Measuring the Economic Gains of Mergers and Acquisitions: Is it Time for a

Change?

Antonios Antoniou
Durham Business School, University of Durham, Durham DH1 3LB, UK
Philippe Arbour
Lloyds TSB Bank, 25 Gresham Street, London EC2V 7HN, UK
Huainan Zhao*
Faculty of Finance, Cass Business School, London EC1Y 8TZ, UK

September 2006

Abstract

In this paper we investigate the issue of measuring the economic gains of mergers and
acquisitions (M&A). We show that the widely employed event study methodology,
whether for short or long event windows, has failed to provide significant insight
regarding the central question of whether mergers and acquisitions create value. We
believe the right way to assess the success and therefore desirability of M&A is
through a thorough analysis of company fundamentals, which implies examining
smaller samples of transactions with similar characteristics.

JEL classification: G14; G34.

Keywords: Mergers & Acquisitions, Economic Gain, Value Creation, Event Study,
Fundamental Analysis.

*
Corresponding author: Tel: +44-(0)20-7040-5253; fax: +44-(0)20-7040-8881.

Email address: h.zhao@city.ac.uk

1. Introduction

The development of the market for mergers and acquisitions (M&A) has gone hand in

hand with that of world capital markets. The corporate landscape is perpetually being

altered by M&A transactions. On a macroeconomic level, mergers come in waves.

The most notable M&A wave was that of the 1990s. During this period, deregulation,

a booming world economy combined with high equity prices and solid growth

prospects propelled the economic significance of M&A to a new height. Indeed,

during the late 1990s, the size, volume and frequency of M&A surpassed anything the

world had ever seen before. On a microeconomic level, mergers represent massive

asset reallocations within or across industries, often enabling firms to double in size in

a matter of months. Adding to the picture that mergers tend to occur in waves and

cluster by industry, it is easily understood that such transactions may radically change

the competitive architecture of the affected industries. It should therefore come as no

surprise that academics have been so intrigued by the merger debate in recent years.

Examining the economic gains (value creation or destruction) of M&A is one of the

most coveted research areas in financial economics. In fact, the spectacular growth of

mergers has prompted many academics and practitioners to investigate whether such

transactions are worth undertaking. More specifically, researchers have sought to find

out whether M&A create or destroy value and how the potential gains or losses are

distributed among transaction participants. Certainly, if potential synergies truly exist,

M&A should therefore have the potential of being value-creating transactions. This

particular question is of utmost importance as its corresponding answer carries

important policy implications for regulators. Furthermore, it is vital to objectively

1
assess the resulting aftermath of such colossal transactions, as lessons learned may

benefit society as a whole.

Although a plethora of research in financial economics has sought to address the issue

of M&A value creation, the examination of the issue from a company fundamentals

standpoint has largely been ignored. The bulk of the existing literature employs event

study methodology as instigated by Fama, Fisher, Jensen, and Roll (1969), which

examine what kind of impact, if any, mergers and acquisitions have on stock prices.

Very simply, a merger is branded successful if the combined entity equity returns

equal or exceed those predicted by some standard benchmark models. For reasons

argued below, this simplistic approach too often leads to a Type II error (i.e., the null

hypothesis is not rejected when in fact it is false and should be rejected) with respect

to the null hypothesis that M&A are value creating transactions. We invite readers to

review the evidence and arguments presented in this article and to judge whether the

event study is an appropriate tool for measuring the economic gains resulting from

mergers on an ex-post basis. We would like to stress here that this article does not

constitute an attack on event study in general, but rather to applying event study to

answer the specific question of whether or not M&A yield economic gains.

2. Short-window Event Study

From a short-run perspective, the most commonly studied event window encompasses

the three or five days surrounding a merger announcement. From a theoretical

standpoint, and in the context of an efficient market, changes in stock market value

around merger announcements should fully capture the economic gains from merging.

2
Researchers to this end have unanimously reported that target firm shareholders enjoy

a significant positive cumulative abnormal return (CAR) around the merger

announcement. These findings, however, mean very little as they should be expected.

Intuitively, target shareholders expect to receive a premium if they are to hand over

their stakes to the acquiring firm. It should therefore come as no surprise that positive

CARs accrue to target firm shareholders during the period surrounding merger

announcements. Put differently, no matter who throws money at target firm

shareholders, the CARs earned by the latter will invariably be positive as long as a

positive premium is offered.

However, the effect of merger announcements on acquiring firms’ share prices is far

from clear. On the one hand, some studies have found that no or small significant

positive abnormal returns accrue to acquiring firm shareholders around merger

announcementsi. On the other hand, others have reported that acquirers experience

significant but small negative abnormal returns during the same periodii. In short, the

general picture that emerges is that, from a stock return standpoint, M&A are clearly

more beneficial to target shareholders than their respective suitors, a fact that is

widely recognized in the literature.

