Conclusion Suggestion Efficiency of Rice Milling Industry, Case Study: Two Sub-Districts in Cianjur Regency, West Java Province

Rice Miller Type Rice Broken Rice Rice Bran Chaff Grain Working hour Fuel 33 1 200.00 0.00 30.22 71.99 302.21 2.00 3.00 34 1 150.00 0.00 38.68 52.21 240.89 2.00 2.00 35 3 1500.00 0.00 534.00 636.00 2670.00 6.00 10.50 36 1 100.00 0.00 35.60 42.40 178.00 6.00 1.00 37 1 70.00 0.00 18.69 35.91 124.60 2.00 2.00 38 3 2500.00 89.00 667.50 1193.50 4450.00 7.00 20.00 39 2 1150.00 0.00 307.05 589.95 2047.00 6.00 30.00 40 1 1500.00 0.00 352.68 498.50 2351.18 5.00 8.00 41 1 100.00 0.00 32.69 30.76 163.45 3.00 1.67 42 1 150.00 0.00 35.27 49.85 235.12 2.00 2.00 43 1 300.00 0.00 31.73 121.58 453.31 4.00 2.00 44 3 2500.00 350.88 438.60 1096.49 4385.96 8.00 30.00 45 2 4000.00 178.18 1781.82 1167.27 7127.27 6.00 43.20 46 2 3500.00 118.28 1182.76 1112.76 5913.79 7.00 31.50 47 1 4000.00 0.00 1147.54 1409.84 6557.38 5.00 20.00 48 1 233.00 0.00 60.09 81.10 374.18 3.50 3.33 49 2 1500.00 288.46 721.15 375.00 2884.62 7.00 35.00 50 1 900.00 0.00 232.10 313.25 1445.34 9.00 12.00 51 3 2300.00 334.55 710.91 836.36 4181.82 6.00 15.00 52 1 600.00 0.00 154.73 208.83 963.56 6.00 10.00 53 2 3000.00 100.00 1000.00 900.00 5000.00 8.00 30.00 54 1 300.00 0.00 77.37 104.42 481.78 3.00 5.00 55 1 100.00 0.00 25.79 34.81 160.59 1.00 1.00 56 1 250.00 0.00 83.33 83.33 416.67 3.00 1.50 57 1 300.00 0.00 77.37 104.42 481.78 5.00 4.00 58 3 2200.00 67.69 338.46 778.46 3384.62 7.00 20.00 59 3 2500.00 104.17 641.03 761.22 4006.41 5.00 25.00 60 1 2000.00 0.00 494.48 802.04 3296.52 4.00 10.00 61 1 2500.00 0.00 595.24 873.02 3968.25 8.00 20.00 62 3 4500.00 142.86 1428.57 1071.43 7142.86 8.00 40.00 63 1 500.00 0.00 128.94 174.03 802.97 7.00 10.00 64 1 200.00 0.00 59.56 53.93 313.49 4.00 5.00 Rice Miller Type Rice Broken Rice Rice Bran Chaff Grain Working hour Fuel 65 1 300.00 0.00 76.50 73.49 449.99 6.00 4.00 66 2 1000.00 85.58 479.23 146.73 1711.54 7.00 13.00 67 3 7000.00 1165.21 1296.30 3501.46 12962.96 8.00 80.00 68 3 2300.00 57.50 383.33 1092.50 3833.33 8.00 43.00 69 2 1000.00 48.39 161.29 403.23 1612.90 8.00 10.00 70 1 125.00 0.00 28.94 38.97 192.91 2.00 2.00 71 1 4800.00 0.00 1142.86 1676.19 7619.05 8.00 70.00 72 1 300.00 0.00 70.31 98.44 468.75 2.00 4.00 73 2 3000.00 26.73 1069.09 1249.64 5345.45 7.00 27.00 74 3 3000.00 33.28 832.08 1681.81 5547.17 6.00 27.63 75 3 6620.00 101.85 712.92 2749.85 10184.62 8.00 60.00 76 1 600.00 0.00 154.73 208.83 963.56 4.00 5.00 77 1 2300.00 0.00 593.13 800.52 3693.65 6.00 20.00 78 3 4000.00 210.00 1050.00 1740.00 7000.00 7.00 44.44 79 1 3000.00 0.00 1147.88 1591.51 5739.39 5.00 22.22 80 3 4000.00 70.00 1050.00 1880.00 7000.00 8.00 44.45 81 3 1000.00 72.59 362.96 379.26 1814.81 4.00 10.00 82 1 600.00 0.00 154.73 208.83 963.56 3.00 4.40 83 1 400.00 0.00 36.67 174.43 611.10 3.00 3.00 84 3 2000.00 38.43 384.31 1420.39 3843.14 8.00 15.00 85 3 2300.00 113.72 568.61 808.41 3790.74 8.00 23.00 86 1 2300.00 0.00 751.88 707.53 3759.41 4.00 25.00 87 1 300.00 0.00 92.31 69.23 461.54 4.00 3.00 88 2 6000.00 588.00 1764.00 3408.00 11760.00 8.00 27.50 89 1 3150.00 0.00 812.33 1096.36 5058.69 8.00 32.00 90 2 4000.00 133.33 1000.00 1533.33 6666.67 8.00 30.00 91 3 7000.00 107.69 3230.77 430.77 10769.23 6.00 70.00 92 2 2000.00 64.52 483.87 677.42 3225.81 4.00 50.00 93 3 5000.00 37.50 800.00 2227.02 8064.52 8.00 50.00 94 3 2000.00 32.26 290.32 903.23 3225.81 4.00 20.00 Note: In the type column, 1= makloon type 2= non-makloon type 3= combination type Appendix 3 Constant Return to Scale Technical Efficiency Scores, Variable Return to Scale Technical Efficiency Scores, Scale Efficiency, and Return to Scale of Each Rice Milling Business Rice Miller Type CRSTE VRSTE SE NIRS 1 1 1.000 1.000 1.000 MPSS 2 1 1.000 1.000 1.000 MPSS 3 3 1.000 1.000 1.000 MPSS 4 1 1.000 1.000 1.000 MPSS 5 1 1.000 1.000 1.000 MPSS 6 3 1.000 1.000 1.000 MPSS 7 1 1.000 1.000 1.000 MPSS 8 1 1.000 1.000 1.000 MPSS 9 3 1.000 1.000 1.000 MPSS 10 1 1.000 1.000 1.000 MPSS 11 1 1.000 1.000 1.000 MPSS 12 1 1.000 1.000 1.000 MPSS 13 2 1.000 1.000 1.000 MPSS 14 3 1.000 1.000 1.000 MPSS 15 1 1.000 1.000 1.000 MPSS 16 3 1.000 1.000 1.000 MPSS 17 3 1.000 1.000 1.000 MPSS 18 2 1.000 1.000 1.000 MPSS 19 2 1.000 1.000 1.000 MPSS 20 1 1.000 1.000 1.000 MPSS 21 1 1.000 1.000 1.000 MPSS 22 1 1.000 1.000 1.000 MPSS 23 1 1.000 1.000 1.000 MPSS 24 1 1.000 1.000 1.000 MPSS 25 1 1.000 1.000 1.000 MPSS 26 1 1.000 1.000 1.000 MPSS 27 1 1.000 1.000 1.000 MPSS 28 3 1.000 1.000 1.000 MPSS 29 1 1.000 1.000 1.000 MPSS 30 1 1.000 1.000 1.000 MPSS 31 1 1.000 1.000 1.000 MPSS 32 3 1.000 1.000 1.000 MPSS 33 1 1.000 1.000 1.000 MPSS 34 1 1.000 1.000 1.000 MPSS Rice Miller Type CRSTE VRSTE SE NIRS 35 3 1.000 1.000 1.000 MPSS 36 1 1.000 1.000 1.000 MPSS 37 1 1.000 1.000 1.000 MPSS 38 3 1.000 1.000 1.000 MPSS 39 2 1.000 1.000 1.000 MPSS 40 1 1.000 1.000 1.000 MPSS 41 1 1.000 1.000 1.000 MPSS 42 1 1.000 1.000 1.000 MPSS 43 1 1.000 1.000 1.000 MPSS 44 3 1.000 1.000 1.000 MPSS 45 2 1.000 1.000 1.000 MPSS 46 2 1.000 1.000 1.000 MPSS 47 1 1.000 1.000 1.000 MPSS 48 1 1.000 1.000 1.000 MPSS 49 2 1.000 1.000 1.000 MPSS 50 1 1.000 1.000 1.000 MPSS 51 3 1.000 1.000 1.000 MPSS 52 1 1.000 1.000 1.000 MPSS 53 2 1.000 1.000 1.000 MPSS 54 1 1.000 1.000 1.000 MPSS 55 1 1.000 1.000 1.000 MPSS 56 1 1.000 1.000 1.000 MPSS 57 1 1.000 1.000 1.000 MPSS 58 3 1.000 1.000 1.000 MPSS 59 3 1.000 1.000 1.000 MPSS 60 1 1.000 1.000 1.000 MPSS 61 1 1.000 1.000 1.000 MPSS 62 3 1.000 1.000 1.000 MPSS 63 1 1.000 1.000 1.000 MPSS 64 1 1.000 1.000 1.000 MPSS 65 1 1.000 1.000 1.000 MPSS 66 2 1.000 1.000 1.000 MPSS 67 3 1.000 1.000 1.000 MPSS 68 3 1.000 1.000 1.000 MPSS Note: In the type column, 1= makloon type 2= non-makloon type 3= combination type Rice Miller Type CRSTE VRSTE SE NIRS 69 2 1.000 1.000 1.000 MPSS 70 1 1.000 1.000 1.000 MPSS 71 1 1.000 1.000 1.000 MPSS 72 1 1.000 1.000 1.000 MPSS 73 2 1.000 1.000 1.000 MPSS 74 3 1.000 1.000 1.000 MPSS 75 3 1.000 1.000 1.000 MPSS 76 1 1.000 1.000 1.000 MPSS 77 1 1.000 1.000 1.000 MPSS 78 3 1.000 1.000 1.000 MPSS 79 1 1.000 1.000 1.000 MPSS 80 3 1.000 1.000 1.000 MPSS 81 3 1.000 1.000 1.000 MPSS 82 1 1.000 1.000 1.000 MPSS 83 1 1.000 1.000 1.000 MPSS 84 3 1.000 1.000 1.000 MPSS 85 3 1.000 1.000 1.000 MPSS 86 1 1.000 1.000 1.000 MPSS 87 1 1.000 1.000 1.000 MPSS 88 2 1.000 1.000 1.000 MPSS 89 1 1.000 1.000 1.000 MPSS 90 2 1.000 1.000 1.000 MPSS 91 3 1.000 1.000 1.000 MPSS 92 2 1.000 1.000 1.000 MPSS 93 3 1.000 1.000 1.000 MPSS 94 3 1.000 1.000 1.000 MPSS mean 1.000 1.000 1.000 Appendix 4 Summary of Output and Input Slacks of All Respondents Rice Miller NIRS Rice Broken Rice Rice Bran Chaff Grain Working hour Fuel 1 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 1.433 2 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 3 MPSS 0.000 0.000 0.000 0.000 0.000 3.896 0.000 4 MPSS 0.000 0.000 0.000 0.000 0.000 1.000 0.000 5 MPSS 0.000 0.000 0.000 0.000 0.000 0.182 1.373 6 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 7 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 