Call it fashion, call it fear of being left behind, call it opportunism -- whatever the reason, every chief executive worth his salt, it seems, is itching to find a company to buy or to merge with. The world's biggest three-way auction, which pitted Worldcom against British Telecom and GTE for the hand of MCI, and the proposed marriage of two pairs of the Big Six accounting firms are just the biggest in what looks like a new wave of megamergers sweeping the United States and Britain. Nor is continental Europe, traditionally resistant to the finance-driven strain of Anglo-Saxon capitalism, immune: in one week last September, no fewer than four large cross-border mergers were announced in France, Germany and Sweden.
Of course, some mergers are justifiable in strategic terms. A few -- although the research says not many -- actually benefit the buyer's shareholders as well as the seller's. But we believe that, paradoxical as it sounds, most companies have more value to gain from corporate subtraction than corporate addition. Instead of seeking to create value by expanding their empires through mergers and acquisitions, managers should be looking at the potential gain in breaking their companies up. In other words, instead of basing their strategy on the formula 1 + 1 = 3, managers should be thinking in terms of a radically different formula: 1 1 = 3.
Note that this is not an asset-stripping or financial-engineering proposition. Our agenda is a management one. Most large companies are too diversified for their own good. There are sound reasons why managing a portfolio of companies covering a number of business sectors is harder than managing companies within a single sector, or, to turn it around, why managers of a focused group of companies can more easily create value than their counterparts in a nonfocused one.
We would go further: as this realization spreads, it will trigger a wave of corporate restructuring that makes the current mergers look like a drop in the ocean. Our research indicates that $1 trillion of extra value could be released in the United States and Britain alone by companies that embrace the breakup regime. More than 50 percent of the largest companies on both sides of the Atlantic should be actively considering breaking themselves up, according to our calculations.
The reason that breakup is such a powerful idea is simple -- and for senior managers of large companies, disturbing. The central management of most large companies destroys more value than it creates.
COMPANIES AS VALUE- DESTROYING MACHINES
To understand why, we need to step back and consider how our dominant companies grew to reach their present size. Most of the top companies in both the United States and Western Europe are multibusiness companies. That is, when the original successful business generated surplus cash, ambitious managers at the corporate center judged they could replicate that success or spread risk or obtain balance by expanding into adjacent business areas. Over time, they built their small, focused companies into large diversified ones, often in more than one industry, spread over several continents. Of the 50 biggest United States corporations, 40 operate in more than one business segment.
Now consider the corporate centers of these companies -- the managers at corporate headquarters and in the divisions who oversee the operating businesses. Corporate centers are curious things. They do not have customers and thus cannot generate wealth. They have costs but not revenues. To outweigh these costs, they must provide some service to the companies in the corporation that the latter need and could not achieve on their own. They must add value to the operating companies.
Now, theoretically speaking, it is not hard for a center to add value. On the financial side, for example, a combined group can probably reduce the cost of borrowing and save on specialist financial administration and act as a liaison with the capital markets. A wise center can also add value through advice and guidance, by attracting and appointing better managers, by providing training, through centralized purchasing and by throwing corporate resources behind promising ventures in the operating companies. These are important advantages. By these and other means almost any center, even if not very well run, is bound to add some value.
The trouble is that most centers destroy more -- substantially more -- than they create. The Ashridge Strategic Management Center, an academic research unit dedicated to understanding multibusiness companies, has been studying value creation and destruction for 10 years. The results are unequivocal: while a few multibusiness companies (including, interestingly, some conglomerates) create substantial net value, the large majority are net destroyers in at least part of their portfolio. We estimate that on average corporate centers destroy between 10 percent and 50 percent of a corporation's entire value.
The idea that a corporation's operating companies would be a tenth or even 50 percent better off without a center to guide them is startling and at first sight improbable. But backing up anecdotal evidence, such as the tangible value being delivered by breakups and the fact that stock markets often value highly diversified companies at less than the total of their parts (in other words, they judge that the center is worth less than nothing), our research shows that there are as many ways for central managers to hold back or harm their subsidiaries as to improve them. What's more, they are often invisible. It is only when they are held up to the light that it becomes clear that the net effect is negative -- often very negative indeed.
