Academic journal article Journal of Real Estate Literature

International Articles: Market Signals Associated with Taiwan Reit Ipos: Reactions of Non-Reit Real Estate Stocks

Academic journal article Journal of Real Estate Literature

International Articles: Market Signals Associated with Taiwan Reit Ipos: Reactions of Non-Reit Real Estate Stocks

Article excerpt


This study examines the influence of real estate investment trust (REIT) initial public offering (IPO) announcements on the returns of non-REIT real estate stocks in Taiwan. In contrast to the Malaysian study of Sing, Ho, and Mak (2002), we find that Taiwan REIT IPOs have favorable effects on these stocks. Moreover, investors can obtain higher abnormal returns from investing in underpriced non-REIT real estate stocks following Taiwan REIT IPO announcements.

(ProQuest: ... denotes formulae omitted.)

Global real estate securities investing has become increasingly popular in recent years (Bigman and Chiu, 2005; Hughes, 2007), as evidenced by the growth of the number of global real estate funds. At the end of the last decade, only 10 such funds existed; by the end of 2003 that number had doubled to 20, and it more than doubled again to 41 in 2004 (Hughes and Arissen, 2006). As the momentum continued to increase, the number of global funds increased to 68 in 2005 and to 93 in 2006 (National Association of Real Estate Investment Trusts, 2007). To the best of our knowledge, there are 155 global real estate mutual funds as of February 28, 2010 (Yahoo! Finance, 2010).1 To obtain the benefits of global diversification without sacrificing liquidity, global real estate mutual funds invest in both real estate investment trusts (REITs) and non-REIT public real estate stocks.

Because REITs and non-REIT real estate stocks have a common link to the private real estate market (Giliberto, 1990; Chen and Guo, 2004; Lee, Lee, and Chiang, 2008) and both are traded in the stock market, there may be an interaction between these two categories of stocks such that developments in the REIT market may attract the interest of investors in non-REIT real estate stocks and thereby influence the market for those stocks. If the market believes that REIT initial public offerings (IPOs) indicate a likelihood of prosperity for the real estate market, then the IPOs can signal prosperity for the non-REIT real estate stocks. In contrast, if the market believes that the timing of REIT IPOs coincides with the peaks of the real estate market and thus indicates an over-valuation of real estate values, or that REIT IPOs may trigger a transfer of funds away from the non-REIT stocks, then the IPOs will have a negative effect on these stocks.

Understanding how REIT IPOs influence the returns of non-REIT real estate stocks can help real estate securities investors, such as global real estate securities funds, to make better investment decisions. We therefore examine this issue with the hope of offering some implications for the increasingly popular practice of global real estate securities investing. As far as we know, this subject has not drawn much attention, except for the Malaysian study of Sing, Ho, and Mak (2002). The objectives of our study are to examine the return reactions of non-REIT real estate stocks to REIT IPOs and their determinants in Taiwan. By so doing, this study offers contributions in the following three aspects.

First, we extend the scant existing studies to another important emerging market. Taiwan is an important emerging real estate securities market in the Far East. To the best of our knowledge, the existing literature contains neither studies on market signals associated with Taiwan REIT IPOs nor literature describing the impact of REITs on non-REIT real estate stocks in Taiwan. At the end of May 2007, in terms of market capitalization, the Taiwan REIT (non-REIT real estate stock) market accounted for about 37.29% (30.70%) of the Far East Emerging REIT (non-REIT real estate stocks) market (Moss, 2007).2 The importance of the Taiwan market is also reflected in the inclusion of Taiwan in the Dow Jones global real estate indices and the recently launched FTSE EPRA/NAREIT emerging markets indices. The empirical results of this study therefore can be of help to global real estate securities investors who are interested in real estate securities in Taiwan.3

Second and more importantly, unlike their counterparts in Malaysia, Taiwan REITs (T-REITs) have limited growth prospects because they can only be constructed as closed-end funds and cannot issue additional units after IPOs (Ooi, Newell, and Sing, 2006).4 This unique feature may make REITs compete less in Taiwan than in Malaysia for investor funds with non-REIT real estate stocks. Therefore, T-REIT IPOs may not have adverse effects on non-REIT real estate stocks similar to those of Malaysian REIT IPOs in the previous study. Consequently, when investing in Taiwan real estate securities, global real estate investors should not blindly apply the knowledge obtained from the Malaysian study. This conclusion highlights the need to study the Taiwan real estate securities market.