8 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.960 9 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 10 MPSS 0.000 0.000 0.000 0.000 0.000 0.840 0.210 11 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 12 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 2.072 13 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 14 MPSS 0.000 0.000 0.000 0.000 0.000 1.977 0.000 15 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.651 16 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 17 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 18 MPSS 0.000 0.000 0.000 0.000 0.000 1.548 0.000 19 MPSS 0.000 0.000 0.000 0.000 0.000 2.890 17.702 20 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 21 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 22 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 23 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 24 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 25 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 26 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 2.134 27 MPSS 0.000 0.000 0.000 0.000 0.000 2.000 0.000 28 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 29 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 30 MPSS 0.000 0.000 0.000 0.000 0.000 0.121 0.248 31 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 32 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 Rice Miller NIRS Rice Broken Rice Rice Bran Chaff Grain Working hour Fuel 33 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 34 MPSS 0.000 0.000 0.000 0.000 0.000 0.060 0.124 35 MPSS 0.000 0.000 0.000 0.000 0.000 2.382 0.000 36 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 37 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 38 MPSS 0.000 0.000 0.000 0.000 0.000 1.351 0.000 39 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 40 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 41 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 42 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.290 43 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 44 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 45 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 46 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 47 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 48 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 49 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 50 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 51 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 52 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 1.501 53 MPSS 0.000 0.000 0.000 0.000 0.000 2.212 4.726 54 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 55 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 56 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 57 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 1.000 58 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 59 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 60 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 61 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 62 MPSS 0.000 0.000 0.000 0.000 0.000 2.326 6.222 63 MPSS 0.000 0.000 0.000 0.000 0.000 2.001 3.501 64 MPSS 0.000 0.000 0.000 0.000 0.000 1.661 2.729 Rice Miller NIRS Rice Broken Rice Rice Bran Chaff Grain Working hour Fuel 65 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 66 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 67 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 68 MPSS 0.000 0.000 0.000 0.000 0.000 3.951 21.317 69 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 70 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 71 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 72 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 1.503 73 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 74 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 75 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 76 MPSS 0.000 0.000 0.000 0.000 0.000 0.691 0.000 77 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 2.949 78 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 6.732 79 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 80 MPSS 0.000 0.000 0.000 0.000 0.000 1.152 10.400 81 MPSS 0.000 0.000 0.000 0.000 0.000 1.458 0.175 82 MPSS 0.000 0.000 0.000 0.000 0.000 0.901 0.000 83 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 84 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 85 MPSS 0.000 0.000 0.000 0.000 0.000 1.809 1.481 86 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 87 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 88 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 89 MPSS 0.000 0.000 0.000 0.000 0.000 1.020 5.181 90 MPSS 0.000 0.000 0.000 0.000 0.000 1.548 0.000 91 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 92 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 9.363 93 MPSS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 94 MPSS 0.000 0.000 0.000 0.000 0.000 0.532 1.239 mean 0.000 0.000 0.000 0.000 0.000 0.420 1.141 Note: The unit with the zero slack of all variables is highlighted in bold in the first column Appendix 5 Summary of Frontier of All Respondents Rice Miller 1 2 3 4 5 6 1 65 22 83 40 55 2 2 3 55 67 6 9 88 16 4 83 39 61 55 22 5 55 25 65 33 6 6 7 7 8 22 83 61 55 40 9 9 10 41 87 36 31 11 11 12 55 86 29 22 48 13 13 14 91 67 88 16 46 9 15 55 61 83 22 39 16 16 17 17 18 91 67 88 16 75 73 19 59 69 55 16 91 20 20 21 21 22 22 23 29 86 79 50 61 31 24 24 25 25 26 65 83 22 43 55 27 29 28 28 29 29 30 55 22 48 86 31 31 32 32 Rice Miller 1 2 3 4 5 6 33 33 34 55 48 86 22 35 87 2 31 86 79 36 36 37 37 38 74 73 88 69 75 55 39 39 40 40 41 41 42 43 65 55 22 83 43 43 44 44 45 45 46 46 47 47 48 48 49 49 50 50 51 51 52 86 50 61 22 54 53 91 55 46 16 9 54 54 55 55 56 56 57 27 58 58 59 59 60 60 61 61 62 75 46 91 16 67 63 22 55 61 50 64 86 55 87 31 Rice Miller 1 2 3 4 5 6 65 65 66 66 67 67 68 75 67 84 74 31 69 69 70 70 71 71 72 40 55 24 33 65 73 73 74 74 75 75 76 55 61 47 86 29 77 47 79 61 86 31 78 73 67 6 75 91 88 79 79 80 73 88 75 93 79 81 45 79 9 73 31 82 40 61 55 47 86 83 83 84 84 85 6 59 44 69 61 86 86 87 87 88 88 89 86 71 47 61 90 91 67 88 16 75 73 91 91 92 9 67 16 50 6 55 93 93 94 67 75 84 74 31 ABSTRACT CILA APRIANDE. Efficiency of Rice Milling Industry, Case Study: Two Sub- Districts in Cianjur Regency, West Java Province. Supervised by RACHMAT PAMBUDY, NUNUNG KUSNADI, and STEPHAN VON CRAMON- TAUBADEL. In rice agribusiness system of Indonesia, rice milling industry has an important role. This industry plays role as connector between paddy producers farmers and consumers of rice. The industry is dominated by small scale and old machine used. There are three types of rice miller business management, namely makloon, non-makloon, and combination of both. This study aims to describe characteristics and determine relative efficiency of rice milling industry. The study was conducted in Gekbrong and Warungkondang, Cianjur Regency