Qualitative Research in Accounting & Management

Strategy, IT and control @ eBay, 1995-2005: The management control system (MCS) as
consumer product
Dan N. Stone Alexei N. Nikitkov Timothy C. Miller
Article information:
To cite this document:
Dan N. Stone Alexei N. Nikitkov Timothy C. Miller , (2014),"Strategy, IT and control @ eBay, 1995-2005",
Qualitative Research in Accounting & Management, Vol. 11 Iss 4 pp. 357 - 379
Permanent link to this document:
Downloaded by UNIVERSITAET OSNABRUCK At 17:11 30 January 2016 (PT)

http://dx.doi.org/10.1108/QRAM-06-2013-0025
Downloaded on: 30 January 2016, At: 17:11 (PT)
References: this document contains references to 83 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 511 times since 2014*
Users who downloaded this article also downloaded:
Bruno Cohanier, (2014),"What qualitative research can tell us about performance management systems",
Qualitative Research in Accounting & Management, Vol. 11 Iss 4 pp. 380-415 http://dx.doi.org/10.1108/
QRAM-06-2013-0023
Richard Baker, (2014),"Qualitative research in accounting: the North American perspective", Qualitative
Research in Accounting & Management, Vol. 11 Iss 4 pp. 278-285 http://dx.doi.org/10.1108/
QRAM-08-2014-0054
Rick Stephan Hayes, Richard Baker, (2014),"A participant observation study of the resolution of audit
engagement challenges in government tax compliance audits", Qualitative Research in Accounting &
Management, Vol. 11 Iss 4 pp. 416-439 http://dx.doi.org/10.1108/QRAM-02-2013-0003

Access to this document was granted through an Emerald subscription provided by emerald-srm:203840 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
 Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.

*Related content and download information correct at time of download.

Downloaded by UNIVERSITAET OSNABRUCK At 17:11 30 January 2016 (PT)
 The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1176-6093.htm

Strategy, IT and control @ eBay, Strategy, IT, and

control @ eBay,
1995-2005 1995-2005
The management control system (MCS) as
consumer product 357
Dan N. Stone
Von Allmen School of Accountancy, University of Kentucky, Lexington,
Kentucky, USA
Downloaded by UNIVERSITAET OSNABRUCK At 17:11 30 January 2016 (PT)

Alexei N. Nikitkov
Faculty of Business, Brock University, St. Catharines, Canada, and
Timothy C. Miller
Department of Accountancy, Xavier University, Cincinnati, Ohio, USA

Abstract
Purpose – This paper aims to adapt Simons’ (1995b) theory of the role of information technology (IT)
in shaping and facilitating the levers of control (i.e. the Levers of Control Applied to Information
Technology – LOCaIT) as a framework for investigating how eBay’s business strategy was realized
through its management control system (MCS) in the first 10 years of the online auction market.
Design and method – The qualitative method uses data from public record interviews, teaching
cases, books, Securities and Exchange Commission filings and other archival sources to longitudinally
trace the realization of eBay’s strategy through its MCS and IT.
Findings – Realizing its strategy through the eBay MCS necessitated a diagnostic control system
unlike any previously seen. This system created a close-knit online community and enabled buyers and
sellers to monitor one another’s performance and trustworthiness.
Research limitations and implications – The LOCaIT theory facilitated understanding the core
aspects of the realization of eBay’s strategy through its MCS and IT. However, LOCaIT largely omits the
strong linkages evident among elements of the MCS, the importance and necessity of building a core IT
infrastructure to support eBay’s strategy and the central role of building consumer trust in the
realization of this strategy.
Practical and social implications – eBay’s MCS is now, perhaps, the world’s most widely imitated
model for creating online trust and user interactions (e.g. Yelp, TripAdvisor, Amazon). In addition,
eBay’s MCS was “sold” as a consumer product that was instrumental in facilitating consumer trust in
the online auction market.
Originality/value – Contributions include: tracing the creation, growth and evolution of, perhaps, the
world’s largest and most widely imitated MCS, which redefined the boundaries of accounting systems

Alexei Nikitkov thanks Brock University for a grant supporting this study. Dan Stone and Tim
Miller thank the University of Kentucky, the Gatton College of Business and Economics and the
Von Allmen School of Accountancy for grants supporting this research. For valuable comments Qualitative Research in Accounting
& Management
and suggestions on previous drafts, thanks to workshop participants at the American Accounting Vol. 11 No. 4, 2014
Association 2006 annual meeting, the University of Florida, the University of California – pp. 357-379
© Emerald Group Publishing Limited
Riverside, the University of Montana, and the 2011 Accounting, Organizations, and Society Fraud 1176-6093
conference. DOI 10.1108/QRAM-06-2013-0025
 QRAM monitoring; and testing the range, usefulness and limitations of Simons’ LOCaIT theory as a lens for
understanding eBay’s use of IT in their MCS.
11,4 Keywords Ebay, MCS, LOCaIT, Information technology, Business strategy
Paper type Research paper

358 That many business transactions are now conducted using small, lighted,
two-dimensional rectangular boxes on poorly understood, abstract systems that exist
“in a cloud” would probably strike anyone unfamiliar with the Internet as strange and
unsafe. While the uncertainty of the “cloud” is mitigated by online accounting systems
and controls, the evolution of these widely implemented and trusted abstract systems is
poorly understood. In this paper, we investigate these issues of trust, accounting
Downloaded by UNIVERSITAET OSNABRUCK At 17:11 30 January 2016 (PT)