Third, Sing, Ho, and Mak (2002) do not investigate the distribution of the IPO effects on non-REIT real estate stocks. Unlike their study, we perform regressions to identify the factors that seemingly influence the returns of real estate stocks. Sing, Ho, and Mak (2002) only examine one REIT IPO announcement in Malaysia. We conduct a much broader study and examine all eight REIT IPOs in Taiwan. Moreover, we explore three events, i.e., filing approval, offering, and listing, for each IPO.

In sharp contrast to Sing, Ho, and Mak (2002), we find that the three REIT IPO events all had favorable impacts on real estate stocks. Furthermore, our regression results suggest that investors can obtain higher abnormal returns from investing in underpriced non-REIT real estate stocks around the time of REIT IPOs in Taiwan. Moreover, the first REIT IPO dilutes the market share of capital for indirect real estate investment and thus depresses the performance of non-REIT real estate stock prices.

The structure of this paper is as follows. There is a review of the REIT IPO literature, followed by a discussion of the empirical methods and samples. Next, the empirical results are presented and analyzed. The paper closes with concluding remarks.


Despite the existence of substantial studies on industrial IPO effects, the unique features of REITs suggest a need for studies to assess their IPO effects (Sahi and Lee, 2001; Akhigbe, Johnston, Madura, and Springer, 2004). Wang, Chan, and Gau (1992) found that in stark contrast to equity IPOs, REIT IPOs in the United States had a negative initial performance in the 1970s and 1980s. The suggested explanation for this unique behavior involves the lack of information about the firm (due to the low participation from institutional investors), the non-operating characteristics of the trust and the underlying real estate asset (Wang, Chan, and Gau, 1992; Chan, Stohs, and Wang, 2001). However, measurement errors are also likely another reason for the overpricing. In particular, Below, Zaman, and McIntosh (1995) provide evidence that REIT IPOs between 1972 and 1988 were, on average, correctly priced when the bidask average or ask price is used to calculate returns.

In contrast, Ling and Ryngaert (1997) find that REIT IPOs in the U.S. were significantly underpriced in the early 1990s. However, the positive returns on the first trading day are much lower than those for equity IPOs. Similarly, Dimovski and Brooks (2006a) identify a significant underpricing for Australian REIT IPOs from 1994 to 2004. Nevertheless, in another study, the authors (2006b) report an insignificant average first day return to subscribers for a sample set of REIT IPOs in Australia from 1994 to 1999. In Japan, Kutsuna, Dimovski, and Brooks (2008) investigate REIT IPOs from 2001 to 2006. These Japanese REIT IPOs do not, on average, show significant underpricing.

Previous REIT studies also explored the signaling effects of IPOs. Ghosh, Nag, and Sirmans (2000) investigate the signal that an issuing REIT IPO sends to investors about the issuing REIT. They find that U.S. REITs that underprice IPOs more are likely to undertake seasoned equity offerings (SEOs) sooner and REITs that underprice IPOs also underprice SEOs. In contrast to that study, Akhigbe, Johnston, Madura, and Springer (2004) focus on whether a REIT IPO and its pricing send the market signals about other REITs. The authors discover that the other REITs have negative cumulative average abnormal returns on the REIT IPO listing dates. Although Akhigbe, Johnston, Madura, and Springer (2004) state that REIT IPOs may also create signaling effects on non-REIT real estate stocks in the U.S., they have not conducted further studies on the subject.

Filling this gap, Sing, Ho, and Mak (2002) study the responses of non-REIT real estate stocks to the issuance of REIT IPOs in Malaysia and find that non-REIT real estate stocks show negative cumulative average abnormal returns after the announcement of one REIT IPO. Following their study, we extend the scant existing studies to another important emerging market: Taiwan. In Taiwan, due to regulations, REITs have limited growth prospects, unlike their counterparts in Malaysia (Ooi, Newell, and Sing, 2006).


