查詢結果分析
來源資料
頁籤選單縮合
| 題 名 | 人工智慧地價稅:演算法外部性對都市土地稅制效率與空間效應的實驗分析=Artificial Intelligence and Land Value Taxation: Algorithmic Externalities and Urban Spatial Effects |
|---|---|
| 作 者 | 劉皓仁; | 書刊名 | 科際整合月刊 |
| 卷 期 | 10:11 2025.11[民114.11] |
| 頁 次 | 頁51-68 |
| 分類號 | 567.241 |
| 關鍵詞 | 土地稅制; 人工智慧稅務; 空間計量經濟學; 演算法外部性; 都市密度; AI監管; 土地使用效率; Land taxation; AI taxation; Algorithmic externalities; Spatial measurement economics; Urban density; AI regulation; Land use efficiency; |
| 語 文 | 中文(Chinese) |
| 中文摘要 | 本研究運用經濟實驗方法,探討人工智慧(AI)導入地價稅制度後,如何重構都市財政治理中效率與公平之間的動態平衡。研究結合實驗經濟學與跨國制度比較,模擬AI輔助地價稅在「演算法外部性」條件下的影響。透過具空間自迴歸(SAR)誤差結構的隨機森林模型,本研究比較傳統地價稅(TPT)與AI地價稅(A-LVT)之差異。結果顯示,A-LVT可提升稅收效率約12%,並使都市緊密度增加5.8%,但同時擴大地主間的收入差距與空間偏誤。跨國分析顯示,透明治理與再分配機制可緩和演算法外部性:愛沙尼亞的開放資料政策最能兼顧效率與公平,德國的多層協調模式與韓國的中央集權模式則加劇區域落差。本研究提出「演算法外部性」作為AI效率與財政正義之間的中介變項,並建構「技術層-制度層-社會層」三層治理模型,以說明AI稅制的制度化條件。研究結論指出,唯有透過法律與公民參與的制度框架,方能將AI的技術效率轉化為土地政策中的分配正義。 |
| 英文摘要 | This study employs economic experiments to investigate how artificial intelligence (AI), when introduced into the land value tax system, can reconfigure the dynamic balance between efficiency and fairness in urban fiscal management. The study combines experimental economics and cross-national comparisons to simulate the impact of AI-assisted land value taxation under conditions of "algorithmic externalities." Using a random forest model with a spatial autoregressive (SAR) error structure, the study compares the differences between the traditional property tax (TPT) and the AI-based land value tax (A-LVT). The results show that the A-LVT increases tax efficiency by approximately 12% and urban density by 5.8%, while simultaneously widening income disparity and spatial bias among landowners. Cross-national analyses indicate that transparent governance and redistributive mechanisms can mitigate algorithmic externalities: Estonia's open-data policy performs as the most efficient and equitable model, whereas Germany's multi-level coordination system and Korea's centralized approach exacerbate regional disparities. This study conceptualizes "algorithmic externality" as a mediating variable between AI efficiency and fiscal justice, and develops a three-tier governance model-"technical layer - institutional layer - social layer"-to illustrate the conditions for institutionalizing an AI-driven tax system. The study concludes that only through an institutional framework grounded in law and citizen participation can the technical efficiency of AI be transformed into distributive justice in land policy. |
本系統中英文摘要資訊取自各篇刊載內容。