Using a weighted-average approach based on firm size as measured by market

capitalization allows for assessing how combined (target and acquirer) stock returns

fair in the same event window. Many believe that this type of study enables us to

evaluate the economic gains of M&A from a net aggregate perspective, thereby

yielding valuable insight regarding the macroeconomic consequences (benefit or

detriment) of M&A. That is, it is believed that this perspective enables us to assess

3
whether or not M&A produce economic gains or whether they simply involve a

wealth transfer from one entity to the other (i.e., a zero-sum game). Overall, the

literature concurs that the combined entity earns a positive CARiii, albeit small,

around the merger announcement. But are conclusions from such studies sufficient to

draw an inference about the true value creation potential or desirability of M&A?

Many believe so. Andrade, Mitchell, and Stafford (2001) refer to short-window event

studies as: “The most statistically reliable evidence on whether mergers create value

for shareholders…”; they go on to conclude that: “Based on the announcement-period

stock market response, we conclude that mergers create value on behalf of the

shareholders of the combined firms”. This however, is surely a premature conclusion

as the hefty premiums paid (which in many cases may turn out to be overpayments

especially under agency or hubris motives) mask the true economic impact of the

transactions under analysis. That is, the premiums offered to target shareholders

distort or bias weighted average return calculations. Let us now explain in closer

detail why these types of studies demonstrate very little with respect to the M&A

value creation issue.

First, examining stock price movements around the merger announcement tells us

little about the sources of economic gains that arise from combining the target and the

acquirer. Shelton (1988) writes: “value is created when the assets are used more

effectively by the combined entity than by the target and the bidder separately”.

Hence, examining short-window stock returns does nothing to test this statement.

Indeed, event study conclusions rely strictly on the assumption of market efficiency.

However, it is possible that investors systematically over or under react to merger

announcements, which would result in stock prices temporarily deviating from their

4
fundamental levels. If this is shown to be the case, the event-study’s ability to

distinguish real economic gains from market inefficiency is compromised. As Healy,

Palepu, and Ruback (1992) put it: “From a stock price perspective, the anticipation of

real economic gains is observationally equivalent to market mispricing”. Indeed, the

mounting body of behavioural finance literature illustrates the need to approach short-

run event study results with scepticism. Furthermore, there is plenty of evidence that

suggests that stock market wealth may temporarily move independently of

fundamentals and the dot.com stock market bubble of the late 1990s is an irrefutable

example of this fact. If the latter premise is accepted, the short-window event study’s

ability to measure real economic gains is compromised.

Further, the short-window event study approach still remains problematic even in the

context of efficient markets. It is important to reiterate that target firm shareholders

generally receive a substantial premium in exchange of their shares. As mentioned

earlier, the premium paid (especially under agency or hubris motives) severely

exacerbates target real returns earned around the merger announcement, thereby

creating an artificially high weighted-average CAR (WACAR). To be more specific,

weighted average calculations are almost sure to generate a positive result when

premiums are offered to target firm shareholders. We therefore believe that examining

combined-entity CARs around merger announcements does not further our

understanding of the M&A value creation debate.

This issue can be illustrated with a simple example. Assume that a tender offer is

made by ‘Bidder Inc.’ for the acquisition of ‘Target Inc.’ One month prior to the

takeover announcement, the market value (MV) of Bidder Inc is $1,000,000, with

5
500,000 common shares outstanding and Target Inc.’s market value is $100,000, with

100,000 common shares outstanding. Therefore, the relative size of the target to the

bidder is 10%, which implies that weights in calculating the WACAR for the

combined entity would be Wbidder = 90.9% and Wtarget = 9.1% respectivelyiv. Now

suppose that the acquirer announces a cash offer at $1.40 per share, which represents

a premium of 40% per share purchasedv. In a relatively efficient market, the price of

the target’s shares will adjust to the offer price quickly. Hence, the market price of

Target Inc’s shares should shoot up to the $1.40 range in the three days surrounding

the announcement. Now suppose that the benchmark model predicts the expected

return of Target’s shares should be 2% for the three days surrounding the merger

announcement, this necessarily implies that the three-day CAR for the target could

reach up to 38%vi. Because the target in this example is relatively small relative to the

acquirer, and the method of payment is cash, it is not unreasonable to assume that the

acquirer should earn the expected rate of return (i.e. 0% excess return) in the three

days surrounding the announcement. Thus, on a weighted average net aggregate basis,

the combined entity’s CAR (WACAR) would be around 3.458 %vii.

Changing the example slightly, assume that Bidder Inc’s shareholders do not share the

same optimism regarding the union as they believe that their management is paying

too much to acquire Target Inc. Thus, shareholders may decide to sell Bidder Inc.’s

shares, which may result in the acquiring firm earning a negative CAR around the

merger announcement (Assume –2%). Using the same weights and premium as

described above, the WACAR for the combined entity will still be approximately

1.6%, thereby leading to the (dubious) conclusion that the acquisition was value-

creating overall. Indeed, in this example, the premium included in the tender for

6
Target Inc.’s shares could be as low as 22% and the transaction would still be

considered value creating regardless of the negative CAR earned by the acquirer. This

illustrates what we call the premium exacerbation problem.

