HWK 3 – Eng. Management

Harvard Business School 9-795-011
August 18, 1994

Research Associate Greg Keller prepared this case under the supervision of Assistant Professor Anita M. McGahan and in
consultation with Visiting Professor Arnoldo Hax as the basis for class discussion rather than to illustrate either effective or
ineffective handling of an administrative situation. The costs associated with each of the investment alternatives described
in this case have been estimated by the casewriters.

Copyright © 1994 by the President and Fellows of Harvard College. To copies or request permission to
reproduce materials, call 1-800-545-7685, write Harvard Business School Publishing, Boston, MA 02163, or go to
http://www.hbsp.harvard.edu. No part of this publication may be reproduced, stored in a retrieval system,
used in a spreadsheet, or transmitted in any form or by any means—electronic, mechanical, photocopying,
recording, or otherwise—without the permission of Harvard Business School.

1

Saturn Corporation’s Module II Decision

In the spring of 1994, Saturn Corporation was setting sales records by attracting more than
25,000 ers per month. These results surprised many observers who had predicted an end to the
company’s growth after a slow first quarter. Since starting production of small cars in 1990, the
newest division of General Motors (GM) had increased sales every year, putting 239,000 of its models
on the road in 1993. Saturn officials believed there was a long-term opportunity to sell 400,000 to
500,000 cars per year in the U.S. and selected international markets.

To reach these sales levels, however, the company needed to convince GM to finance
additional production capacity. Since 1985 GM had spent $3.5 billion to build a dedicated factory and
establish the small car brand. Although Saturn earned its first operating profit in 1993, several
analysts questioned whether GM could earn an adequate return on its investment. Saturn president
Richard “Skip” LeFauve argued that the best way to build profits was by committing to opportunities
that leveraged the company’s fixed costs of manufacturing and marketing.

LeFauve and other Saturn managers had been reviewing options for a second assembly plant
(known as “Module II”) with GM since the beginning of the year. One possibility was to expand
capacity at Saturn’s existing production facilities in Spring Hill, Tennessee. The plant had been
constructed between 1986-1990 after GM’s management endorsed a “clean sheet” approach for the
new nameplate. A second set of options involved refitting one of several GM plants that had been
mothballed or was scheduled to close shortly. This approach required a lower fixed investment, but
would create additional operating challenges as well as added operating costs.

Saturn’s Achievements

By 1994, Saturn had achieved “cult car” status. Most ers were aware that the company
had been founded to make GM competitive with Japanese imports in the small car market.
Advertisements explained how manufacturing and retailing practices had been restructured to
improve the overall ownership experience. Saturn’s union workers exchanged hourly wages for
salaries and bonuses tied to product quality and company profitability. A “no haggle” pricing system
and extensive training of salespeople helped build positive relationships with customers. When
something did go wrong, the company and its dealers made unprecedented efforts to solve the

795-011 Saturn Corporation’s Module II Decision

2

problem in a positive way. As a result, Saturn regularly earned a place alongside luxury brands in
rankings of product quality, dealership satisfaction, and overall customer satisfaction.

Saturn was attracting a new audience for GM. About 45% of Saturn owners listed Japanese
imports (primarily Toyotas and Hondas) as the car they would have purchased otherwise. Another
30% listed Ford and Chrysler models. Further research revealed that Saturn owners had higher than
average education levels and household incomes. With a median age of just over 40, the typical
Saturn owner was likely to 4 to 5 more new cars over his or her lifetime than an owner of GM’s
Cadillac, Buick or Oldsmobile models (see Exhibit 1).

The company also served as a proving ground for new ideas. On the technical side, for
example, Saturn’s engine plant was the first to adopt an innovative casting method developed by
GM’s Technical Center during the mid-1980s. Similarly, the Saturn was the only car with durable
thermo-plastic panels that covered most of its body. These panels resisted minor damage from rust or
dents, and were inexpensive to replace if they were cracked or scraped. At the retail level, the
company implemented new approaches to inventory tracking and service management by using
information systems to link dealerships with the factory and corporate offices.

