房地合一稅對房屋市場之影響

Abstract

本研究探討房地合一稅實施對房屋市場之影響。新制與舊制皆依持有時間長短適用不同的稅率，並且日出條款導致施行新制後市場上仍存在適用舊制之交易。本研究透過持有時間、前一筆交易日期以及交易日期為邊界將稅制分為六類，接著利用多項式估計稅制邊界不存在下的反事實分配以辨認稅制邊界附近的交易是否有聚集而導致交易次數分配扭曲，並以此分析買方與賣方的行為。 實證結果發現新制與舊制的持有時間分配兩年處皆有顯著的扭曲，但新制的扭曲小於舊制，顯示舊制下賣家透過延長持有時間來躲避高稅率的誘因較新制大。其次，持有時間兩年以下的賣家會選擇在2016年以後出售，顯示他們寧願負擔房地合一稅的高稅率也不願繳交奢侈稅，買方則沒有顯著偏好特定的時間點購入房屋，顯示他們沒有為了未來出售時可以適用特定的稅制而改變交易時間。最後，本研究使用特徵價格模型估計稅制對房價的影響，發現無論新制或舊制持有時間兩年以下的房價都顯著高於持有時間兩年以上，但新制與舊制間並無顯著差異。

This study empirically investigates the effect of integrated housing and land tax on the Taiwanese housing market. The new and the old tax regimes impose different tax rates according to the length of the holding period. The Sunrise Clauses leads to the coexistence of the old and new tax regimes in the housing market after the implementation of the new tax regime. First, all the transaction data were split into six tax regimes based on their holding period, and current and previous transaction dates. Next, flexible polynomials were applied to estimate the counterfactual distribution under the absence of the notches between different tax regimes and compared with the actual distribution to identify whether transactions bunch near tax regime notches. The distortions can also be used to analyze the behavior of the buyers and sellers. The results show that significant distortions exist in the distribution of the holding period in both the new and old tax regimes at two years. However, the distortion in the new tax regime is less significant compared to the old tax regime. Furthermore, sellers who held the real estate for less than two years will choose to sell them after 2016, indicating that they would rather pay the high rate of the integrated housing and land tax than the luxury tax. On the other hand, buyers did not have a significant preference for a particular time of purchase, suggesting they did not manipulate the time of the transaction to apply a specific tax regime for future sales. Finally, the hedonic price model is used to estimate the impact of tax regimes on housing prices, and the results show that housing prices under the new and old tax regimes for a holding period of fewer than two years are significantly higher than those of more than two years. However, there is no significant price difference between new and old tax regimes.