An Analysis of a Merger’s Profitability

Abstract

Introduction

In the last few years we have observed a revived surge in the number of mergers. They are more often observed in countries with better accounting standards. Companies undergo mergers for a number of reasons. The primary reason is the proper allocation of resources and thus, increasing cost efficiency. A small amount of research has been done in the past years analyzing the short and long term effects of mergers in creating value for the companies. It has been seen that most mergers result in failures but not much research has been done in analyzing the reasons behind it. My research would be based on few of the biggest mergers that have taken place in the last few years. Firstly, my analysis of a merger’s profitability would be based on standard event study methodology. It would take into account the return to shareholders. Secondly, it would also aim to provide evidence regarding the signalling theory and the synergistic and the agency views. This would be based on an in-depth analysis of various determinants such as the excess returns around the announcement of the merger bid and around the termination of the merger and the significant differences in the responses of firms attempting focusing versus diversifying mergers respectively. Lastly, this would be followed by a thorough analysis of the valuation effects of mergers. There have been varied views but no one conclusion has been reached. I would thus, like to investigate deeper into it

2. Literature Review and Hypotheses

Analysis revealed

My research concentrates on the effects of a focusing and diversifying merger on the abnormal returns around the announcement period of the merger. The study by Delong (1999) can be seen as an extension to my research. He based his research on evaluating the stock pricing behaviour of the bidder and the target in bank mergers. He further studied the abnormal returns according to the nature of the merger i.e. focusing or diversifying. In my analysis, I take into account only activity focused mergers, whereas Delong (1999) considers mergers which focus on both activity and geography. His results show an enhancement in value of a focusing firm of about 2 % to 3 % as compared to a diversifying firm irrespective of the time period. On investigating further, he found that the relative market size of the target to the bidder and the pre-merger performance of the targets show an apparent relationship with the cumulative abnormal returns so calculated.

Wong and Cheung (2009) analyse the changes in the stock prices of the bidding and target firms in Hong Kong, China, Taiwan, Singapore, South Korea and Japan, following a merger or an acquisition announcement. It can be seen from their analysis that such an announcement yields positive results for the bidding firms but does not prove to be very beneficial for the target firms. Their hypotheses considered the consequences of the mode of payment, the type of acquisition and the type of the target firm on the stock pricing. Out of them only the second variable seems to have a direct effect on the post announcement returns of the bidding firm.

Huang and Walkling (1987) conducted similar research by extracting a sample of acquisitions from the Wall Street Journal which consisted of all initial front-page acquisition announcements. But this, took into account slightly different variables as compared to the other analyses discussed above. They determined the effect of tender offers vs. mergers; cash offers vs stock offers and resisted offers vs. unresisted offers. Their analysis revealed higher abnormal returns for tender offers which were quite insignificant once the effect of the extent of resistance and the payment type were isolated from it. The deals which faced resistance during a merger or tender offer showed higher abnormal returns irrespective to the mode of payment. All the results obtained above were either insignificant or marginally significant, but the effects of the third variable i.e. the form of payment showed some concrete results. The cumulative abnormal returns obtained from cash offers were radically higher than those obtained from stock offer. This research carried out by Huang and Walkling gave quite a holistic overview of the effect of the announcement of an acquisition, as it took into consideration, variables which are affected by both the bidding and target firms decisions.

All the literature discussed earlier in this paper, has illustrated some positive effects of an acquisition for both the bidders and the target firms. But, the analysis conducted by Bruner (2001) showed a little variation to the above. It suggested that only the target shareholders draw upon the benefits of the acquisition. No such profitable return is observed for the bidding firms. But, the combined returns of the bidder and the target yield positive results.

The approach followed by Bruner differs significantly from most of the research discussed earlier. He measures the performance of a merger and acquisition based on the investors’ required returns.

After observing the abnormal returns of the acquirer around the announcement date of the merger as per my research, the most obvious next step would be to analyse the long term effects of the merger. Various studies have been done in order to rightfully determine the outcome .The study by Asquith (1983) showed drastic negative returns after about three years of the merger. One of the best analyses that I found was by Agrawal, Jaffe and Mandelkar (1992) in their paper ‘The Post-Merger Performance of Acquiring Firms: A Re-examination of an Anomaly.’ Their results are based on a thorough analysis of a number of mergers that took place from 1955 to 1987. They explored the effect of the size of the firm and its beta risk, and found a loss of 10 % in the total wealth of the acquiring firm, five years after the merger was completed. An attempt was also made to find the additional NPV which is not captured by the announcement returns analysis. But, it was seen that the modification of the market was similar for both the announcement and post merger analyses.

Cole et al (2006)

Investigate a number of unsuccessful mergers in order to determine if they create or destroy value for acquirers by using mainly two approaches. Their signalling approaches show that the value of the bidding firm is reduced by a large margin in the market, which is a form of a punishment for considering the acquisition of a low NPV project. They also find that horizontal mergers yield negative CAR.

Hypothesis 1: The average abnormal returns (AAR) yield positive results for all sub-periods in the event

Hypothesis 2: The Cumulative average abnormal returns (CAAR) yield positive results for all sub-periods in the event.

Hypothesis 3: The type of acquisition, kind of acquisition, the mode of payment and the type of target firms affects the value of the cumulative abnormal returns (CAR) around the announcement day t=0 in the event.

Take into account focusing and diversifying…

3. Methodology

We begin by classifying the effective sample into two categories – Focusing and Diversifying. The classification approach has been adopted by Mann and Sicherman (1991).This can be done by comparing the two-digit SIC Codes of the acquirer and the target firm respectively. If both the firms involved in a deal have the same two-digit SIC Code, it can be classified as a focusing acquisition, whereas, if both firms have different codes, it is classified as a diversifying acquisition.