In reality, Moeller, Schlingemann, and Stultz (2003) report that the mean premium

paid for over 12,000 US takeover transactions with announcement dates between

1980 and 2001 was 68% for large firms and 62% for small firms, which necessarily

implies that, according to our illustration, most (if not all) M&A transactions

evaluated using a short-window event study approach will be branded as being

successful, or value creating, in spite of the negative returns to acquiring firms. That

is, the WACAR is almost invariably positive. Furthermore, the problem is severely

compounded as the relative size of the target to the acquirer increases.

Indeed, premiums offered may easily represent overpayments (Roll, 1986). During

times of high M&A activity, firms that are potential targets are likely to carry an

important takeover premium in their stock prices. Hence, during M&A waves,

acquiring firms are likely to pay a premium on top of what may turn out to be an

already excessively high stock price. Ironically, the WACAR approach to evaluating

M&A rewards overpayments, which systematically leads to the conclusion that M&A

are invariably value-creating transactions. But it is widely believed that overpaying

may be a leading cause of deal failures. Besides, the controversy surrounding the

desirability of M&A suggests that these results must be viewed with great scepticism.

Moreover, Mitchell, Pulvino, and Stafford (2004) recently examine price pressure

effects around merger announcements and find that on average, acquirers earn a

significant negative abnormal return of –1.2% in the three days surrounding the

7
announcement. However, they also find that a substantial proportion of this negative

announcement period abnormal return is due to merger arbitrage short-selling, rather

than information, thereby contradicting the premise that stock returns solely reflect

value changes or expectations. After controlling for price pressure effects, acquirers’

announcement period abnormal return increases to –0.47% and becomes statistically

insignificant. Their findings demonstrate that conventional short-run event studies in

M&A can produce poor estimates of shareholder wealth effects. That is, stock returns

reflect more than investor expectations regarding the desirability of mergers taking

place. The implication here, once again, is that relying on short-window event studies

to gain insight on the M&A value creation issue is a potentially dangerous practice.

In their most recent study, Moeller, Schlingemann, and Stulz (2004) examine

acquiring firm returns in recent merger waves. In addition to gauging abnormal

percentage returns, they also measure aggregate dollar returnsviii. Strikingly, they find

that between 1998 and 2001, acquiring firms’ three-day announcement period (day –

1, 0, and +1) average CAR is 0.69%, while the aggregate dollar return measure

indicates that acquiring firm shareholders lose a total of $240 billion over the same

three-day period. Also, upon further investigation, they find that the losses to

acquirers exceeded the gains to targets, resulting in a net aggregate dollar loss of $134

billion during the same window. These findings provide interesting but rather painful

evidence that if we merely rely on the short-run event study result (i.e., the three-day

CAR 0.69%), we will unavoidably conclude that the sample merger transactions are

value creating, regardless of the troubling assertion that acquiring firm shareholders

suffer a massive loss of $240 billion and that targets and bidders together sustain a net

loss of $134 billion during the same three-day period. The latter authors also show

8
that between 1980 and 2001, the average three-day announcement period CAR for

acquirers is positive every year except for 2 out of 21 years; however, the three-day

aggregate dollar returns are negative for 11 out of 21 years. Once again, the three-day

CARs tell us mergers create value for acquirers in almost every year between 1980

and 2001 regardless the massive dollar losses realized in half of the period. In short,

Moeller, Schlingemann, and Stulz’s research is very important in that it illustrates just

how unreliable short-window event study results are in explaining whether or not

M&A really yield economic gains.

Thus far, we have shown that according to short-window event studies, the results are

quite clear: mergers and acquisitions are value-creating transactions. But if this

conclusion is so clear cut and obvious, then why is there so much controversy

surrounding the desirability of M&A? That is, why do event study results stand in

such sharp contrast with the growing rhetoric that creating value through M&A is

easier said than doneix. It is widely recognized that many recent mergers have proven

to be total disasters. Even consultancy firms, which derive important fees in advising

companies on M&A issues, have documented the widespread nature of these failuresx.

Academic studies have also recognized the high divestiture rate subsequent to

acquisitions taking placexi. If mergers are truly value creating transactions due to real

economic gains and not market mispricings, it is highly unlikely that firms would

divest acquisitions made at such a high frequency, nor would there be so much

controversy surrounding the desirability of M&A.

