THE EFFECTS OF MACROPRUDENTIAL POLICY ON HOUSING MARKET: EVIDENCE FROM 30 PROVINCES IN CHINA

Transcription

1 THE EFFECTS OF MACROPRUDENTIAL POLICY ON HOUSING MARKET: EVIDENCE FROM 30 PROVINCES IN CHINA LINA WANG Master Student of Faculty of Economics, Chulalongkron University, Bangkok, Thailand Abstract- Thehousing market plays a significant role in Chinese economic growth, especially housing-related consumption, investment and tax revenue. However, a house price (boom or bust) will affect the stability and the safety of the financial system. In order to manage the potential risks from the housing market, macroprudential policy has been popularly used in economies all over the world in recent years as both a regulatory and supervisory instrument. Additionally, China also implemented macroprudential policy to maintain the soundness of the financial system and the housing market. The purpose of this research is to investigate whether the macroprudential policy enacted by the government is effective in managing housing prices in 30 provinces within China since 2006 via the panel fixed-effects model. The results showed that most of macroprudential instruments are effective. This can help control and contain increases in house prices in both newly constructed housing markets, as well as second-hand housing markets. It also found that the reserve requirement ratio is the most effective instrument. Index Terms- China, Housing market, Macroprudential policy, Policyeffectiveness I. INTRODUCTION The housing market is an essential sector related to a large share of investments, loans and the wealth of the national economy. Since 1978, under Deng Xiaoping s lead, the People s Republic of China embraced the first real estate reform. Till now, housing markets played a significant role in Chinese economic growth; especially housing-related consumption, investment and tax revenue. In China, real estate investments, loans for developers, mortgages of households and house prices increased continually in recent years. The data showed in 2013, China s real estate investment reached up to 15.1 percent of GDP, more than 1 in 3 of total investment, in comparison to only 4.9 percent 13 years ago. As regards to real estate loans; it rose to nearly 30 percent of GDP and more than 20 percent of total loans. Until 2013, housing prices have increased by almost 4 times their values of ten years prior in some large cities. House mortgage loans went up nearly six-fold in the last 10 years. However, when looking back at the last century until now, housing market has been the source of crises [1].House prices, boom or bust, form a crucial potential risk in the financial sector, which will have serious consequences for the stability and the safety of the financial system. The experience showed that if a house price booms, it requires financial institutions (such as banks) to keep enough or much more reserve requirement, capital adequacy and liquidity to provide more loans and mortgages. However, if a house price goes bust, those institutions face a great loss from the loss value of the house and big numbers of non-performing loans. As a consequence, house price values become a potential housing market risk and might lead to instability of the financial system. In general, housing markets have certain characteristics that intrinsically link them to financial stability [2], as well as the national economy. Not only China, but various countries all over the world are facing booming asset prices. In order to regulate and supervise the whole systematic risk of the financial system, macroprudential policy is widely used. Macroprudential is defined as the safety and soundness of the broad financial system and payments mechanism [27]. With frequent financial crisis, the importance of macroprudential regulations have been emphasized by many countries. Macroprudential regulation even has an objective that sometimes is assigned to avoid asset price bubbles directly [9]. As for China, the People s Bank of China and the China Banking Regulatory Commission adopted macroprudential polices to sustain the housing market risk and the housing price booms in main cities from 2009 on. And different instruments have been used until now. In this research, the objective is to evaluate the effects of individual macroprudential policies on house price management in 30 provinces of China, as well as to point out the differences between newly constructed houses and second-hand houses. II. LITERATURE REVIEW A. Asset Price Boom An asset price boom is defined a positive deviation of an aggregate asset price indicator from its recursively estimated trend by at least 10 percent [11]. The features of an asset price boom are that equity and real estate prices rise strongly during the period of boom, but crash after that. Real GDP growth is quickly and constant during the boom, but is followed by a lower growth. Following the definition of Alessi and Detken [10], a "costly" asset-price boom is usually 13