as one of largest paddy producer in West Java. 94 rice millers were selected purposively as sample. Mostly, owner of rice millers were male, rice miller business as main job, and ownership of rice miller was private. Rice milling industry was dominated by makloon type. This type offered milling service to consumers and had small capacity. Variable return to scale DEA output orientated model was used to determine relative efficiency of rice milling industry. This study concluded that rice milling industry in study site was inefficient. Key words: efficiency, rice milling industry, DEA ABSTRAK CILA APRIANDE. Efisiensi Industri Penggilingan Padi, Studi Kasus: Dua Kecamatan di Kabupaten Cianjur, Provinsi Jawa Barat. Dibimbing oleh RACHMAT PAMBUDY, NUNUNG KUSNADI, dan STEPHAN VON CRAMON-TAUBADEL. Pada sistem agribisnis beras di Indonesia, industri penggilingan padi memiliki peranan penting. Industri ini berperan sebagai penghubung antara produsen padi petani dan konsumen beras. Industri ini didominasi oleh penggilingan padi skala kecil dan kondisi mesin yang digunakan sudah tua. Terdapat tiga jenis manajemen bisnis penggilingan padi, yaitu makloon, non- makloon, dan kombinasi keduanya. Penelitian ini bertujuan untuk mendeskripsikan karakteristik dan menentukan efisiensi relatif industri penggilingan padi. Penelitian dilakukan di Kecamatan Gekbrong dan Warungkondang, Kabupaten Cianjur sebagai salah kabupaten penghasil beras terbesar di Jawa Barat. 94 penggilingan padi dipilih secara purposive sebagai sampel. Sebagian besar, pemilik penggilingan padi adalah pria, penggilingan padi adalah sebagai pekerjaan utama, dan kepemilikan penggilingan padi adalah privat. Industri penggilingan padi didominasi oleh jenis makloon. Penggilingan jenis ini menawarkan jasa giling kepada konsumen dan berkapasitas kecil. Model variabel return to scale DEA berorientasi output digunakan untuk menentukan efisiensi relatif industri penggilingan padi. Penelitian ini menyimpulkan bahwa industri penggilingan padi di lokasi penelitian inefisien. Kata kunci: efisiensi, industri penggilingan padi, DEA SUMMARY CILA APRIANDE. Efficiency of Rice Milling Industry, Case Study: Two Sub- Districts in Cianjur Regency, West Java Province. Supervised by RACHMAT PAMBUDY, NUNUNG KUSNADI, and STEPHAN VON CRAMON- TAUBADEL. Rice is one of the main staple foods in Indonesia. In 2012, Indonesian consumption of rice is high, approximately 139 kilogram per capita per year Pambudy, 2012. Rice is a political commodity that can be an indicator for national stability. Rice is also an important economic indicator in which, rice price is able to reflect the inflation rate and the minimum income in Indonesia. Rice has an important role for the life of Indonesia’s society. Hence, availability of the rice must be able to be guaranteed. Governmental efforts to keep the availability of rice are done through established policies, ranging from the production, distribution, and consumption of rice. Rice milling industry as an important link in paddy processing into rice is required to contribute provision of national rice in terms of quantity and quality. Performance of rice milling industry needs to be developed and improved Budiharti, Harsono, Juliana, 2003. In 2002, the number of the rice milling is 109,000 units. It is dominated by small-scale by 95 percent and the rest is large- scale. The average yield that is produced by small rice milling is still low at only 60 percent, medium rice milling is 64 percent, and large rice milling is 65 percent of each dry milled grain that milled in each rice milling. Generally, small scale rice millings are an investment in the 1960s until 1980s Sawit, 2011. While in 2008, the number of rice milling is decrease. It is about 108,512 units Thahir, 2010. Rice milling industry plays an important role in the processing side, is expected to work efficiently and effectively, in order to increasing a national rice production. This is especially with respect to start attainment surplus by 10 million tons of rice in 2014. This is evidenced by milling ratio and quality of rice produced Nazaruddin, 2012. This study is aimed to describe characteristic and determine relative efficiency of rice milling industry in Gekbrong and Warungkondang, Cianjur Regency, West Java Province. This study used 94 rice millers that selected purposively, which in 44 units are in Gekbrong and 50 units are in Warungkondang by used questionnaire. Data processing was conducted by used output orientated and variable return to scale VRS DEA Data Envelopment Analysis model. Determination of efficiency was reffered to Koopmans definition. Rice miller was efficient if operates on the frontier efficiency scoreES equal to one and achieves zero slack of all variables used. In addition, to overcome the difference between to sub-districts, a comparative analysis was conducted by using statistical Minitab Release 13.20. Mostly, owner of rice millers were male, rice miller business as main job, and ownership of rice miller was private. Rice milling industry was dominated by makloon type. This type offered milling service to consumers and had small capacity. This study concluded that rice milling industry in study site was inefficient. This study was case study of rice milling industry. So, it could not be generalized to general condition of Indonesian including differences of sample size, observations type, location, and so forth. Further research can uses other method or tools programs to examine the efficiency of rice miller and also determine factors affecting inefficiency of rice miller, conducting research in a different location, and so forth related to efficiency of rice milling industry to obtain information that are not captured in this study. Government is expected to conduct research on rice mill industry efficiency nationally. The research is addressed to obtain information the efficiency of various types of rice miller in all provinces. So, it can be used as consideration in determining the appropriate policy for this industry. Policies are not only considering producers farmers and consumers but also rice milling industry as an industry that linking producers and consumers in rice agribusiness system. Key words: efficiency, rice milling industry, DEA 1 INTRODUCTION Rice is one of the main staple foods in Indonesia. In 2012, Indonesian consumption of rice is high, approximately 139 kilogram per capita per year Pambudy, 2012. Each household, on the average, spends 25 percent of their income for rice. Even, the percentage will get greater to the poor. Although there has been diversification program that issued by the government, rice is the most preferred food by most of Indonesian. Therefore, the staple food has become strategic for politics, economics, and business in Indonesia. Rice is a political commodity that can be an indicator for national stability. Rice is also an important economic indicator in which, rice price is able to reflect the inflation