Hitherto, we have shown that it is counterintuitive to conclude that M&A are value-

creating transactions based on the prevalence of positive abnormal returns, especially

WACARs. In many cases, target firm shareholders may have been the only ones who

9
gained anything from the transactions, and possibly to the detriment of acquiring

firms. But on an ex-post basis, examining how target shareholders fare has very little

relevance to examining the economic gains from M&A. Undeniably, at the merger

announcement and the few days surrounding it, we know little about any future

possible negative drifts in the acquirer’s stock prices, nor do we know whether

acquiring firm managers succeed at unlocking synergies with the target firm. Indeed,

it would be unreasonable to conclude anything at all from these short-window event

study results apart from the fact that target firm stock prices react rather quickly to

tender offers made and that weighted-average calculations (i.e., WACAR) are almost

invariably positive regardless of the actual fate of the combined firms. It thus becomes

clear that it is less relevant and perhaps even counterintuitive to actually include the

target into event-study merger calculations when investigating the question of value

creation as we are much more concerned about the firm that makes the investment and

ultimately carries on: the acquiring firm.

3. Long-window Event Study

A second strand of the literature examines the long-run post-merger stock

performance of the combined entity. We believe that long-run event studies also lack

effectiveness for tackling the issue of M&A value creation for the following reasons.

First and foremost is the methodological problem associated with long-run event

studies. For instance, bad model problems imply that it is not possible to accurately

measure expected returns, thus rendering futile the analysis of long-run abnormal

returns. In addition to the bad model problem, a number of researchers have most

10
recently pointed out that the process used in calculating and testing the long-run

abnormal returns is itself biased. For example, recent papers by Barber and Lyon

(1997) and Kothari and Warner (1997) address misspecification problems in long-

horizon event studies. They argue that commonly used methods for testing for long-

run abnormal returns yield misspecified test statistics, and invite researchers to be

extremely cautious in interpreting long-horizon test results.

In order to eliminate bad model problems and biased test statistics, Barber and Lyon

(1997) and Lyon, Barber, and Tsai (1999) advocate the use of a single control firm or

a carefully-constructed reference portfolio approach. The idea is to select a matching

firm or a portfolio of matching firms that have approximately the same size (MV) and

book-to-market ratio (BE/ME) as the sample firms. This approach has been shown to

eliminate some well-known biases and leads to a better test statisticxii. Although some

recent M&A studies have applied this updated approach, it remains problematic for

reasons discussed below.

Until the adoption of Statement of Financial Accounting Standard (SFAS) 141 in

2001 in the US, mergers were accounted for using either the purchase or the pooling

method, therein creating a hurdle in selecting control firms. First, the pooling method

consolidates target and acquiring firm accounts from the beginning of the year,

regardless of when the merger is actually completed; the purchase method on the

other hand, consolidates accounts at merger completion. Consequently, post merger

book values (common stock + retained earnings) will be larger under the pooling

method than those under the purchase for cash-based transactions because the

operating results are added together regardless of when the merger is actually

11
completed, thereby increasing retained earnings for the period. Under the purchase

method, target and acquiring firm operating results are only added together from the

date of the transaction completion onwards. Therefore, an acquirer that has an

accounting year end of Dec. 31st, but that manages to complete an acquisition on June

30th of that year will only consolidate half the target’s operating results in its year-end

financial statements under the purchase method. Furthermore, under the latter method,

mandatory inventory write-ups for firms using a last-in-first-out (LIFO) cost flow

assumptions will increase reported cost of goods sold (COGS) and further depress

retained earnings for the period. Similarly, retained earnings are also depressed by in-

process research and development (IPR&D) write-offs which are once again

mandatory under US generally accepted accounting principles (GAAP). The picture

that emerges is that one cannot bundle deals that were accounted by using different

merger accounting methods into the same sample.

Another problem is that even under the same method (i.e., the purchase method) the

resulting equity account of the combined firm will vary depending on whether the

acquisition is financed with cash or with stock. In short, resulting combined book

equity values should generally be higher for stock-financed transactions than for

acquisitions paid in cash.

Thus, according to the above arguments, the equity book value of the merged firm

will generally be smaller under the purchase method, and even smaller for cash

financed transaction under the same method, thereby illustrating that the selection of

control firms (i.e., the expected returns) becomes tricky and potentially flawed. This

in turn compromises the calculation of long-run abnormal returns, as well as the

12
cross-sectional comparability of results. Although merger accounting intricacies go

beyond the scope of this article, it is easily understood that such issues must be

carefully analysed when applying popular long-run event study methodology. But the

literature has largely failed to control for these key issues.

For instance, the above merger accounting discussion also invalidates the use of the

so-called state-of-the-art bootstrap approachxiii advocated by Ikenberry, Lakonishok,

and Vermaelen (1995), Kothari and Warner (1997), and Lyon, Barber, and Tsai

(1999) and applied by Rau and Vermaelen (1998) in mergers and acquisitions. Indeed,

the 1000 pseudo-portfolios matched in size and book-to-market ratio at the time of

merger completion do not control for the aforementioned accounting issues, thereby

nullifying the validity of obtained matches and thus the empirical distribution of

abnormal returns generated under the bootstrapping approach.

