C H A P T E R 2

We cannot overcome the problem of unpredictability.

[It] represents the fundamental and universal
situation of life on earth.
— P E R B A K , H O W N AT U R E W O R K S

UNPREDICTABILITY
The Only Sure Thing

I N THE EARLY 1980S, IN AN EFFORT TO PREPARE FOR DEREGULA-

tion and the opening of Australia to foreign banks, executives of
Australia’s largest and most profitable bank consulted extensively
with colleagues in major financial institutions around the world to
determine what financial products and services the industry
should offer over the next five years. Expectation of deregulation
and foreign competition had already led to the creation of Westpac
in a 1982 merger in which the Bank of New South Wales, Aus-
tralia’s oldest bank, acquired the Commercial Bank of Australia.
Combined, the firms enjoyed a 20 to 25 percent market share of
most retail and corporate banking segments in Australia. Westpac
executives now sought greater understanding of the world in
which they would be operating.
They found, much to their surprise, that none of the execu-
tives to whom they spoke believed reliable predictions of new
products and services were possible even one year into the future,
23
24 C H A P T E R 2

much less five. The variety and pace of new product introduction
in the financial services industry made a mockery of forecast as-
sumptions. The “continuous discontinuity” of marketplace change
made it impossible to place strategic bets on new product develop-
ment with any confidence.
Deregulation alone did not explain this state of flux in a once
stable industry. More fundamental was a radical change in the
products themselves. Traditionally, the wealth handled by finan-
cial institutions had been embodied in precious metals. These
could be physically shifted from vault to vault and represented by
certificates and banknotes. By 1980, both wealth and the condi-
tions for its exchange could be represented electronically. Finan-
cial products were coming to exist only as symbols in cyberspace.
And because computers can manipulate symbols very rapidly, new
products could go from concept to launch in weeks or even days.
More than anything else, the uncertainty faced by Westpac and
other banks was driven by the possibility of creating almost in-
stant new products on computer workstations.
The first of these products—examples of what became
known as financial derivatives—were mortgage-backed obligations
introduced in the United States by the Federal National Mortgage
Association (Fanny Mae) a few years before Westpac asked its
questions about the future of the industry. Fanny Mae did so to
transfer some of its mortgage loan risk to buyers and sellers in the
securities market. Banks developed their own version of this inno-
vation: the so-called interest rate mismatches that later developed
into interest rate swaps and still later into swaptions. These and
other derivatives have a purely electronic existence, with all the
dynamic potential for rapid change that implies. They can be al-
most as varied as the imaginations of their creators. By the early
1990s, financial derivatives had become a $30 trillion global mar-
ket. Their spectacular success came even though no one ever has
seen or ever will see a financial derivative.
We often hear repeated the adage that we live in the Informa-
tion Age, yet rarely do we stop to ponder what that really means.
But one symptom we all experience is this: Not only are we
swamped by unprecedented amounts of information, but, increas-
ingly, we work with and generate wealth from information rather
 U N P R E D I C T A B I L I T Y 25

than from things. Financial derivatives provide only one of many

examples.

T HE MOTHER OF ALL D ISCONTINUITIES : CODIFIED

I NFORMATION AND KNOWLEDGE

In a speech given in 1989 at an IBM Process Industry Confer-

ence, Wharton professor and former Assistant Secretary of Com-
merce Bruce Merrifield reported that 90 percent of the world’s
codified information had been produced since 1960. (See Figure
2.1.) Merrifield went on to project that the then current amount
would double in the next fifteen years. As a result, he predicted,
more change would occur during the next quarter century than
humanity had seen in the entire span of its history.
Merrifield’s curve started its almost vertical climb in the early
1960s, when solid state computers were first introduced for com-

FIGURE 2.1

M ERRIFIELD ’ S C URVE

Percentage
of
Information
and
Knowledge
that is
Codified

8,000 B.C. 2,000 A.D.

Source: Bruce Merrifield.

