Csac06[1].p

CSAC06 1/13/07 9:22 Page 169

6
Organization Structure
and Management
Systems

Ultimately, there may be no long-term sustainable advantage other than the
ability to organize and manage.
—JAY GALBRAITH AND ED LAWLER

I’d rather have first-rate execution and second-rate strategy anytime than
brilliant ideas and mediocre management.
—JAMIE DIMON, CEO, JP MORGAN CHASE & CO.

OUTLINE

l Introduction and Objectives l Hierarchy in Organizational Design
Hierarchy as Coordination: Modularity
l The Evolution of the Corporation
Hierarchy as a Control: Bureaucracy
Firms and Markets Mechanistic and Organic Forms
Emergence of the Modern Corporation Rethinking Hierarchy
Organizational Change Since the
Mid-Twentieth Century l Applying the Principles of
Organizational Design
l The Organizational Problem:
Defining Organizational Units
Reconciling Specialization With
Organizing on the Basis of Coordination
Coordination and Cooperation
Intensity
Specialization and Division of Labor Other Factors Influencing the Definition
The Coordination Problem of Organizational Units
The Cooperation Problem: Incentives
and Control

169

CSAC06 1/13/07 9:22 Page 170

170 PART II THE TOOLS OF STRATEGY ANALYSIS

l Alternative Structural Forms Human Resource Management Systems
The Functional Structure Corporate Culture as a Control
The Multidivisional Structure Mechanism
Matrix Structures Integrating Different Control
Beyond Hierarchy? Mechanisms

l Management Systems for l Summary
Coordination and Control
l Self-Study Questions
Information Systems
Strategic Planning Systems l Notes
Financial Planning and Control Systems

Introduction and Objectives
“Great strategy; lousy implementation,” is an epithet applied to organizational failures
from Philip II of Spain’s disastrous attack on England with the Spanish Armada1 to the
unravelling of the global strategy of Vodafone, the world’s biggest mobile telecom com-
pany.2 The idea that the formulation of strategy can be separated from its implementation
has become institutionalized by the numerous strategic management texts that devote
separate sections to strategy formulation and strategy implementation.
This supposed division between formulation and implementation is fiction. At the most
obvious level, formulating a strategy without taking into account the conditions under
which it will be implemented will result in a poorly designed strategy. A fundamental flaw
in the corporate planning systems of 25 years ago was separating strategy formulation –
the task of corporate executives and strategic planners – from its implementation by
divisional heads and middle managers.
The design of organization structure and management form key components of strat-
egy implementation. Hence, the view of strategy formulation and strategy implementa-
tion as a sequential process is summed up in the adage “structure follows strategy.” Yet,
management guru Tom Peters argues the reverse: if capabilities are the primary basis of
strategy, and if capabilities are a product of organizational structure, then strategy follows
structure.3 The key point, however, is not whether strategy or structure takes precedence,
but the recognition that the two are closely interdependent.4 For companies, such as
Benetton, with its closely coordinated network of local suppliers and worldwide network
of franchised retailers, or Amway, with its pyramid of commission-based, independent
distributors, strategy is defined by these firm’s organizational structures.
Having established that how companies organize themselves is fundamental to their
strategy and their performance, the goal of this chapter is to introduce the key concepts
and ideas necessary to understand and design companies’ structures and systems. The
approach is concise and selective. I do not intend to offer a potted overview of organiza-
tional theory. My aim is to introduce some basic principles of organizational design and
to apply these to key aspects of firm structure. The principles outlined here will be further

CSAC06 1/13/07 9:22 Page 171

CHAPTER 6 ORGANIZATION STRUCTURE AND MANAGEMENT SYSTEMS 171

developed in later chapters when we consider strategies within particular business
contexts. For example, Chapter 11 considers the organizational conditions conducive to
innovation; Chapter 12 considers organization and organizational change within mature
industries; Chapter 13 discusses vertical structures and outsourcing; Chapter 14 examines
the structure and systems of the multinational corporation; and Chapter 16 deals with
organizing the multibusiness company.

By the time you have completed this chapter you will be able to:

l Recognize the key organizational innovations that have shaped the evolution of
the modern corporation.

l Understand the basic principles that determine the structural characteristics of
complex human organizations.

l Apply the principles of organizational design to recommend the types of
organizational structure suited to particular tasks and particular business
environments.

l Understand the role of information systems, strategic planning, ﬁnancial
control, and human resource management in the coordination and control of
corporations.

l Appreciate the forces that are causing companies to seek new organizational
structures and management systems.

The Evolution of the Corporation
Firms and Markets
Most of the world’s production of goods and services is undertaken by corporations
– enterprises with a legal identity that is distinct from the individuals that own the
enterprise. The main exceptions include agriculture and crafts in the developing
world, where family-based production predominates, and services such as defense,
policing, and education that are usually provided by government organizations.
This has not always been so. Until the late 19th century, the world’s only large-scale
organizations were the Roman Catholic church and national armies (see Strategy
Capsule 6.1). The only large ﬁrms were colonial trading companies such as the Dutch
East India Company, Hudson’s Bay Company, and the United Africa Company. As
late as the 1850s, the largest enterprises in the US in terms of numbers of workers were
agricultural plantations.5 Most manufacturing was organized through networks of
self-employed, home-based workers. The English woolen industry consisted of home-
based spinners who purchased raw wool (on credit) from a merchant to whom they

CSAC06 1/13/07 9:22 Page 172


STRATEGY CAPSULE 6.1

Reorganizing the Prussian Army, 1857–1870

In 1857, Helmuth von Moltke was the situations, even without direct instruction from
appointed commander-in-chef of the Prussian the general staff. Moreover, individual officers
army. During the next 12 years he completely where interchangeable.
reorganized it. His new structure was based on The test of the new organization came with
divisions that were controlled and coordinated the Franco-Prussian War of 1870. The French
by a general staff. Each division was a stand- army was one of the biggest and most experi-
ardized unit with the same composition of enced in Europe, yet within weeks it was
infantry, cavalry, and artillery and the same size, routed. The French were defeated not by
structure, equipment, and training methods. superior numbers, armaments, or valor, but by
The general staff comprised headquarters gen- a superior organizational system.
erals and specialist units: engineering, intelli- Like the Prussian army, the French army had
gence, training, and supply. decentralized by creating formations that in-
A key role of the general staff was officer cluded units of infantry, cavalry and artillery.
training. Each year 120 young officers were Yet no standardization had been achieved.
selected from the whole officer corps for in- The units had different sizes, structures, and
tensive training at a war academy that placed methods. Most serious were the deficiencies
a strong emphasis on strategic and tactical of the French general staff, which comprised
planning. Each year, the top 12 graduates of the commander-in-chief supported by messen-
the war academy were selected for several gers and clerks, but with no effective means of
years’ further training and development with coordinating the different army units.
Moltke at the general staff before being
Sources: M. Howard, The Franco-Prussian War: The Ger-
assigned to one of the divisions. The idea was
man Invasion of France, 1870–1871 (London: Routledge,
that, through common training, each officer 1991); R. Stark, Sociology, 10th edn (Wadsworth Publish-
would react with similar responses to new ing, 2006).

sold the yarn; the merchant resold the yarn to home-based weavers from whom he
purchased cloth. This “putting-out” system survived until the introduction of powered
looms, when weavers relocated to factories and eventually became employees rather
than independent contractors.
The business corporation is one of the greatest innovations of modern civilization.
The rise of the corporation as the predominant institution for organizing production
is one of the central features of modern economic development. This rise reflects the
efficiency and effectiveness of corporations – relative to other institutions – in organ-
izing economic activity. In the capitalist economy, production is organized in two
ways: in markets – by the price mechanism – and in firms – by managerial direction.
The relative roles of firms and markets are determined by efficiency: if the adminis-
trative costs of firms are less than the transaction costs of markets (as occurred in
the English textile industry after the introduction of the factory system), transactions

CSAC06 1/13/07 9:22 Page 173


will tend to be organized within firms rather than across markets. We shall revisit the
transaction cost theory when we consider vertical integration in Chapter 13.

Emergence of the Modern Corporation
According to business historian Alfred Chandler, the modern corporation emerged as
a result of two “critical transformations.”6

Line-and-Staff Structure Initially, most companies were small and operated
from a single plant or office. Lack of transportation limited each firm’s market to its
immediate vicinity, while lack of communication prevented firms from operating in
multiple location. The railroad and the telegraph changed all that – but to operate over
a wider geographical area, firms needed new organizational structures and manage-
ment techniques. In the US, the railroad companies were the first to create geo-
graphically separate operating units managed by an administrative headquarters. This
organizational form was termed a line-and-staff structure. Employees are either line,
allocated to operational tasks within the operating units, or staff, administrators and
functional specialists located at head office.
By the end of late 19th century, simple line-and-staff structures had developed into
more complex functional structures – companies such as Du Pont, Sears Roebuck and
Company, and Shell Transport and Trading managed a number of separate operating
units with large functional departments that conducted sales, finance, R&D, legal
affairs, and other specialist activities. Other large business enterprises were organized
as holding companies – Standard Oil (of the US), Mitsui (of Japan), and the British
South Africa Company were created by a series of acquisitions in which the parent
company bought controlling equity stakes in a number of other companies.

