Dubious Logic of Global Megamergers

Most executives think it’s good to be big in a globalizing economy. You can’t look at the front pages of the Wall Street Journal or the Financial Times without seeing yet another megadeal in the headlines. Companies seem to be combining at a rate almost unprecedented in history—and on a global scale.

In the oil sector, there’s Exxon and Mobil, not to mention BP’s mergers with Amoco and Atlantic Richfield. In the automobile industry, Daimler-Benz and Chrysler have joined forces, Ford has taken over the automobile operations of Volvo, and Renault has acquired a significant stake in Nissan. Similar merger examples can be found in industries as diverse as telecommunications, entertainment, financial services, soft drinks, and even cement.

Pushing these huge—and pricey—cross-border deals is the almost universal belief that industries will inevitably become more concentrated as the world’s markets become more globalized. The spoils of the market are supposed to go to a select few in each industry. And companies believe that if they are going to be among the winners, they will have to shore up economies of scale in manufacturing, branding, and research and development. That’s how they hope to scare off potential competitors and sew up new markets. From this perspective, cross-border mergers are a do-or-die proposition. If you want to survive, let alone thrive, you must be one of the world’s biggest players.

That idea has a long-established pedigree. More than 100 years ago, Karl Marx wrote, “One capitalist always kills many,” meaning that a constantly dwindling number of capitalists would eventually monopolize everything. This theory has persistently dominated business, though usually in less pernicious terms than Marx used. In the 1970s, Bruce Henderson, the founder of management firm Boston Consulting Group, formulated his famous rule of three: a stable competitive market never has more than three significant competitors. Similar thinking underscores CEO Jack Welch’s widely quoted dictum that General Electric must be either number one or two in its various businesses. Not to be outdone, Mercer Management Consulting in the late 1990s popularized the plight of the silver medalist: companies are either number one, Mercer argued, or they’re nobody.

The assumption that the global economy is a winner-take-all economy has become common wisdom—but there’s no evidence to support this premise. The theoretical links between the globalization of an industry and the concentration of that industry are weak. Empirical research indicates that global—or globalizing—industries have actually been marked by steady decreases in concentration in the post—World War II period. Executives, then, need to break free of the biases that lead them to pursue larger and larger cross-border deals. There are better, more profitable strategies for dealing with globalization than relentless expansion.

A Fresh Look at Theory

To properly understand the relationship between globalization and concentration, it’s necessary to look at the underlying economics. The granddaddy of all globalization models is the theory of comparative advantage, which was laid out by David Ricardo at the beginning of the nineteenth century. Ricardo studied two countries, England and Portugal, which produce two distinct goods, cloth and port, respectively. Ricardo demonstrated that as long as Portugal was in the comparatively better position to make port, and England to make cloth, then both countries would do better by specializing. This sort of specialization would be productive even if one of the countries was more efficient across the board.

Many business thinkers assume that this theory points toward industry concentration. But in fact, it simply predicts the geographic concentration of production, not concentration of the number of companies in an industry. Consider Ricardo’s example again. Portugal is indeed the center of the port business today, but more than 30,000 small companies and 70 shippers engage in this export trade. It is difficult to describe this as industry concentration.

Of course, the theory of comparative advantage is abstract and doesn’t take into account economies of scale—perhaps the biggest driver of industry concentration. So what happens if we modify the theory to include the economies of scale that most companies are scrambling for today? The line of research that is best developed in this regard, monopolistic competition, hardly offers a ringing endorsement for industry concentration. Let’s look at an example from Paul Krugman and Maurice Obstfeld’s International Economics: Theory and Policy (Addison-Wesley, 2000) in which they examine what happens to an industry when two producing countries integrate their economies. Prior to integration, Country A has six producers in a certain industry; Country B has eight. After integration, only ten of the 14 total producers survive—yet that still represents an increase in the number of distinct product lines available to customers in both countries. This is hardly a rule of three, let alone of one.