26 C H A P T E R 2

mercial use. Many people, seeing such evidence, assume that infor-
mation technology alone brought about the Information Age. The
role of computers such as the IBM 1401 (1959) and the IBM S/360
(1964) did play an undeniable role in the explosion of information,
but technology is at most half the story. Equally astounding, but of-
ten overlooked, has been the improvement in our ability to create
rich symbolic representations of real world objects and actions.
Building and using these representations may depend on the ca-
pacity of computers to store and rapidly manipulate vast quanti-
ties of data, but the representations derive from efforts of human
thought and imagination. The advent of the Information Age re-
quired more than the collection of ever larger quantities of data. It
required the abstraction of real entities and behaviors into sym-
bols, and the use of that codification of reality to change reality.
In the late 1950s, limitations imposed by the eighty-column
IBM card restricted computer representation of the entity cus-
tomer to a very few attributes, such as name, address, amount of
transaction, and so on. Today, an MCI database contains rich ab-
stractions of up to 10,000 items of information each on 140 mil-
lion households. Technology made such a database feasible, but
humans determined the essential information elements of what
customer means to MCI. Abstractions of behavior have developed
to a similar degree. In 1960, cutting a payroll check could be repre-
sented symbolically with sufficient rigor to automate that process.
Today, the design, virtual construction, and testing of something
as complex as a Boeing 777 can occur entirely in cyberspace. The
777 prototype was an electronic abstraction; the first physical
777 built went into commercial service. Symbolic abstractions,
which can be manipulated at electronic speeds, can be created,
modified, and moved much faster than can their physical coun-
terparts. In this lies their principal source of economic advantage.
Behaviors thus codified have yielded over the past four decades a
25 percent annual improvement in productivity.
These changes have put intangibles at our culture’s economic
center. In 1979, Harvard sociologist Daniel Bell wrote that, just as
capital and energy had replaced land and labor as the basic
wealth-creating economic resources, “codified information and
knowledge are replacing capital and energy as the primary trans-
 U N P R E D I C T A B I L I T Y 27

forming economic resources.”1 He supported this assertion by cit-

ing Marc Porat’s research on changes in the distribution of work.
(See Figure 2.2.)
The information and knowledge economy not only moves
faster than the capital and energy economy, it works differently.
The economic properties of information differ from those of the
material resources that fueled the agricultural and industrial ages.
Traditional capital assets have maximum value before they are
used. Until used, however, information has no value at all. Land,
labor, capital, and energy are appropriable, that is, once given
away, they are no longer possessed by the giver. Moreover, their
value generally increases as they become scarcer. These economic
laws do not govern information. Those who give information to
others still have it, and the value of knowledge tends to increase
rather than decrease with sharing and use.2
Bell points out that an economic value theory of information
does not currently exist. By this he means that, as yet, no one can

FIGURE 2.2

D ISTRIBUTION OF W ORK IN THE U.S. E CONOMY

100
Percentage of Work in the United States

Agricultural Sector

Industrial Sector

Service Sector

Information Sector

0
1860 1880 1900 1920 1940 1960 1980 2000

Source: From Michael Dertouzos and Joel Moses, The Computer Age: A Twenty Year View (Cambridge, MA:
The MIT Press, 1979) Figure 9.2.
28 C H A P T E R 2

predict how a given change in information input will affect the out-
put of a productive process.3 Without a theoretical basis for mea-
suring the economic value of knowledge, databases costing billions
of dollars as well as most of the knowledge in employees’ heads
cannot and do not appear on corporate balance sheets. Though
many executives like to say that “people are our most important
assets,” these particular assets are marooned on income state-
ments as expense and cost. Because the value of knowledge cannot
be measured, the term knowledge management remains more a
wish than a reality. Without such valuations, in addition, uncer-
tainty over company worth increases. What is happening when the
stock market values a company at many times the value of its cap-
ital assets? Is the market reflecting the underlying value of the
company’s intellectual assets? Or is it manifesting the kind of
speculative exuberance that drove tulip bulb prices to astronomi-
cal, unsustainable heights in seventeenth century Holland?
Similarly, the theory of decreasing returns to scale, another
so-called law of classical economics, does not always apply in the
Information Age. The theory holds that the finite size of the mar-
ket, the entry of competitors, the limited availability of resources,
and similar factors limit a firm’s profitable growth. Growth of even
the most successful company will eventually be curtailed by nega-
tive feedback, driving it toward economic equilibrium. Industrial
Age experience has largely borne out the theory, but economist
Brian Arthur says that the theory of increasing returns with posi-
tive feedback cycles is more likely to apply to knowledge-intensive
firms.

