Zhang Ming and Liu Yao | Analysis of the driving factors of housing price trends and fluctuations in China's cities

Note: This article is published in Nanjing Social Sciences, No. 6, 2021, in collaboration with doctoral student Liu Yao, please be sure to indicate the source when reprinting. In order to save space, this paper omits the quantitative research part, references, English abstracts and footnotes, and the full text can be found on CNKI.

【Abstract】:This paper constructs an analytical framework for studying the trend and fluctuation of china's housing prices, divides the variables affecting housing prices into demand, supply, finance and policy, empirically analyzes the driving factors of housing prices in 31 provinces and autonomous regions and 70 large and medium-sized cities in China from 2005 to 2017, and further explores the heterogeneity of housing price drivers in first- and second-tier cities and five regional core city clusters Population inflow and the number of high-quality resources, personal loan growth rate and household sector leverage ratio are the most significant driving factors for the trend of house prices, per capita income, money supply growth and policy regulation have the most obvious impact on house price fluctuations; second, demand and financial variables drive the trend of housing prices in 70 large and medium-sized cities across the country, in the demand variables, the proportion of permanent population is a significant driving factor for the rise in house prices, among financial variables, the reduction of personal housing loan interest rates has the greatest effect on house prices. Almost all financial variables drive the fluctuation of housing prices in large and medium-sized cities; third, the trend and fluctuation of house prices in first- and second-tier cities are more significantly driven by financial factors, and the per capita land purchase area is also an important factor in the change of house prices; fourth, in recent years, real estate regulation has led to reverse fluctuations in house prices to a certain extent, and the policy is difficult to achieve the expected effect. The above shows that population or resource flows, credit and leverage ratios are important reasons for the rise and fall of national housing prices in this round, while the widening of per capita income, excessive agglomeration of funds and imbalance of land supply are the important reasons for the widening of urban housing price differentiation in this round.

I. Introduction

Since the reform of China's urban housing system in 1999, the level of residents' housing consumption has continued to increase, and housing has become the highlight of household sector consumption, and even the "barometer" of China's economy. In particular, in the more than ten years from 2005 to 2015, China's real estate market has shown a remarkable feature: on the one hand, the general acceleration of housing prices in various provinces and cities in China, and on the other hand, the regulatory authorities represented by the Ministry of Housing and Urban-Rural Development have frequently regulated house prices. After 2015, China's real estate showed a new round of recovery and recovery, and house prices also quietly appeared a significant differentiation between regions and cities: first-tier cities led the rise, followed by second-tier cities in coastal areas, and the price increases in third- and fourth-tier cities varied dramatically, and this change gradually formed a new trend of China's housing price changes. This seems to indicate that the use of a single national level of macro data analysis of China's housing price trends will no longer be applicable, and exploring the heterogeneity between provinces and cities will be crucial to grasping the future of China's housing prices.

At the same time, house prices play a key role in assessing financial stability and warning systemic risks. When the housing bubble is over-concentrated and even leads to a sudden and significant decline in housing prices, it will have a huge impact on the balance sheet of the household sector, corporate confidence and the borrowing capacity of commercial banks. The data shows that in 2018, the house price-to-income ratio of China's 50 large and medium-sized cities was 11.36, and behind the high house prices supported by such high leverage also hides the risk of a significant downward trend in house prices. Then, classifying the drivers of housing prices has positive implications for maintaining economic growth, maintaining financial stability, and stabilizing consumer confidence.