In fact, for theoretical models to predict a significant concentration of producers, you have to assume very high economies of scale. That tends to happen only in extreme cases. When an industry undergoes a big technological change, for instance, the companies that invest the most money in the new technology can reap huge payoffs. Consequently, those companies drive out competitors, leaving only a few players standing. Consider the color film revolution in photography. Opportunities to improve black-and-white film had largely been tapped out by the 1960s as the technology matured. But the increasing consumer preference for color film, which requires physical and chemical interactions of a dozen or more layers of emulsions, triggered an intense R&D race. Kodak, the industry leader, won the race by outspending its competitors. Today, Kodak and Fuji are the top two global competitors at the cutting edge of photo technology.¹

What the Data Show

Is the experience of the color film industry a common one? To answer this question, we examined more than 40 years of data on the global market shares of companies in more than 20 industries. We used these data to calculate a Herfindahl index for each industry. The Herfindahl index measures the relative degree of industry concentration. The index number is the sum of the squares of the market shares of the companies in an industry. If a few key players have high market shares, the Herfindahl index is higher than if the industry’s market shares are spread more evenly across more companies. Three companies splitting a market evenly translates into a Herfindahl index of .33. Ten companies splitting a market evenly translates into a Herfindahl index of .10. For this article, we’ve used a modified Herfindahl index based on the market shares of the top ten companies in each industry we looked at.² Earlier work that we conducted using other concentration measures—for instance, the market share of the largest company or the combined shares of the four largest companies—produced comparatively similar results.

Let’s first consider the oil industry, which most businesspeople think of as global throughout the postwar period because of the reach of the major oil companies. The exhibit below tracks the Herfindahl index for the oil industry during the past half-century—except during the late 1970s and the 1980s because of the unreliability of the data in the wake of oil shocks at that time. No, those steadily declining lines from the 1950s through the 1970s don’t reflect a dwindling number of industry players. The Herfindahl index clearly demonstrates an increase in producers and refiners in the first 20 years of the oil industry data. And while the modest uptick in the graph reveals slight concentration in the early 1990s, the earlier downward trend seems to have resumed. Even if we assume that all the oil deals announced in the last few years happen, they won’t greatly affect the index. Thus, the oil industry is actually far less concentrated today than it was 50 years ago. The industry’s Herfindahl index of less than .05 today corresponds to more than 20 equal-sized competitors in the field—a significantly greater number of players than is predicted by the rules of one, two, or three.

Of course, it could be argued that the oil industry presents a special case because of its political importance. The governments of oil-exporting countries, particularly OPEC countries, effectively nationalized large chunks of their oil production and refining capacities. But we don’t think deconcentration in the oil industry is an anomaly. Consider other industries that are based on natural resources. Analysis of the Herfindahl indices in industries such as zinc, bauxite, and copper also reveals decreased global concentration since World War II.

But it’s not just commodity industries that show a long-term trend away from concentration. Take the automobile industry: based on its cross-border trade and investment flows, this has undeniably become a more global business since World War II. But it hasn’t become more concentrated. As the auto industry exhibit below illustrates, the long-run trend has been toward dispersing market share and power across a greater number of players. That’s borne out by the experience of individual companies. Consider General Motors. Its worldwide market share has fallen to its lowest level in more than 50 years. And the market share it has lost has been spread among a large number of smaller players. It’s true that recent international mergers—including the $40 billion Daimler-Chrysler deal—have contributed to increased industry concentration in the 1990s. But that increase is still relatively modest.

In high-tech industries, the case for concentration may be somewhat stronger but hardly overwhelming. According to Standard & Poor’s Compustat data from 1988 to 1998, the top five high-tech companies’ shares of worldwide sales in each of three industries—computer hardware, computer software, and long-distance telephony—actually declined by 15 to 30 percentage points. This would point toward less concentration in high-tech industries. Those who believe in global concentration are likely to protest that the Compustat calculations ignore the powerful global alliances that dominate long-distance telephony and other industries—and which arguably represent a form of industry concentration. In principle, they are right, but let’s conduct a reality check.

Consider for a moment the major telecommunications alliances that were forged between 1992 and 1996 to compete for the profitable business of multinational corporations. First, Unisource was created by major telecom companies in Sweden, the Netherlands, and Switzerland and was later folded into the WorldPartners alliance, which involved about 25 telecom companies. Then there was the Concert alliance, which started out as partnership between British Telecom and MCI, and grew to include almost 50 telcos by 1998. Finally, there was Global One, launched in 1996 by Sprint, Deutsche Telekom, and France Telecom. By the end of the 1990s, all these alliances were in disarray. BT had partnered with AT&T, WorldPartners was largely defunct, and Global One was wracked by financial problems and by disagreements between its German and French parents over Deutsche Telekom’s unsuccessful bid for Telecom Italia. Musical chairs, anyone?