Empirical Methods

This study, like the prior Malaysian study, makes use of the event study mythology to examine the impacts of T-REIT IPOs. Following Sing, Ho, and Mak (2002), we employ the event-parameter model to calculate the abnormal returns of non-REIT real estate stocks.5 For this study, this model is formulated as follows:

... (1)

where Rit is the return on non-REIT real estate security i on day t, Rmt is the return on the stock market portfolio on day t, Djk is a dummy variable that takes the value of one for the day k within the event window [t1, t2] of event j and is zero otherwise, and εit is the error term. Among the coefficients, βi0 is the intercept, βim is the traditional market coefficient for non-REIT real estate security i, and βijk is the parameter for dummy variable Djk and measures the abnormal returns of non-REIT real estate security i for day k during the event j.

For non-REIT real estate stock i, the cumulative abnormal returns (CARij) for event j over the event window period [t1, t2] can be represented as follows:

... (2)

The average cumulative abnormal returns (ACAR) for N stocks over J event windows can be specified as follows:

... (3)

The variance of (ACAR) is estimated by:

... (4)

The signaling effect from the REIT events on non-REIT real estate stock returns is tested for significance by the t-statistic, defined as:

... (5)

The variation in stock price responses of non-REIT real estate stocks to the REIT IPO announcements is assessed with the following model:

CCAR^sub ih^ = α + β^sub 1^MTB^sub ih^ + β^sub 2^IPORETURN^sub h^ + β^sub 3^EXCH^sub h^ + β^sub 4^RELSIZE^sub ih^ + β^sub 5^MARKETUP^sub h^ + β^sub 6^FIRST^sub h^ + ε^sub ih^ (6)


CCAR^sub ih^ = The combined CAR of non-REIT real estate stock i in response to the IPO announcements of REIT h over its filing approval date, offering date, and listing date;

MTB^sub ih^ = 1 if the market-to-book value ratio of non-REIT real estate stock i at the previous quarter before the IPO filing approval date of REIT h is less than 1; otherwise, MTBih = 0;

IPORETURN^sub h^ = The percentage change between the offering price and the closing price on the first trading day for REIT h;

EXCH^sub h^ = 1 if REIT h is listed on the exchange market; otherwise, EXCHh = 0,

RELSIZE^sub ih^ = The relative size of the non-REIT real estate stock i, measured as the market value of non-REIT real estate stock i relative to the market value of the REIT h on its listing date;

MARKETUP^sub h^ = The cumulative returns in the stock market portfolio from 100 days to 20 days prior to the first trading day of REIT h;

FIRST^sub h^ = 1 if REIT h is the first REIT in Taiwan; otherwise, FIRSTh = 0; and εih = The error term.

Following Akhigbe, Johnston, Madura, and Springer (2004), we analyze the responses of non-REIT real estate stocks to the combined CARs over the announcements during an REIT IPO process because the stocks could respond to the news of the filing approval, the offering, and the listing. In particular, this regression uses the combined CARs (CCAR) over the three dates of each IPO as the dependent variable.

Vaidyanathan and Chava (1997) argue that a stock may be underpriced if its MTB is less than 1. If investors are attracted to non-REIT real estate stocks due to the announcement of a REIT IPO, then the undervalued non-REIT real estate stocks should have greater appreciation potential. In this case, we anticipate a positive value of MTB, i.e., lower-valued non-REIT real estate stocks should have greater cumulative abnormal returns.

The first trading day return of a REIT IPO may represent market sentiment for the indirect real estate security market. If this market sentiment spreads to non-REIT real estate stocks, then there will be a positive relationship between IPORETURN and non- REIT real estate stock returns (Akhigbe, Johnston, Madura, and Springer, 2004). Conversely, IPORETURN is also likely positively linked to the attracting power for real estate capital, which moves funds from non-REIT real estate stocks to the REITs. This case expects a negative relationship between IPORETURN and non-REIT real estate stock returns (Liziere and Satchell, 1997; Sing, Ho, and Mak, 2002; Lee, Lee, and Chiang, 2008).

REIT IPOs that are traded in the exchange market or the over-the-counter market may have different effects on non-REIT real estate stocks. Favored by institutional investors, exchange-listed REIT IPOs attract more market attention and thus produce larger impacts (Kohers, 1999). On the other hand, OTC REIT IPOs may bring about greater impacts if they deliver information that was formerly more unclear to the market (Akhigbe, Borde, and Whyte, 2003). Thus, we do not have a prior hypothesis on the effect of EXCH.

