The current issue and full text archive of this journal is available at

www.emeraldinsight.com/0144-3577.htm

IJOPM
28,6 A systems perspective
on the death of a car company
Nick Oliver
562 University of Edinburgh Business School, University of Edinburgh,
Edinburgh, UK
Received 19 June 2007 Matthias Holweg
Revised 3 February 2008
Accepted 19 February 2008
Judge Business School, University of Cambridge, Cambridge, UK, and
Mike Carver
Oxford, UK

Abstract
Purpose – The aim of this paper is to understand how large and apparently successful organizations
enter spirals of decline that are very difficult to reverse. The paper examines the case of Rover, once
one of the largest car producers in the world, which collapsed in 2005. An analysis of strategic and
operational choices made over a period of 40 years investigates the reasons for, and consequences of, a
growing mismatch between the context faced by the company (industry dynamics, market conditions)
and its operational capabilities, a mismatch that ultimately brought about the company’s demise.
Design/methodology/approach – The paper is based on interviews with 32 people, including
senior managers (including four chief executives), government ministers and union officials who were
key decision makers within, or close to, the company during the period 1968 and 2005. Secondary
sources and documentary evidence (e.g. production and sales data) are used to build up a historical
picture of the company and to depict its deteriorating financial and market position from 1968
onwards.
Findings – The company was formed from a multitude of previously independent firms as part of a
government-sponsored agenda to build a UK National Champion in the car industry. The merged
company failed due to several factors including poor product development processes, poor
manufacturing performance, difficult labour relations, a very wide product portfolio and a lack of
financial control. Although strenuous efforts were made to address those issues, including periods of
whole or part ownership by British Aerospace, Honda and BMW, the company’s position deteriorated
until eventually production volumes were too low for viable operation.
Practical implications – The case of Rover highlights the importance of what has been termed “the
management unit” in complex systems. The management unit comprises processes and routines to
deal with challenges such as managing product portfolios, connecting strategic and operational
choices, and scanning and responding to the environment. In the case of Rover, a number of factors
taken together generated excessive load on a management unit frequently operating under conditions
of resource scarcity. We conclude that viewing corporate failure from a systems perspective, rather
than in terms of shortcomings in specific subsystems, such as manufacturing or product development,
yields insights often absent in the operations management literature.
Originality/value – The paper is of value by showing corporate failure from a systems perspective,
rather than in terms of shortcomings in specific subsystems, such as manufacturing or product
International Journal of Operations &
development; and yields insights often absent in the operations management literature. The Rover
Production Management case featured in the paper demonstrates the usefulness of systems ideas to understanding at least some
Vol. 28 No. 6, 2008 types of failure, not as an alterative to capability-based approaches, but in addition to them.
pp. 562-583
q Emerald Group Publishing Limited Keywords Automotive industry, Corporate strategy, Operations management, Business failures
0144-3577
DOI 10.1108/01443570810875368 Paper type Case study
Introduction The death
The start of the second automotive century has been interesting for the global motor of a car company
industry. The industry has seen both record profits and losses, as well as bankruptcies
amongst global suppliers and manufacturers, some of the largest industry mergers and
de-mergers, and – largely thanks to emerging new markets – increased global demand
for automobiles.
Recent dynamics within the global motor industry have several implications for the 563
academic study of the industry. Regional stereotypes that in the past have
predominantly compared Japanese producers to their western counterparts (Womack
et al., 1990; Cusumano and Takeishi, 1991; Hines, 1998) now only partially hold. All but
two Japanese vehicle manufacturers have been taken over, or had major stakes
acquired by, western manufacturers. Nissan, for example, which was once the subject
of best practice case studies (DETR, 1998), faced near-bankruptcy in the late 1990s and
merged with Renault.
The changing fortunes of the automotive industry have affected firms in Japan,
Europe and the US alike. Most prominently, the cost of restructuring (and downsizing)
in the US automotive industry has impacted on US domestic manufacturers in
particular. The Ford Motor Company (various years) posted losses of $12.6 bn in 2006,
and $2.67 bn loss in 2007; General Motors (various years) faced similar losses of $8.6 bn
in 2005, but has since returned to marginal profitability (Company Annual Reports).
The future of Chrysler is uncertain after its 2007 de-merger from Daimler, a joint
venture that entered the record books in Germany as destroying the most value
(approx. e40 bn) in the country’s post-war history. The Western European industry has
seen a stream of production facilities relocating to Eastern Europe, putting further
strain on national champions such as Renault, Volkswagen and Fiat. Moreover,
incumbent manufacturers are threatened by a wave of low-cost imports from China
and India, following the same pattern as the Japanese makers in the 1970s, and the
Koreans in the 1990s. In short, the level of competition in this very mature industry has
reached a level where it threatens the survival of new-entrant and established players
alike. In 2005 Rover, as one of the weakest of the established players, was the first
“national champion” in the western world market to exit the industry.
In this paper, we will explore how a large corporation entered a dynamic that ultimately
led to its demise. We do so by telling the story of the Rover Car Company, which went from
being one of the largest car producers in the world to oblivion over a period of 40 years.
Our point of departure is that the automotive industry, like other industries, follows
a technology cycle. More or less all countries with indigenous car industries have been
through a pattern of proliferation and consolidation of numbers of producers
(Utterback, 1994). The emergence of a new technology (such as the internal combustion
engine at the end of the nineteenth century) typically creates an influx of players, all
keen to try their hand in the new arena. This period usually lasts until a dominant
design emerges (for example, in the case of passenger cars, the use of four wheels,
propulsion by an internal combustion engine, an enclosed steel body, etc). At this point
the basis of competitive advantage shifts away from issues of basic design and
configuration, around which producers converge, and towards product and
organizational attributes such as production efficiency, style, marketing or other
criteria. Less capable companies are forced to exit the market, or are taken over by the
stronger ones. In the USA, for example, in the mid-1920s there were 75 vehicle
IJOPM producers, but by 1960 this was down to ten, and in 2006 there were just three. In the
28,6 UK, no fewer than 221 vehicle producers entered the market between 1901 and 1905, of
which 90 per cent had either exited the sector or gone out of business altogether by
1914 (Saul, 1962). In 1920, in the UK there were 90 manufacturers of passenger cars;
by 1929 this had dropped to 41; by 1939 to 33; by 1946 to 32; and by 1950 there were
just 20 – about the same as the number of producers in the USA at that time, but with a
564 much smaller total production volume (Church, 1994).
The significance of these patterns to this paper is that our analysis of the failure of
Rover should be seen against a background of industry consolidation and
rationalization (at least in the mature economies) that has been occurring over many
decades, and in which there are inevitably winners and losers. Our purpose in the
paper is to understand why one particular company, once in a position of apparent
strength, disappeared from sight. Did Rover lack capabilities that other car companies
possessed? Or did the company face particular circumstances, that other car companies
did not, that brought about its demise?
This paper and the research on which is based did not start out as a conventional
piece of hypothesis or theory testing. When Rover collapsed in 2005, we became
curious about the reasons for this, partly because of the explanations for the collapse
that appeared in the media. These largely focused on recent management decisions,
and placed the blame for the collapse on the Phoenix Consortium, who owned the
company for the final five years of its life. Although not explicitly stated, we set out
with the implicit assumption that at least some of the factors that brought
about the ultimate demise of Rover predated the Phoenix ownership; we also assumed
that these dynamics could not be understood purely on the basis of secondary data,
and that we would therefore need to gather data on the perceptions and context that
senior decision makers faced throughout the history of the firm.
We therefore interviewed a cross-section of key people involved in the company
from its creation in the late 1960s through to its collapse in 2005, complemented with
an analysis of secondary sources, as described in the section on the research approach.
Preliminary analysis of these data revealed that the company faced many challenges.
At various times in its life these included deficiencies in product development
processes and manufacturing systems; chronic labour relations problems; a very
challenging economic environment due to trade liberalization and unfavourable
exchange rates; and multiple changes of ownership, including a substantial period
under government ownership. Consequently, identifying a conceptual framework that
was simple enough to be comprehendible and yet at the same time comprehensive
enough to cope with such a diverse set of issues was a major challenge.
In practice, this occurred in two stages. First, a brief review of the operations
management literature was used to construct a basic framework to understand the core
processes and capabilities of a car company. Secondly, ideas from the literature on
systems thinking, in particular those on viable systems (Beer, 1984) were used
to provide insight into how the company failed as a complete system. We turned to
systems thinking because it became clear that the company faced multiple challenges
and failings in a variety of subsystems, not just one or two. Moreover, it was also clear
that many people in the company were aware of these failings at the time, but for a
variety of reasons appeared incapable of correcting them. This suggested that
a systems perspective might be the most appropriate one to take.
Capabilities of a car company The death
There is clearly much more to being a car company than simply assembling vehicles, of a car company
even though this is the most visible part of the process of vehicle production.
Car making comprises a complex set of processes, and orchestrating these successfully
is a major organisational challenge. A typical vehicle comprises 10-15,000 individual
parts, sourced by the manufacturer in the form of 2,000-4,000 distinct components
(Holweg and Pil, 2004), and a typical volume car model will have a production run 565
comprising of 500,000 units or more.
In this section, we develop a framework for analysing the core capabilities that car
companies must possess if they are to be effective. We use Heller et al. (2006) definition
of a “fully capable” car company as one that is “able to independently design,
manufacture and market a vehicle”. These capabilities are shown diagrammatically in
Figure 1 and their key characteristics are summarized in Table I.
Figure 1 shows the classic process of product development, production and
distribution typical of an automotive value chain. New product development typically
begins with analyses of both market needs and available technologies; the relative