rate and the minimum income in Indonesia. In addition, rice is one of the five commodities that are commodities focus of development Directorate General of Marketing and Processing of Agriculture, 2010. Rice has an important role for the life of Indonesia’s society. Hence, availability of the rice must be able to be guaranteed. Governmental efforts to keep the availability of rice are done through established policies, ranging from the production, distribution, and consumption of rice. In production side, in order to meet domestic demand for rice, the government continues to increase Indonesian national production through a variety of policies. This corresponds with UU NO.7 Th. 1996 Indonesian law concerning food. This policy is carried out in two ways. First is intensification way. This is done by increasing the productivity of crops and cropping index. Second is extensification way. This is more emphasis on increasing crop land area. In implementation, government by means of Ministry of Agriculture makes a few programs to increase production of rice. This program is known as Program Peningkatan Produksi Padi nasionalP4 National Rice Production Enhancement Program Baga, Pratiwi, Firdaus, 2008. First program of P4 is Program Padi Sentra Centers Paddy Program, started in 1959. However, this program is unsuccessful. This is caused by less number of farmers as participants. Hence, government changes to the other program, namely Program Bimibingan MasalBimas Mass Mentoring Program and Program Intensifikasi KhususInsus Special Intensification Program. The programs started in 1965 through SK Mentan No. 003 Tahun 1979. In 1984, Indonesia achieves self-sufficiency of rice through Panca Usahatani technology Five Farming technology. Government continues to evaluate the P4 program corresponds with environment change, both of nature and socioeconomic Baga, Pratiwi, Firdaus, 2008. Recently, production policy is known as Program Peningkatan Beras NasionalP2BN National Rice Improvement Program. This program started in 2007 with target to increase the production per year either of rice by 2 million tons or of grain by 5 percent for domestic stock. This is also to reduce import in order to achieve self-sufficiency in 2015. Some provinces in Indonesia plays role as rice producers. Based on data of Statistics Indonesia, paddy is widely produced in the island of Java. It shows on Table 1, where the three provinces on Java Island are the big paddy producers in Indonesia. Table 1 Harvest Area Ha, Productivity quintalHa, and Production Ton of 10 Provinces Biggest Paddy Producers in Indonesia 2011 Province Harvest Area Ha Productivity quintalHa Production Ton West Java 1,964,457.00 59.22 11,633,836.00 East Java 1,926,796.00 54.89 10,576,543.00 Central Java 1,724,246.00 54.47 9,391,959.00 South Sulawesi 889,232.00 50.73 4,511,336.00 North Sumatera 757,428.00 47.62 3,607,036.00 South Sumatera 784,820.00 43.09 3,381,751.00 Lampung 606,973.00 48.45 2,940,795.00 West Sumatera 461,711.00 49.37 2,279,442.00 West Nusa Tenggara 418,062.00 49.45 2,067,137.00 South Kalimantan 489,134.00 41.67 2,038,309.00 National 13,201,316.00 49.80 65,740,946.00 Source: Statistics Indonesia, 2011 In the 2011, West Java is the biggest paddy-producing province and it produced paddy was 11,633,836.00 tons. While Indonesian national production of paddy is equal to 65,740,946.00 tons. It means that West Java accounts for 17.70 percent of national paddy production. In addition, West Java also has the largest harvest area and the highest productivity. East Java and Central Java placed second and third as a province of paddy producers. Each province produced paddy was 10,576,543.00 tons and 9,391,959.00 tons. Cianjur is one of regencies that are produce paddy in West Java province. In 2010, Cianjur produced paddy about 862,229.00 tons. It has harvest area about 159,229.00 hectares with the productivity of 54.15 quintal per hectare. The biggest province in West Java that produced paddy was Indramayu with total production by 1,358,441.00 tons. It shows on Table 2. Harvest Area Ha, Productivity quintalHa, and Production Ton of 10 Provinces Biggest Paddy Producers in West Java Province year of 2010. Table 2 Harvest Area Ha, Productivity quintalHa, and Production Ton of 10 Provinces Biggest Paddy Producers in West Java Province 2010 Regency Harvest Area Ha Productivity quintalHa Production Ton Indramayu 240,716.00 56.43 1,358,441.00 Karawang 187,892.00 59.29 1,113,978.00 Subang 169,462.00 54.28 919,789.00 Garut 147,426.00 60.65 894,197.00 Cianjur 159,229.00 54.15 862,229.00 Tasikmalaya 138,247.00 61.65 851,108.00 Sukabumi 146,825.00 54.89 805,924.00 Ciamis 117,295.00 61.8 724,842.00 Bekasi 100,966.00 58.44 590,043.00 Majalengka 103,392.00 56.16 580,638.00 Source: Statistics Indonesia of West Java, 2010 In distribution side, government set a distribution policy. The objective of this policy is to guarantee stock of food along the year smoothly and achievable. This is important because Indonesia has large population and also wide and spread geographical coverage. Hence, since 1967, government provides authority to BULOG to regulating domestic rice stock and stabilizing rice price. In Indonesia, rice distribution process is implemented by BULOG and market mechanism. BULOG has only market share by 10 percent of total national market and the rest is controlled by market mechanism Baga, Pratiwi, Firdaus, 2008. In price side, policy is made with respect to protect farmers as producer and consumer through price stabilization mechanism. Since 1970, government issued a floor price policy for grain and rice. The objective of this policy is to provide price guarantee for farmers that the production will be purchased in accordance with the price set by the government. So it becomes an incentive for farmers to increase the production. The policy will further help for farmers in the harvest. Farmers will not get a lowest price because of excess supply occur in the market. On the other hand, government sets a ceiling price to protect consumers. A ceiling price is the maximum price that might be applied by producers to the consumers. This policy will further help for consumers in the bad season. The consumers will not get a highest price because of excess demand occur in the market. Referring to Inpres No.9 Tahun 2002 President Instruction concerning of determination of rice policy, government change Harga Dasar GabahHDG Floor Price of Grain into Harga Dasar Pembelian PemerintahHDPP Floor Price of Government Purchasing is also known as Harga Pembelian PemerintahHPP Government Purchasing Price. Fundamental difference of both policies is on government guaranteed price level. In HDPP HPP policy, government only guarantee price of grain on the certain level at the site which has been set. In order to implement both HPP policy and ceiling price, government provides authority to BULOG by SK Mendag No.1111 Tahun 2007 to keep price domestic price stabilization of rice. Based