OVERVIEW OF THE BOOK

I frame the response to these questions in the three parts of this book. Part I (chapters 2, 3, and 4) offers perspectives on merger failure from the standpoints of previous research. Chapter 2 summarizes what we know about merger failure and success based on more than 130 studies drawn from research in business and financial economics. The research shows that the field of M&A is highly segmented; there are attractive and unattractive neighborhoods. This is the foundation for my argument that all M&A is local. Chapter 3 offers a summary profile of the best and worst deals from 1985 to 2000. Chapter 4 gives a perspective on the processes of failure, drawn from analyses of real disasters and from concepts in a number of disciplines, including cognitive psychology, sociology, and engineering. Thus, Part I constructs a lens through which to view the causes of merger failure.

Part II lends texture to our understanding of M&A failure through 10 case studies of big M&A disasters. I have paired each of the 10 with a counterpoint or complementary case so that each chapter is actually a paired comparison of what can go wrong and how. The comparison cases are not intended to be deals from heaven; they merely differ in some instructive way. The comparisons suggest how little the situation must differ in order to deliver rather different results.

These 10 cases are not necessarily the worst in any absolute sense. However, judged on common standards there can be little disagreement that they belong on a short list of bad deals. One has a large pool of candidates from which to select. I chose these 10 for several reasons.

• Size of damage. Big bad deals certainly get one’s attention and have face validity. I looked for losses in the billions of dollars, for layoffs, CEO change, tarnished reputation, and possibly, bankruptcy.

• Diversity of industry, deal type, and challenges. I could fill a book of M&A disasters drawn from any one of a number of industries. However, as I argue here, industry conditions have a powerful effect on merger success and failure. Thus, diversity of settings sharpens our understanding about how industry has an impact and intrigues us with its local surprises.

• Access to information. In half of these cases, I was fortunate to find insiders or knowledgeable outside observers to interview and inform the discussion. However, most participants do not want to discuss their M&A failures. In five instances, I was able to tap valuable archives that lent some insight into senior management’s views. For the rest, I relied on a diverse collection of investigative journalism, security analysis, and open commentary. Regardless of sources of information, I aimed to bring a fresh lens through which to view these cases, informed by economics and a scientific mind-set.

I tried to avoid M&A disaster cases caused mainly by crimes, looting, fraud, and sabotage.These are more appropriate for a book on white-collar crime. Such cases are a small fraction of the larger sample of messes we can find and by their notoriety, tend to obscure more important lessons for CEOs and the public. Even so, criminal litigation followed two of the cases described here.

The research and framework from Part I and the case studies in Part II open the door to Part III. There I offer some summary implications for CEOs, investors, and those concerned with public policy.

This is an exercise in inductive research, the generation of a way of thinking about failure in M&A, drawn from a detailed look at the research and cases. These were all failures that could have been avoided or sharply mitigated. I hope to show why and how. The result is a volume that seeks to teach, rather than harangue, titillate the reader, or humiliate the protagonists. Where the facts do not fit with sympathetic explanations, I speak plainly; but generally, my bias is to view the challenges facing executives as extremely difficult, arenas in which scholars and the casual reader could easily have done worse. Other analysts may beg to differ on the interpretation of specific events in certain cases or of detailed points in the research stories summarized here.Yet such differences should not obscure the larger point that there are considerable similarities among merger failures and that such similarities lend insight into the causes of failure and the implications for managers.

OVERVIEW OF THE FINDINGS

The key message of this book is that mergers fail because of a “perfect storm” of factors that combine to destroy the new firm. This message invites consideration of the definition, frequency, profiles, and process of failure.

What Is Merger Failure?

The first question is definitional. Though “failure” is commonly understood, it has several differing applications. For instance, the Oxford English Dictionary defines “failure” as:

1. A failing to occur, to be performed, or be produced; an omitting to perform something due or required; default . . . 2. The fact of becoming exhausted or running short, giving way under trial, breaking down in health, declining in strength or activity, etc. . . . 3. The fact of failing to effect one’s purpose; want of success . . . 4. The fact of failing in business; bankruptcy, insolvency . . .²

As the dictionary suggests, “failure” connotes both a process (“a failing to occur . . . omitting to perform . . . giving way . . . breaking down”) and an outcome (“fact of . . . want of,” and bankruptcy). In this book, I aim to discuss both process and outcome. My method is to begin with identifiable outcomes and induce from them some insights about process.

