環境品質特徵需求函數及其內生性之研究

2015年的巴黎協議取代京都議定書，訂立更明確的氣候環境保護協議，顯現人們愈重視愈趨極端的氣侯與環境議題，本研究之目的為估計台灣地區氣候與空氣品質之需求函數，並推估其價值，供政府作為氣候變遷與環境保護政策之參考。本研究運用民國97年至100年之住宅需求動向調查資料，將房屋交易市場區隔為台北都會區、台中都會區、台南高雄都會區三大市場，應用特徵價格法理論，採用半對數模型進行最小平方法迴歸分析，計算各都會區環境品質之邊際隱含價格，並以兩階段最小平方法估計特徵需求函數及解決其內生性問題，進而推估台灣地區氣候與空氣品質之價值，並以三種特徵需求函數模型設定進行比較。　　透過本研究實證結果可知，未處理內生性問題之特徵需求函數估計結果，大部份高估環境品質價值，因此應以調整後之特徵需求函數估計價值較為適當。此外，本研究發現，在第一階段估計中較不顯著之特徵變數，無法進行第二階段之估計，故本研究最後並未估計空氣品質之需求函數及價值。而在調整內生性問題的二模型中，將購屋者個人特徵變數納入之模型（2SLS-2），其估計結果皆低於僅納入環境特徵變數之模型（2SLS-1）。估計結果顯示，台灣家戶喜居於冬暖夏涼且空氣清新之地區，邊際價值之評估，當一月均溫上升攝氏0.1度時，每家戶均增加新台幣9,070元之效益，當七月均溫下降攝氏0.1度時，每家戶均增加新台幣7,474元之效益。而非邊際價值之評估，本研究以2075至2099之預估氣溫，估計下一世紀之環境品質價值，可知一月月均溫之上升，帶給民眾新台幣58,626元之效益，而七月月均溫之上升，帶給民眾新台幣146,654元之損失。綜上可知全球氣候變化更為嚴峻，全球平均氣溫逐漸上升，夏季愈來愈炎熱，使得七月月均溫下降帶來之效益較大。關鍵字：特徵價格法、內生性、氣候與空氣品質之價值

關鍵字

特徵價格法；內生性；氣候與空氣品質之價值

並列摘要

The Paris agreement, an agreement within the United Nations Framework Convention on Climate Change in 2015, replaced the Kyoto Protocol in 1997. Because of that, more people concern about the extreme environmental issues. The purpose of this study is to estimate the demand for climate and air quality in Taiwan, evaluate their value, and provide the information for the government. We use the Taiwan Housing Demand Survey data from 2008 to 2011, and separate markets for the Taipei, Taichung, and Tainan-Kaohsiung. In this research, we use the semi-logarithmic model to estimate the hedonic price function, calculate the marginal implicit prices of each environment variable, and address the endogeneity problem via two stage least squares method and ordinary least squares method. In addition, we separately estimate the inverse demand functions of each environment variable by using their marginal implicit prices, and evaluate households’ benefits when the improvement of climate or air quality exists. Finally, we compared with the results, and find the better one. 　　From the empirical results, we find that using two stage least squares method is more appropriate than the OLS estimates which underestimates the households’ benefits. The empirical results suggest that Taiwan households seem to prefer warmer winter, cooler summer, and better air quality. Evaluating the marginal value, every household will get NT$ 9,070 benefit on average from a 0.1 degree Celsius higher January temperature, NT$ 7,474 benefit on average from a 0.1 degree Celsius lower July temperature. And evaluating the non-marginal value, when the temperature becomes the estimated temperature of 2075 to 2099, every household will get NT$ 58,626 benefit on average from the change of Jan. temperature, NT$ 146,654 benefit on average from the change of July temperature. We also find that the benefit from lower July temperature is more than others, because of the rising earth’s surface temperature that make people worry about the effects of global warming.