Printer Friendly

The relationship between Chinese real estate market and stock market.

INTRODUCTION

The real estate industry in China has boomed since 1998 (Fung, Jeng, & Liu, 2010). The real estate market is an important part of China's economy. Real property composes major part of social wealth. Real estate price index is one of national economy vanes. Meanwhile, Chinese stock market develops from a weak efficient market to a semi-strong efficient market, and becomes an important barometer of national economy. Research on interaction mechanism of Chinese real estate market and stock market will help investors choose reasonable assets and establish efficient portfolios, and help Chinese government carry out effective supervision on capital markets. For example, Chinese government should control the amount of "hot money" that flows into stock market in order to avoid stock bubble while squeeze real estate bubble.

This study applies time series analysis to divide Chinese real estate market into three sub-periods based on real estate sales price indexes from January 1999 to November 2009. ADF test, co-integration test, and Granger Causality test results show that the fluctuations of Chinese real estate prices and stock prices have stage correlation, in some sub-periods the real estate market led the stock market. It might provide helpful information for investors to establish effective portfolios and for Chinese government to make relevant policies.

LITERATURE REVIEW

The arguments about the relationship and interaction mechanism of real estate market and stock market are mainly divided into two sides: segmented or integrated. Liu, Hartzell, Greig, & Grissom (1990) researched the relationship between the U.S real estate market and stock market with asset pricing model and concluded that two markets were segmented. However, Ibbotson & Siegel (1984) analyzed the relationship between the U.S real estate prices and S&P 500 stock index and found the existence of negative correlation. Studies of Okunev & Wilson (1997), Okunev, Wilson & Zurbruegg (2001), and Ullah & Zhou (2003) showed the existence of correlation between the U.S real estate market and stock market, and the stock market played a leading role. Quan & Titman (1999) studied relationship between real estate prices and stock prices of 17 countries, and concluded there was significant positive correlation in the long run. Studies of Stone & Ziemba (1993), Liow (2006), and Shen & Lu (2008) separately showed positive correlations between real estate markets and stock markets in Japan, Singapore and China. Hence it is unclear whether real estate market and stock market are segmented or integrated.

METHODOLOGY AND EMPIRICAL RESULTS

Data Description

The monthly data of Chinese Real Estate Sales Price Index (CRPI) is selected to analysis Chinese real estate fluctuating cycles. Data is from China Economic Information Network Statistics Database.

Shanghai Composite Index HCPI and Shenzhen Component Index SCPI are chosen as the indexes to measure prices changes in Chinese stock market. HCPI and SCPI are published by Shanghai Stock Exchange and Shenzhen Stock Exchange.

In order to remove heteroscedasticity and reduce volatility, the indexes are made dimensionless and taken logarithm to get the corresponding new variables which are LCRPI, LHCPI and LSCPI.

Time Series Analysis and Test

The basic principle of time series analysis is that any economic time series can be composed of its first order differential sequence. To increase the symmetry of differential sequence, mean of all differential values are calculated. Then a new time series can be generated with the first order differences and the mean.

First, we define Y(t) as time series of Chinese real estate price index (CRPI). Then the first order differential sequence of Y(t) is generated:

Y[??](t)= Y(t + 1)- Y(t) (t = 1, 2, 3, ...., n) (n=131)

The results of ADF test indicate that Y[??](t) is stationary at the 1 percent level of significance.

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

H is the average value of Y[??](t) covered t - 1 years. Here, H = 0.1323> 0, which indicates the trend of Y(t) is upward.

Y[??](f)= Y[??](t)- H (1[pounds sterling] t [pounds sterling] n - 1)

Then the new differential time series Y[??](t) is generated and shown in figure 1.

[FIGURE 1 OMITTED]

Overall, the fluctuating trend of Chinese real estate market is upward. Combined with major events during the period, Chinese real estate market can be divided into three sub-periods: January 1999 to April 2005, May 2005 to January 2008, and March 2008 to October 2009.