What, then, is the outcome of M&A failure? Consider at least these six dimensions:

1. Destruction of market value. Harnessing the perspective of the providers of capital, we measure the destruction of value by the percentage change in share values, net of changes in a benchmark, such as a large portfolio of stocks.

2. Financial instability. Some of the saddest M&A deals are those that, rather than making the buyer stronger, actually destabilize it. In most of these cases, the buyer overreaches its financial capacity. Degree of financial stability is reflected in debt ratings, earnings coverage ratios, probability of default, and other measures of the ability of the firm to bear risk.

3. Impaired strategic position. Many M&A transactions are motivated by a strategic purpose that seeks to improve the firm’s competitive position, acquire new capabilities, improve agility, or obtain resources that are vital to future prosperity. Indications of failure in this dimension would include loss of market share, and involuntary abandonment of products, geographic markets, or research and development (R&D) programs.

4. Organizational weakness. Knitting together two firms is especially challenging from an organizational perspective. Most CEOs would agree with the old slogan “People are our most important asset.” In essence, one could measure organizational strength in terms of depth of talent and leadership, effectiveness of business processes, and the transmission of culture and values. Adverse changes in human resources appear in unanticipated workforce layoffs, involuntary changes of leadership in senior management and the board, and defections of talented individuals to competitors.

5. Damaged reputation. The M&A deal should improve the reputation of the acquirer and its deal architects. Usually, the realization of these other aims will do just that. But one can imagine deals that depend on acrimony, subterfuge, and a win-lose mentality—in a world of repeated play, the executive must consider how these qualities might affect one’s M&A success in future deals. Measurable outcomes in this dimension would include changes in name recognition, reputation, analyst sentiment, and press coverage.

6. Violation of ethical norms and laws. You can gain financial, organizational, and strategic objectives in M&A, but in ways that violate norms such as equity, duty, honesty, and lawful observance. After the corporate scandals of recent years, any assessment of outcomes would be incomplete without consideration of laws and ethics. Adverse judgments in criminal and civil litigation would be a rough measure of M&A failure, though they usually follow an extended lapse in ethics.

I was influenced in my selection of the 10 case studies by all of these factors. However, some of these criteria are difficult or impossible to benchmark. As a result, the discussion that follows gives somewhat more attention (though not necessarily more weight) to the financial dimension.

How Prevalent Is M&A Failure?

M&A failures amount to a small percentage of the total volume of M&A activity. Investments through acquisition appear to pay about as well as other forms of corporate investment. The mass of research suggests that on average, buyers earn a reasonable return relative to their risks. M&A is no money-pump. But neither is it a loser’s game. Conventional wisdom seems to think otherwise, even though the empirical basis for such a view is scant.

Of more interest to the thoughtful practitioner is not the average result, but the distribution around it. The wide variation in findings about the profitability of M&A suggests that something may be going on to tilt the odds of success and failure. Chapter 2 identifies 18 neighborhoods or dimensions of the M&A market, along which returns to buyers vary significantly. For instance, acquisitions of public companies tend to be much less profitable to buyers than are acquisitions of privately held companies.

The existence of neighborhoods of return suggest that all M&A is local . That is, managerial choice can have a huge impact on the results of M&A: Quite simply, where you choose to do business will influence success or failure.

What Causes M&A Failure?

That M&A is local also implies that we might find constructive or destructive patterns of managerial choice in the deals from hell or heaven that they produce. Chapter 3 reports that the extreme outliers of deals, both good and bad, differ from each other and from the middle in at least four general ways.

The first regards strategy. In the best deals, buyers acquire targets in industrially related areas. In the worst deals, targets are in areas that are more distant. This may reflect the benefits of sticking to your knitting: Better knowledge of a related industry may yield fewer surprises and more opportunities to succeed.

In addition, the fit of the buyer and target matter significantly. In successful deals, buyers acquire from strength—the performance attributes of buyers are stronger than their targets suggesting that in good deals, the buyer brings something important to the success of Newco. In the worst deals, buyers acquire from weakness. There, the target is significantly stronger, and one can guess that the buyer hoped to fix some problem by means of the acquisition.