Smaller IPOs draw less attention. Consequently, a smaller IPO generates weaker signals and has less impact on the re-evaluation of larger non-REIT real estate stocks. Nevertheless, a larger REIT IPO can have a greater dilution effect, i.e., a greater supply of real estate stocks may lead to greater downward pressure on non-REIT real estate stocks. Each of the two predictions about RELSIZE has support from Akhigbe, Borde, and Whyte (2003) and Akhigbe, Johnston, Madura, and Springer (2004), respectively.

Following Akhigbe, Johnston, Madura and Springer (2004), this study utilizes MARKETUP as the proxy for recent stock market sentiment. When the market sentiment is more favorable, investors may evaluate REIT IPO information more optimistically. If the optimism spreads to non-REIT real estate stocks, these stocks may generate more cumulative abnormal returns.

Following a period of dormancy, the first IPO usually has the strongest impact on revaluation in the same industry sector (Akhigbe, Borde, and Whyte, 2003). REITs and non-REIT real estate stocks are both real estate stocks. Consequently, we anticipate a significant coefficient for FIRST that may be positive or negative, depending on the market anticipation about the influence of REIT IPOs on non-REIT real estate stocks.

The Samples

Presently, there are eight publicly traded REITs in Taiwan, as presented in Exhibit 1.6 Fubon REIT No.1 was the first REIT introduced in the market. Its public offering was oversubscribed by more than a factor of five. The information on the REIT and non-REIT real estate stocks in this study is from the Taiwan Economic Journal (TEJ) and is supplemented by the Market Observation Post System (MOPS) of the Taiwan Stock Exchange, as well as, which is a leading historical news provider in Taiwan.

This study, like the study of Liow and Sim (2006), employs construction stocks as non-REIT real estate stocks in Taiwan because construction companies undertake real estate building and development activities and thus obtain their profits from real estate markets.

To facilitate empirical investigation, we first collect the daily returns of construction stocks from January 2, 2004 to June 29, 2007 from the TEJ. Following Howe and Jain (2004), the period is chosen in order to cover all event dates and to estimate Equation (1). After eliminating the firms without all of the required information, we have 42 construction stocks, which include 25 real estate development stocks and 17 building stocks, for each year in this study.

Exhibit 2 presents the descriptive statistics for the sample construction stocks and the relevant variables. The sample construction stocks have an average capitalization of 4.625 billion New Taiwan dollars and are slightly smaller than the T-REITs, as indicated by mean value of RELSIZE, which is about 0.989. The construction stocks have an average market-to-book ratio of 1.084. Thus, the construction stocks on average appear to be fairly priced with respect to their asset values. Nevertheless, the mean value of the MTB dummy reveals that about 58% of the construction stocks are priced at a discount compared to their asset values.

One possible reason for this phenomenon stems from the poor demand prospects for new real estate construction works, which is strongly influenced by political instability (Tsenkova, 2009; Fisher, 2010). In particular, the construction firms probably reacted to the favorable proposals put forward by candidates during the 2004 presidential election campaign and produced more supply for the real estate market. These firms, however, face poor demand prospects because of intensified Pan-Blue and Pan-Green conflicts during the study period to a level hardly seen before.7

The average CCAR is 2.387%;8 thus, investors in construction stocks on average earn 2.387% abnormal returns from each REIT IPO process. The mean of MARKETUP is 4.225; thus, the stock market on average has a cumulative return of 4.225% from 100 days to 20 days prior to a REIT IPO. The average return of a REIT IPO on the first trading day is 0.016%, which indicates that T-REIT IPOs are not significantly underpriced. EXCH indicates that seven REITs are listed on the Taiwan Stock Exchange among the existing eight T-REITs.9


Impacts of REIT IPO Announcements on Construction Stocks

Exhibit 3 presents the average cumulative abnormal returns (CARs) and the percentages of positive CARs generated from construction stock on the filing approval dates, the public offering dates, and the listing dates. Clearly, construction stocks enjoy significantly positive average CARs on each of the three dates.10 The returns reveal that REIT IPO events deliver net favorable messages to construction stocks. These outcomes contrast with the findings of Sing, Ho, and Mak (2002), who show that the announcement of REIT IPOs in Malaysia had a negative effect on non-REIT real estate stocks. Moreover, the patterns of average CARs and the percentages of positive CARs are consistent with the notion that media coverage affects stock price reaction to information (Vega, 2006).