The Multidivisional Corporation The second critical transformation was the
emergence during the 1920s of the divisionalized corporation, which, over time,
replaced both the centralized, functional structures that characterized most industrial
corporations and the loose-knit holding companies created in the merger wave of the
early 20th century. The pioneers were DuPont and General Motors.

l At DuPont, increasing size and a widening product range strained the
functional structure and top management became overloaded:
. . . the operations of the enterprise became too complex and the problems
of coordination, appraisal and policy formulation too intricate for a small
number of top officers to handle both long-run, entrepreneurial and
short-run, operational administrative activities.7
The solution devised by Pierre Du Pont was to decentralize: ten product
divisions were created, each with their own sales, R&D, and support
activities. The corporate head office headed by an Executive Committee took
responsibility for coordination, strategy, and resource allocation.8
l General Motors, which had grown by acquisition into a loose holding
company, adopted a similar structure as a solution to the problems of weak
financial control and a confused product line. The new structure (shown in
Figure 6.1) was based on two principles: the chief executive of each division
was fully responsible for the operation and performance of that division,

CSAC06 1/13/07 9:22 Page 174


FIGURE 6.1 General Motors Corporation: organizational structure, 1921

Board of Directors

President Executive Committee

Financial GM Acceptance Legal General
Staff Corporation Department Advisory Staff

SOURCE: A. P. SLOAN, MY YEARS WITH GENERAL MOTORS (ORBIT PUBLISHING, 1972): 57.
Chevrolet Sheridan Oldsmobile Buick Cadillac GM Export
Division Division Division Division Division Company

Canadian
Division

GM Truck Samson Oakland Inter- Scripps
Division Tractor Division company Booth
Division Parts Corp.
Division

while the general office, headed by the president, was responsible for the
development and control of the corporation as a whole, including:
– monitoring return on invested capital within the divisions;
– coordinating the divisions (including establishing terms for interdivisional
transactions);
– establishing a product policy.9
The primary feature of the divisionalized corporation was the separation of oper-
ating responsibilities, which were vested in general managers at the divisional level,
from strategic responsibilities, which were located at the head office. The divisional-
ized corporation reconciled central coordination with the efficiencies and respon-
siveness of operational decentralization.

Organizational Change Since the Mid-Twentieth Century
Since the end of the Second World War, business enterprises have continued to evolve
their structures and systems at a rapid rate. Increased scope and complexity has
resulted in the multidivisional form developing into the matrix organization – where
separate hierarchies coordinate around products, functions, and geographical areas.
The quest for flexibility and responsiveness has resulted in the delayering of hier-
archies, the shift from functionally organized headquarters staff to shared services
organizations, and the creation of flexibility and responsiveness through alliances,
networks, and outsourcing partnerships.

CSAC06 1/13/07 9:22 Page 175


Most striking has been the rapid evolution of management systems – operational and
capital expenditure budgeting, corporate planning, and management-by-objectives –
during the 1950s and 1960s, through to knowledge management and corporate
social responsibility during recent years.
However, our purpose is not to review history not to identify best practice. Our
challenge is to appreciate the basic principles of organizational design so that we can
design organizations that are appropriate to specific purposes and circumstances and
recognize the fit between strategy, structure, and the business environment.

The Organizational Problem: Reconciling
Specialization With Coordination and Cooperation
According to Henry Mintzberg:
Every organized human activity – from making pots to placing a man on the
moon – gives rise to two fundamental and opposing requirements: the division of
labor into various tasks, and the coordination of these tasks to accomplish the
activity. The structure of the organization can be defined simply as the ways in
which labor is divided into distinct tasks and coordination is achieved among
these tasks.10
We begin with these two fundamental organizational requirements: specialization and
coordination.

Specialization and Division of Labor
Firms exist because they are efficient institutions for the organization of economic
activities, particularly the production of goods and services. The fundamental source
of efficiency in production is specialization, especially the division of labor into
separate tasks. The classic statement on the gains due to specialization is Adam Smith’s
description of pin manufacture:
One man draws out the wire, another straightens it, a third cuts it, a fourth points
it, a fifth grinds it at the top for receiving the head; to make the head requires two
or three distinct operations; to put it on is a peculiar business, to whiten the pins
is another; it is even a trade by itself to put them into the papers.11
Smith’s pin makers produced about 4,800 pins per person each day. “But if they
had all wrought separately and independently, and without any of them having been
educated to this peculiar business, they certainly could not each have made 20, per-
haps not one pin, in a day.” Similarly, Henry Ford experienced huge productivity gains
by installing moving assembly lines and assigning individuals to highly specific pro-
duction tasks. Between the end of 1912 and early 1914, the time taken to assemble
a Model T fell from 106 hours to just over six hours. More generally, the difference
in human productivity between modern industrial society and primitive subsistence
society is the result of the efficiency gains from individuals specializing.
But specialization comes at a cost. The more a production process is divided be-
tween different specialists, the greater are the costs of coordination. The more volatile
and unstable the external environment, the greater the number of decisions that
need to be made and the higher are coordination costs. Hence, the more stable is the

CSAC06 1/13/07 9:22 Page 176


environment, the greater is the optimal division of labor. This is true both for firms
and for entire societies. Civilizations are built on increased division of labor, which is
only possible through stability. As Somalia, Afghanistan, and the Congo have demon-
strated so tragically, once chaos reigns, societies regress toward subsistence mode
where each family unit must be self-sufficient.

The Coordination Problem
No matter how great the specialist skills possessed by individuals, unless these indi-
viduals can coordinate their efforts, production doesn’t happen. The current chal-
lenge for every coach of a national soccer team is how to coordinate the efforts of a
group of talented individuals within a limited time before the 2010 World Cup finals.
Conversely, the exceptional performance of organizations such as Wal-Mart, the
Cirque du Soleil, and the Berlin Philharmonic Orchestra are primarily the result of
superb coordination between organizational members. How do individuals within
organizations coordinate their efforts? Let us look at the operation of four different
coordination mechanisms:
l Price. In the market, coordination is achieved through the price mechanism.
Price mechanisms also exist within firms. Different departments and divisions
may trade on an arm’s-length basis, where internal prices (transfer prices) are
either negotiated or set by corporate headquarters.
l Rules and directives. A key feature of firms is the existence of employment
contracts. Unlike self-employed workers, who negotiate market contracts for
individual tasks, employees enter general employment contracts where they
agree to perform a range of duties as required by their employer. Authority is
exercised by means of general rules (“Employees will report for work not later
than 8.30 a.m.”) and specific directives (“Miss Moneypenny, show Mr. Bond
his new cigarette case with 3G communication and a concealed death ray”).
l Mutual adjustment. The simplest form of coordination involves the mutual
adjustment of individuals engaged in related tasks. In soccer or doubles tennis,
each player coordinates with fellow team members without any authority
relationship among them. Such mutual adjustment occurs in all teams and
work groups where there is no formal leader.
l Routines. Where activities are performed recurrently, coordination based on
mutual adjustment and rules becomes institutionalized within organizational
routines. As we noted in the previous chapter, these “regular and predictable
sequences of coordinated actions by individuals” are the foundation of
organizational capability. If organizations are to perform complex activities at
extreme levels of efficiency and reliability, coordination by rules, directives, or
mutual adjustment is not enough – coordination must become embedded in
routines.
The relative roles of these different coordination devices depend on the types of
activity being performed and the intensity of collaboration required. Price mechanisms
work well in situations of “arm’s-length” coordination. For example, in coordinating
production and sales, it may be sufficient to offer sales personnel simple price incen-
tives such as higher commission rates on those products where inventories are high.
Rules tend to work well for activities where standardized outcomes are required

CSAC06 1/13/07 9:22 Page 177


and the decision-making abilities of the operatives involved may be limited – most
quality control procedures involve the application of simple rules. Routines form the
basis for coordination in most activities where close interdependence exists between
individuals, whether a basic production task (supplying customers at Starbucks) or
a more complex activity (performing a heart by-pass operation or implementing a
systems integration project for a multinational corporation).

The Cooperation Problem: Incentives and Control
The discussion of coordination has dealt only with the technical problem of integrat-
ing the actions of different individuals. However, coordination problems are not
entirely solved by implementing coordination mechanisms, there is also the problem
of different organizational members having conflicting goals. This is referred to as the
cooperation problem. Overcoming goal conflict requires creating incentives and controls.
The economics literature analyzes goal misalignment in terms of agency problems.12
An agency relationship exists when one party (the principal) contracts with another
party (the agent) to act on behalf of the principal. The problem is ensuring that the
agent acts in the principal’s interest. Within the firm, the major agency problem
is between owners (shareholders) and managers. The problem of ensuring that
managers operate companies to maximize shareholder wealth is at the center of
the corporate governance debate. During the 1990s, changes in top management
remuneration – in particular the increasing emphasis given to stock options – were
intended to align the interests of managers with those of shareholders.13 However, at
Enron, WorldCom, and other companies, these incentives encouraged managers to
manipulate reported earnings rather than to work for long-term profitability.
Agency problems exist throughout the hierarchy. For individual employees, sys-
tems of incentives, monitoring, and appraisal are designed to encourage pursuit of
organizational objectives and overcome employees’ tendency to do their own thing or
simply shirk. The organization structure may create its own problems. Organizational
departments create their own subgoals that do not align with one another. The
classic conflicts are between different functions: sales wishes to please customers,
production wishes to maximize output, R&D wants to introduce mind-blowing new
products, while finance worries about profit and loss.
Several mechanisms are available to management for achieving goal alignment
within organizations:
l Control mechanisms typically operate on the basis of managers supervising
groups of subordinates. Managerial supervision involves monitoring behavior
and performance, while subordinates are obliged to seek approval for actions
that lie outside their area of authority. Such hierarchical supervision and
control rests on both positive and negative incentives. Positive incentives are
typically the reward of promotion up the hierarchy in return for compliance;
negative incentives are dismissal and demotion for failing to acquiesce to rules
and directives.
l Financial incentives are designed to reward performance. Such incentives
extend from piece-rates for production workers to stock options and profit
bonuses for executives. Such performance-related incentives have two main
benefits: first, they are high powered – they relate rewards directly to output –
and second, they economize on the need for costly monitoring and

