Chapter 3 Research Method and Data Information
3.3 T-REITs Index presentation and data analysis
國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
30
3.3 T-REITs Index presentation and data analysis
Most of the researchers use exponential pattern while studying REITs, however, T-REITs index has not yet been officially established in Taiwan. In order to better understand the overall performance of T-REITs and to support the follow-up researches, the T-REITs index should be established. This section reviews the REITs index of United States and Japan, and applies the establishment method proposed by Zheng and Chang (2007) to generate a longer T-REITs index. Further, we intend to outline the basic information on T-REITs and illustrate the past performance and trend of T-REITs.
3.3.1 U.S. REITs Index
The base period of REITs index sets up by National Association of Real Estate Investment Trust (NAREIT) is on December 31, 1971, and the base value is set to be 100. The NAREIT includes all the REITs from New York Stock Exchange (NYSE), the NASDAQ Stock Market and American Stock Exchange (AMEX). In addition to a composite REITs index, NAREIT also provides index of equity REITs, mortgage REITs and hybrid REITs. The methodology is as:
（ . ）
∑ , , ,
∑ , , （ . ） where n=the number of companies; = price of the current day; , =stock price of the previous day; , = stock numbers of the previous day; ∑ ,
, = total market value.
3.3.2 Japan’s REITs Index
There are two prestigious REITs index in Japan, one is the STBRI J-REITs index established by Sumitomo Trust & Banking Co., which is the same method as the U.S..
The other one is the Topix REITs established by Tokyo Stock Exchange, the same one that established TOPIX, which is from March 31, 2003 and set a value of 1000. The methodology is as:
‧
numbers of the current day; , =stock price of the base period; , = stock numbers of the base period.3.3.3 Establish T-REITs index
This study is based on the establishment of T-REITs index method proposed by Zheng and Chang (2007), using Japan's Topix REITs index mode. The T-REITs index contents all the REITs listed on OTC (Over-the-counter) since the launched date, including Fubon No.1, Cathay No.1, Shin Kong No.1, Fubon No.2, Trident, Kee Tai Star, Cathay No.2 and Gallop No.1.
The method of establishing T-REITs index by Zheng and Chang is a circulation-weighted price index, using each REITs’ number of shares as weighted values. REITs index for the day of the date is the total issue value divided by the base value, and the base value is market value of total issued during base period. Total market value is the sum of the daily market price (closing price in this study) multiplied by the number of shares issued. In this study, the launched date of Fubon No.1, March 10, 2005, is the base period, and the base value is 100. The methodology is:
∑ , ,
∑ , , （ . ） where n=the number of companies; , = price of the current day; , = stock numbers of the current day; , =stock price of the base period; , = stock numbers of the base period. Because this study uses monthly data as research units, we will first establish the daily T-REITs index then calculate the average of each month's index as the study used data on T-REITs.
3.3.4 Basic Information of T-REITs
Taiwan has a total of eight REITs listing. Since the time of this study ranges from March 2005 to December 2008, all the eight REITs are the subjects under study. The basic information of all T-REITs are shown in Table 3-2, the basic information of
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
32
T-REITs are obtained from the files of the prospectus, and the index are established by using the closing price index from Taiwan Economic Journal (TEJ) database.
We can learn from the basic information of T-REITs that the highest percentage of current total market capitalization is Cathay No.1; Shin Kong No.1 is second place, and the proportion of total market capitalization of Fubon No.1, Fubon No.2 and Cathay No.2 are close. The least percentage of the total market value is Kee Tai Star, with only 4.19%, which may be related to the relatively small number of combinations on the content.
Table 3-2 Basic Information of T-REITs
T-REITs Listing Date Market value (billion)
Percentage of total market capitalization
Fubon No.1 2005. 03. 10 80.9 13.85%
Cathay No.1 2005. 10. 03 139.3 23.85%
Shin Kong No.1 2005. 12. 26 113 19.35%
Fubon No.2 2006. 04. 13 73.02 12.50%
Trident 2006. 06. 26 38.52 6.59%
Kee Tai Star 2006. 08. 14 24.5 4.19%
Cathay No.2 2006. 10. 13 72 12.33%
Gallop No.1 2007. 05. 15 42.84 7.33%
Total 584.08 100%
3.3.5 Variables Descriptive Statistics
The descriptive statistics of the variables are in Table 3-3, including the mean, standard deviation, maximum, minimum, and J-B values. The commercial rent is calculated in the average price per ping, so the mean value of the price is 53.622. We can learn from the Jarque-Bera value that all the variables are at normal distribution because the statistics can reject the null hypothesis of not being under normal distribution at 1% and 10% significant level.