The bad news, however, is that even if we control for differences between the pooling

and purchase methods in matching the reference firms and all other possible sources

of misspecificationxiv, we are still far from obtaining an accurate and reliable long-run

test result. In one recent attempt, Lyon, Barber, and Tsai (1999) recommend two

general approaches that control for common misspecification problems in long-run

event studies. However, their simulation results show that, despite all efforts and

intentions, well-specified test statistics (i.e., where empirical rejection levels are

consistent with theoretical rejection levels) are only guaranteed in random samples

while misspecified test statistics are pervasive in non-random samples. But we know

that mergers and acquisitions cluster in time and by industry, which necessarily

implies that well-specified test statistics in long-run M&A event studies should hardly

13
exist. The central message in their study, however, is that: “the analysis of long-run

abnormal returns is treacherous”.

Second, on a more general framework, Viswanathan and Wei (2004) provide a

mathematical proof that the usual abnormal return (CAR/BHAR) calculated in event

studies has a negative expectation. They prove that, in any finite sample, the expected

event abnormal return will invariably be negative and becomes more negative as the

event window is lengthened. Coming back to M&A, we thus understand that long-run

negative abnormal returns will be observed for most finite samples. The implication

of utmost importance here is that these negative results do not discriminate between

successful or unsuccessful mergers; that is, the observed abnormal return will be

negative no matter what the fate of the combined firm. This would explain why a vast

majority of studies have reported a long-run underperformance of acquiring firms.

This further highlights the event study’s failure in resolving the M&A value creation

debate.

In addition, Viswanathan and Wei go on to examine the above problem in infinite

samples. They prove that, asymptotically, the event abnormal return converges to zero

and hence conclude that the negative long-run event abnormal return is simply a small

sample problem. This again offers a good explanation as to why some M&A studies

using very large samples have reported insignificant results. If the small sample

problem is the long-run event study’s only flaw, one can easily get around this by

increasing the sample size. But what would such studies contribute to our

understanding of whether M&A yield economic gains? By averaging abnormal

returns across a near-infinite number of cross-sectional observations, we end up with

14
a normal distribution of abnormal returns with a mean of zero. Therefore, nothing can

be concluded from this result apart from the fact M&A have a 50/50 probability of

creating or destroying value. Surely, this type of conclusion is of no use to regulators

or investors. We believe that the value creation potential of M&A represents more

than a crapshoot. However, when using large data samples in empirical research, all

interesting insight to be gained about mergers and acquisitions appears to cancel out.

Finally and perhaps most importantly, the recent development of a series of new

methodologies has given rise to a new wave of long-run event studiesxv. Mitchell and

Stafford (2000), for instance, reexamine the long-run anomaly associated with

corporate takeoversxvi. Their results suggest that acquirers (combined firms) earn a

fair rate of return over the long run, thereby implying that mergers do not create nor

destroy value, an idea consistent with that found in latter paragraph. But we know that

stock price is forward looking and the general idea here is that in a relatively efficient

market, the price of an asset should reflect the present value of the underlying asset’s

future cash flows. Very simply, this means that the observed price of an asset should

reflect (discount) what is expected about the future today. Therefore, any longer-term

price movements discount the future, which may extend to anticipated events that

reach far beyond the merger or acquisition under analysis. For example, stock returns

occurring 5 years subsequent to an M&A transaction (t+5) should reflect only the

changes of estimation/expectation in the cash flow generating ability of that firm in

years t+6, t+7, t+8 etc. But this extends far beyond the actual impact of the M&A

transaction that occurred in year t. Furthermore, various confounding events may

ensue within the 5 years following the merger. Hence, anticipated future events will

be incorporated into the combined firm’s stock price, thereby undermining the

15
researcher’s ability to isolate and quantify the benefits that occurred as a result of the

original event (the merger or acquisition under analysis). All in all, long-term stock

return analyses fail to provide a means of identifying and isolating the effects of the

actual merger that has taken place and thus do not provide much insight about the

micro or macro economic impact of M&A.

In light of the arguments presented above, we therefore believe that event study

methodology, for both short and long event windows, falls short of proposing an

economically sound tool for measuring merger performance on an ex-post basis. The

picture that emerges is that, in attempting to resolve the M&A value creation

quagmire, financial economists have overindulged in event studies, which have

largely yielded biased and thus unreliable results. The failure of the event study to

resolve the question of whether M&A represent desirable transactions leads us to

approach this topic from a different perspective: that of company fundamentals.