In view of this, this paper constructs an analytical framework for studying the trend and fluctuation of China's housing prices, divides the driving factors of China's housing prices into supply factors, demand factors, financial factors and policy factors, and discusses the driving factors of China's housing price trends and fluctuations from 2005 to 2017 by region and city, taking 31 provinces and autonomous regions and autonomous regions and 70 large and medium-sized cities as samples. If this round of house price changes is mainly driven by supply and demand factors, then it means that the housing prices in the region are still determined by the economic fundamentals, and the government can jointly build a long-term mechanism for adjusting house prices from the supply side and the demand side to guide the house to "only live and not speculate"; if this round of house prices is more determined by financial factors and policy factors, then it means that the housing prices in cities in the region have deviated from the fundamentals, have certain financial attributes and even asset bubbles, once the strict policies to suppress house prices or residents significantly increase leverage, Such areas will have higher downside risks for house prices. The structure of the remainder of this article is as follows: the second part is a literature review, the third part constructs an analytical framework for studying the trend and fluctuation of China's housing prices, the fourth part is a quantitative empirical analysis of the driving factors of China's housing price trends and fluctuations, the fifth part is the robustness test, and the last part is the conclusion and policy recommendations.

2. Literature review

Literature on real estate drivers is not uncommon. Muth (1971) constructed a model from a microscopic perspective, showing that variables such as construction cost per unit area, distance from CBD, and land use area are all influencing factors driving housing price trends. As the degree of population and capital flows increases significantly, more literature is beginning to focus on the macro factors driven by real estate. Financial conditions are an important variable that more literature has paid attention to, and financial conditions can drive house price fluctuations through a variety of channels. For example, rising risk premiums on corporate bonds will drive credit supply contractions, which will affect asset price volatility, including house prices (Gilchrist & Zakrajšek, 2012); Adrian et al. (2019) analyzed the potential impact of financing conditions on GDP growth and found that tight financing conditions in the short term usually lead to increased downside risks to the economy, with lower housing consumption due to lower household incomes and increased unemployment; Ortalo-Magne & Rady (2005) found that household income is an important variable driving house prices and has important implications for predicting house price trends.

Household sector indebtedness is another key variable affecting the trend of house prices. The IMF (2019) found that in both developed and developing countries, the high credit boom is highly correlated with the downward trend in house prices; the transmission mechanism of household sector debt on house prices is similar to the transmission mechanism of financial conditions, and the increase in household sector leverage will have a negative impact on economic growth, household income and employment, and household debt through the above three channels has a potential impact on the real estate market. After the 2008 global financial crisis, some scholars began to explore the relationship between household sector leverage, housing bubbles and financial crises in developed countries, but they came to the opposite result. Dang et al. (2010) found that the outbreak of the financial crisis was directly related to the accumulation of subprime loans in the real estate market, while the increase in subprime loans was closely related to the high leverage of low-quality borrowers. However, Bhutta (2015) questioned this view, finding that real estate market capital inflows from real estate investors grew faster than capital inflows from first-time buyers with lower credit scores, suggesting that the debt of subprime borrowers was actually limited in its contribution to the financial crisis.

Capital flows and demographic structure are also one of the factors driving the trend of housing prices. Cross-border capital flows typically drive housing prices in advanced economies, such as Caballero et al. (2008), which empirically found that excess savings in emerging markets chased safe, quality assets in the United States, which, combined with long-term low interest rates in the United States, led to cross-border capital inflows pushing up U.S. house prices. For China, housing prices in a city are often driven by demographics and interregional capital inflows because the capital account is not fully open. Xu Jianwei et al. (2012) believe that the increase in the dependency ratio of young people will push up house prices; Zou Jin (2014) believes that the age structure of the population will determine the fluctuation of house prices in the long run.

In addition, supply, demand and cyclical factors also play an important role in the determination of house prices. Philips (1988) argues that both income and expectations drive house prices. Gattini & Hieber (2010) used housing supply indicators to predict housing market developments in the euro area, finding that housing supply indicators including residential investment and real interest rates correlate with fluctuations in house prices. Yu Huayi (2010) found that land policy is an important factor driving house prices, and the increase in land supply helps to drive house prices up. Aherene et al. (2005) found that house price changes are related to the credit cycle, and that house prices will first face an increase after the money supply in OECD countries increases.