CSAC06 1/13/07 9:22 Page 178


supervision of employees. The problems of pay-for-performance arise where
employees work in teams or on activities where output is difficult to measure.
Linking pay to individual performance may discourage collaboration.
l Shared values. Some organizations are able to achieve high levels of
cooperation and low levels of goal conflict without extensive control
mechanisms or performance-related incentives. Churches, charities, clubs,
and most voluntary organizations fall into this category. The reason is the
commonality of goals between organizational members. Since Peters and
Waterman’s In Search of Excellence, the role of shared core values in sustained
organizational success is well documented.14 The role of culture as a control
mechanism that is an alternative to bureaucratic control or the price
mechanism is central to Bill Ouchi’s concept of clan control.15 The role of
corporate culture in encouraging conformity to organizational goals has long
been recognized among Japanese corporations. However, in western
companies too – in Wal-Mart, Four Seasons Hotels, Amway, and the Shell
Group – the presence of shared values and principles encourages the
alignment of individual and corporate goals without necessarily undermining
the individuality of organizational members. Such control saves on monitoring
costs: self-control and informal monitoring by co-workers substitute for
managerial supervision and financial incentives. Similar observations can be
made about companies driven by a common technological vision. At Apple
Computer in the 1980s, the belief that Apple was leading a computer
revolution that would transform and democratize society permitted intense
cooperation with very little formal control. As one cynic noted: “What’s the
difference between Apple and the Boy Scouts? In the Boy Scouts, the kids
have adult supervision!”
We shall return to these issues of incentives and control when we consider the
management systems of companies.

Hierarchy in Organizational Design
How have companies addressed these basic needs for specialization, coordination,
and cooperation? The traditional approach to large-scale organization has been to
create hierarchy. Despite the negative associations that currently attach to hierarchy,
I shall argue that hierarchical structures are essential for creating efficient and flex-
ible coordination in complex organizations. The critical issue is not whether or not to
organize by hierarchy – there is little alternative – but how the hierarchy should be
structured and how the different parts of it should relate to one another. Hier-
archies come in many forms. Traditionally, hierarchy is associated with bureaucratic
approaches to management control. However, hierarchical structures may also be
organized along organic lines. The past decade has seen important changes in how
companies structure and manage hierarchical structures.

Hierarchy as Coordination: Modularity
Hierarchy is fundamental to the structure of all organizations; indeed, according to
Herbert Simon, hierarchy is present in virtually all complex systems.16 If a hierarchy

CSAC06 1/13/07 9:22 Page 179


is defined as a system composed of interrelated subsystems, examples of hierarchy
include:
l The human body, which is composed of a hierarchy of cells, organs, and
subsystems such as the respiratory system, nervous system, digestive system,
and so on.
l Physical systems are composed at the macro level of planets, stars, and
galaxies, and at the micro level of subatomic particles, atoms, and molecules.
l Social systems consist of individuals, families, communities, tribes or
socio-economic groups, and nations.
l A book consists of letters, words, sentences, paragraphs, and chapters.
Note that this is a broader concept of hierarchy than that encountered in most
discussions of organization design, where hierarchy is identified with administrative
hierarchy, in which organizational members are arranged in superior–subordinate
relationships and authority flows downward from the top.
Viewed in this broad context of subsystems and component units, there are two key
advantages to hierarchical structures:
1 Economizing on coordination. As we have noted, the gains from specialization
come at the cost of coordination. Suppose there are five programmers
designing a piece of customized computer software. If they are structured
as a “self-organized team,” where coordination is by mutual adjustment
(see Figure 6.2a), ten bilateral interactions must be managed. Alternatively,
suppose the programmer with the biggest feet is selected to be supervisor.
In this simple hierarchy (Figure 6.2b), there are only four relationships to be
managed. Of course, this says nothing about the quality of the coordination:
if the programmers’ work is highly interdependent, hierarchical relationships
may not allow for the richness of communication and collaboration that a
team structure would permit. As an organization increases in size and
complexity, so the communication-economizing benefits of hierarchically
arranged modules increase:
By breaking up a complex system into discrete pieces – which can then
communicate with one another through standard interfaces within a
standardized architecture – one can eliminate what would otherwise be
an unmanageable spaghetti tangle of interconnections.17

FIGURE 6.2 How hierarchy economizes on coordination

(a) Self-organizing Team: (b) Hierarchy:
ten interactions four interactions

CSAC06 1/13/07 9:22 Page 180


2 Adaptability. Hierarchical, modular systems are able to evolve more rapidly
than unitary systems that are not organized into subsystems. Such adaptability
requires some degree of decomposability: the ability of each component
subsystem to operate with some measure of independence from the other
subsystems. Modular systems that allow significant independence for each
module are referred to as loosely coupled.18 In developing a new model of
automobile, a modular structure permits different subassemblies (engine,
brakes, steering, electricals, etc.) to be developed by separate teams that do
not need constant communication and coordination with the designers of
every other unit.19 Once developed, defects can be corrected by replacing a
single subunit – the engine, the gearbox, or the exhaust system – without
having to scrap the entire car. Similar advantages exist for modular
organizations. In a divisionalized firm, such as GE, decisions can be made in
GE’s jet engines business that do not require coordination with GE’s other
business areas. Similarly, GE can acquire a new business or dispose of an
existing subsidiary without requiring organizational changes throughout the
company.20
The efficiency and flexibility advantages of modularity and hierarchical commun-
ication are evident in Nelson Mandela’s restructuring of the ANC (see Strategy
Capsule 6.2). Let’s look more closely at administrative hierarchies associated with
bureaucratic or mechanistic organizational forms.

Hierarchy as a Control: Bureaucracy
I have shown that hierarchy is an efficient solution to the problem of coordination in
organizing complex tasks. To the extent that hierarchy is also a device for exercising
control, it is also one solution to the problem of cooperation in organizations. The
administrative hierarchy, in which power is located at the apex of the hierarchy
and delegated downward, has been the basic design for large organizations since the
Ch’in dynasty of China in 220 BC. Administrative hierarchies operate as bureaucracies.
According to Max Weber, writing at the end of the 19th century, bureaucracy is based
on the following principles:
1 Specialization through a “systematic division of labor” with clear job
definitions and individual authority limited to the sphere of work
responsibilities.
2 Hierarchical structure with “each lower office under the control and
supervision of a higher one.”
3 Coordination and control through rules and standard operating procedures.
4 Standardized employment rules and norms.
5 Separation of management and ownership.
6 Separation of jobs and people, where the organization is defined by positions
and their associated responsibilities and authority, not by individuals; there is
no ownership of the position by the individual.
7 Rational-legal authority based on “belief in the legality of enacted rules and
the right of those elevated to authority under such rules to issue commands.”
8 Formalization in writing of “administrative acts, decisions, and rules.”21

CSAC06 1/13/07 9:22 Page 181


STRATEGY CAPSULE 6.2

Hierarchical Structures: The 1952 Mandela Plan for the ANC

Along with many others, I had become con- cell, which in urban townships consisted of
vinced that the government intended to roughly ten houses on a street. A cell steward
declare the ANC (African National Congress) would be in charge of each of these units. If a
and the SAIC (South African Indian Congress) street had more than ten houses, a street
illegal organizations, just as it had done with steward would take charge and the cell stew-
the Communist Party. It seemed inevitable that ards would report to him. A group of streets
the state would attempt to put us out of busi- formed a zone directed by a chief steward,
ness as a legal organization. With this in mind, who was in turn responsible to the secretariat
I approached the National Executive with the of the local branch of the ANC. The secretariat
idea that we must come up with a contingency was a subcommittee of the branch executive,
plan . . . They instructed me to draw up a plan which reported to the provincial secretary. My
that would enable the organization to operate notion was that every cell and street steward
from underground. This strategy came to be would know every person and family in his area,
known as the Mandela-Plan, or simply, M-Plan. so that he would be trusted by his people and
The idea was to set up organizational know whom to trust. The cell steward arranged
machinery that would allow the ANC to take meetings, organized political classes, and col-
decisions at the highest level, which could then lected dues. He was the linchpin of the plan.
be swiftly transmitted to the organization as The plan was accepted and was imple-
a whole without calling a meeting. In other mented immediately. Word went out to the
words, it would allow the organization to con- branches to begin to prepare for this covert
tinue to function and enable leaders who were restructuring . . . As part of the M-Plan, the
banned to continue to lead. The M-Plan was ANC introduced an elementary course of polit-
designed to allow the organization to recruit ical lectures for its members throughout the
new members, respond to local and national country. These lectures were meant not only to
problems and maintain regular contact be- educate but to hold the organization together.
tween the membership and the underground They were given in secret by branch leaders.
leadership. Those members in attendance would in turn
I worked on it for a number of months and give the same lectures to others in their homes
came up with a system that was broad enough and communities.
to adapt itself to local conditions and not
fetter individual initiative, but detailed enough Source: Nelson Mandela, Long Walk to Freedom (London:
to facilitate order. The smallest unit was the Little, Brown, 1994): 134–5.