systems, information technology (IT) and controls by documenting the evolution of the
eBay MCS viewed through Simons’ (1995b) levers of control framework.
Our work considers a system that predates, and was the prototype for, the system
considered by Jeacle and Carter (2011), who examined the development of management
control and the use of rankings and calculations at TripAdvisor. Jeacle and Carter (2011)
base their work in Gidden’s (1990, 1991) analysis of abstract systems and Miller’s (2001)
exploration of calculative practices. The present paper uses a different theoretical focus
and investigates how eBay’s business strategy was realized through its development of
a unique management control system in the first 10 years of the online auction market.
eBay’s MCS facilitated the creation of a close-knit online community and enabled buyers
and sellers to monitor one another’s market performance and trustworthiness. The
success of this system is evident in its subsequent imitation by many eBay competitors,
including Yelp, the aforementioned TripAdvisor (Jeacle and Carter, 2011), and Amazon.
The development of eBay’s MCS, and its use of IT, affords investigation of the
strategies that lay beneath an emergent assemblage of social relations and technology
changes that facilitated the creation of abstract, calculative systems. Our investigation
explores the following questions: What strategies enabled eBay to pioneer and sustain
an unprecedented level of consumer trust in an unfamiliar, abstract e-commerce system
(cf. Giddens, 1990, 1991) and how did eBay’s MCS support and enable the success of
these strategies? How did IT support eBay’s construction of an innovative MCS that was
scaled for global reach? This study unpacks the structures and strategies that lay
beneath the creation and growth of the eBay MCS and the central role of IT in creating
a MCS to support and realize eBay’s strategy and mission.
To explore these issues, this study applies and extends Simons’ (1995b) theory of the
levers of control as applied to information technology (LOCaIT) as a framework for
investigating the first 10 years of the eBay strategy and management control system
(MCS).
This is the first investigation of which we are aware that is based in a test and
extension of Simons’ LOC framework regarding IT. Our choice of a theoretical
perspective is partially motivated by a desire to explore a largely neglected portion of a
theory that has been widely used among professionals, and widely investigated by
scholars. Specifically, while more than 2,000 investigations use the core principles of
Simons’ LOC model, surprisingly, none that we could locate investigate Simons’
extension of his theory to the role of IT in supporting and enabling management’s
realization of its business strategies. This extension of the LOC into IT, which we refer
 to as LOCaIT, is a mapping of the levers of control onto two information attributes – Strategy, IT, and
information codification and diffusion – and the degree to which each attribute should
persist, or not.
control @ eBay,
eBay’s business strategy was realizable only through creating and implementing an 1995-2005
MCS and supporting IT infrastructure, which makes the present investigation an
appropriate application of Simons’ LOCaIT theory. Malsch and Gendron (2013, p. 874)
observe, in quoting (Ketokivi and Mantere, 2010, p. 319), that: 359
[…] it is always the researcher who selects the ”best“ from among competing explanations, and
the de facto criteria for best are defined as pragmatic virtues such as interestingness,
usefulness, simplicity, or conservativeness – not truth value or even empirical adequacy.
Our choice of LOCaIT explores its contributions and failings in a well-matched context,
Downloaded by UNIVERSITAET OSNABRUCK At 17:11 30 January 2016 (PT)

with an intention of guiding scholars who are considering theoretical groundings for
exploring the linkages among accounting systems, trust, control and IT.
Few studies investigate the eBay MCS (Boyd, 2002; Duh et al., 2002); the present
study contributes to the eBay MCS literature by being the first, to our knowledge, to
explore the creation and evolution of the eBay MCS and to explore the role of IT in its
realization. Many published studies investigate aspects of the online auction market,
which eBay created and dominated (see Dellarocas, 2003a and Hasker and Sickles, 2010
for literature reviews). Several books trace the origins and history of eBay (Bunnell and
Luecke, 2000; Lewis, 2008; Gilbert, 2012); one includes eBay management cooperation
(Cohen, 2002). No investigations, of which we are aware, trace the relation of the eBay
MCS to its business strategy or the dynamics and evolving nature of the interplay
between eBay’s business strategy, its MCS, IT and the evolving online auction market.
We next describe our theoretical lens, i.e. Simons’ LOC model and his LOCaIT
extension of this model, followed by a brief description of the method of investigation.
Following this, we briefly summarize eBay’s business history, strategy and the
development of its core IT capabilities. Linking LOCaIT with eBay’s systems follows,
with a concluding section that considers the implications of the results.

2. Simons’ LOCaIT model

2.1 The basic LOC model
The LOC model was first articulated by Simons (1994), although several previous
papers (Simons, 1987, 1990, 1991, 1992) articulate model elements. The basic LOC model
derived from interviews, conducted over 18 months, with ten newly appointed C-suite
managers, i.e. Presidents and chief executive officers (CEOs), regarding their use of
controls to direct and realize corporate strategy (Simons, 1994). The four types of control
systems, using citations to the Simons (1995b) book, are (emphasis added to highlight
system types):
(1) “A belief system is the explicit set of organizational definitions that senior
managers communicate formally and reinforce systematically to provide basic
values, purpose and direction for the organization” (Simons, 1995b, p. 34). In
short, the belief system articulates an organization’s purpose.
(2) Boundary systems – “[…] delineate the acceptable domain of activity for
organizational participants […] and establish limits […] to opportunity seeking”
(Simons 1995b, p. 39). The two types of boundary systems are: business systems,
which are guided by, “[…] society’s laws, the organization’s beliefs systems and
 QRAM codes of behavior promulgated by industry and professional associations
(Gatewood and Carroll, 1991),” (Simons, 1995b, p. 42) and strategic systems,
11,4 which limit the opportunity space (e.g. only invest in projects with a ROI of 15
per cent (Simons, 1995b, p. 47-51).
(3) “Diagnostic control systems (DCSs) are the formal information systems that
managers use to monitor organizational outcomes and correct deviations from
360 preset standards of performance” (Simons, 1995b, p. 59). DCSs are defined by the
following features:
• the ability to measure the outputs of a process;
• the existence of predetermined standards against which actual results can be
compared; and
Downloaded by UNIVERSITAET OSNABRUCK At 17:11 30 January 2016 (PT)

• the ability to correct deviations from standards.

(4) “Interactive control systems (ICS) are formal information systems managers use
to involve themselves regularly and personally in the decision activities of
subordinates” (Simons, 1995b, p. 95). These systems have the following
characteristics:
• system output is an important and recurring agenda addressed at the highest
levels of management;
• by operating managers at all levels of the organization
• system output is interpreted and discussed in face-to -face meetings of
superiors, subordinates and peers; and
• the system is a catalyst for the continual challenge of, and debate about,
underlying data, assumptions and action plans (Simons, 1995b, p. 97).