FOUR KINDS OF VALUE DESTRUCTION
Consider the oil companies. When Exxon, Shell and British Petroleum diversified into minerals in the 1980's, it seemed a sensible choice. Minerals and oil appeared to be two of a kind, involving exploration, extraction, tricky political negotiations, technically complex projects and large amounts of capital. Commentators agreed.
Yet after a decade of disillusionment, not a single oil company has made a success of minerals. Most companies have tossed in the towel. On average, during the mid-1980's the minerals businesses of Atlantic Richfield, B.P., Exxon and Shell were earning a pretax return on sales of minus 17 percent. At the same time, independent minerals companies were making plus 10 percent.
With hindsight, the similarities between oil and minerals were deceptive. Success factors were quite different. In fact, the oil majors have been unsuccessful not just in minerals but in practically every other attempted diversification. Given the number of tries, this cannot have been just a matter of bad luck. The outcome can only have been the result of value destruction of truly heroic consistency and comprehensiveness.
The experience of the oil companies dramatically demonstrates Value Destroyer No. 1: the potentially disastrous effects of central executive influence on the companies it is in charge of. Put crudely, the oil company centers did not understand the minerals business and imposed on the companies worse decisions than they would have made on their own. Logically, this is hardly surprising. Why should an oil-company chief executive, in 10 percent of his or her time, be able to see better strategies for a minerals business than energetic managers who have devoted 100 percent of their time to the sector and its nuances for years?
The delusion is the greater since the chief executive depends for information on the operating managers in the first place. By definition, that information will not only be incomplete; it will also be subtly distorted because there are so many insidious incentives for managers to withhold or "manipulate" the data (particularly at budget and bonus time). For all the good intentions, influence and action derived from skewed information can hardly help being flawed.
Or consider the LBO/MBO experience. Why do buyouts have such a strong overall record of success? Because they work with the grain of a simple truth. However enlightened the parent, most people work harder and smarter if they are in charge of their own destinies. Improvisation, going the extra mile, adapting quickly to change are all more natural and effective when managers are free to do it in their own way. By shifting management control and eliminating the corporate center, a buyout forces the business to grow up and make its own decisions -- almost always leading to radical changes in behavior.
The second powerful destructive element -- Value Destroyer No. 2 -- is the pursuit of synergy. Synergy has always been one of the most elusive concepts in management, dazzlingly attractive in theory, consistently disappointing in practice. Our research shows why. Most synergies are mirages, the product of corporate wishful thinking. There are powerful reasons that corporate managers find the idea of synergies irresistible -- after all, academics and consultants have taught them that is what the center is there for. But our investigations demonstrate that this is looking through the wrong end of the telescope. In ordinary circumstances, the self-interested, profit-seeking behavior that is the whole basis of our successful market system will insure that independent managers cooperate to capture available synergy benefits. Only in special cases, therefore, should the center expect to create value by intervening in what motivated managers would do on their own account. Think of it like this: to create value from synergies, the corporate center must first overcome its natural information disadvantage, then find areas where enlightened self-interest is not working, next confirm that the benefits are not a mirage and finally bring the parties together in a successful transaction. Each of these steps by itself involves difficult judgments. Together, they are a tall order. Not surprisingly, there are more failures than successes.
Value Destroyer No. 3 is the behavior of corporate staffs. The central functions -- finance, treasury, public relations, legal and human resources -- are all well-documented sources of frustration for the operating businesses. Their one-size-fits-all policies are often unsuited to parts of the corporation. Their privileged position makes the services they provide unresponsive or uncompetitive or both. Information technology departments for instance often reduce the competitiveness of the business units they are supposed to serve (which is why so many of them are being hived off in outsourcing deals). Finally, the overhead that they impose in terms of unwanted policies and continual demands for information are actively disempowering for business unit managers. These influences are pervasive and, by the nature of the multibusiness company, almost impossible to avoid.