Saturn was governed through a team structure that included suppliers, dealers and union
workers. Each member of a Saturn team participated in training programs that emphasized breaking
down barriers among people with competing interests. Several Saturn officials said that on the best
functioning teams, it was difficult to distinguish members of management from union representatives
because of their focus on a common goal.

As a corollary to teamwork, the company committed to exchange what it learned with GM.
Saturn’s founders had reinforced this objective by including it in the company’s mission statement.
Information initially spread through corporate meetings, plant visits and training sessions. In
particular, GM’s Oldsmobile group tried to incorporate styling features and customer service ideas to
appeal to Saturn ers who wanted larger cars. GM supported the knowledge transfer by
promoting engineers and managers from Saturn into other divisions.

Despite these accomplishments, industry observers were divided over whether Saturn could
be considered a success. Several Wall Street analysts questioned whether the company could sell
more cars without competing with GM’s other divisions. They argued that GM should devote its
resources to its established car brands. Others believed Saturn should be judged by criteria other
than just financial returns. Small cars had never been very profitable, and Saturn was breaking even
after three-and-a-half years in production. Moreover, GM relied on Saturn sales to boost the average
fuel efficiency of its fleet above the 27.5 mile per gallon threshold required by U.S. law.

Options for Expansion

Saturn finished 1993 with its first annual operating profit, estimated at $100 million on sales
of about $3 billion. Officials said the company had been consistently profitable since May, excluding
a two-week summer shutdown for retooling. The retooling would allow Saturn to offer updated
models in 1995 without changing platforms. Toward the end of 1993, Saturn added a third shift of
production workers, which allowed the Spring Hill complex to operate at near its capacity of 320,000
cars per year. This move relieved pressure on the first two shifts, but gave the company additional
production capacity in the midst of the slower winter sales period. Company sources said they had
not expected the third shift to reach its production targets so quickly. Analysts suggested that Saturn
had overestimated demand for its models; first quarter sales increased just 4% over 1993. Regardless,
dealer inventories grew beyond their typical 30-40 day levels to more than 90 days in February 1994,
and Saturn discontinued the third shift to limit output.

Saturn Corporation’s Module II Decision 795-011

3

The production slowdown coincided with meetings between company managers and GM
officials to review proposals for expanding manufacturing capacity. Saturn’s setback highlighted two
of the key issues in the discussions—first, whether sales would continue to grow, and second, what
sales level the company could maintain over the long run. Saturn’s managers believed that the Spring
Hill complex would be running near capacity levels during the late summer and fall. Longer term,
they thought the company was capable of selling 400,000 cars in the U.S. and another 100,000
internationally.

These estimates depended on customer perceptions of the value of the Saturn models. When
the cars were introduced in 1990, the company priced them at $1,000 to $2,000 less than their target
competition, the Honda Civic and Toyota Corolla. New products such as Chrysler’s Neon were
introduced with claims that they redefined the price/value relationship in the small car market. In
addition, surveys of dealers showed that Saturn’s product quality rating had declined from a high of
9.8 in 1993 to 9.2 in 1994 (on a scale of 1 to 10).1

The company’s managers argued that Module II would provide the necessary economies of
scale to remain competitive. By increasing production and sales levels, Saturn could leverage its
investments in design, engineering, sales and marketing. New assembly lines would also reduce the
complexity of introducing new platforms. To change platforms with the current capacity, Saturn
would have to build inventory in anticipation of an 8-10 week shutdown for retooling. Then the
startup phase would have to run smoothly in to restock the retail network. Saturn’s
competitors faced similar risks when they introduced new products, but were generally able to
support dealers with other models during the changeover. With a second module, Saturn could
match this capability by shutting down one plant at a time.