4. Fundamental Analysis

‘Unlocking synergies’ is the most commonly cited managerial motivation for

undertaking M&A (Walter and Barney 1990). As such, if synergies truly exist,

economic gains from mergers should thus show up in the combined firm’s

fundamentals. Coming back to Shelton’s definition of value creation, we see that the

concept has less to do with share price movements, and more to do with asset

reorganizations, managerial performance and the efficient utilization of these

reorganized assets. Jarrell, Brickley, and Netter (1988) postulate that gains to

shareholders must be real economic gains via the efficient rearrangement of resources.

16
In other words, consistent with basic finance principles, fundamental improvements

should drive capital gains. As mentioned earlier, one major problem with event

studies is that stock market wealth may change independently of fundamentals. Thus,

the analysis of the desirability of M&A begins not with stock returns, but with the

underlying factors that power these returns.

In addition to short and long-window event studies, there is a small body of literature

that examines pre and post-merger operational performance. The objective of such

studies is to look for improvements in company fundamentals in order to assess

whether a merger is value creating or not. Roughly speaking, M&A labelled

successful or value-creating should exhibit some form of cash flow, margin, asset

productivity, profitability and/or efficiency improvement. Stated differently, a value

creating merger should be associated with some measurable change in company

fundamentals. These studies are more conducive to performance evaluation on an ex-

post basis. Furthermore, if purported synergies are real, and cash flows do improve,

we should be able to identify the sources of any such real economic gains. Clearly,

managers that undertake M&A for synergistic motivations, and not for agency reasons

or managerial hubris, must have identified possible sources of economic gains before

proceeding with the transaction.

In a 1992 journal article, Healy, Palepu, and Ruback find that the 50 largest mergers

between US public industrial firms completed between 1979-1984 experienced higher

operating cash flow returns, mainly due to post-merger improvements in asset

productivity. They also report that such improvements do not come at the expense of

cutbacks in capital expenditures and research and development (R&D) investments,

17
thereby undermining the claim that the improvement in post merger cash flows is

achieved at the expense of the acquiring firms’ long-run viability/competitiveness.

These results are similar to Kaplan (1989). However, Kaplan finds that merged firms

decrease capital expenditures (CAPEX) in the three years following the merger. If

firms decrease CAPEX relative to their industry peers, this may undermine their

future competitiveness. In short, both studies report that merging was a desirable

course of action, as evidenced by the fact that firms that engaged in M&A appeared to

enjoy better economic strength relative to their peers post-merger. This point

illustrates the importance of using an industry-adjusted benchmark in interpreting

study results. If the aforementioned results are pervasive on a time series and cross-

sectional basis, then we feel more confident that mergers and acquisitions produce

economic gains. However, a large literature gap remains to be filled in this particular

area of research.

Further, Healy, Palepu, and Ruback’s analysis and many of its kind also suffer from

methodological problems stemming from complexities in analysing and interpreting

financial statement accounting data. For instance, the existence of different

permissible merger accounting methods, such as the pooling versus the purchase

method, render time series and cross-sectional comparisons difficult. For instance,

pre and post-merger accounting results are simply not comparable under the purchase

method. This is so because the purchase method consolidates bidder and target firm

accounts from the acquisition date onwards, and prior accounts are not restated.

Performing ratio analysis and comparing pre and post-merger results is therefore not

very useful as the asset base used to generate these operating results is no longer the

same subsequent to the merger taking place.

18
Another problem is that computing meaningful industry averages is complicated by

the fact that firms operate in multiple industries, which obscures the process of

identifying relevant benchmarks for computing industry-adjusted results and ratios.

The selected benchmark should consist of firms that closely mimic the sample

transaction firms being analysed. A related point is that benchmarks should be

composed of firms that have chosen not to merge. We know that mergers come in

waves and that such waves may cause entire industries to consolidate. Little insight

will be gained by comparing merged entities to similar firms that have also engaged in

M&A. Research should thus seek to determine whether an operating performance

improvement or deterioration occurs relative to similar firms that have chosen not to

merge. Employing such a course of action will provide superior insight as to whether

or not M&A represent beneficial events. Despite the aforementioned problems

however, fundamental analysis is much closer in spirit to helping resolve the M&A

value creation enigma on an ex-post basis. We would like to see researchers propose

new ways of tackling the M&A value creation debate, but from a fundamentals

perspective.

5. Conclusions: Is it Time for a Change?

Since its birth in 1969, the event study has predominately held a monopoly in the area

of investigating the M&A value creation issue. In this article, we argue that the event

study may have stagnated in terms of its ability to evaluate the desirability of mergers

on an ex-post basis. Among other things, the hefty premiums offered to target

19
shareholders exacerbate the short-run weighted-average CAR (WACAR) for

combined entities, and this in turn has misled many researchers into concluding that

most M&A represent value creating events. Furthermore, we show that short-term

changes in stock prices around M&A announcements can produce poor estimates of

shareholder wealth. We also discuss various problems inherent with long-window

event studies. We believe that long-run event study analyses are also unsuitable for

measuring the economic gains of M&A due to various methodological problems and

the forward-looking nature of stock returns. In short, both short and long-window

event studies provide biased results and undependable insight regarding the question

of M&A value creation.