From the perspective of research methods, except for a few literature studies that study the impact of exogenous shocks on housing prices under the framework of the DSGE model, most of them are empirical studies. For example, Liu et al. (2011) constructed a DSGE model that integrates house prices and fixed investment, discussing how credit constraints affect macroeconomic variables such as house prices; Hirata (2012) analyzed the drivers of global house price fluctuations by constructing a FA-VAR model containing interest rates, monetary policy, output, credit and uncertainty factors, and found that the house price trend in major countries around the world is synchronous, and the global interest rate shock will have a negative impact on house prices, but monetary policy has little impact on it. The impact of uncertainty is the most important factor driving house prices; Liu Jinquan and Lu Mengfei (2018) studied the impact of monetary policy, GDP growth and resident leverage on house prices by constructing a VAR model, and found that loose monetary policy is the main force of China's house price rise. Kuang Weiwei (2013) used a questionnaire of 1040 urban housing owners in Beijing to investigate the impact of the introduction of real estate tax on expected house prices using the logit model, and found that the effect of the introduction of real estate tax on suppressing house prices is limited, and the earlier the levy, the better; Ding Ruxi and Ni Pengfei (2015) used exploratory data analysis methods and spatial measurement techniques to study the regional spatial pattern and characteristics of urban housing prices in China. It was found that china's inter-city housing prices depend on the spatial positive spillover effect of housing price fluctuations in surrounding cities, and this effect is heterogeneous between regions.

Combined with the above research, it is not difficult to find that most of the literature focuses on exploring the impact of a single variable on the housing price drive, mostly using the VAR model to regard the house price as an endogenous variable, and few documents have made a panoramic exploration and consideration of the degree of contribution to the driving factors of urban housing prices in China. Based on this, this paper constructs an analytical framework for studying the trend and fluctuation of China's housing prices, explores the driving factors of housing prices in 31 provinces and autonomous regions in China and 70 large and medium-sized cities across the country from 2005 to 2017, and the main contributions of this paper are as follows: First, the analysis framework of the influencing factors of China's housing prices is constructed, which divides the factors driving housing prices into supply, demand, finance and policy variables, and discusses the main factors driving the trend and fluctuation of house prices and the degree of contribution; second, from the provincial level to the prefecture and city, from the whole to the region. Third, the results of this paper reflect to a certain extent whether housing prices in the region or city are driven by economic fundamentals or financial factors, providing a certain reference value for policy makers and home buyers.

Third, the analysis framework of China's housing price trends and fluctuations

In order to construct an analytical framework for china's housing price trends and fluctuations, based on the fact that Chinese residential housing may have both commodity and financial attributes, and housing prices are greatly affected by policy changes, with reference to the research of Ping An Securities (2017), we divide the driving factors of China's housing price trends and fluctuations into four categories: the first is demand factors, the second is supply factors, the third is financial factors, and the fourth is policy factors. These four types of factors basically include the main causes affecting China's housing prices.

(1) Demand factors

The demand factors driving China's housing prices can be roughly classified into three categories: one is the resident income variable, one is the demand for the stock population, and the other is the demand for the flow of population, as follows:

Income level per capita: Expressed in terms of GDP per capita in a region or city. According to the wealth effect of housing residents, the per capita income level is the most critical variable affecting the demand for housing purchasing power, and the higher the income level, the higher the demand for housing, and therefore the higher the house price.

Resident/Registered Population: The local household registration population usually has a stable number of family housing. In contrast, the foreign resident population is a rigid source of demand for home purchases in the region or the city. The larger the value of this ratio, the stronger the demand for housing, and the higher the house price. For large and medium-sized cities with rapid population inflows, the positive correlation between the increase in this proportion and the rise in house prices will be more significant.

Working population/permanent population: In the permanent population, the larger the proportion of the working population, indicating a stronger rigid demand for housing. Cities with the highest price increases in house prices typically have a higher proportion of the working population. This variable is usually positively correlated with house prices.

The number of high-quality public resources: The more high-quality public resources in a region or city, the higher the potential population inflow, indicating that the potential demand for housing is more exuberant, which will push up house prices for a long time. Education and health care are usually the highest quality public resources, and we use the number of "211 universities + top three hospitals" as a tool variable for the number of high-quality public resources.