2 followed by a three-year period that the real GDP growth will lower than potential growth at least 3 percent. The cause of this result is that, during the boom period, real GDP growth is mainly driven by housing-related investments and loans with looser monetary policy. However, not all asset price booms lead to a bust and not all busts lead to a financial crisis [11]. It still will be very dangerous that if government do nothing but watch it boom. B. Macroprudential Policy Macroprudential regulations that can address systematic risk in both time dimension and cross section dimension exist in financial system [17], [22]. Macroprudential regulations even have an objective whereby sometimes macroprudential policies are assigned so as to avoid asset price bubbles [9]. It seems that macroprudential instruments are used more in managing the housing market than the stock market. With different type of risks applying to different macroprudential instruments, the instruments used include countercyclical capital buffers, time-varying loan-to-value (LTV) ratios, systemic capital surcharges, liquidity surcharges, etc.because of the difference of national conditions, according to disparate objectives, national government or policy maker need to select instrument combinations suitable. Housing-related measures, credit measures and reserve requirement, for example, are very different used between Asia, Latin America, Middle East and Africa, Advanced Europe and North America. Among all of regions, the greatest number of macroprudential instruments are used in Asia, especially utilizing housing related measures. Since that Asia has more developing countries than other continents. In the century of financial crisis frequent burst, real estate sector will lead national economy, and will be the necessity goods of people as well. Then it seems to be very serious to control and manage real estate market. In general, for Asian region, loan-to-value ratio is the major housing related measures and reserve requirement is major credit measures and reserve requirements. In the regular meeting of monetary policy committeeof China s central bank, held at the fourth quarter of 2009, explicit stated that: Macroprudential regulation should be established to prevent and reduce potential financial risks effectively. Under these conditions, macroprudential regulation had been formally incorporated into the central bank's agenda.and in the real estate market, the two main instruments China tends to focus on in macroprudential policies are Reserve Requirement Ratio and Loan-to-value Ratio, which also serve as the two most commonly used indicators in housing related measures and credit related measures in Asia. C. Related Research Most of the empirical studies are trying to identify the effectiveness of macroprudential policies in reducing systemic risk and the relevant macroeconomic variables in different regions in the same period, such as Lim et al., [25], Vandenbussche et al., [28]. Lim et al., [25] use the fixed-effect dynamic panel regression with GMM estimator and data covering 49 countries during The analysis shows that many instruments are effective in reducing systemic risk. Furthermore, Asian economies appear to use macroprudential policies extensively much more than the other regions, especially housing-related measures are most effective in sustaining the housing price growth and housing market risk. Zhang and Zoli [18]investigate how macroprudential policy and capital flow measures affect the relevant macroeconomic variables (i.e., real credit growth, real housing prices growth, and capital inflows). The data used in fixed-effect dynamic panel regressions with GMM covers 46 countries all over the world from 2000Q1 to 2013Q1. The results show that macroprudential policy tools and capital flow measures can help curb housing price inflation, equity flows and credit growth, particularly the LTV ratio and housing tax which are the housing-related measures. However, the empirical analysis in China are still very limited.because that most of current researches on macroprudential policy in China are only theoretical analysis. Empirical analysis like Wang and Sun [13] examine the effectiveness of macroprudential policies in controlling the loan growth and housing prices by panel fixed-effect models, using a sample of 171 banks and 31 provinces during The results show that the RRR and housing-related policies are useful. But the policies adopted now cannot guarantee curb the current systemic risk. Then this research gives a suggestion that policy maker or governor should make targeted macroprudential policies according to different regions and levels. As for other related researches, there are many papers research the housing bubble appearing in China. Two approaches can be broadly used. First approach investigates the existence of bubble by using indicators, such as price-to-income ratio, housing vacancy rate, growth rate of housing price, growth rate of GDP etc.(hou and Feng [31],Lv[32], Hou[6]). The second focuses on modeling the fundamental price to compare gap between real market price and fundamental price with a under and upper control line (Noguchi [29], Gao[30], Ahuja et al. [26]). III. RESEARCH METHOD This research needed to identify whether the major individual macroprudential policy is effective in managing house prices in 30 provinces of China. Following the literature review, in order to investigate how these instruments affected the relevant macroeconomic variables and reduced systemic risk, Lim et al., (2011) and Zhang and Zoli (2014) use fixed-effect dynamic panel regression. Zhang and Zoli 14