Augment Dickey-Fuller Test (ADF test)

ADF test is applied respectively on time series of LCRPI, LHCPI, and LSCPI and their first order differences DLCRPI, DLHCPI, and DLSCPI in three sub-periods. ADF statistic values are less than critical values at the 1 percent significance level after the first order difference. So the first order difference sequences are stationary.

Cointegration Causal Test

First, we estimate equations with OLS method.

LCRPI = [c.sub.1] + a xLHCPI + [e.sub.1]

LCRPI = [c.sub.2] + b xLSCPI + [e.sub.2]

a, b are parameters, and [e.sub.1], [e.sub.2] are residuals.

In the total period, there are equations as follows:

LCRPI = 4.384371+ 0.033060LHCPI + [e.sub.1]

(61.07283) (3.472573) [R.sup.2] =0.085488 S.E=0.043307 F=12.05876

LCRPI=4.437370+0.023084LSCPI + [e.sub.2]

(75.04605) (3.320789) [R.sup.2] =0.078753 S.E=0.043466 F=11.02764

Then, we test the unit roots of residuals series.

Residuals series are stable. The equations are not spurious regressions. So LCRPI is integrated with LHCPI, and LSCPI.

Granger Causality Tests

n order to obtain the Granger causes of stock market and real estate market, the lags of first to tenth orders are calculated based on the Granger test method of Vector Auto Regression (VAR) model. Results are showed in table 4.

The results show that the fluctuations of Chinese real estate prices and stock prices appear stage correlation. In the period of 1999 to 2007, LCRPI was Granger cause of LHCPI and LSCPI. So the real estate market led the stock market in the sub-period.

CONCLUSION

This study examines the relationship between Chinese real estate market and stock market. It applies time series analysis to divide Chinese real estate market into three sub-periods based on real estate sales price indexes. ADF test, co-integration test, and Granger Causality test results show that Chinese real estate market and stock market are integrated, in some sub-periods the real estate market led the stock market.

This relationship can be explained by many factors and theories. First, many economic factors including national economic development, inflation rates, interest rates and others affect real estate prices and stock prices at the same time, so the changes of real estate prices and stock prices might exist some correlation. Then, real estate and stock are symbols for both wealth and investment tools. The relationship between real estate and stock can be interpreted by wealth effect, crowding-out effect, substitution effect and portfolio theory.

REFERENCES

Fung, H., J. Jeng & Q. Liu (2010). Development of China's real estate market. The Chinese Economy, 43(1), 71-92.

Gao, X. H. & J. Y. Li (2009). Research summary on the relationship between real estate prices and stock prices. Shanghai Real Estate, issue 10, 39-41.

Ibbotson. R. G. & L. B. Siegel (1984). Real estate returns: a comparison with other investments. Real Estate Economics, 12(3), 219-242.

Liow, K. H. (2006). Dynamic relationship between stock and property markets. Applied Financial Economics, 16(5), 371-376.

Liu, C. H., D. J. Hartzell, W. Greig & T. V. Grissom (1990). The integration of the real estate market and the stock market: some preliminary evidence. Journal of Real Estate Finance and Economics, 3(3), 261-282.

Okunev, J. & P. Wilson (1997). Using nonlinear tests to examine integration between real estate and stock markets. Real Estate Economics, 25(3), 487-503.

Okunev, J., P. Wilson & R. Zurbruegg (2001). The causal relationship between real estate and stock markets. Journal of Real Estate Finance and Economics, 18(2), 257-278.

Okunev, J., P. Wilson & R. Zurbruegg (2002). Relationships between Australian real estate and stock market prices: a case of market inefficiency. Journal of Forecasting, 21(3), 181-192.

Quan, D. C. & S. Titman (1999). Do real estate prices and stock prices move together: an international analysis. Real Estate Economics, 27(2), 183-207.

Shen, Y. & W. Lu (2008). Relevance research on Chinese stock prices and real estate price. Modern Economic Science, 30(4), 87-92.

Stone, D. & W. T. Ziemba (1993). Land and stock prices in Japan. Journal of Economic Perspectives, 7(3), 149-165.

Tse, R. Y. C. (2001). Impact of property prices on stock prices in Hong Kong. Review of Pacific Basin Financial Markets and Policies, 4(1), 29-43.