GM was also considering the flexibility a plant expansion would provide for meeting
Corporate Average Fuel Efficiency (CAFE) ratings. Even though the standard had remained
relatively stable for 10 years, the Clinton Administration had hinted that the level could be raised
significantly. New capacity at Saturn would put GM in a better position to respond to a CAFE
increase since small car models were among the most fuel efficient vehicles in the corporation’s fleet.
In addition, GM had not determined the future of its other leading small car, the Geo Prizm, which
was produced in collaboration with Toyota at NUMMI. This joint venture agreement was scheduled
to expire in 1996.

At the beginning of 1994, Saturn favored building Module II on extra land at the Spring Hill
complex but was having trouble convincing GM that this was the best option. GM’s senior managers
focused on several issues, including:

• the initial investment and long run operating costs for each option
• the preferences of the United Auto Workers (UAW) leadership
• Saturn’s relationship with other NAO divisions
• the risks for Saturn

Option 1: Expand at Spring Hill

Locating the Module II site alongside the existing complex was viewed as the least disruptive
option for expansion. Excess capacity at the engine systems and body systems factories would supply
major components, while the company’s current vendors would provide most smaller parts. The
increased volume was expected to lower the production costs for these inputs. A second advantage
was the presence of trained assembly workers and managers who understood the team management
system. These seasoned employees would lead training sessions and could reinforce the lessons by

1NADA dealer survey, Winter 1993/1994.

795-011 Saturn Corporation’s Module II Decision

4

example on the factory floor. Company officials also emphasized the value of informal
communication between Module II and the other groups at the Spring Hill complex. They hoped this
process would encourage innovation across the sites and reinforce Saturn’s team-oriented culture.

UAW vice president Stephen Yokich was not as enthusiastic about the Spring Hill option.2
He contended that it did not make sense to uproot workers from communities where factories had
been shut down or were scheduled to close; GM should be spending any new investment to renovate
these plants and retrain local workers. Yokich had been an outspoken critic of Saturn’s flexible work
rules and had asked to renegotiate the contract on several occasions. This position differed from that
of Saturn’s local union president, Mike Bennett, who had not given Yokich permission to re-open the
agreement. Saturn could expand at Spring Hill without further negotiations, but the Saturn contract
could not be applied to another site without the national UAW’s approval.

Yokich was particularly interested in renegotiating Saturn’s overtime policy. At other GM
plants, assembly employees were paid overtime wages for working more than 8 hours per day. At
Spring Hill, union workers did not have a similar provision, allowing schedules of four ten-hour days
per week. Bennett said this flexibility made it easier for Saturn to schedule three shifts per week,
while saving the company up to $60 million per year in overtime pay.

Cost had been another deterrent to building Module II next to the existing complex.
Constructing a new factory to raise capacity by 178,000 units would require an initial investment of
$900 million, an amount GM was reluctant to spend during its turnaround phase. On the other hand,
the advantages of lean supply lines and concentrated volumes were expected to reduce variable
manufacturing costs to $7,200 per car by 1998,3 a year after the module would be completed.
Variable manufacturing costs currently averaged about $7,600 per car (excluding factory overhead,
preproduction expenses, and depreciation charges). An incremental expansion of Spring Hill over a
longer period would cost proportionately more per unit of capacity (see Exhibit 2 for a summary of
options).

Option 2: Convert an Existing GM Factory

By investing at an existing GM site, Saturn and GM could reduce the initial cost of the project
and generate goodwill with the UAW. This choice would signal a commitment to adapt Saturn’s
management and manufacturing lessons into GM in cooperation with the national union leadership.
The process of “Saturnizing” an established plant would also provide vital information about
programs and practices that could be transferred to other GM factories. Saturn expected to transfer
about 1,500 of its current employees to serve as a core for the new assembly site. Another 1,500
would be recruited from within GM and from the UAW rolls of laid off workers. These additional
hires could come from any GM plant, although the applicant pool was likely to include a high
concentration of people who had previously worked at the site Saturn chose. Saturn’s trainers would
lead all employees through classes to emphasize the company’s values and initiate the team-building
process. Teams would make recommendations on the factory’s design, but they would be
encouraged to replicate the assembly processes of the Spring Hill plant.

