Running a basic-materials company can be thankless. Whether it competes in pulp and paper, in energy, in chemicals, or in mining and metals, it inevitably confronts mature markets, intense global competition, and continual pressure to cut costs. To make the task more trying, however hard such a company struggles to make its products distinctive, those products are invariably saddled with the dread "C" word: commodity.

Steel producers face the greatest challenge of all. Over the past decade, the global industry has earned only a 4 percent annual operating return on assets—half the rate of aluminum and pulp and paper. The entry of new players explains the low returns. Steel companies—including such market leaders as Posco (South Korea), China Steel (Taiwan), and Techint (Argentina)—have emerged as local champions in developing countries with relatively high rates of economic growth. Meanwhile, technological change has lowered barriers to entry for such companies as Nucor (United States), Riva (Italy), and Ispat International (based originally in Indonesia and now in the Netherlands).

The result is that almost 30 percent of the world’s steel now comes from new plants belonging to companies that didn’t exist 25 years ago. Such upstarts have entered a global market that since 1980 has grown by less than 1 percent a year—an average combining annual growth of 4.7 percent in the developing world, 0.5 percent in the developed world, and -5.5 percent in Eastern Europe and the former Soviet Union. During that period, operating rates for producers in Europe, Japan, and North America have averaged only 80 percent of capacity, a rate that barely permits them to make a profit and thus creates intense pressure to cut costs.

Paradoxically, continual cost cutting has tended to trap the industry in a vicious cycle of excess capacity and poor returns. For most companies, the easiest way to reduce costs is to eliminate bottlenecks, either through incremental investments or changes in operating practices. Such efforts provide economies of scale but also increase capacity by some 1.5 percent a year, more than twice the market’s growth rate. They are essential for companies that wish to survive but reinforce the underlying structural problems facing the industry as a whole.

Initiatives that go beyond this conventional response are needed. The most promising of them link the themes of consolidation and specialization. Consolidation, in the form of mergers and acquisitions, offers potential synergies and a chance to rationalize the industry’s structure and to control self-defeating behavior—for instance, the duplication of investment by smaller companies. Specialization, whether in products, technical processes, or functional capabilities, offers an opportunity for differentiation and thus a defense against the industry’s poor fundamentals. Which of these paths offers a better opportunity to overcome the vicious cycle of excess capacity?

The case for consolidation

Many steel executives view consolidation as the more promising response because it reflects the trend toward consolidation in industries, like automotive, that use steel products or, like aluminum, compete with them. Consolidation has advanced furthest in Europe, where privatization has created an active market. As privatized companies become more efficient, competitors have been forced to recognize the potential benefits of scale, which in turn promotes further concentration. A similar dynamic is apparent in South America, where nationalized steel industries have been broken up and replaced by networks of affiliated companies spanning the continent. In Japan, a few giants have long dominated the industry. In each of these regions, the four largest enterprises generate more than half of total output.

North America is the exception. Its long-fragmented steel industry is actually becoming more fragmented, mainly because of the minimills; indeed, the share of production claimed by the four largest US companies fell to one-third in 1998, from more than half in 1985. Yet in the United States as well, the air rings with talk of consolidation: senior executives of US companies, for example, have explicitly admitted the need for it at industry conferences and in discussions with securities analysts. It hasn’t taken place so far mainly as a result of factors such as the weak balance sheets of most traditional US steel companies. But the current structure of the US industry probably will not endure now that leading competitors in other continents have global aspirations backed up by revenue bases that are up to three times bigger than those of the largest North American companies.

Consolidation has a twofold logic: direct cost and revenue synergies, and improved industry fundamentals

Consolidation has a twofold logic: first, direct cost and revenue synergies, beating the vicious cycle; and, second, improved industry fundamentals, breaking it. Purely scale-related synergies might amount to between 1 and 2 percentage points in the return on consolidated sales if the merged entities are roughly the same size. Cost reductions, such as overhead consolidation and purchasing, typically provide around a 1-point improvement, again assuming a merger of equals. Opportunities to increase revenues through devices like cross-selling can give the combined entity another percentage point in return on sales. But customers may seek alternative suppliers, and competitors may try to gain market share by exploiting the disturbance of established relationships and expectations, thereby diluting revenue-related benefits.

