史瓦濟蘭酪農旱季技術效率之評估

Assessing the Technical Efficiency of Dairy Farms during Dry Season in Swaziland

涂莉蕾(Thulile Phasha Zwane)

指導教授 ： 張文宜

國立屏東科技大學/國際學院/熱帶農業暨國際合作系/碩士(2016年)

摘要

史瓦濟蘭的旱季特點為降雨量少，使得天然牧草的可及性受到限制。為了保持較高的牛奶產量，農民採用不同的飼料型態，如青貯飼料、乾草和牧草。本研究之目的即在瞭解旱季時牧場使用青貯飼料、乾草或牧草之中的哪一種最為有效，同時確認其飼料利用的效率性。本研究以史瓦濟蘭的兩個行政區域為範圍，從史瓦濟蘭乳品局登記的農民名單中進行立意隨機抽樣，透過問卷調查收集41家牧場的相關資料，訪談焦點以旱季營運為主。為了評估史瓦濟蘭酪農的技術效率(TE)值，本研究採用DEAFrontier的差額基礎模型，利用以投入要素為導向的方式在報酬不變的狀況進行衡量。被考量的投入要素組合包含畜群數量、勞動力、土地面積和飼料；產出則是牛奶的公升數。為了衡量這些農場餵飼型態的子向量效率，飼料被作為唯一的投入要素，並保持所有的變項不變。所有樣本農場的技術效率TE平均值為78.9%，顯示所有投入要素可能降低21.1%且仍維持相同的產出水平。飼料的子向量效率平均值為65.2%，低於技術效率TE平均估計值，代表相對高的TE值某種程度掩蔽了飼料使用的低效率，而樣本中的低效率農民似乎能夠善加應用其它投入要素來彌補飼料的較低使用效率。測定結果指出，在旱季期間，相
較於使用任何乾草或牧草的牧場，使用青貯飼料當原料的牧場有較高的平均技術效率值和較高的子向量值。飼料的使用量也與牧場的效率呈負相關。此外，飼料總成本的係數最高(0.501)，即每增加飼料費用就會提高效率0.501，統計證實可顯著促進史瓦濟蘭酪農的效率。這也意味著如果農民購買優質飼料，將可改善牛奶產量、提高報酬，因此對效率具有正向作用。然而，全職勞工數與牧場技術效率呈負相關，顯示部份牧場的勞動力超過所需，每增加勞動量反而會降低牧場效率0.07。旱季低效牧場的平均投入要素差額分別為：畜群數量3.68、全職勞動力0.38、土地0.36公頃和飼料5,000公斤。和低效牧場相比，高效牧場顯著具有較高(p <0.05)的經驗、牛奶產量、飼養規模、飼料用量和飼料花費。改善當地酪農技術效率預期將可降低所進口的牛奶及奶製品量，國內牛奶生產量的提高也將有益於史瓦濟蘭當地的糧食安全。

關鍵字

DEAFrontier ； 差額基礎模型 ； 子向量效率 ； 史瓦濟蘭酪農 ； 技術效率

並列摘要

The dry season of Swaziland is characterized by little rainfall, which limits the availability of natural forages. To maintain high milk yield production, farmers utilize different types of feeding systems such as silage, hay and managed pasture. The goals of this study were to determine which farms among those using silage, hay or pasture are more efficient in the dry season and also determine their feed use efficiencies. The study was conducted in two administrative regions of Swaziland, and purposive random sampling was performed from the Swaziland Dairy Board registered farmers list. Data were collected through questionnaire interviews from 41 farms, with questions focused on dry season operations. To evaluate technical efficiency (TE) scores for dairy farms in Swaziland, a slacks-based model of DEAFrontier was applied, using input-oriented method under constant returns to scale. The set of inputs considered were herd size, labour, land area and feed; and the output was milk litres. To measure the subvector efficiency for feed in these farms, we used feed as the only input and kept all variables
constant. The mean TE score for all the farms in the sample was 78.9%. This result means that a 21.1% decrease in all inputs is possible and still maintain the same level of outputs. The average score for subvector efficiencies for feed (65.2%) was lower than the average TE estimated. This reveals that the relative high TE scores masked a degree of inefficiency in feed usage and the inefficient farmers in the sample seem to be able to compensate for the less efficient use of feed by a better use of other inputs. It was determined that farms that use silage as a source of feed during the dry season have a higher mean TE score and a higher a subvector score for feed use than the farms that use either hay or pasture. The quantity of feed used is negatively associated with the efficiency of the farms. On the other hand, the total cost of feed has the highest coefficient (0.501) and this indicates that an increase of money spent on feed will increase the efficiency by 0.501, and it is statistically significant in contributing to the measurement of efficiency of the dairy farms in Swaziland. This emphasizes that if farmers spend money on purchasing quality feed; there will be better returns from improved milk production, hence a positive effect on efficiency. The number of full time labour is negatively associated with the TE of farms and this suggests that some farms have more labour than necessary and increasing the amount of labour reduces the efficiency of the farms by 0.07. Average input slacks for the dry season in inefficient farms were: 3.68 for herd size; 0.38 full-time labourers; 0.36 hectares of land and 5000kgs of feed. The efficient farms also had significantly higher (p < 0.05) experience, milk volumes, herd sizes, use a larger quantity of feed and spend more money on feed than inefficient farms. It is anticipated that improved TE in the local dairy farms can lower the amount of milk and milk products that are imported, and the higher domestic production of milk may also result in lowered milk prices in the local (Swazi) markets.

並列關鍵字

DEAFrontier ； slacks-based model ； subvector efficiency ； Swazi dairy farms ； technical efficiency

參考文獻

Ahmad, M. and B. E. Bravo-Ureta. 1996. "Technical Efficiency Measures for Dairy Farms Using Panel Data: A Comparison of Alternative Model Specifications," The Journal of Productivity Analysis, 7: 399-415.

Amanloo, A. and G. H. Mobtaker. 2012. "Energy Efficiency Improvement in Forage Maize Production Using Data Envelopment Analysis Approach," African Journal of Agricultural Research, 7(41): 5571-5577. doi: 10.5897/AJAR12.317.

Banker, R., A. Charnes, and W. W. Cooper. 1984. "Some Models for Estimating Technical and Scale Inefficiences in Data Envelopment Analysis," Management Science, 30: 1078-1092.

Bardhan, D. and M. L. Sharma. 2013. "Technical Efficiency in Milk Production in Underdeveloped Production Environment of India," Springerplus, 2. doi: Artn 65.10.1186/2193-1801-2-65.

Byma, J. P. 2006. Efficiency of New York Dairy Farms: Exploring the Role of Managerial Ability. Master Faculty of the Graduate Cornell University.