馬子量 郭志儀 馬丁丑
摘要 基于西部地區(qū)12個省域2000-2011年城市化相關數(shù)據(jù),通過空間計量經(jīng)濟學方法對西部地區(qū)省域城市化動力機制進行了分析??臻g相關性檢驗結果顯示,西部地區(qū)省域間城市化存在顯著的空間相關性,有著空間集聚的趨勢,利用傳統(tǒng)OLS模型進行解釋會造成空間因素的遺漏,空間計量模型更為適用,進一步利用空間杜賓模型對西部地區(qū)省域城市化進程中動力因素的空間交互效應進行驗證,并在此基礎上對其空間效應進行分解,結果顯示:西部地區(qū)省域產(chǎn)業(yè)結構轉(zhuǎn)變對其城市化的正向直接效應最強,但是省域間存在較強的產(chǎn)業(yè)競爭,周邊省域產(chǎn)業(yè)結構轉(zhuǎn)變會對省域城市化產(chǎn)生明顯的負向間接效應;經(jīng)濟發(fā)展水平的提高對省域城市化的直接效應顯著,同時由于西部地區(qū)省域間經(jīng)濟發(fā)展水平差異不明顯,周邊省域經(jīng)濟發(fā)展水平對省域城市化的負向間接效應并不顯著;城市用地規(guī)模擴張對省域城市化的正向直接效應僅次于產(chǎn)業(yè)結構轉(zhuǎn)變,而且這種城市化推動行為在省域間效仿度較高,導致其對省域城市化的正向間接效應最強;基礎設施投入對省域城市化也具有顯著的正向直接效應,省域間基礎設施投入雖存在效仿行為,但是財政水平較低導致其對省域城市化的正向間接效應并不顯著;由于區(qū)位劣勢,對外開放程度對西部地區(qū)省域城市化的正向直接效應不顯著,但是周邊省域?qū)ν忾_放程度的增強卻一定程度上導致省域自身經(jīng)濟要素外溢,對省域城市化產(chǎn)生了顯著的負向間接效應。因此,在未來西部地區(qū)省域城市化進程中不僅要關注省域自身城市化動力因素,還應注意省域間的城市化動力因素的協(xié)調(diào)。
關鍵詞 空間效應;西部地區(qū);城市化;空間杜賓模型
中圖分類號 F061.5 文獻標識碼 A 文章編號 1002-2104(2014)06-0009-07 doi:103969/jissn1002-2104201406002
城市化進程是經(jīng)濟發(fā)展的必然路徑,伴隨著我國經(jīng)濟穩(wěn)步增長,城市化進程的重要性不斷凸顯,城市化在提高資源效率、提升產(chǎn)業(yè)層次、增加就業(yè)容量等方面起到了至關重要的作用。在既往的研究中,學者們從不同角度對于城市化動力進行了考察[1-5],但是這些研究往往孤立地去分析樣本區(qū)域的城市化動力因素,忽視了區(qū)域間的相互影響。城市化實質(zhì)是包括人口在內(nèi)的經(jīng)濟要素從鄉(xiāng)村向城市的空間流動過程,而這種流動不僅僅限于區(qū)域內(nèi),也存在著區(qū)域間的流動,周邊區(qū)域的城市化進程和其他影響因素對于區(qū)域城市化有著重要的影響,所以僅從區(qū)域內(nèi)孤立地去研究其城市化是不合理的,還應結合區(qū)域間的空間效應對區(qū)域城市化動力機制進行考察分析。
1 文獻綜述
一個區(qū)域的某種經(jīng)濟現(xiàn)象與周邊區(qū)域是相關的,忽略空間效應對經(jīng)濟問題進行分析是存在遺漏的,而現(xiàn)代空間計量經(jīng)濟學為捕捉空間效應提供了相應的理論基礎和技術方法,從Fisher [6]率先提出空間計量經(jīng)濟學之后,大量的學者不斷地進行深入研究,代表性的有Anselin [7]、Heikkila [8]、Lee [9]、 LeSage [10]和Elhorst [11]等,空間計量分析方法的不斷進步使現(xiàn)代經(jīng)濟學對于空間因素影響的把握越來越全面。隨著空間計量經(jīng)濟學引入國內(nèi),部分學者嘗試著將這種方法應用到城市化研究方面。蔣偉利用2005年數(shù)據(jù)對我國31個省域的城市化影響因素進行了空間計量分析,分析結果顯示省域間城市化存在空間依賴性,同時發(fā)現(xiàn)空間計量方法的解釋力強于一般計量方法[12];姜磊和季民河利用截面數(shù)據(jù)建立空間誤差模型對全國的城市化與創(chuàng)新活動進行了分析,發(fā)現(xiàn)在城市化過程中存在明顯的知識外溢現(xiàn)象[13];連飛以2008年東北地區(qū)34個城市為研究對象,通過建立全局和局部空間計量模型,發(fā)現(xiàn)城市間工業(yè)生產(chǎn)效率存在明顯的空間溢出[14];胡玉敏和杜綱通過近二十年的數(shù)據(jù)對我國29個省區(qū)的城市增長進行了分析,通過建立空間面板誤差模型發(fā)現(xiàn)城市增長具有空間外溢性,城市間增長具有空間示范效應[15]??v觀這些研究,大部分是通過截面數(shù)據(jù)去分析區(qū)域城市化,截面數(shù)據(jù)由于選取的時間點固定,所以無法反映城市化的動態(tài)變化,同時采取的方法為空間滯后模型或空間誤差模型,前者只能反映城市化進程的空間交互效應,忽略了城市化動力因素的空間交互效應,而后者把空間交互效應“黑箱化”處理,只能反映區(qū)域間存在空間效應,但是對空間效應發(fā)生的途徑卻無法解釋,利用二者無法對區(qū)域間城市化空間效應進行全面捕捉。
全國城市化水平不斷提高的同時,城市化進程的區(qū)域差異也越發(fā)明顯,西部地區(qū)作為經(jīng)濟最為落后的區(qū)域,其城市化水平也低于全國水平,對于西部地區(qū)各省區(qū)來說,提高自身城市化水平對于提高經(jīng)濟效率和促進其經(jīng)濟增長有著重要的作用,所以對西部地區(qū)省域城市化進程中的動力機制進行研究有著重要的現(xiàn)實意義。本文以西部12個省區(qū)為研究區(qū)域,利用其2000-2011年城市化相關數(shù)據(jù),在空間相關性檢驗的基礎上,利用空間杜賓模型對西部地區(qū)省域城市化水平及其動力因素進行模型估計,并對省域間城市化進程及其動力因素的空間效應交互路徑進行分析和整理,以期為推動西部地區(qū)省域城市化進程提供相應的理論基礎。
2011年西部地區(qū)只有4個省區(qū)位于一、三象限,說明空間正相關性在逐步下降,同時8個省區(qū)存在空間負相關,而在造成這一狀況的直接原因就是城市水平的集聚性不斷增強,造成省域間的城市化水平差異不斷拉大。
3.2 非空間面板模型估計
首先采用一般面板模型對西部地區(qū)城市化動力機制模型進行估計,其目的是為了對模型估計殘差序列進行空間相關性檢驗,以檢驗是否有必要建立空間計量模型,同時可以通過檢驗判斷哪種模型更為適用。對西部地區(qū)城市化動力機制模型進行面板OLS模型估計,結果如表2所示,總體上模型擬合良好,且通過顯著性檢驗,但是DW值為2.46,說明存在序列相關性,進一步檢驗后發(fā)現(xiàn),殘差序列的Morans I在1%的水平下顯著,說明殘差序列存在空間自相關性,說明利用面板OLS模型進行估計是存在不足的,應當建立空間計量模型。兩種拉格朗日乘子檢驗LM test no spatial lag和LM test no spatial error均在1%的水平下顯著,說明空間滯后和空間誤差均存在,所以應通過空間杜賓模型來進行分析。
3.3 空間面板數(shù)據(jù)計量分析
由于模型存在空間滯后項,所以OLS估計是有偏的,通過Matlab R2012b軟件和空間計量軟件包將式(8)的SDM模型進行了ML估計,其中依據(jù)面板模型固定效應分為三種形式:空間固定效應、時期固定效應和空間時期雙向固定效應。估計結果如表3所示,三種效應下SDM模型的擬合度均比非空間面板模型高,極大似然值也高于非空間面板模型,同時三種效應下的模型在瓦爾德檢驗和似然比率檢驗時均拒絕模型簡化為SLM或SEM,說明SDM模型是適用的。從三種效應的模型估計來看,雙向固定模型的擬合度更優(yōu),極大似然值最高,各變量的顯著性也得到了明顯的增強,同時利用似然比率檢驗對固定效應進行檢驗時,時期固定效應和空間固定效應均在1%的水平下顯著,所以依據(jù)雙向固定效應的SDM模型進行相應分析。在雙向固定效應的SDM模型下,ρ的系數(shù)為-0.13,且在1%的水平下顯著,說明西部地區(qū)省域城市化進程中存在顯著的集聚效應,城市化水平的提高一定程度上依賴于周邊省區(qū)人口的流出,城市化進程存在空間非均衡性,這與之前的空間相關性檢驗結果是吻合的。
3.4 空間效應分解
在雙向固定效應SDM模型估計基礎之上,通過偏微分的方式對各動力因素對于城市化水平的影響進行了空間分解。如表4所示,各動力因素對城市化水平的整體影響用總體效應表示,總體效應包括直接效應和間接效應,其中直接效應表示省區(qū)自身動力因素對其城市化水平的影響,間接效應表示鄰接省區(qū)動力因素的對本省區(qū)城市化水平的影響。從直接效應來看,又可分為兩種影響路徑,一種是動力因素對本省區(qū)自身城市化水平直接的影響,可以SDM模型中各動力因素的系數(shù)進行估計,另一種是動力因素對鄰接省區(qū)城市化水平造成影響,進而又對本省區(qū)自身城市化水平產(chǎn)生的空間回饋效應。間接效應也可以分為兩種影響路徑,其一是鄰接省區(qū)動力因素發(fā)生變化直接對樣本省區(qū)的城市化水平產(chǎn)生影響,另一種是鄰接省區(qū)動力因素發(fā)生變化首先使鄰接省區(qū)自身的城市化水平產(chǎn)生變動,進而對本省區(qū)的城市化水平產(chǎn)生影響。
產(chǎn)業(yè)結構轉(zhuǎn)變對西部地區(qū)省域城市化水平的直接效應和間接效應均顯著,且為最強,但其直接效應為正,間接效應為負。直接效應為正且較顯著,說明省域自身的產(chǎn)業(yè)結構演變是推動其城市化水平的有效手段,伴隨著產(chǎn)業(yè)結構的升級,二、三產(chǎn)業(yè)吸納的就業(yè)日益增多,會導致人口向城市集聚,帶來城市化水平的提高。間接效應為負,說明省域間產(chǎn)業(yè)類型趨同,省域間產(chǎn)業(yè)發(fā)展存在競爭性,鄰接省區(qū)的產(chǎn)業(yè)結構的升級會導致本省區(qū)包括人口在內(nèi)的要素外流,導致省域城市化進程更加遲滯。產(chǎn)業(yè)結構發(fā)展的非均衡性,會導致城市化水平發(fā)展速度存在差異的同時產(chǎn)生空間集聚,這是導致西部地區(qū)省域間城市化水平存在空間差異的主要原因。
經(jīng)濟發(fā)展水平對西部地區(qū)省域城市化水平的直接效應為正,間接效應為負但不顯著。直接效應為正,說明經(jīng)濟發(fā)展水平對城市化水平有正向推動作用,二者呈現(xiàn)正向變動關系,伴隨著經(jīng)濟發(fā)展水平的提高,生產(chǎn)能力不斷增強,城市集聚能力不斷提升,將推動其城市化進程,這與一般經(jīng)濟規(guī)律是相符的。間接效應不顯著,說明西部地區(qū)省域間經(jīng)濟發(fā)展水平差異相對較小,省域間經(jīng)濟發(fā)展差距并不能導致經(jīng)濟要素流動進而使城市化水平產(chǎn)生空間差異。
城市用地規(guī)模擴張對西部地區(qū)省域城市化水平的直接效應和間接效應均顯著,且都為正。直接效應為正且其影響程度僅次于產(chǎn)業(yè)結構轉(zhuǎn)變,說明著城市用地規(guī)模的擴張對于西部地區(qū)城市化水平的提高起到了很大的推動作用,同時也從另外一個角度說明目前西部地區(qū)城市水平的提高很大程度上依賴于城市外延式的擴張。其間接效應為正且為最大值,說明城市用地規(guī)模擴張這種主要依賴于政府的城市化行為在省域間具有較強示范效應,當鄰接省區(qū)采用外延式的城市規(guī)模擴張路徑時,由于其速度較快,且效果明顯,樣本省區(qū)會效仿這種路徑。