3 (2014) find that macroprudential policy tools can help curb housing price inflation, particularly the LTV ratio and housing tax, which are the housing-related measures. Wang and Sun (2013) use panel fixed-effect models, finding that the RRR and housing-related policies are useful. D. Regression This research used the panel fixed-effect model. There are 2 parts of regression. First, estimating the regression with the newly constructed house price index as a dependent variable and using a different individual policy instrument for each regression. Second, estimating the regression with a second-hand house price index as a dependent variable and using a different individual policy instrument for each time. E. Data The resource of data obtained in this research came from the National Bureau of Statistics of China, PBOC, CEIC and Bloomberg. The data is quarterly: from January 2006, to December According to the literature, the macroprudential instruments will be reserve requirement ratio (RRR), loan-to-value ratio (LTV), capital adequacy ratio (CAR) and liquidity ratio. In addition, the newly constructed house price index and the second-hand house price index of each province are dependent variables. The provincial GDP, provincial CPI and provincial fixed Asset Investment Index are also used as macroeconomic variables. F. Statistic Model The model of Wang and Sun [15] is used to evaluate the effectiveness of individual macroprudential instrument. The fixed effect panel model can be defined as: House Price Index, = β + β MPolicy, + β House Price Index, +β GDP, + β In lation, + β Investment, + ε, For each province i, at time t: House Price Index, is matrix of sale price indexes of newly constructed houses and sale price indexes of second-hand houses. MPolicy, isa matrix of macroprudential policy variables, which contains the time-series value of one of the macroprudential instruments, (e.g. the LTV ratio or Reserve Requirement Ratio) or a dummy variable that is given the value of 1 when the policy is used in the same period. Otherwise, it would be zero. GDP,, In lation, and Investment, represent macroeconomic control variables. Following the literature, they are provincial GDP, provincial CPI and the provincial fixed-asset investment index, respectively. IV. RESULTS This research aimed to evaluate whether macroprudential policy enacted by the government can have an impact on house prices in terms of 15 helping manage house prices, as well as strengthen the stability of the financial system. G. The results for the impact of each individual macroprudential policy instrument on house price growth (overall) of 30 provinces. 1 The results showed that most individual macroprudential policy instruments are significant in managing both newly constructed, as well as second-hand house price growth (overall); especially the RRR of large and small banks, LTV ratio of the first house (housing space less than 90 square meters), CAR and the liquidity ratio. 1) Reserve requirement ratio (RRR) The reserve requirement ratio of both large and small banks has a negative relationship with newly constructed, as well as second-hand house price growth (overall). The rise of RRR represents that banks will keep more capital and decrease releasing loans. Then consumers don t have enough money to buy houses which makes the house prices drop. However, RRR is more effective in containing newly constructed house prices versus the second-hand house prices (overall), due to increased significance in newly constructed house prices. 2) Loan-to-value ratio (LTV) There are three kinds of loan-to-value ratios that meet different requests in lending mortgage loans. These are LTV for the first house with housing space totaling less than 90 square meters, the first house with housing space totaling more than 90 square meters and the second house. Theoretically, all kinds of loan-to-value ratios should be positive values. Regardless of house space, size of a first house or a second house; when LTV rises, banks will release more loans for mortgages. Consumers pay less than before and demand increases. At the same time, prices increase. The results showed that only the LTV of the first house with a housing space totaling less than 90 square meters is positively associated with newly constructed house price growth and the second-hand house price growth (overall). LTV is more effective in managing newly constructed house prices than the second-hand house prices (overall), due to a more significant newly constructed house price. 3) Capital adequacy ratio (CAR) Capital adequacy ratio has a strongly negative relationship with both newly constructed, as well as second-hand house price growth (overall), representing that when CAR rises up, banks need to keep more capital and release less loans. If it is more difficult to lend loans from financial institutions for people who are willing to buy a house, then demand of housing will decrease. Thus, price decreases. 4) Liquidity ratio Liquidity ratio has a strongly negative association with both newly constructed as well as the second-hand house price growth, (overall) too. When 1 See Table.1 and Table. 2.

