The results for the cumulative abnormal returns (CARs) are reported in Table 3.
Panel A reports the CARs for the all-acquirer portfolio, panel B reports the CARs for acquirers in the public-to-public mergers, and panel C reports the CARs for acquirers in the public-to-private mergers. In each portfolio, abnormal returns are calculated for
+2) event windows.
[Insert Table 3 Here]
In panel A Table 3, the results for the all-acquirer portfolio show that when announcing a merger with other firms, acquirers earn small and significant positive announcement returns no matter the value-weighted or equal-weighted market index is used.
These positive results are quite different from those of other literature, for example Campbell et al. (2001) find that when REIT acquirers have significant negative abnormal returns, -0.6%, and Travlos (1987) finds that acquirers experience small negative abnormal returns of -1% to -2% for firms in conventional industries.
Cmpared to Campbell et al. (2001), we used different market indices. Campbell et al.
(2001) use the Russell 2000 index as it includes more REITs. The difference in these results might be due to the structural change of REITs mentioned by Ling and Ryngaert (2001).3
In this research, we further separate the sample into two different kinds of groups.
One is public-to-public portfolio, and the other is public-to-private portfolio.
The results for the public-to-public mergers are reported in panel B Table 3. In panel B, all results in this portfolio are negative and significant, ranging from -0.6% to -1.4%, and consistent with those found in Campbell et al. (2001). In Campbell et al.
(2001), they argue that the results are in line with the information signaling hypothesis, which states that whenever managers choose to use stocks as the main method to finance a merger, they send a signal to investors that the firm’s stocks have been
3 In Ling and Ryngaert (2001), they have argue that there is a significant structural change between REITs prior to 90’s and after 90’s.
overvalued, consequently the market reacts negatively.
We further separate this portfolio into two subgroups based on whether acquirers and targets belong to the same conglomerate. The results are reported in Table 4.
[Insert Table 4 Here]
In Table 4, only the CARs for the different conglomerate portfolio are significantly negative. The CARs for the acquirers merging with private targets are reported in panel C Table 3. In this panel, the results show that when an acquirer announces to merge with another private target, their shareholders earn significantly positive abnormal returns, ranging from 0.13% to 0.32%. These CARs are also consistent with those found in Campbell et al. (2001, 2005) and other literature such as Travlos (1987) and Fuller et al. (2002) that study on conventional firms. To find out the sources of these positive abnormal returns, we separate these samples into two different parts. One includes only events that are classified as mergers and acquisitions, in which those acquirers are acquiring other companies, and the one in which the acquirers are only acquiring others’ assets. We not only separate these public-to-private events into the two subsamples presented above but also further separate these acquirers into same conglomerate and the different conglomerate parts as we have done in the public-to-public portfolio. The results for these portfolios are reported in Table 5. In panel A and B, we report the results for the Acquire Company (AC) portfolio and the results for the Acquire Assets (AA) portfolios respectively.
[Insert Table 5 Here]
insignificant, and these results are different from the findings in Campbell et al.
(2001)4. Again, we attribute this difference in the results to the fact that we have used different market indices or to the structural change in REITs.
In panel B, the results for Acquire Assets portfolio are positive and significant. It is worth noting that the CARs are insignificant in the same conglomerate portfolio, but significantly positive in the different conglomerate portfolio. We attribute these results to that the investors might take for granted that the firms acquiring assets from others are doing this for the benefit of acquiring firms, so they react positively toward these events. However, when it comes to the “buying their own assets” events, the market seems do not regard this kind of events as beneficial as those of buying assets from others, therefore the abnormal returns for those same conglomerate assets acquirers are insignificant.
To check the robustness for the results presented above, we exercise two different tests. Firstly, in order to check whether the CARs presented above are correct, we use the same sample period used in Campbell et al. (2001), which is from 1994 to 1998. If the CARs are calculated correctly, we should have similar results like those reported in Campbell et al. (2001).