基礎設施投入對西部地區(qū)省域城市化水平的直接效應和間接效應均為正,但間接效應不顯著。直接效應為正,說明基礎設施投入可以提高西部地區(qū)城市的承載力,帶來城市化水平的提高,但其效應較弱反映出目前西部地區(qū)城市的承載力還未制約其城市化進程,所以基礎設施投入對城市化的推動力較弱。其間接效應為正但不顯著,說明基礎設施投入在西部地區(qū)省域間有示范效應,但政府財政能力的相對薄弱限制了這種效應對城市化進程的影響力度。
對外開放程度對西部地區(qū)省域城市化水平的直接效應為正,間接效應為負,但直接效應不顯著。直接效應為正但不顯著,說明對外開放程度的加強有利于西部地區(qū)改善其資本和技術狀況,對其城市化有正面影響,但是西部地區(qū)開放程度較低,外來資本和技術投入未能對城市化產(chǎn)生有效的促進。間接效應為負,說明西部地區(qū)省域間在引入外來資本時存在競爭效應,外來資本和技術注入鄰接省區(qū)導致其產(chǎn)業(yè)升級和經(jīng)濟增長,引致本省區(qū)經(jīng)濟要素向周邊省區(qū)外溢,使其城市化進程受阻。
4 結論及建議
西部地區(qū)省域城市化水平存在空間相關性,有著空間集聚趨勢,利用空間杜賓模型可以有效地對省域間城市化和城市化進程的動力因素進行空間效應分析,實證結果表明:①西部地區(qū)省域自身產(chǎn)業(yè)結構轉(zhuǎn)變、經(jīng)濟發(fā)展水平、城市用地擴張和基礎設施投入對其城市化水平具有顯著的直接推動,其中產(chǎn)業(yè)結構轉(zhuǎn)變對城市化水平的貢獻度最高,而對外開放程度的增強對城市化水平的促進作用并不顯著。②通過擴張城市用地規(guī)模提高城市化水平在省域間具有示范效應,某省區(qū)通過擴張城市用地規(guī)模加速其城市化進程時,會使鄰接省區(qū)進行學習和效仿,基礎設施投入雖也在省域間具有示范效應,但受制于西部地區(qū)省區(qū)薄弱的財政水平其效應并不顯著。③通過產(chǎn)業(yè)結構轉(zhuǎn)變和增強對外開放程度提高城市化水平的行為在省域間競爭效應明顯,某省區(qū)通過這兩種方式提高其城市水平,會使鄰接省區(qū)的經(jīng)濟要素外溢,使其城市化進程受阻,經(jīng)濟發(fā)展水平雖也在省域間城市化進程中具有競爭效應,但是由于西部地區(qū)省域間經(jīng)濟發(fā)展差異不大導致其效應不顯著。
西部地區(qū)省域間城市化進程中的空間交互效應對各省區(qū)的城市化進程產(chǎn)生了重要的影響,未來西部地區(qū)省域在加速城市化進程中不僅要關注省域內(nèi)的動力因素,同時也要注意省域間動力因素間的交互效應。在通過加快產(chǎn)業(yè)結構轉(zhuǎn)變、提高經(jīng)濟發(fā)展水平、合理擴張城市用地規(guī)模和加大基礎設施投入增強的途徑促進各省區(qū)城市化水平的基礎上,應對省域間產(chǎn)業(yè)類型和外資投入進行合理布局,規(guī)避產(chǎn)業(yè)趨同和外資投入引發(fā)的省域間城市化空間競爭效應,同時各省區(qū)應在合理范圍內(nèi)加大基礎設施財政投入力度和擴張城市用地規(guī)模,增強城市化進程中省域間的示范效應。
(編輯:李 琪)
參考文獻(References)
[1]陳明星,陸大道,張華. 中國城市化水平的綜合測度及其動力因子分析[J].地理學報,2009,64(4):387-398.
[2]房國坤,王詠,姚士謀. 快速城市化時期城市形態(tài)及其動力機制研究[J].人文地理,2009,160(2):40-43.
[3]汪段泳,朱農(nóng). 中國城市化發(fā)展決定因素的地區(qū)差異[J].中國人口·資源與環(huán)境,2007,17(1):66-71.
[4]王新娜. FDI在發(fā)展中國家城市化中的動力作用:基于國外研究的綜述[J].云南財經(jīng)大學學報,2010,(6):59-65.