Transferring best practices represents the most significant potential form of synergy; indeed, this has been the chief way Ispat, the industry’s most successful acquirer, has created value for its shareholders. The scale of the impact on costs and revenues depends on the difference in skills between the two companies and the ease (or otherwise) of transferring them, but the return on sales could rise by as much as 2 to 3 percentage points. An optimal merger might generate an improvement in gross profits equal to about 4 percent of the combined entity’s return on sales—a substantial improvement. Against this must be weighed the deal’s transaction costs (including any premium), which might reduce the gross benefit by 1 or 2 percentage points; less than optimal fit and execution can diminish the benefits as well. Nonetheless, the most direct and persuasive argument for consolidation is the possibility of improved margins from gains in efficiency.

The second argument for consolidation is the desirability of cutting the number of competing enterprises. Today, few companies can afford to consider industry-wide benefits in their internal decision making. Closing outmoded plants rapidly and limiting new investment, for example, would restrain supply and thus help alleviate the problem of excess capacity; investments in market development—helping steel gain share at the expense of aluminum, concrete, and plastics—would boost demand. Unfortunately, the industry is now so fragmented that the benefits any individual company generates from such initiatives flow overwhelmingly to competitors. As a result, these decisions are not made, and the vicious cycle of excess capacity persists. Larger companies, so the argument goes, could make decisions that benefit their own shareholders, the industry as a whole, and even their customers.

Improved conduct within the industry—more rapid closure of redundant capacity, more disciplined investment, and more successful market development—could generate results equal to 3 points in return on sales. These benefits, quite importantly, reflect better asset utilization rather than anticompetitive measures or price increases (Exhibit 1). Together with the increase contributed by direct synergies, conduct-related gains imply that consolidation could generate improvements in profit exceeding the industry’s average returns over the past decade.

However, what we have modeled in these estimates are changes in behavior, which doesn’t answer the question of how such changes might actually come about. Steel companies would have to be substantially larger than they are now to generate these benefits. Indeed, for improved conduct to break the vicious cycle, leading companies would have to manage about 50 million metric tons of capacity—some 2.5 and 5 times the output of today’s largest producers in Europe and North America, respectively. It remains to be seen whether companies will have the vision to pursue such targets, or governments the willingness to accept them.

The case for specialization

Whatever the difficulties, the potential benefits of consolidation suggest that all leading steel companies should consider it. Nonetheless, those that have earned good returns over the past decade have been relatively small and highly focused. Exhibit 2 documents the financial performance of 50 large steel companies from 1988 to 1997, as measured by returns (operating profit on operating assets) and revenue growth. Of the handful of companies that achieved strong profits and growth over this period, most specialized in some way (Exhibit 3). None of the traditional industry leaders—large, integrated companies such as Nippon Steel (Japan), U.S. Steel, and British Steel—makes it into this group.

Specialization can involve focusing on any one of three dimensions: specific product or market niches, specific steps in the value chain, or specific functions, such as environmental or maintenance services. Böhler-Uddeholm (Austria) and Carpenter Technology (the United States), for example, have succeeded by focusing on specific product or market niches. Companies that take this route can typically build technical or organizational barriers to entry, such as patents or proprietary distribution channels, respectively.

At first glance, this model might not seem applicable to larger entities whose production capacity exceeds the maximum potential demand of a regional niche market. But larger companies can succeed as niche players in selected markets. Consider the case of the Swedish integrated producer SSAB. In the late 1980s, it was a distressed state-owned enterprise. Lately, it has become a strong value creator by achieving global leadership in heat-treated plate, used in applications for earth-moving equipment and the like. Although these niche products account for less than 20 percent of SSAB’s sales, they have such distinctive value in the market that they provide more than half of the company’s profits, which are among the best in the industry. On the commercial front, SSAB maintains a separate sales force for these particular products and distinguishes itself from its competitors by distributing them itself. As for operations, SSAB has developed distinctive skills for making these products thanks to the large and growing demand for them. This "virtuous cycle" of commercial and operational performance has helped SSAB brand its heat-treated plate products, Weldox and Hardox.

Larger companies could be even better equipped to specialize in specific stages of the value chain, the second kind of specialization. Companies shaped by their upstream operations, which must be vast to be efficient, often have difficulty concentrating on the needs of customers in the final stages of processing. Could such companies boost the profitability of the value chain as a whole by breaking it apart? Rather than handle the final processing themselves, it might make sense for them to ship semifinished and relatively fungible products such as slabs or hot bands to stand-alone finishing customers, which could then adapt those products to particular applications. Just as important, specialization would make it easier for the industry to manage the vicious cycle, since the core of the excess-capacity problem lies in upstream (smelting) operations.

