Business Journal 2

53

Pankaj Ghemawat Sustainable advantage

"If a man., .make a better mouse-trap than his neighbour, tho' he build his house in the woods, the world will make a beaten path to his door." Attributed to one of Emerson's lectures in the nine- teenth century, these words seem to have anticipated the exhortations of the twentieth: manage for unique- ness, develop a distinctive competence, create compet- itive advantage.

"For outstanding performance, a company has to beat the competition.

The trouble is the competition has heard the same message."

But that's not all Emerson had to say about investment in better mousetraps; he also re- marked, "Invention breeds invention." What will re- strain rivals from imitating or even improving on an invention? That question preoccupies the real mouse- trap industry as it staggers from imported copies of its innovative glueboards and repeating traps.

That question is also central to compet- itive strategy. Strategists insist that for outstanding performance, a company has to beat out the competi- tion. The trouble is that the competition has heard the same message. Deadlocks ensue. Look at the cross- industry findings about three competitive hot spots:

1 Product innovation. Competitors secure detailed information on 70% of all new products with- in a year of their development. Patenting usually fails

Mr. Ghemawat is an assistant professor of business administration at the Harvard Business School, where he teaches courses on industry and competitive analysis. His research and consulting focus on competitive dynamics and strategic investments. This is his second ar- ticle for HBR.

to deter imitation. On average, imitation costs a third less than innovation and is a third quicker.'

2 Production. New processes are even harder to protect than new products. Incremental improvements to old processes are vulnerable t o o - if consultants are to be beheved, 60% to 90% of all "learning" ultimately diffuses to competitors. Produc- tion often blurs competitive advantage: recem studies show that unionized workers pocket two-thirds of the potential profits in U.S. manufacturing.-

3 Marketing. Nonprice instruments are usually ascribed more potency than price changes, part- ly because they are harder to match. Rivals often react to a particular move, however, by adjusting their entire marketing mix. Such reactions tend to be intense; lim- ited data on advertising suggest that the moves and countermoves frequently cancel out.'

In principle, threats like these have al- ways been part of doing business. In practice, they have multiplied with the intensification of domestic and in- temational competition. How should a business cope with such competitive pressure? For guidance, we can tum from cross-industry findings to cases.

Keeping the edge

This study of sustainable success grew out of a sample of 100 businesses that far outperformed their industries in the recent past. Not all of them promise to be as successful in the coming decade. The vulnerable ones have a lesson to impart.

Analog Devices, which focuses on spe- cialized applications for analog semiconductors, has invested countercyclically to cash in on business up- turns. The results: 80% faster growth and 50% higher profitability than the rest of the semiconductor indus- try. But existing competitors seem set to copy Analog's investment policy, and new ones-notably the lapanese - a r e invading its profitable niches.

54 Harvard Business Review September-October 1986

Nike's leadership in athletic shoes was built on cheap Far Eastern labor and massive invest- ments in product development and marketing. Over the last five years, Nike averaged thrice the profitabil- ity and four times the growth of the rest of the U.S. shoe industry. But competitors are busy cloning its strategy. Reebok Intemational, for one, sources 95% of its shoes from South Korea, spends heavily on product styling, and has won endorsements from rock stars as well as athletes. Reebok's sales and profits expanded fivefold in 1985, while Nike's actually declined.

Piedmont Aviation's hubs at Baltimore, Charlotte, and Dayton tie together dozens of small and mid-sized cities. Since the major airlines had neglected these routes, Piedmont grew three times as fast as the rest of the industry and was six times as profitahle. But others are now muscling in: People Express has started to encroach, and American Airlines' planned hub at Raleigh will hurt Piedmont's operations out of Balti- more and Charlotte.

Analog Devices, Nike, and Piedmont are very different from one another, yet they face the same threat: copying by competitors. Their competi- * tive advantages are insecure, or contestable, because each can be duplicated. These examples also show that some success stories do revolve around contestable ad- vantages: all of a company's competitors may be stupid some of the time. But can you count on your competi- tors being stupid all of the time? The historical record suggests otherwise. That is why sustainable advan- tages-advantages anchored in industry economics- command attention.

The literature on strategy is crammed with accounts of why a sustainable competitive advan- tage is A Good Thing To Have. But all those accounts beg two key questions: Which advantages tend to be sustainable, and why?

Sustainable advantages fall into three categories: size in the targeted market, superior access to resources or customers, and restrictions on competi- tors' options. Note that these advantages are nonex- clusive. They can, and often do, interact. The more of them, the better.