[5]吳建峰,周偉林. 新時期我國城市化動力機制及政策選擇[J].城市發(fā)展研究,2011,18(5):21-26.
[6]Fisher W D. Econometric Estimation with Spatial Dependence[J]. Regional and Urban Economics,1971,(1): 19-40.
[7]Anselin L. Spatial Econometrics: Methods and Models[M]. Dordrecht, Netherlands: Kluwer, 1988.
[8]Heikkila E J, Kantiotou C. Calculating Fiscal Impacts Where Spatial Effects Are Present[J]. Regional Science and Urban Economics,1992,(22): 475-490.
[9]Lee L F. Asymptotic Distributions of Quasi-maximum Likelihood Estimators for Spatial Autoregressive Models[J]. Econometrica,2004, 72 (6):1899-1925.
[10]LeSage J P. Spatial Econometrics [J]. Encyclopedia of Social Measurement, 2005,(3): 613-619.
[11]Elhorst J P. Applied Spatial Econometrics: Raising the Bar [J].Spatial Economic Analysis,2010, 5 (1), 9-28.
[12]蔣偉. 中國省域城市化水平影響因素的空間計量分析[J].經(jīng)濟地理,2009,29(4):613-616.
[13]姜磊,季民河. 城市化、區(qū)域創(chuàng)新集群與空間知識溢出:基于空間計量經(jīng)濟學模型的實證[J].軟科學,2011,25(12):86-90.
[14]連飛. 工業(yè)集聚與勞動生產(chǎn)率的空間計量經(jīng)濟分析:來自我國東北34個城市的經(jīng)驗證據(jù)[J].中南財經(jīng)政法大學學報, 2011,(1):108-114.
[15]胡玉敏,杜綱. 中國城市增長的空間計量經(jīng)濟學研究[J].科學·經(jīng)濟·社會,2012,30(1):50-56.
[16]Elhorst J P. Dynamic Spatial Panels: Models, Methods and Inferences [J]. Journal of Geographical Systems,2012,(1):5-28.
[17]LeSage J P, Pace R K. Introduction to Spatial Econometrics [M]. Boca Raton,US:CRC Press/Taylor & Francis, 2009: 20-44.
Abstract Based on the urbanization data of 12 provinces in the western region from 2000 to 2011, provincial urbanization dynamic factors in the western region are analyzed through spatial econometrics methods. According to the results of the spatial correlation test, there is a spatial correlation in the western region provincial urbanization,and there is a spatial agglomeration trend. Therefore, using the traditional OLS model will miss the space factors, and spatial econometric model is more suitable. Spatial interaction effect of provincial urbanization in the western region is verified by spatial durbin model; on the basis,the spatial effects are decomposed.The results show that the transformation of industrial structure in western provinces has the most positive direct effects to provincial urbanization,and due to the industrial competition between provinces,transformation of industrial structure in neighboring provinces has negative indirect effects to provincial urbanization; level of economic development in western provinces has significant positive direct effects to provincial urbanization,but due to the gap of economic development between western provinces is small, and the indirect effects of economic development in neighboring provinces to provincial urbanization is not significant;the positive direct effects of urban land scale expansion to provincial urbanization are the second only to transformation of industrial structure,and due to the imitation between western provinces,it is has the most positive indirect effects to provincial urbanization;infrastructure investment of western provinces has significant positive direct effects to provincial urbanization, but subject to the low level of provincial fiscal revenue,the indirect effects of infrastructure investment between western provinces are not significant;because of the disadvantage of location, the direct effects of openingup degree to provincial urbanization is not significant,but the enhancement of openingup degree in neighboring provinces will cause the provincial economic element spillover,it is has significant negative direct effects.Therefore, in the process of provincial urbanization in the western region, decision makers shouldnt only enhance provincial own motivation factors, should also coordinate the urbanization motivation factors between provinces.
Key words spatial effects; westerns region; urbanization; spatial Durbin model
(編輯:李 琪)
參考文獻(References)
[1]陳明星,陸大道,張華. 中國城市化水平的綜合測度及其動力因子分析[J].地理學報,2009,64(4):387-398.
[2]房國坤,王詠,姚士謀. 快速城市化時期城市形態(tài)及其動力機制研究[J].人文地理,2009,160(2):40-43.
[3]汪段泳,朱農(nóng). 中國城市化發(fā)展決定因素的地區(qū)差異[J].中國人口·資源與環(huán)境,2007,17(1):66-71.
