查詢結果分析
來源資料
頁籤選單縮合
題名 | 高雄縣市財政收入結構比較與未來稅源預測=The Comparison of Financial Structure and Prediction of Tax Resource in the Future between Kaohsiung City and Kaohsiung County |
---|---|
作者 | 洪東煒; 汪明生; | 書刊名 | 公共政策學報 |
卷期 | 18 1997.09[民86.09] |
頁次 | 頁33-56 |
分類號 | 566.933 |
關鍵詞 | 高雄縣市; 財政收入結構; 稅源預測; |
語文 | 中文(Chinese) |
中文摘要 | 高雄縣市行政區域合併的支持論點,除行政效率提高、避免投資重複浪費外, 強 調地方經濟的自主及區域發展的綜效。由於高雄縣與高雄市的產業結構不盡相同,都市型態 的高雄市,其特色是蓬勃的工商業、服務業,而高雄縣則是以農業為其主軸,在財政收入結 構自然有其差異。高雄縣市合併雖有其政治之考量,但合併對財政收入所造成的影響亦不容 輕忽。本文採 G.E.P. BOX 和 G.M. Jenkins 所共同發展而出的「綜合性自我迴歸移動平均 模型」, (簡稱 ARIMA 模型),與迴歸分析分別對高雄市、高雄縣未來的財政收入做預測 。本文雖利用時間序列分析與迴歸分析分別預測高雄市、縣各項稅收,就各項檢定值來看, 模型所得的結果尚稱合理。但不可諱言的,稅收(尤其是地方稅收部份)以數理模型來預測 ,存有實證上的困難。舉例而言,時間序列分析無法掌握區域性經濟活動特性,加之以結構 的變化,更無法全然以時間的變化來闡述分析,時間序列分析僅大致上能描繪各項稅收成長 的趨勢,而且各項稅收(尤其是有關財產稅的部份),若涉及交易行為與移轉行為,更無法 以時間序列分析窺其全貌。其次,以迴歸分析進行預測,也有合理解釋變數的數據資料取得 問題,因為本文是進行地方稅收的預測,許多影響經濟活動的變數資料,因無地域性的資料 ,只得以較為相關的替代變數資料替代,對於掌握區域特性上,就受到了某種呈度的限制, 故結果不盡理想,但就參考的目的而言,該項分析仍具有掌握未來稅收變化趨勢的價值。市 縣合併後,除市的稅課收入得以維持現行全歸市庫外,縣的稅收除原歸縣庫者將歸入市庫外 ,並將增加原撥省庫之稅收及劃歸鄉鎮(市)稅收,扣除由中央統籌部份外,餘均歸市庫。 即使省對縣補助中尚有部分係中央透過省預算轉撥予縣者,此部分即使合併,補助款總額未 必因而減少,因此對財政收入將是有利而無害。況且縣境除緊鄰市縣交界的鄉鎮已較開發外 ,其餘鄉鎮尚多未開發,深具發展潛力。但由於目前隸屬不同行政轄區,無法作整體規劃, 且各項公共設施興建仍較落後,合併後若能予以妥善規劃,除可緩和市區人口集中的壓力, 更可因腹地的增大,帶動產業的發展,對未來稅收的增加亦將有所助益。本文所採資料來源 均為八十四年度以前正常成長之數據,唯八十五年度已近結束,卻發現不但營業稅呈現二十 年來首度負成長,土地增值稅又短徵一半。由於缺乏高雄市、縣的產業別的細項資料,使本 研究對產業的成長與稅源之預測造成相當大的限制。未來政府發展亞太營運中心對高雄市縣 合併後財政收入的影響,由於受限現存資料數據的缺乏,產業結構改變對各種稅收預測分析 仍付之闕如,此是後續研究有待努力的方向。 |
英文摘要 | The argument for combination of the Kaohsiung City and County has been, additional to enhancement of administrative efficiency and avoidance of repetitive investment, independence of local economy and synergy in regional development. With different industrial structures, the city is strong in manufacturing and service industries, while the county is strong in agricultural activities, the revenue structure in local finance is therefore quite different. Although with mainly political considerations for the combination, its impact on local finance is nevertheless of particular importance. To forecast the future revenue structure of the combination decision, the models of Auto- Regressive- Integrated Moving-Average (ARIMA) and Regression Analysis were used. Although with various examinations of statistics the analysis results seem reasonable, forecasting of local tax revenue with mathematical modeling exist certain uncertainties. For example, characteristics of local economy structure can not be fully reflected in the model, in patroller, the change of the economy structure. The model can only describe the general trend of tax revenue growth but not the various transactions and transfers (especially when property tax is involved). Next, the availability of data becomes a problem when local level analysis is conducted. Nevertheless, the analysis results could still be used for general trend diagnosis for the future. When the city and the county are combined, additional to the original city tax revenue, the revenue to the county, the portion submitted to the province, the portion distribution to the townships, except for those submitted to the central government, would all be added to the city tax revenue. Even the protion of revenue distributed to the county by the central government through the province would not be reduced for the city. Additional benefits for the combination would also include development of those townships that are currently underdeveloped for running by different jurisdictions. This would help also the decentralization of population and the increase of available lard for future development for the metropolitan area, which would further help the future tax revenue. The data used in the study were those before and including the 1995, in the 1996 the business tax shows first reduction in 20 years and property tax revenue is reduced by 50%. Because more detailed data for industrial structure are not available for the city and the county, which has limited the accounting for future tax revenue effects of industrial change, including the APROC policy under planning now. These should be included in future studies. |
本系統之摘要資訊系依該期刊論文摘要之資訊為主。