Secondly, in Campbell et al. (2008), they discover that REITs which participated in mergers underperform in one year following the merger. With regard to this underperformance, we exclude those overlapping events and keep only the first event for each firm to examine whether those results found above still exist5. The intuition is that if the one-year underperformance really bias the results both for public-to-public and public-to-private portfolios, we expect that after dropping those overlapping
4 In Campbell et al. (2001) they find significantly positive abnormal returns for the public-to-private subsample.
5 In this robustness test, the time length between two events that were done by the same acquiring firms must exceed one year.
events, the negative CARs for the public-to-public portfolio should become insignificant or the magnitude should be reduced and the positive CARs for the public-to-private portfolio should become larger. The results for these two robustness tests are reported in Table 6 and 7.
[Insert Table 6 Here]
In table 6, the results for all three portfolios are similar to those found in Campbell et al. (2001). The CARs for the public-to-public portfolio are significantly negative, and results for the public-to-private portfolio are significantly positive.
[Insert Table 7 Here]
For the second robustness test, the results are reported in Table 7. After dropping those overlapping events, the results for period from 1983 to 2007 are partially consistent with the assumption that we have assumed in previous paragraph. The magnitude of the results for the public-to-private portfolio becomes greater than those reported in Table 4. Moreover, the results using events from 1994 to 1998 provide better evidence.
The significance of the CARs for the public-to-public portfolio become insignificant and the magnitude of CARs for the public-to-private portfolio become greater than those in Table 4.
IV. Conclusion
We show that when REIT firms announce public-to-public mergers, the market reacts negatively toward these events, but when REIT firms announce
announcement returns only exist when the targets are belong to other conglomerates.
The insignificant results for the same conglomerate might due to those investors take these events not as bad as those different conglomerate mergers. They might think that those parent firms are trying to cut-off parts that are losing money or they are trying to refocus on their core business.
Other than the CARs around announcements for merger of company events, in this research, we also find that when REITs announce to acquire other firms’ assets, the market also reacts positively toward these events and the insignificant results for the subsample of same conglomerate in Acquire Asset (AA) portfolio reveal that the market does not react so positively because they know that these kinds of events do not have too many beneficial impacts on acquiring firms’ future operations.
Reference
z Allen, Paul R. and C. F. Sirmans., 1987, “An Analysis of Gains to Acquiring Firm’s Shareholders: The Special Case of REITs” Journal of Financial Economics 1, 18:
175-184
z Ang, James S. and Yung Ling Lo, 2007, “The Market-Timing Ability of Low Transparency Firms in the Acquisition Market”, working paper
z Campbell, Robert D., Chinmoy Ghosh, and C. F. Sirmans, 2001, “The Information Content of Method of Payment in Mergers: Evidence from Real Estate Investment Trusts (REITs)”, Journal of Real Estate Economics 29, 3: 361-87
z Campbell, Robert D., Chinmoy Ghosh, Milena Petrova, and C. F. Sirmans, 2006,
“Corporate Governance and Performance in the Market for Corporate Control: The Case of REITs”, working paper