[4]王新娜. FDI在發(fā)展中國家城市化中的動力作用:基于國外研究的綜述[J].云南財經(jīng)大學學報,2010,(6):59-65.
[5]吳建峰,周偉林. 新時期我國城市化動力機制及政策選擇[J].城市發(fā)展研究,2011,18(5):21-26.
[6]Fisher W D. Econometric Estimation with Spatial Dependence[J]. Regional and Urban Economics,1971,(1): 19-40.
[7]Anselin L. Spatial Econometrics: Methods and Models[M]. Dordrecht, Netherlands: Kluwer, 1988.
[8]Heikkila E J, Kantiotou C. Calculating Fiscal Impacts Where Spatial Effects Are Present[J]. Regional Science and Urban Economics,1992,(22): 475-490.
[9]Lee L F. Asymptotic Distributions of Quasi-maximum Likelihood Estimators for Spatial Autoregressive Models[J]. Econometrica,2004, 72 (6):1899-1925.
[10]LeSage J P. Spatial Econometrics [J]. Encyclopedia of Social Measurement, 2005,(3): 613-619.
[11]Elhorst J P. Applied Spatial Econometrics: Raising the Bar [J].Spatial Economic Analysis,2010, 5 (1), 9-28.
[12]蔣偉. 中國省域城市化水平影響因素的空間計量分析[J].經(jīng)濟地理,2009,29(4):613-616.
[13]姜磊,季民河. 城市化、區(qū)域創(chuàng)新集群與空間知識溢出:基于空間計量經(jīng)濟學模型的實證[J].軟科學,2011,25(12):86-90.
[14]連飛. 工業(yè)集聚與勞動生產(chǎn)率的空間計量經(jīng)濟分析:來自我國東北34個城市的經(jīng)驗證據(jù)[J].中南財經(jīng)政法大學學報, 2011,(1):108-114.
[15]胡玉敏,杜綱. 中國城市增長的空間計量經(jīng)濟學研究[J].科學·經(jīng)濟·社會,2012,30(1):50-56.
[16]Elhorst J P. Dynamic Spatial Panels: Models, Methods and Inferences [J]. Journal of Geographical Systems,2012,(1):5-28.
[17]LeSage J P, Pace R K. Introduction to Spatial Econometrics [M]. Boca Raton,US:CRC Press/Taylor & Francis, 2009: 20-44.
Abstract Based on the urbanization data of 12 provinces in the western region from 2000 to 2011, provincial urbanization dynamic factors in the western region are analyzed through spatial econometrics methods. According to the results of the spatial correlation test, there is a spatial correlation in the western region provincial urbanization,and there is a spatial agglomeration trend. Therefore, using the traditional OLS model will miss the space factors, and spatial econometric model is more suitable. Spatial interaction effect of provincial urbanization in the western region is verified by spatial durbin model; on the basis,the spatial effects are decomposed.The results show that the transformation of industrial structure in western provinces has the most positive direct effects to provincial urbanization,and due to the industrial competition between provinces,transformation of industrial structure in neighboring provinces has negative indirect effects to provincial urbanization; level of economic development in western provinces has significant positive direct effects to provincial urbanization,but due to the gap of economic development between western provinces is small, and the indirect effects of economic development in neighboring provinces to provincial urbanization is not significant;the positive direct effects of urban land scale expansion to provincial urbanization are the second only to transformation of industrial structure,and due to the imitation between western provinces,it is has the most positive indirect effects to provincial urbanization;infrastructure investment of western provinces has significant positive direct effects to provincial urbanization, but subject to the low level of provincial fiscal revenue,the indirect effects of infrastructure investment between western provinces are not significant;because of the disadvantage of location, the direct effects of openingup degree to provincial urbanization is not significant,but the enhancement of openingup degree in neighboring provinces will cause the provincial economic element spillover,it is has significant negative direct effects.Therefore, in the process of provincial urbanization in the western region, decision makers shouldnt only enhance provincial own motivation factors, should also coordinate the urbanization motivation factors between provinces.
Key words spatial effects; westerns region; urbanization; spatial Durbin model
(編輯:李 琪)
參考文獻(References)
[1]陳明星,陸大道,張華. 中國城市化水平的綜合測度及其動力因子分析[J].地理學報,2009,64(4):387-398.