Bureaucracies attempt to minimize most of the traits that characterize human
beings and their interaction: creativity, personality, variation, and emotion. For this
reason, Burns and Stalker describe bureaucratic organizations as mechanistic,22 while
Mintzberg calls them machine bureaucracies.23

CSAC06 1/13/07 9:22 Page 182


Mechanistic and Organic Forms
During the first half of the 20th century, the bureaucratic model dominated thinking
about organizational structure. This reflected Weber’s clear articulation of the prin-
ciples of bureaucracy and the fact that most large-scale organizations – the military and
civil service in particular – embodied these principles. However, as management
theory developed, interest grew in alternatives to bureaucracy.
During the 1950s and 1960s, the human relations school recognized that coopera-
tion and coordination within organizations was about social relationships as well as
bureaucratic principles. A study of Scottish engineering companies by Burns and
Stalker identified two organizational forms: mechanistic forms, characterized by
bureaucracy, and organic forms that were less formal, where coordination relied on
mutual adjustment and interaction was more flexible. The mechanistic form was found
mainly in stable markets; the organic form predominated in unstable markets with
rapid technological change.
Table 6.1 contrasts key characteristics of the two forms.
The relative merits of the two organizational forms depend on the activities
undertaken and the surrounding environment. Where an organization is supplying
standardized goods or services (beverage cans, blood tests, or haircuts for army
inductees) using well-understood processes, in an environment where change is slow
and predictable, the bureaucratic model with its standard operating procedures and
high levels of specialization offers substantial efficiency advantages. The problems
occur when the bureaucratic model has to produce heterogeneous outputs from
heterogeneous inputs, using poorly understood technologies, in an environment
where change requires constant adjustment. Here, the bureaucracy fails because
greater organizational flexibility is required.
But even when faced with variability in the outside environment, firms may
attempt to retain the advantages of bureaucracy by trying to control variation.
McDonald’s business system is highly mechanistic, relying heavily upon standardized,
formalized working practices that are carefully documented in the company’s
operating procedures. Making this system work requires that McDonald’s carefully
controls its inputs to reduce variation: potatoes are carefully selected for size and

TABLE 6.1 Mechanistic vs. Organic Organizational Forms
SOURCE: ADAPTED FROM RICHARD BUTLER, DESIGNING ORGANIZATIONS: A DECISION-

Feature Mechanistic Organic

Task definition Rigid and highly specialized Flexible and less narrowly defined
Coordination and Rules and directives vertically Mutual adjustment, common
MAKING PERSPECTIVE (LONDON: ROUTLEDGE, 1991): 76.

control imposed culture
Communication Vertical Vertical and horizontal
Knowledge Centralized Dispersed
Commitment and To immediate superior To the organization and
loyalty its goals
Environmental context Stable with low technological Unstable with significant
uncertainty technological uncertainty and
ambiguity

CSAC06 1/13/07 9:22 Page 183


shape, managers are carefully selected and trained, consumer tastes and expectations
are carefully managed through advertising and promotion.
Within companies, the organization of different functions and departments
depends on these same variables. Stable, standardized activities such as payroll, treas-
ury, taxation, customer support, and purchasing activities tend to operate well when
organized along bureaucratic principles; research, new product development, mar-
keting, and strategic planning require more organic modes of organization.

Rethinking Hierarchy
Hierarchical organizations add layers as they get bigger. Thus, with a fixed span of
control of three, a firm with four employees (including the CEO) is organized into two
layers, five to 13 employees requires three layers, from 14 to 41 employees requires
four layers, and 42 to 122 employees requires five layers. (Sketch this for yourself.)
If the hierarchy is run as a bureaucracy with centralized power, growth implies an
increasing ratio of managers to operatives, slower decision making, and increased loss
of control.24
In a stable environment with limited decision-making pressure on top manage-
ment, such ponderousness is of little consequence. However, in a fast-paced business
environment, the slow movement of information up the hierarchy and decisions down
the hierarchy can be fatal. As the business environment has become increasingly tur-
bulent, so administrative hierarchy organized along bureaucratic principles has
become increasingly unpopular.
At the same time, efforts to reform and restructure corporate hierarchies do not
amount to a rejection of hierarchy as an organizing principle. So long as there are
benefits from the division of labor, hierarchy is inevitable.25 The critical issue is to
reorganize hierarchies in order to increase responsiveness to external change. The
organizational changes that have occurred in giant corporations such as BP and
General Electric have retained the basic multidivisional structures of the companies,
but reduced the number of hierarchical layers, decentralized decision making, shrunk
headquarters staffs, emphasized horizontal rather than vertical communication, and
shifted the emphasis of control from supervision to accountability.26
The trend towards decentralization has not been one way. Some companies engage
in decentralization followed by a phase of centralization. Thus, BP pursued radical
decentralization during 1994–8, but by 2000– 4 was re-centralizing decision making
and control. Nickerson and Zenger argue that this type of “structural modulation” in
a company’s formal structure is effective in achieving an optimal balance between
centralization and decentralization.27

Applying the Principles of Organizational Design
We have established that the fundamental problem of organization is reconciling
specialization with coordination and cooperation. The basic design for complex
organizations – whether they are business enterprises, religious orders, political
associations, or criminal organizations – is hierarchy. The essence of hierarchy is
creating specialized units coordinated and controlled by a superior unit. But this does
not take us very far. On what basis should specialized units be defined? How should
decision-making authority be allocated? And what kind of relationships should there
be between different organizational units?

CSAC06 1/13/07 9:22 Page 184


In this section, we will tackle the first two of these questions: the basis of group-
ing and the allocation of decision-making power. In the next section, we identify some
typical organizational structures found in business enterprises. Then, in the following
section we shall look at structuring relations between units – the operation and design
of management systems.

Defining Organizational Units
In creating a hierarchical structure, on what basis are individuals assigned to organ-
izational units within the firm? This issue is fundamental and complex. Multinational,
multiproduct companies are continually grappling with the issue of whether they
should be structured around product divisions, country subsidiaries, or functional
departments, and periodically they undergo the disruption of changing from one
to another. Some of the principal bases for grouping employees are common tasks,
products, geography, and process:
l Tasks. Organizational units can be created around common tasks. This usually
means grouping together employees who do the same job – thus, a firm might
create a machine shop, a maintenance department, a secretarial pool, and a
sales office.
l Products. Where a company offers multiple products, these can provide a
basis for structure. In a department store, departments are defined by
products: kitchen goods, bedding, lingerie, and so on. PepsiCo comprises
three main product groups: PepsiCo Beverages, Frito-Lay (snack foods),
and Quaker Foods (cereals and processed foods).
l Geography. Where a company serves multiple local markets, organizational
units can be defined around these localities. Wal-Mart is organized by
individual stores, groups of stores within an area, and groups of areas within
a region. The Roman Catholic church is organized into parishes, dioceses,
and archdioceses.
l Process. A process is a sequence of interlinked activities. An organization
may be viewed as a set of processes: the product development process, the
manufacturing process, the sales and distribution process, and so on. A
process may correspond closely with an individual product, or a process
may be dominated by a single task. Functional organizations tend to combine
task-based and process-based grouping.

Organizing on the Basis of Coordination Intensity
How do we decide whether to use task, product, geography, or process to define
organizational units? The fundamental issue is achieving the coordination necessary
to integrate the efforts of different individuals. This implies grouping individuals
according to the intensity of their coordination needs. Those individuals whose
tasks require the most intensive coordination should work within the same organiza-
tional unit.
l In a geographically dispersed organization where communication across
distance is difficult, the organization must be built on local units. The ANC is
an example (see Strategy Capsule 6.2).

CSAC06 1/13/07 9:22 Page 185


l Where an organization is not particularly diversified in relation to products
and does not need to be differentiated by location, but possesses strong
functional specializations, then a grouping around functional tasks is
appropriate. For example, British Airways is organized primarily around
functions: flight operations, engineering, marketing, sales, customer service,
human resources, information, and finance.
l Where a company is diversified over many products and these products are
substantially different in terms of technology and markets, it is vital that
individuals who work on the same product should interact closely – a
product-based organization is the appropriate structure. Virtually all
diversified companies – General Electric, 3M, Sony, Siemens, and Unilever –
are organized by product divisions.
Having created organizational units that bring together individuals whose coordina-
tion needs are most intense, the next challenge is to create hierarchical control
that permits effective coordination while giving as much operational autonomy as
possible to the subordinate units. Oliver Williamson refers to this as the principle
of hierarchical decomposition. At the operating level (where decision making is high
frequency), organization units are created where the interactions are strong. At the
strategic level (where decision making is low frequency), a separate organization unit
is created to exercise coordination and direction. Hence:
The hierarchical decomposition principle can be stated as follows: Internal
organization should be designated in such a way as to effect quasi-independence
between the parts, the high frequency dynamics (operating activities) and low
frequency dynamics (strategic planning) should be clearly distinguished, and
incentives should be aligned within and between components so as to promote
both local and global effectiveness.28
To organize according to coordination needs requires understanding the nature of
interdependence within an organization. James Thompson distinguished three levels
of interdependence: pooled interdependence (the loosest), where individuals operate
independently but depend on one another’s performance; sequential interdependence,
where the output of one individual is the input of the other; and reciprocal inter-
dependence (the most intense), where individuals are mutually dependent. Thompson
argued that organizational design needed to begin with creating organizational units
where interdependence was the most intense.29
Over time, the relative importance of these different dependencies change. Hence,
companies need to change the basis on which they define their structure. For exam-
ple, as trade and communication between countries has become easier and consumer
preferences between countries have become more homogeneous, multinational
corporations have shifted from geographically based structures to worldwide product
divisions.

