A Mass Valuation Model Proposal for Residential Property Taxation in Türkiye

Abstract

Real estate valuation, which is an important part of land management functions, is encountered in many applications. One of these application areas is property taxation. In Türkiye, property tax for lands, minimum unit values on street basis are determined by municipalities, and minimum tax value is determined for residential properties. Since spatial, physical and socio-economic characteristics of the properties are not analyzed adequately in the practical studies for taxation purposes, a significant income loss occurs in tax amounts on behalf of the public. In order to ensure fair property taxation in Türkiye, it is clear that the value of each property should be appraised as stated in the property tax regulations, instead of making a generalized valuation on the basis of streets. However, considering the vast amount of properties, it is not practical to carry out the valuation using classical methods. In this study, a mass valuation model based on Nominal Valuation method, which takes into account the criteria expressed in the "Regulation on the Appreciation of Tax Values to be Subject to Property Tax" was developed. The performance of the developed model was tested with 1,000 residential properties whose tax values and current market values are known by the municipality in Gaziosmanpaşa district of Istanbul. According to the results, it has been observed that there is an approximate 6 to 7 times difference between the value of the taxable real estate applied by the Municipality and the market values. On the other hand, several performance metrics of the regression analysis showed that prediction accuracy was high enough. It is evaluated that the developed model will give more accurate results with the increase of data density and the use of real market values instead of property listing data.