Concentration Can Destroy Value

Even when concentration happens in an industry, it is often unclear whether the trend makes economic sense. To profit from dominating in a concentrating industry, a company needs to extract value by pushing certain economic levers—for example, reducing production costs, reducing risk, or increasing volume. But the problem with these levers is that they are harder to manipulate than they are to identify. (See the sidebar “Levers of Value.”) And deals that do not offer advantages along these dimensions may drastically reduce value because of takeover premiums, transaction costs, and the like.

Levers of Value

Companies that want to dominate in a concentrating industry try to extract value by seeking market advantage along certain dimensions—in other words, they push certain economic levers. But those levers have their limits.

For instance, companies that pursue consolidation to improve industry profitability may have a hard time preventing competitors from freeloading on their investments.

Companies that want to increase production and sales volume through a merger may face cross-country differences in profitability that could reduce the benefits of scale. Increasing customers’ willingness to pay by offering a deeper product line is another reason to merge, but customer preferences may vary across countries, thus limiting margins. And multinational customers can have greater bargaining power than regional customers do.

Reducing costs is another reason to consolidate, but savings are often overestimated and don’t account for take-over premiums. Cross-border adaptation, complexity, and size can actually increase costs. And multinational suppliers can have greater bargaining power than regional suppliers do. Finally, companies may seek to reduce risk through consolidation, but many merger moves are irreversible. And countries may share similar risk profiles that won’t diminish with a merger of businesses in those countries.

Consider the aluminum industry (see the exhibit below), which experienced deconcentration through the last quarter of the twentieth century—until a wave of deals in the summer of 1999. First, there was the proposed three-way merger of Switzerland’s Alusuisse-Lonza Group, France’s Pechiney Group, and Canada’s Alcan Aluminum into APA. Then Alcoa acquired Reynolds Metals. The press releases accompanying both mergers were effusive about the benefits of size. As Alain Belda, Alcoa’s CEO, described the Reynolds acquisition: “The new company will be better positioned to address the ongoing globalization of the metals industry and the new competitive landscape this is creating. It will permit the greater efficiencies and cost reductions required by an environment that has seen the lowest prices in many years for our commodity products.”

Such rationalizations, however, raise serious questions. For a start, the profitability of the aluminum industry is dreadful. Increasing volume doesn’t make a lot of sense unless you can improve margins, which in recent years have generally failed to cover the cost of capital in this sector. Given the commodity nature of the products, it’s hard to increase customers’ willingness to pay, especially since large-scale aluminum buyers have centralized their procurement functions to squeeze suppliers. As for the projected reductions in operating costs, they are relatively modest—in the case of the Alcoa-Reynolds merger, only about 1% of the combined entity’s sales. Finally, the risk-reducing effects of the mergers are limited; they do not significantly expand the companies’ geographic scope outside traditional markets, and they leave the scope of the companies’ products relatively unchanged.

Stories like the aluminum industry’s are common. But then why are cross-border consolidations pursued even when they destroy economic value? It seems there is often a pathology involved. Management appears to suffer from one or more of several motivational and cognitive biases toward mega-mergers, which can lead to irrational decision making and large-scale destruction of value. Let’s consider those biases, because awareness is the first step toward overcoming them.

Top-Line Obsession.

One of the most powerful biases is an overemphasis on growing revenues as opposed to growing profits. That bias is often embedded in a company’s stated goals. How many industries do you know in which the leading players have a 10% to 15% growth target? And how many of those industries have average growth rates of only 2% to 3%? The obsession with top-line growth also appears in compensation systems, where growth drives financial and other rewards—for example, opportunities for promotion. In still other cases, nothing more than bragging rights are at stake. The aluminum example is suggestive in this regard: the APA alliance threatened to dethrone Alcoa as the world’s largest aluminum producer, but buying Reynolds allowed Alcoa to retain the top spot over the proposed three-way merger of its rivals.

Stock Price Exploitation.

Another common bias is a desire to exploit a company’s high stock price. It has long been known that stock prices tend to fuel mergers. There are two reasons for this. First, companies can afford a merger—they have overvalued stock that they can use to finance global industry rollups. For example, Alcoa first expressed its interest in acquiring Reynolds in spring 1999 after its stock price had crossed the $60 mark—twice the level at which it had been trading in fall 1998. And second, companies sometimes see a megadeal as a way to maintain the popularity of their shares. Take the pharmaceutical industry. Leading drug companies that have engaged in megamergers hope their share prices will be buoyed for at least two or three years by all the talk of the operating benefits that will arise from consolidation. Yet a recent study by management consulting firm AT Kearney concludes that merged pharmaceutical companies create significantly less economic value than independent pharmaceutical players.