Other Factors Influencing the Definition of
Organizational Units
Coordination requirements are not the only consideration in deciding how to group
together employees and activities within the firm. Additional factors that influence
the efficiency of different organizational arrangements include:

CSAC06 1/13/07 9:22 Page 186


l Economies of scale. There may be advantages in grouping together activities
where scale economies are present. Thus, it may be desirable to group
together research activities even if there is little coordination among different
research projects, simply to exploit scale economies in specialized facilities
and technical personnel.
l Economies of utilization. It may also be possible to exploit efficiencies from
grouping together similar activities that result from fuller utilization of
employees. Even though there may be little need for individual maintenance
engineers to coordinate with one another, establishing a single maintenance
department permits maintenance personnel to be utilized more fully than
assigning a maintenance engineer to each manufacturing cell.
l Learning. If establishing competitive advantage requires building distinctive
capabilities, firms must be structured to maximize learning. Typically, it was
assumed that learning was best achieved by grouping together individuals
doing similar jobs – creating a manufacturing engineering department, a
quality control department, and a finance function. More recently, it has been
observed that the specialized functional and discipline-based knowledge
may be less important than architectural knowledge – knowing how to link
together specialized knowledge from different fields. This implies the creation
of multifunctional work groups comprising experts from different knowledge
bases.
l Standardization of control systems. Tasks may be grouped together to achieve
economies in standardized control mechanisms. An advantage of the typing
pool and the sales department was that employees doing near-identical jobs
could be subject to the same system of monitoring, performance
measurement, training, and behavioral norms. Creative activities such as
research and new product development need to be managed in a different way
from routine activities such as manufacturing and accounting – hence they
should be located in different organizational units.30

Alternative Structural Forms
On the basis of these alternative approaches to grouping tasks and activities, we can
identify three basic organizational forms: the functional structure, the multidivisional
structure, and the matrix structure.

The Functional Structure
Single-business firms tend to be organized along functional lines. Grouping together
functionally similar tasks is conducive to exploiting scale economies, promoting learn-
ing and capability building, and deploying standardized control systems. Since cross-
functional integration occurs at the top of the organization, functional structures are
conducive to a high degree of centralized control by the CEO and top management
team.
However, even for single-product firms, functional structures are subject to
the problems of cooperation and coordination. Different functional departments
develop their own goals, values, vocabularies, and behavioral norms which make

CSAC06 1/13/07 9:22 Page 187


cross-functional integration difficult. As the size of the firm increases, the pressure
on top management to achieve effective integration increases. Because the different
functions of the firm tend to be tightly coupled rather than loosely coupled, there is
limited scope for decentralization. In particular, it is very difficult to operate individual
functions as semiautonomous profit centers.
The real problems arise when the firm grows its range of products and businesses.
As we noted with DuPont during the early 20th century, once a functionally organized
company expands its product range, coordination within each product area becomes
difficult.
Although the long-term trend among very large companies has been for product-
based, divisionalized companies to replace functionally organized companies, the
trend is not entirely one way. As companies mature, the need for strong centralized
control and well-developed functional capabilities has caused some companies to
revert to functional structures. For example:
l When John Scully became CEO of Apple in 1984, the company was
organized by product: Apple II, Apple III, Lisa, and Macintosh. Despite
strong cross-functional coordination within each product group, there was
little integration across products. Each product was completely incompatible
with the others, and the structure failed to exploit scale economies within
functions. Scully’s response was to reorganize Apple along functional lines to
gain control, reduce costs, and achieve a more coherent product strategy.
l General Motors, pioneer of the multidivisional structure, has adopted a more
functional structure. As its strategic priorities have shifted from differentiation
and segmentation toward cost efficiency, it has maintained its brand names
(Cadillac, Oldsmobile, Chevrolet, Buick), but merged the separate divisions
into a more functionally based structure to exploit scale economies and faster
technical transfer (see Figure 6.3 and compare it with Figure 6.1).

The Multidivisional Structure
We have seen how the product-based, multidivisional structure emerged during the
20th century in response to the coordination problems caused by diversification. The
key advantage of divisionalized structures (whether product based or geographically
based) is the potential for decentralized decision making. The multidivisional struc-
ture is the classic example of a loose-coupled, modular organization where business-
level strategies and operating decisions can be made at the divisional level, while the
corporate headquarters concentrates on corporate planning, budgeting, and provid-
ing common services.
Central to the efficiency advantages of the multidivisional corporation is the
ability to apply a common set of corporate management tools to a range of different
businesses. At ITT, Harold Geneen’s system of “managing by the numbers” allowed
him to cope with over 50 divisional heads reporting directly to him. At British
Petroleum, John Browne’s system of “performance contracts” allows direct reporting
by over 20 “strategic performance units.” Divisional autonomy also fosters the
development of top management leadership capability among divisional heads – an
important factor in CEO succession.
The large, divisionalized corporation is typically organized into three levels: the
corporate center, the divisions, and individual business units, each representing a

CSAC06 1/13/07 9:22 Page 188


FIGURE 6.3 General Motors Corporation: organizational structure, 1997

Board of Directors

President’s Council Corporate Functions

North Delphi GM International Hughes
American Automotive Acceptance Operations Electronics
Operations Systems Corporation

GM Europe
Midsize Small GM Vehicle Development
and Car Power Sales, and and Technical
Asian and
Luxury Group Train Marketing Cooperation
Pacific
Car Group Group Group
Operations
Group

Latin
American,
African, and
Middle East
Operations

FIGURE 6.4 General Electric: organizational structure, 2002

Corporate Executive Office
Chairman & CEO

Corporate Staff
Service Divisions Finance Business R&D Human Legal
Development Resources
SOURCE: BASED ON INFORMATION IN GENERAL ELECTRIC ANNUAL REPORT, 2001

GE Aircraft GE Trans- GE GE GE GE
Engines portation Industrial Plastics Appliances Supply
Systems

GE Power GE Medical GE GE NBC GE
Systems Systems Lighting Specialty Capital
Materials

26 businesses organized into 5 segments:
Consumer Mid-market Specialized Specialty Equipment
Services Financing Financing Insurance Management

CSAC06 1/13/07 9:22 Page 189


distinct business for which ﬁnancial accounts can be drawn up and strategies formu-
lated. Figure 6.4 shows General Electric’s organizational structure at the corporate and
divisional levels.
In Chapter 16, we shall look in greater detail at the organization of the multi-
business corporation.

Matrix Structures
Whatever the primary basis for grouping, all companies that embrace multiple
products, multiple functions, and multiple locations must coordinate across all three
dimensions. Organizational structures that formalize coordination and control across
multiple dimensions are called matrix structures.
Figure 6.5 shows the Shell management matrix (prior to reorganization in 1996).
Within this structure, the general manager of Shell’s Berre reﬁnery in France reported
to his country manager, the managing director of Shell France, but also to his business

FIGURE 6.5 Royal Dutch Shell Group: pre-1996 matrix structure

CSAC06 1/13/07 9:22 Page 190


sector head, the coordinator of Shell’s refining sector, as well as having a functional
relationship with Shell’s head of manufacturing.
Many diversified, multinational companies, including Philips, Nestlé, and Unilever,
adopted matrix structures during the 1960s and 1970s, although in all cases one
dimension of the matrix tended to be dominant in terms of authority. Thus, in the
old Shell matrix the geographical dimension, as represented by country heads and
regional coordinators, had primary responsibility for budgetary control, personnel
appraisal, and strategy formulation.
During the past two decades, most large corporations have dismantled or reorgan-
ized their matrix structures. Shell abandoned its matrix during 1995–6 in favor of a
structure based on four business sectors: upstream, downstream, chemicals, and gas
and power. During 2001–2, the Swiss-Swedish engineering giant ABB, abandoned its
much-lauded matrix structure in the face of plunging profitability and mounting debt.
In fast-moving business environments companies have found that the benefits from
formally coordinating across multiple dimensions have been outweighed by excessive
complexity, larger head office staffs, slower decision making, and diffused authority.
Bartlett and Ghoshal observed that matrix structures “led to conflict and confusion;
the proliferation of channels created informational logjams as a proliferation of com-
mittees and reports bogged down the organization; and overlapping responsibilities
produced turf battles and a loss of accountability.”31
Yet, all complex organizations that comprise multiple products, multiple functions,
and multiple geographical markets need to coordinate within each of these dimen-
sions. The problem of the matrix organization is not that it attempts to coordinate
across multiple dimensions – in complex organizations such coordination is essential
– but that this multiple coordination is over-formalized, resulting in excessive corporate
staffs and over complex systems that slow decision making and dull entrepreneurial
initiative. The trend has been for companies to focus formal systems of coordination
and control on one dimension, then allowing the other dimensions of coordination
to be mainly informal.32 Thus, while Shell is organized primarily around four business
sectors and these sectors exercise financial and strategic control over the individual
operating companies, Shell still has country heads, responsible for coordinating all
Shell activities in relation to legal, taxation, and government relations within each
country, and functional heads, responsible for technical matters and best-practice
transfer within their particular function, whether it is manufacturing, marketing,
or HR.

