標題: | 臺灣各縣市環保機關資源回收之績效評估 Performance Evaluation of Resource Recycling in Taiwan’s Local Government Services |
作者: | 張子漙 胡均立 經營管理研究所 |
關鍵字: | 資料包絡分析;隨機邊界法;異質性非單調隨機邊界法;資源回收;績效評估;data envelopment analysis;stochastic frontier;heteroscedasticity and non-monotonic stochastic frontier;recycling performance |
公開日期: | 2005 |
摘要: | 自民國八十六年推動資源回收四合一政策後,各縣市資源回收成效逐年遞增。然目前尚未有一客觀評估各縣市資源回收績效的標準,故本研究旨在評估1998年至2004年間各縣市環保機關之資源回收效率、各縣市環保機關效率變化情形、並瞭解各縣市環保機關發生無效率之原因及其改善目標。
本文採用了人均環保經費、單位面積清運人力、參與環保教育宣導活動率與人均垃圾處理車輛等四個投入項;而產出項則考慮人均回收量與回收率。以多產出多投入的評估模型,將較能達到公平客觀之原則。並以資料包絡分析、隨機邊界法之一階段聯立估計法和異質性非單調隨機邊界法三種模型評估。
本研究發現三種評估模式所得出的結果類似,僅排序上有些微差異。整體而言表現最好的縣市為台中市、基隆市,而以高雄市、台北縣表現最差。至於研究環境變數對無效率因子的影響也可由三種模型求得,研究發現人均可支配所得、高等教育人口比例、老年人口比例、都市人口比例越高的縣市,其整體資源回收效率也隨之提高;而人口密度越高的縣市其資源回收績效將越低。最後本研究以異質性非單調隨機邊界法求出環境變數之邊際效果,提供最適之環境變數水準。
資源回收乃一長期、持續關注的政策施行,本文僅就1998〜2004年為分析期間,後續應該繼續對此議題做深度與廣度的研究,以期能為政府部門建立長期可信的效率評比方式。 Taiwan started promoting resources recycling policies in 1997. The recycling amounts in all counties and cities increased progressively year by year. However, overall assessment methods for recycling performance still need to be developed. This research aims at assessing resources recycling performance in Taiwan’s local government services from 1998 to 2004. Both data envelopment analysis and stochastic frontier method are applied in this research for regional recycling efficiency measurement. Especially, the heteroscedasticity and non-monotonic stochastic frontier approach as a new stochastic frontier proposed by Wang (2003) is also adopted to find out the appreciate levels of environmental variables for improving recycling performance. There are two outputs (recycling volume per capita and recycling rate) and four inputs (budget of environmental protection per capita, waste clearance vehicles per capita, waste clearance persons per squared kilometer, and participators in the promoting activities per capita frequency) in each efficiency model. The results of three methods are consistent: (1) The best recycling performers are Taichung City and Keelung City. (2) Increases in real income per capita, proportions of population of high-education, old proportions of population, and urban population ratio will improve the regional recycling efficiency. (3) An increase in population density will decrease the regional recycling efficiency. (4) The appreciate levels for improving recycling performance are the per capita real income about 225,000 NT dollars, high-education population proportion up to 36.37%, old population proportion about 8.39%, and population density about 2,500 person/km2. |
URI: | http://140.113.39.130/cdrfb3/record/nctu/#GT009337517 http://hdl.handle.net/11536/79646 |
顯示於類別: | 畢業論文 |