Effects Of Housing Prices On China

1068 WordsApr 5, 20175 Pages
Introduction Since the estate market has experienced a dramatic development and the soaring speed of housing price has been warned by scholars and entrepreneurs (Jethro Mullen, A. S. 2016). According to the data from National Bureau of Statistics of China, Zhaoyingzi Dong, Eddie C.M. Hui and Shenhua Jia find that the consumption rate reduced while the housing price in China increased significantly during the period from 2000 to 2014. So, does the rise in housing price suppress consumption? To address it, the writers introduced two main points in academia, wealth effect and substitution effect. After observing the data and estimating models, they combine these two points and provide a new understanding of the impact of housing price to…show more content…
In this study, the model with two estimated threshold parameters, γ1 and γ2, is chosen because of the heterogeneity of estate market and financial market in different cities. So, the model is following: y_(it=) u_i+β_1^ ' x_it∙1(q_it≤γ_1 )+β_1^ ' x_it∙1(γ_1γ_2 )+ε_it where qit is the threshold variable as well as γ1 and γ2 represent the ratio of housing price-to-income (rem) and the ratio of deposit to GDP (fd). And the models in this case are: 〖ln⁡cons〗_(it=) μ_i^1+β_11^1 ln⁡〖hp_it 〗∙1(rem_it≤γ^1 )+β_12^1 ln⁡〖hp_it 〗∙1(rem_it>γ^1 )+β_2^1 ln⁡〖〖inc〗_it 〗+β_3^1 ln⁡〖〖saving〗_it 〗+ε_it^1 〖ln⁡cons〗_(it=) μ_i^2+β_11^2 ln⁡〖hp_it 〗∙1(fd_it≤γ^2 )+β_12^2 ln⁡〖hp_it 〗∙1(fd_it>γ^2 )+β_2^2 ln⁡〖〖inc〗_it 〗+β_3^2 ln⁡〖〖saving〗_it 〗+ε_it^2 After confirming the stationarity of all variables by undertaking unit root test, the writers test the model when assuming there is zero, one or two thresholds in the fixed effect model to determine the number of thresholds of rem and fd. The output indicates that rem has two thresholds and fd only has one. Then, the further tests show that γ1 and γ2 are 5.0882 and 5.9625 respectively. The single threshold value of fd is 1.8827. The findings of threshold estimation and grouping regression suggest that on the one hand, for the areas where the housing price-to-income ratio is between 5.0882 and 5.9625, the substitution effect become dominant. On the other hand, the wealth effect is more
Open Document