z Campbell, Robert D., Erasmo Giambona, and C.F. Sirmans, 2008, “The
Post-Merger Underperformance Anomaly: Evidence from REITs”, working paper z Eichholtz, Piet M.A., and Nils Kok, 2008, “How Does Market for Corporate Control
Function for Property Companies?” Journal of Real Estate Finance Economics 36, 2: 141-163
z Fuller, Kathleen, Jeffry Netter, and Mike Stegemoller, 2002, “What Do Returns to Acquiring Firms Tell Us? Evidence from Firms That Make Many
z Acquisitions”, Journal of Finance 57, 4: 1763-1793
z Ghosh, Chinmoy and C. F. Sirmans, 2003, “Board Independence, Ownership Structure and Performance: Evidence from Real Estate Investment Trusts”, Journal
of Real Estate Finance and Economics 26, 2/3:287-318
z Ghosh, Chinmoy, John Harding, Özcan Sezer, and C. F. Sirmans, 2006, “The Role of Executive Stock Options in REIT Repurchases”, working paper
Takeovers”, The Journal of American Review 76, 2:323-329
z Jensen, Michael C., and William H. Meckling, 1976, “Theory of the Firm:
Managerial Behavior, Agency Costs and Ownership Structure”, Journal of
Financial Economics 3, 4:305-360
z Lang, Mark H. and Russell J. Lundholm, 1996, “Corporate Disclosure Policy and Analyst Behavior”, The Accounting Review 71, 4:467-492
z Moller, Sara B., Frederik P. Schlingemann, René M. Stulz, 2004, “Firm Size and The Gains from Acquisitions”, Journal of Financial Economics 73, 2: 201-228 z Myers, Stewart C., and Nicholas S. Majluf, 1984, “Corporate Financing and
Investment Decisions When Firms Have Information That Investors Do Not Have”,
Journal of Financial Economics 13, 2: 187-221
z Opler, Tim, and Sheridan Titman, 1993, “The Determinants of Leveraged Buyout Activity: Free Cash Flow vs. Financial Distress Costs”, Journal of Finance 48, 5:1985-1999
z Petersen, Mitchell A., 2008, “Estimating Standard Errors in Finance Panel Data Sets:
Comparing Approaches”, forthcoming, Review of Financial Studies
z Shen, Yang-pin, Li-Ching Chiu and Chiuling Lu, 2008, “A Discrete Random Effect Logit Model of the Determinants of Asset-Backed Securitization”, Journal of
Financial Studies 16, 2: 69-100
z Sudarsanam, Sudi Peter Holl, and Ayo Salami, 1996, “Shareholder Wealth Gains in Mergers: Effect of Synergy and Ownership Structure”, Journal of Business Finance
and Accounting 23, 5-6: 673-698
z Travlos, Nickolaos G., 1987, “Corporate Takeover Bids, Methods of Payment, and Bidding Firms’ Stock Returns”, Journal of Finance 42, 4:943-963
z Trautwein, Friedrich, 1990, “Merger Motives and Merger Prescriptions”, Strategic
Management Journal 11,283-295
z Vincent, Linda, 1999, “The Information Content of Funds From Operations (FFO) for Real Estate Investment Trusts (REITs)”, Journal of Accounting and Economics 26, 1-3: 69-104
Table 1
Distribution of the Merger Events and Number of Firms Being Acquirers During 1983-2007
There are a total 1887 REITs merger events during 1983 to 2007. In panel A, we report the distribution for the events used in calculating abnormal returns across the sample period; we also present the number of events for public-to-public portfolio (acquirer and target are publicly traded companies) and public-to-private portfolio (acquirer are publicly traded and target are privately traded companies).
In panel B, we report the distribution for firms that being acquirers and those didn’t. There are a total 413 REITs conducting 4195 merger observations during 1983 to 2007. After dropping those observations which have some missing data from the CRSP database, the Compusata-North America database, and the IBES database, we finally have 979 samples, with 260 observations which have once being acquirers, and 719 non-acquirers.
