A case study on finding efficient monetary policy to solve Shenzhen's excessive-priced housing

The relationship between open market operation and housing prices

In an open market operation, the central bank swap currency for bonds (Rocheteau, Wright, and Xiao, 2018). In China, this included the typical purchasing and selling of government bonds to expand or tighten base money. In addition, the People’s Bank of China (PBC), the central bank in China, controlled the scale and frequency of the short-termed issuance to depository institutions, which was seen as a method to reduce the supply of base money (Xu and Chen, 2012). The money supply is the main aim of open market operation, the more frequent the open market operations, the lower the money supply. Then, the small amount of money supply will make the housing price with a lower increasing level. Liang and Cao (2007) investigated the relationship between property price and bank leading in China from 1999 to 2016 and found that there was a unidirectional causality running from bank lending to property price. In this case, the housing prices were directly connected with the open market operation, and prior researchers had done numerous studies on the negative correlation between housing prices and open market operation. Xu and Chen (2012) examined the monetary policy variables, including money supply growth, had the impact on the dynamic growth of real estate price in China. They explained that the contraction in the growth of money supply may decline the loan-making ability of commercial banks and altering the credit supply to the real estate sector, then going on to elaborate on the influence on the public’s inflationary expectation and the demand for real estate assets. Finally, they summarized that all these channels result in a positive effect of money supply growth on the change in real estate growth. In further study, Lastrapes (2002) simulated the dynamic response of real house prices to exogenous money supply shocks as predicted by theoretical models. In conclusion, the empirical outcomes consistently demonstrate that the slower money supply growth contributes to the deceleration in the housing price growth subsequently. Therefore, the money supply can act as the key driving force behind the real estate price growth in China.

The relationship between discount rate and housing prices

The discount rate is the interest rate charged by the central bank for loans of reserve funds to commercial banks and other financial intermediaries. In the analysis of the interest rate approach, the discount rate has shown a negative relationship with housing prices. When the discount rate has increased, then the rate of increase in house price has slowed. Kau and Keenan (1980) found that when the currency authorities raised the interest rate, it caused real estate investment to fall, and then the price of commercial housing to also fall. Abraham and Hendershott (1994) revealed that the rise of housing price was negatively correlated with interest rate by using the model of housing price change considering lag process. Similarly, in the UK, Iwayama, Akiyama and Ishigaki (2003) established a vector autoregressive (VAR) model about the influence of monetary policy on the real estate market. This illustrated that the rise by 50 basis points on interest rate led to a fall in housing prices of about 0.8% in five quarters in the UK. After several years, Meltzer (1995) set up an asset price model which suggested that monetary policy could directly push housing prices up. However, at the same times, he argued that if a central bank fixes as interest rate, which was the common practice in China, and it will then let property prices to rise. Therefore, the results illustrated that the interest rate should reduce when dealing with the high level of housing prices. Despite the economic situations and developments being different compared to China, the linkage between housing prices and discount rate is relatively the same. Tan and Wang (2015) claimed that monetary policy and housing prices interacted with each other through direct and indirect channels. The impact of interest rates on urban housing prices was negative in the long run and gradually weakened (Wang & Guo, 2007). This study established a VAR model to reach the conclusion. According to Chen and Zhou (2011), the increase in interest rates had a restraining effect on the rise in urban housing prices. Meanwhile, interest rates had limitations existing when taking into account regional differences. A study by the Research Group of the Investigation and Statistics Office of Kunming Central Sub-branch (2018) pointed out that the interest rate regulation policy had obvious location differences. The role in regulating the region where the real estate market development was relatively immature and the rigid demand for real estate was relatively weak was more obvious. Nevertheless, combining with the housing market is relatively well established and mature in Shenzhen, the interest rate should still be regulated to an even lower level due to the lack of effectiveness of implementation when the regulation of interest rates is not significant enough.

The relationship between legal deposit reserve and housing prices

Legal deposit reserve refers to the reserve that an insurance enterprise must deposit in accordance with the relevant laws and regulations of the state. In China, it has been a frequently used and seen as a highly important tool recently, while in many western countries the tool is viewed differently as they are rarely used. In general, higher reserve ratio leads to reduced money supply (Xu and Chen, 2012). In China, there was a negative causality between legal deposit reserve and housing price as the increase in legal deposit reserve caused the money supply down and then led to the drop in housing price. Collyns and Senhadji (2002) found that credit growth had an important impact on housing prices among Asian countries. According to the analysis of Beijing housing price, Hong-ling (2008) stated that the increase in the statutory reserve requirement ratio has reduced the reserves and loanable funds of commercial banks, reducing the ability of banks to issue loans, and housing consumption loans, thereby reducing the effective demand for real estate and leading to a decrease in real estate prices. By using the error correction model (ECM) model, Pi and Wu (2004) showed that there was a linear causal relationship between the real estate market and the bank credit market in China, as well as a bi-directional association in both the long- and short- term mission effect of money supply. However, there are certain deficiencies in this tool. Firstly, to implement the change, there were many steps that need to undergo for decreasing the housing price (Hong-ling, 2008). Also, Hong-ling (2008) analyzed that the result of adjusting legal deposit reserve in a low rate was not obvious for decreasing the housing price because it ultimately depended on the power of real estate supply and demand. The housing price only decreased after its demand went down. Therefore, individual willingness to purchase houses was important as well. Secondly, the legal deposit reserve sometimes may not be an effective monetary tool. It was believed that when the legal deposit reserve was raised to a certain level, its function in liquidity management and monetary control was weakened (Zhang, Ji and Cui, 2008).

