s t r a t e g y

November 2008

Using ‘power curves’ to assess industry dynamics
A new way of looking at industry structures reveals startling patterns of inequality among even the largest companies.

Michele Zanini

Major crises and downturns often produce shakeouts that redefine industry

structures. However, these crises do not fundamentally change an underlying structural trend: the increasing inequality in the size and performance of large companies. Indeed, a financial crisis—for example, the one that erupted in 2008—is likely to accelerate this intriguing long-term tendency. The past decade has seen the rise of many “mega-institutions”—companies of unprecedented scale and scope—that have steadily pulled away from their smaller competitors.1 What has received less attention is the striking degree of inequality in the size and performance of even the mega-institutions themselves. Plotting the distribution of net income among the global top 150 corporations in 2005, for example, doesn’t yield a common bell curve, which would imply a relatively even spread of values around a mean. The result instead is a “power curve,” which, unlike normal distributions, implies that most companies are below average. Such a curve is characterized by a short “head,” comprising a small set of companies with extremely large incomes, and drops off quickly to a long “tail” of companies with a significantly smaller incomes. This pattern, similar to those illustrating the distribution of wealth among ultrarich individuals, is described by a mathematical relationship called a “power law.” 2 The relationship is simple: a variable (for example, net income) is a function of another variable (for example, rank by net income) with an exponent (for example, rank raised to a power). Exhibit 1 shows the top 30 US banks and savings institutions in June 1994, 2007, and 2008, measured by their domestic deposits (the 2008 shares of different institutions were adjusted to reflect the surge of banking M&A in the autumn of 2008). The exhibit shows that inequality has been increasing from 1994 (when the number-ten bank was roughly 30 percent of the size of the largest one) to 2008 (when it was only 10 percent as large as the first-ranked institution). It also shows how in 2008, the financial crisis accelerated the growth of the top five compared with the other banks in the top ten as the largest financial institutions took advantage of their relatively healthy balance sheets and absorbed banks in the next tier. Regulation could put a damper on this crisis-driven acceleration of inequality, but power curve dynamics suggest that it will not reverse the trend. Indeed, we found long-term patterns of increasing inequality in size and performance in a variety of industries and markets when we used metrics such as market value, revenues, income, and assets to plot the size of companies by rank.

1

EX HI B IT 1 Increasing inequality in banking

Our analysis suggests that an industry’s degree of openness and competitive intensity is an important determinant of its power curve dynamics. You would expect a bigger number of competitors and consumer choices to flatten the curve, but in fact the larger the system, the larger the gap between the number-one and the median spot. As Exhibit 1 shows, after the liberalization of US interstate banking, in 1994, deposits grew significantly faster in the top-ranking banks than in the lower-ranking ones, creating a steeper power curve. Greater openness may create a more level playing field at first, but progressively greater differentiation and consolidation tend to occur over time, as they did when the United States liberalized its telecom market. Power curves are also promoted by intangible assets—talent, networks, brands, and intellectual property—because they can drive increasing returns to scale, generate economies of scope, and help differentiate value propositions. Exhibit 2 shows a significant degree of inequality, across the board, in the size and performance of companies in a number of sectors we researched. But the more labor- or capital-intensive sectors, such as chemicals and machinery, have flatter curves than intangible-rich ones, such as software and biotech.

2

EX HI B IT 2 Sector variations

The fact that industry structures and outcomes appear to be distributed around “natural” values opens up an intriguing new field of research into the strategic implications. Notably, the extreme outcomes that characterize power curves suggest that strategic thrusts rather than incremental strategies are required to improve a company’s position significantly. Consider the retail mutual-fund industry, for example. The major players sitting atop this power curve (Exhibit 3) have opportunities to extend their lead over smaller players by exploiting network effects, such as cross-selling individual retirement accounts (IRAs), to a large installed base of 401(k) plan holders as they roll over their assets. The financial crisis of 2008 may well boost this opportunity further as weakened financial institutions consider placing their asset-management units on the block to raise capital.

3

EX HI B IT 3 A steep slope

When executives set strategy, power curves can be a useful diagnostic tool for understanding an industry’s structural dynamics. In particular, there may well be commonalities across sectors in the way these curves evolve, and that might make it possible to gain better insights, based on the experience of other industries, into an industry’s evolution. As the importance of intangible assets increases across sectors, for example, will power curves in media and insurance resemble the currently much steeper ones found in today’s intangible-rich sectors such as software and biotech? Power curves could also benchmark an industry’s performance. Curves for specific industries evolve over many years, so the appearance of large deviations from a more recent “norm” can indicate exceptional performance, on one hand, or instability in the market, on the other. Unlike the laws of physics, power curves aren’t immutable. But their ubiquity and consistency suggest that companies are generally competing not only against one another but also against an industry structure that becomes progressively more unequal. For most companies, this possibility makes power curves an important piece of the strategic context. Senior executives must understand them and respect their implications.

4

About the Author Michele Zanini is an associate principal in McKinsey’s Boston office.
Notes

1

See Lowell L. Bryan and Michele Zanini, “Strategy in an era of global giants,” mckinseyquarterly.com, November 2005.

2 The power laws phenomenon has been explored in the recent books The Black Swan (Nassim Nicholas Taleb, Random House, 2007) and The Long Tail (Chris Anderson, Hyperion, 2006).

Related Articles on mckinseyquarterly.com “The winner-takes-all economy” “Enduring ideas: The SCP Framework” “The do-or-die struggle for growth”

Copyright © 2008 McKinsey & Company. All rights reserved.

5