經環境效果及統計噪音調整後的亞太經濟體總要素能源效率

Abstract

台灣參與的亞太經合組織（APEC）包含了許多成長快速的經濟體，並且擁有眾多的外資、先進的技術及管理知識。隨著迅速的經濟發展，同時也迅速地增加能源消耗。如何有效率地節約能源，而又不妨礙包含台灣在內的APEC會員體的經濟發展，是一個非常重要的學術與政策研究課題。 Hu and Kao (2007)、Hu (2006)、Hu and Wang (2006) 的一系列文獻建構出總要素能源效率架構，然而所求得的總要素效率值及減量比率目標值，會受到環境效果及統計噪音的影響。本研究將採用Fried et al. (2002) 的並用資料包絡分析法及隨機邊界法的三階段方法，來計算經環境效果及統計噪音調整後的各區域總要素能源效率。 第一年計畫將以包含台灣在內的APEC經濟體為主要研究對象，預計將至少包含澳洲、加拿大、智利、中國、香港、印尼、日本、韓國、馬來西亞、墨西哥、紐西蘭、秘魯、菲律賓、新加坡、台灣、泰國、美國等17個經濟體、1991-2004年之年度資料。將使用經過購買力平價條件轉換後、以1995年基期價格計算的實質變數資料。將採用單一產出-多投入模型，其中產出項為實質GDP，投入項則為實質資本存量、勞動僱用量、和能源消費量。此外，我們將正式在總要素能源效率衡量中納入環境變數，能源依存度、人均所得、貿易開放程度、政府貪腐程度、國家競爭力指標、是否為OECD會員國等。 第二年計畫研究對像為台灣本島及澎湖在內之23縣市，採取縣市別年度資料，研究期間為1999-2006年。產出變數有四項：各縣市年度所得、人口數、高中以上人口比例、垃圾處理量。投入變數有六項：各縣市政府歲出、就業人數、家庭用電量、其他用電量、汽油銷售量、柴油銷售量等。主要模型是在考慮能源投入下，來計算台灣各縣市各年度的效率。環境變數包括：人口結構（含就業結構與年齡結構）、區位、是否為都會區、縣市政府預算、人口密度、都市人口比例等。所有名目變數將平減為以1999為年基期之實質變數。

There are many fast-growing member economies in the Asia-Pacific Economic Cooperation in which Taiwan is a member economy, with abundant foreign capital, advanced technology, and managerial knowledge. However, fast economic development also ensues more energy consumption and exhaustion. It becomes a very important academic and policy issue to efficiently save energy without impeding the economic development of APEC』s member economies including Taiwan. The serial works by Hu and Kao (2007), Hu (2006), and Hu and Wang (2006) construct the total-factor energy efficiency framework. However, the total-factor energy efficiency scores and efficient savings ratios can be affected by environmental effects as well as statistical noise. This research will adopt the three-stage approach consisting of both data envelopment analysis and stochastic frontiers proposed by Fried et al. (2002), in order to calculate the regional total-factor energy efficiency after adjusting environmental effects and statistical noise. The first-year project will research seventeen APEC economies including Taiwan: Australia, Canada, Chile, China, Hong Kong, Indonesia, Japan, South Korea, Malaysia, Mexico, New Zealand, Peru, the Philippines, Singapore, Taiwan, Thailand, and the United States. We will construct am annual data set from 1991 to 2004. All variables will be real after the purchasing power parity adjustment to be in 1995 prices. This research will adopt model of a single output and multiple inputs. The single output is the real GDP and the inputs are real capital stock, labor employment, and energy consumption. Environmental variables will be formally incorporated into the total-factor framework, including energy dependence, per capita income, trade openness, governmental corruption index, national competitiveness index, OCED membership, etc. The second-year project will study twenty-three administrative regions in Taiwan, including those on the Taiwan Island and Penhu County. The research period will be from 1999 to 2006. There are four outputs: regional total income, population, population ratio of high graduation and more, processed trash. There are six inputs: local government expenditure, employment, processed trash, household electricity consumption, non-household electricity consumption, gasoline sales volume, and diesel sales volume. Environmental variables include population structure (containing both employment structure and age structure), location, municipality, local government expenditure, population density, urban population ratio, etc. All nominal variables will be transformed into real variables in the 1999 price level.