on SK Mendag No. 1109 Tahun 2007 declares that control both of monopoly of price and of import policy are return back to BULOG Baga, Pratiwi, Firdaus, 2008. The other form of price policy is market operation both of Operasi Pasar MurniOPM Pure Market Operation and Operasi Pasar KhususOPK Special Market Operation. OPM is part of general price subsidy which is used at the highest price occur because of excess demand in the market. It is done by cutting of price by 10-15 percent below market price. Otherwise, OPK is implementation of targeted price subsidy to distribute food aid to poor people. Since 2002, this operation changes to Beras untuk Keluarga MiskinRaskin Rice for Poor Families Baga, Pratiwi, Firdaus, 2008. The policies made by the government are expected to strengthen the rice agribusiness system. The rice agribusiness system consists of several subsystems from upstream to downstream. The system is built by some industries, one of which is the rice milling industry. The system is depicted in Figure 1. Figures 1 illustrates the rice milling industry is on the processing subsystem. This subsystem is between farmers and consumers. Farmers as producers provide input in the form of grain and then this industry processes it into the rice that either ready to cook by consumers or ready to stock by seller. The government has role to give an incentive for farmers to produce qualified rice. It is done through policy implementation, such as HPP. This policy gives an incentive because there Figure 1 Rice Agribusiness System is guarantee of price for farmers. This policy makes more influence in the harvest season while production of paddyrice is higher. The guarantee of price increases the incentive for all parts in the rice agribusiness system to produce better quality rice and reduce the yield loss that occurs at every level. This point is done to strengthen linkages between primary industries on farm and a processing industry milling, to be equally strong. So, farmers can increase income in a sustainable manner as well as the milling business. Then, it can create a good and solid rice agribusiness system Sawit, 2011. Rice milling industry as an important link in converting paddy into rice is required to contribute provision of national rice in terms of quantity and quality. This industry increase value added of paddygrain. Rice has a higher economic value when sold than sold in the grain form. This industry converts paddygrain which is produced by farmers into rice that is ready to be cooked or stored. The role of government is needed to improve performance and develop this industry in order to create an integrated rice agribusiness. Rice production both quality and quantity will give a large effect on national rice trade. The quality of rice produced would affect the competitiveness of the national rice products. Good quality rice will become a force for Indonesia in competing with the other countries in both domestic and foreign markets. In addition, the quantity of rice produced would affect the amount of rice supply in the market. If the number of nationally produced rice cannot cover the needs of the community, then this will affect the rice import policy. Improvement of rice quality is determined by a few factors in every subsystem. In upstream subsystem, agricultural inputs quality produced in this subsystem affect quality of paddygrain produced on the subsystem on farm, especially quality of seed. In on farm subsystem, farming technique and input factors used affect the quality of paddy produced both quality and quantity. In off farm subsystem, paddygrain plays role as one of input factors. Good quality of paddygrain produces also good quality of rice. In addition, machine condition also affects both quality and quantity of rice produced. All processes in every subsystem affect the quality and yield of rice quantity. Rice millers can be divided into several categories based on a few distinguishing features, including production capacity, level of technology, business management and so forth. Based on production capacity, rice millers are grouped into three categories J., 2010. First is small rice miller consisting of two units of machines in separated pairs, namely husker and polisher. In general, transfer of material from husker to polisher is done by human manually. Real production capacity is about 0.3-0.7 ton of rice per hour. Second is large rice miller comprising of a complete set of cleaner, husker, separator, polisher, shifter, grader, and so on. Material transfer is done by using elevator. Real production capacity is more than 0.7 ton of rice per hour. Third is rice milling unit consisting of husker and polisher inseparable. Real production capacity is 0.3-0.7 ton of rice per hour. Based on level of technology, rice milling in Indonesia is divided into two types Winarno, 2004. First is medium-large rice mill commercial. Second is service custom rice miller offering milled service in small quantities. In 2002, the number of the rice milling in Indonesia is 109,000 units. This is dominated by small-scale by 95 percent and the rest is large-scale. The average yield that is produced by small rice milling is still low at only 60 percent, medium rice milling is 64 percent, and large rice milling is 65 percent of each dry milled grain that milled in each rice milling. Generally, small scale rice millings are an investment in the 1960s until 1980s Sawit, 2011. While in 2008, the number of rice milling is decrease. It is about 108,512 units Thahir, 2010. In 2009, there are 2,028 units of rice miller in Cianjur Regency Kabupaten Cianjur, 2009. This number is dominated by small rice miller about 90.24 percent. The rest is large rice miller. Gekbrong and Warungkondang are sub- districts in Cianjur being study site. In study site, based on preliminary survey data, there are three types rice miller, namely makloon, non-makloon, and combination of both. This is divided by management business. Explanations above raise some questions related to this industry in this study site. What is characteristic of rice milling industry? What kind of rice millers scattered? Are production factors efficient? Is rice milling industry efficient? According to those research questions, the objectives of this study are: 1. To describe characteristics and types of rice miller. 2. To determine relative efficiency of rice milling industry. This study was case study of rice milling industry. So, it could not be generalized to general condition of Indonesian including differences of sample size, observations type, location, and so forth. 2 LITERATURE REVIEW Generally, this chapter reviews some previous studies on rice milling and efficiency relative. First, it starts with overview of rice milling industry in Indonesia. Second, it describes a few previous studies about efficiency of production function. Finally, the last section of this chapter gives information about efficiency study in the previous studies.

2.1 Rice Milling Industry Overview

Based on production capacity, rice milling business was divided into three categories. First, Small Scale Rice Mill SSRM had a production capacity of 0.5 tons per hour. Most of that consist of Engelberg Rice Mill ERM. Second, Medium Scale Rice Mill MSRM had a production capacity between 0.5-1.0 tons per hour. This category consists of Rice Mill Unit RMU and SSRM. Third, larger scale rice mill LSRM had a production capacity of more than 10 tons per hour Winarno, 2004. In the Karawang regency, production capacity of milling was the main factor that distinguishes between large and small-scale rice milling Arief, 2008. Other differing factors were owned capital, milling machine capacity, partnerships with BULOG Mainstay of Food Security, and the owner education level. In Indonesia, there were a few types of rice milling. It was divided by production capacity, technology, business activity, and so on. In Cikarawang Village, rice millers tend to only offer milling service Chaerunnisa Sd, 2007. Rice miller cannot take chances in selling rice. It was related to a few reasons, including opportunities, funding, and management. Based on the technology usage, rice mill industry in Indonesia was still using simple technology Amsari, 2006. This became one of the causes of low quality and yield of rice produced. Rice produced was less than the production capacity of rice miller in Indonesia. Most rice milling business did not work optimally. Even the average working was approximately one third of maximum capacity. The production capacity was one characteristic to distinguishing rice miller Arief, 2008. This referred to the ability of rice miller to produce rice a day. This was closely related to the engine used, capital, amount of grain, and other related factors. In his research, located in Karawang, the average production capacity of the rice miller was equal to 29.23 tons of rice per day. The largest production capacity was about 60 tons of rice per day and the smallest production capacity was about 20 tons of rice per day. This was related to the ability of large miller to buy grain in large quantities and also supported by the ability of big capital. In addition, small rice miller had an average production capacity about 5.91 tons of rice per day. The largest production capacity was about 15 ton of rice per day and the smallest production capacity was about 2 tons of rice per day. Most of rice miller in Karawang Regency did drying and processing activities only for 9-10 months a year. The remaining time was used to store and repurchase stock for the next process because at that time farmers had a period of rest or famine. When viewed from business activities described in this study, rice millers were a private rice miller. Rice miller did activity start from purchase grain, drying, processing, packaging or storing, until sale of rice.