Ullah, A. & Z. G. Zhou (2003) Real estate and stock returns: a multivariate VAREC model. Journal of Forecasting, 21(1), 181-192.

Wilson, P. & J. Okunev (1999). Long-term dependencies and long run non-periodic co-cycles: real estate and stock markets. Journal of Real Estate Research, 18(2), 257-278.

Zhang, Y. & J. Wu (2008). A study of phase correlation on fluctuations between China real estate market and stock market, China Real estate market, issue 1, 29-31.

Zhou, X. (2009). Analysis on Chinese real estate cycle fluctuation. Study & Exploration, issue 3, 134-137.

Xiaohui Gao, Shanghai University of Finance and Economics

Jingyi Li, Shanghai Taxation Bureau

Anthony Yanxiang Gu, State University of New York at Geneseo
Table 1: The Results of ADF Tests On Y(t) and Y'(t)

 ADF Test PROB.
 Statistic

Y (t) -3.156089 0.0983
Y <t) -5.274925 0.0001

Test 1% level -4.033727
 critical 5% level -3.446464
 values 10% level -3.148223

Table 2: The Results of ADF Tests on LCRPI, LHCPI,
LSCPI and their First Order Differences
DCRPI, DHCPI, DSCPI

TIME VARIABLE ADF CRITICAL
 STATISTIC VALUE (1%)

1/1999- LCRPI -3.173166 -4.033727
11/2009 LHCPI -1.99034 -4.031309
 LSCPI -1.698804 -4.031309
 DLCRPI -5.475863 -4.038365
 DLHCPI -5.967869 -4.031309
 DLSCPI -5.947102 -4.031309

1/1999- LCRPI -1.731925 -3.525618
4/2005 LHCPI -1.773146 -3.520307
 LSCPI -2.051703 -3.520307
 DLCRPI -4.109851 -3.525618
 DLHCPI -7.944620 -3.521579
 DLSCPI -8.793404 -3.521579

5/2005- LCRPI -0.064007 -2.639210
1/2008 LHCPI 2.670071 -2.639210
 LSCPI 3.100284 -2.639210
 DLCRPI -3.486200 -2.644302
 DLHCPI -3.783803 -2.641672
 DLSCPI -3.611474 -2.641672

2/2008- LCRPI 0.717293 -2.679735
11/2009 LHCPI -0.564451 -2.679735
 LSCPI -0.291309 -2.679735
 DLCRPI -4.010954 -2.685718
 DLHCPI -4.813476 -2.685718
 DLSCPI -4.209537 -2.685718

TIME VARIABLE TEST RESULT
 EQUATION *

1/1999- LCRPI (C, T) Not stationary
11/2009 LHCPI Not stationary
 LSCPI Not stationary
 DLCRPI Stationary
 DLHCPI Stationary
 DLSCPI Stationary

1/1999- LCRPI (C, N) Not stationary
4/2005 LHCPI Not stationary
 LSCPI Not stationary
 DLCRPI Stationary
 DLHCPI Stationary
 DLSCPI Stationary

5/2005- LCRPI (N, N) Not stationary
1/2008 LHCPI Stationary
 LSCPI Stationary
 DLCRPI Stationary
 DLHCPI Stationary
 DLSCPI Stationary

2/2008- LCRPI (N, N) Not stationary
11/2009 LHCPI Not stationary
 LSCPI Not stationary
 DLCRPI Stationary
 DLHCPI Stationary
 DLSCPI Stationary

* C, T and N denote separately test equations including
intercept, trend and neither of them.