Beyond Hierarchy?
For several decades consultants and management scholars have proclaimed the death
of hierarchical structures in business firms. In 1993, two of America’s most promin-
ent scholars of organization announced: “. . . the new organizational revolution is
sweeping one industry after another . . . quantum changes in manufacturing and
computer-mediated communication technologies have given managers radical new
options for designing organizations.” The new organizations featured “. . . flatter hier-
archies, decentralized decision making, greater tolerance for ambiguity, permeable
internal and external boundaries, empowerment of employees, capacity for renewal,
self-organizing units, self-integrating coordination mechanisms.”33
As I noted in the earlier section on “Rethinking Hierarchy,” there have been
substantial changes in the way in which corporate hierarchies have been organized.

CSAC06 1/13/07 9:22 Page 191


Layers have been removed; mechanistic formality has been replaced by organic in-
formality. Yet, hierarchy remains as the basic structural form of almost all companies.
Are there alternative modes of organization?
Several organizational forms have been identified which, although they comprise
some hierarchical elements, are sufficiently distinctive to be regarded as alternative
organizational forms:
l Adhocracies. In some organizations, the presence of shared values, motivation
and willingness to participate, mutual respect, and communication
effectiveness may allow a high level of coordination with little need for
hierarchy, authority, or tools of control. These organizations, which Henry
Mintzberg calls adhocracies,34 feature flexible, spontaneous coordination and
collaboration around problem solving and other nonroutine activities.
Adhocracies tend to exist among organizations where expertise is prized.
In research organizations, new product development groups, jazz bands, and
consulting firms, each specialist is valued for his or her expertise and there is
little exercise of authority.
l Team-based and project-based organizations. Adhocracies are one example of
an organizational form based on informal structure with flexible patterns of
coordination. Flexibility and adaptability can also be achieved in project-based
organizations – common in sectors such as construction, consulting, oil
exploration, and engineering services – where business takes the form of
projects of limited duration. Because every project is different, and every
project goes through a changing sequence of activities, each project needs to
be undertaken by a closely interacting team that relies on problem solving and
mutual adjustment as well as rules and routines. Increasingly, companies are
introducing elements of team- and project-based organizations into their
conventional divisional and functional structures. For example, in most
divisionalized corporations, new product development, change management,
knowledge management, and research is organized in projects and undertaken
by teams.
l Networks. Localized networks of small, closely interdependent firms have
been a feature of manufacturing for many hundreds of years. In Italy such
networks are prominent in the clothing industry of Prato, near Florence, and
in packaging equipment.35 Hollywood movie making36 and microelectronics
in Silicon Valley have similar structures – highly specialized firms that
coordinate to design and produce complex products. Often these networks
feature a central firm that acts as a “systems integrator,”37 as in the case of
Benetton and Toyota.38 In fast-moving industries, the ability of highly
specialized, know-how intensive firms to reconfigure their relationships can
be conducive to innovation, product differentiation, and rapid new product
development. In the developing world, such networks can be a viable
alternative to industrial development where large enterprises are lacking.39
These different organizational forms share several common characteristics:
1 A focus on coordination rather than control. In contrast to the “command-
and-control” hierarchy, these structures focus almost wholly upon achieving
coordination. Financial incentives, culture, and social controls take the place
of hierarchical control.

CSAC06 1/13/07 9:22 Page 192


2 Reliance on coordination by mutual adjustment. Central to all nonhierarchical
structures is their dependence on voluntaristic coordination through bilateral
and multilateral adjustment. The capacity for coordination through mutual
adjustment has been greatly enhanced by information technology.
3 Individuals in multiple organizational roles. Reconciling complex patterns of
coordination with high levels of flexibility and responsiveness is difficult if job
designs and organizational structures are rigidly defined. Adhocracies and
team-based organizations feature individuals switching their organizational
roles and occupying multiple roles simultaneously. For example, for most of
the 1990s, AES had no finance function, no HR function, no safety or
environmental affairs functions, and no public relations department. These
functions were performed by teams of operatives and line managers.

Management Systems for Coordination and Control
The relationship between management systems and organizational structure is sim-
ilar to that between the skeleton and bodily systems in the human body. The skeleton
provides the framework; the respiratory system, digestive system, and nervous system
are the means by which the body operates. Computer networks offer another analogy:
the hardware provides the structure and the software provides the systems that make
the network operational.
Management systems provide the mechanisms of communication, decision making,
and control that allow companies to solve the problems of achieving both coordina-
tion and cooperation. Four management systems are of primary importance: the
information systems, the strategic planning systems, the financial systems, and the
human resource management systems.

Information Systems
Information is fundamental to the operation of all management systems. Commun-
ication technology – the telephone and telegraphy – were essential for the emergence
of the modern corporation. The computer has had an equally dramatic impact
during the past half century. Accounting systems are key components of firms’
information systems. They collect, organize, and communicate financial information
to top management and other parts of the organization.
Administrative hierarchies are founded on vertical information flows: the upward
flow of information to the manager and the downward flow of instructions. The trend
towards decentralization and informality in organizations rests on two key aspects
of increased information availability: information feedback to the individual on job
performance, which has made self-monitoring possible, and information networking,
which has allowed individuals to coordinate their activities voluntarily without
hierarchical supervision. For example, a central element of total quality management
has been recognition that regular, real-time, performance feedback to employees per-
mits them to take responsibility for quality control, reducing or eliminating the need
for supervisors and quality controllers. During the past decade, corporate intranets,
web-based information systems, and groupware have transformed organizations’
capacity for decentralized coordination.

CSAC06 1/13/07 9:22 Page 193


Strategic Planning Systems
Small enterprises can operate successfully without an explicit strategy. The firm’s
strategy may exist only in the head of the founder and, unless the founder needs to
write a business plan in order to attract outside financing, the strategy may never be
articulated. Most large companies have a regular (normally annual) strategic planning
process. Multibusiness companies typically have systematic strategic planning pro-
cesses, the outcome of which is a documented corporate plan that integrates the
business plans of the individual divisions.
Whether formal or informal, systematic or ad hoc, documented or not, the strat-
egy formulation process is an important vehicle for achieving coordination within a
company. As discussed in Chapter 1, the strategy process brings together knowledge
from different parts of the company, ensures consistency between the decisions being
made at different levels and in different parts of the company, and commits managers
to ambitious performance targets.
The system through which strategy is formulated varies considerably from company
to company. Even after the entrepreneurial startup has grown into a large company,
strategy making may remain the preserve of the chief executive. Functional managers
may provide key inputs such as financial projections and market analysis, but the
key elements of strategy – goals, new business developments, capital investment,
and key competitive initiatives – are often decided by the chief executive.40 At MCI
Communications former CEO Orville Wright observed: “We do it strictly top–down
at MCI.”41 The first director of strategic planning was warned: “If you ever write a
strategic plan, you will be fired!”
As companies mature, their strategic planning processes become more systematized
and typically follow an annual cycle. Strategic plans tend to be for three to five years
and combine top–down initiatives (indications of performance expectations and
identification of key strategic initiatives) and bottom–up business plans (proposed
strategies and financial forecasts for individual divisions and business units). After
discussion between the corporate level and the individual businesses, the business
plans are amended and agreed and integrated into an overall corporate plan that is
presented to and agreed by the board of directors. Figure 6.6 shows a typical strategic
planning cycle.
The resulting strategic plan typically comprises the following elements:

l A statement of the goals the company seeks to achieve over the planning
period with regard to both financial targets (e.g., targets for revenue growth,
cost reduction, operating profit, return on capital employed, return to
shareholders) and strategic goals (e.g., market share, new products, overseas
market penetration, and new business development). For example, in BP’s
February 2006 strategy statement, the company established its “primary
objective is to deliver sustainable growth in free cash flow,” which it would
achieve through “growing production by about 4% a year to 2010” and
“delivering further improvements in return on average capital employed
relative to our peer group.”42
l A set of assumptions or forecasts about key developments in the
external environment to which the company must respond.
For example, BP’s 2006–10 strategic plan assumed an oil price of
$40 a barrel.

CSAC06 1/13/07 9:22 Page 194


FIGURE 6.6 The generic strategic planning cycle

Corporate
Guidelines
Draft Discuss Revised
Approval
Business with Business
Forecasts/ by Board
Plans Corporate Plans
Scenarios/
Planning
Assumptions
Corporate
Plan

Capex
Annual
Performance Budget
Performance
Review
Targets

l A qualitative statement of how the shape of the business will be changing in
relation to geographical and segment emphasis, and the basis on which the
company will be establishing and extending its competitive advantage. For
example, BP’s 2006–10 strategy emphasized capital discipline (capex to
increase by $0.5 billion per year) and upgrading BP’s asset portfolio by
selective divestments of about $3 billion per year.
l Specific action steps with regard to decisions and projects, supported by a set
of mileposts stating what is to be achieved by specific dates. For example,
BP’s strategic commitments included 24 start-ups of upstream projects during
2006–8, $6 billion capital expenditure on its Russian joint venture, and
growth in wind power to 450 MW .
l A set of financial projections, including a capital expenditure budget and
outline operating budgets. For example, BP’s 2006 strategy statement set a
capital expenditure budget of $15–16 billion per year, shareholder
distribution of $50 billion during 2006–8, and operating costs to increase
at less than the rate of inflation.
Although strategic planning tends to emphasize the specific commitments and
decisions that are documented in written strategic plans, the most important aspect
of strategic planning is the strategy process: the dialog through which knowledge is
shared and ideas communicated, the consensus that is established, and the commit-
ment to action and results that is built.
Increasing turbulence in the business environment has caused strategic planning
processes to become less formalized and more flexible. For example, among the
world’s largest petroleum majors, the key changes have been as follows:
l Strategic plans have become less concerned with specific actions and became
more heavily focused on performance targets, especially on financial goals
such as profitability and shareholder return. Planning horizons have also
shortened (two to five years is the typical planning period).