If one product or nation in a competitive marketplace gets ahead by

“chance,” it tends to stay ahead and even increase its lead. Pre-
dictable, shared markets are no longer guaranteed. . . .
[T]he parts of the economy that are resource-based (agricultural,
bulk-goods production, mining) are still for the most part subject to
diminishing returns. Here conventional economics holds sway. The
parts of the economy that are knowledge-based, on the other hand,
are largely subject to increasing returns.4

Arthur uses the metaphor of a gambling casino to describe

the experience of knowledge-intensive companies competing in an
 U N P R E D I C T A B I L I T Y 29

increasing-returns economy. Imagine you want to enter a new,

high stakes game, digital communications. The ante is a few billion
dollars. You begin to play, even though you don’t know who you
will be playing against. The rules won’t be established until all
players join in, and they will be subject to change during the game.
You won’t know what new technologies will be developed or how
the government might decide to regulate play. But you cannot stay
out of the game until you have this information because only those
in it from the start have a reasonable shot at winning. You play,
trying to make sense out of what happens and acting accordingly,
but other players’ actions and sensemaking efforts affect and
alter the way the game proceeds. Pure chance usually determines
who gains sufficient initial advantage to ride the positive feed-
back cycle. If you are dealt a losing hand, you have to get out
quickly and cut your losses. Even if you’re a winner, you must quit
while you’re ahead, moving on to the next game, and the next after
that.5
This vivid, unsettling, but very apt metaphor captures the
unprecedented challenge firms face in trying to conduct business
in an information economy.
Consider once again these historic, irreversible develop-
ments of the last few decades.

• The quantity of information and knowledge has increased ex-

plosively.

• To an ever greater extent, information and knowledge consist

of symbolic representations of reality subject to high-speed
manipulation by technology, a process that both reflects and
exacerbates rapid change in the real world.

• Information and knowledge have become primary economic

resources but remain difficult to measure, with economic be-
havior differing vastly from that of traditional capital assets.

These changes underlie the discontinuity recognized by the

banking executives Westpac consulted and by the executives
surveyed at the Advanced Business Institute. They characterize
the Information Age, and they have created a new economic world.
The organizations that thrive in this world will be as different from
30 C H A P T E R 2

those of today as the factory was from the farm in the nineteenth
century.

T HE NEED FOR A N EW O RGANIZATION

For an organization to work as an efficient making-and-sell-

ing machine requires above all a clear understanding of how things
should be done, a specification that should not change frequently
or unexpectedly. Frederick Taylor, the famous pioneer of machine-
like efficiency in the work place, unequivocally assumed that this
kind of certainty was not only necessary, but possible. These two
quotations reveal how Taylor and his followers thought about the
conduct of business:

The high-priced man does just what he is told to do, and no back
talk. When [the foreman] tells you to walk, you walk; when he tells
you to sit down, you sit down.

It is necessary in any activity to have a complete knowledge of what

is to be done and to prepare instructions. The laborer has only to
follow instructions.6

As dated as this now sounds, the approach works extremely

well if leaders have “a complete knowledge of what is to be done.”
If not, it breaks down. Make-and-sell firms clearly lack the agility
to play successfully in Brian Arthur’s increasing-returns casino,
where even a moderate chance of winning requires sensitivity to
the game’s changing complexion and sufficient creativity to make
sense out of unexpected events.
Westpac began to perceive the changes overtaking the bank-
ing industry at much the same time that new technological devel-
opments were redefining the computer industry. Westpac was not
ready for a world of discontinuity and rapid change. When it ac-
quired the Commercial Bank of Australia, the Bank of New South
Wales had been an innovator for thirty years, introducing private
savings banks, automated check clearing, bank investment ser-
vices, credit cards, and ATMs to Australia. In 1985, however, West-
pac was taking eighteen to twenty-four months to develop new fi-
 U N P R E D I C T A B I L I T Y 31