In the wake of multiple merger waves that appear to be growing in strength every

time the tide comes in, financial economists can no longer afford to restrict its

research activities to solely using popular event studies for assessing whether M&A

represent beneficial events. Given the escalating macroeconomic significance of

merger transactions in recent years, advancing the M&A value creation debate has

never been more critical. Thus, new methods/techniques are very much needed to

advance our knowledge on this critical issue and we believe that fundamental-based

approaches represent more prolific avenues for new and future research. Despite the

flexibility offered by GAAP, accounting data remains the best proxy of company

economic performance available to investors, analysts and academics alike. We

therefore believe the M&A value creation puzzle can be better understood by

returning to fundamentals. We thus call on researchers to develop new analysis

techniques based on operating performance and fundamentals as these may yield

20
superior insight on whether the quasi-mystical idea of ‘synergies’ represents a valid

argument for engaging in M&A, or a mere excuse for empire building.

Footnotes:
i
See, for example, Dodd and Ruback (1977); Asquith (1983); Asquith, Bruner and Mullins, Jr. (1983);

Dennis and McConnell (1986); Bradley, Desai, and Kim (1988); Franks and Harris (1989).
ii
See, for example, Firth (1980); Dodd (1980); Sudarsanam, Holl, and Salami (1996); Draper and

Paudyal (1999).

21
iii
See, for example, Bradley, Desai, and Kim (1988); Mulherin and Boone (2000); Andrade, Mitchell,

and Stafford (2001).

iv
We make the simplifying assumption that both firms are 100% equity financed and therefore the

market value of assets is equal to the market value of equity. Weights are determined as follow: Wt =

MVt /(MVt+MVb) = Wt = 100,000 /(1,100,000) = 9.1%, and Wb = 1-Wt = 90.9%.

v
MVt = $100,000 divided by the number of shares outstanding (100,000) equals $1 per share. (1.40-

1)/1 represents a 40% premium on the share.

vi
In three-day window, Andrade, Mitchell, and Stafford (2001) find that the expected return is

approximately 0%, regardless of the benchmark model used. An expected return of 2% therefore

represents a very ‘generous’ estimate.

vii
WACAR = Wb (CARb) + Wt (CARt) = (9.1%)*(38%) + 0 = 3.458%.
viii
Malatesta (1983) argues that the widely used percentage abnormal returns do not capture the real

wealth changes of acquiring firm shareholders. However, dollar returns capture the wealth change of

acquiring firm shareholders.

ix
“Evidence suggests that the majority of acquisitions do not benefit shareholders in the long term.

Valuations and premiums tend to be excessively high and targets impossible to achieve”. - - Financial

Times 2004.
x
See for instance, Lajoux and Weston (1998) for an overview of several practioner study results.
xi
See for instance, Kaplan and Weisbach (1992).
xii
Barber and Lyon (1997) identify three biases: the new listing bias, the rebalancing bias, and the

skewness bias.
xiii
As noted in Ikenberry, Lakonishok, and Vermaelen (1995), the bootstrap approach avoids many

problematic assumptions associated with conventional t-tests over long time horizons, namely

normality, stationarity, and the time independence of sample observations.

xiv
Lyon, Barber, and Tsai (1999) document that the misspecification of test statistics can generally be

traced to 5 sources: the new listing bias, the rebalancing bias, the skewness bias, cross-sectional

dependence, and bad models problems.

xv
For these new methodologies, refer to Barber and Lyon (1997), Lyon, Barber, and Tsai (1999), Brav

(2000), and Mitchell and Stafford (2000).

22
xvi
For these long-run event studies, see, for example, Mitchell and Stafford (2000), Brav, Geczy, and

Gompers (2000), Eckbo, Masulis, and Norli (2000), and Boehme and Sorescu (2002).

References:

Andrade, G., M. Mitchell and E. Stafford, 2001, New evidence and perspectives on mergers,
Journal of Economic Perspectives 15, 103-120.

Asquith, P., 1983, Merger bids, uncertainty, and stockholder returns, Journal of Financial
Economics 11, 51-83.

23
Asquith, P., R. F. Bruner, and D. W. Mullins, Jr., 1983, The gains to bidding firms from
merger, Journal of Financial Economics 11, 121-139.

Barber, B.M. and J.D. Lyon, 1997, Detecting long-run abnormal stock returns: the empirical
power and specification of test statistics, Journal of Financial Economics 43, 341-372.

Boehme, R. D. and S. M. Sorescu, 2002, The long-run performance following dividend

initiations and resumptions: underreaction or product of chance? Journal of Finance 57, 871-
900.

Bradley, M., A. Desai, and E. H. Kim, 1988, Synergistic gains from corporate acquisitions
and their division between the stockholders of target and acquiring firms, Journal of
Financial Economics 21, 3-40.