Benefits of size

Size advantages exist because markets are finite. If a business can commit to being large, com- petitors may resign themselves to remaining smaller. What holds them back is the fear that if they matched the leader's size, supply might exceed demand by enough to make the market unprofitable for everyone.

Commitment to being large means mak- ing durable, irreversible investments. To exploit com- mitment opportunities, a business must be able to pre-

"My big mistake was sticking to my last."

empt its competitors. Caveat preemptor: first-movers have to be especially wary of environmental changes that can erode the value of their early investments.

Size is an advantage only if, net net, there are compelling economies to being large. Such econo- mies have three possible bases: scale, experience, and scope.

Scale economies usually summon up a vision of a global factory running flat out. But it is im- portant to remember that scale can work on a national, regional, or even local level, and that its effects need not be confined to manufacturing.

Wal-Mart, the discount merchandiser, illustrates the power of local and regional scale econo- mies. Historically, it focused on small Sunbelt towns that its competitors had neglected. Most of these towns could not support two discounters, so once Wal-Mart made a long-lived, largely unrecoverable investment to service such a town, it gained a local monopoly. The company reinforced this advantage by wrapping its stores in concentric rings around regional distribu- tion centers. By the time competitors realized that this policy cut distribution costs in half, Wal-Mart had pre- empted enough store sites to render competing region- al warehouses unviable. Now you know why Sam Walton is one of the richest men in America.

The Wal-Mart story also shows the lim- its to scale economies. K mart and other discounters are beginning to enter some of Wal-Mart's larger loca- tions. The problem, ironically, is market growth: be- cause of the boom in the Sunbelt, some of these tovras

Sustainable advantage 55

can now accommodate two discounters. And even the warehousing advantages look insecure in the regions into which Wal-Mart is expanding; it probably won't be able to blanket these areas with stores before com- petitors move in.

Experience effects are based on size over time, rather than size at a particular point in time. If you think about it, experience is a kind of irreversible, market-specific investment. While it is usually cited in the context of the experience curve—the inverse rela- tion between cumulative production and average cost-its ambit is actually much broader. For example, experience has been shown to increase the operating reliabihty of processing plants, the success rate of prod- uct introductions, and the marketability of high-tech products.

Experience effects-especially experi- ence curves-have come under heavy fire recently be- cause they were oversold in the 1970s. Yet some com- panies have parlayed them into competitive success. Take Lincoln Electric's experience in the electric weld- ing industry. Ever since John Lincoln developed the portable arc welder in 1895, Lincoln Electric has out- raced its competition down the experience curve. In its ninth decade it still commands a 7% to 15% cost ad- vantage over its four largest rivals.

Lincoln is an object lesson about when and how to exploit experience effects. As the product pioneer it had a first-mover advantage; and that lead has proved durable because of the incrementalism of technological change. Lincoln has also kept its experi- ence proprietary by integrating backward, customizing its production machinery, and holding annual worker turnover under 3%. Finally, it has continued to invest in experience by sharing cost reductions with custom- ers. Competitors complain publicly that they have trouble matching Lincoln's prices, let alone undercut- ting them.

Scope economies are derived from activ- ities in interrelated markets. If they are strong, a sus- tainable advantage in one market can be used to build sustainability in another. The term scope economies isn't just a newfangled name for synergy; it actually defines the conditions under which synergy works. To achieve economies of scope, a company must be able to share resources across markets, while making sure that the cost of those resources remains largely fixed. Only then can economies be effected by spreading assets over a greater number of markets.

Cincinnati Milacron, the largest U.S. machine tool manufacturer, shows how companies can capitalize on scope economies. For the last two decades it has led the U.S. machine tool industry in both R&D and the size of its sales and service networks-activi- ties that account for a third of the value added by the

industry. In the 1980s it has pushed hard into robotics, with good reason. Its cumulative R&D experience gives the company such a big lead in the machine tool segment that no domestic challenger can rationally commit to matching its R&D, sales-force, or service expenditures. Furthermore, the technologies and cus- tomers for its computerized, numerically controlled machine tools overlap with those for robotics. And the company's R&D, sales, and service activities are all very volume sensitive, which slashes the incremental costs of moving into robotics. These factors make it a formidable contender in the industry.

A company pursuing a sustainable scope advantage cannot afford to run its businesses as isolated units. Activities have to be coordinated and al- lowances must be made for contributions from one business to the success of another. This makes scope economies especially hard to implement.

Access advantages

Preferred access to resources or custom- ers can award a business a sustainable advantage that is independent of size. The advantage persists because competitors are held back by an investment asymme- try; they would suffer a penalty if they tried to imitate the leader.