Third, the worst deals show a propensity to occur in “hot” market conditions. The preeminent hot market in business history was the equity market bubble from 1998 to 2000 associated with the emergence of the Internet. Industries can also be hotbeds of activity caused by sudden deregulation, technological change, shifts in consumer demand, and so on. Recent research suggests that many mergers are motivated by strategic turbulence in their industries. The best deals occur in cooler market conditions. Chapters 2 and 3 discuss the possible reasons for this disparity.

Finally, deal tailoring pays. One size does not fit all. Better deals are associated with payment by cash and earnout schemes and the use of specialized deal terms. The worst deals are associated with payment by stock.

Chapters 2 and 3 show that managers’ choices matter. However, as the dictionary definition of “failure” suggests, there is more to the story, namely, what can go wrong between the managers’ choices and the outcomes. Here is where the case studies of merger failure become vitally important. Chapter 4 shapes a lens through which to view these cases. In that chapter, I turn to research on real disasters (those involving the loss of life and property) to highlight six factors that lend traction in understanding the cases of M&A failure:

1. The businesses and/or the deal were complicated.This made it difficult for people on the scene to understand what was going on or to take quick and effective action.

2. Flexibility was at a minimum. Little slack or inadequate safety buffers meant that problems in one part of the business system would radiate to other parts.Trouble would travel.

3. Deliberately or inadvertently, management made some choices that elevated the risk exposure of the new firm.

4. The thinking of decision makers was biased by recent successes, sunk costs, pride, overoptimism, and so on.

5. Business was not as usual. Something in the business environment departed from expectation causing errors or problems.

6. The operational team broke down. Cultural differences between the buyer and target, unresolved political issues, and generally overwhelming stress prevented the team from responding appropriately to the unfolding crisis.

The case studies in Chapters 5 through 14 illustrate these six factors at work in producing the “perfect storm” of merger failure. Each case highlights one or more special aspects of the storm. Chapters 15 and 16 discuss important implications for deal-doers, operating managers, and CEOs.

CONCLUSION: ANTICIPATING M&A FAILURE CAN ENHANCE SUCCESS

I wrote this book to lift the practice of M&A. Given the sheer complexity of these deals, best practice cannot be reduced easily to a sound bite. For this reason, case studies are excellent for helping the thoughtful practitioner envision what can go wrong and what to do about it. More importantly, the book suggests that success depends vitally on adopting the right attitude about M&A as a path of corporate growth. It is no formula for surefire success; rewards are extremely difficult to sustain over time; risks are legion; it is to be undertaken only with very serious planning and preparation; and the effort should be motivated by the right values and respect for investors. As William Blake said, “Execution is the chariot of genius.”

NOTES

1 See Sarasvathy and Menon (2002) and Gartner (1988) for overviews of research on entrepreneurial failure. The research on failure rates of new firms shows wide variation from one study to the next.This is most likely attributable to variations in the length of time after founding from which failure is measured.

2 Oxford English Dictionary (Oxford, England: Oxford University Press, 2d ed. 1989), volume V, 667.

THE FOUNDATIONS OF M&A FAILURE

Where M&A Pays and Where It Strays: A Survey of Research

How you assess an M&A deal or the whole flow of M&A activity depends on your frame of reference, on beliefs that help you decide whether specific deals represent the average, or what statisticians call the “tails of the distribution.” This frame of reference is a hugely important filter for decision makers and their advisers and is typically built on a blend of personal experience, anecdotes, conventional wisdom, and facts. The aim of this chapter is to enrich your frame of reference about success and failure in M&A with the findings of scientific research.¹

I have two basic criticisms of the way most people think about M&A. First, the conventional wisdom is poorly grounded in the scientific evidence on the subject. The fashionable view seems to be that M&A is a loser’s game.Yet an objective reading of more than 130 studies supports the conclusion that M&A does pay.² These studies show that the shareholders of the selling firms earn large returns from M&A, that the shareholders of the buyers and sellers combined earn significant positive returns, and that the shareholders of buyers generally earn about the required rate of return on investment.