Management System:
Corporate Management and Control; Financial Control; Portfolio Management; Brand
Management and Marketing; Resolution of Trade-Offs; Cross-Unit Learning and Improvement

Technology Product Retail and

Development Manufacturing Distribution
Development

Market and Component

Competitor Figure 1.
Supply Capabilities framework
Analysis

Capability The ability to . . .

Market and competitor analysis Identify market trends and competitor actions and feed this
intelligence into strategic decisions
Technology development Acquire, develop and deploy technology
Product development Develop competitive new products in a timely and cost-effective
way
Manufacturing Match market demand with supply, to competitive cost and
quality levels
Component supply Identify suppliers who can develop competitive subsystems
and components and supply these to the right levels of cost,
quality and delivery
Retail and distribution Establish and operate systems of retail, distribution and
aftermarket support in the appropriate markets
Management unit Allocate and control resources, establish and enforce standards, Table I.
manage the diversity/uniformity across subsystems (e.g. The seven core
functions and divisions) and over time capabilities
IJOPM influence of these determine whether a process is “technology push” or “market-pull”.
28,6 Such intelligence informs choices in product design and development, although the
extent to which this happens in practice can vary considerably.
In the automotive industry, developing new products involves many functions within
the car companies themselves, and dozens, possibly hundreds of suppliers. These
include engineers and designers who design the product, suppliers who provide a
566 variety of specialist parts, and the manufacturing function who have to produce the
vehicles in high volume and consistently to the required levels of quality and cost (Clark
and Fujimoto, 1991). The development of a new vehicle demands many thousands of
choices and decisions, large and small, by many people. Over successive development
cycles these choices and decisions combine to give companies and their products
particular attributes – capabilities, scale, brand values and other sources of competitive
advantage – or disadvantage. Strong brands, such as the BMW brand with its image of
the “ultimate driving machine” or Audi’s “Vorsprung durch Technik” do not develop
overnight, but rather require years, perhaps decades, of consistency in how choices in
design, manufacturing and marketing are made and presented to consumers (Bayley,
1986). Similarly, companies that lack consistency in such decisions, for example across
models or over time, for example due to changes in personnel or ownership, may suffer
from weak or confused brand attributes (Aaker, 1991).
The core capabilities in Table I are well recognized in the operations management
community. They include processes of product development (Clark and Fujimoto,
1991), manufacturing (Schonberger, 1982, 1986; Womack et al., 1990), buyer-supplier
relations (Lamming, 1993; Sako, 1992) and logistics and distribution and retail (Kiff,
1997; Holweg and Pil, 2004; Reichhart and Holweg, 2007). Other processes, such as
market intelligence and the management of retail systems have generally received less
attention by the operations management community, although such analyses do exist
(Delbridge and Oliver, 1991; Oliver and Delbridge, 1991).

Viable systems
Although the core capabilities framework goes a considerable way to providing a
diagnostic for success and failure in the auto industry, preliminary analysis of the Rover
data indicated that of itself the core capabilities framework did not explain the whole
picture. As we shall see, many people within the company, at various points in its life,
were acutely aware of shortcomings in areas such as product development,
manufacturing and labour relations and there were strenuous efforts to correct these.
The problem seemed to be that addressing all of these issues together and sufficiently
rapidly appeared to be beyond the reach of successive management teams.
This appeared to be a function of the totality of the parts, rather than any individual
element, and this realization led us towards systems theory, specifically the work of
Beer (1984) on systems viability, in search of an appropriate additional conceptual
framework.
The chief proponent of theory on systems viability is Beer. Beer’s ideas have been
developed in a number of books and articles (Beer, 1972, 1979, 1984; Jackson, 2001) and we
shall simply summarize the key points here. Beer’s first concept is that the major challenge
in managing any system is variety – that subsystems must be controlled by a
meta-system, or “management unit” and that the capacity of the metasystem
to process information must be commensurate with the ability of the subsystems to
generate it – otherwise overload will result, and the system is likely to enter a catatonic The death
state, and become paralysed. This builds on Ashby’s “Law of requisite variety” (1956) and of a car company
is similar to the information-processing perspective on organizational design (Galbraith,
1974), which maintains that the critical limiting factor of an organization is its ability to
process information. Thus, managers within an organization that faces too many
problems simultaneously may be acutely aware of the shortcomings in particular areas,
but be unable to devote sufficient attention to resolve these, due to demands in others. 567
A second key concept is that below the meta-system are four subsystems, the
precise details of which need not concern us in this paper, but which comprise critical
functions or operational subsystems and that absence or inadequacy on the part of
these can threaten system viability. Examples include subsystems to monitor the
environment, to make policy, and to integrate the activities of operational subsystems.
A third concept is that of diversity of goals. Most organizations serve multiple
stakeholders and must therefore of necessity pursue multiple goals. However, Beer
argues that if there are too many divergent views of what a system should do goals
may be incommensurate, leading to the problems of paralysis described above. Partly
this is a problem of attention – the “management unit” is forced to meter out its
attention to too many lines of activity. However, it is also a political issue, in that
building support for a particular strategy or direction is that much more difficult when
views as to the purpose and priorities of the organization are widely divergent.
With this background the questions and objectives of this paper can therefore be
summarized as follows:
.
How can the failure of Rover best be understood?
.
Through analysis of the failure of Rover can we identify patterns that may apply
to other examples of corporate failure?
.
To what extent can systems theory provide a useful framework to think about
failure in large complex organizations?