Saturn’s financial planners attempted to determine the optimal distance between a new site
and the Tennessee complex. Many auto makers built cars in separate regions to reduce delivery time
and transportation expense to dealerships. Saturn weighed these factors against the costs of
stretching its supply lines. Any Module II plant would rely on Spring Hill for major components.

2Local unions were covered by an umbrella contract negotiated by the national UAW, but were given flexibility
to modify some areas of the agreements through bargaining with local plant managers.
3The figures in this paragraph are estimates by the casewriter.

Saturn Corporation’s Module II Decision 795-011

5

The company also wanted to avoid having different suppliers for the new site. Expanding on the
West coast had been ruled out, even though this alternative offered advantages for exporting to Asia.

Alternatives to expanding Spring Hill were suggested and explored by Saturn management.
One choice was a closed plant in Willow Run, Michigan—20 miles west of GM’s Detroit headquarters.
This assembly plant had produced larger cars prior to closing in September 1993. Willow Run’s local
union had negotiated one of the most flexible contracts among GM plants in an attempt to avert a
shutdown. This selection would signal a strong symbolic commitment to GM’s constituency in the
Detroit area. Another possibility was an underutilized Chevrolet Corvette factory in Bowling Green,
Kentucky. Proponents said that this location, 50 miles north of Spring Hill via a major highway,
would allow Saturn to retain many of its suppliers. Company officials also felt that Bowling Green’s
proximity would encourage cross-plant communication.

Saturn estimated the costs of converting Willow Run at $850 million, including relocation and
training expenses. The facility could be operational in 36 months and would be capable of
manufacturing 178,000 cars annually. Bowling Green would require a $900 million investment and
would be ready for production within 30-36 months. Because of its smaller size, however, this site’s
capacity would be limited to 100,000 cars per year.

As a prerequisite to converting Willow Run or Bowling Green, the company had to negotiate
a separate contract with the UAW. Differences in the contracts at Spring Hill and the new site could
lead to variability in the terms and levels of compensation. Assuming some sort of agreement could
be reached, Saturn would have to adapt its organizational structure in response to the challenges of
coordinating two locations. For example, the company had attempted to push decision-making down
to the factory floor, but with two plants, the choices made by teams might vary. If these differences
were not reconciled over time, the modules might eventually disagree on larger issues, such as
choices of suppliers or designs for the cars. Similarly, managing the array of cooperative
relationships could be more difficult across two plants, especially if the labor-management
partnership deteriorated at either one.

From a marketing and sales perspective, expanding outside of Spring Hill would probably
enhance interest in the company. Saturn had attracted the public’s attention in the late 1980s and
early 1990s by advertising its ambitions to change the way cars were manufactured and sold.
Spreading its methods to another production location would be an intriguing second chapter for “The
Saturn Story.” The potential disadvantage besides higher costs, however, was that any setbacks—
such as quality problems or internal tensions—would occur under a spotlight of media attention. If
these issues were not resolved to the public’s satisfaction, Saturn could lose many of its admirers.

Option 3: Source Models from Another GM Plant

Company officials noted that this option would leverage Saturn’s brand equity and sales
network. Industry studies showed that customers had such a positive perception of the Saturn name
that they would be willing to pay a premium for any car the company sold. Therefore, even though
GM was likely to sell a hybrid of the new model through Chevrolet or Pontiac dealerships, Saturn’s
marketing and customer service would help distinguish the products.

Analysts offered mixed opinions of this option. While the financial details looked positive,
some questioned the long term effects of diluting the Saturn brand. Customer enthusiasm could
diminish if the company was seen as deviating from its original goal of doing things differently. The
move would also represent a shift in Saturn’s mission of “market[ing] vehicles developed and
manufactured in the United States,” a fact that might be highlighted by the news media.