Specific functions, the industry’s third area of specialization, can be outsourced to reduce costs or developed as growth businesses. Although more than half of the value added in steel operations could be outsourced, less than half actually is. A "virtual" steel company might have an original business concept developed by consumers, government authorities, and investors, with plant designed and built by engineering firms and equipment suppliers and operated by contract operators, maintenance providers, and providers of overhead services. Output would be allocated by contract, perhaps to a part owner, and marketed by trading companies, distributors, and brokers. Large parts of this model are already being adopted in the developing world—for instance, by new slab-making ventures in Mozambique and Venezuela. Companies operating as service providers to the steel industry often earn higher returns than their customers, partly because they are less asset-intensive and partly because successful functional specialists have distinctive capabilities (Exhibit 4).

Balancing consolidation and specialization

Consolidators and specialists will increasingly dominate the global steel industry, and this will provoke a thorough restructuring. In the process, the huge middle ground—geographic incumbents, including household names like U.S. Steel and BHP Steel (Australia) that are large only in the context of their regions—will have to move toward one role or the other (Exhibit 5). Choosing to remain an undifferentiated regional player will ensure only average returns, an unattractive option given the persistence of the vicious cycle.

Over the next decade, geographic incumbents will probably see their shares of the worldwide market fall as low as 10 percent, from 40 percent today. Most of this loss will come at the hands of consolidators, whose average annual production will grow to 30 million metric tons or more, surpassing the output of today’s biggest companies. The obvious candidates to drive this process are the large regional leaders, such as Ispat, Pohang (South Korea), Nippon Steel and Kawasaki Steel (Japan), Thyssen-Krupp (Germany), Usinor (France), Arbed (Luxembourg), and BSKH (the United Kingdom and the Netherlands).

Some geographic incumbents may specialize as well. Indeed, we can already identify several companies—AK Steel (the United States) and Techint—that produce a wide range of products but have differentiated themselves in some way and will move toward further specialization. The trend will embrace not only mergers of whole companies but also discrete segments of the industry value chain, thus combining the themes of specialization and consolidation. This is most likely to happen in upstream operations, where slow market growth and high capital intensiveness have forced down returns; a business that limited itself to upstream operations conducted on a large scale could probably improve on them. The slab or hot-band units of Usinor and Thyssen-Krupp, for example, might be combined and operated as a utility serving various downstream operations.

The resulting improvements in overhead, asset utilization, and investment could improve returns on upstream assets by up to 3 percentage points, almost matching the world industry’s average return on all assets over the past ten years. Moreover, the economies of scale and superior capacity management such a utility could achieve might enable it to become the main supplier of semifinished products to specialists, which would increasingly seek to divert managerial and financial resources from value-destroying upstream operations.

Specialization by function will help to break up the current structure of the steel industry

Specialization by function will help break up the industry’s current structure. A company that develops, for internal use, distinctive capabilities in an important functional area, such as maintenance or energy management, will be able to market them as services to other producers; Nucor, for instance, could probably build a service business on its international reputation for outstanding operational and plant-building skills. Similarly, SSAB’s success in establishing Weldox and Hardox suggests that it might be able to manage the brands of other steel companies, perhaps even products it doesn’t make itself.

Choosing the right approach

How can steel companies adapt to these trends? Successful ones are likely to balance consolidation and specialization. Specialists will seek to grow by consolidating globally around their areas of focus. Consolidators will aim to offset the bureaucratic tendencies of large enterprises by creating more specialized units that can respond to market signals and earn rewards accordingly. The "best of both worlds" is thus an appropriate aspiration. Nevertheless, we believe that specialization is the more appropriate starting point for most companies.

In the steel industry, the case for specialization is frequently drowned out by investment banks and business publications promoting consolidation as a strategic imperative. Moreover, the idea that bigger (as reckoned by facilities, output, or the size of headquarters) is better has always been an article of faith for the industry. And companies can grow through acquisitions without dramatically challenging the business-as-usual mindset.

Specialization, by contrast, entails wrenching organizational changes and substantial risks. Yet the market often rewards companies that can transform themselves in this way with high share multiples, which can finance further investment in specialized assets and related acquisitions. Although the sheer scale and complexity of traditional steel plants can make specialization seem impractical, there are ways out: if the home market is too small to support a specialist strategy, think globally; if an enterprise is too large or too complex to achieve adequate focus, break it up.

Ultimately, our preference for the path of specialization reflects a concern that the industry’s traditional business model—the integrated firm that starts with dirt in the ground and ends with fabricated products—is more a part of the problem than a part of the solution. Indeed, this model has been crumbling for most of the years since the end of World War II. Given the industry’s continuing low returns, why not consider radical solutions? Instead of resisting a "disintegrated" future, why not embrace it?