[2]房國坤,王詠,姚士謀. 快速城市化時期城市形態(tài)及其動力機制研究[J].人文地理,2009,160(2):40-43.
[3]汪段泳,朱農(nóng). 中國城市化發(fā)展決定因素的地區(qū)差異[J].中國人口·資源與環(huán)境,2007,17(1):66-71.
[4]王新娜. FDI在發(fā)展中國家城市化中的動力作用:基于國外研究的綜述[J].云南財經(jīng)大學學報,2010,(6):59-65.
[5]吳建峰,周偉林. 新時期我國城市化動力機制及政策選擇[J].城市發(fā)展研究,2011,18(5):21-26.
[6]Fisher W D. Econometric Estimation with Spatial Dependence[J]. Regional and Urban Economics,1971,(1): 19-40.
[7]Anselin L. Spatial Econometrics: Methods and Models[M]. Dordrecht, Netherlands: Kluwer, 1988.
[8]Heikkila E J, Kantiotou C. Calculating Fiscal Impacts Where Spatial Effects Are Present[J]. Regional Science and Urban Economics,1992,(22): 475-490.
[9]Lee L F. Asymptotic Distributions of Quasi-maximum Likelihood Estimators for Spatial Autoregressive Models[J]. Econometrica,2004, 72 (6):1899-1925.
[10]LeSage J P. Spatial Econometrics [J]. Encyclopedia of Social Measurement, 2005,(3): 613-619.
[11]Elhorst J P. Applied Spatial Econometrics: Raising the Bar [J].Spatial Economic Analysis,2010, 5 (1), 9-28.
[12]蔣偉. 中國省域城市化水平影響因素的空間計量分析[J].經(jīng)濟地理,2009,29(4):613-616.
[13]姜磊,季民河. 城市化、區(qū)域創(chuàng)新集群與空間知識溢出:基于空間計量經(jīng)濟學模型的實證[J].軟科學,2011,25(12):86-90.
[14]連飛. 工業(yè)集聚與勞動生產(chǎn)率的空間計量經(jīng)濟分析:來自我國東北34個城市的經(jīng)驗證據(jù)[J].中南財經(jīng)政法大學學報, 2011,(1):108-114.
[15]胡玉敏,杜綱. 中國城市增長的空間計量經(jīng)濟學研究[J].科學·經(jīng)濟·社會,2012,30(1):50-56.
[16]Elhorst J P. Dynamic Spatial Panels: Models, Methods and Inferences [J]. Journal of Geographical Systems,2012,(1):5-28.
[17]LeSage J P, Pace R K. Introduction to Spatial Econometrics [M]. Boca Raton,US:CRC Press/Taylor & Francis, 2009: 20-44.
Abstract Based on the urbanization data of 12 provinces in the western region from 2000 to 2011, provincial urbanization dynamic factors in the western region are analyzed through spatial econometrics methods. According to the results of the spatial correlation test, there is a spatial correlation in the western region provincial urbanization,and there is a spatial agglomeration trend. Therefore, using the traditional OLS model will miss the space factors, and spatial econometric model is more suitable. Spatial interaction effect of provincial urbanization in the western region is verified by spatial durbin model; on the basis,the spatial effects are decomposed.The results show that the transformation of industrial structure in western provinces has the most positive direct effects to provincial urbanization,and due to the industrial competition between provinces,transformation of industrial structure in neighboring provinces has negative indirect effects to provincial urbanization; level of economic development in western provinces has significant positive direct effects to provincial urbanization,but due to the gap of economic development between western provinces is small, and the indirect effects of economic development in neighboring provinces to provincial urbanization is not significant;the positive direct effects of urban land scale expansion to provincial urbanization are the second only to transformation of industrial structure,and due to the imitation between western provinces,it is has the most positive indirect effects to provincial urbanization;infrastructure investment of western provinces has significant positive direct effects to provincial urbanization, but subject to the low level of provincial fiscal revenue,the indirect effects of infrastructure investment between western provinces are not significant;because of the disadvantage of location, the direct effects of openingup degree to provincial urbanization is not significant,but the enhancement of openingup degree in neighboring provinces will cause the provincial economic element spillover,it is has significant negative direct effects.Therefore, in the process of provincial urbanization in the western region, decision makers shouldnt only enhance provincial own motivation factors, should also coordinate the urbanization motivation factors between provinces.
Key words spatial effects; westerns region; urbanization; spatial Durbin model