Research approach
The demise of the British Motor Industry has been the subject of several analyses
(Dunnett, 1980; Church, 1994; Whisler, 1999). Commonly based on secondary sources,
these contributions describe in detail the patterns of how BLMC/British Leyland/Rover
(the company was re-named a number of times during its life) and the British Motor
Industry in general, declined over time. However, most provide only limited explanations
as to why British Leyland/Rover entered such a spiral of decline. While this paper also
draws extensively on secondary sources (production data, annual reports, government
select committee meeting notes, official reports, previous published studies), our
assessment as to why BL/Rover failed is complemented by a total of 32 interviews with a
cross section of people involved with the company in a variety of roles. These include 19
senior executives, including several CEOs who ran the company during the period
1968-2005, nine executives from Honda, Rover’s partner for 15 years, a major figure in the
company’s retail and distribution network, a senior union official, and two former
government ministers responsible for industrial policy. The interviews were conducted
between October 2005 and August 2007, after the company had collapsed, and typically
lasted between one and three hours. They focused on the choices and events that, in the
view of the interviewee had proved critical to the fortunes of the company.
IJOPM The interviews followed a common format, with variations according to the position
28,6 of the particular interviewee and the period with which they were most familiar with
the company. Figures 2-4, which show Rover’s volume of output and market share
between 1970 and 2005 were produced from secondary sources before the interviews
commenced, and these were presented to the interviewees, who were asked to tell their
version of “the story behind the numbers”, with particular reference to the period in
568 which they were most closely associated with the company.
Interviews followed a common schedule. Each interview commenced by asking
interviewees to describe:
.
critical events – both internal and external to the company;
.
key opportunities that presented themselves to the company – possibly
opportunities that were not recognized as such at the time;

1,000,000
Nationalisation British BMW Phoenix
Aerospace
800,000
Output [units]

600,000

400,000

200,000
Figure 2.
Car production
at BL/Rover, 1970-2005
0
(excludes Jaguar, Land
1970 1975 1980 1985 1990 1995 2000 2005
Rover and New Mini)
Source: Company Accounts

50
Nationalisation British BMW Phoenix
Aerospace
40

30
%
20

10

Figure 3.
0
BLMC/Rover’s UK market
1968 1972 1976 1980 1984 1988 1992 1996 2000 2004
share, 1968-2005
Source: Company Accounts
 1,000,000 The death
British
Nationalisation Aerospace BMW Phoenix of a car company
800,000
Output [units]

600,000
Jaguar 569
400,000
New
Mini
200,000
Rover volume cars Land
Rover Figure 4.
Production of Rovers,
0 Land Rovers and New
1970 1975 1980 1985 1990 1995 2000 2005
Minis, 1970-2005
Source: Company Accounts

.
key choices, both correct and incorrect (and possibly only recognizable as such
with hindsight);
.
key market and competitor dynamics; and
.
conclusions about this period, including how the company compared to its
competitors during this time.

Interviewees were asked to do this period by period, the periods being:

.
Mid-1960s to 1975 (up to nationalization).
.
1975 to 1987 (Nationalization/Honda).
.
1987 to 1994 (Bae/Honda).
.
1994-2000 (BMW).
.
2000-2005 (Phoenix).
.
2005 – brief predictions for the future (e.g. Rover under Chinese ownership).

As the research covered a period of 40 years, no single interviewee had detailed

information about the whole period, but several overlapped in their times with the
company. Where this was the case, differences in perspective were surfaced and tested,
by feeding back the accounts of interviewees (unascribed) to others and asking them to
comment. Some of these differences in perspective were profound. An example of this
were labour relations and the role of organized labour in the company’s decline; those
speaking from a union perspective had a very different view of history from those
speaking from a management perspective.
Throughout this historical account, interviewees were also asked questions about
particular processes within the company and how these were enacted at different
times. These included perceptions of the company’s strengths and weaknesses in
different areas, including marketing, product development, manufacturing, suppliers,
distribution and sales and so on. All interviewees were also asked when, in their view,
the company passed the point of no return, a question which elicited an extraordinarily
wide range of responses. All interviews were recorded and professionally transcribed.
IJOPM The account presented in this paper thus represents “a best fit” of what is in fact a
28,6 complex and, in some areas, contested story.

The Rover story

Rover’s history in many ways reflects the classic pattern of industry proliferation and
consolidation described by Utterback (1994). As the UK auto industry matured, there
570 were closures and mergers, but in the post-WWII period the industry was still
relatively small in terms of its aggregate output, and fragmented in its ownership and
production organization. There were successive mergers and attempts at consolidation
from 1950 to 1970 as explicit efforts were made to build a British automotive company
capable of holding its own in the world. Ford in particular was regarded as the model
to emulate. Although there were a number of amalgamations during the first half of the
twentieth-century, this began in earnest with the merger of Austin and Morris in 1952
into the British Motor Corporation (BMC). Jaguar joined in 1966 and the company
became British Motor Holdings (BMH). In 1960, Standard-Triumph became part of
Leyland Motors, as did Rover, in 1967. These two groups, BMH and what was by then
the Leyland Motor Corporation, joined together to form the British Leyland Motor
Corporation (BLMC) in 1968. Thus, consolidation took place over a period of years but
the life of the consolidated firm was relatively short. By 1984, under the wing of a
conservative government committed to privatization, a process of disaggregation was
underway, with several operations either floated off as independent enterprises or sold
to other car companies (e.g. Jaguar, Leyland trucks, Unipart). The main steps in this
process of consolidation and disintegration are shown in Figure 5.
Thus, the Rover Car Company had not one, but many different starting points.
Rover itself was established, as a bicycle manufacturer, in 1878 with its first car
produced in 1904. Triumph, Austin and Morris were all established between 1885 and
1915. Austin and Morris joined forces in the 1950s and formed BMC, the idea being to
create a British car manufacturer to rival Ford. However, the Austin-Morris merger
was largely defensive and the animosity that existed between the two companies prior
to merger continued post-merger. This rivalry impeded efforts to build a single
company and lingered for many years. Despite this, BMC pioneered a number of
influential innovations, such as the Mini, launched in 1959. This was an Austin product
and was designed by Alec Issigonis, who had also designed the Morris Minor. The
Mini was the first car to combine front-wheel drive and a transversely mounted engine
using continuous velocity joints in the drivetrain. This breakthrough subsequently
became the industry standard in layout for small to medium-sized vehicles and the
Mini was to stay in production for over 40 years. Commercial success, however, was
more elusive; relatively high-costs and a low-selling price meant that the margins on
the Mini were wafer thin, and at some points in its life, negative.
For Rover, not yet part of BMC, the 1950s and 1960s were fruitful years, with the
success of Land Rover. Rover also conducted pioneering research into gas turbine
vehicles. In 1967, Rover became part of the Leyland Motor Company, marking the end
for Rover as an independent company.
When BLMC was formed in 1968 the combined company enjoyed production
volumes of close to one million units in the early 1970s, but struggled to integrate its
operations across the different constituent companies. Consolidation occurred, but the
integration upon which the success of the consolidation depended, did not.
 Daimler Motor
1910 Birmingham
Syndicate/
Company Small Arms Co. 1960:
(est. 1893/96) (est. 1861) Daimler part of Longbridge
BSA sold to 1979–1994: operations
Jaguar Collaboration with Honda; sold to
Jaguar Honda takes 20% Phoenix;
Austin Motor
(Swallow Sidecar of Rover equity, Rover takes a 20% renamed
Company
Company prior to stake in Honda Motor Europe. MG Rover
(est. 1905) 1959: Mini is
WWII; est. 1922) Five joint models are developed. Group.
introduced
Ceases
by BMC
Morris Motors operations
Company 1994: in 2005,
1952: 1966: 1975:
(Nuffield 1968: BMW assets are
British Motor British Motor Ryder Report, 1988:
Organisation) British 1981: buys 2000: sold to
Company (BMC) Holdings BLMC 1986: Rover
(est. 1913) Leyland BL Ltd. Rover BMW sells Chinese
formed formed nationalised BL Ltd Group plc
Motor renamed Group, individual firm NAIC.
Brands: and renamed renamed sold to
Wolseley, Corporation BL Public col- parts of
to British Rover British
MG (Morris 1960 Limited Limited laboration Rover Mini brand
1877: Standard Leyland Motor Leyland Ltd., Group plc Aerospace
Garages), (BLMC) Company with Group and Cowley
Motor Company Company in 1978 to (BAe)
Riley formed Honda plant
(est. 1877) (est. 1896) BL Ltd.
ends retained by
BMW
Triumph Motor
1945
Company
(est. 1885, first 1967 1986-87:
car in 1921) 1978: Land Rover Leyland and Unipart Land Rover
becomes own are sold off and its
1948: Land Rover
is introduced subsidiary Solihulll
Rover plant &
(est. 1877, first 1981: 1984: Gaydon
car in 1904) Alvis sold to Jaguar floated off R&D facility
United Scientific and bought by Ford sold to Ford
1965 Holdings in 1989