Last -- Value Destroyer No. 4 -- corporate centers destroy value by their efforts to develop the group portfolio, especially by their attempts to grow by acquisition. Countless academic studies have been carried out on the subject of acquisitions. The conclusions are consistent and irrefutable: most acquisitions destroy value for the purchasing company. Any value released by the change of ownership goes to the seller, not the buyer. Part of the reason is the well-established "acquisition premium" -- companies usually have to overpay to persuade the seller to sell. But, as we can now see, this is only the half of it. To justify the premium paid, the buyer intervenes to try to improve operations -- by (guess what) exerting executive influence, launching synergy initiatives and applying the expertise of central staffs -- thus compounding the initial error. AT&T's disastrous excursion into computers is a good example of these forces at work. So until recently was Sony's Hollywood venture. Rough calculations indicate that Sony paid up to $2 billion more than it should have for the ailing Columbia Studios. A few billion more disappeared as the parent company chose the wrong people, exercised the wrong controls and pursued wrongheaded ideas about hardware-software synergies. Although Sony is now making money, it has a long way to go to recoup the $4 billion to $8 billion it previously lost -- a truly Oscar-winning demonstration of how acquisitions wreck value.
As should be clear from these examples, far from being an isolated occurrence, value destruction is endemic to the multibusiness company. It is not primarily a question of management incompetence or ill will, but part of the fabric and structure of the corporation. Our research shows that most companies -- not just some -- are subject to these kinds of value destruction. They are overdiversified, and on balance the corporate center does more harm than good. There is an implacable corollary to this finding. To release the value that is currently being destroyed and to prevent it from being eroded in the future, value-destroying companies should break themselves up, or at least put a breakup on the agenda.
PUTTING BREAKUP ON THE AGENDA
For many managers, whose whole careers have been built on the assumption that success involves mounting the ladder to power at corporate headquarters, this is a tough proposition even to consider. Like men with chest pains who avoid going to see the doctor, they prefer not to think about the possibility of bad news. To bring objectivity to an emotive issue, we offer a four-part self-diagnostic, designed to bring out the hard facts that will either confirm the health of the current portfolio or cause corporate managers to ask themselves searching questions about their roles and the lifestyle of the present grouping.
What are the natural clusters in the portfolio? Some businesses naturally go together. They are so similar that they respond to the same managerial influences. Two McDonald's franchises form a natural cluster, as do businesses manufacturing similar kinds of metal parts for similar kinds of customers. The feature of a natural cluster is that the businesses have similar critical success factors and similar opportunities for improvement.
In every business, there are a few things that must be done really well for a company to succeed. These are the critical success factors. In the oil exploration business, for example, negotiating with governments, technical exploration skills, project management and the ability to mobilize huge amounts of capital are all critical to success. It is in these areas that leading companies distinguish themselves from their rivals. As it turns out, these are quite different from the things that make for success in minerals. In minerals the key to success is low-cost deposits, which are often better accessed through joint venturing than through exploration. Despite some superficial similarities, oil and minerals are not natural bedfellows: they do not respond to the same managerial influences.
Clustering by critical success factors is no science, and reasonable people can disagree about how similar the businesses are. The purpose of the exercise is not to arrive at a precise result but rather to take the current portfolio of businesses and group it into clusters of similar types. The aim is to create some separate clusters rather than to agonize about which business fits into which cluster. In a portfolio of food-manufacturing businesses, the clusters might be frozen foods, ambient foods and fresh foods -- or they might be premium foods and standard-quality foods. Normally, the way a company is currently organized into divisions is a good indication of what the natural clusters are likely to be.
A second dimension of the clustering is about opportunities to improve. Think of the center as a doctor. If all the doctor's patients have similar health problems, it is easier for her to diagnose accurately and develop specialist remedies for her patients. If each problem is different, however, she will need orders of magnitude more time, effort and resources to develop effective remedies.
Improvement opportunities exist in all shapes and sizes. The easiest way to get at them is to examine existing business plans. What are the major tasks facing the management teams? Is there a reason to suppose that a business's managers will underperform in any of them? If so, which and why? Clusters can then be developed of businesses with similar areas of likely underperformance -- much as a doctor might group patients with asthma together, separately from those with broken limbs.