The model argues for a productive tension among these control systems that balances
control with empowerment. In Chinese philosophy, opposing and interdependent forces,
i.e. the yin and yang give rise to and complement one another. Within Simons’ model,
belief and IC systems create positive and inspirational forces, i.e. the yang, representing
sun, warmth and light, while the remaining two, boundary and DC systems represent
yin, i.e. constraints, darkness and cold (Simons, 1995b, p. 7-8).
2.1.1. LOC model applications. With over 2,000 citations to the LOC, the present work
does not aspire to a review of the LOC literature. Instead, we conducted a literature
search of two online databases (EBSCO Business Source Premier, Google Scholar) for
published papers that included the phrase “levers of control” or “Simons” in either the
title, keywords or abstract. After omitting false positives (e.g. citations to work other
than Simons’ LOC model), we identified nine direct applications of the LOC model and
nine papers that extend the LOC model. None of the identified papers used or applied the
LOC model or framework to IT issues.

2.2 Application and extension of the LOCaIT model

The present work applies and extends the LOCaIT model found in Appendix B of
Simons’ (1995b) Levers of Control book. In this Appendix, Simons predicts the value,
role and limitations of information technology in creating, disseminating and applying
the proposed LOC, i.e. LOCaIT model. A Google Scholar search to identify literature that
applies the LOCaIT model, using the author and key terms of the LOCaIT model: “’levers
 of control’, Simons, codification, diffusion” resulted in 50 hits that were either false Strategy, IT, and
positives, unpublished papers or works in which the discussion of the Simons LOCaIT
model is incidental to the works’ primary focus. Accordingly, the evidence suggests that
control @ eBay,
the present work is among the first applications of the LOCaIT model in scholarly or 1995-2005
professional work. Hence, while Simons’ LOC model has been widely adopted and
applied, the LOCaIT model remains largely uninvestigated.
The central concern of the LOCaIT model is the capacity of IT to codify and diffuse 361
the levers of control to extend their power and influence in creating and implementing
business strategy. According to Simons, codification and diffusion are fundamental to
understanding the communication and control needs of senior managers and to
mapping IT innovations to their potential use as MCS enhancements. IT innovations
often fail to support and enhance MCS “[…] because the designers of these systems do
Downloaded by UNIVERSITAET OSNABRUCK At 17:11 30 January 2016 (PT)

not understand how senior managers use information for control purposes (p. 184).”
Leveraging IT innovations to enhance MCSs requires an understanding how, “[…]
different levers of control require different configurations of information systems.”
Information codification concerns the structuring of information by categorizing and
compressing data into more readily used forms. For example, quantitative financial
statement information and descriptive statistics illustrate highly codified information.
In contrast, gossip about a new technology is difficult to codify because it is vague,
ambiguous and dependent on context. Information diffusion concerns the degree to
which information is shared and available internally. Diffusion is high when
information is transmitted easily to everyone in the organization and is low when
information is available only to a small subgroup within the organization. Continuing
the previous example, an income statement is easily diffused while gossip about a new
technology is not, as the credibility and relevance of gossip requires face-to-face
re-telling. Simons argues that, while codification and diffusion often co-vary, a
positively co-varying relation is not universal. For example, information provided to the
Board of Directors may be highly codified but evidence low diffusion, while stories
about a charismatic founder may evidence high diffusion but low codification.
Table I summarizes the LOCaIT model, links the four control systems to the potential
uses of IT to codify and diffuse their influence and provides illustrative uses of IT to
achieve greater codification and diffusion. According to the model, IT can help diffuse
belief and boundary systems throughout the organization through the use of
well-designed technologies such as conferencing, e-mail and reminders. However, IT is
likely to be of low value in codifying belief and boundary systems, as the effectiveness
of these systems does not depend on codification. DCSs can benefit from IT by both
increasing their codification and by diffusing exception reporting to managers, while
ICS can benefit from IT by increasing codification and regular, periodic diffusion to
managers (c.f. Simons, 1995b, pp. 183-195).
We next describe the research method.

3. Case method
Clear selection criteria are important to establish the contribution of research that
focuses on critical cases (Eisenhardt and Graebner, 2007; Miles and Huberman, 1994).
We chose eBay as the market, and the eBay MCS as the focus of the study because of
eBay’s unique position as the creator of the world’s largest online MCS and its dominant
online auction market share. Longitudinally tracing the evolution of the eBay MCS
 Downloaded by UNIVERSITAET OSNABRUCK At 17:11 30 January 2016 (PT)

11,4

362

Table I.
QRAM

control (LOCaIT)
diffuse the levers of
Simons’ model of use of
technology to codify and
System Application Desired codification Desired diffusion Examples

Belief Widely diffuse personal Low, highly personal Throughout organization, Audio-video conferencing,
message periodically Email
Boundary Ensure internalization Low, unambiguous Throughout organization, “Mischarging is illegal” popup
of unambiguous periodically on data screen entry
message
DCS Increase efficiency and High, measure and monitor Managers for exceptions Color-coded exception reporting,
effectiveness of critical critical performance drill-down matrix, balanced
performance measures variables scorecard reports
and transmit exception
information
ICS Real-time data, Medium, transform raw All managers, Database access to market
including market data into accessible periodically trends, profitability, models,
dynamics; action formats internal technology projects
planning and testing
scenarios

Source: Simons (1995b) Appendix B

 constitutes the qualitative, and “netnographic” (Jeacle and Carter, 2011; Jayanti, 2010; Strategy, IT, and
Rokka, 2010) investigation methods. The evolving eBay MCS is both a focus of, and a
challenge to, this investigation.
control @ eBay,
One data challenge in investigating eBay’s strategy and MCS is secretiveness – eBay 1995-2005
declines most requests for interviews and data (Lewis, 2008; Bunnell and Luecke, 2000,
p. 97) and, declined our requests for interviews with eBay staff and executives regarding
its MCS and for access to historical documents related to its evolving user agreements 363
and interface. Fortuitously, eBay’s growth and development is well documented in the
business press, in books about the company, in online blogs and ecommerce Web sites
and in business school cases. We worked to ensure that our secondary data sources
included eBay company documents (e.g. Securities and Exchange Commission [SEC]
Downloaded by UNIVERSITAET OSNABRUCK At 17:11 30 January 2016 (PT)