Alvis
(est. 1919)
of a car company
The death

Rover’s “family tree”

Figure 5.
571
IJOPM With worsening finances, BLMC was nationalised in 1975 and renamed British
28,6 Leyland. It received a £2 bn government cash injection (an equivalent of £11.4 bn in
2005 terms, calculated by GDP deflation) to modernise its plants and its products, but
matters continued to worsen. Michael Edwardes was brought in as CEO in 1977 to
steer the company forward (renaming it “BL Ltd”) and the company joined forces with
Honda in 1979, as it became clear that the company was incapable of developing
572 sufficient new models on its own. In 1979 Honda signed a collaboration agreement, and
granted Rover the right to produce one of its models, sold as the Triumph Acclaim.
Honda later agreed that Rover should produce other Honda designs for sale as Rovers
and produce some Honda vehicles in the UK on Honda’s behalf. Subsequently, the two
companies arranged a cross-shareholding with Honda taking a 20 per cent stake in
Rover, and Rover taking a 20 per cent stake in Honda’s European manufacturing
operation.
In 1982, there was another name-change; British Leyland became Austin Rover.
The Rover badge was used on a range of cars co-developed with Honda: the first
Honda-sourced model to carry the Rover badge, released in 1984, was the Rover 200,
which, like the Triumph Acclaim that it replaced, was based on the Honda Ballade.
In 1986, the Rover SD1 was replaced by the Rover 800, which was based on the
Honda Legend. That year Graham Day became head of the company, which in 1987
was renamed “the Rover Group” and adopted a one-brand strategy. The Austin
Maestro and Montego, by then badged as Rovers, were replaced by the Rover 400 and
Rover 600 which were based on Honda’s Civic and Accord platforms.
In 1988 the company was sold to British Aerospace (BAe). It remained under BAe’s
ownership for six years, until in 1994 BAe put Rover up for sale as part of a move to
focus effort on their core aerospace business; Rover was sold to BMW. Shortly before
the deal was closed, strenuous attempts were made to convince Honda to increase its
stake from 20 per cent and take over Rover, but Honda refused to do this. Even on the
morning when the deal with BMW was due to be closed, Rover executives were in
Tokyo trying to persuade Honda to take a majority stake in Rover, but Honda would
not increase its share to more than 47.5 per cent. This was not sufficient for BAe, and so
Rover passed to BMW for £800 m. BMW invested considerably in Rover, in particular
the development of the Rover 75 and the New Mini.
Six years on, in May 2000, after two consecutive years of heavy losses and failure to
secure sufficient government subsidies for the replacement for the mid-range 25 and 45
models, BMW broke up the business and Rover was sold. One of the would-be buyers
was a group of venture capitalists, Alchemy partners, but there was widespread
resistance to Alchemy in favour of the Phoenix Group, headed by former Rover CEO
John Towers, who had left the company after the takeover by BMW. Alchemy had
proposed to convert Rover into a low-volume sports car company, focussing on
MG-branded sports cars, a concept initially favoured by the government, but
abandoned after workers’ protests in London. Subsequently, the Phoenix offer was
supported. Phoenix bought Rover for £10 and pledged to keep all employees in work,
aiming to return a profit within two years. Renamed “MG Rover”, the company
received a 49-year loan of £470 m from BMW and a licence to use the Rover brand.
At the same time, BMW sold Land Rover to Ford for £1.8 bn, including the Gaydon
R&D facility and the Solihull plant. BMW retained the New Mini, launched in 2001,
and the facilities to build it at Cowley, near Oxford. As a result, Rover 75 production
was moved to Longbridge. Rover now produced the 25, 45, and 75 at Longbridge, The death
but was now deprived of two of its key brands (Mini and Land Rover), and having lost of a car company
most of its R&D personnel due to the sale of Gaydon.
From 2000 onwards the decline of MG Rover continued. In 2001, eight month
operating losses of £254 m were reported. Amidst the acquisition of the Italian
sportscar maker Qvale (which led to the creation of the SV sportscar, of which a total of
25 were sold by April 2005), the company announced an alliance with the Chinese 573
group China Brilliance, to help fund investment in new models. However, this deal was
not completed, despite an initial cash injection by the Chinese company. In 2003 there
were losses of £77 m and production output fell further. In an effort to raise cash, the
Longbridge site was sold for £45 m, and leased back. In November 2004 a plan for a
£1 bn joint venture with the Shanghai Automotive Industrial Corporation (SAIC) was
announced, and the rights for the 25, 75 models and the K-series engines were sold to
SAIC (another Chinese automotive firm) for £67 m.
By March 2005 sales had continued to fall, and some suppliers were demanding
cash payment upon delivery of components. A rescue mission to save the SAIC deal in
April 2005 failed, although a bridging loan of £100 m by the British Government was
offered, and some suppliers stopped their deliveries of components. On 15 April 2005,
all 5,100 Longbridge workers faced redundancy after production was halted there. A
century of car production at Longbridge had ended. In July 2005 Rover’s remaining
assets were sold to Nanjing Automotive, who dismantled the majority of production
assets and shipped them to China. Since then, “token” low-volume production of MGF
sports cars at Longbridge has resumed. In September 2006 there was another twist;
Ford exercised its right to buy the Rover trademark under the terms of its purchase of
Land Rover in 2000, largely as a defensive measure to protect the Land Rover brand.
Nanjing responded by creating the “Roewe” brand (“Rong Wei” or “Grand Prestige” in
Chinese), with a logo strongly resembling the old Rover badge.
In June 2007, Ford announced its intention to sell both Land Rover and Jaguar, with
Tata emerging as a likely buyer. Shanghai and Nanjing automotive corporations
announced a government-supported merger in early 2008, reuniting the IPR and
production assets needed to build the previous Rover models and engines in China.

Analysis
Having sketched out the story of Rover, in this section, we examine each period in the
company’s history in more detail. Of necessity, our treatment is selective, in that it is
clearly difficult to cover all the details of a period of nearly 40 years in a short paper.
We highlight those issues that emerged repeatedly across multiple interviewees as
significant in the company’s decline. The declining output of the company between
1968 and 2005 is shown in Figure 2. We shall refer to this repeatedly as we analyse
events at Rover.