Now we progress towards analysing the cumulative value created by a focusing and diversifying acquisition around the announcement date, using a standard event-study methodology described by MacKinlay (1997), Huang and Walkling (1987) and Wong and Cheung (2009). The Market Return Model is used in this case, to calculate the abnormal returns of the sample using a linear relationship between stock returns and market return.

Rit = αi + βiRmt + εit (1)

E (εit = 0) var (εit) = σεt2

Where:

Rit : Return on security i on day t

Rmt : Return on market portfolio on day t

εit : Zero mean disturbance term

αi, : expected value of the difference between Ri and βiRmt

βi : covariance between Rit and Rmt divided by the variance of Rmt

σεt2 : variance of the error term

We use the market model instead of the constant mean return model as it gives us a more accurate judgement of the effect of the event. This is true as it does not take into account the variation of the market return , thus, giving us more accurate abnormal returns.

( if any words remain add how to calculate rit and rmt)

In order to calculate the abnormal returns, we use the market model parameter estimates.

^ ^

ARit = Rit – (αi + βiRmt) (2)

Where:

ARit : the abnormal return for security i on day t

^ ^

αi and βi : estimates of αi and βi

In order to calculate the abnormal returns we use a maximum of 351 daily observations (Huang and Walkling, 1987). We start collecting data from t -300 to t +50 days, with t = 0 being the announcement date of the acquisition. These 351 days include non-trading days as well. In other words, we actually gather data from t -214 to t +36, taking only trading days into account. We use different time periods of an event for a complete comparative analysis of abnormal returns in each sub period which is described as below:

Event period : day t -10 → day t +30 (41 days)

Pre-announcement period : day t -10 → day t -2 ( 9 days)

Announcement period : day t -1 → day t 0 ( 2 days)

Post Announcement Period : day t +1 → day t +30 (30 days)

diagram

To analyse the effect of the event , we now calculate the average abnormal return (AAR) for all the securities for a time period t. AAR is the sum of all abnormal returns of firms on day t divided by N( the number of firms):

(3)

The t statistic, Ï•, is calculated by dividing AARt by the standard deviation of the average abnormal returns. This is final step of the model, which helps in determining the significance of the AARt in the event period.

(4)

While calculating the standard error, an estimator is used to calculate the variance of the abnormal returns in the absence of but in this case we use the sample variance measure of that we derive from the market model regression. The estimator is as follows:

In order to establish a more holistic viewpoint, the cumulative average abnormal returns (CAAR) are calculated:

Where T1 to T2 is the duration of the event in which the AARt is collected.

According to our hypotheses we have to calculate one more variable, the cumulative average abnormal return (CAAR) over a certain period. In order to find out the significance of CAAR we calculate its t statistic as follows:

Where var(CAAR) is the variance of the cumulative average abnormal returns.

We could use a variety of formulas to calculate the standard deviation and t statistic such as those described in Campbell, Lo and MacKinlay (1997) and Brown and Warner (1985). But we calculate using the method adopted by Kothari and Warner (1985):

Where:

: Variance of the average abnormal return for one period.

L : Longer the L, the higher is the variance of CAAR

To test the third hypothesis, another variable is taken into consideration – the Cumulative abnormal returns (CAR). We now develop a regression model using dummy variables to test the effect of the type of acquisition, kind of acquisition, the type of the target firm and the mode of payment on the CAR of the acquirers. The control variables are the relative market size of the market value of the target to acquirer (RMV) and the market size of the acquiring firm (M) (Wong and Cheung, 2009).

Where:

: Cumulative abnormal return from day d1 → day d2

D1 : 1 if the type is acquisition

D1 : 0 otherwise i.e. merger

D2 : 1 if it is focusing

D2 : 0 otherwise i.e. diversifying

D3 : 1 if target firm is private

D3 : 0 otherwise i.e. public

D4 : 1 if mode of payment is cash

D4 : 0 otherwise i.e. stock

M : Market Value of the acquiring firm=Number of outstanding share *closing price on the announcement date

The tests of hypotheses 1, 2 and 3 can be described as the following tests:

H1 : H0 : AARt = 0

H1: AARt ≠ 0

H2 : H0 : CAARt = 0

H1: CAARt ≠ 0

H3 : H3i :β1 = 0 (Acquisitions vs. Mergers)

H3ii :β2 = 0 (Focusing vs. Diversifying)

H3iii :β3 = 0 (Public vs. Private target firms)

H3iv :β4 = 0 (Cash offer vs. Share offer)

4. Data Description

The number of mergers and acquisitions carried out in India has been quite extensive. Hence, certain criterion has been used to select a suitable sample.

The deals carried out with Morgan Stanley, JP Morgan, Goldman Sachs, UBS, Deustche Bank and Citi as their financial advisors should be included. These banks have been chosen as they deal with high valued mergers which are perfect for highlighting the true effects of a focusing or diversifying merger.

All deals should have been completed from January, 2003 to March, 2010. The sample consists of only 178 completed transactions.

All the acquirer firms must be publicly listed in the Bombay Stock Exchange.

The SIC Codes for the target and acquirer should be available in the CRSP Database. This helps in dividing the sample into focusing and diversifying mergers.

Because of these restrictions, the sample reduces to 70 firms, three of which have some information missing regarding the stock returns etc and hence our effective sample is 67. It has been further classified into 44 focusing and 23 diversifying deals.

Using only publicly listed firms enables us to extract information about these deals such as – announcement dates, termination dates, stock returns, market returns etc. from the Thomson One database, company websites and the Bombay Stock Exchange.

The Bombay Stock Exchange Sensitivity Index or the BSE Sensex (30) has been used to gather the market returns of the firms.

 

Professor

You must be logged in to post a comment