filings) and sources written by both company supporters (Cohen, 2002) and critics
(Kenny, 2008; Klink and Klink, 2005).
Data on eBay’s MCS exist in its SEC filings, in occasional eBay MCS-related
announcements (interviews and press releases) and in the analysis of the MCS itself at
the eBay Web site, including its user forums (eBay, 2008; Vendio Services, 2008) and
transaction records. Indirect (secondary) data on the eBay MCS exist in histories written
by journalists mostly praising eBay’s business model and success (Cohen, 2002),
first-person histories of eBay crime and fraud (Walton, 2006), Web sites that evaluate
and critique online auctions and eBay (Kenny, 2008; Klink and Klink, 2005) and business
school case studies, most of which were written with the cooperation of high-level eBay
executives who were alumni of the business schools whose faculty were granted access
(i.e. Harvard and Stanford).
The study’s qualitative method mapped the relevant events in eBay’s MCS history to
the LOCaIT model and to the theoretical constructs therein (i.e. codification and
diffusion). Our investigation focused on those aspects of eBay’s MCS for which public
data exists, e.g. the DCS feedback ratings and fraud strategies. Our investigation
excludes those aspects of the eBay MCS that are exclusively internal systems, e.g. the
budgetary system, as we are unaware of relevant data sources. We next present a brief
history of eBay and a description of its business strategy and core IT capabilities.

4. eBay history, strategy and core IT capabilities

4.1 eBay history and strategy: 1995-2005
eBay’s mission was to “[…] to develop the world’s most efficient and abundant online
trading platform in which anyone, anywhere can buy practically anything” (eBay 10K,
2002). eBay’s success in achieving these goals inspired even otherwise skeptical
financial analysts. For example, Albert Meyer, in a December 2002 analysts report
stated:
[…] if we forget about true bottom-line profitability for a while, eBay is undoubtedly the most
astonishing story since the dawn of the Internet.[…] It is impossible not to get animated by the
eBay phenomenon. It reads like a modern-day fairy tale (Hawkins and Cohen, 2003).
EBay’s growth was exceptional (Table II). From 1997 to 2005, the number of registered
users grew from about 34,000 to 180 million, listings grew from about 44,000 to 1.9
billion and sales grew from about $95,000 to $44 billion. eBay’s dominance of the online
auction market was sufficient that its main competitors (i.e. YAHOO! Auctions and
 QRAM Amazon Auctions) eventually closed their auction sites (Yahoo in 2007 and Amazon in
2008) (Cohen, 2002; Skogøy, 2010).
11,4 Simons envisions a corporation’s MCS as a means for realizing its business strategy.
eBay’s business strategy, which is explicitly stated in its first SEC filing (S-1 1998),
includes the following elements[1]:
• growing the community, brand, product categories and markets served, such that
364 eBay’s goal was growth, and that this goal was achieved (Helft, 2001) is evident in
the above summary of its first 10 years (Table II). As intended, this growth was of
its US community, brand, product categories and markets.
• Fostering eBay community affinity; eBay was and is a unique online community
(Boyd, 2002). The creation, growth and evolution of this community is discussed
Downloaded by UNIVERSITAET OSNABRUCK At 17:11 30 January 2016 (PT)

more below in relation to the company’s belief and boundary systems .

• Enhancing site features and functionality, and, introducing value-added services.
The role of IT and its web-based delivery platform is discussed more in the section
below related to IT.
• Leveraging eBay’s business model as a zero inventory, zero sales force and
net-based business (eBay Form S-1, 1998). From inception, eBay’s strategy
unprecedented levels of outsourcing. For example, eBay never held inventory, has
no supply chain, no managed distribution network and no warehouses (Frei and
Rodriguez-Farrar, 2005; eBay, 1998). This unique business model, as a net-centric
market facilitator gave it several distinct advantages over competitors (e.g.
Amazon and Yahoo) by reducing demands for: cash, logistics for product delivery
and for inventory and sales force management. This business model partially
enabled eBay’s extraordinary early growth, while other businesses, e.g. Amazon,
struggled to create those aspects of e-commerce, e.g. inventory and logistics
management, that eBay, in its early years, avoided.

4.2 Information technology

Creating its core IT functionality was a critical, and distinctive, aspect of eBay’s
business strategy. For example, upon learning about eBay, Meg Whitman, the
company’s first CEO observed that a critical aspect of eBay was that it could not have
existed offline. However, the centrality of IT to a business, i.e. that some businesses
cannot realize their strategy without IT, is absent from the LOCaIT model. But not all IT
systems were central to eBay’s business strategy. Rather, the priority and importance of

1997-2005
Year 1997 2000 2002 2005 % change

Registered users 0.34 22.5 61.7 180.6 53,018

Listings 0.0439 264.7 638.3 1876.8 4,275,071
Sales $0.0953 $5,422 $14,868 $44,299 46,483,636
Auction listing value
Table II. (Average) $21.68 $20.48 $23.29 $23.60 8.9
eBay activity volume and
listing value–selected Note: Registered users, listings and sales in millions
years from 1997 to 2005 Source: eBay SEC 10K filings and Annual Reports
 eBay’s IT bifurcated into front- and back-end systems. The stability, reliability, Strategy, IT, and
scalability and capacity of the front-end system, by which customers accessed and
traded on eBay, was a core system to generate revenue. Hence, creating reliable and
control @ eBay,
stable front-end systems was essential to eBay’s success and growth. For example, in 1995-2005
the month of January 1997 alone, eBay hosted about 200,000 auctions compared with
250,000 auctions for the 1996 calendar year (Cohen, 2002). Further, the complexity of
eBay Web site traffic was significantly greater than that of other sites where content 365
was merely posted, as the eBay system processed complex transaction data. Because of
this demanding transaction load, Omidyar’s cobbled-together shareware and
hand-written PERL (Practical Extraction and Reporting Language) code frequently
crashed the site. This led to eBay, ironically, asking in January 1997 that users avoid the
site during peak use hours. Omidyar hired Mike Wilson to re-engineer the site, from
Downloaded by UNIVERSITAET OSNABRUCK At 17:11 30 January 2016 (PT)