1968-1975 – the first years of the conglomerate

As we have seen, in 1968, the company, then named BLMC, was formed out of a
disparate set of previously independent companies, some of them themselves formed
from previous amalgamations. This paper deals only with the car operations, but for
the first 15 years of its existence the conglomerate produced much more than just cars.
Its output included trucks, buses, construction equipment, commercial refrigeration
IJOPM equipment and a miscellany of other products. This wide range of products greatly
28,6 complicated the overall management task and reduced the focus – and resources –
available for the development of the car business, consistent with Beer’s observations
on requisite variety. Indeed, until 1975 there was no single unit in control of all car
operations. There was a history of strong concept engineering, which generally
stemmed from exceptional single designers, such as Alec Issigonis in the case of the
574 Mini and 1,100/1,300 series, and Spen King in the case of the Range Rover. One of the
product development executives who we interviewee reported how Issigonis resisted
evolutionary changes to his original design, even though market analysis indicated
that these would increase the vehicles’ appeal. Another interviewee, a member of the
board during the 1970s and 1980s, reported the same pattern with respect to King and
the Range Rover, which was not produced in a four-door version until many years after
its launch. A side effect of such behaviour was an emphasis on what might be termed
“intuitive engineering” and a consequent lack of process and discipline. This had at
least three effects. First, with increasing vehicle complexity and sophistication,
intuitive approaches to engineering were progressively less capable of delivering the
consistency, thoroughness and follow-through necessary to produce well-engineered
vehicles.
Secondly, the presence of a small number of outstanding engineers generated and
perpetuated a myth that engineering competence was actually higher than in fact it
was – previous strength in what was essentially concept engineering was
over-generalized to other aspects of the company’s operations. This is quite a subtle
process, and hinged on the existence of high-profile examples that run counter to a
general trend, therefore allowing participants to believe that criticism of vehicle design
and quality was less warranted than it actually was. In Rover’s case this perpetuated a
belief than things were much better than they actually were in terms of basic standards
of engineering and manufacturing. Third, “hero-engineers” felt a strong sense of
ownership of their concepts and were resistant to subsequent changes to the products
that were the embodiment of these, even if this was clearly what the market was
demanding. For example, Issigonis resisted the replacement of his original sliding
windows with wind-down windows on the Mini, and King blocked the addition of a
four-door model to the Range Rover series (his original had only two doors) for many
years. As a product development executive explained:
The engineer was held as a bit of an idol, not only in the company, but outside as well. It was
really critical, so if anybody challenged their territory it was a matter of life and death,
almost. A lot of the energy of the company was actually involved in this internal positioning,
rather than saying, “the Japanese are producing some interesting things, Fiat have got some
stuff, VW are producing the Golf”. These things were going on outside, but a lot of the energy
was about how to hold on to your territory, and create things in your part of the company.
Many car companies have grown through a process of relatively strong, large
companies acquiring relatively weaker, smaller ones. There have, of course, also been
examples of mergers of pairs of relatively equal companies such as Renault and Nissan
and Chrysler and Mercedes-Benz, but these are relatively unusual. In the case of BLMC,
what effectively occurred was the simultaneous amalgamation of many previously
independent companies, some with niche strengths (such as Rover, Jaguar) but
combined with the volume business of Austin-Morris which was described by several
interviewees as “a mess”. As one of the ex-Rover CEOs put it:
 When I joined in ’69 I joined a sprawling, fragmented, disparate entity called BLMC, which The death
had just been formed. Looking back, Donald Stokes [the first CEO] had been handed an
impossible restructuring task [. . .] It was the mother of all integration projects. of a car company
When companies grow organically, they gradually develop the capabilities necessary to
manage enormous scale and complexity – what we earlier termed the “management unit”
– techniques of financial control, costing, budgeting, marketing and portfolio
management commensurate with their scale. In Beer’s terms, this ensures that the 575
sophistication of the coordinating mechanisms is commensurate with the variety of
operations that have to be coordinated. Yet BLMC was essentially a collection of relatively
small owner-managed enterprises (in ethos, if not necessarily legally so), each of which
individually lacked these capabilities. Moreover, some of these were themselves in a very
weak condition, so it was not that they could be left to their own devices with any great
prospect of prospering. With hindsight, combining them made the problems worse, not
better. John Egan, who was later to become CEO of Jaguar described it thus:
You had all of these currents running through the BL story of how they didn’t know how to
design new products, poor industrial relations, a poor rationalisation program, really quite
anarchic in terms of managerial process. You had government programs going on like
regionalisation and stop-go, which didn’t give them much of a chance either, and a relatively
ill-educated management team. You add it all up and the chance that possibly they had was
frittered away between 1970 and 1974. There was no coherence in anything.
There were attempts to redress these shortfalls in the management system, one
manifestation of which was significant recruitment from Ford, particularly in the areas
of finance and product planning. However, in the early days the central staff who were
tasked with doing this were very few in number – an estimated 20 people, which for a
very diverse organization of nearly 200,000 was extremely small.
Even before the creation of BLMC, Austin-Morris had been generating insufficient
funds for model renewal, and the amalgamation created another layer of problems and
complexity with which the corporation was ill-equipped to deal. Whilst struggling with
its post-merger issues, in the early 1970s BLMC was hit by two forces over which it had
no control – the removal of tariffs on vehicle imports into the UK with Britain’s entry
into the European Union, and the first oil shock. These, combined with the company’s
weaknesses in design and production quality, produced a catastrophic fall in output,
from around 900,000 units a year in 1971 to just 600,000 in 1976. The oil shock hit all
car producers of course, and output dipped across the industry. However, for most
producers output recovered, but for BLMC it did not, apart from a brief uplift in 1976,
as shown in Figure 2.

