Competition Demystified Chapter 3
ECONOMIES OF SCALE AND CUSTOMER CAPTIVITY
The competitive advantages we have described so far are uncomplicated. An incumbent firm may defeat entrants either because it has sustainably lower costs or, thanks to customer captivity, it enjoys higher demand than the entrants. Together, these two appear to cover fully the revenue and cost elements that determine profitability. But there is an additional potential source of competitive advantage. In fact, the truly durable competitive advantages arise from the interaction of supply-and-demand advantages, from the linkage of economies of scale with customer captivity. Once the firm understands how these operate together—sometimes in ways that are surprisingly contrary to commonly held beliefs about the attractiveness of growing markets—it can design effective strategies to reinforce them. The competitive advantage of economies of scale depend not on the absolute size of the dominant firm but on the size difference between it and its rivals, that is, on market share. If average costs per unit decline as a firm produces more, then smaller competitors will not be able to match the costs of the large firm even though they have equal access to technology and resources so long as they cannot reach the same scale of operation. The larger firm can be highly profitable at a price level that leaves its smaller competitors, with their higher average costs, losing money. The cost structure that underlies these economies of scale usually combines a significant level of fixed cost and a constant level of incremental variable costs. An apparel company, for example, needs the same amount of fabric and labor to make each unit and very little in the way of complicated machinery, so its level of variable to fixed costs is high. A software publisher, by contrast, has almost all fixed costs, which are the expenses of writing and checking the software code. Once the program has been finished, the costs of producing an additional unit are miniscule. So its total expenses increase very slowly, no matter the number of customers.
As the scale of the enterprise grows, the fixed cost is spread over more units, the variable cost per unit stays the same, and the average cost per unit declines. But something in addition to this cost structure is necessary for economies of scale to serve as a competitive advantage. If an entrant has equal access to customers as the incumbents have, it will be able to reach the incumbents’ scale. A market in which all firms have equal access to customers and common cost structures, and in which entrants and incumbents offer similar products on similar terms, should divide more or less evenly among competitors. This holds true for differentiated markets, like kitchen appliances, as well as commodity markets. All competitors who operate effectively should achieve comparable scale and therefore comparable average cost. For economies of scale to serve as a competitive advantage, then, they need to be coupled with some degree of incumbent customer captivity. If an efficient incumbent matches his competitors on price and other marketing features, then, thanks to the customer captivity, it will retain its dominant share of the market. Though entrants may be efficient, they will not match the incumbent’s scale of operations, and their average costs will be permanently higher. The incumbent, therefore, can lower prices to a level where it alone is profitable and increase its share of the market, or eliminate all profit from competitors who match its prices. With some degree of customer captivity, the entrants never catch up and stay permanently on the wrong side of the economies of scale differential. So the combination of even modest customer captivity with economies of scale becomes a powerful competitive advantage. The dynamics of situations like this are worth a closer look. It seems reasonable to think that a persistent entrant will sooner or later reach an incumbent’s scale of operation if it has access to the same basic technologies and resources. If the incumbent is not vigilant in defending its market position, the entrant may indeed catch up. The Japanese entry into the U.S. car market, the success of Fuji Film in taking on Kodak, and the initial significant market share captured by Bic disposable razors from Gillette in the 1980s are testimony to the vulnerability of poorly safeguarded economies of scale advantages. Still, if an incumbent diligently defends its market share, the odds are clearly in its favor. This is why it is important that incumbents clearly understand the nature of their competitive advantages and make sure that their strategies adequately defend them. Think of Microsoft in the operating systems market, Boeing versus McDonnell-Douglas in the commercial airframe business, or Pitney-Bowes in postage equipment. A simple example should help explain why small markets are more hospitable than large ones for attaining competitive advantages. Consider the case of an isolated town in Nebraska with a population of fifty thousand or less. A town of this size can support only one large discount store. A determined retailer who develops such a store should expect to enjoy an unchallenged monopoly. If a second store were to enter the town, neither would have enough customer traffic to be profitable. Other things being equal, the second entrant could not expect to drive out the first, so its best choice would be to stay away, leaving the monopoly intact.