scratch, and to repair the existing code sufficiently to keep the site functioning until the
new code could be written, tested and implemented. Site stability and scalability
remained a critical problem however, even after Wilson implemented the new system in
September, 1997. Shortly after Whitman assumed the CEO’s job in March 1998, eBay’s
site crashed for eight hours. Whitman described the crash as, “[…] a wake-up call for
me” that motivated a significant investment in expanding system capacity.
System stability remained a critical problem, however, because of the rapid
expansion of US users and the aggressive expansion of eBay into non-US markets.
eBay’s site crashed repeatedly in May and June 1999, culminating in a 22-hour crash on
June 10, followed by a tentative, unsteady recovery over the next week (Cohen, 2002).
eBay’s stock dropped approximately 20 per cent following the June crashes (Yahoo
finance). Following these crashes, eBay replaced its Chief of Technology, hired three
new high-ranking engineers and increased investment in “product development”, i.e. in
creating a stable, useful, scalable Web site. Efforts to stabilize the system and expand its
capacity increased eBay’s “product development” costs from $4.6 million in 1998 to
$55.9 million in 2000 (eBay 10-K, 2000). These investments paid off in enhanced system
reliability, with only minor stability issues (e.g. in January 2001). Although subsequent
crashes occurred, they were much less disruptive than the 1999 failures.
While eBay’s front-end systems were deemed essential to its success, the resources
devoted to its accounting, i.e. back-end systems, lagged those devoted to its front-end
system. At inception, Omidyar used QuickBooks as eBay’s accounting system (Cohen,
2002). In May 1996, Omidyar began formalizing the eBay accounting system when, nine
months after launching his business, he hired a friend to come to his home twice a week
to open envelopes containing checks and organize them for deposit. In December 1996,
Omidyar hired an office manager, called the “Head of Finance”, who was assigned the
duties of customer service, billing, filing and human resource. One year later (December
1997), in anticipation of the company’s Initial public offering (IPO), eBay hired a chief
financial officer, Gary Bengier, who replaced the eBay accounting system with, “[…] a
new robust, scalable ‘backend’ transaction processing architecture (eBay 10-Q
September, 1998)”.
In short, eBay’s core IT functionality was central to its strategy, but this role for IT is
not found in the LOCaIT model. Instead, the assumption of LOCaIT is that management:
has a strategy and may consider how to use IT to implement it. That IT could be
essential to a core business strategy is a possibility that is unconsidered in the model. In
addition, while eBay devoted considerable resources to its front-end systems,
 QRAM investment in its back-end accounting systems seemed to occur only in anticipation of
eBay’s IPO.
11,4
5. “Locating” eBay’s MCS
Its position as the market leader in creating the world’s largest online auction market led
eBay to also be the industry leader in establishing an MCS (Cohen, 2002; Hoyt and
366 Baron, 2001, eBay S-1 filing, 1998). The following sections discuss the role of eBay’s
belief, boundary, DCS and ICS in realizing its business strategy. Each section discusses
the role of IT in codifying and diffusing the separable LOC systems, including the
following aspects of eBay’s MCS:
• Its core belief and boundary system.
Downloaded by UNIVERSITAET OSNABRUCK At 17:11 30 January 2016 (PT)

• The DCS that provided feedback to users.

• The boundary and DCS related to fraud prevention and detection.
• The ICS.

5.1 Core belief and boundary systems

Central to eBay’s belief system was its goal to “pioneer new communities” (eBay 10K,
2007; eBay, 2010a, 2010b, 2011) within which buyers and sellers could trade with
confidence in, and personal engagement with, one another (Boyd, 2002). An alternative
wording of this belief system is seen in Omidyar’s claim that the most important lesson
from eBay is, “[…] that 135 million people have learned they can trust a complete
stranger (Maney, 2005)”. This vision necessitated community creation among
participants which would be “[…] sustained by trust and inspired by opportunity”
(eBay, 2010a, 2010b, 2011, 2007). A key to creating the community of trust, and a viable
MCS, was Omidyar’s (1996) “founder’s letter” to the eBay community, which was posted
to eBay’s simple black-text-on-gray-background home page. It stated, “Most people are
honest. […] But some people are dishonest. Or deceptive. […] But here, those people
can’t hide. We’ll drive them away. Protect others from them”.
Similarly, eBay’s Code of Business Conduct and Ethics echoes the founder’s letter in
stating: “We believe people are basically good” (eBay, 2010a, 2010b). If trust comprises
three elements: benevolence, integrity and ability (Jeacle and Carter, 2011; Mayer et al.,
1995; Schoorman et al., 1996), then these documents surely contain two of these
elements: benevolence and integrity. These beliefs are consistent with Omidyar’s
libertarian, laissez-faire philosophy, within which controls impeded the full realization
of community self-governance (Cohen, 2002). Pragmatically, this philosophy meant that
Omidyar and eBay sought minimalist controls over users and transactions meaning
that, initially, there was no eBay boundary system beyond the threat that dishonest
people will be driven away. For example, before February 1996, Omidyar mediated
conflicts among users by sending them a personal e-mail stating, “You guys work it out”
(Cohen, 2002). This level of personal interaction and informality helped build trust in a
nascent system of a few hundred dedicated, “techie” users, who willingly tolerated
recurring system outages and an unfriendly and unattractive user interface. However,
such a system lacked a means of combating the large-scale frauds that would come with
a much larger, heterogeneous user base.
eBay’s early philosophy, i.e. to “not be corporate”, to avoid appearing to invoke
top-down management decisions for the community and to instead actively engage its
 community in its self-creation may partially explain eBay’s creation of a mythical Strategy, IT, and
founder’s legend: that eBay was founded to allow Omidyar’s then girlfriend (later wife)
to trade Pez candy dispensers (Cohen, 2002). The importance and success of this legend
control @ eBay,
is, well, legendary (Maney, 2005); even Omidyar acknowledges its critical role in 1995-2005
promoting the company (Lacy, 2010). According to eBay’s first publicist (Mary Lou
Song), journalists were uninterested in the story of a tireless French–Iranian immigrant
to the USA, who was hell-bent on creating a company that grew rapidly to make him and 367
his co-founders wealthy. Far more attractive was a cute story about a guy starting an
on-line hobby to please his girlfriend. The success of this legend is evident in its
propagation, including in many journalistic accounts of eBay, and in two business
school cases about eBay (Hoyt and Baron, 2001; Burnett and Schill, 1999).
Downloaded by UNIVERSITAET OSNABRUCK At 17:11 30 January 2016 (PT)