Panel A : Numbers of Events
Year Total Obs. Public-to-Public Public-to-Private 1983 6 1 5
Table 1 (continued)
Panel B : Numbers of Acquirers and Non-AcquirersYear Total Obs. Acquirers Non-Acquirers
1983 8 1 7
1984 12 1 11
1986 19 2 17
1988 33 2 31
1989 29 2 27
1990 30 3 27
1991 26 2 24
1992 28 4 24
1998 14 5 9
1999 27 6 21
2000 18 1 17
2001 58 10 48
2002 83 19 64
2003 90 21 69
2004 120 35 85
2005 135 52 83
2006 129 60 69
2007 121 34 86
Total 979 260 719
Table 2 Summary Statistics
This table presents the summary information about the firm characteristics for all 979 observations in the sample. Data are obtained from the CRSP database, the Compustat North-America database, and the IBES database. SIZE is the stock price multiplies the outstanding shares of each observation at the end of year t-1, and take nature log. In this table, we report the MV (the market value) instead. Tobin’s Q ratio proxies firms’ growth opportunities and is calculated by adding the market value of assets and book value of debt, and divided by the book value of each observation at the end of year t-1. LVG proxies firm’s level of leverage, and is the long-term debt divided by the total assets for each observation at the end of year t-1. ROA is the variable that proxy for firm’s profitability, which is the Net Income (NI) divided by the total assets for each observation at the end of year t-1. TRANS proxies for firm’s transparency and this variable is the sum of the number of analyst following during year t-1 for each observation. CF proxies for cash owned by management for each observation, it is calculated by using the earnings before interests, taxes, depreciations, and Amortizations (EBITDA) divided by the book value of assets for each observation at the end of year t-1. The variable HQ*HL is the interaction term of two dummies. One is that if the Tobin’s Q ratio for an observation is higher than median then this dummy variable will equal to 1, and zero otherwise, and the other is that if the leverage for one observation is higher than median then this dummy equal to 1, and zero otherwise. Y is the dummy variable that equals 1 if a firm acquires other firms in a given year; and zero otherwise.
N indicates the number of observations.
Variables N Mean S.D. Minimum Maximum
Y 979 0.27 0.44 0 1
MV (mil.) 979 1544.19 2420.64 1217.95 22413.25
Tobin’s Q 979 0.65 0.46 0.01 3.78
LVG 979 2.25 0.46 -13.14 132.65
ROA 979 0.03 0.07 -1.57 0.40
TRANS 979 3.96 3.51 1 26
CF 979 0.06 0.07 -1.54 0.29
HQ*HL 979 0.17 0.38 0 1
Table 3
The Cumulative Abnormal Returns (CARs) for Acquirers Around the Merger Announcements
This table reports the cumulative abnormal returns (CARs) in percentage for 1887 acquiring firms in Real Estate Investment Trust (REITs) mergers. We apply the standard event study methodology that mainly follows Campbell et al. (2001). The detail in calculating the CARs are as follows. We use the market model to estimate the parameters that are needed in calculating the normal returns for each acquiring firm. The estimation period is from day -190 to day -10. Moreover, for the market index, we use two different kinds of market indices, one is CRSP NYSE/AMEX/NASDAQ value-weighted index, and the other is CRSP NYSE/AMEX/NASDAQ equal-weighted index.
We separate these events into three categories, the all acquirer portfolio, the public-to-public portfolio, and the public-to-private portfolio. For each portfolio, we calculated the CARs for 4 different event windows, 5-day (day -2 to +2), 3-day (day -1 to +1), 2-day (day 0 to day +1), and 1-day (day 0). For each event window, we first calculate the abnormal return by subtracting the returns for each observation on each day with the normal returns estimated from the market model, and add them together to get the CARs. For example, 5 days CARs are calculated as follow:
N
To test the significance of these CARs, we report the t-statistics. In panel A, B, and C, we report the results for the all acquirer portfolio, the public-to-public portfolio, and the public-to-private portfolio respectively and both the results using value- and equal-weighted market index are reported for each event window.
Panel A All Acquirer portfolio
Value-Weighted Equal-Weighted
Panel B Public-to-Public portfolio
N=139 N=139
***, **, and * indicate significance at the 1%, 5%, and 10% level, respectively.
Table 3 (continued)
Panel C Public-to-Private portfolioN= 1748 N= 1748
CARs t-value CARs t-value
5-DAY(-2,2) 0.32% *** 3.04 0.34%*** 3.22 3-DAY(-1,1) 0.22% *** 2.54 0.24%*** 2.75 2-DAY(0,1) 0.27% *** 3.54 0.29%*** 3.75 1-DAY(0) 0.13% ** 2.35 0.13%** 2.43
***, **, and * indicate significance at the 1%, 5%, and 10% level, respectively.
Table 4
The Cumulative Abnormal Returns (CARs) for Acquirers Around the Merger Announcements for the Public-to-Public Portfolio
We report the CARs in percentage for the public-to-public portfolio in this table. We further separate the public-to-public portfolio by using whether these acquirers and targets belong to same conglomerate or not.
The results for same conglomerate portfolio are reported in panel A, and the results for different conglomerate portfolio are reported in panel B. For each portfolio, the CARs for 4 different event windows are reported, 5-day (day -2 to +2), 3-day (day -1 to +1), 2-day (day 0 to day +1), and 1-day (day 0), and for each event window the results using value- and equal- weighted indices are reported.
Panel A Same Conglomerate
Value-Weighted Equal-Weighted
N= 23 N= 23
CARs t-value CARs t-value
5-DAY(-2,2) 0.63% 0.59 0.79% 0.76
3-DAY(-1,1) 0.60% 0.68 0.72% 0.82
2-DAY(0,1) 0.16% 0.21 0.22% 0.3
1-DAY(0) 0.42% 0.61 0.48% 0.7
Panel B Different Conglomerate
N= 116 N= 116
CARs t-value CARs t-value 5-DAY(-2,2) -1.80% *** -2.93 -1.70%*** -2.83
3-DAY(-1,1) -1.60% *** -2.96 -1.60%*** -2.96 2-DAY(0,1) -1.40% *** -2.83 -1.30%*** -2.85 1-DAY(0) -0.90% ** -2.12 -0.90%** -2.15
***, **, and * indicate significance at the 1%, 5%, and 10% level, respectively.
Table 5
The Cumulative Abnormal Returns (CARs) for Acquirers Around the Merger Announcements for the Public-to-Private Portfolio
This table reports the CARs in percentage for the public-to-private portfolio. We separate the portfolio into two different subsets. One is that the purpose of these events are classified as merger and acquisitions (the Acquire Company portfolio, AC), the other is that the purpose of these events are to acquire assets from others (Acquire Assets Portfolio, AA). After separating these events into these subsets, we further separate them into two different portfolios by whether these acquirers and targets belong to same conglomerate or not. For each portfolio, the CARs for 4 different event windows are reported, 5-day (day -2 to +2), 3-day (day -1 to +1), 2-day (day 0 to day +1), and 1-day (day 0), and for each event window the results using value- and equal- weighted indices are reported.
Panel A Acquire Company (AC) portfolio
All Acquirer Same Conglomerate Different Conglomerate
Value-Weighted Equal-Weighted Value-Weighted Equal-Weighted Value-Weighted Equal-Weighted
N= 195 N= 195 N= 34 N= 34 N= 161 N= 161
CARs t-value CARs t-value CARs t-value CARs t-value CARs t-value CARs t-value 5-DAY(-2,2) -0.01% 0.00 0.09% 0.19 1.47% 0.81 1.36% 0.75 -0.30% -0.8 -0.20% -0.5 3-DAY(-1,1) 0.20% 0.56 0.27% 0.78 1.11% 0.77 1.01% 0.69 0.00% 0.01 0.12% 0.39 2-DAY(0,1) 0.38% 1.33 0.45% 1.54 1.29% 1.06 1.28% 1.03 0.19% 0.8 0.27% 1.14 1-DAY(0) 0.19% 0.74 0.22% 0.87 1.34% 1.09 1.35% 1.09 -0.06% -0.4 -0.02% -0.1 Panel B Acquire Assets (AA) portfolio
All Acquirer Same Conglomerate Different Conglomerate
N= 1553 N= 1553 N= 46 N= 46 N=1507 N=1507
CARs t-value CARs t-value CARs t-value CARs t-value CARs t-value CARs t-value 5-DAY(-2,2) 0.37%