In addition, these changes would have major implications for the relationships among the
various interest groups within and outside the company. Saturn’s retailers might not be willing to

795-011 Saturn Corporation’s Module II Decision

6

maintain their “no haggle” pricing approach when a customer could negotiate a lower price for
virtually the same car at a Chevrolet or Pontiac dealership. In addition, engineers would be likely to
consider the outsourcing choice as an indication that they would eventually be transferred to the
centralized NAO design center.

Option 4: Delay Expansion

“We can be profitable without [Module II], but we can be far more profitable over time by
having added capacity,” said Don Hudler, Saturn’s vice president of sales, service and marketing.4
David Cole, the director of the University of Michigan’s Center for the Study of Automotive
Transportation, seconded this opinion while discussing Saturn’s labor contract, the need for new
products, and the plant expansion decision: “These are all important things that need to be done… but
if they’re not done in the next six months, it’s not the end of the world.”5

GM and Saturn had been trading proposals regarding Module II since the end of 1993, but no
decision had been announced. Analysts suggested three likely reasons for the delays. First, when
Saturn began meeting with its parent, GM’s decision-making ability was affected by its limited
resources. Consequently, GM seemed to focus on other priorities while waiting for its cash balances
to recover. Second, Saturn’s sales forecasts may have been questioned because of concerns about
Chrysler’s Neon introduction. The Neon had received strong endorsements from critics and was
advertised at $1,000 less than Saturn’s base model. Third, GM was pursuing closed-door negotiations
with the national UAW to see whether the two sides could reach a compromise related to a Saturn
contract for a converted GM facility.

By June 1994, two of these three constraints had relaxed somewhat. GM’s strong
performance in the first half of the year generated capital to invest in the car divisions. Saturn had set
a sales record in May, while the Neon’s momentum had been slowed by a recall shortly after the
model’s introduction. One source said GM’s final decision hinged on the outcome of bargaining with
the UAW.

Meanwhile, every month of delay shifted the relative advantages of the options. Each one
had a window of opportunity influenced by the required lead time (options 1 and 2) or by the
planning processes at other divisions (option 3). In addition, Saturn incurred extra costs related to
planning around several different outcomes. Mike Bennett, Saturn’s local union president, noted that
as the company had adjusted its business plans since 1992, several projects had been postponed or
canceled. This process was especially difficult for teams preparing Saturn’s international strategy
since they did not know what types of new products the company would be able to provide.

General Motors’ North American Automotive Operations

GM’s North American Automotive Operations (NAO) encompassed seven car and truck
lines6, as well as numerous components suppliers. In 1993, the group accounted for 70% of GM’s
$134 billion in revenues and contributed $190 million in profits following three consecutive years of
negative net income (see Exhibit 3).

4“A Falling Star; Mission Accomplished, Saturn Joins GM’s Struggles,” Chicago Tribune, May 1, 1994, p. 3.
5“Saturn Showing Growing Pains,” Gannett News Service, June 23, 1994.
6Buick, Cadillac and Saturn were strictly car brands, while Chevrolet, Pontiac and Oldsmobile marketed
passenger trucks in addition to cars. GMC Truck, the seventh division, sold larger utility trucks and vans.

Saturn Corporation’s Module II Decision 795-011

7

The 15 years between 1979 and 1994 had been difficult for NAO divisions. In 1979, GM sold
nearly half of all new cars in the U.S. By 1993, its share had slipped to under 35%. Initial losses were
attributable to competition from low-priced Japanese models that offered better fuel economy and
superior reliability. In the late 1980s and early 1990s, GM faced increasing competition from its
domestic rivals, Ford and Chrysler. These companies had taken steps to improve product
development and manufacturing quality that were beginning to pay off.