CSAC06 1/13/07 9:22 Page 195


l Companies recognized the impossibility of forecasting the future and based
their strategies less on medium- and long-term economic and market forecasts
of the future and more on more general issues of strategic direction (in the
form of vision, mission, and strategic intent) and alternative views of the
future (e.g., using scenario analysis).
l Strategic planning shifted from a control perspective, in which senior
management used the strategic planning mechanisms as a means of controlling
decisions and resource deployments by divisions and business units and
departments, toward more of a coordination perspective, in which the strategy
process emphasized dialog involving knowledge sharing and consensus
building. As a result, the process became increasingly informal and put less
emphasis on written documents.
l A diminishing role for strategic planning staff as responsibility for strategic
decisions and the strategy-making process become located among senior
managers.43

Financial Planning and Control Systems
Financial flows form the life blood of the enterprise. Revenues from customers pro-
vide the funds to pay suppliers and employees and any surplus remunerates owners.
If inflows are insufficient to cover outflows, the firm becomes insolvent. Hence, finan-
cial systems are inevitably the primary mechanism through which top management
seeks to control the enterprise. At the center of financial planning is the budgetary
process. This involves setting and monitoring financial estimates with regard to
income and expenditure over a specified time period, both for the firm as a whole
and for divisions and departments. Budgets are in part an estimate of incomes and
expenditures for the future, in part a target of required financial performance in terms
of revenues and profits, and in part a set of authorizations for expenditure up
to specified budgetary limits. Two types of budget are set: the capital expenditure
budget and the operating budget.

The Capital Expenditure Budget Capital expenditure budgets are established
through both top–down and bottom–up processes. From the top down, strategic plans
establish annual capital expenditure budgets for the planning period both for the com-
pany as a whole and for individual divisions. From the bottom up, capital expenditures
are determined by the approval of individual capital expenditure projects. Companies
have standardized processes for evaluating and approving projects. Requests for fund-
ing are prepared according to a standardized methodology, typically based on a fore-
cast of cash flows discounted at the relevant cost of capital (adjusted for project risk).
The extent to which the project’s returns are sensitive to key environmental uncer-
tainties is also estimated. Capital expenditure approvals take place at different levels
of a company according to their size. Projects up to $5 million might be approved by
a business unit head, projects up to $25 million might be approved by divisional top
management, larger projects might need to be approved by the top management com-
mittee, while the biggest projects require approval by the board of directors.

The Operating Budget The operating budget is a pro forma profit and loss state-
ment for the company as a whole and for individual divisions and business units for

CSAC06 1/13/07 9:22 Page 196


the upcoming year. It is usually divided into quarters and months to permit continual
monitoring and the early identification of variances. The operating budget is part fore-
cast and part target. It is set within the context of the performance targets established
by the strategic plan. Each business typically prepares an operating budget for the
following year that is then discussed with the top management committee and,
if acceptable, approved. At the end of the financial year, business-level divisional
managers are called upon to account for the performance over the past year.

Human Resource Management Systems
Strategies may arise from principles, formulae, or divine inspiration, but their imple-
mentation depends on people. Ultimately, strategic and financial plans come to noth-
ing unless they influence the ways in which people within the organization behave. To
support strategic and financial plans, companies need systems for setting goals, creat-
ing incentives and monitoring performance at the level of the individual employee.
Human resource management has the task of establishing an incentive system that
supports the implementation of strategic plans and performance targets through align-
ing employee and company goals, and ensuring that each employee has the skills
necessary to perform his or her job. The general problem, we have noted, is one of
agency: how can a company induce employees to do what it wants?
The problem is exacerbated by the imprecision of employment contracts. Unlike
most contracts, employment contracts are vague about employee performance ex-
pectations. The employer has the right to assign the employee to a particular category
of tasks for a certain number of hours per week, but the amount of work to be per-
formed and the quality of that work are unspecified. Employment contracts give the
right to the employer to terminate the contract for unsatisfactory performance by the
employee, but the threat of termination is an inadequate incentive: it imposes costs
on the employer and only requires the employee to perform better than a new hire
would. Moreover, the employer has imperfect information as to employees’ work
performance – in team production, individual output is not separately observable.44
The firm can ensure the employee’s compliance with organizational goals using
direct supervision of the type that administrative hierarchies are designed to do. The
weaknesses of such administrative supervision are, first, there is little incentive for
performance in excess of minimum requirements, second, supervision imposes costs,
and third, the system presupposes that the supervisor has the knowledge required to
direct the employee effectively.
The key to promoting more effective cooperation is for more sophisticated incen-
tives than the threat of dismissal. The principal incentives available to the firm for
promoting cooperation are compensation and promotion. The key to designing
compensation systems is to link pay either to the inputs required for effective job
performance (hours of work, punctuality, effort, numbers of customers visited) or to
outputs. The simplest form of output-linked pay is piecework (paying for each unit
of output produced) or commission (paying a percentage of the revenue generated).
Relating pay to individual performance is suitable for tasks performed individu-
ally. However, firms exist primarily to permit complex coordination among indi-
viduals; encouraging such collaboration requires linking pay to team or departmental
performance. Where broad-based, enterprise-wide collaboration is required, there
may be little alternative to linking pay to company performance through some form
of profit sharing.

CSAC06 1/13/07 9:22 Page 197


Corporate Culture as a Control Mechanism
We have already noted how shared values can align the goals of different stake-holders
within the organization. More generally, we can view the culture of the organization
as a mechanism for achieving coordination and control. Corporate culture comprises
the beliefs, values, and behavioral norms of the company, which influence how em-
ployees think and behave.45 It is manifest in symbols, ceremonies, social practices,
rites, vocabulary, and dress. It is embedded within national cultures, and incorporates
elements of social and professional cultures. As a result, a corporate culture may be
far from homogeneous: very different cultures may be evident in the research lab, on
the factory floor, and within the accounting department. To this extent, culture is not
necessarily an integrating device – it can contribute to divisiveness.
Culture can play an important role in facilitating both cooperation and coordina-
tion. In companies such as Storbucks, Shell, Nintendo, and Google, strong corporate
cultures create a sense of identity among employees that facilitates communication
and the building of organizational routines, even across national boundaries. The
unifying influence of corporate culture is likely to be especially helpful in assisting
coordination through mutual adjustment in large cross-functional teams of the type
required for new product development. One of the advantages of culture as a co-
ordinating device is that it permits substantial flexibility in the types of interactions it
can support.
The extent to which corporate culture assists coordination depends on the char-
acteristics of the culture. Salomon Brothers (now part of Citigroup) was renowned for
its individualistic, internally competitive culture; this was effective in motivating drive
and individual effort, but did little to facilitate cooperation. The British Broadcasting
Corporation has a strong culture that reflects internal politicization, professional
values, internal suspicion, and a dedication to the public good, but without a strong
sense of customer focus.46 The culture of a leading British bank was described as
one of complaint, negativity, and pessimism.47 However, culture is far from being a
flexible management tool. Cultures take a long time to develop and cannot easily be
changed. As the external environment changes, a highly effective culture may become
dysfunctional. The Los Angeles Police Department’s culture of professionalism
and militarism, which made it one of the most admired and effective police forces
in America, later contributed to problems of isolation and unresponsiveness to com-
munity needs.48

Integrating Different Control Mechanisms
The past ten years have seen substantial progress in integrating different control
systems. As strategy has become more and more focused on creating shareholder
value, so financial planning has become more closely integrated with strategic
planning. Performance management systems have also done much to link strategic
and financial planning with human resource management – especially in terms of
goal setting and performance appraisal. The central aspect of the “metrics” move-
ment within management is the ability not just to establish quantitative goals for
individual employees and groups, but to create mechanisms for measuring and
reporting the attainment of these targets. The balanced scorecard system outlined in
Chapter 2 is but one approach to this linking of employee goals to company-wide
goals.

CSAC06 1/13/07 9:22 Page 198


Summary
The internal structure and systems of the firm are this area. Organization theory is an exceptionally
not simply a matter of “strategy implementation,” rich field that still lacks adequate integration of
which can be separated from the hard analytics of its component disciplines: sociology, psychology,
strategy formulation. Not only is strategy imple- organizational economics, systems theory, popula-
mentation inseparable from strategy formulation, tion ecology, and organizational evolution. While
but issues of structure and systems are central to business enterprises continue to experiment with
the fundamental issues of competitive advantage new organizational forms, we business school
and strategy choice – the existence of organiza- academics are still struggling to articulate general
tional capability in particular. principles of organizational design.
Despite the importance of these issues, this The chapters that follow will have more to say
chapter provides only a brief introduction to some on the organizational structures and management
of the key issues in organization design. Subse- systems appropriate to different strategies and
quent chapters develop many of the themes more different business contexts. In the final chapter
fully in relation to particular areas of strategy (Chapter 17) we shall explore some of the new
and particular business contexts. Nevertheless, our trends and new ideas that are reshaping our
progress is limited by the weakness of theory in thinking about organizational design.