Table 3: ADF Tests Results of Residuals[R]

TIME RESIDUAL ADF CRITICAL
 STATISTIC VALUE

1/1999- [[epsilon].sub.1] -2.702823 -2.583744
11/2009 [e.sub.2] -2.771089 -2.583744

1/1999- [[epsilon].sub.1] -1.976559 -1.945456 **
4/2005 [e.sub.2] -3.175017 -2.597476 *

5/2005- [[epsilon].sub.1] -4.834757 -2.644302 *
1/2008 [e.sub.2] -3.279594 -2.644302 *

2/2008- [[epsilon].sub.1] -4.511116 -2.685718 *
11/2009 [e.sub.2] -4.128769 -2.685718 *

TIME RESIDUAL RESULT

1/1999- [[epsilon].sub.1] Stationary
11/2009 [e.sub.2] Stationary

1/1999- [[epsilon].sub.1] Stationary
4/2005 [e.sub.2] Stationary

5/2005- [[epsilon].sub.1] Stationary
1/2008 [e.sub.2] Stationary

2/2008- [[epsilon].sub.1] Stationary
11/2009 [e.sub.2] Stationary

(@) Test equations do not include intercept and trend.

* , ** represent separately the critical values at
the 1 percent and 5 percent level of significance.

Table 4: The Results of Pairwise Granger Causality Tests

TIME NULL HYPOTHESIS LAG F-STATISTIC

1/1999- LCRPI does not 1 0.97012
11/2009 Granger Cause LHCPI

 LHCPI does not 1 1.67593
 Granger Cause LCRPI

 LCRPI does not 1 0.77620
 Granger Cause LSCPI

 LSCPI does not 1 2.26086
 Granger Cause LCRPI

1/1999- LCRPI does not 1 7.79510
4/2005 Granger Cause LHCPI

 LHCPI does not 1 0.92441
 Granger Cause LCRPI

 LCRPI does not 1 5.27988
 Granger Cause LSCPI

 LSCPI does not 1 0.91306
 Granger Cause LCRPI

5/2005 - LCRPI does not 2 22.4719
 Granger Cause LHCPI

1/2008 LHCPI does not 2 0.32282
 Granger Cause LCRPI

 LCRPI does not 2 22.1897
 Granger Cause LSCPI

 LSCPI does not 2 0.23536
 Granger Cause LCRPI

2/2008- LCRPI does not 1 0.98016
11/2009 Granger Cause LHCPI

 LHCPI does not 1 0.00785
 Granger Cause LCRPI

 LCRPI does not 1 0.96507
 Granger Cause LSCPI

 LSCPI does not 1 0.06231
 Granger Cause LCRPI

TIME NULL HYPOTHESIS PROB. RESULT

1/1999- LCRPI does not 0.32652 Not refusal
11/2009 Granger Cause LHCPI

 LHCPI does not 0.19782 Not refusal
 Granger Cause LCRPI

 LCRPI does not 0.37997 Not refusal
 Granger Cause LSCPI

 LSCPI does not 0.13516 Not refusal
 Granger Cause LCRPI

1/1999- LCRPI does not 0.00673 Refusal *
4/2005 Granger Cause LHCPI

 LHCPI does not 0.33958 Not refusal
 Granger Cause LCRPI

 LCRPI does not 0.02452 Refusal **
 Granger Cause LSCPI

 LSCPI does not 0.34288 Not refusal
 Granger Cause LCRPI

5/2005 - LCRPI does not 0.00001 Refusal *
 Granger Cause LHCPI

1/2008 LHCPI does not 0.72607 Not refusal
 Granger Cause LCRPI

 LCRPI does not 0.00001 Refusal *
 Granger Cause LSCPI

 LSCPI does not 0.63052 Not refusal
 Granger Cause LCRPI

2/2008- LCRPI does not 0.32797 Not refusal
11/2009 Granger Cause LHCPI

 LHCPI does not 0.92984 Not refusal
 Granger Cause LCRPI

 LCRPI does not 0.33167 Not refusal
 Granger Cause LSCPI

 LSCPI does not 0.80412 Not refusal
 Granger Cause LCRPI

*, ** represent separately the critical values at the 1 percent
and 5 percent level of significance.
COPYRIGHT 2012 The DreamCatchers Group, LLC
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2012 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Gao, Xiaohui; Li, Jingyi; Gu, Anthony Yanxiang
Publication:Journal of International Business Research
Geographic Code:9CHIN
Date:Jan 1, 2012
Words:2363
Previous Article:Managerial accounting as a tool for corporate strategy: synergy creation and anergy inhibition.
Next Article:Investment and trade between West Africa and China: the Senegalese case.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters