7. CONCLUSIONS AND RECOMMANDATION
5.1 A NALYSIS FOR M ERGER I MPACTS IN THE S HORT - TERM AND THE C ONNECTION TO R ELATIVE
To prove the arguments of the short-term post-merger performance within the first quarter after mergers, two analyses are done in this section. Section 5.1.1 presents the percentage change analysis using event studies. Section 5.1.2 presents the results of regression analysis.
5.1.1 Percentage Change Analysis
When analyzing the probable merger impacts in the short-run (the 1st quarter after mergers), it is to calculate the change of performance to the average performance in the estimation windows in both market and operating assessments, as discussed in chapter 3. When analyzing the impact by using market assessment, change in stock returns will be divided by average stock returns in the estimation windows. On the other hands, when analyzing the impact by using operating performance assessment, change in ROA and ROE will be divided by average ROA and ROE in the estimation
windows.4
In the short-term size impact analysis, there are 58 samples in total, excluding a few outliers. Two parts are included in: absolute size analysis and relative size analysis.
Absolute Size Analysis
In the analysis of absolute size, it was broken into five different groups, the whole sample, the first, the second, the third, and the fourth quartiles. The quartiles are arranged by absolute sizes of acquiring companies from small to large in order. The analysis for merger impacts in the short term is presented in table 5.1 below.
Table 5.1
Percentage Change in Stock Returns and Operating Performance for the First Quarter after Mergers- Arranged by Absolute Sizes of Acquiring Companies
Mean Percentage change of Post-merger Performance for the First Quarter Quartile Group Stock Return ROA ROE
Total Sample -81.45% -42.66% -61.22%
(N=58) (-1.4140)*** (-1.3986)*** (-1.4079)***
1st Quartile -145.44% -37.49% -118.09%
(N=14)) (-1.1037)*** (-0.2929) (-0.5003)*
2nd Quartile -29.50% -52.07% -58.11%
(N=15) (-0.1265)* (-0.3959) (-0.5030)*
3rd Quartile -104.13% -40.83% -35.10%
(N=15) (-1.0395)*** (-0.4230)** (-0.2848)
4th Quartile -40.55% -40.75% -31.38%
(N=14) (-0.3075) (-0.3574) (-0.3002)
***,**, * Statistical significance in 2-tailed tests at 1%, 5%, and 10% level, respectively.
Source: Author
4 Operating Cash Flow Return (OCFR) discussed in chapter 2 is not used in this dissertation due to the data incompleteness for large number of firms in COMPUSTAT.
The whole-sample analysis was shown in the first line in table 5.2. The mean percentage changes for stock returns, ROA, and ROE are -81.45%, -42.66%, and -61.22% respectively, all significantly negative.
According to table 5.1, when the analysis was broken into four different quartiles corresponding to the absolute of the acquiring firms, most of the performances are still significantly negative. It is especially can be seen that stock returns of the first quarter after mergers for most groups are significantly negative. On the other hands, for ROA and ROE, though the average percentage change for ROA and ROE are mostly negative, there is no significant evidence to prove the results.
In spite of the most insignificance for all data, the bar chart in figure 5.1 below can still help observe the percentage change for those three performance measurements.
For stock returns and ROA, there is not clear phenomenon observed between different quartiles. However, for ROE, the percentage changes get less negative, from -118.09%, -58.11%, -35.10% to -31.38%, when absolute sizes of the acquiring firms increase. This phenomenon shows that in the first quarter, with the larger size of acquiring companies, the percentage change of ROE gets less negative.
-160.00%
Figure 5.1 Bar chart of merger impact as percentage change of performance
for the absolute size groups from the first to the fourth quartileSource: Author
Relative Size Analysis
In the analysis of relative size, similarly, it was broken into five different groups, the whole sample, the first, the second, the third, and the fourth quartiles. The quartiles are arranged by relative size between combining firms from small to large in order.
The analysis for merger impacts in the short term is presented in table 5.2 below.
Table 5.2
Percentage Change in Stock Returns and Operating Performance for the First Quarter after Mergers- Arranged by Relative Sizes of the Two Combining Companies
Mean Percentage change of Post-merger Performance for the First Quarter Quartile Group Stock Return ROA ROE
Total Sample -81.45% -42.66% -61.22%
(N=58) (-1.4140)*** (-1.3986)*** (-1.4079)***
1st Quartile -88.80% -15.30% -33.52%
(N=14)) (-0.6426)** (-0.3660) (-0.1310)*
2nd Quartile -62.83% -57.45% -62.61%
(N=15) (-0.3924)* (-0.3316) (-0.3908)*
3rd Quartile -125.14% -56.75% -68.58%
(N=15) (-0.9215)** (-0.5301)** (-0.6362)**
4th Quartile -47.24% -20.86% -97.78%
(N=14) (-0.2338) (-0.3309) (-0.4437)*
***,**, * Statistical significance in 2-tailed tests at 1%, 5%, and 10% level, respectively.
Source: Author
According to table 5.2, when the analysis was broken into four different quartiles corresponding to the relative size between combining firms, most of the performances are still significantly negative. The bar chart in figure 5.2 below helps observe the percentage change for those three performance measurements more clearly. For stock returns and ROA, there is not clear phenomenon observed between different quartiles, but it can be observed that for stock returns, the fourth quartile has the smallest negative merger impact; while for ROA, the first quartile has the smallest negative merger impact.
However, for ROE, it can be seen that all percentage changes in ROE for different quartiles are all significantly negative; moreover, the percentage changes get more negative, from -33.52%, -62.61%, -68.58% to -97.78%, when relative size increases.
This phenomenon shows that in the first quarter, with increasing the size of target companies, the percentage change of ROE gets more negative.
-140.00%
Figure 5.2 Bar chart of merger impact as percentage change of performance
for the relative size groups from the first to the fourth quartileSource: Author
5.1.2 Regression Analysis
Multivariate OLS analysis for the first-quarter percentage changes of stock returns, ROA, and ROE are presented in this section. Panel A in Table 5.3 reports the descriptive statistics of the variables included in the regression model and Panel B presents the results of the regression analysis.
The significance of the association between independent variables and dependent variables are only found in change in stock returns, where absolute size of the acquiring companies and relative size of combining firms are found to be significantly and positively associated with the percentage change of stock returns. The results may be due to the large standard deviation and to an unsystematic relation between the dependent and the independent variables.
Table 5.3
Multivariate OLS Analysis for the First-quarter Percentage Changes of Stock returns, ROA, and ROE
Panel A: descriptive statistics of variables
Average Median Max. Min. S.D. Sum
△Stock Returns -81.45% -116.51% 585.79% -384.49% 159.34%
△ROA -42.66% -33.18% 303.46% -345.39% 115.12%
△ROE -61.22% -39.97% 331.99% -832.18% 155.58%
Absolute Size (Ln) 8.20 8.25 12.7 3.72 1.93 Relative Size 25.86% 12.75% 231.57% 0.05% 43.10%
B_T_M (Acquirers) 41.01% 39.44% 152.16% 4.26% 28.50%
CONGLOMERATE
dummy 37
CASH dummy 11
STOCK dummy 46
Panel B: regression results
Market Assessment Operating Performance Assessments Independent Variables △Stock Returns % △ROA % △ROE %
Intercept -3.6986 0.8897 -1.3029
(-2.8663)*** (0.9001) (-0.9449)
Absolute Size of Acquirer (Ln) 0.2808 -0.1468 0.0760
(2.0020)** (-1.3657) (0.5071)
Relative Size (St/Sa) 2.0600 -0.4514 0.0815
(3.2324)*** (-0.9246) (0.1197)
B-T-M (Acquirers) 0.0860 -0.0037 0.0774
(0.1116) (-0.0063) (0.0940)
CONGLOMERATE dummy -0.17551 0.3218 0.2984
(-0. 5815) (1.3959) (0.9252)
CASH dummy -0. 4528 0.9614 0.9656
(-0.37375) (1.0359) (0.7459)
STOCK dummy -0.0663 1.0826 1.0879
(-0.0594) (1.2668) (0.9127)
F(t-test) 2.1032** 0.7964 0.2578
Adjusted
R
2 0.1057 0.0066 0.07451. N=58, the explanation of variables is in section 4.3 of this dissertation.
2. ***,* Statistical significance in 2-tailed tests at 1%, 5%, and 10% level, respectively.