Inevitably, the clustering by opportunities to improve will give some slightly different combinations than clustering by critical success factors. Here again, our advice is to take a broad-brush approach rather than a precision analysis approach. The objective is to divide the portfolio into different kinds of clusters. If more than one way of clustering is evident, it may be best to try looking at the portfolio in both ways to see if one concept stands up to the subsequent steps of analysis better than another.
* Does the center fit with each cluster? The purpose of cluster analysis is to identify groups of businesses with similar success factors and similar improvement opportunities. The next issue to address is whether the center has the specialist skills to be a good parent of each of the clusters.
The traditional response was to dismiss this question as irrelevant or unimportant: general managers are capable of handling anything, and if the center lacks some needed skills, they can be bought in. To the contrary, our research says clearly that corporate centers, like individual managers, cannot be good at everything. They must specialize. If managers with the required corporate-center skills were freely available, the improvement opportunities would not exist -- the businesses would already be improved. Moreover, managers with different skill sets do not sit easily together, especially in a small corporate center. Rationalizers and cost cutters have different beliefs and values from business builders. Retailers think differently from manufacturers. Imagining that they can work together without a) debilitating disagreement or b) constraining compromise, is naive.
Judging whether skills at the center fit the needs of a cluster, however, is hard; and who is to make the call? Fortunately, some objective analysis can be done. An analysis of the successes and failures of the last 5 or 10 years can throw up compelling data on what the center is good or bad at. Additional objectivity can be gained by asking the opinions of business-unit managers -- who are normally vocal about the strengths and weaknesses of their corporate masters. Finally, some process of forced ranking of the clusters by degree of fit will identify those that conform least well.
Is there a logic for keeping the clusters together? For many years, the British chemicals company Imperial Chemical Industries (I.C.I.) argued that there were major technology synergies that justified keeping its chemical and pharmaceutical operations under common ownership. In 1991, Britain's most feared corporate raider, Lord Hanson, bought a 2.8 percent stake in the company. For some months, a full bid appeared possible but was repelled, courtesy of the synergy argument. Shortly afterwards, however, I.C.I. reversed the logic and spun off its life-sciences division, Zeneca. Now, it claimed, managing two such different businesses had been holding both back. (In light of market response to the demerger, this was true.)
As the I.C.I. example suggests, many traditional arguments for diversification -- or owning different clusters -- do not hold up under scrutiny. Here are some frequently encountered statements that make us feel uncomfortable: "If we broke up, we would be too small." "We need diversification to provide growth opportunities." "We don't want all our eggs in one basket." "Although we are good at x, the future is in y, so we must be in y too." "We need cyclical balance." "By being widely spread we can save overhead." Although common, such remarks as these are not just unsatisfactory reasons for diversifying; they are also likely to lead to value destruction and are a strong indication that breakup should be on the corporate agenda.
Of course, there are many successful companies with more than one cluster in their portfolio. Take Procter & Gamble and Unilever, for example, clearly multi-cluster businesses. And what about General Electric, perhaps the most admired company in the world, or ABB -- both multibusiness companies if not conglomerates? Or, even more extreme, with clusters ranging from travel to retailing, media to financial services, Virgin?
The answer is that all these companies have developed ways of adding value that are relevant to a broad range of clusters. All of these companies have their limitations, as G.E. demonstrated when it ventured into investment banking. However, the medicine that these companies have developed has proved suitable to a range of clusters, and the management system they employ is sufficiently decentralized to reduce the dangers of value destruction. Their exceptional performance is just that, however, and strongly underlines our fundamental point: to pass the multicluster test takes a logic that goes far beyond conventional notions of synergy, spread or balance.
Does the multicluster logic stand up to challenge? It will be clear enough by now that we believe multicluster strategies should be viewed with great skepticism. If managers still insist on such strategies, their reasoning should be subject to intense questioning. Why do they believe they can buck the usual rules? Will investors buy the argument? Does past performance support it? What hard evidence can be brought in favor?
One obvious witness is the shareholder verdict. If the company is worth more than the breakup value of the businesses, shareholders implicitly believe that central management is adding value sufficient to justify the premium. If management complains that the company's shares are undervalued -- and particularly if the whole is worth less than sum of the parts -- then the market is giving central management the thumbs down. The corporation may be worth more dead than alive.
