattens the supply curve. It will he useful in the di2cussion beloQ to remember that an increase in al/d, the ratio of labor share to capital share in the industrial sector, increases s0 and flattens the supply curve. We can now combine^the deimand curve in (1) and the supply curve in (6) to obtain solutions for PQ and Q. Again, this can be done by substitution or in matrix form. Using the sQ notation, (1) and (6) can be combined in matrix form: r ~~B.. I; s 0 ° 1('+lYE) *1 1 X (6') - Q 0 SQ1 10 - ~0 pA~ L~~~~~ I - 53 - Appendix V The solution is Q 1 sQ Bi B L The solution for PQ can be written out as PQ -Ml+5Q + (a2+ a2Yl)PE + -f OdYAPA] (7) From the demand curve (1), movement in output is given by Q = M - P. (8) With equation (7), we are now in a positiofi to estimate the effects^of an increase in energy prices, represented by PE, on industrial prices P and the CPI. The ratio of output price increase to energy price increase fr2m (7) is given by ^ ^ ~2 + alYE P/P E =' (9) Q a~C2 + (1-Y ) al, + Remember the definition of S in (6) in trying to go from (7) to (9). The numerator of (9) is the direct effect of an increase in energy prices on costs in industry. a2 is the energy share and yEal gives the increase in labor costs through wages. The denominator is a multiplier. A large share of Q in the CPI raises the multiplier because wages rise in the Q sector. A large capital share f reduces the multiplier because profits are not assumed to be marked as PE rises. An increase in the labor share al reduces the sensitivity of PQ to PE in (9) because it increases the denominator more than the numerator. In terms of the supply curve of Figure 1, when PE rises, and increase in a, gives a bigger shift upward but a flatter curve, with the latter effect dominating. Table V.1: GROSS OUTPUT AND INPIJTS IN MIANUFACTURING, 1979 (billions R ) VA plus Output Less Fuel VA at Fuel, Gross Indirect Input Elec., Factor Elec., Employmernt Sector Output Taxes Cost Gas Cost Gas Costs 31 1653.2 1398.8 965.6 21.7 433.1 454.8 82.3 32 673.5 664.9 466.7 24.0 198.2 222.2 61.6 33 188.0 186.1 126.2 3.9 59.9 186.1 19.5 34 120.2 118.1 73.3 6.2 44.7 50.9 13.4 35 886.3 872.5 647.5 16.1 224.9 241.0 57.9 36 211.0 207.0 93.8 30.1 113.2 143.3 20.1 37 68.4 67.3 50.3 5.1 16.9 22.0 2.9 38 690.1 665.3 470.2 9.9 195.1 205.0 56.2 37 19.5 18.8 14.7 0.2 4.2 4.4 1.6 Source: Biro Pusat Statistik, Industrial Statistics, 1979. - 55 - Appendix V Table V.2: SHARES OF VALUE ADDED Sector Full VA + Labor Capital Fuel Share 31 .05 .18 .77 32 .11 .28 .61 33 .02 .10 .78 34 .12 .26 .62 35 .07 .22 .71 36 .21 .14 .65 37 .23 .13 .64 38 .05 .27 .68 39 .05 .36 .59 Mean .08 .21 .71 Source: Table 1