Access will lead to a sustainable advan- tage if two conditions are met: it must be secured un- der better terms than competitors will be able to get later, and the advantage has to be enforceable over the long run. Enforceability can come from ownership, binding contracts, or self-enforcing mechanisms such as switching costs. Without enforceability, the terms of access shift in line with overall market conditions, wiping out any competitive differences.

Enforceability can be a two-edged sword, however. The risk of pursuing sustainable access ad- vantages is that they may saddle a business with worse terms than those available to its rivals.

Know-how. Superior access to informa- tion may reflect the benefits of scale or experience. Boeing, for instance, has acquired superior know-how about commercial jet aircraft t h r o u ^ billions of dol- lars of cumulative investments in R&D. More often, though, sustainability hinges on hidden know-how- what your rivals don't see. For example, IBM's size and the complexity of its operating environment make it hard for competitors to figure out exactly what makes it tick. If the cost of surmounting that kind of informa- tional barrier exceeds the payoffs, rivals may not even attempt imitation.

Consider Du Pont's preemption of all the capacity expansion in the U.S. titanium dioxide in-

56 Harvard Business Review September-October 1986

Research method I have drawn many of my examples from research reports prepared by MBA students for the Industry and Competitive Analysis {ICA) course at the Har- vard Business School. Constraints of space pre- clude individual citations; I should, however, empha- size the collective contribution of ICA students to this article.

In fall 1984 and 1985, we asked our students to ana- lyze companies identified as outstanding perform- ers in their industries. The purpose was threefold: to understand the sources of competitive advan- tage on which these companies relied, to determine why the advantages had proved sustainable, and to assess their future security.

Our students submitted reports on well over 150 companies. Some, obviously, dealt with companies that had been canonized publicly; many others, however, were unfamiliar to the ICA teaching group.

Even more important, the quality of the reports sur- passed our expectations. Taken together, they sug- gest answers to some of the most fundamental questions about strategy formulation. To give these reports the broader circulation they deserve, Douglas Anderson, Michael Porter, and I are editing 20 of them for publication. The compendium will be available in 1987.

dustry in the 1970s. Thanks to a production process based on low-cost feedstock, Du Pont enjoyed a 20% cost advantage over competitors' processes. Mastering the cheaper feedstock technology was a black a r t - it could be aceomphshed only by investing $50 million to $ 100 million and several years of testing time in an efficiently scaled plant. The cost and risk of this alter- native kept Du Pont's competitors from trying to imi- tate its demonstrably superior technology.

An obvious but important point; know- how must be kept secret if it is to yield an advantage. Many high-tech and service companies have been devastated by the defection of key personnel in whom their know-how is vested. The Boston Consulting Group, for instance, has suffered more than a dozen spin-offs, eroding its competitive advantage in manage- ment consulting and its client base. Other sources of leaks include suppliers, customers, reverse engineer- ing, and even patent doeuments.

Inputs. Tying up inputs will lead to a sustainable advantage only if the commodity's supply is bounded and the company has the right to use it on favorable terms. Boundedness here is interpreted broad- ly: it may imply either a strictly limited supply of the input or a supply that is elastic but of varying quality.

In both cases, the supply of the preferred input is hmit- ed; as a result, tying it up can be very profitable.

This description covers a wide range of phenomena. Courtaulds' 10% to 15% cost advantage over its competitors in the viscose industry can be traced to its backward integration into dissolving pulp, which accounts for a third of the finished produet's cost. Courtaulds gets its pulp from a well-located sub- sidiary for half what its competitors pay. fames River Corporation has averaged a 24% ROE by buying obso- lete commodity paper machines at fire-sale prices and converting them to specialty products, a stratagem that has held its assets-to-sales ratio to two-thirds the industry average. In diamonds, the Central Selling Or- ganization (controlled by De Beers] has built up its marketing muscle by tying up contracts to market 80% of the western hemisphere's supply.

Companies can also secure preferred ac- cess through their reputations or established relation- ships. In the record industry, for example, CBS has at- tracted promising artists because of its reputation for being able to take them to the top - at least partially a self-fulfilling prophecy.

Access advantages are vulnerable to shifts in input availability or prices. Courtaulds' cost advantage in dissolving pulp will wither as infrastruc- tural development opens up more tropical and subtrop- ical forests. And James River's competitors, particu- larly Hammermill, have begun to bid up the prices of the second-hand paper machines that it traditionally bought for a song. This constrains James River's growth, even though its cheap asset base will prop up profitability for years to come.