4 capital has much more liquidity, the economy runs well and house prices should decrease. Table 1. Effectiveness of Macroprudential Policy Instruments in managing newly constructed house price growth (overall) Note: The table presents panel fixed-effects regressions. ***, **, * indicate statistical significance at 1%, 5%, and 10%, respectively. Table 2.Effectiveness of Macroprudential Policy Instruments in managing second-hand house price growth (overall) Note: The table presents panel fixed-effects regressions. ***, **, * indicate statistical significance at 1%, 5%, and 10%, respectively. 16

5 Table 3. Effectiveness of Macroprudential Policy Instruments in managing newly constructed house price growth (divided by housing space) Note: The table presents panel fixed-effects regressions. ***, **, * indicate statistical significance at 1%, 5%, and 10%, respectively. Table 4. Effectiveness of Macroprudential Policy Instruments in managing second-hand house price growth (divided by housing space) Note: The table presents panel fixed-effects regressions. ***, **, * indicate statistical significance at 1%, 5%, and 10%, respectively H. The results for the impact of reserve requirement ratio as a macroprudential policy instrument on house price growth (classified with housing space, which will be housing space totaling less than 90 square meters, between 90 to 144 square meters, as well as more than 144 square meters) of 30 provinces. 2 From previous results, it can obviously be found that the reserve requirement ratio (RRR) is the most effective macroprudential policy instrument, versus others. At this subsection, the main purpose is to investigate the performance of RRR in house price growth, which is divided into 3 different space types. The results show that only part of reserve 2 See Table. 3 and Table. 4 requirement ratio is significant in managing both newly constructed and second-hand house price growth (classified with 3 types of housing space). It indicated that the reserve requirement ratio of both large banks and small banks have a strongly negative relationship with newly constructed, as well as second-hand house price growth, which is classified with 3 types of housing spaces. Especially, according to the coefficient, reserve requirement ratio can help restrain newly constructed house price growth much more than second-hand house price growth in each type level. As for a newly constructed house, the reserve requirement ratio can help reduce the house price much more, when the housing space is between 90 and 144 square meters. In contrast, it can help 17

6 reduce those second-hand house prices much more, when housing space is less than 90 square meters. CONCLUSION During 2006 to 2014, those key instruments of macroprudential policy were significant in managing both the newly constructed housing market, as well as the second-hand market. RRR and LTV ratio, as the important housing-related instruments, can help control and contain increases of house prices in different levels. Especially, RRR was the most significant and effective instrument. Sometimes, some of the LTV s do not have impact. This might come from a lack of data, or when the adjustment of ratio is not frequent enough. Another important finding is that there are some different impacts between newly constructed house price growth and second-hand house price growth. It showed that the average impact on newly constructed house price growth is more significant than on second-hand house price growth. And most policy instruments can reduce newly constructed house price growth significantly as well. The reason might be that these two kinds of housing have different shares in the market and people might prefer to purchase newly constructed houses. SUGGESTION These key instruments should be routinely used. But sometimes, if the function instruments are too similar to distinguish, then it s hard to estimate the impact at the same time. And there are still many other factors making the policy useful in managing house prices. Additionally, China is not a completely open economy yet. The impact of government control could be much more obvious. Furthermore, it would be better if policy is made considering the differences of each province, and then customize implementation to each region. SUGGESTION FOR FUTURE RESEARCH The vast land of China leads to a large difference between regions. The indicators representing the regional differences are inadequate due to insufficient open data, as well as difficulty in data collection.in addition, the implementation of macroprudential policies has just started in China. Most researches are still relying on theoretical analysis. Yet very limited quantitative analysis and empirical analysis exists. Hopefully, there can be much more empirical macroprudential-related analysis in the case of China. REFERENCES [1] Zhu, M. (2014). Housing Markets, Financial Stability and the Economy. Remarks at the Bundesbank/German Research Foundation/IMF Conference, June. Retrieved from [2] Praet, P. (2011). Housing cycles and financial stability the role of the policymaker. Speech at the EMF Annual Conference, Brussels, November. Retrieved from 4.en.html [3] Deng, Y., Zheng, D., & Ling, C. (2005). An early assessment of residential mortgage performance in China. In Journal of Real Estate Finance and Economics (Vol. 31, pp ). doi: /s [4] Stability, H. F., & Basics, B. T. O. (2011). Housing finance and financial stability back to basics? International Monetary Fund, (April), [5] Galati, G., &Moessner, R. (2012). Macroprudential Policy - a Literature Review. Journal of Economic Surveys, 27(5), doi: /j x [6] Hou, Y. (2010). Housing price bubbles in Beijing and Shanghai?:A multi-indicator analysis. International Journal of Housing Markets and Analysis, 3(1), doi: / [7] HKMA (Hong_Kong_Monetary_Authority). (2010). Loan-to-value ratio as a macroprudential tool Hong Kong SAR s experience and cross-country evidence. Retrieved from [8] Kuttner, K. N., & Shim, I. (2013). Can non-interest rate policies stabilise housing markets? Evidence from a panel of 57 economies (p. 41). [9] Landau, J.P. (2009). Bubbles and macro-prudential supervision. Remarks at the Joint conference on the "Future of Financial Regulation", Banque de France and Toulouse School of Economics, Paris, 28 January [10] Alessi, L., &Detken, C. (2009). Real Time Early Warning Indicators for Costly Asset Price Boom/Bust Cycles: A Role for Global Liquidity. ECB Working Paper No [11] Detken, C., &Smets, F. (2004). Asset price booms and monetary policy. In H. Sibert (ed.) Macroeconomic Policies in the World Economy. Berlin: Springer. [12] Liu, L. (2014). Is China s Property Market Heading toward Collapse? Peterson Institute for [13] International Economics, November. Retrieved from [14] Policy, M., & Group, A. (2007). China Monetary Policy Report Quarter Two, Policy, [15] Wang, B., & Sun, T. (2013). How Effective are Macroprudential Policies in China? IMF Working Papers, 13(75). [16] Borio, C. (2009). Implementing the Macroprudential Approach to Financial Regulation and Supervision. Banque de France Financial Stability Review, Vol. 13. [17] Bank of England. (2011). Instruments of Macroprudential Policies. Discussion Paper. [18] Zhang, L., &Zoli, E. (2014). Leaning Against the Wind: Macroprudential Policy in Asia. IMF Working Papers, 14(22). doi: / [19] Clement, P. (2010). The term macroprudential : origins and evolution 1. BIS Quarterly Review, (March), [20] Brockmeijer, J., Moretti, M., Osinski, J., & Blancher, N. (2011). Macroprudential policy: An organizing framework. International Monetary Fund. [21] Weistroffer, C., (2012). Macroprudential supervision. Deutsche Bank Research, Current Issues. May 24, [22] Borgy, V., Clerc, L., &Renne, J. (2009). Asset-price boom-bust cycles and credit: what is the scope of macroprudentialregulation? Working Pa per 263, Banque de France. [23] Kaufman, G. (2000). Banking and currency crises and systemic risk: Lessons from recent events. Federal Reserve Bank of Chicago Economic Perspectives, [24] Wooldridge, J. M. (2001). Applications of Generalized Method of Moments Estimation. Journal of Economic Perspectives, 15(4), doi: /jep [25] Lim, C., Columba, F., Costa, A., Kongsamut, P., Otani, A., Saiyid, M.,Wezel, T.,&Wu, X. (2011). Macroprudential Policy : What Instruments and How to Use Them? Lessons from Country Experiences. Group, 11(238),

7 [26] Ahuja, A., Cheung, L., Han, G., Porter, N., & Zhang, W. (2010). Are House Price Rising too Fast in China? IMF Working Papers (Vol. 10). [27] Bank for International Settlements. (1986). Recent innovations in international banking. Report prepared by a study group established by the central banks of the G10 countries (Cross Report), Basel, April. [28] Vandenbussche, J., Vogel, U., &Detragiache, E. (2012). Macroprudential Policies and Housing Prices A New Database and Empirical Evidence for Central, Eastern and Southeastern Europe. Working Paper 12/303. International Monetary Fund. [29] Noguchi, Y. (1989). The Bubble and Economic Policies in the 1980s.[J] Japan: Journal of Japanese Studies, Vol. 20, No.2(Summer,1994), [30] Gao, B., Wang,HL.,& Li, WJ. (2012). Real Estate Bubbles in 30 Large and Medium- sized Cities in China: Based on the Model of Expected Equilibrium Price. Nanjing: Industrial Economics Department, Nanjing University. [31] Hou, YS.,&Feng, CC. (2009). An Empirical Research on the Measurements of the Real Estate Bubbles in Chinese First-tier Cities. Beijing: Peking University, the Centre of Real Estate Studies and Appraisals and Peking University, College of Urban and Environmental Sciences. [32] Lv, JL. (2010). The Measurement of the Bubble of Urban Housing Market in China. Economic Research Journal, (06)