2.2 Efficiency of Variables Used in Study

There were three performance models in the study of technical and scale efficiencies Australian universities Avkiran, 2001. All model showed slack whether input and output. Input slack means an over-utilized resource. While output slack means less production by uses a set input. Overall, slacks were small. Both slacks represent potential improvements for universities. The other study about efficiency was done in Italian National Parks Bosetti Locatelli, 2006. The study used management costs, variable costs, and area extension as input variables. While in output side, they used number of visitors, number of park employees, number of economic businesses, number of protected species, and number of students. The results showed that partially the Italian National Parks in this study still have potential improvements to obtain efficiency. It means whether input used and output produced still inefficient. The other study was about benchmarking productive efficiency of selected wheat areas. The study used three inputs and single output in her study. There were slacks in two inputs used. It means that areas could reduce both input usage to produce output in the same level. Generally, the wheat areas were inefficient in input used. In addition, it produced output efficiently. It can be seen by zero slack on the output variable.

2.3 Efficiency Relative

The study described the industry in Central Java Rejekiningsih, 2011. Measurement of efficiency for each sector in the industrial processing was performing using the DEA assuming variable returns on scale VRS. It will efficient when the industry has a 100 of efficient score. The study was uses a sample of 21 industries. The data used some of the data output and several inputs used during the period 2000-2005 from each sector in the manufacturing of medium and large. Therefore, the input and output were varies, then the efficiency was calculated by transforming a single input and output through the appropriate weighting. Based on research results, for the period 2000-2005 concluded that the efficiencies were achieved in a very diverse industry. Average efficiency scores for each industry did not reach 100 which means that each sector has not been working efficiently. This suggests of no proper allocation of resources that lead to the less achievement of industrial output. These results are different when compared with the analysis performed for each year. For each year, there are some efficient industries with 100 efficient values. In 2005, the number of the efficient industries more than inefficient industries. The relative efficiency Among South African universities was done Taylor Harris, 2004. This study also used the DEA and the 10 universities as the samples. Relative efficiency analysis performed for each year from 1994 to 1997. The analysis begins by specifying DEA models to be used. Of the seven DEA models are analyzed, the model DEA6 was chosen as the model used in the study. Model selection based on the degree of consistency and stability in the efficiency measures for all universities as samples. Besides, the model was chosen DEA6 because it has high correlation between input and output variables. Based on the results of the study, the average efficiency is different every year. The samples is said to be efficient when its position in appropriate efficiency frontier. However, the limitations of the DEA are not able to explain the reason for the difference efficient and inefficient. The study adds the cost accounting professions analytical review technique to complement the DEA. The study are attempted to explain differences in the efficiencies between the sample characteristics by connecting the inputs used in terms of both quantity and quality of the efficiencies that may occur positive or negative. Generally, South African universities were inefficient in every year. The research is technical and economic efficiency measures under short run profit maximizing behavior Cherchye, Kuosmanen, Leleu, 2010. Differ with previous researches; this research has examined the measurement of economic and technical efficiency in the framework of short run profit maximization behavior, drawing special attention on the corresponding duality relationship. Research is dividing into two steps. First, is by searching alternative profit efficiency PE measures thr ough literature review. Then identify Varian’s percentage profit efficiency measure for evaluating short run profit efficiency. Second, is establishing the dual link between the economic efficiency measure and the quantity based McFadden gauge function. 1 The technical efficiency TE should be interpretable as the PE at the input and output shadow price vectors; 2 The TE should provide an upper bound for the PE at all prices. The McFadden gauge as such a technical efficiency measure. This is capturing the maximal radial expansion of the variable input and output vectors simultaneously. It has considered the choice of measurement direction from the perspective of the evaluated firm within the general directional distance function framework. The McFadden gauge function is the optimal measurement direction for the evaluated firm in the sense that the corresponding reference production plan and associated shadow prices implies minimal profit inefficiency. DEA was used to examine the relative efficiency of Australian universities Avkiran, 2001. This study used cross section data for 1995 of 36 Australian universities. Three performance models were developed, namely overall performance, performance on delivery of educational services, and performance on fee-paying enrolments. This study is under both of output maximization and Variable Return to Scale VRS. The results of this study are list of universities which is efficient and inefficient, list of universities which is operate at Increasing Return to Scale IRS, Most Productive Scale Size MPSS, and Decreasing Return to Scale DRS, and slack of each university. Generally, three performance models showed that universities were inefficient. The other research analyzed the level of bank efficiency on using approach of DEA Jemric Vujcic, 2002. This study analyzed the level of bank efficiency in Croatia. Usage of the DEA method is able to classify efficient and inefficient banks based on a certain size that have been previously defined. The research is also able to identify the factors of leading cause of inefficiency in the banking system in Croatia. DEA method was also used to analyze the technical efficiency of usage of production input in on-farm of paddy rice in Central Java Purnomo, 2006. Usage of the DEA method can determine efficient and inefficient production inputs. In this study, he conducted a study comparison between regression and DEA methods. Based on the study, it can be concluded that the comparative study with many efficient data is bad to estimate the regression mode of production function. Finally, he advised not to use regression to assess technical efficiency. 3 CONCEPTUAL FRAMEWORK This chapter is divided into five parts. First describes about concept of efficiency. Second describes about concept of measurement concept. This part is divided into two sections, namely output and input orientated. Third describes about constant return to scale DEA model. Fourth describes about variable return to scale DEA model. The last part describes about slack.

