透過您的圖書館登入
IP:3.145.151.141
  • 期刊

空間外溢:直接效果與間接效果-以住宅地價之空間外溢為例

Spatial Spillovers: Direct Effects and Indirect Effects-A Case Study on the Spatial Spillover Effects of Residential Land Prices

摘要


空間外溢效果一直是區域科學、都市經濟學及經濟地理學等的一個重要研究核心。在過往很多實證研究中,大多是以空間自我迴歸係數的顯著與否做為判定空間外溢效果是否存在的唯一指標。隨著空間計量經濟學近期新文獻的提出,我們發現空間計量經濟模型對於空間外溢效果的捕捉能以更深入的層面反應出來,主要分為直接效果與間接效果。直接效果能系統性的反應出:「當一單位解釋變數變動時,會有多少單位的相依變數因而發生改變。」,直接效果能考量到系統中的空間回饋效果,精確的反應彈性係數的概念。間接效果則反應出當一個地區之解釋變數變動時,其他地區之相依變數會應而改變的程度。本文有兩個重點,第一點是從計量經濟學理論的角度,參考國際近期文獻,以簡捷的方式重新闡述幾個新發展之空間計量經濟模型對於空間外溢效果之捕捉。期能補足國內在這方面文獻之有限性。第二點是運用上述觀點進行實證研究。探討的對象為住宅土地價格之空間外溢,在解釋變數的選擇上,我們依據Tiebout所提出之公共投資水準觀點,選用教育支出作為代表。相較於多數過往研究只以空間自我迴歸模型來反應住宅地價和其周邊區域地價之空間相依性,本研究更進一步的估計了直接效果和間接效果。實證結果顯示公共投資水準能顯著的解釋住宅土地價格之空間變動,如理論所預期。空間自我迴歸縱橫模型及空間杜賓縱橫模型都能顯著的反應直接效果與間接效果。直接效果中的20%~25%是由空間回饋效果所貢獻,這意味著如果以模型估計之係數為彈性係數,則低估了20%~25%的效果。

並列摘要


Spatial spillover effects are a research focus of fields such as Regional Science, Urban Economics, Economic Geography. Many previous empirical studies used the significant level of spatial autoregressive coefficient to determine the existence of spatial spillover effects. The recent development of spatial econometrics has led to models that can capture spatial spillover effects in a more detailed and advanced manner than previously. Spatial spillovers can be classified as direct effects and indirect effects. Direct effects can systematically reflect how many units of a dependent variable change when one unit of an explanatory variable changes. Direct effects consider the spatial feedback effects existing in the system, and thus are capable of reflecting the concept of elasticity precisely. Indirect effects can reflect the degree of change of a dependent variable in neighbouring regions when a one degree of an explanatory variable changes in its own region. This study has two focuses. The first focus is to re-expound spatial spillover effects captured by several newly developed spatial econometrics models, based on the theoretical perspectives of econometrics, referencing recently published international papers. We hope that this part of theoretical discussion can complement the shortage of domestic spatial econometrics literature. The second focus is to undertake an empirical study that applies these perspectives. The research subject is on spatial spillovers of residential land price. Regarding explanatory variables, education expenditure is adopted as a main variable, reflecting Tiebout's proposition of public investment level. This study builds upon previous studies, which modeled spatial dependence of residential land price across regions by only estimating the spatial autoregressive model, to estimate direct effects and indirect effects. Empirical results indicate that public investment level can significantly explain the spatial variation of residential land prices as the expectation of theory. The spatial autoregressive panel model and spatial Durbin panel model can significantly reflect direct effects and indirect effects. Spatial feedback effects contribute 20%~25% of direct effects. Thus, if the estimated coefficients of spatial panel models are regarded as coefficients of elasticity, then this empirical study underestimates 20%~25% of effects.

參考文獻


Anselin, L. 1988. Spatial econometrics: Methods and models. The Netherlands: Kluwer Academic Publishers.
Anselin, L. 2003. Spatial externalities, spatial multipliers, and spatial econometrics. International Regional Science Review 26 (2): 153-66.
Anselin, L., J. Le Gallo, and H. Jayet. 2008. Spatial panel econometrics. In: The econometrics of panel data: Fundamental and recent developments in theory and practice, eds. L. Matyas, and P. Sevestre, 625-57. Berlin Heidelberg: Springer-Verlag.
Arbia, G. 2014. A primer for spatial econometrics with applications in R. London: Palgrave Macmillan.
Baltagi, B. H. 2008. Econometric analysis of panel data (4th Edition). New Jersey : John Wiley & Sons.

延伸閱讀