The construction stocks have the most favorable responses to REIT IPOs on the offering dates, with a significant 2-day ACAR of 1.099% and a significant 3-day ACAR of 1.423%. Such timing is likely due to greater attention being paid to the offering dates by the investors and media, as these are the days on which the public can purchase REIT shares. This reasoning is backed by the presence of more news reports at on the offering dates than on the filing approval dates or listing dates.

According to Article 14 of the Real Estate Securitization Statute, T-REITs must offer their shares to the public within three months of their filling approval dates. Correspondingly, it is possible to anticipate the public offering dates. The news archive of supports this conjecture. This may be the reason that construction stocks accumulate more abnormal returns over the interval (-1, +1) than over the interval (0, +1) on the REIT offering dates. The proportions of construction stocks experiencing positive CARs, i.e., 83.33% for the three-day window and 80.90% for the two-day window, are also consistent with this possibility.

On the other hand, the market has no information on when the Financial Supervision Commission will approve the REIT IPO applications. News reports are issued after each filing is approved. Construction stocks are therefore less likely to respond before the REIT IPO filing approval dates. The stocks have a significant two-day ACAR of 0.478% and an insignificant three-day ACAR of 0.198%; 69.05% of construction stocks experience positive cumulative abnormal returns over the (0, +1) interval, and only 57.14% have positive CARs over the (-1, +1) interval. The effects of the magnitude and significance of the ACARs, as well as fractions of positive CARs on the filing approval dates appear to agree with this story.

The news archives show that the listing dates draw less media attention than the offering dates. A dearth of information may thus be one of the reasons that the ACARs are smaller on the listing dates than those on the offering dates. The news coverage pattern for the REIT listing dates is similar to that for the filing approval dates: news is only reported on the listing dates, and the media pays little attention before or afterwards. Similarly, the two-day ACAR of 0.486% is larger than the threeday ACAR (0.320%) on the listing dates. The percentage of construction stocks that have positive CARs is 71.43% for the two-day window and 69.05% for the three-day window. This pattern is similar to that of the filing approval dates.

Variations in Construction Stock Responses

The regression of construction stock responses is estimated with weighted least squares to account for heteroscedasticity.11 Exhibit 4 presents the results. The variance inflation factors (VIFs) are all below 10, which indicate that this regression does not suffer from severe multi-collinearity (Kennedy, 1993; Guijarati, 2003; O'Brien, 2007).

As hypothesized, the MTB coefficient is positive and highly significant at the 1% level. Explicitly underpriced construction stocks have 2.397% more abnormal returns than other construction stocks. Thus, by investing in under-valued construction stocks, investors can double the returns of average investors, who earn about 2.387% during each REIT IPO process.12

The coefficient for the REIT IPO first-day return variable is not significant. This coefficient may be interpreted as the net outcome for both the diluting force and the market sentiment. To be precise, while positively correlated to the indirect real estate market sentiment, higher REIT IPO first-day returns also reflect more capital attraction and larger depressing power on construction stock prices. The two forces offset each other and lead to the insignificance of the IPORETURN coefficient.

EXCH has a significant and negative coefficient, which indicates that Taiwan REIT IPOs traded on the GreTai Securities Market (GSM), i.e., the over-the-counter (OTC) market, have greater impacts on construction stock prices than those traded on the Taiwan Stock Exchange (TSE).13 This finding contrasts with that of Akhigbe, Borde, and Whyte (2003) on industrial firm IPOs in the U.S. Nevertheless, it is consistent with their thought that OTC-traded IPOs with greater asymmetric information could release more valuable incremental information than exchange-listed IPOs (Akhigbe, Borde, and Whyte, 2003).