GM’s CEO Roger Smith had pursued a range of plans to revitalize NAO but achieved mixed
results during his 9 years at the helm. When he retired in August 1990, he told reporters he wanted to
be remembered as someone who helped prepare GM for the next decade. He had invested heavily in
plant technology and led GM into joint ventures with other auto makers. However, he had also
presided during the period of NAO’s share declines that forced GM to close factories and lay off
thousands of union workers.

NAO’s problems escalated under Smith’s successor, Robert Stempel. The continuing U.S.
recession, coupled with increases in labor and benefits expenses, contributed to GM’s losses of $2.0
billion in 1990 and $4.5 billion in 1991. In April 1992, the Board of Directors promoted John F. “Jack”
Smith (no relation to Roger) to the presidency and placed John Smale, Procter & Gamble’s former
CEO, in charge of the GM executive committee. Stempel retained his duties as chief executive but his
influence waned. In October of 1992, the board asked Stempel to step down and elevated Smith to
CEO. Jack Smith had managed GM’s international auto operations during the mid-1980s. His efforts
to cut costs and streamline decision-making were widely credited for transforming GM-Europe into a
consistent profit center. As CEO, he issued a similar blueprint for GM’s North American divisions.

By the spring of 1994, NAO had regained profitability. Customers were returning to
showrooms as the U.S. economy rebounded from a long recession. GM’s divisions posted gains of
more than 19% in the first quarter, while the industry as a whole grew by about 14% on a unit basis.
Analysts attributed the auto maker’s better-than-average results to improved products and a strong
yen, which eased price competition from Japanese imports. Smith also predicted that NAO’s share
would grow as a result of new product introductions planned for later in 1994 and 1995.

The Car Divisions

Alfred Sloan organized GM around car divisions in the early 1920s to offer “a car for every
purse and purpose.” Chevrolet targeted entry-level ers, while Pontiac, Oldsmobile and Buick
offered models that moved progressively up the price scale. Cadillac staked out the high-end market
by providing the largest and most luxurious cars on the road.

From the 1930s to the late 1950s, each of GM’s five divisions sold a distinct model. While the
cars shared basic components to save costs, every brand offered a different engine, chassis and body
style to create individual personalities. The corporation began deviating from this formula in 1959
when it introduced smaller cars that shared “platforms” (i.e. engine and chassis). This practice
became known as “badge engineering.” A decade later the company offered 18 models, many of
which differed in only cosmetic features. For instance, the Buick Skylark, Oldsmobile F-85, Pontiac
Tempest, and Chevrolet Chevelle were virtually identical except for their body panels and
upholstery.

Analysts believed GM’s increasing reliance on badge engineering and other cost-saving
measures during the late 1960s and early 1970s reflected concerns about antitrust scrutiny by the U.S.
government. The firm had achieved a 50% market share and could no longer aggressively increase
sales without the threat of prosecution. Management therefore focused on increasing profits by
taking costs out of the vehicles. Over time, organizational units from different divisions were merged
until their boundaries were indistinguishable. This approach produced strong financial results until
the late 1970s.

795-011 Saturn Corporation’s Module II Decision

8

In the face of new competitive challenges, GM attempted to revive the divisions’ separate
identities under a major reorganization plan in 1984. The company’s leaders considered
reconstituting the five divisions, but eventually decided to separate Buick, Oldsmobile and Cadillac
(BOC) from Chevrolet, Pontiac and Canadian operations (CPC). One executive recalled, “as we got to
looking at the individual areas of expertise…, it was clear that you couldn’t cut them into fives
without losing your critical numbers in any one of the capabilities. We finally concluded that two
[groups] was as many as you could do.”7

Saturn was formed outside the BOC/CPC framework to insulate the new division from
budget pressure and tensions between brands. This independence was intended to provide flexibility
for testing new ideas. Saturn was allowed to produce its own engines and transmissions to infuse the
models with a unique feel. New features and styling choices reinforced the claim that Saturn was
offering “a different kind of car.”

The Turnaround Plan

Starting in 1992, Jack Smith pursued an aggressive plan for improving …