Study description and data collection

This study is established on the real estate market in Shenzhen, China. As the prefectural-level city in Guangdong Province and the national economic center, Shenzhen is on the eastern bank of the Pearl River, and is bordered with Hong Kong across the Shenzhen River, making it one of the three largest national financial centers in China. Transforming from a fishing village to one of the most developed cities in China, Shenzhen has made a great effort on its own development and experienced a series of miracle changes, such as the establishment of Special Economic Zones in 1980. As a resident who grew up in Shenzhen, the author would like to make a feasible suggestion for the development of Shenzhen’s housing market. In addition, due to its strong economy and high population density, the demand for houses is also quite high, so the price of houses in my investigation can well demonstrate the effectiveness of the monetary policies approach under the wide range of demands in Shenzhen. Since the economic development in Shenzhen is among the highest in China, the city can provide a better place for research on the housing market, and it serves as examples that other cities can also study and imitate.

To explore how the money supply could influence the housing price in Shenzhen, this paper uses secondary data from the National Bureau of Statistics of China, rather than survey data from questionnaires due to the lack of feasibility. The variables used in this research are divided into two main categories, the money supply, measured in the amount of M2, and the sales price index of newly built commercial properties in Shenzhen, also called the housing index (set the sale price in the base year to 100). In economics, monetary supply is often proxied by M2, which refers to the sum of cash in circulation, outside the entire banking system, owned by enterprises, and by individuals plus the corporate demand deposits, and then add the quasi money which is the time, resident savings, and other deposits. For this research, it is better to use the amount of M2 as it can typically reflect the variations in aggregate social demand and the future state of inflationary pressures. M2 can exclude the effect of other variables for the study and then can be regarded as a target for money supply regulation. For housing index, it is better to choose the sale price index because it can show a more visual representation of the post-regulation price of the property in question, and it also reduces the need to consider the impact of other conditions on the price of the property. The newer data is more relevant to current developments and can provide better reliability for studying excessive housing prices, whereas the older data is often outdated and lacks accuracy.

This paper uses the monthly data from 2011 to 2017, 84 months in total. One reason to consider their trends from 2011 to 2017 is the more stable economic development at that time, which could avoid distractions by other factors. The data used in this research were obtained from the National Bureau of Statistics of China.

Overall, the sales price index ranges from 98.9 to 107.1, with a mean of 100.9 and a standard deviation of 1.7. While for the money supply, measured in the amount of M2, its mean is 11.9 with a deviation of 2.9 ten trillion.

Figure 1 plots the correlation between housing price index and M2 from 2011 to 2017. M2 grows steadily over time, and the index has increased at a steady growth rate and at the same time, it has fluctuated in the study period.

Figure 1. The relationship between M2 and housing price obtained from the National Bureau of Statistics of China (2022).

Figure 2 considers the change of the logarithmic form of the amount of M2 ($\mathrm{\Delta }\mathit{log}\left(M{2}_t\right)$) and the change of housing price ($\mathrm{\Delta }H{I}_t$), illustrating that there is a positive correlation between them. In other words, more money supply is associated with an increase in the housing index.

Figure 2. Relationship between growth rate of lnM2 and change of housing index obtained from the National Bureau of Statistics of China (2022).

Regression model for M2 and housing price

This research uses the open-source software RStudio, which is mainly based on Java, C++, and a little bit of JavaScript programming languages, and its current latest version is 2023.06.0+421. The software was used to input the code and insert the data from the National Bureau of Statistics of China, correlation coefficients, error terms, and intercepts of the estimated equations between M2 and housing price can be obtained. The ordinary least square was used to find the best functional match for the data by minimizing the sum of squares of the errors. Then the linear regression in research can estimate the mathematical correlation between the money supply and the housing price in Shenzhen. As the money supply, M2 shows the similar increasing trend with housing price index from 2011 to 2017, the author proposes the following model to explore the correlation between the change rate of monetary supply and the change of housing index in Shenzhen:

In this model, the change of the housing index in period t is called $\mathrm{\Delta }H{I}_t$, and $u_t$ is the error term. Then ${\mathrm{\beta }}_0$ is the intercept of this estimation equation, while ${\mathrm{\beta }}_1$ indicates the correlation between the change of housing index and change of logM2. If ${\mathrm{\beta }}_1$is expected to be positive, then the increase in M2 for country can increase the housing price; On the other hand, if it is expected to be negative, then the increase in M2 for country can show the decrease in housing price.

However, considering another equation that might discuss the relationship between housing index and monetary supply, this paper also shows another equation: