住宅負擔能力的空間差異性與影響因素探討—以台灣六都為例

Abstract

台灣在近年來房價不斷上升是有目共睹的現象，而根據內政部營建署公告之111年第1季房價負擔能力統計發布內容，全台房價所得比為9.58倍，相對於國外有著較高的比例且在各城市間又有極大的差異，由此可見台灣可能存在著住宅負擔能力分布不均且不足的問題。本研究欲透過相關資料並以空間統計方法，探討全台各村里之可負擔房價與住宅負擔能力空間分布差異。本文使用2018至2019年內政部不動產實價登錄交易資料以及綜合所得申報之家戶所得推算全台各村里之貸款負擔率以衡量住宅負擔能力狀況。並透過空間自相關分析方法，觀察與分析各村里家戶住宅負擔能力以及不可負擔房屋單價的空間分布，再以地理加權迴歸模型（GWR）探討影響區域住宅負擔能力差異的相關因素。研究結果顯示不可負擔房屋單價在六都均具有空間自相關的現象，且多集中在改制前的舊市區，以區域型空間自相關來看發現在各都市不可負擔房屋單價的熱區內仍有較低住宅負擔壓力的村里，此外在舊台中市的內部住宅負擔能力問題也具有空間的異質性，透過地理加權迴歸模型探討住宅負擔能力問題較OLS表現較佳，調整後R2達到0.77，對住宅負擔能力有顯著影響的變數有屋齡、建築面積及老化指數等非距離因素，以及轉運站距離、文化設施距離、百貨公司距離、工廠距離、金融設施距離及捷運站距離等距離因素，各影響變數對住宅負擔能力問題的影響在空間上均具有差異性，因此建議在規劃與實施解決住宅負擔能力問題的政策時，應將空間的相關影響因素考慮進去，才可針對不同的區域規劃以及住宅政策的制定進行合理的調整。In recent years, it is obviously to say that housing price in Taiwan has been rising continuously. According to the housing affordability statistic by Construction and Planning Agency in Q1 2022, Price to Income ratio（PIR）was 9.58, which is a relative high ratio compared to Foreign countries. There are also great differences among Taiwan cities, which shows that Taiwan may have the problem of uneven distribution and insufficient housing affordability. This study intends to explore the differences in the spatial distribution of affordable housing prices and housing affordability in villages across Taiwan through the use of relevant data and spatial statistical methods.This thesis uses real estate transaction data and the household income to calculate the Mortgage Burden in each village in Taiwan to measure the housing affordability from 2018 to 2019. And through the spatial autocorrelation analysis method, observe and analyze the spatial pattern of housing affordability and unit price of unaffordable housing in each village, finally using the geographically weighted regression model (GWR) to explore the relevant factors that affect the difference of the spatial distribution of regional housing affordability.The research results show that the unit price of unaffordable housing has spatial autocorrelation in the six Municipalities, and most of them are concentrated in the old urban areas before the restructuring. Through regional spatial autocorrelation analysis, it is found that there also have lower housing affordability pressures in the hot spot of unaffordable housing unit price in each city. In addition, the internal housing affordability problem in former Taichung city also has spatial heterogeneity. Using GWR to explore the housing affordability problem performed better than OLS, and the adjusted R2 reached 0.77. Factors that have a significant impact on housing affordability can be categorized into non-distance and distance-related variables. The former includes housing age, building area, and aging index, while the latter comprise proximity to transit stations, cultural facilities, department stores, factories, financial institutions, and MRT stations. The influence of each impact factors still has spatially difference. Therefore, it is recommended to incorporate spatial factors when formulating and executing policies to address housing affordability concerns, leading to more tailored regional planning and housing strategies.