Although Omidyar’s then girlfriend did collect Pez dispensers (Lacy, 2010), the
creation and propagation of this canard is an example of an “ironical process” (cf.
Wegner, 2009) in relation to eBay’s belief system. That is, a company whose success, and
MCS, was founded on trusting “[…] in the kindness of strangers” and the eBay MCS and
management (Lewis, 2008, p. 76), deliberately falsifying a founder’s legend to publicize
their community that was purportedly “founded on trust”. Cynicism about this irony,
called by Boyd (2002), eBay’s “community/commerce tension”, was a recurring theme on
eBay’s chat boards (e.g. comment “If eBay’s a community, where’s the voting booth?”).
Such “ironical process” are not present in the LOC or LOCaIT models which focus
instead of management’s objectives for its MCS.
To summarize, eBay’s espoused belief system reflected a “communitarian” ideal
(Baron, 2002; Hoyt and Baron, 2001), within which self-interest was integrated with
community building, specifically including tenants that online buyers and sellers are
basically good and that one can create an engaged, active online community with a
structure and system that appeals to “[…] the better angels of our nature (Lincoln,
1861)”. Rules are implemented only if demanded by behavior that must be curtailed.
While growing the company, and achieving profitability, was eBay’s dominant strategy
and goal (eBay S-1, 1998), masking or disguising this goal required spinning a mythical
founder’s legend to “humanize” the company. Hence, eBay’s belief system included a
core irony and tension, i.e. while eBay’s dominant goal was growth, this goal had to be
masked in a “user-friendly” form within which eBay’s actions were community driven.
Consistent with Omidyar’s libertarian ideals, eBay’s boundary system was initially
minimalist and vague, with an exhortation to, “Above all, conduct yourself (sic) in a
professional manner. Deal with others the way you would have them deal with you”.
Omidyar’s initial boundary system, i.e. of saying “you guys work it out”, when problems
arose between users gave rise to a vigilante system in which six users, called “The
Posse” monitored listings and chat boards and quickly damaged the reputations, using
eBay’s feedback system, of any users they deemed to have acted against the
community’s best interests (Hoyt and Baron, 2001; Baron, 2002). The first evidence of an
eBay user agreement, indicating greater formalization of the boundary system, is dated
March 1998 (Hoyt and Baron, 2001). While the primary purpose of the (and perhaps all)
user agreement(s) would seem to be to release eBay from multiple legal duties,
obligations of users under this agreement are to complete transactions (whether as
bidder or seller), to not manipulate prices (e.g. through “shill” bidding) and to not
interfere with eBay’s system integrity, including its feedback system.
 QRAM Growth and expansion brought high-publicity fraud cases which required more rules
and formalization to the boundary system. eBay formed its Safe Harbor group, i.e.
11,4 in-house staff charged with investigating violations of community standards, in
February 1998 (Hill and Farkas, 2000). In 1999, following the successful IPO, eBay
expanded the Safe Harbor group, which began more aggressively pursuing fraud cases
and more actively monitoring for violations of eBay policy on allowable items for sale.
368 Several product bans occurred in 1999, including gun sales (February), body parts and
drugs (August) and alcohol and cigarettes (September). Prohibitions came to include
threatening others; the aforementioned shill bidding; and buying, selling or trading
feedback (i.e. exchanging positive feedback ratings for money or reciprocal actions with
no goods being sold) (eBay 10K filings, 1999/2007; Bunnell and Luecke, 2000; Cohen,
2002; Guadamuz González, 2003). However, evidence from knowledgeable users,
Downloaded by UNIVERSITAET OSNABRUCK At 17:11 30 January 2016 (PT)

including statements on the eBay user forums, indicated minimal enforcement of these
rules, except for the most egregious violators (Brunker, 2002; Walton, 2006). Hence, over
the period of this case, the eBay boundary system could perhaps best be characterized as
expanding the number of rules with lax, though increasing, levels of enforcement.

5.2 Core belief and boundary system – diffusion and codification

5.2.1 Diffusion That eBay sought to create a new role for users, in which the experience
of the company was directly a function of users’ treatment of one another, created a
special need to diffuse the belief and boundary systems throughout the eBay
community. These exceptional levels of diffusion to users, accomplished through eBay’s
use of its Web site and e-mail systems, are consistent with LOCaIT’s predictions that the
belief and boundary systems should be diffused throughout the organization. Perhaps
the single most innovative aspect of the eBay’s diffusion of its systems, including its
belief and boundary systems, was the intended recipient of these communications.
LOCaIT assumes that the desired MCS diffusion will be internal to the organization
(Simons, 1995b). In contrast, eBay, because of its unique business model, sought and
succeeded in diffusing its belief and boundary systems to all users, as users’ experience
of the company was contingent upon their experience of one another. Hence, eBay’s
diffusion was consistent with LOCaIT “throughout the organization” but this also
included diffusion to those outside of the organization.
5.2.2 Codification. The extent of codification of the eBay belief system was consistent
with LOCaIT, largely personalized through narratives and the persona of its founder.
The LOCaIT model suggests that boundary systems should seek a low level of
codification but be unambiguous. In contrast, the codification of the eBay boundary
system was relatively ambiguous. That is, while unambiguous rules existed, the
enforcement of these rules varied both with respect to the nature of the transgression
(e.g. selling illegal drugs vs. shill bidding) and when the transgression occurred, with
increasing enforcement of some rules over the studied period. Hence, the eBay boundary
system, in contrast to the LOCaIT model, seemed to exist as a set of unambiguous rules
with ambiguous and shifting levels of enforcement.