nancial products, a dangerously long time, given the impossibility

of knowing what the industry would be doing in twelve months.
Westpac’s competitors, on the other hand, had development cycle
times of less than a year.
The acquisition, meanwhile, intended to increase the bank’s
competitiveness, had created its own problems. Integrating
the banks proved difficult. Each bank had its own product
lines, and each product line had its own processes and IT appli-
cations. Developing a new product thus required developing a
new IT system. When Merrill Lynch, a non-bank competitor,
arrived with its sweep function that integrated information about
all products utilized by individual clients, Westpac could not re-
spond in a timely way. The bank’s management saw the need “to
get IT off the critical path.” IT contributed significantly to West-
pac’s 15 percent competitive disadvantage in structural costs and
to the growth penalties associated with being chronically late to
market.
The strategic challenges facing Westpac in 1985 boiled down
to these: impending deregulation, new offshore niche competitors,
an unpredictable market for financial services, an uncompetitive
cost structure, and a two-year product cycle time. Despite its size
and past success, Westpac was unprepared to compete in the In-
formation Age that was now transforming banking. CEO Robert
White characterized the trend and magnitude of the transforma-
tion: The industry had been “banking in the 1960s, became finan-
cial service in the 1970s, was becoming financial intermediation
in the 1980s, and would mature into information intermediation
in the 1990s.”7
In other words, Westpac needed to learn how to participate
in a new game; the old one was not being played any more. The
new game would move faster and would be more unpredictable.
Westpac’s executives came to realize that unpredictability was no
longer a problem: It was the problem.
That unpredictability was due to the growing importance of
information. Why does information as the basis of wealth creation
wreak such havoc on market predictability? In the 1980s, Berke-
ley professor Rashi Glazer developed principles that provide im-
portant insights about the answer to this question.
32 C H A P T E R 2

GLAZER ’S L IST

At a time when little empiric evidence was available, Glazer

derived his principles directly from the economic properties of in-
formation. He published them as logical implications of doing busi-
ness in information-intensive environments. Experience now cor-
roborates Glazer’s insights, as information-intensive organizations
increasingly come to resemble his predictions. He addressed both
the nature of the conditions firms face in the Information Age and
the necessary responses to those conditions. His definition of in-
formation intensity captures the connection between information
and change.

[A] firm is information intensive to the degree that its products and
operations are based on the information collected and processed as
part of exchanges along the value-added chain. Whereas traditional
products and operations are relatively static, information-intensive
products and operations change as new data from the environment
become incorporated into them.8

Because the information component of products can change

faster, it will change faster. Because information can be rapidly
disseminated, it will be rapidly acquired by others. As a result,
product life cycles will continue to shrink, and the pace of change
will continue to accelerate.
Glazer has more recently described interactive commerce us-
ing electronic media, the result of a transition from what he calls
dumb (that is, not interactive) environments to smart environ-
ments, in a way that also puts rapid change at center stage. He iden-
tifies four basic types of change for Information Age commerce.9

• Frequent turnovers in the stock of knowledge or information

• Flexible, adaptive roles and boundaries that change as infor-

mation is exchanged

• Smart products that change as users interact with them

• Smart prices that change through negotiations, auctions, and

yield management techniques
 U N P R E D I C T A B I L I T Y 33

Taken together, the following hypotheses about information-

intensive organizations point toward a markedly different way of
doing business than that of traditional make-and-sell firms. Glazer
emphasizes the following key transitions.10

• Product life cycles will shorten

As products become more information intensive and as infor-
mation technology accelerates the rate at which information
changes and spreads, product change becomes easier and
more necessary. Given the short economic shelf life of most
information, product life cycles shorten as their information
component increases.

Where speed, flexibility, and variety are possible, they be-

come necessary. If you don’t offer them, your competitors will,
and your customers will disappear.

• Markets and industries will be defined in terms of customers rather

than products
Interactive transactions will give businesses more current in-
formation about their customers’ wants and needs. Busi-
nesses will use this to discover more opportunities to create
customer value.