Unlike previous studies, this study does not find a significant coefficient of RELSIZE, which indicates that the relative sizes have no significant effect on the abnormal returns of construction companies. In other words, large construction stocks experience price increases that are similar to those of small construction stocks when T-REITs offer IPOs. The lack of significance could be because their revaluation effects (Akhigbe, Borde, and Whyte, 2003) offset their dilution effects (Akhigbe, Johnston, Madura, and Springer, 2004).

Similarly, the MARKETUP coefficient is not significant, which is not consistent with the empirical results presented by Akhigbe, Johnston, Madura, and Springer (2004). Specifically, construction stocks do not accumulate more abnormal returns on the announcements of REIT IPOs when the stock market sentiment is more favorable. This finding implies either that investors do not evaluate REIT IPO information more optimistically or that the favorable sentiment does not spread to construction stocks.

As anticipated, the FIRST coefficient is significant at the 1% level. However, its negative sign is contrary to the finding of Akhigbe, Borde, and Whyte (2003). The first Taiwan REIT IPO, Fubon REIT No.1, either conveys stronger but negative revaluation information or has a greater fund-transfer effect. After a closer look, the 500% over-subscription of the first T-REIT IPO clearly provides an explanation of the fund-transfer effect.

Robustness Checks

To check the robustness of the above results, we exclude the first Taiwan REIT IPO and repeat the analyses. Exhibits 5 and 6 present the empirical results. The announcements of the seven remaining REIT IPOs clearly deliver positive messages to construction stocks. The patterns of CARs even now are consistent with the media coverage pattern. In the regression, MTB and EXH still have significant coefficients with the same signs and magnitudes similar to those of the results from all eight REITs. Explicitly, the evidence still supports the claim that OTC-traded IPOs could release more valuable incremental information than exchange-listed IPOs and that investors can double the returns of average investors when investing in under-valued construction stocks.

We further divide construction firms into building companies and real estate development companies as another robustness check.14 The former are contracted and model-based production building-construction companies (Zhang, 2008). The latter commission building companies to build residential properties or operate a commercial property rental business (Zhang, 2008). As shown in Exhibit 7, both building companies and real estate development companies enjoy positive CARs on the announcements of REIT IPOs. Exhibit 8 shows that the coefficients of MTB and EXH are statistically significant. Again, the evidence supports our hypothesis.


This study examines the reactions of non-REIT real estate stocks to Taiwan REIT IPO announcements. In contrast to the Malaysian study of Sing, Ho, and Mak (2002), we find that the IPOs signify profit and prosperity for non-REIT real estate stocks in Taiwan. Furthermore, underpriced stocks have more abnormal returns than other non- REIT real estate stocks following the IPO announcements.

The immediate implication of our results is that investors could potentially benefit from investing in non-REIT real estate stocks on REIT IPO announcements in Taiwan. In particular, investors should invest in the underpriced stocks to gain more significant extra returns in the Taiwan securitized real estate market. For global real estate securities investors, our study highlights the danger of blindly applying knowledge from one market to another market. Consequently, one possible future extension of this study is the examination of other real estate securities markets to provide valuable information for the increasing market of global real estate securities investors.


1. The global real estate mutual funds originated across at least 19 countries and had total assets under management that were close to US $39 billion at the end of 2006 (NAREIT, 2007).

2. The Far East emerging market includes Taiwan, Malaysia, South Korea, and Thailand (Moss, 2007).

3. Examples of the real estate funds currently authorized in their prospectus to invest in Taiwan real estate securities are DFA Global Real Estate Securities, TLG Asia Pacific REITs Fund, Yuanta Global Realty and Infrastructure Fund, and ING Asia-Pacific Real Estate Fund.

4. Important T-REIT regulations include: (1) At least 75% of the REIT asset value must be composed of bank deposits, government or financial bonds, Treasury bills, and negotiable certificates of time deposits, while the proceeds derived from the REIT investment, which are to be distributed pursuant to the REIT contract, must be distributed within six months after the end of each fiscal year; (2) REITs must be closed-end funds; (3) a REIT's investment in securities should not exceed 40% of its assets or NT$6 million; (4) the trust property's rent and lease terms are not subject to existing laws; (5) REITs are organized as special-purpose trusts; and (6) the trustees cannot issue additional beneficial securities.