5.3 Boundary and DCS – disclaiming, preventing and detecting fraud

Wedded to eBay’s remarkable success was a parallel growth in online fraud and
deception that resulted from the unregulated nature of “hyper-real” online activity and
ecommerce (Baker, 2002). Despite a small volume of online auction transactions, online
 auctions accounted for 87 per cent of the online complaints to the National Consumers Strategy, IT, and
League in 1999 (National Consumers League, 2000). The Internet Crime Complaint
Center (ICCC), a partnership between the USA FBI and the White Collar Crime Center,
control @ eBay,
began tracking online crime in 2001. Between the ICCC’s 2014 inception and 2005, online 1995-2005
auction fraud complaints grew by over 500 per cent, from 21,576 to 133,380. Despite
eBay’s persistent claims that online auction fraud was rare (Cohen, 2002; Frei and
Rodriguez-Farrar, 2005; Walton, 2006; Warner, 2003), during the period of this case, 369
online auction fraud was the most frequent complaint filed with the ICCC and eBay
dominated the online auction market.
Most research assumes that the eBay market, and its MCS, was static (Baron, 2002;
Duh et al., 2002; Dellarocas, 2003a, 2005; Gu, 2007; Hoyt and Baron, 2001; Steiglitz, 2007).
In fact, the eBay MCS evolved in relation to emerging seller and buyer abuses and ICS
Downloaded by UNIVERSITAET OSNABRUCK At 17:11 30 January 2016 (PT)

feedback from users (Cohen, 2002; Bunnell and Luecke, 2000). As such, the design of the
eBay MCS is characteristic of managerial efforts to structure “unstructured” problems
(Mintzberg et al., 1976). Consistent with Mouritsen’s (1999) description of achieving
management control through the “virtual organization”, a balance, and tension, of
improvised action was evident in the creation and evolution of eBay’s DCS in relation to
preventing and detecting fraud. Despite Omidyar visionary claim that “[…] here, those
people can’t hide. We’ll drive them away […]”, eBay, in its earliest years, had few
resources for addressing undesirable members of the eBay community (Walton, 2006).
One pillar of eBay’s boundary and DCS was a disclaimer of responsibility for fraud, i.e.
that “eBay is only a venue”, and therefore not liable for fraud on the site. EBay
encouraged “[…] victims of fraud to report the incident to local law enforcement” (Hill
and Farkas, 2000, p. 12) and in its users’ agreement (quoted in Bunnell and Luecke, 2000,
p. 54) and 10K (Wired, 1998; Warner, 2003), disclaimed responsibility. This strategy
appeared to initiate with eBay’s October 1997 hiring of an attorney (Cohen, 2002). The
assertion of a lack of responsibility contradicted the traditional roles and responsibilities
of auctioneers (Albert, 2002; Baker, 2002) and outsourced, i.e. disclaimed, a formerly
important auctioneer responsibility. Importantly, USA courts consistently upheld
eBay’s disclaimer of responsibility for the sale of fraudulent products (Guadamuz
González, 2003).
The first well-publicized eBay fraud emerged in December 1998 when Sonny Stemple
“sold” $30,000 of undelivered merchandise (Wired, 1998; Cohen, 2002). The scale and
scope of Stemple’s fraud and perhaps, more importantly, its publicity, made clear that
Omidyar’s e-mails to disputants stating “you guys work it out” was an embarrassingly
inadequate response if eBay was to continue its meteoric growth. Following the
accelerating negative publicity from the Stemple and other eBay frauds, a further
improvisation occurred in January 1999 – eBay expanded and perhaps, more
importantly, publicized its earlier mentioned “Safe Harbor” group, which was then
charged with investigating fraud, and terminating (delisting) the most egregious
offenders of its policies (Hill and Farkas, 2000; eBay 10K 1999/2007; Shaughnessy, 2004).
Simultaneous improvisations, designed to address expanding fraud on eBay, included
adding a service for users to identify themselves (voluntarily), a minimalist insurance
program (maximum $200 loss coverage with a $25 deductible) for proven cases of fraud
and third-party escrow (Escrow.com, 2013) services (Hill and Farkas, 2000; eBay 10K,
1999/2007). Hence, eBay’s fraud-related improvisations, and its boundary and DCS
related to fraud, were a three-pronged strategy:
 QRAM (1) Officially disclaim responsibility for fraud.
11,4 (2) Refute publicly available evidence that fraud on eBay was widespread and
increasing.
(3) Improve fraud controls.

5.3.1 Diffusion and codification. The diffusion and codification of eBay’s boundary
370 system related to fraud is consistent with the previous discussion of its core boundary
system. However, the diffusion and codification of its DCS related to fraud contrasts
with that predicted by LOCaIT. Specifically, LOCaIT presents a high level of
codification in DCSs and diffusion to managers to highlight and report exceptions.
However, the eBay DCS related to fraud was an odd mix of a disclaimer of responsibility
Downloaded by UNIVERSITAET OSNABRUCK At 17:11 30 January 2016 (PT)

by eBay, threats to users and an evolving boundary system where the enforcement
criteria remained a secret known only to eBay. Hence, eBay’s DCS related to fraud would
seem to have assumed a complexity and characteristics that are inconsistent with
LOCaIT’s relatively simple predictions.

5.4 DCS – monitoring and feedback

EBay’s outsourcing strategy extended to the creation of perhaps the world’s most
imitated DCS. Monitoring in a MCS is sufficiently important that the Committee of
Sponsoring Organizations (COSO, 2009) recently issued a pronouncement specifically
intended to help organizations improve MCS system monitoring. Omidyar’s great DCS
insight, first disseminated in his founder’s letter (Figure 1), created the possibility for a
DCS scale and scope that exceeded all others (Dellarocas, 2003a, 2003b) and is now
widely imitated in online markets (Dellarocas, 2003a; Jeacle and Carter, 2011). eBay’s
feedback system effectively outsourced the monitoring of suppliers and customers to
[…] suppliers and customers (eBay, 1998). It includes the three essential elements of a
DCS:
(1) Measurement of outputs.
(2) Standards for evaluating outputs.
(3) The ability to correct deviations from standards (Simons, 1995a, 1995b).

While eBay’s DCS evolved considerably over its life, its initial and still primary,
performance variable, i.e. standard for evaluating buyers and sellers, was a simple
rating called the “Feedback score” (initially, the sum of a user’s positive minus negative
feedback). In its initial and simplest form, any account holder could leave eBay feedback
for anyone else, regardless of whether they had transacted business. Hence, early eBay
deception strategies included creating multiple accounts and leaving glowingly positive
feedback among the accounts and in one’s network of account-holding friends (Walton,
2006; Brown and Morgan, 2006; Steiglitz, 2007).
In February 1998, eBay’s Trust and Safety Department first used the feedback DCS
to monitor and discipline the most egregious violating traders (Cohen, 2002), but the
criteria by which users were identified for punishment was never disclosed (cf. Brivot
and Gendron, 2011). Beginning in March 2000, eBay restricted feedback to actual
transactions, which led to the creation of a new “market for deception” on eBay, i.e. the
proliferation of trivial transactions (usually with a sales price of $0.01) executed solely to
generate positive feedback for participating buyers and sellers (Brown and Morgan,
 Strategy, IT, and
control @ eBay,
1995-2005

371
Downloaded by UNIVERSITAET OSNABRUCK At 17:11 30 January 2016 (PT)

Figure 1.