Markets. In many ways, preferred access to markets is the mirror image of preferred access to in- puts. But access to markets relies less on vertical inte- gration or contracts and more on self-enforcing mecha- nisms such as reputation, relationships, switching costs, and product complementarities.

That is not to say that vertical integra- tion and contracts are entirely missing from the pic- ture. Tele-Communications' strategy in cable televi- sion systems shows otherwise. While competitors outbid eaeh other in their scramble to secure big fran- chises, Tele-Communications concentrated on ac- quiring small, contiguous systems in areas that were poorly served because they were hard to get to or far from large population centers. Tele-Communications' current network faces no serious threats from substi- tutes or competitors, limits its exposure to the whims of any one regulatory authority, and allows the com- pany to spread the costs of its microwave common- carrier network over several communities.

Still, self-enforcing meehanisms for market access crop up far more frequently. Let us look at just two examples. Tandem, which pioneered expan-

Sustainable advantage 57

sible, fault-tolerant computers for processing transac- tions, has gained preferred access to demand for up- grades and replacements because changeovers from one system to another are very costly. And Borden's brand of processed lemon juice, ReaLemon, attained a 50% premium over identical competing brands be- cause as the pioneer, it benefited from consumer risk- aversion: lemon juice doesn't cost very much and you don't buy it very often, so why take a chance with an untried brand?

You have probably already figured out that market access advantages are very sensitive to customer preferences. Even slight, apparently iimocu- ous shifts in preferences can weaken an entrenched brand, dispel accumulated switching costs, or undercut long-standing relationships.

Exercising options

Sometimes the sustainability of an ad- vantage cannot be pinned on either size or access. In- stead, competitors' options may differ fundamentally from yours, hamstringing their ability to imitate your company's strategy. Rivals may be frozen into their current positions for several reasons:

Public policy. Govemmcnt intervention always affects the workings of markets; that is its avowed purpose. Sometimes its actions percolate so far as to affect competitive positions within an industry. The examples are familiar: patents (try to) protect in- novators from imitators, antitrust laws prevent large businesses from being as aggressive as smaller compet- itors, some companies get handouts while others do not. The lesson, strategically, is that a company that is on the right side of public policy can exploit its posi- tion to build sustainability against companies that are not.

Heileman Brewing exemplifies both the leverage from this source of sustainability and its lim- its. The shakeout in the U.S. brewing industry during the 1970s endangered many small, regional brewers. Antitrust laws prevented the national brewers- Anheuser-Busch, Miller, and Schlitz-from acquiring them. Heileman, then one of the larger regional brew- ers, faced no such constraints. It grew throughout the decade by buying out numerous smaller brewers lock, stock, and barrel. These cheap assets fed right through to the bottom line: with an average ROE of 29% over the last five years, Heileman is still ahead of its com- petitors. But the bloom on this particular strategy is fading. By 1982, Heileman's market share had tripled, and the Justice Department blocked its proposed ac- quisitions of Schlitz and Pabst. Other takeovers by Heileman are improbable.

Remember that what the govemment gives, the govemment can take away. Treat an advan- tage based on public policy as sustainable only if you are sure you will continue to be on the right side. If not, try a different route.

Defense. A business can also sustain an advantage if its competitors are restricted by past in- vestments. If imitation threatens the cash flow from those investments, disadvantaged competitors may ra- tionally stay put and defend them, thereby giving the innovator an opportunity to take the lead.

Examples of defensiveness are legion. Bic used its 19-cent Crystal to wrest leadership from Gillette in the U.S. pen market in the 1950s, when Bic's aggression went unmatched because Gillette did not want to decimate the sales of its more expensive Paper Mate line. In plain-paper copying, a number of compet- itors successfully attacked Xerox in the 1970s. Recog- nizing that its multibillion-dollar rental base was be- coming obsolete, Xerox milked it by dragging its feet on price cuts and product innovation - even though these lags caused its share of new placements to fall from nearly 100% in 1972 to 14% by 1976.

Time often erodes defensiveness, how- ever, as it depreciates the value of past investments. In 1970, Gillette introduced a low-priced line of Write Bros, pens to arrest Bic's advance. And since 1977, Xerox has restored its share of new copier placements to the 40% to 50% range by matching competitors' prices and product features.

"All of your competitors may be stupid some of the time-

but can you count on them being stupid all of the timel"

Response lags. The final restriction on rivals' options comes from response lags. One business can be every bit as efficient as another in terms of po- tential size or access without being equally prepared to make a specific move. In that event, the nimbler of the two can count on a lag in its competitor's response, or a period of sustainability.