1975-1987 – state ownership

In 1975, nationalization brought a substantial injection of funds, and the slogan was
“product-led recovery”. However, although the injection of resources could have
addressed the shortage of funds to develop new products, the underlying capability to
do so was still seriously lacking. In the two years following nationalization little or no
discernible improvement occurred and Michael Edwardes was brought in as CEO in
1977. He took improvement of industrial relations as his major project – poor
industrial relations had been a major factor in the problems that the company was
experiencing. As Edwardes described:
IJOPM What one had to do at that point was slim down the company to its competitive position,
which meant a lot of people going, and then you’d got to slim down to the point when there
28,6 wasn’t the volume, and so it went on [. . .] There had been enormous industrial relations
problems and I think that I was doing a mopping-up job when I came in.
For 10 years there were no product profit and loss accounts. It took us into ’78, I went in in
November, we got our first stab at product profitability in the first half of ’78. I couldn’t
576 believe it. For 10 years they were flying blind.
This further illustrates the difficulty with the management system of the enterprise,
but the emphasis on industrial relations, however essential, produced another side
effect, which Edwardes freely acknowledges, and that was a diversion of attention
away from other pressing matters:
We got distracted. Where the Germans were spending 80 per cent of their time on product,
I was spending 80 per cent of my time on industrial relations, so that was a total waste.
By the end of the Edwardes period in 1982, output stabilized at around 400-500,000
units per annum, and industrial relations had improved, due to Edwardes’ efforts to
break the power of the more militant trade unions on the shopfloor. The company had
been rationalized considerably, had been renamed “BL”, there was now an integrated
model programme, and there was considerably more coherence than there had been ten
years before. The company was developing a new range of models in the the Metro,
Maestro and Montego, but sales of current models were falling, and BLMC’s new
executive car (the Rover SD1, launched in 1977) on which much depended, was beset
with quality problems and failed to meet its sales targets. This failure followed the
failures of the Allegro (a small to mid-sized car) and the TR7 (a two-seater sports car).
Combined with product inadequacies, these failures reduced sales volumes to levels
that made independent viability impossible, as the company lost the resources and
capacity to renew itself through new product development.
Recognizing the product development shortfalls in the company, and the acute need
for a model to plug the mid-range gap until the Maestro came on stream, BL sought a
partner who could provide it with a new model. An alliance was struck with Honda of
Japan in 1979. At the time BL was actually the larger car company, a sign of the stark
contrast between Honda’s growth and BL’s decline since that time. Honda was a
partner to Rover for 15 years, and provided Rover with a number of successful new
models throughout the 1980s and early 1990s, but the partnership ended abruptly
when Rover’s owner, BAe, sold Rover to BMW in 1994. It is clear that BL/Rover learnt
a great deal from Honda, although there were obvious tensions as the different
standards of the two companies were reconciled. As one of our Honda interviewees
commented:
We found that on average each Rover-produced car had well over 100 faults. Many of these
faults were very small, but Honda standards were high and our attitude was that in order to
maintain our standards, every fault however small, had to be dealt with [. . .] Of course we
reported to Rover what we had found. Their response was one of disbelief. They said that, in
general, their customers were finding few faults in the Rover versions of the cars in question
and they could not accept the Honda results. There was a lot of discussion between the two
companies and we explained the Honda attitude and tried to demonstrate the faults.
The same interviewee also commented that there was, at best, ambivalence on the part The death
of some very senior people at Rover about the Honda relationship, and a continuing of a car company
reluctance to accept and act upon Honda’s advice:
I was aware that there wasn’t wholehearted support, among quite a lot of senior Rover people,
for the relationship. I think that the position broadly was that people didn’t like the
relationship very much; they didn’t necessarily want to work very closely with Honda, but
they did see – a lot of them saw – that it was necessary. 577
The consequence of these different approaches was graphically illustrated by the
reception of the Honda Legend and Rover 800 (Sterling) in the USA. This was a joint
development between the two firms, and was the model that was supposed to re-launch
Rover in the US market. The underlying design of the two derivatives was essentially
the same, yet the Honda version was virtually at the top of the JD Power quality index
when the Rover version was almost at the bottom of the same ranking. Analysis of the
failures revealed that it was largely the Rover-designed or Rover-modified parts that
were the source of these problems.

1987-1994 privatisation and BAe

When the company was nationalized in 1975, a labour government, sympathetic to
state ownership, was in power. From 1979 onwards there were a succession of
conservative governments, led by Margaret Thatcher, who were deeply opposed to the
idea of state ownership, and whose agenda was to return the company to private
ownership – or close it down – as rapidly as possible. The right hand side of Figure 5
shows this process in action, with those parts of the business that were considered
saleable sold off between 1984 and 1987. Jaguar was privatized in 1984 and then
bought by Ford in 1989, Unipart and Leyland Trucks were sold in 1985-1986 and then
finally the cars group, by then including Land Rover, was sold to BAe in 1987. In the
same way that Edwardes described how industrial relations issues were a distraction
in the 1970s, interviewees from this period described how the priorities when preparing
a business to be sold for sale are quite different to those when one is planning to grow
and develop the business in the long-term. Moreover, despite the Thatcher
Government’s commitment to market forces, the matter of privatization was not
purely a commercial one. The near sale of the commercial vehicle operations, which at
that time included Land Rover, to General Motors in 1986 foundered on a political
decision that Land Rover should remain under British ownership.
Despite some rhetoric at the time about synergies between BAe’s aerospace
operation and Rover’s car operations, by all accounts BAe was “a hands-off” owner.
Under the leadership of Graham Day, serious attention was given to image and
branding, and in what was know as the “Roverization” period all non 4 £ 4 models
were badged as Rover, the MG brand being revived in the mid-1990s with the MGF
sports car. The aim was to position Rover as an upmarket, premium brand.
However, as BAe came under increasing financial pressure their continued
ownership of Rover was called into question. A Board member from this period
described the situation as follows:
[BAe] ran out of money, and went to the shareholders to raise more money. The offer to the
shareholders was underwritten by banks, and the shareholders declined to give BAe any
more money, so the banks were stuffed and had to put a lot of money into BAe. And that’s
IJOPM why BAe subsequently didn’t feel it could support the Rover cash flows, and looked for an
owner for Rover. [. . .] The Chairman, and the Chief Executive, and subsequently the Finance
28,6 Director of BAe all left as a result of that failed rights issue, and the new blood who came into
BAe looked around and said, “Rover’s got to go”.
The decision by BAe to sell to BMW was controversial, as many observers felt that
Rover’s interests would be better served by a continuing relationship with Honda.
578 Honda interviewees reported that at the time of the sale, Honda felt that Rover had
made progress, that Rover’s standards of manufacturing were rising, and might reach
Honda’s expectations within two years. However, Rover was very dependent on Honda
for their product engineering by this stage, a factor that BMW seem to have
underestimated.
During the Honda period BL/Rover’s output had remained fairly stable although now
below half a million units a year – too low a volume to be sustainable in the volume
market where margin per unit is relatively low. However, this stability of output
occurred against a backdrop of a rising market, and domestic market share actually
slipped from around 20 to 15 per cent during BAe’s ownership, as Figure 3 shows.

1994-2000 BMW
Perhaps, the single biggest mystery in the Rover story is why things went so wrong
during the BMW period. When BMW bought Rover, Rover was not in a particularly
strong position, but losses had been stemmed, some small profits had been returned
and there was a reasonable range of models, though the small car (the Metro) was
desperately overdue for replacement. Under BAe’s ownership, in order to save money,
it had been decided to try to combine coverage of the Rover 200 category with the
Metro category in the form of the 1995 Rover 200, larger than a Metro, but smaller than
the previous 200 series, a strategy which turned out to be misguided.
BMW appeared to over-estimate Rover’s basic capabilities when they first took
them over. Moreover, they also seem to have thought that the Rover-Honda
relationship would continue after BMW’s purchase of Rover, something which Honda
was not prepared to do. As a Honda Board member from the period explained:
[. . .] a lot of people have asked why Honda was willing to work with Rover, but not with
BMW. The official position was that we were willing to help a British company with the aim
of that company becoming a self-sustaining, independent entity. We had no desire to own the
company. BMW was a German company with no need of help. The circumstances were
different. In reality, we did not want to work with BMW. It was a big, successful company
and we did not see where getting involved would lead.
Thus, although Honda continued to allow Rover to use Honda designs for a fee
(something which they were legally under no obligation to do) the relationship and the
associated support effectively ended at the point of sale. BMW did not immediately
step into replace this. As a senior Rover product planner described to us:
[BMW] were extremely professional, to the extent that they’d all had their British cultural
training, and they understood that we liked to start all meetings with a joke, so we would get
a standard joke at the beginning of every meeting, which was embarrassing to the extreme
[. . .] but the real reason is, I don’t think they could countenance coming into a company,
and coming in with jackboots. They would be so unpopular, and culturally it would be bad
karma, whatever, for them. They couldn’t do it, so they let us carry on, trying to influence us,
almost like they would only come in if we really failed. It was a disaster. They should have
 come in, they should have come in and really shaken us up, really been critical, but they The death
didn’t. They’d left the management as was, left the plan as was [. . .]
of a car company
Thus, if the problem in the past had been high-variety coupled with the lack of a
capable management unit, the situation was now of a much slimmed-down set of
operations, but an owner that either did not understand the operations of the company
it had bought, or which understood these, but perceived social and cultural obstacles to
intervention. 579
Differences within BMW about Rover eventually led to the resignations of both
Pischetsreider, the CEO of BMW and a supporter of BMW’s relationship with Rover
and Reitzler, the number two, who was opposed to it. BMW began to look for ways to
rid itself of Rover, and entered in negotiations with Alchemy, an equity capital firm
specializing in turnarounds. This revealed that Rover’s management, such a source of
weakness for so many years, still posed major problems. As Jon Moulton, Alchemy’s
CEO described to us:
By the time we [Alchemy] got to it with BMW, the patient was terribly ill, the volumes were far
too low, but still the financial control was as bad as it had ever been. BMW were unable to tell
us, during negotiations, anything about the management accounts of Rover. We didn’t believe
them, we thought they were just hiding it. The only figures they could give us were the amount
of money they were putting in each month, they gave us some numbers on the numbers of cars
being sold by Rover; those numbers, unfortunately – as reported to the BMW senior
management – were not actually the numbers of cars they were selling, they were the numbers
on the spreadsheet out of the five year plan. And that’s the honest truth. So the senior
management of BMW didn’t even know what volume of cars Rover were producing.
Alchemy’s bid to buy Rover was turned down in favour of the Phoenix Consortium
headed by John Towers, former CEO of Rover. Land Rover was sold to Ford for £1.8 bn,
while BMW retained the Cowley plant to produce the New Mini, which became part of
the BMW Group as an independent brand.