3.1 Efficiency Concept

Theoretically, efficiency is one of the performance parameters that underlying all of the performance of an organization. Efficiency related to the ability to produce the optimal output to the input that measure the expected performance Abidin Endri, 2009. The efficiency concept was first introduced by Farrell in 1957. The efficiency of a firm usually has definition in two sides. First, when the efficient defines as a successful of a firm in producing many possible outputs from a given set of input technical efficiency. Second, the efficiency will achieve when a firm success to choose an optimal set of input with the same output allocative efficiency. Efficiency is a concept, in which it is related to profit or loss earned by a business. This is concluded from the comparison between the numbers of generated outputs results with the usage of inputs from production activities. From the explanation, the concept of efficiency is related to the concept of production function. A method of production is called efficient if the method produces more output for the same sacrifices, opportunity cost. In other conditions, a method of production is succeeding if the method produces the same output for the smaller sacrifices. Therefore, it is concluded that the concept of efficiency is a relative concept Harianto, 1989. Efficiency is defined by looking at how a work unit achieved an outcome by comparing the inputs used and output produced Suwandi, 2004. Efficiency related to the number of sacrifice spent by the work unit in achieving goals. If it is considered too large, then the work unit is inefficient. Efficiency must always be quantifiable and measureable. Efficient in input used, will produce the high productivity as objective of all organizations in every activity. Efficiency concept contains three terms, namely technical efficiency TE, allocative efficiency AE, and economic efficiency EE. According to Farrell, 1957, EE will occur by itself if the TE and AE can be achieved. Technical and Allocative Efficiencies describe in Figure 2. Source: Farrell, 1957 The line AA’ is the line that describes the relative price of the relative cost of input used in a production process. Relative cost of usage of input will be minimum if the line AA offends isoquant curve SS ΄. The isoquant represents the various combinations of the two factors that a perfectly efficient firm might use to produce output. The point P shows the input of two factors, per unit of output, that the firm is observed to use. The point Q shows an efficient firm using two factors in the same ratio as point P. It produces the same output as P using only 0Q0P of each factor. It also describes of as producing 0P0Q times as much output from the same inputs. It could be defined 0Q0P as the technical efficiency of the firm P Farrell, 1957. Figure 2 Technical and Allocative Efficiencies y x S΄ Q΄ Q R P A ’ A A concept of efficiency is approached from two side approaches, namely the input and output orientations Farrell, 1957. It needs the availability of input price information and an isoquant curve showing combinations of inputs used to produce output at maximum level raises the approach on the input side. The approach on the output side is the approach used to see the extent of the output amount is proportionally increased without changing the number of inputs used.

3.2 Efficiency Measurement Concepts

Firm is efficient while it successes in producing maximum outputs using a given inputs Farrell, 1957. The efficiency of firm consists of two components, technical efficiency TE and allocative price efficiency AE. TE reflects the ability of the firm to achieve maximum outputs from a given set of inputs. AE reflects ability of firm to use the inputs in optimal proportions, given price and production technology respectively. The achievement of both efficiencies implies economic efficiency EE.

3.2.1 Input Orientated

The input orientated TE is measure to answer the ques tion “How much quantity of inputs can be proportionally reduced without changing the output quantity produced?” TE and AE of the case represented in Figure 3. Figure 3 illustrates the unit of isoquant of fully efficient firms, represented by SS΄. If the firm uses a given set of input, defined by the point A, to produce a unit of output, the technical inefficiency of that firm could represent by the distance DA. This is the amount by which all inputs are proportionally reduced without a reduction in output. It also expressed in percentage terms by the ratio DA0A. This is represents the percentage by which all inputs need to be reduced to achieve technically efficient production. The TE of a firm is commonly measured by the ratio as on Equation 3.1. TE i = 0D0A Where: TE i is the input orientated technical efficiency 3.1 Source: Coelli, Rao, Battese, 1998 Equation 3.1 is equal to 1 – DA0A. It will take value between zero and one, and therefore provides an indicator of the degree of technical inefficiency of the firm. A value of one indicates the firm is fully technically efficient. The point D is technically efficient because it lies on the efficient isoquant. If the input price is also known, represent ed by CC΄ the slope of isocost in Equation 3.1, permits the measurement of AE. The AE of the firm operating at A is defined to be ratio as on Equation 3.2. AE i = 0B0D Where: AE i is the input orientated allocative efficiency 3.2 The distance BD represents the reduction in production costs. If the production were to occur at point D, the allocatively and technically efficient would occur. While the point D΄ shown the technically efficient but allocatively inefficient. The achievement of the TE and AE implies EE. The EE is defined to be ratio as Equation 3.3. x 2 y S A D B D΄ S΄ x 1 y C΄ C Figure 3 Technical and Allocative Efficiencies of Input Orientation EE i = 0B0A Where: EE i is the input orientated economic efficiency 3.3 The distance BA could also be interpreted in terms of cost reduction. The EE is also could be calculated by Equation 3.4. TE i x AE i = 0D0A x 0B0D 3.4 It could be done because the product of the TE and AE measures provides the measures of overall economic efficiency. All of three measures bounded by zero and one.

3.2.2 Output Orientated

The alternative question about eff iciency is “How much quantity of outputs can be proportionally increased without changing the input quantity used?” The output oriented measures is opposed to the input oriented measures. The difference between the input and output oriented measures can be seen in Figure 4, which is using a simple example involving one input x and one output y. Source: Coelli, Rao, Battese, 1998 Figure 4 Input and Output Orientated TE Measures and Return to Scale a b fx fx P P R R Q Q S S y y x x A A