The longer the response lag the better, as existing advantages stretch out and opportunities to create new ones multiply. Lag times vary enormously, of course, but I can make some broad generalizations. Responses to most pricing moves come in weeks if not days, while responses to nonprice competition and to R&D usually take a few years. And it may take a de-

58 Harvard Business Review September-October 1986

cade or more to match a competitor's scope economies or superior organization.''

Kodak vividly illustrates how you can sustain a lead by exploiting competitors' response lags. In the late 1950s, Du Pont and Bell & Howell formed a joint venture to challenge Kodak's dominance of the color film market. They found product development ex- asperating, however, because each time they improved their film, Kodak seemed, as if by magic, to make its film even better. When the Du Pont-Bell & Howell film was finally ready, Kodak administered the coup de grace by introducing the vastly superior Kodachrome n slide film. The competing entry never reached the market.

Guidelines for strategy

I have outlined a set of factors that af- fect the sustainability of competitive advantages. How should these factors-and the broad notion of sustain- ability-be integrated into strategy formulation? Here are several points to remember:

1 Managers cannot afford to ignore con- testable advantages. For one thing, even moves that of- fer ephemeral advantages may be worth making, if only to avoid a competitive disadvantage. For another, some contestable advantages may survive uncon- tested: disadvantaged competitors may be tied up try- ing to meet their profit targets, constrained by their corporate strategies, or just ineptly managed.

2 The distinction between contestable and sustainable advantages is a matter of degree. Sus- tainability is greatest when based on several kinds of advantages rather than one, when the advantage is large, and when few environmental threats to it exist.

3 Not all industries offer equal opportuni- ties to sustain an advantage. First-mover advantages tend to be most potent in industries characterized by durable, irreversible, market-specific assets, either tan- gible or intangible. Industries that evolve gradually of- fer more room to sustain advantages than those that are regularly rocked by drastic changes in technology or demand. And sustainability is more accessible in industries with more than one dominant strategy because competitors may not have the same options you do.

4 To create a sustainable advantage, you must either be blessed with competitors that have a re- stricted menu of options or be able to preempt them. Propitious times to preempt occur when an industry is undergoing wrenching changes in technology, demand patterns, or input availability. Scan the environment

actively. If you notice any changes, see whether they play to your particular strengths.

Ultimately, the search for sustainability involves a series of decisions about the degree to which you are willing to commit your business to a particular way of doing things. You have to pick the relative em- phasis you are going to place on two things: commit- ment to competing a particular way and retaining the flexibihty to compete effectively in other ways.

References

1 Edwin Mansfield, "How Rapidly Does New Industiial Technology Leak Out?" loumal of Industrial Economics, December 1985, p. 217; Richard C, Levin et al,,

"Survey Research on R8iD Appropriabiiiiy and Technological Opportunity, Part I," Yale University Working Paper (New Haven: July 19841; and Edwin Mansfield, Maik Schwartz, and Samuel Wagner.

"Imitation Costs and Patents: An Empirical Study," Economic/ournai, December 1981, p. 907.

2 Michael A. Salinger, "Tijbin's q, Unionization, and the Concentration-Profits Relationship," Rand loumal of Economics, Summer 1984, p. 159i and Thomas Karier,

"Unions and Monopoly Profits," Review of Economics and Statistics, February 1985, p. 34.

3 M.M. Metwaliy, "Advertising and Competitive Behavior of Selected Australian Firms," Review of Economics and Statistics, November 1975, p. 417>

JcaD-|acques Lambin, Advertising, Competition, and Market Conduct in Ohgopoly Over Time (Amsterdam: Notth-Holland, 1976)̂ and (effrey M. Netter,

"Excessive Advertising: An Empirical Analysis," louinal of Industrial Economics, |unel982.p.361.

4 See, for example, David I.Teece,

"Tbe Diffusion of an Administrative Innovation," Management Science, May 1980, p. 464.

Editor's note. Additional references are available from the author.

Harvard Business Review Notice of Use Restrictions, May 2009

Harvard Business Review and Harvard Business Publishing Newsletter content on EBSCOhost is licensed for

the private individual use of authorized EBSCOhost users. It is not intended for use as assigned course material

in academic institutions nor as corporate learning or training materials in businesses. Academic licensees may

not use this content in electronic reserves, electronic course packs, persistent linking from syllabi or by any

other means of incorporating the content into course resources. Business licensees may not host this content on

learning management systems or use persistent linking or other means to incorporate the content into learning

management systems. Harvard Business Publishing will be pleased to grant permission to make this content

available through such means. For rates and permission, contact [email protected].