• Market power will shift from suppliers toward customers

As customers become the main source of information about
value, shorter product cycles and customer-defined markets
will give consumers more options and lower switching costs.

• More reliance will be placed on flexible, customized marketing and

intangible elements of the value proposition
Customer-specific information will lead to more varied, tar-
geted marketing and the embodiment of that information in
products and services better tailored to customer needs.

• The need to choose between a high volume/low cost strategy and a

niche/differentiation strategy will disappear
Flexible, modular marketing systems, informed by increased
customer knowledge and complemented by modular, infor-
mation-driven manufacturing systems, will make possible
34 C H A P T E R 2

systematic, low-cost customization. Sometimes called “mass

customization,” this development will allow companies to
pursue both high volume/low cost and niche/differentiation
strategies simultaneously.

• Maximizing the number of transactions with the same loyal cus-

tomer by offering a diverse array of products and services will be-
come increasingly important
As market power shifts to customers and modular production
and marketing systems become more important, products
will cease to be the basis of lasting differentiation.

• The most productive business strategies will be cooperative, not

competitive
Because it changes rapidly, customer information has most
value if it is acted upon quickly. By sharing it with firms with
complementary competencies, a business can more rapidly
bring appropriate offerings to market, to the benefit of all.
Hypercompetition for customers thus increases the need for
collaboration among competitors.

Competition in the Information Age will become situational

rather than institutional. Quick and flexible response to ever-
changing customer needs will require businesses to form multiple
simultaneous cooperative relationships.

• Organizations will rely more on decision teams and parallel informa-

tion processing and less on individual decision-making and sequen-
tial information processing
Parallel processing—different brains working simultaneously
on different elements of a problem—can more efficiently and
naturally extract meaning from data than can sequential pro-
cessing. Because organizations make meaning out of huge
quantities of diverse data, parallel processing will become es-
sential.

Glazer’s propositions of the 1980s are rapidly becoming the

realities of the 1990s. CIBC Insurance, a fast-growing and very
profitable Canadian company founded in 1993 is one example.
 U N P R E D I C T A B I L I T Y 35

It sells all of its property and casualty insurance and 80 percent

of its life insurance directly to clients by telephone. CIBC repre-
sentatives completely customize all term-life policies, building
each from modular components to meet individual customers’
needs.11

W HAT W ILL REPLACE M AKE - AND-S ELL ?

Some of the changes described by Glazer were already affect-

ing the banking industry in 1985, when Westpac was trying to
develop its future strategy. Notable were shorter product cy-
cles and increased choices for consumers. Banking was on its
way to becoming an electronic business, with products and ser-
vices existing more in the world of information than in the world
of things.
As the century ends, Glazer’s hypotheses now read like
a prophecy of many organizations’ initiatives. Modular customiza-
tion, customer relationship management, rapid product devel-
opment, responsiveness to customers, strategic alliances, and
parallel information processing (another way to describe the
distributed, collaborative decision-making known by the rubric
empowerment) have become central to business conversation
and—to a lesser extent—to business action. Glazer’s description
of interactive, electronic commerce summarizes the way some
of his earlier hypotheses are now playing out on the World Wide
Web.
But these innovations, when attempted, are often adopted
piecemeal. If taken together, Glazer’s insights about information-
intensive businesses constitute more than a description of initia-
tives and improvements that can help make-and-sell businesses
survive in the Information Age. They imply the need for a radically
new business model—the sense-and-respond model.
Consider once more this question: What can planning and
strategy mean when unpredictability becomes the norm? This
chapter covered some of the disruptions created by the Informa-
tion Age. The center of gravity of economic value creation shifts
from tangibles to intangibles. Information arrives in an ever-
36 C H A P T E R 2

increasing flood. New economic rules arise and old ones mutate.
The relationship of companies to both their customers and their
competitors alters dramatically. Strategy, clearly, must be flexible
and adaptive. The next chapter will look at the effect these
changes should have on the concept of business strategy. Does
strategy have any meaning at all in the casinos of the Informa-
tion Age?