2000-2005 Phoenix
When Phoenix bought Rover in 2000 it was already obvious to many industry experts
that the company could not survive, unless a joint venture partner could soon be found.
In terms of our capabilities framework, Rover was no longer a fully capable car
company, and in fact had not been for some time. In retrospect, one could criticize the
Phoenix directors for portraying the illusion that MG Rover was viable enough to
survive, but given their need to secure a partner, this was perhaps the only line that
they could take. By 2000 the merger wave that the auto industry had seen throughout
the 1990s had subsided, and it became clear that many large-scale mergers such as the
one between Daimler and Chrysler were not yielding the hoped-for benefits. In May
2007 the DaimlerChrysler merger was largely unwound, with the sale of 80 per cent of
Chrysler to Cerberus Capital Management.
In addition, overcapacity had become a key concern in the industry. As a
consequence, there was no rationale for any of the established players to buy Rover,
who by this time offered little other than some aging models and assembly capacity of
average quality in a region with little prospect for growth in sales. The only possibility
for Rover was to seek partners in markets that showed real growth, such as China.
Chinese manufacturers, while plentiful in number, are short on technology, and all
foreign manufacturers that established operations in China have been compelled to set
IJOPM up joint ventures with Chinese companies. Chinese and foreign manufacturers alike are
28,6 mainly concerned with establishing their position in the Chinese domestic market, so
the interest in Rover by Brilliance in 2002 and SAIC in 2004 was geared towards
getting access to Rover’s technology (which was still marketable in China and other
developing regions), and possibly its brands.
The Alchemy vs Phoenix question demonstrates a tension that is repeatedly visible
580 in the history of Rover, and which illustrates the significance of Beer’s observation on
how the divergence of goals impacts on system viability. Until the reforms in the late
1970s, the union agenda of protecting employment (in the short-term) impeded reforms
and factory closures that may otherwise have occurred on the basis of purely
commercial criteria. It is arguable whether creating a successful car company was ever
particularly high on BAe’s agenda, given their focus on short-term financial issues. The
sale to Phoenix promised fewer job losses in the short-term, but it was never clear how
the company could attain viability, and there was deep hostility to Alchemy’s more
radical proposal to create a significantly smaller, but possibly longer-lived, business.
As Batchelor (2001) points out, the union’s position over redundancy payment
liabilities made a deal with Alchemy very difficult. The Alchemy bid, which would have
meant downsizing the company to a small-scale niche sports car maker under the MG
brand, would nonetheless have been a much more sustainable option. On the other hand
this would have meant immediate redundancies at Longbridge of at least 4,500 workers,
most likely even more. Under Phoenix, the 9,060 Longbridge workers of the total 32,070
Rover employees all kept their jobs in the first instance (Rover Task Force Report, 2000),
although under BMW the workforce had already reduced by 7,000 workers, most of
whom took voluntary redundancy (Batchelor, 2001), and again under Phoenix the Rover
workforce at Longbridge reduced to 5,100 by 2005. Hence, Phoenix cushioned
redundancies from a projected 9,060 in 2000 to 5,100 in 2005. The latter should also be
seen in perspective, for as we have seen, the Rover Group was in fact gradually split up
from 1984 onwards, and many of these businesses (Jaguar, Land Rover, Mini, Unipart)
are still in operation, but under the protective wing of foreign capital. Figure 4 shows that
when the output of Jaguar (Ford), Land Rover (Ford) and Mini (BMW) is taken into
account, output is about the same as it has been since the late 1970s, at around 400,000
units per annum. These operations are not without their problems, particularly Jaguar,
but the fact that they have all continued to exist, and indeed to grow, under alternative
owners, supports the idea that the new owners brought capabilities that Rover itself
lacked. Even so, the picture has not been one of unqualified success, with Ford putting
Jaguar and Land Rover up for sale in June 2007, as part of a restructuring plan.