At the other extreme from our Nebraska town is downtown New York City. This large market can support many essentially similar stores. The ability of even a powerful, well-financed incumbent to prevent entry by a newcomer will be limited. It cannot, in other words, establish effective barriers to entry based on economies of scale relative to its competitors. Markets of intermediate size and density, as we would expect, fall between small and large cities regarding the ability to establish and maintain barriers to entry. This general principle applies to product as well as to geographic space; the special-purpose computer in a niche market has an easier time in creating and profiting from economies of scale than the general-purpose PC competing in a much larger market. Long before it became the global powerhouse in retailing, Wal-Mart enjoyed both high levels of profitability and a dominant market share in the south-central United States due to regional economies of scale in distribution, advertising, and store supervision. It defended its territory with an aggressive policy of “everyday low prices.” Southwest Airlines, with a regional franchise in Texas and the surrounding states, was similarly profitable, as have been a lot of other strong local companies in service industries like retailing, telecommunications, housing development, banking, and health care.
DEFENDING ECONOMIES OF SCALE
The best strategy for an incumbent with economies of scale is to match the moves of an aggressive competitor, price cut for price cut, new product for new product, niche by niche. Then, customer captivity or even just customer inertia will secure the incumbent’s greater market share. The entrant’s average costs will be uniformly higher than the incumbent’s at every stage of the struggle. While the incumbent’s profits will be impaired, the entrant’s will be even lower, often so much lower as to disappear altogether. The incumbent’s competitive advantage survives, even under direct assault. The combination of economies of scale coupled with better access in the future to existing customers also produces an advantage in the contest for new customers and for new technologies. Consider the competition between Intel and Advanced Micro Devices (AMD)—or any other potential entrant, like IBM or Motorola—to provide the next-generation microprocessor for Windows-compatible personal computers. Computer manufacturers are accustomed to dealing with Intel and are comfortable with the level of quality, supply stability, and service support they have received from it. AMD may have performed nearly as well in all these areas, but with a much smaller market share and less interaction, AMD does not have the same intimate association with personal computer manufacturers. If AMD and Intel produce next-generation CPUs that are similarly advanced, at equal prices, and at roughly the same time, Intel will inevitably capture a dominant market share. All Intel need do is match AMD’s offering to retain the roughly 90 percent share it currently commands. In planning its next-generation chip, Intel can afford to invest much more than AMD, knowing that its profits will be much greater, even if its CPU is no better. A rough rule of thumb should lead Intel and AMD to invest in proportion to their current market shares. If each company invests 10 percent of current sales in R&D, Intel will outspend AMD $2.6 billion to $300 million. That enormous edge makes Intel the odds-on favorite in the race for next-generation technology. In fact, the situation is even more unequal for AMD. Should it manage to produce a better new chip, computer manufacturers would almost certainly allow Intel a significant grace period to catch up, rather than switch immediately to AMD. The history of competition between the two has seen instances both of Intel’s larger investments usually paying off in superior technology and of its customer captivity allowing it time to catch up when AMD has taken a lead. Thus, economies of scale have enabled Intel to sustain its technological advantage over many generations of technology. Economies of scale in distribution and advertising also perpetuate and amplify customer captivity across generations of consumers. Even if smaller rivals can spend the same proportion of revenue on product development, sales force, and advertising as, for example, Kellogg’s, McDonald’s, and Coca-Cola, they can’t come close to matching the giants on actual dollars deployed to attract new customers. Because of the edge it gives incumbents in both winning new generations of customers and developing new generations of technology, the combination of economies of scale and customer captivity produces the most sustainable competitive advantages.
Three features of economies of scale have major implications for the strategic decisions that incumbents must make. First, in order to persist, competitive advantages based on economies of scale must be defended. Any market share lost to rivals narrows the leader’s edge in average cost. By contrast, competitive advantages based on customer captivity or cost advantages are not affected by market share losses. Where
economies of scale are important, the leader must always be on guard. If a rival introduces attractive new product features, the leader must adopt them quickly. If the rival initiates a major advertising campaign or new distribution systems, the leader has to neutralize them one way or another. Unexploited niche markets are an open invitation to entrants looking to reach a minimally viable scale of operations. The incumbent cannot concede these niches. When the Internet became a major focus of personal computing, Microsoft had to introduce its own browser to counter Netscape and offer network alternatives to niche players like AOL. When Pepsi-Cola targeted supermarkets in the 1950s as an alternative distribution channel, Coca-Cola was too slow to respond, and Pepsi picked up market share. The American motorcycle industry did not challenge Japanese companies like Honda when they began to sell inexpensive cycles in the 1960s. That was the beginning of the end for almost all the American firms. Harley-Davidson survived, though barely and with government help, in part because the Japanese allowed it to control the heavyweight bike niche. Economies of scale need to be defended with eternal vigilance 警戒心.