5. This approach is widely adopted by non-REIT studies such as Binder (1985) and Thompson (1985).

6. All the eight T-REITs are infinite life REITs. Five of them are sponsored or co-sponsored by construction firms.

7. The Pan-Blue force tends to favor a Chinese nationalist identity over a separate Taiwanese one, and favors a softer policy and greater economic linkage with the People's Republic of China, as opposed to the Pan-Green force (Wikipedia, 2010a,b).

8. The CCAR here and hereafter is the combined CAR over the two-day CAR on the filing approval date, the three-day CAR on the offering date, and the two-day CAR on the listing date. The empirical results are qualitatively similar when using the three-day CAR instead of the two-day CAR on the filing date or when the listing date is used to compute the CCAR.

9. Due to its size, Kee-Tai Star REIT is the only T-REIT not listed on the TSE.

10. The offering date of San-Ding REIT is also the filing approval date of Kee-Tai Star REIT. We split the abnormal returns in half for each of the two REITs to reconcile the conjunction effect. We obtain qualitative similar results when we try some other slightly different splitting methods or ignore the overlapping.

11. We follow the estimation procedure of WLS stated in Wooldridge (2006, p. 292) to correct heteroscedasticity.

12. The MTB coefficient remains significant when IPORETURN was dropped from the regression.

13. The TSE has more strict listing requirements on size, minimum years since establishment, ownership, and so on than GSM.

14. We also perform the analyses with the first IPO excluded and obtain qualitatively similar results.



Akhigbe, A., S.F. Borde, and A.M. Whyte. Does an Industry Effect Exist for Initial Public Offerings? The Financial Review, 2003, 38:4, 531-51.

Akhigbe, A., J. Johnston, J. Madura, and T.M. Springer. Market Signals Associated with REIT IPOs. Journal of Real Estate Finance and Economics, 2004, 28:4, 355-67.

Below, S., M.A. Zaman, and W. McIntosh. The Pricing of Real Estate Investment Trust Initial Public Offerings. Journal of Real Estate Finance and Economics, 1995, 11, 55-64.

Bigman, T. and C. Chiu. The Case for a Strategic Allocation to Global Real Estate Securities. Morgan Stanley Investment Management, 2005, 1:2, 15-34.

Binder, J.J. Measuring the Effects of Regulation with Stock Price Data. The Rand Journal of Economics, 1985, 16:2, 167-83.

Chan, S.H., M.H. Stohs, and K. Wang. Are Real Estate IPOs a Different Species? Evidence from Hong Kong IPOs. Journal of Real Estate Research, 2001, 21, 201-20.

Chen, M.C. and C.J. Guo. An Analysis of Long-term Performance of Taiwan Property Stock Comparative Study of Direct and Indirect Property Investment. Management Research, 2004, 4:2, 144-68.

Dimovski, W. and R. Brooks. Factors Influencing Money Left on the Table by Property Trust IPO Issuers. Journal of Property Research, 2006a, 23, 269-80.

_____. The Pricing Property Trust IPO in Australia. Journal of Real Estate Finance and Economics, 2006b, 32, 185-99.

Fisher, Y. Peace and the Economy, 2010. Retrieved from

Ghosh, C., R. Nag, and C.F. Sirmans. The Pricing of Seasoned Equity Offerings: Evidence from REITs. Real Estate Economics, 2000, 28, 363-84.

Giliberto, S.M. Equity Real Estate Investment Trusts and Real Estate Returns. Journal of Real Estate Research, 1990, 5:2, 259-63.

Gujarati, D.N. Basic Econometrics. Fourth Edition. Singapore: McGraw-Hill, 2003.

Howe, J.S. and R. Jain. The REIT Modernization Act of 1999. Journal of Real Estate Finance and Economics, 2004, 28, 369-88.

Hughes, F. International REIT-Market. Institutional Investment Real Estate Magazine, 2007, February.

Hughes, F. and J. Arissen. Top Real Estate Fund Managers Go Global. Real Estate Portfolio, 2006.

Kennedy, P.A. Guide to Econometrics. Third Edition. Oxford: Blackwell, 1993.

Kohers, N. The Industry-wide Implications of Dividend Omission and Initiation Announcements and the Determinants of Information Transfer. Financial Review, 1999, 34, 137-58.

