Evaluating The Effectiveness Of Three Restrictive Housing Policies In China

The Global Financial Crisis,resulted from over expansion of credit and asset prices bubbles,caught policymakers'attention in macro-prudential policies as vehicles for stabilizing the credit and housing markets,in developed countries as well as emerging ones.As housing market plays a vital role in residences'livelihood and many industries'development,controlling housing prices within the appropriate range becomes one of the most significant targets that the relevant administrations are trying to achieve.In China,restrictive housing policy as an important means of regulating house prices is widely implemented in major cities.Using monthly city-level data,this paper investigates the effectiveness of three typical restrictive housing policies in China,including limited purchase policy,limited sales policy and limited credit policy.Firstly,event study analysis is employed to analyze this issue in the cities that respectively implemented these policies.Making the month of policy implementation as the event month,and the housing price index is used to reflect the response of housing markets to these policies.Based on this preliminary analysis,after incorporating other control variables,conventional panel regression models are introduced.The policy dummy variables are created to quantify the impact of each restrictive housing policy.The results show that the effect of the limited purchase policy is not obvious in the initial stage after its implementation,but the negative impact on the price will be magnified in the medium and long term,and becoming stronger.Similarly,the immediate effect of the limited sales policy is not strong,and it begins to work in a few months,but the effect becomes weak in sustainability and inadequate in the later period.In contrast,the effect of the limited credit policy on inhibiting increasing housing prices is the most stable among these three policies,working the fastest and the most persistently.The innovation of this paper can be summarized into three points.As for data selection,local policies are selected randomly and collected manually one by one instead of using a unified national policy,which can capture policies'release dates and effective dates more accurately.Also,by collecting samples for three kinds of policies respectively then using the same research methods to analyze,the effectiveness of these different policies can be compared,which has important reference significance for future policymakers.As for the data frequency,using moderate frequency monthly data can more accurately capture the immediate,medium and long-term impacts of policies.In addition,event months are employed in the panel regression models instead of calendar months.This can make the acting time length of the policy in each sampled city constant,so that each individual contributes the same degree to the final results,then improving the accuracy of the research.