Brav, A., 2000, Inference in long-horizon event studies: a Bayesian approach with application
to initial public offerings, Journal of Finance 55, 1979-2016.

Brav, A., C. Geczy, and P. Gompers, 2000, Is the abnormal return following equity issuances
anomalies? Journal of Financial Economics 56, 209-249.

Dennis, D. K. and J. J. McConnell, 1986, Corporate mergers and security returns, Journal of
Financial Economics 16, 143-187.

Dodd, P., 1980, Merger proposals, management discretion and stockholder wealth, Journal of
Financial Economics 8, 105-137.

Dodd, P. and R. Ruback, 1977, Tender offers and stockholders returns: An empirical analysis,
Journal of Financial Economics 5, 351-373.

Draper, P. and K. Paudyal, 1999, Corporate takeovers: Mode of payment, returns and trading
activity, Journal of Business Finance and Accounting 26, 521-558.

Eckbo, B. E., R. Masulis, and O. Norli, 2000, Seasoned public offerings: resolution of the
“new issue puzzle,” Journal of Financial Economics 56, 251-291.

Fama, E.F., L. Fisher, M.C. Jensen, and R. Roll, 1969, The adjustment of Stock Prices to New
Information, International Economic Review 10, 1-21.

Firth, M., 1980, Takeovers, shareholder returns, and the theory of the firm, Quarterly Journal
of Economics, 235-260.

Franks, J. R. and R. S. Harris, 1989, Shareholder wealth effects of corporate takeovers: the
U.K. experience 1955-1985, Journal of Financial Economics 23, 225-249.

Healy, M.H., K.G. Palepu and R.S. Ruback, 1992, Does corporate performance improve after
mergers? Journal of Financial Economics, 31, 135-175.
Ikenberry, D., J. Lakonishok and T. Vermaelen, 1995, Market underreaction to open market
share repurchases, Journal of Financial Economics, 39, 181-208.

Lajoux, A. R. and J. F. Weston (1998), Do deals deliver on postmerger performance ?,

Mergers and Acquisitions 33 (2), 34-37.

Jarrell, G.A., J. Brickley and J. Netter, 1988, The market for corporate control: The empirical
evidence since 1980, Journal of Economic Perspectives, 2, 49-68.

24
Kaplan, S., 1989, The effects of management buyouts on the operating performance and
value, Journal of Financial Economics, 24, 217-254.

Kaplan, S. and M.S Weisbach, 1992, The success of acquisitions: Evidence from divestitures,
Journal of Finance 47, 107-138.

Kothari, S.P. and J.B. Warner, 1997, Measuring long-horizon security price performance,
Journal of Financial Economics 43, 301-339.

Loughran, T. and A.M. Vijh, 1997, Do long-term shareholders benefit from corporate
acquisitions? Journal of Finance 52, 1765-1790.

Lyon, J.D., B.M. Barber and C. Tsai, 1999, Improved methods for tests of long-run abnormal
stock returns, Journal of Finance 54, 165-201.

Malatesta P. H., 1983, The wealth effect of merger activity and the objective functions of
merging firms, Journal of Financial Economics 11, 155-181.

Mitchell, M., T. Pulvino and E. Stafford, 2004, Price Pressure Around Mergers, Journal of
Finance 58, 31-63.

Mitchell, M. and E. Stafford, 2000, Managerial decisions and long-term stock price
performance, Journal of Business 73, 287-329.

Moeller, S.B., F.P. Schlingemann and R.M. Stulz, 2003, Do shareholders of acquiring firms
gain from acquisitions? Working Paper, Ohio State University.

Moeller, S.B., F.P. Schlingemann and R.M. Stulz, 2004, Wealth Destruction on a Massive
Scale? A Study of Acquiring-Firm Returns in the Recent Merger Wave, Journal of Finance,
Forthcoming.

Mulherin, J. H. and A. L. Boone, 2000, Comparing acquisitions and divestitures, Journal of

Corporate Finance 6, 117-139.

Rau, P.R. and T. Vermaelen, 1998, Glamour, value and the post-acquisition performance of
acquiring firms, Journal of Financial Economics 49, 223-253.

Roll, R., 1986, The hubris hypothesis of corporate takeovers, Journal of Business 59, 197-
216.

Shelton, L.M., 1988, Strategic business fit and corporate acquisitions, Strategic Management
Journal 9, 279-287.

Sudarsanam, S., P. Holl, and A. Salami, 1996, Shareholder wealth gains in mergers: effect of
synergy and ownership structure, Journal of Business Finance and Accounting 23, 673-698.
Visvanathan, S. and B. Wei, 2004, Endogenous events and long run returns, Working Paper,
Fuqua School of Business, Duke University.

Walter, G.A. and J.B. Barney, 1990, Management objectives in mergers and acquisitions,
Strategic Management Journal 11, 79-86.

25