(2) Supply factors

Housing as an ordinary commodity, the price is determined by the law of supply and demand. Supply factors affecting housing prices usually include land supply area and housing inventory.

Land acquisition area: Land supply is an important driver of house price fluctuations. However, due to differences in regional or urban area size, the overall land supply has little correlation with housing prices. However, the per capita land acquisition area usually has a strong negative correlation with house prices, that is, the smaller the increase in house prices in areas or cities where land supply is more sufficient. This variable is expressed in terms of land acquisition area/permanent population.

Per capita commercial housing inventory: expressed in inventory/permanent population. Theoretically, the lower the inventory of the area or city, indicating that the more supply of housing is in short supply, the greater the room for house price growth. This variable is usually negatively correlated with house price increases.

(3) Financial factors

The financial factors driving China's housing prices are mainly related to the size of credit, the level of interest rates and the level of valuation. On the one hand, house prices are usually related to the rotation of the credit cycle and interest rate cycle, on the other hand, the house as an asset, has both financial attributes, so the house price is also greatly affected by financial factors.

Housing loan balance growth rate: The credit cycle and the housing price trend have a high homogeneity and mutual reinforcement effect. For example, the expectation of rising house prices has increased the willingness of residents to buy houses, resulting in an acceleration of the growth rate of housing loan balances, while conversely, the excess willingness of residents to buy houses has pushed up the upward expectations of house prices. Therefore, the faster the housing loan balance grows, the faster house prices will rise.

M2 year-on-year growth: The faster M2 grows year-on-year, it usually indicates that the liquidity environment will be more relaxed, which will boost house prices. At the same time, according to historical data, the year-on-year growth rate of M2 is usually 5-6 months ahead of house price changes, which is a leading indicator of house price trends.

Home Loan Rate: The level of lending rates is an important factor driving house price trends. In theory, rising housing loan interest rates will increase the financing costs of home buyers and inhibit the rise in house prices; conversely, a decline in housing loan interest rates will reduce home buyers' costs and boost house prices. It is worth noting that the change in the interest rate of housing loans has a limited effect on buyers who just need it, and has a stronger effect on buyers with investment needs.

House purchase leverage level: The house purchase leverage level is regarded as the "price-to-earnings ratio" of house prices as a valuation indicator. We usually take the house price income ratio as a representative variable of the leverage level, under the condition of forming unilateral expectations, the higher the house price income ratio, the easier it is to be amplified by the house price rise expectations, so the two usually show a positive correlation. At the same time, the leverage ratio of the household sector can also be regarded as another alternative indicator of the leverage level of house purchases, and the leverage of the household sector usually has a certain role in promoting the price of assets such as houses.

(4) Policy factors

Policy regulation: In practice, we find that the driving effect of policy factors on the trend of China's housing prices should not be underestimated. For example, when the Ministry of Housing and Urban-Rural Development promulgates the real estate regulation and control policies of "National Six Articles" and "National Eight Articles", the trend of house prices may be reversed by exogenous shocks. Since there are direct policies and indirect policies for formulating plans for real estate regulation and control policies, we only include direct policies that affect house price expectations into the analytical framework. In the case of rational expectations, if a strict regulation and control policy is introduced, it is regarded as a policy to suppress house prices; if there is a loose policy such as the cancellation of the restriction order, it is regarded as a policy to boost the rise in house prices. We treat each policy as a dummy variable, with boosting house prices being recorded as +1 and suppressing house prices as -1.

