Determinants of housing price evolutions for the French sub-markets

Real estate price dynamics are well described at a large scale, but it suffers from a poor understanding of the fundamentals behind these dynamics, especially locally. Indeed, many indicators produced by different actors describe the fluctuations of prices well but fail to explain the mechanisms driving these changes at a local scale. This work aims at measuring the relevancy and the explanatory power of a set of potential economic, demographic and financial fundamentals on the housing price evolutions at a local scale, covering all France. The main objectives are to highlight the underlying factors of sub-markets in France and quantify their relative influence. The determinants of the housing markets have been studied in several papers at a country scale or for specific big cities. Indeed, some work focused on the Europe scale (Banerji, A., Shi, H., Hilbers, M. P. L. C., & Hoffmaister, M. A. W. (2008), Égert, B., & Mihaljek, D. (2007)), the country scale (Bessone, A. J., Heitz, B., & Boissinot, J. (2005)), or directly in specific cities (Baroni, M., Barthélémy, F., & Mokrane, M. (2008)). If some economic or financial variables like the interest rates or the average disposable income of a household seem to be designated as key factors explaining the fluctuations in prices, the relative influence of each of these variables on housing prices might be different in all sub-markets in France. In this work, the influence of a set of potential determinants is measured thanks to statistical methods which are regularly used for this kind of purpose, such as simple linear regression, panel VAR methods, or cointegration measures. These methods have been tested on a database of indices covering all France. This study is based on a set of more than 800 indices built on a regionalization of France based on socio-demographic and economic data. These indices offer a good representation of house price evolutions at a local scale by construction of regions as fine as the available data allows. The fundamentals tested for this work are a set of time series data of economic, demographic, and financial variables covering the entire period 2010-2020 and are retrieved at the finest geographical scale. These variables were chosen among the more often cited in academic literature as potential underlying factors of real estate dynamics. The fundamentals tested are the following : long term interest rates, short term interest rates, unemployment rates, evolutions of population, inflation, new built construction, income evolution, savings rates, disposal income. The contribution of this paper is twofold. First, we are, to our knowledge, the first to systematically test for the impacts of potential determinants of the evolution of housing prices, over a whole country, at a local scale. From a methodological perspective, it forces us to introduce a new custom econometric approach in order to produce meaningful and intelligible results. Second, we measure significant differences in the role played by the drop of long term interest rates in the price dynamics among local markets. In our opinion this is a proof of the asymmetric transmission of the accommodating monetary policy decided by the European Central Bank from 2015.