‧
Table 3-3 The Variables Descriptive Statistics Variable T-REITs Stock Price
Index
Minimum 82.300 4496.383 -1.230 2.261 1624 36.340
Jarque-Bera 0.876 *** 0.658 *** 0.646 *** 5.467 * 5.383 * 1.055 ***
Observations 46 46 46 46 46 46
Note 1: *, **, and *** represent significant level of 10% 、5%、 1% respectively
3.3.6 Performance of T-REITs
Figure 3-1 Trend of T-REITs Index（Daily）
We can learn from figure 3-1 that before July 2006, T-REITs show a steady growth trend with slight fluctuations, which was during the intensive distribution time for the former 6 REITs. The main reason for the decline in September and October 2006 was the poor performance of Trident and Kee Tai Star, thereby affecting the overall performance of T-REITs. Thereafter, the price index rose sharply until May 2007, and the second half of 2007 was the downturn phase of the performance of T-REITs, mainly due to the U.S. subprime mortgage crisis. January 2008 began another wave of growth. In addition, the overall performance of T-REITs will reach the pinnacle around March each year. We can also learn the figure that the obvious life cycle occurs after 2007.
70
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
34
3.3.6 Trends of T-REITs, Macroeconomy and Commercial Real Estate Market
This section uses monthly data to show the historical trend between T-REITs and other variables, and conducts preliminary study about whether there is long-run equilibrium relationship existed.Figure 3-2 Trend of T-REITs and Stock Price Index
Figure 3-2 indicates the past trends of T-REITs and the stock price index. It can be seen from the figure that T-REITs and stock price have the same volatility over time, and that T-REITs could have a positive correlation associated with stock price.
Besides the widening gap between the two lines and slightly different fluctuations from March to September 2007, the long-run trend between the two variables is still similar.
0.00 2,000.00 4,000.00 6,000.00 8,000.00 10,000.00 12,000.00
0.00 20.00 40.00 60.00 80.00 100.00 120.00 140.00
2005/3 2005/5 2005/7 2005/9 2005/11 2006/1 2006/3 2006/5 2006/7 2006/9 2006/11 2007/1 2007/3 2007/5 2007/7 2007/9 2007/11 2008/1 2008/3 2008/5 2008/7 2008/9 2008/11
Reits Stock
‧
Figure 3-3 Trend of T-REITs and Interest Rate
Figure 3-3 shows the T-REITs and interest rate volatility was similar before May 2007, and the fluctuations were small. There was a severe and significant inverse relationship between T-REITs and interest rate by May 2007, but after that, the direction of correlation vary with time. For example, T-REITs and interest rate were negatively correlated between September and December 2007, while the correlation became positive from July through November 2008. Therefore, it is unable to determine whether the long-run relationship exists between T-REITs and interest rate, and their correlation direction.
Figure 3-4 Trend of T-REITs and Inflation Rate
From figure 3-4, we can see there may not be a consistent relationship between
0.000
2005/3 2005/5 2005/7 2005/9 2005/11 2006/1 2006/3 2006/5 2006/7 2006/9 2006/11 2007/1 2007/3 2007/5 2007/7 2007/9 2007/11 2008/1 2008/3 2008/5 2008/7 2008/9 2008/11
Reits Int
2005/3 2005/5 2005/7 2005/9 2005/11 2006/1 2006/3 2006/5 2006/7 2006/9 2006/11 2007/1 2007/3 2007/5 2007/7 2007/9 2007/11 2008/1 2008/3 2008/5 2008/7 2008/9 2008/11
Reits Inf
‧
T-REITs and inflation, and the inflation volatility is more dramatic and obvious relative to the volatility of T-REITs, which is more gentle. Therefore, there may not be a long-run equilibrium relationship between T-REITs and inflation rate.
Figure 3-5 Trend of T-REITs and Commercial Rent
Figure 3-5 indicates that the rent shows stable growth until May 2008, when there was a slight decline, and its fluctuations are more similar to the T-REITs trends before 2008, but is not obvious after 2008. The possible reason is that there have not been significant changes in the commercial rent in the past few years, but T-REITs are daily trading so that the price volatility is more obvious. Therefore, we can’t determine whether the two variables have a clear long-term relationship.
Figure 3-6 Trend of T-REITs and Commercial Price
It can be seen from figure 3-6 that the volatility of commercial price is greater than T-REITs. It may be the transaction price data itself that causes the larger degree
1500
2005/3 2005/5 2005/7 2005/9 2005/11 2006/1 2006/3 2006/5 2006/7 2006/9 2006/11 2007/1 2007/3 2007/5 2007/7 2007/9 2007/11 2008/1 2008/3 2008/5 2008/7 2008/9 2008/11
Reits Rent
2005/3 2005/5 2005/7 2005/9 2005/11 2006/1 2006/3 2006/5 2006/7 2006/9 2006/11 2007/1 2007/3 2007/5 2007/7 2007/9 2007/11 2008/1 2008/3 2008/5 2008/7 2008/9 2008/11Reits Cprice
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
37
of price volatility. However, the trend between T-REITs and commercial price is still similar, and thus they are likely to have long-run relationship.
Summarizing the results from Figure 3-2 to 3-6, the preliminary view of this study shows that T-REITs are more likely to have a long-run equilibrium relationship with stock price index and with commercial price, and are less likely to have a long-run relationship with inflation rate. We could not explicitly determine whether T-REITs have long-term equilibrium relationship with interest rate and commercial rents. However, we should not conclude whether there is a real long-run equilibrium relationship between the two variables just by their trends. Therefore, it still needs to apply various models to analyze the data, and use empirical results to determine whether the long-run equilibrium relationships exist between T-REITs and other variables.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
38