Efficiency of livestock-oil palm integration under small holder scheme in Johor, Malaysia

Malaysia has dominated global palm oil production for many decades. Even after Indonesia surpassed Malaysia in 2007 to become the current world leading producer, Malaysia’s role in the global scene is still substantial and will continue to be for a long time to come. However, the Malaysia’s oil palm policy of economy of scale has resulted to scarcity of land for agricultural use; upon which the poor performance of the ruminant animals is partly attributed to. This study analyzed production costs, returns and efficiency issues based on five estimators under goat-oil palm (GOI) and cattle-oil palm integrated (COI) systems. Data were collected from 255 livestock-oil palm smallholders across the 10 districts of Johor,Malaysia. Descriptive statistics, gross margin model and net income model for estimating production costs and returns. The Data Envelopment Analysis (DEA), DEAbootstrap,Free Disposal Hull (FDH), order-α and order-m for estimating efficiency of resource use. The box and whiskers plot was used to eliminate outliers in the data set in order to make the result more robust. The tobit and Ordinary Least Squares (OLS) regressions were used to analyze the determinants of Technical Efficiency (TE) under both goat-oil palm (GOI) and cattle-oil palm (COI). Depending on age of the palms, results on costs and returns show that the goat-oil palm (GOI) farms realized between 6.90mt/ha/yr and 22.84mt/ha/yr relative to the goat-oil palm (COI) between 13.17mt/ha/yr and 25.70mt/ha/yr of FFB. On the whole, while the goat-oil palm (GOI) farms realized between RM6562.94/ha/yr and RM17268.78/ha/yr, the cattle-oil palm (COI) farms obtained between RM11529.07/ha/yr and RM21034.17/ha/yr in net income. While the goat-oil palm (GOI) system predicted mean of 94% and 23% reduction in weeding cost and saved cost from total cost of production, the goat-oil palm (COI) system estimated 100% and 20% reduction in weeding cost and saved cost from total cost of production respectively, relative to none-integrated farms. The Data Envelopment Analysis (DEA) results found mean TE of 0.997, 1.000 and mean Scale Efficiency (SE) 0.802, 1.000 for goat-oil palm (GOI) and cattle-oil palm (COI) systems respectively. Similarly, 1.000, 0.998, 0.990 and 1.000, 0.998,0.972 were estimated as mean TE under FDH, order-α and order-m estimators under goat-oil palm (GOI) and cattle-oil palm (COI) systems respectively. Mean CE and AE were estimated 0.867, 0.869 and 0.865, 0.865 for goat-oil palm (GOI) and cattle-oil palm (COI) respectively. Analysis of input and output slacks detected higher slack in the cattle-oil palm (COI) relative to the goat-oil palm (GOI) system; just as the noise estimate for factors beyond farmers’ control found lower bias (mean=0.047) components in the cattle-oil palm (COI) relative to the goat-oil palm (GOI) (mean=0.065). The bootstrap estimator reports that the plantations operate at sub-optimal phase of production under increasing returns to scale and estimated 0.888, 0.764-0.950 and 0.891, 0.776-0.937 as bias-corrected TE and confidence interval for goat-oil palm (GOI) and cattle-oil palm (COI) systems respectively. The t-test result found significant statistical difference between Fresh Fruit Bunches (FFB) yield and stocking rate under the two systems,also between the DEA and DEA-bootstrap estimators and between Federal Land Development Authority (FELDA) and independent farms. Finally, the tobit and OLS results found farmers’ age, years of integration,farmers’ education, credit and farming association to have positive and significant impact consistently across both integration systems. Policy decision encouraging increased farm size and one that can mitigate the effect of detrimental exogenous factors (flood, diseases) among others will help increase their efficiency status