Grooved Thinking.

Even when a company doesn’t have undue bias toward top-line growth, an industry’s mind-set can get stuck. Consider the telecommunications industry. The telecom companies’ common desire to maximize the number of telephone lines they control has driven many recent mergers and deals. But data communications are expected to make up as much as 95% of telecommunications traffic by 2010, and there is no consensus about what kind of architecture will accommodate that massive increase in data and video transmissions. Telecom companies have enjoyed their status as regulated, protected monopolies for decades. Now, when faced with the double threat of deregulation and rapid technological change, they are finding it hard to change. Conducting business as usual seems much easier.

Herd Behavior.

Firms in oligopolistic industries often play follow the leader. Imagine that you manage a European bank, and many of your larger competitors are going pan-European. Even though studies suggest that size tends to increase complexity rather than savings, you’re uncomfortable being the odd man out. It’s hard nowadays to say, “No, this consolidation stuff isn’t grounded in reality. My company will sit this wave out.” Managers are far more likely to join the stampede, either to signal that they are “with it” or to hide in the herd. This pack mentality is often reinforced by corporate incentive systems that measure and reward CEOs based on how well the company does relative to its competitors.

Personal Commitment.

Senior managers’ personal commitments to a particular theory can often drive a company’s consolidation strategy for a long time—even in the face of persistent underperformance. A classic example is Whirlpool, whose global strategy for major home appliances was initiated by David R. Whitwam when he became the company’s CEO in 1987. As Whitwam put it: “The market is more global than ever and growing more so every day. Global barriers are rapidly deteriorating. So you have to integrate all your operations into a total global strategy.” By the late 1990s, the company was reeling from a triple whammy. First, there was the fallout from Whirlpool having paid too much for NV Philips’ European operations, as well as initial delays in restructuring those units. Second, there was economic chaos in Latin America. And third, Whirlpool was forced to retreat from China as domestic competitors there made significant inroads. Despite these reverses, Whitwam was slow to shift Whirlpool’s strategy because of his deep-seated commitment to global consolidation.

Trust in Interested Parties.

Managers have to examine the motivations of all parties involved in a potential merger or acquisition. As an investment banker bluntly explains, “Our client is doing this $50 billion deal so he can do a $100 billion deal next time.” Could the prospect of large contingency fees—not to mention the assumption that the stock market will rise forever—have influenced this businessperson’s perceptions of what would be best for his client? The better Wall Street firms seem to be aware that this is a dangerous bias and, unusual for the times, tend to be merger-shy.

Alternatives to the Big Deal

In spite of the general skepticism about concentration expressed so far in this article, you may still think it is happening in your industry. (“The Global Concentration Matrix” presents a simple tool to help managers determine that.) But even if you establish beyond a reasonable doubt that your industry is becoming more concentrated, that’s still a far cry from saying that you have to get involved in a global M&A deal. There are a range of alternatives that may offer your company much more value than global consolidation. We can suggest at least seven possibilities.

The Global Concentration Matrix

This matrix can help managers evaluate the imperatives for consolidation in their industry. The columns denote two possible industry scenarios: either the industry is consolidating or it isn’t. The rows denote the basic strategic choices a company has: consolidate or don’t. Now there are four possible outcomes. In each case, the company will have a different result.

Filling out dollar values in the four cells of the matrix sheds powerful light on incentives and disincentives to consolidate. And once the payoff numbers have been filled in, you can figure out how high the probability p has to be for consolidation to be your best strategic choice. For a deal to make sense, the ratio of the probability that consolidation is happening divided by the probability that it isn’t must be greater than the ratio (V₄–V₂)/(V₁–V₃). Do your estimates of V₁ through V₄ and p satisfy this condition? The probability p implied by the dollar values of V₁ through V₄ is sometimes implausibly high.

Besides these quantitative benefits, one of the qualitative benefits of the matrix is that it forces managers to think about how their industries might develop. Managers should compare what an industry will look like in terms of concentration levels, average margins, and so on, if it concentrates globally—or not. It may be useful to look at more than one alternative to consolidation, but remember that matters quickly get more complicated as the number of alternatives multiply.

It is often helpful to refine the analysis several times instead of working through it just once. It forces managers to keep thinking more carefully through time. When will you find out whether the industry scenario that each strategic option was based on is coming to pass or not? What kinds of backtracking costs and lags will you incur with mistaken choices, or will backtracking be impossible? How can you increase flexibility, and what will that cost? Can you influence the probability that a particular industry scenario will come to pass through your choice of strategy? And so on.