Another tell-tale sign is past performance. It is easy for managers to become emotionally committed to their ideas. Simple performance analysis, benchmarking a company's figures against those of rivals, often comes as a shock. At an international leisure company, a senior planner was surprised to find that the corporation's high-profile acquisitions in a number of areas were comprehensively outperformed by a peer group of good independents. Despite the evidence of value destruction, however, he was unable to slow down the pace of acquisition. "Growth ambitions have their own momentum," he explained, "and we seem to be doing fine at the moment." As you would expect, we do not expect this company to be "doing fine" for very long.
THE FOCUSED FUTURE
As should now be abundantly clear, we believe that for many companies, the answers to the four questions will confront managers with a radical decision: breaking up the corporation or at least breaking off one or more clusters. One measure of the pressures building on them is the contortions that companies are going through to justify their diversity with focus-sounding phrases that in fact mean something quite different: "leisure and related businesses," "chemical and speciality products," "electrical, electromechanical and electronic products." None of these justify diversity; they merely hide it.
Do not be fooled. The multibusiness companies of today will survive in the long term only if they can demonstrate that they consistently add more value than they subtract. Few pass this test in 1998; even fewer will do so in the new millennium, as competitive pressures continuously raise the performance bar.
Some multibusiness companies will live on, but they will be so skilled at what they do that they will in effect become focused-business companies. Some, like Unilever, will be so much better at marketing fast-moving consumer goods that they will more than compensate for being spread over several clusters. Some, like Virgin, will be so good at image building and using Richard Branson's brand that they will find improvement opportunities where others would reap only value destruction. Others again, like Emerson and Kohlberg Kravis Roberts, will find a niche for their particular corporate-center skills, seeking out businesses that need them, acquiring them and passing them on when they have no more value to add. For the majority of companies that are overdiversified, however, radical solutions are needed. If managers do not do the arithmetic themselves, the market will do it for them.
The Virgin Group is a $3 billion family of private companies, largely majority-owned by Richard Branson, the group's founder and head. Virgin consists of a number of "clusters" of related concerns. For example, a travel cluster contains two airlines, an aviation services business and a travel company. A trading cluster includes vodka and cola. An entertainment cluster has cinemas, music "megastores," radio stations, a record label and film interests. There is a financial-services cluster along with many other businesses ranging from a ballooning company to a railway and from bridal dresses to nightclubs.
Virgin looks like a ragbag, with all the value-destruction potential that conglomeration usually entails. In fact, Mr. Branson succeeds in adding value to all the group's diverse businesses using a recipe with four ingredients: the Virgin brand, the group's public relations and marketing skills, its understanding of the opportunity presented by "institutionalized" markets, and its experience with green-field start-ups. In Virgin lingo, an institutionalized market is one dominated by a few competitors who fail to give good value to customers because they have become inefficient over time or overabsorbed with one another. The Virgin brand enables the group to scale normally insurmountable barriers to entry, and its low-cost marketing skills insure an excellent cost structure even where the product is produced by a third party.
Mr. Branson and his two-man business-development team review about 50 business proposals a week. At any one time they have about four new prospects under discussion. Good prospects are those that address institutionalized markets, fit the Virgin brand (genuine and fun, contemporary and different, consumers' champion and first class at business-class prices), will respond to the Virgin recipe, offer an enticing reward-to-risk ratio and are presented by a capable management team. Mr. Branson has been less successful when he has had a good idea and has gone out to look for a manager to run it. The best proposals have come to him from managers who want to run the business themselves.
Mr. Branson avoids all four value destroyers by maintaining an exceptionally decentralized structure. The Virgin brand is controlled by licensing agreements with each business, and Mr. Branson's interests are protected by insuring he almost invariably has an ownership stake of 50 percent or greater. Apart from these controls, the businesses are run as independent companies by their own boards of directors. The corporate headquarters, if it can be called that, contains fewer than 20 people. Virgin has little executive influence, no corporate staffs, no desire to push synergies and little appetite for acquisitions. The recipe Mr. Branson offers his businesses (the brand, marketing, public relations and start-up skills) is highly specific, and the group has been designed to minimize any other interference.
Reprint No. 98306