Conclusions and implications

The data presented in this paper illustrate how Rover’s collapse was a culmination of a
process that started more than four decades previously: a series of failed attempts at
consolidation and rationalisation, a persistent inability to develop products that hit
the right markets, and missed opportunities with different partners all led to a situation in
which the company eventually consumed its own capital in order to stem its operating
losses.
At the outset, this paper set out to explore three questions. The first question
concerned how the failure of the company could be understood, and in this respect the
company’s history carries a number of lessons. First, the company was clearly much
more than just a commercial enterprise, and had, at various times in its life, to reconcile The death
the wishes of multiple stakeholders with very different agendas, something which Beer of a car company
has identified as a threat to system viability. From a commercial point of view, in the
1970s reform was delayed because of the politically unpalatable consequences of job
losses. In the 1980s, the privatization agenda led to some decisions that did not
necessarily make sense from a commercial point of view, including the vetoing of the
potential sale of the company to non-British buyers. Even in 2000, considerations of 581
employment protection seem to have outweighed commercial ones.
This lack of focus was aggravated by what might be termed “the missing
development step”. The formation of BLMC, which later became Leyland and then
Rover, created an enormous conglomerate from a set of previously relatively small and
independent firms. Some of these were in weak and precarious positions well before the
merger. In Beer’s terms, the conglomerate lacked a management unit of the scale and
sophistication to deal with the combination of overall complexity and weaknesses in
the operating units. Hence, the situation became worse, not better, with the relatively
strong firms being pulled down by the weaker ones.
That a large, complex and difficult-to-manage organization was created without an
adequate management unit was one part of the problem. However, in the case of Rover,
several other factors hit at the same time, putting additional load on the already
over-pressed centre. The company was running out of cash at the time of the merger, and
was not generating sufficient funds to resource its own product development; the UK’s
entry into the common market exposed the firm to the full force of foreign competition at a
time when it was weak; labour relations issues preoccupied Michael Edwardes who was
perhaps the strongest of the firm’s CEOs; and so on. Following Ashby (1956), Beer (1984)
argues that “the management unit” must have the capacity to process the inputs that it
receives from its own operating units as well as other sources; selective attention is the
inevitable consequence of a mismatch in capacity and load. In this paper, we have followed
Beer in terming this key capability the “management unit”, but this is really a shorthand
term for a whole series of capabilities – in financial control, portfolio management,
marketing and so on – that Rover lacked. A capable management system also delivers
sufficient consistency over time to permit consistent brand values, something that Rover
struggled to do.
In our view, this high-level integration is a generally neglected area by operations
management as a discipline – operations management researchers often focus on very
specific processes or subsystems such as manufacturing, product development, supply
chain management or technology acquisition and tends to overlook the governance
systems that connect all of these subsystems together. This may be because the traditional
business disciplines tend to carve up the world in ways that allows detailed, nuanced
dissection of particular functions or processes (“subsystems”, in the parlance of Beer) to
the neglect of issues around how the integration of these subsystems occurs. Operations
management researchers clearly do at times consider issues that span functional areas
(e.g. the relationship between manufacturing and product development; the role of
suppliers in product development) but holistic analyses of how all the pieces fit together –
or fail to – are unusual.
The second question, to what extent do the patterns we find in the Rover case apply
to other examples of failure, is more difficult to answer. The Rover case is unusual, in
that such a large enterprise was created so quickly from so many weak units.
IJOPM However, we would argue that although it is an extreme case, it demonstrates a
28,6 number of principles that apply across many situations. These apply particularly to
organizations or other systems that are in crisis, where there are failings in several
organizational subsystems simultaneously, and where undue load is placed on the
management unit. One interpretation of the Rover case is that there was a race between
developing the management unit and simplifying the company’s operations on the one
582 hand and the rate of decline on the other. In the end, it was the rate of decline that
triumphed. The same processes may also be seen in turnarounds, mergers and
acquisitions, and other situations in which companies face major shocks.
Finally, the Rover case demonstrates the usefulness of systems ideas to understanding
at least some types of failure, not as an alterative to capability-based approaches, but in
addition to them. Despite being perceived in public as a fully capable car company and
commended by commentators on its manufacturing improvements as recently as 1994
(Pilkington, 1996), in many ways Rover lost the ability to be a fully capable car company
well before that. Successive other capabilities were lost over the years, but right up to the
end, because the vehicles were still “rolling off the line”, many still thought of the company
as a “complete” car company. If anything, our study shows that judging a firm’s
competitiveness by defining it as a set of capabilities at the level of subsystems is not
sufficient. Analysis of the “management units” that coordinate these subsystems can, we
argue, offer considerable insight into the reasons behind corporate success and failure.

References
Aaker, D. (1991), Managing Brand Equity: Capitalizing on the Value of a Brand Name,
The Free Press, New York, NY.
Ashby, W.R. (1956), An Introduction to Cybernetics, Chapman & Hall, London.
Batchelor, J. (2001), “Employment security in the aftermath of the break-up of the Rover group”,
Warwick Business School Working Paper Series, No 342.
Bayley, S. (1986), Marketing Vorsprung Durch Technik. S. Bayley, Sex, Drink and Fast Cars:
The Creation and Consumption of Images, Chapter 7, Faber and Faber Limited, London,
pp. 87-112.
Beer, S. (1972), The Brain of the Firm – The Managerial Cybernetics of Organization, Wiley,
Chichester.
Beer, S. (1979), The Heart of the Enterprise, Wiley, New York, NY.
Beer, S. (1984), “The viable system model: its provenance, development, methodology and
pathology”, The Journal of the Operational Research Society, Vol. 35 No. 1, pp. 7-25.
Church, R. (1994), The Rise and Decline of the British Motor Industry, Cambridge University
Press, Cambridge, MA.
Clark, K. and Fujimoto, T. (1991), Product Development Performance: Strategy, Organisation and
Management in the World Auto Industry, Harvard Business School Press, Boston, MA.
Cusumano, M.A. and Takeishi, A. (1991), “Supplier relations and management: a survey of
Japanese, Japanese-transplant, and US auto plants”, Strategic Management Journal, Vol. 12
No. 8, pp. 563-88.
Delbridge, R. and Oliver, N. (1991), “Just-in-time or just the same? Developments in the auto
industry: the retailers views”, International Journal of Retailing and Distribution
Management, Vol. 19 No. 2, pp. 20-6.
DETR (1998), Efficient JIT Supply Chain Management: Nissan Motor Manufacturing (UK) Ltd, The death
Good Practice Case Study 374, Department of the Environment, Transport and the
Regions, London. of a car company
Dunnett, P. (1980), The Decline of the British Motor Industry: The Effects of Government Policy,
1945-1979, Croom Helm, London.
Ford Motor Company (various years), Annual Reports, Ford Motor Company, Dearborn, MI.
Galbraith, J.R. (1974), “Organizational design: an information processing view”, Interfaces, Vol. 4 583
No. 3, pp. 28-36.
General Motors Corporation (various years), Annual Reports, General Motors Corporation,
Detroit, MI.
Heller, D.A., Mercer, G. and Fujimoto, T. (2006), “The long term value of M&A activity that
enhances learning organizations”, International Journal of Automotive Technology and
Management, Vol. 6 No. 2, pp. 157-76.
Hines, P. (1998), “Benchmarking Toyota’s supply chain: Japan vs UK”, Long Range Planning,
Vol. 31 No. 6, pp. 911-8.
Holweg, M. and Pil, F.K. (2004), The Second Century: Reconnecting Customer and Value Chain
through Build-to-Order, MIT Press, Cambridge, MA.
Jackson, M.C. (2001), “Critical systems thinking and practice”, European Journal of Operational
Research, Vol. 128, pp. 233-44.
Kiff, J. (1997), “Supply and stocking systems in the UK car market”, International Journal of
Physical Distribution & Logistics Management, Vol. 27 Nos 3/4, pp. 226-43.
Lamming, R. (1993), Beyond Partnership: Strategies for Innovation and Lean Supply,
Prentice-Hall, Englewood Cliffs, NJ.
Oliver, N. and Delbridge, R. (1991), “Beyond customer satisfaction: the changing face of car
retailing”, International Journal of Retailing and Distribution Management, Vol. 19 No. 3,
pp. 29-39.
Pilkington, A. (1996), Transforming Rover: Renewal Against the Odds 1981-1994,
Bristol Academic Press, Bristol.
Reichhart, A. and Holweg, M. (2007), “Lean distribution: concepts, challenges, conflicts”,
International Journal of Production Research, Vol. 45 No. 16, pp. 3699-722.
Sako, M. (1992), Prices, Quality and Trust, Cambridge University Press, Cambridge, MA.
Saul, S.B. (1962), “The motor industry in Britain to 1914”, Business History, Vol. 5 No. 1, pp. 22-44.
Schonberger, R. (1982), Japanese Manufacturing Techniques, The Free Press, New York, NY.
Schonberger, R. (1986), World Class Manufacturing, The Free Press, New York, NY.
Utterback, J. (1994), Mastering the Dynamics of Innovation, Harvard University Press, Boston, MA.
Whisler, T.R. (1999), The British Motor Industry, 1945-94: A Case Study in Industrial Decline,
Oxford University Press, Oxford.
Womack, J.P., Jones, D.T. and Roos, D. (1990), The Machine that Changed the World:
The Triumph of Lean Production, Rawson Macmillan, New York, NY.

Corresponding author
Nick Oliver can be contacted at: nick.oliver@ed.ac.uk

To purchase reprints of this article please e-mail: reprints@emeraldinsight.com

Or visit our web site for further details: www.emeraldinsight.com/reprints
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.