Analyzing a major commitment of this kind requires more than just routine financial modeling. It requires detailed industry and competitive analysis, ideally leavened with imagination and creativity.¹

1. For a more detailed discussion, see chapters 6 and 7 in Pankaj Ghemawat’s Commitment: The Dynamic of Strategy (Free Press, 1991).

Pick Up the Scraps.

As Peter Drucker recently noted, for every megamerger or large acquisition, there are usually several spin-offs, divestments, or asset sales that can give companies—especially smaller ones—lots of growth opportunities. The 1998 merger of BP and Amoco, for example, led to the disposal of 12 oil-storage terminals scattered across North America. The terminals were purchased by the Williams Companies, a small business compared with BP Amoco but hardly an insignificant one. In 1998, Williams had nearly $8 billion in sales. The oil industry is becoming more competitive in part because of such cast-off purchases by companies like Williams.

Stay Home.

For many companies, it still makes a lot more sense to grow domestically or regionally than to try and establish a global presence. Think of Maytag and Lloyd’s Bank; both companies have prospered by focusing on their home markets of the United States and Great Britain respectively. Telefónica de España, a midsize telecommunications company, has rejected the role of global consolidator as being too expensive. Instead, it has focused on building a strong regional presence in Latin America. Even if Telefónica has to sell its assets, or itself, its regional position will command a high price.

Keep Your Eye on the Ball.

A big deal takes lots of time and consumes lots of managerial attention. If others in your industry are busy with their mega-deals, you can exploit that fact to improve your own competitive position. For instance, while pharmaceutical players such as Glaxo Wellcome and SmithKline Beecham have made headlines with their merger news, one large competitor, Merck, has so far refrained from making any such moves itself. According to industry observers, Merck intends to use this period in which its competitors are busy with postmerger integration to improve its position through aggressive marketing and other initiatives.

Make Friends.

Another alternative to the mega-deal is to build scale through relationships. In many cases, partnerships are a more appropriate way to grow than M&A. Even a company as large as IBM uses alliances in its PC business, which has been deemed unprofitable but still important given its links with other IBM businesses. In 1996, IBM entered into a standing agreement to purchase low-priced PCs from Acer to resell under the IBM brand. In many cases, alliances are easier to bring about than acquisitions because they encounter less resistance within companies, from the government, and from other interested parties.

Appeal to the Referee.

If your competitors’ mega-deals may actually hurt you, and if you can’t respond adequately on your own, you can put a spanner in the works by forcing regulators in your industry to institute antitrust proceedings. For example, among telecommunications companies, the nonconsolidators have bought some time by bringing in regulators to examine the legality of deals such as Bell Atlantic’s acquisition of GTE and AT&T’s acquisition of MediaOne. The current rash of international antitrust cases—such as those in Europe around MCI-WorldCom’s purchase of Sprint and Vodafone’s acquisition of Mannesmann—suggests this strategy may also be effective outside the United States.

Stalk Your Target.

Even if it seems to make sense for your organization to pursue global consolidation, ask yourself if your industry offers significant first-mover advantages. If it doesn’t, it may be smarter for you to let someone else go ahead and clear a path. Take the case of Tricon, the owner-franchiser of the KFC, Pizza Hut, and Taco Bell fast-food chains. Tricon’s international strategy explicitly targets markets in which McDonald’s has already established a significant presence because that’s an indicator of the potential for its own chains. Also, McDonald’s has already shouldered some of the costs of establishing quick-service restaurant formats locally.

Sell Out.

Even in cases where the big deal makes sense, it may be better for your company’s shareholders if you’re the seller rather than the buyer. America Online’s proposed acquisition of Time Warner is a case in point. Time Warner’s shareholders did well in the immediate aftermath of the announcement; many of them rushed to cash out. AOL’s shareholders, by contrast, did not do as well. Some think AOL will eventually recover what it paid, but others believe this may be the deal that brings some rationality to the valuation of Internet stocks.• • •

These are just some generic strategies, of course. Your viability as a nonconsolidator will depend directly on the analysis and imagination you bring to the table as you devise alternatives to buying up shops. In many cases, consolidation strategies are wrong. In an era that is witnessing technological discontinuities, managers must not focus so much on size as a goal but rather on the development of new business models that help them compete.