Second, conventional wisdom seems to hold that failure is the average outcome of all classes and varieties of M&A, and that, in this sense, M&A, is regrettably homogeneous. Yet the research reveals wide dispersion of returns both within and across studies. And such variation suggests that, like powerful turbulence under the smooth surface of a river, something is going on in the world of M&A that differentiates deals and predisposes them to success or failure—and that, like the varying river conditions below the surface, the world of M&A is not homogeneous. In this chapter, I argue that to see the varying states of this world is to build a view that more readily grasps the circumstances and approaches that give rise to success and failure.

Tip O’Neill, the Boston pol who rose to be Speaker of the U.S. House of Representatives, explained that in trying to understand the workings of Congress, it made no sense to focus on lofty national issues or policy debates within the Washington Beltway. Instead, he said,“All politics is local.” The mindset of the successful politician begins with his or her constituency and the hopes and fears in town halls, school boards, and police precincts. It is the same in M&A:The best foundation for pursuing success and avoiding failure in M&A lies in seeing the important ways in which individual deals differ from one another. In other words, all M&A is local.

CONVENTIONAL WISDOM ON M&A FAILURE: STILL HAZY AFTER ALL THESE YEARS

The popular view is that M&A is a loser’s game. The following excerpts from a recent book are representative:

The sobering reality is that only about 20% of all mergers really succeed . . . most mergers typically erode shareholder wealth . . . the cold, hard reality that most mergers fail to achieve any real financial returns . . . very high rate of merger failure . . . rampant merger failure . . .³

I have lost count of the references—in newspaper columns, magazines, and consultants’ reports—to this 20 percent success rate: Much is made of how small it is, though it would dwarf success rates of other business activities frequently lauded including new business startups, new product introductions, expansions to new markets, and investments in R&D and new technology. All business is risky. Our purpose in studying failure should be to manage risk better, not eliminate it.Yet there is no body of empirical research that documents this low success rate of M&A activity with a consistency and level of care equal to the gravity of the assertion.What is missing in the popular discussion is a rigorous definition of what the conventional wisdom means by “failure” in M&A. Here, too, the writers tend to refer vaguely to a shortfall between the goals of a merger and its outcomes, without any consideration of the appropriateness of goals or the general conditions in which the merger took place.

Here’s an example of how the conventional wisdom on merger failure takes shape. In December 2002, a columnist wrote in the Wall Street Journal ,“Most mergers don’t work . . . A mountain of academic research shows most acquisitions end up costing shareholders . . .”⁴ Later, in correspondence with me, he cited as proof a BusinessWeek article entitled “Why Most Big Deals Don’t Pay Off” and reporting that “61% of buyers destroyed shareholder wealth.”⁵ That article studied 302 large mergers of public companies from 1995 to 2001 and looked at the changes in the buyer’s share price in the year following the bid adjusted for changes in the share prices of peer firms or an industry average.

This example reveals a number of problems in making inferences about the profitability of M&A. First, the conventional wisdom generalizes too readily the findings of a single study (the columnist’s use of “most mergers”). Second, there is a tendency to exaggerate the extent of failure. By the terms of the Business Week study, a share price decline of just a penny would constitute a “failure.” But recognizing that most share price movements are subject to a certain amount of noise, a more sensible approach would focus on significant failure and would exclude the noise. What’s more, the study’s period of observation was a once-in-a-lifetime outlier in capital market performance, which should discourage generalization to other time periods. Especially interesting during this time period was the feverish M&A activity in certain industry segments. For instance, a number of the large deals in the sample involved high-tech or Internet-related companies. To what extent was the general finding of the study affected by the price collapse of the technology-media-telecommunications sector that had started in 2000? Most of the deals in the sample were stock-for-stock exchanges, which, as past research has shown, tend to be worse for buyers than cash deals. The deals in the sample were also big deals, which by their nature have more integration and regulatory problems. And the study focused on deals between two public companies while acquisitions of private companies—which are far more common than public deals—tend to be much more profitable for buyers.