Second, the company has to understand that pure size is not the same
thing as economies of scale, which arise when the dominant firm in a market can
spread the fixed costs of being in that market across a greater number of units
than its rivals. It is the share of the relevant market, rather than size per
se, that creates economies of scale. The relevant market is the area—geographic or otherwise—in which the fixed costs stay fixed. In the case of
a retail company, distribution infrastructure, advertising expenditures, and
store supervision expenses are largely fixed for each metropolitan area or
other regional cluster. If sales are added outside the territory, fixed costs
rise and economies of scale diminish. When it was still in the cellular
business, AT&T’s cellular
operations in the Northeast and Atlantic states had larger fixed costs per
dollar of revenue in that region than Verizon’s, which controlled a far greater share of the
territory. The fact that AT&T cellular may have been larger nationally than
Verizon cellular is irrelevant. The same conditions apply when the relevant
geography is a product line rather than a physical region. Research and
development costs, including the start-up costs of new production lines and
product management overhead, are fixed costs associated with specific product
lines. Though IBM’s total sales
dwarf those of Intel, its research and development expenses are spread over a
far greater range of products. In CPU development and production, which has its
own particular technologies, Intel enjoys the benefits of economies of scale.
Network economies of scale are similar. Customers
gain by being part of densely populated networks, but the benefits and the
economies of scale extend only as far as the reach of the networks. Aetna’s HMO has many more subscribers nationally than
Oxford Health Plans. But because medical services are provided locally, what
matters is share in a local market. In the New York metropolitan region, Oxford
has more patients and more doctors enrolled than Aetna. Its 60 percent share of
doctors makes it more appealing to new patients than Aetna’s 20 percent share. The fact that Aetna also has 20
percent in Chicago, Los Angeles, Dallas, or even Philadelphia is irrelevant.
The appropriate measure of economies of scale is comparative fixed costs within
the relevant network. There are only a few industries in which economies of
scale coincide with global size. The connected markets for operating systems
and CPUs is one example; Microsoft and Intel are the beneficiaries of global
geographic economies of scale. The commercial airframe industry, now shared
between Boeing and Airbus, is another. However, despite some other interests, each
of these four companies concentrates on a single product line and hence on
local product space economies of scale. General Electric, the most successful
conglomerate, has always focused on its relative share within the particular
markets in which it competes, not on its overall size.
Third, growth of a market is generally the enemy of
competitive advantages based on economies of scale, not the friend. The
strength of this advantage is directly related to the importance of fixed
costs. As a market grows, fixed costs, by definition, remain constant. Variable
costs, on the other hand, increase at least as fast as the market itself. The
inevitable result is that fixed costs decline as a proportion of total cost.
This reduces the advantages provided by greater incumbent scale. Consider two
companies, an incumbent and an entrant, competing in a market in which fixed
costs are $100,000 per year. If the entrant has sales of $500,000 and the
incumbent $2,500,000, then fixed costs consume 20 percent of the entrant’s revenue versus 4 percent of the incumbent’s, a gap of 16 percent. Now the market doubles in
size, and each company doubles as well. The gap in fixed cost as a percentage
of sales declines to 8 percent. At a level ten times the original, the gap
drops to 1.6 percent. See table 3.1. Moreover, growth in the market lowers the
hurdle an entrant must clear in order to become viably competitive. Let us
assume that the entrant can compete with the incumbent if the economies of
scale advantage is no more than 2 percent against it. With fixed costs at
$100,000 per year, the gap drops to that level if the entrant has sales of $5
million. So if the size of the market were $25 million, the entrant would need
to capture a 20 percent share; in a market of $100 million, it would only need
a 5 percent share, clearly a much lower hurdle. Even if the incumbent were the
only other firm in the industry and thus had sales of $95 million, the entrant
would still face less than a 2 percent competitive gap. There are some highly
visible instances of how economies of scale advantages have dwindled as markets
have become international and thus massive. The global market for automobiles
is so large that many competitors have reached a size, even with a small
percentage of the total, at which they are no longer burdened by an economies
of scale disadvantage. For very large potential markets like Internet services
and online sales, the relative importance of fixed costs are unlikely to be
significant. If new entrants can capture a share sufficient to support the
required infrastructure, then established companies like Amazon will find it
difficult to keep them out.