Kutsuna, K., W. Dimovski, and R. Brooks. The Pricing and Underwriting Costs of Japanese REIT IPOs. Journal of Property Research, 2008, 25, 221-39.

Lee, M.-L., M.-T. Lee, and K.C.H. Chiang. Real Estate Risk Exposure of Equity Real Estate Investment Trusts. The Journal of Real Estate Finance and Economics, 2008, 36:2, 165-81.

Ling, D. and M. Ryngaert. Valuation Uncertainty, Institutional Involvement, and the Underpricing of IPOs: The Case of REITs. Journal of Financial Economics, 1997, 43, 433- 56.

Liow, K.H. and M.C. Sim. The Risks Return Profile of Asian Real Estate Stocks. Pacific Rim Property Research Journal, 2006, 12:3, 283-310.

Liziere, C. and S. Satchell. Interactions between Property and Equity Markets: An Investigation of Linkages in the United Kingdom 1972-1992. Journal of Real Estate Finance and Economics, 1997, 15:1, 11-26.

Moss, A. Asia REIT Market Overview. AME Capital, 2007, June.

O'Brien R.M. A Caution Regarding Rules of Thumb for Variance Inflation Factors. Quality and Quantity, 2007, 41, 673-90.

National Association of Real Estate Investment Trusts. Global Mutual Funds Hit Their Stride. Real Estate Portfolio, 2007, March/April.

Ooi, J.T.L., G. Newell, and T.-F. Sing. The Growth of REIT Markets in Asia. Journal of Real Estate Literature, 2006, 14:2, 203-22.

Sahi, W. and S.L. Lee. The Initial Returns of U.K. Property Company IPOs. Journal of Property Investment and Finance, 2001, 19, 127-39.

Sing, T.F., K.H.D. Ho, and M.F. Mak. Real Estate Market Reaction to Public Listings and Acquisition News of Malaysian REITs. Journal of Real Estate Portfolio Management, 2002, 8: 3, 209-27.

Thompson, R. Conditioning the Return-generating Process on Firm-specific Events: a Discussion of Event Study Methods. The Journal of Financial and Quantitative Analysis, 1985, 20:2, 151-68.

Tsenkova, S. Housing Policy Reforms in Post Socialist Europe Lost in Transition. Heidelberg: Physica-Verlag, 2009.

Vaidyanathan, R. and S. Chava. Stock Returns and Price to Book Ratio. ICFAI Journal of Applied Finance, 1997, 3:2, 45-53.

Vega, C. Stock Price Reaction to Public and Private Information. Journal of Financial Economics, 2006, 82:1, 103-33.

Wang, K., S.H. Chan, and G.W. Gau. Initial Public Offerings of Equity Securities: Anomalous Evidence Using REITs. Journal of Financial Economics, 1992, 31, 381-410.

Wikipedia. Pan-Blue Coalition. Retrieved from /Pan-Blue Coalition. 2010a.

Wikipedia. Pan-Green Coalition. Retrieved from http: / / Coalition. 2010b.

Wooldridge, J.M. Introductory Econometrics: A Modern Approach. 2006.

Yahoo! Finance. Mutual Fund Top Performers-Global Real Estate. Retrieved from http: / / /GR.html. 2010.

Zhang, Z.Y. Growth-Restricted Industry: Forecasting the Construction Industry under the Current Regulation. Professional Engineer Journal, 2008, 50, 10-5.

The authors would like to thank Graeme Newell (the editor) and two anonymous referees for their helpful comments and suggestions. Thanks are also due to the Taiwan National Science Council for its research support with Grant NSC 98-2410-H-259-073 and The Trust Association of the Republic of China for its financial support.

[Author Affiliation]

Ming-Long Lee

National Dong Hwa University

Shew-Huei Kuo

National Yunlin University of Science and Technology

Ming-Te Lee

Ming Chuan University

Chia-Wei Lin

National Yunlin University of Science and Technology

[Author Affiliation]

Ming-Long Lee, National Dong Hwa University, Shoufeng, Hualien, Taiwan 974 or

Shew-Huei Kuo, National Yunlin University of Science and Technology, Taiwan. Ming-Te Lee, Ming Chuan University, Taiwan

Chia-Wei Lin, National Yunlin University of Science and Technology, Taiwan.

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.