So was Business Week’s sample representative of all M&A? No. It simply gives us insight into the profitability of big public-company deals in a “hot market.” To vault from there to the assertion that “most mergers don’t work” is unwarranted.⁶

Two recent studies (both by the same authors) offer a cautionary counterpoint to the WSJ/BW articles. In the first of the two, the authors examined the two-day abnormal stock market returns to buyers using a large sample of deals announced from 1980 to 2001.⁷ They found that the adjusted returns to buyers measured in dollars (not percentage returns) over this period were significantly negative (-$25.2 million), on average, consistent with the argument that “most mergers don’t work.” But the authors also report three other important details. First, the average percentage adjusted return to buyers was a significantly positive 1.1 percent and their research shows that the inconsistency between the dollar and percentage returns is due to the extreme unprofitability of a few large deals. Also, in the second study, the authors report that most of the losses from 1980 to 2001 were concentrated in just 87 deals, out of a total sample of 12,023; without these deals, the whole sample would have showed a significantly positive dollar return. Third, the 87 culprits were concentrated in the hot M&A market of 1998 to 2001. Thus, but for a relatively small number of deals in a limited market episode, one reaches a very different conclusion about the profitability of M&A.

The conventional wisdom on M&A thus tends to be either hazy on the evidence or, where solid evidence is offered, too ready to generalize from the findings based on localized conditions.There is more to the story of M&A success and failure. An informed view depends on mastering the scientific findings on the profitability of M&A and on understanding how profitability varies by types of deals and companies.

MEASUREMENT OF M&A PROFITABILITY: BETTER THAN WHAT?

Before looking at the findings, you need to define the tests. The benchmark for measuring performance is investors’ required returns, commonly defined as the return investors could have earned on other investment opportunities of similar risk. Against this benchmark, we can see three possible outcomes:

1. Value is destroyed. In this case, investment returns are less than those required by investors. Investors could have done better investing in another opportunity of similar risk, and they are justifiably unhappy.

2. Value is created. The investment earns a rate of return higher than required. Investors should be happy.

3. Value is conserved. The investment just earns its required rate of return. Economically speaking, investors earn “normal” returns. They should be satisfied because wealth is growing at a rate that just fairly compensates the investor. If an investor requires a return of 15 percent and receives it, the value of the investment will double in five years. But value is being just “conserved” in the sense that the investor is merely doing as well as required.

Judgments about the success or failure of M&A transactions should be linked to these measurable economic outcomes. In economic terms, an investment is “successful” if it does anything other than destroy value.

Why do we focus so narrowly on economics? After all, many managers describe a complex set of motives for acquisitions: shouldn’t the benefit of M&A activity be benchmarked against all of them? In an ideal world, it would be nice to draw on a range of concerns, much as corporations do today with the balanced scorecard. But managers’ motives may have little to do with shareholder returns, or with building capabilities that might help achieve them over the longer run. One often hears of M&A deals that are struck for vague strategic benefits, the creation of special capabilities, the achievement of competitive scale, or simply because two organizations or their CEOs are especially friendly. The only way one can prove that such benefits add value is by measuring the economic outcomes rigorously, using a common metric across all deals. Second, special deal-specific definitions of success limit our ability to generalize from the research findings. Increasing shareholder wealth of shareholders is a fundamental objective of all firms; indeed, in the United States, corporate directors are required to implement policies consistent with shareholders’ interests, which is generally synonymous with creating value.⁸ Fortunately, benchmarking against value creation does allow us to make generalizations. In what follows, I focus on economic outcomes in the hope of saying something meaningful grounded in scientific research.

One of the basic conclusions of economics is that where markets are reasonably competitive, players will earn a “fair” rate of return; that is, you just get paid for the risk you take, but no more. The intuition for this is simple:Where information is free-flowing and entry is easy, a firm earning very high returns will draw competitors as surely as honey draws flies.The entry of these other firms will drive returns down to the point where the marginal investor gets just a fair rate of return. This idea, which has been tested extensively in financial markets, leads directly to the concept of market efficiency, which says that prices reflect news quickly and without bias. Whether a free lunch exists in M&A hinges on returns to investors, and, as with tests of capital market efficiency, that hypothesis can be tested using one of three classes of measures:

1. Weak form. Did the share price rise? Are the shareholders better off after the deal than they were before? Most simply, was the buyer’s stock price higher after the deal than before? Such a before-and-after comparison is widespread, especially in the writings of journalists and securities analysts. But it is a weak test in the sense that it fails to control for other factors unrelated to the deal that might have triggered a price change.

