Agricultural sector is considered the bedrock of Pakistan economy in terms of its contribution in GDP, export earning, labour force absorption and supply of raw materials for industrial units. However, its productivity was not sustain to meet the dietary requirements of the growing population. Because of various limitations faced by farming community, per unit yield in Pakistan has been graded in the lower to middle ranged economics of the world. At national level due to rapid population growth per capita land and water availability is squeezing with the passage of time, therefore sustainable growth rate in agricultural sector is utmost necessary to ensure food security and overall national development. Sugarcane is a major cash crop that serves as a building block for sugar industry in Pakistan. At national level about 8.76 million farmers are involved in sugarcane production along with employment of 100,000 people in sugar industry. It was grown on an area of 1.14 million hectares with production of 62.65 million tonnes during 2014-15. Sugarcane national yield level (50 tonns/hectare) was noted low compare to international level (70 tonnes/hectare), however, sugar consumption (25.83 kg per capita) during the year 2004-05 was highest in Pakistan compare to other South Asian countries. The present study was designed to calculate per acre gross revenue, total cost and net return along with technical, allocative and economic efficiency of sugarcane production in three major sugarcane growing districts (Mardan, Charsadda and D.I. Khan) of Khyber Pakhtunkhwa. Determinants of technical, allocative and economic efficiency were also investigated. Primary data on sugarcane production, inputs and its cost was collected for crop year 2014-15 from 303 respondents. Out of total 303 respondents, 109 sugarcane growers were interviewed in district Mardan, 118 in Chardada and 76 in D.I. Khan through proportional allocation technique. Stochastic frontier function approach was used to estimate technical, allocative and economic efficiency scores of sugarcane farms in study area. In order to identify inefficiency sources the technical and allocative inefficiency score was regressed separately on socio-economic characteristics of sugarcane growers that were considered for this study. Cobb-Douglas type stochastic frontier production function was estimated by OLS and MLE techniques. District wise profile of study area shows that district Mardan occupy 1632 km2 area with population of 2.36 million. Its major crops are tobacco, sugarcane, wheat, rice and maize. It share 28.52 % area and 29.78% production in overall provincial area under sugarcane crop and its production. Main sources of irrigation are canals, tube wells and lift irrigation system. District Charsadda is the second district of study area that occupies 996 km2 areas with 1.02 million populations. 86% of its area is irrigated. Its major crops are sugarcane, sugar beet, maize, tobacco along with orchards and vegetables. It shares 32.41% and 31.49% in overall provincial sugarcane area and production respectively. Third district included in study area was D.I. Khan that is the southernmost district of Khyber Pakhtunkhwa. District D.I. Khan has the characteristics of extremely unequal land distribution, low land utilization and high tenancy ratio. Sugarcane is considered a major crop along with cotton, rice wheat and maize. In study area land use intensity was found high in district Mardan (95%) followed by Charsadda (92%) and D.I. Khan (42%). In study area respondent’s average age, education and farming experience was noted 38.38, 4.96 and 12.78 years with standard deviation of 11.87, 3.77 and 7.02 respectively. Majority (45.21%) of respondents were found in age group of 31 to 45 years. Average family size and extension contacts were arrived at 10.26 and 7.14 with standard deviation of 3.18 and 2.80 respectively. In study area average farm to home distance was 623.14 meters with standard deviation of 263.85 meters. 60.40 percent respondents lied within the distance of one kilometer (500-1000 meter). In study area 52 % respondents were found involved in off-farm activities with almost 50 % standard deviation. Out of total 303 respondents 36.63 % were found tenants while 42.24 % and 21.12 % were owners and owner-cum tenants respectively. Profitability of sugarcane growers was estimated district wise as well as combines. Per acre overall average yield was noted 24,420.73 kg with total cost of production Rs.83, 473/- . Results of the study also revealed that per acre average cost was high in district D.I. Khan (Rs. 80,397.43) followed by district Mardan (Rs.77, 771) and Charsadda (Rs.77, 083). Among inputs labour days, seed and tractor hours applied for various field operations were found major constituents of total cost of production, its share was more than 50% in overall per acre cost of production. Per acre average labor days employed were high in district Mardan (57.90) followed by D.I. Khan (46.98) and Charsadda (44.02). Results indicate that per acre average seed quantity used was high in D.I. Khan (3267.37 Kg) followed by district Charsada (2471.86 Kg) and Mardan (1739.06 Kg). Other results predict that per acre average tractor hours were 23.92, 20.74 and 18.36 respectively in district D.I. Khan, Charsadda and Mardan while overall it arrives at 20.68. According to summarized results overall per acre gross revenue arrived at Rs. 122103.65. Among districts per acre gross revenue was highest in district Mardan (Rs. 1, 26,950.45/-) followed by district Charsada (Rs. 1, 22, 833.90/-) and D.I. Khan (Rs. 1, 13, 571.05/-). Net revenue was calculated by subtracting per acre total cost from gross revenue. Per acre net revenue from sugarcane production was found high in district Mardan (Rs.49, 179.45/-) followed by district Charsadda (Rs. 45,750.90/-) and D.I. Khan (Rs. 33,173.65/-). Overall per acre net revenue was found Rs.38, 630/-. According to result of stochastic production function and technical inefficiency model in overall data the derived production elasticity for seed was higher (0.138) compare to other explanatory variables i.e. labor (0.092), irrigation (0.085), urea (0.079), land (0.022), FYM (0.025), DAP (0.02) and pesticides (0.005). All inputs had positive and significant elasticities except DAP, FYM and pesticides that are positive but statistically non-significant. Primary reason for non-significance was low than recommended level of application. Growers specific socio-economic characteristics were included in the model to ascertain factors affecting technical efficiency of sugarcane production. The effect of grower’s age, farming experience and off-farm income had found significant and negatively related to technical inefficiency. Years of schooling was found positive and significantly correlated with technical inefficiency. While, the coefficient of family size, farm to home distance, extension contacts were negative but non-significant. Similarly tenure status was positive but non-significant. The gamma parameter arrived at 0.61, which illustrate that 61 variations in overall production were due to technical inefficiency. Stochastic frontier production function and technical inefficiency model for district Mardan show that all inputs had positive and significant elasticities except land variable, FYM and pesticides, which are positive but non-significant. The estimated production elasticity for seed was found high (0.133) followed by labor days (0.121), irrigation (0.091), DAP (0.057), urea (0.022), tractor hours (0.011), FYM (0.047), pesticides (0.020) and land (0.012) respectively. In inefficiency model growers specific to sugarcane experience and contacts with extension workers were found significant and negatively correlated with technical inefficiency. The other socio-economic variables (age, education, family size, off-farm income, farm to home distance and tenure status) have mix signs but non-significant. In district Mardan gamma parameter arrived at 0.59 which mean that 59% variability in sugarcane production among grower is due to technical inefficiency. Estimated production elasticities for all variables in district Charsadda were found positive and significant except DAP and pesticides which were found non-significant. Primary reason for non-significance was lower quantity of application. Derived production elasticity for irrigation variable was high (0.156) followed by labour days (0.121) and tractor hours (0.111). These variables were found important in term of its contribution toward sugarcane yield. In technical inefficiency model the effect of experience, education and extension contacts was found negative and statistically significant at 5 % level. The signs of coefficients of age, off-farm income and farm to home distance were found positive and significant, which show that with increase in grower age, off-farm income and farm to home distance technical inefficiency increase. Coefficient for family size was positive but non-significant. Results predict that 69 percent variation in sugarcane yield among growers is due to technical inefficiency. Derived production elasticity for seed was higher (0.171) in district D.I. Khan followed by labour days (0.132), land (0.13), irrigation (0.088), urea (0.075), pesticides (0.073), FYM (0.028) and DAP (0.014). However, production elasticities for land, DAP and FYM were found statistically non-significant at 5% level. Findings of the study showed that the effect of age and off-farm income was significant and negatively correlated with inefficiency. Mix signs were noted for the remaining variables, but its effect was non-significant except education with positive signs. The estimated gamma parameter for district D.I. Khan was noted 0.82 which means that 82% variation in sugarcane production was due to technical inefficiency. Empirical findings of stochastic frontier cost functions for combine as well separate district wise data show that sugarcane yield, labor days, seed, tractor hours, irrigation and urea are important constituents in total cost of production. While DAP, FYM and pesticides are positive but non-significant. In inefficiency model education and extension contacts were found significant factors that effect allocative inefficiency negatively in combine data. In district Mardan growers experience and off farm income show negative and significant correlation with allocative inefficiency. Family size, extension contacts along with tenure status were noted significant contributors to allocative efficiency in districts Charsadda and D.I. Khan. Mix and non-significant results were noted for other socio-economic variables. The estimated gamma parameters arrived at 0.62, 0.72, 0.79 and 0.84 for combine data and district Mardan, Charsadda and D.I. Khan respectively. Gamma value points out allocative inefficiency problem in study area. In study area mean technical efficiency varied from 0.53 to 0.98 with mean value of 0.70, which means that per acre sugarcane yield, could be increased with existing inputs level. The estimated technical efficiency score was found high in district Mardan (0.83) followed by district Charsadda (0.78) and D.I. Khan (0.63). In district Mardan majority of the farmers were found in the range of 0.80 to 0.90, while in Charsada and D.I. Khan most of respondents lies between 0.50-0.70 technical efficiency level. According to empirical results average allocative efficiency of sample respondents was noted 0.57 with range from minimum 0.20 to maximum 0.91. Mean allocative efficiency level was found high for district Charsadda (0.63) followed by district Mardan (0.60) and D.I. Khan (0.44). Empirical results showed that there is scope to reduce cost and increase productivity. The study shows that technical and allocative efficiency scores and factors influencing the sugarcane productivity are different among the districts. Mean economic efficiency of 041, 0.49, 0.40 and 0.28 noted for overall combine data and for district Mardan, Charsadda and D.I. Khan respectively showed that sugarcane growers are not efficient in production as well as in resource allocation. In study area none of the respondents has achieved above 60 % of economic efficiency level, therefore, for enhancing economic efficiency level both technical as well as allocative efficiency appears to be consider important." xml:lang="en_US