Empirical analysis: the driving factors of China's housing price trends and fluctuations

Fifth, the robustness test

6. Conclusions and Policy Implications

Since 2005, China's real estate market has shown an overall upward trend and intensified differentiation between cities. By constructing an analytical framework for the trend and fluctuation of house prices in China, this paper divides the influencing factors driving house prices into demand, supply, finance and policy variables, from the inter-provincial to the urban level, constructs the panel regression equation, examines the driving factors of house price trends and fluctuations in 31 provinces and autonomous regions and 70 large and medium-sized cities in China from 2005 to 2017, and further explores the heterogeneity of the driving factors of house prices in 35 core cities and five regional core city clusters, the main conclusions are as follows:

First, from a nationwide perspective, the trend and fluctuation of housing prices in various provinces are determined to a certain extent by the economic fundamental factors of supply and demand, and housing still has a large degree of commodity attributes. The increase in the supply of housing land, the increase in per capita income level, the inflow of permanent and working population, and the accumulation of high-quality resources will all drive house price changes, but the "total" policy regulation may cause reverse fluctuations in house prices in the short term, and it is not easy to meet the ideal policy expectations.

Second, from the perspective of housing prices in 70 large and medium-sized cities across the country, the trend and fluctuation of urban housing prices are mainly driven by demand and financial variables, which may indicate that the housing of major cities in the country has both commodity and financial asset attributes. The increase or decrease of land supply and commodity housing inventory is no longer a significant factor affecting house prices, and the changes in residents' leverage ratios, the growth rate of money supply, and the adjustment of personal loan interest rates are the biggest contributors to the trend and fluctuation of house prices. This may indicate that the future trend of housing prices in major cities across the country has greater uncertainty, and it is necessary to be vigilant that private sector debt may form linkage and spillover effects, push up house prices or lead to falling house prices, triggering financial risks.

Third, from the perspective of housing prices in 35 core cities across the country, the financial attributes of housing in first- and second-tier cities may be much greater than the attributes of commodities, and the leveraged variables of house price-to-income ratio are the biggest contributors to the housing price trend in core cities. This means that house prices in first- and second-tier cities have departed from traditional economic fundamentals, and the volatility of house prices will rise significantly in the future, and the risk of the real estate market will increase significantly. At the same time, land supply drives the trend of housing prices in first- and second-tier cities to a certain extent, which means that increasing the supply of housing land can effectively "cool down" the "overheated" property market.

Fourth, from the perspective of housing prices in the five major regional core cities in the country, the housing prices in the Beijing-Tianjin-Hebei region and the core cities in the Yangtze River Delta region are mainly driven by financial factors, in view of the normalized high housing prices in the two regions, the housing prices in the two regions will face more uncertainty risks in the future; the housing prices in the core cities of the Pearl River Delta region are mainly driven by demand and financial variables, in view of the preferential policies for the introduction of talents and the concentration of high-quality resources in the region, the real estate market of the core city cluster in the Pearl River Delta will continue to be active in the future The housing prices of the core cities in the central triangle and the western triangle region are still dominated by the fundamental factors of supply and demand, and the growth potential and spatial spillover effect of several core cities will lead to demand factors that may push up the house prices in the two regions in the future, and the core cities in the central triangle and the western triangle region in the future will be the "potential stocks" of the real estate market.

The conclusions of this article have certain policy implications:

First of all, China's real estate market should strengthen structural regulation and control policies, and should also avoid the reverse expectations caused by frequent changes in similar policies or policy lags. Under the new normal, real estate regulation and control should build a differentiated regulation path and system that divides regions and cities, and "implement policies according to the city". The real estate regulation and control policies of first- and second-tier cities should focus on increasing supply, while third- and fourth-tier cities should avoid the formation of large-scale inventory;

Second, regulators should focus on monitoring housing prices in first- and second-tier cities, especially to curb speculative housing speculation in the Beijing-Tianjin-Hebei region and the Yangtze River Delta region;

Third, government departments should implement the affordable housing and low-cost housing system in hot cities, guide the housing prices in hot cities to gradually return to rationality, and form a demonstration effect and positive externalities in a certain area;

Finally, it is necessary to avoid inducing the residential sector to increase leverage too quickly, and commercial banks should carefully determine the down payment ratio and loan amount of housing loans to avoid the occurrence of default risks and systemic financial risks.