2. Semistrong form. Did the firm’s returns exceed a benchmark? Are shareholders better off compared to the return on a comparable investment? Introducing a benchmark like the return on the S&P 500 index, or the return on a matched sample of peers that did not merge, strengthens the analysis. This kind of test, which is commonly used in academic research, is more reliable than the weak form tests because it controls for the possibility that the observed returns were actually driven by factors in the industry or the entire economy, rather than were due to the merger. But this kind of test is at best semistrong because benchmarks are imperfect.

3. Strong form. Are shareholders better off after the deal than they would have been if the deal had not occurred? This is the true test of the cost of lost opportunity, the economists’“gold standard” of comparison. And it is what most people think they are finding when they look at weak and semistrong form test results. But the true strong form test could tell a very different story. Consider the case of America Online (AOL)’s merger with Time Warner in January 2001. The weak and semistrong tests would show large losses to AOL’s shareholders in the two years following the deal. As Chapter 12 reports, AOL Time Warner outperformed a benchmark portfolio of pure Internet stocks, while underperforming the S&P 500 index and media and entertainment indexes. Given the bursting of the Internet bubble, it seems likely that AOL’s shareholders would have been much worse off without the merger. It would appear that AOL’s acquisition of Time Warner was shrewd and successful for the buyer, despite what the weak and semistrong results show (though Time Warner’s shareholders appear to have cushioned the collapse of the Internet bubble for AOL’s shareholders). The problem is that strong form results are difficult, if not impossible, to observe.

The distinction among these three kinds of tests is important to bear in mind. The studies summarized in this chapter are, at best, semistrong. Therefore we must exercise humility in drawing conclusions about performance against economic opportunity. We are looking through a glass darkly.

Two main research approaches offer findings that can help us in forming a view of M&A profitability. The first are so-called “event studies,” which examine the abnormal returns to shareholders in the period surrounding the announcement of a transaction.The “raw” return for one day is simply the change in share price plus any dividends paid, divided by the closing share price the day before.The abnormal return is the raw return less a benchmark of what investors required that day. Typically, the benchmark is the return specified by the capital asset pricing model (CAPM) or, more simply, the return on a large market index like the S&P 500.These studies are based on the assumption that stock markets are forward-looking and that share prices are simply the present value of expected future cash flows to shareholders. Since the 1970s, these studies have dominated the field.

By contrast, so-called “accounting studies” examine the reported financial results (i.e., accounting statements) of acquirers before and after acquisitions to see how financial performance changes. The focus of these studies is on variables such as net income, return on equity or assets, earnings per share (EPS), leverage, and liquidity. The best of these studies are structured as matched-sample comparisons in which acquirers’ performance is set against that of nonacquirers of similar size that operate in the same industry. In these studies, the question is whether the acquirers outperformed their nonacquiring peers.

If scientific inquiry means anything, it is to frame a hypothesis and test it rigorously against the possibility that the result is due merely to chance. Researchers accordingly test for the statistical significance of a result. But statistical significance is not the same as economic materiality. To say that M&A transactions create or destroy value on average requires not only the proof of significance (i.e., that the result is not due to chance) but also some compelling evidence of materiality, namely that the wealth effects of M&A transactions are large enough that shareholders or society should worry about them. Many of the significant abnormal returns reported in event studies are as small as 1 percent or 2 percent. But since these returns occur over just a few days, they are large enough, when viewed on an annualized basis, to cause concern or elation among institutions or other sophisticated investors whose performance can be greatly affected by such events.

THE BROAD FINDING: M&A PAYS

Event studies yield insights about market-based returns to target firm shareholders, buyers, and the combined entity.

Abnormal returns to target firm’s shareholders. The findings of 25 studies⁹ show that target firm shareholders earn returns that are significantly and materially positive, despite variations in time period, type of deal (merger vs. tender offer), and observation period. In short, the M&A transaction delivers a premium return to target firm shareholders.

Abnormal returns to buyer firm’s shareholders. The pattern of findings about market-based returns to buyer firms’ shareholders is more problematic. About 40 percent of roughly 50 studies report negative announcement returns to buyers; 60 percent report positive returns. When statistical significance is taken into account, the studies of returns to buyer firm shareholders show an even stronger positive bias: Twenty-six percent (14) show value destruction¹⁰ (significantly negative returns); 31 percent (17) show value conservation¹¹ (insignificantly different from zero); and 43 percent (23) show value creation ¹² (significantly positive returns).