Self-Study Questions
1 As DuPont expanded its product range (from explosives into paints, dyes, plastics, and
synthetic fibers) why do you think that the functional structure (organized around
manufacturing plants and other functions such as sales, finance, and R&D) became unwieldy?
Why did the multidivisional structure based on product groups facilitate administration?

2 Explain (with reference to a diversified, divisionalized company such as General Electric) the
extent to which the multidivisional company may be regarded as a modular organization? To
what degree is each division an independent entity? What are the “standardized interfaces”
that allow the divisions to fit together into a coherent whole?

3 Within your own organization (whether a university, company, or not-for-profit
organization), which departments or activities are organized mechanistically and which
organically? To what extent does the mode of organization fit the different environmental
contexts and technologies of the different departments or activities?

4 The examples of Apple Computer and General Motors (see section on “Functional
Structure”) point to a more general feature of organizational structure over the product life
cycle. During the growth phase many companies adopt multidivisional structures, during
maturity and decline many companies revert to functional structures. Why might this be?

CSAC06 1/13/07 9:22 Page 199


(Note: you may wish to refer to Chapter 10, which outlines the main features of the life cycle
model.)

5 Draw an organizational chart for a business school that you are familiar with. Does the
school operate with a matrix structure (e.g. are there functional/discipline-based departments
together with units managing individual programs)? Which dimension of the matrix is more
powerful, and how effectively do the two dimensions coordinate? How would you reorganize
the structure to make the school more efficient and effective?

Notes
1 G. Parker, The Grand Strategy of Philip II (New Haven: 18 J. D. Orton and K. E. Weick, “Loosely Coupled Systems:
Yale University Press, 1998). A Reconceptualization,” Academy of Management
2 “Vodafone: What Went Wrong?” Business Week ( June 6, Review 15 (1990): 203–23.
2006). 19 M. Sako and F. Murray, “Modular Strategies in Car
3 T. Peters, “Strategy Follows Structure,” California and Computers,” Financial Times, Mastering Strategy
Management Review, 26 (Spring 1984): 114–28. Part 11 (December 6, 1999): 4–7.
4 R. Whittington, A. Pettigrew, S. Peck, E. Fenton, and 20 Modularity in organizations is explored in a number of
M. Conyon, “Change and Complementarities in the articles. See R. Sanchez and J. T. Mahoney, “Modularity,
New Competitive Landscape,” Organization Science 10 Flexibility, and Knowledge Management in Product and
(1999): 583–96. Organizational Design,” Strategic Management Journal
5 A. D. Chandler, The Visible Hand: The Managerial 17, Winter Special Issue (1996): 63–76; M. A. Schilling,
Revolution in American Business (Cambridge, MA: “Toward a General Modular Systems Theory and its
MIT Press, 1977): Chapter 2. Application to Interfirm Product Modularity,” Academy
6 A. D. Chandler, Strategy and Structure (Cambridge: of Management Review 25 (2000): 312–34; C. Baldwin
MIT Press, 1962); Chandler, The Visible Hand, op. cit. and K. Clark, “Managing in an Age of Modularity,”
7 Chandler, Strategy and Structure, op. cit.: 382–3. Harvard Business Review (September–October 1997):
8 http://heritage.dupont.com/floater/fl_management/ 84–93.
floater.shtml. 21 The quotes in this section are from Max Weber’s
9 A. P Sloan, My Years at General Motors (London:
. Economy and Society: An Outline of Interpretive
Sidgwick & Jackson, 1963): 42–56. Sociology (Berkeley: University of California Press,
10 H. Mintzberg, Structure in Fives: Designing Effective 1968).
Organizations (Englewood Cliffs: Prentice Hall, 22 T. Burns and G. M. Stalker, The Management
1993): 2. of Innovation (London: Tavistock Institute,
11 A. Smith, The Wealth of Nations (London: Dent, 1910): 5. 1961).
12 S. Ross, “The Economic Theory of Agency,” American 23 H. Mintzberg, op. cit.: Chapter 9.
Economic Review, 63 (1973): 134–9; K. Eisenhardt, 24 The control loss phenomenon in hierarchies is analyzed
“Agency Theory: An Assessment and Reviews,” Academy in O. E. Williamson, “Hierarchical Control and Optimal
of Management Review, 14 (1989): 57–74. Firm Size,” Journal of Political Economy 75 (1967):
13 M. Conyon, S. Peck, G. Sadler, and L. Read, “The 123–38.
Structure of Executive Compensation Contracts: UK 25 H. J. Leavitt, “Why Hierarchies Thrive,” Harvard
Evidence,” Long Range Planning 33 (2000): 478–503. Business Review (March 2003): 97–102.
14 T. Peters and R. Waterman, In Search of Excellence 26 R. Whittington and A. Pettigrew, “New Notions of
(New York: Harper & Row, 1982). Organizational Fit,” Financial Times, Mastering Strategy
15 W G. Ouchi, Theory Z (Reading, MA: Addison-Wesley,
. Part 10 (November 29, 1999): 8–10.
1981). 27 J. Nickerson and T. Zenger, “Being Efficiently Fickle:
16 H. A. Simon, “The Architecture of Complexity,” A Dynamic Theory of Organizational Choice,”
Proceedings of the American Philosophical Society 106 Organization Science 13 (September–October 2002):
(1962): 467–82. 547–67.
17 R. N. Langlois, “Modularity in Technology and 28 O. E. Williamson, “The Modern Corporation: Origins,
Organization,” Journal of Economic Behavior and Evolution, Attributes,” Journal of Economic Literature
Organization 49 (September 2002): 19–37. 19 (1981): 1537– 68.

CSAC06 1/13/07 9:22 Page 200


29 J. D. Thompson, Organizations in Action (New York: 38 J. H. Dyer and K. Nobeoka, “Creating and managing
McGraw-Hill, 1967). The nature of interdependence in a high-performance knowledge-sharing network: the
organizational processes is revisited in T. W Malone,
. Toyota case,” Strategic Management Journal 21 (2000):
K. Crowston, J. Lee, and B. Pentland, “Tools for 345–67; A. Camuffo, P Romano, and A. Vinelli, “Back
.
Inventing Organizations: Toward a Handbook of to the Future: Benetton Transforms Its Global
Organizational Processes,” Management Science 45 Network,” Sloan Management Review 43 (Fall 2001):
(March 1999): 489–504. 46–52.
30 The need for organizations to differentiate management 39 D. Wheeler, K. McKague, J. Thomson, R. Davies,
and organization between different functional J. Medalye, and M. Prada, “Sustainable Local Enterprise
departments and product units is discussed in Networks,” Sloan Management Review (Fall 2005):
P R. Lawrence and J. W Lorsch, Organization and
. . 33– 40.
Environment (Boston: Harvard Business School Press, 40 W C. Finnie, Hands-On Strategy: The Guide to Crafting
.
1986). Your Company’s Strategy (New York: John Wiley, 1994).
31 C. A. Bartlett and S. Ghoshal, “Matrix Management: 41 MCI Communications: Planning for the 1990s, Case
Not a Structure, a Frame of Mind,” Harvard Business No. 9-190-136 (Boston: Harvard Business School,
Review ( July–August 1990): 138– 45. 1990): 1.
32 “A Survey of the Company: The New Organization,” 42 BP Strategy and Fourth Quarter 2005 Results
Economist ( January 21, 2006). Presentation, February 7, 2006 (www.bp.com).
33 R. Daft and A. Lewin, “Where are the theories for the 43 R. M. Grant, “Strategic Planning in a Turbulent
new organizational forms?” Organization Science 3 Environment: Evidence from the Oil Majors,” Strategic
(1993): 1–6. Management Journal 24 (2003): 491–518. Similar
34 H. Mintzberg, op. cit.: Chapter 12. findings have been reported by the American
35 M. H. Lazerson and G. Lorenzoni, “The Firms that Feed Productivity and Quality Center among “best practice
Industrial Districts: A Return to the Italian Source,” companies” (Strategic Planning: Final Report, Houston:
Industrial and Corporate Change 8 (1999): 235– 66; APQC, 1996).
G. Lorenzoni and A. Lipparini, “The leveraging of 44 A. Alchian and H. Demsetz, “Production, Information
interfirm relationships as a distinctive organizational Costs, and Economic Organization,” American
capability: a longitudinal study,” Strategic Management Economic Review 62 (1972): 777–97.
Journal 20 (1999): 317–38; A. Grandori, Interfirm 45 E. H. Schein, “Organizational Culture,” American
Networks (London: Routledge, 1999). Psychologist 45 (1990): 109–19.
36 R. J. DeFilippi and M. B. Arthur, “Paradox in Project- 46 Tom Burns, The BBC: Public Institution and Private
based Enterprise: The Case of Film Making,” California World (London: Macmillan, 1977).
Management Review 42 (1998): 186–91. 47 J. Weeks, Unpopular Culture: The Ritual of Complaint in
37 G. Lorenzoni and C. Baden-Fuller, “Creating a a British Bank (Chicago: University of Chicago Press,
Strategic Center to Manage a Web of Partners,” 2004).
California Management Review 37, no. 3 (1995): 48 “LAPD: Storming the Rampart,” Economist (December
146–63. 2, 2000): 72.

Csac06[1].p

More Related Content

What's hot (18)

Similar to Csac06[1].p (20)

More from Kinshook Chaturvedi (20)

Recently uploaded (20)

Csac06[1].p