***3.47 0.37%
***3.56 0.22% 0.42 0.06% 0.11 0.37%
***3.45 0.38%
***3.58 3-DAY(-1,1) 0.23%
***2.56 0.24%
***2.69 0.59% 1.26 0.48% 1.02 0.21%
**2.39 0.23%
***2.56 2-DAY(0,1) 0.26%
***3.29 0.27%
***3.42 0.60% 1.58 0.54% 1.39 0.25%
***3.09 0.26%
***3.25 1-DAY(0) 0.12%
**2.3 0.12%
**2.31 0.39% * 1.63 0.33% 1.41 0.11%
**2.09 0.11%
**2.14
***, **, and * indicate significance at the 1%, 5%, and 10% level, respectively.
Table 6
The Cumulative Abnormal Returns (CARs) for Acquirers Around the Merger Announcements Using Sample Period from 1994 to 1998.
In this table, we report the CARs in percentage for acquiring firms using data from 1994 to 1998.
There are a total of 1001 merger events during 1994-1998, and 76 events belong to the public-to-public portfolio, and 925 events belong to the public-to-private portfolio. In panel A, B, and C, we report the results for the all acquirer portfolio, the public-to-public portfolio, and the public-to-private portfolio respectively. For each portfolio, the CARs for 4 different event windows are reported, 5-day (day -2 to +2), 3-day (day -1 to +1), 2-day (day 0 to day +1), and 1-day (day 0), and for each event window the results using value- and equal- weighted indices are reported.
Panel A All Acquirer portfolio
Value-Weighted Equal-Weighted Panel B Public-to-Public portfolio
N= 76 N= 76
Panel C Public-to-Private portfolio
N= 925 N= 925
***, **, and * indicate significance at the 1%, 5%, and 10% level, respectively.
Table 7
The Cumulative Abnormal Returns (CARs) for Acquirers Around the Merger Announcements, Excluding One-Year Overlapping Events
In this table, we report the CARs in percentage for acquiring firms by using the sample that have corrected for the overlapping problem. We drop those events that were done by one firm in same year, and only keep the fist event for each firm in one year during the sample period. To be more specific, the time length between two events that were done by same acquirer must over one year or above. In panel A, we report the results using sample period from 1983 to 2007, and in panel B, we report the results using data from 1994-1998. The results for the public-to-public portfolio and the public-to-private portfolio are reported. Initially, we have total 1887 (1001) events from 1983-2007 (1994-1998), after we drop those overlapping events, we have a total of 542 (303) events.
For each portfolio, the CARs for 4 different event windows are reported, 5-day (day -2 to +2), 3-day (day -1 to +1), 2-day (day 0 to day +1), and 1-day (day 0), and for each event window the results using value- and equal-weighted indices are reported.
Original Results Overlapping dropped Results Panel A Using sample period 1983-2007
1. Public-to-Public portfolio
Value-Weighted Equal-Weighted Value-Weighted Equal-Weighted N=139 N=139 N= 59 N= 59 Panel B Using sample period 1994-1998
1. Public-to-Public portfolio
***, **, and * indicate significance at the 1%, 5%, and 10% level, respectively.
第三年 摘要
本文探討不動產投資信託外部公司治理的議題。以往的研究著重在不動產投資信 託購併後主併公司與被併公司股價行為與異常報酬原因的探討，然而對主併公司 的動機卻鮮少討論。本文發現公開發行的不動產投資信託購併公開交易公司時有 顯著負向異常報酬，然而如果被併公司為未上市則股價反應為正向。另外，規模 較大，獲利情形較好，沒有成長機會，舉債比例較低，較少分析師追蹤的公司頃 向從事購併活動。
關鍵字： 不動產投資信託、公司購併、成長機會、透明程度、槓桿