2006). One implication of this “market for deception” is that users perceived eBay
feedback scores as sufficiently valuable that a secondary market developed for falsely
inflating them.

5.5 DCS – monitoring and feedback – codification and diffusion

5.5.1 Diffusion. Unlike its predecessors, eBay’s feedback DCS was not primarily
intended as a means for the company to monitor its employees, suppliers or customers
 QRAM (cf. Caglio and Ditillo, 2008; Malmi and Brown, 2008). Instead, the eBay MCS enabled
customers to monitor suppliers and suppliers to monitor customers. Hence, although the
11,4 eBay MCS included the standard expected elements of a DCS (i.e. measures of outputs,
standards and means to correct deviations) it was, Zen-like, a DCS without a
“controller”, a system designed primarily for use by users and not its owner and creator
(cf. Brivot and Gendron, 2011). In this sense, the eBay DCS blurred the boundaries
372 between internal and external reporting, by creating a neo-external reporting system
focused on the self-reporting of customer and supplier performance. It recast the
traditional DCS roles to expand beyond the organization to allow external parties to
monitor one another. Hence, the eBay MCS was an innovative blending of market
coordination, as buyers and sellers, using the system, chose one another for market
transactions, with clan coordination, as the system heavily relied on users trusting one
Downloaded by UNIVERSITAET OSNABRUCK At 17:11 30 January 2016 (PT)

another and the system (Mouritsen and Thrane, 2006; cf. Giddens, 1991) and adopting,
“[…] common values and beliefs” (Hakansson and Lind, 2004, p. 55; Ouchi, 1980).
5.5.2 Codification. eBay’s use of “symbolic tokens” illustrates one process of
codifying its feedback. “Symbolic tokens” are “[…] media of exchange which have
standard value, and thus are interchangeable across a plurality of contexts” (Giddens,
1991, p. 18; Boyd, 2002; Jeacle and Carter, 2011). EBay’s feedback ratings are a
standardized metric for evaluating seller performance (and buyer performance until
February 2008); eBay feedback ratings range from a minimum of - 2, for a
nonperforming seller whom eBay will delist to a 2012 maximum observed value of over
2.2 million (October 4, 2012; eBay, 2012). In 1999, eBay added an additional symbolic
token – colored “stars”, which were described as follows:
Users who have developed positive reputations over time will have a star symbol displayed
next to their user name, which is color coded to indicate the amount of positive feedback as
compared to negative feedback received by the user (eBay 10K, 1999/2007).
Hence, eBay stars provided an intuitive, visual indicator of buyer and seller performance
and means of disseminating and codifying the eBay feedback system to the masses.
To summarize, the eBay feedback DCS system was diffused beyond the internal
organizational boundary limits envisioned in LOCaIT. It was, instead, a process for
creating an online market, rather than a system to enable a company to monitor its
customers and suppliers. Although eBay did, in fact, monitor customers and suppliers,
and take disciplinary action under certain (vaguely specified) conditions, the primary
purpose of the eBay feedback DCS was to diffuse the DCS to enable customers and
suppliers to monitor one another. The eBay DCS did follow the LOCaIT model
predictions of a highly codified feedback system, which used technology, including
using the emerging capability of Internet graphics, to increase the level of codification of
its feedback DCS.

5.6 Interactive control systems

Improvisation was also evident in the development of eBay’s ICS which created a
mechanism for user comments to guide site features and development (Frei and
Rodriguez-Farrar, 2005; Bunnell and Luecke, 2000; Farkas and Hill, 2001). At inception,
the eBay Web site was unattractive, difficult to navigate and unreliable (Hoyt and
Baron, 2001); early users expected and accepted downtime (Cohen, 2002) and used it to
correspond with one another, at first, by e-mail, and later, on eBay chat rooms, which
 often remained accessible even when the transaction processing system failed. In the Strategy, IT, and
first six months of the company, Omidyar improvised changes to the system in response
to user e-mail comments, in the evening after finishing his day job. User e-mail evolved
control @ eBay,
into chat rooms and discussion boards (in February 1996), which, with development and 1995-2005
expansion, were monitored by eBay community relations staff (Cohen, 2002), thereby
meeting the criteria of a technology-enabled ICS dialog (Simons, 1995a, 1995b) between
eBay managers and users. For example, when Mary Lou Song, eBay’s public relations 373
department, posted proposed colors for a new system of feedback stars, based on
performance, on an eBay bulletin board, the response was immediate and negative,
including, “Do you know how this place works?” (Frei and Rodriguez-Farrar, 2005;
Cohen, 2002).
The eBay ICS also included, as predicted by the LOC model, a more traditional
Downloaded by UNIVERSITAET OSNABRUCK At 17:11 30 January 2016 (PT)

face-to-face interaction process among high-level eBay management. This process

originated with the hiring of Whitman (1998):
[…] to prepare her management team to be able to make decisions in ‘Internet time’, Whitman
mandated frequent and lengthy (two to four hour) team meetings right from the start (Hill and
Farkas, 2000, p. 6).
In addition, Meg Whitman repeatedly sought to create innovative measures to quantify
and summarize relevant activity on eBay.
5.6.1. Diffusion and codification. The diffusion of the eBay ICS, in the form of its chat
rooms and bulletin boards, mirrored the extended reach of its feedback DCS, i.e. a focus
on bringing customers and suppliers into the process of changing some of the operating
processes and directions of the company. Consistent with LOCaIT, the level of
modification of the ICS was moderate and evolving to respond to user needs. For
example, with an increasing user and available product base, eBay expanded the
number of product categories, partially based on user suggestions (Cohen, 2002; Lewis,
2008). Similarly, it revised the structure and content of its bulletin boards and help
facilities, based on user feedback.

6. Summary, limitations and conclusions

6.1 Case summary
EBay’s emergence as the dominant online auction facilitator affords the opportunity to
investigate the growth of a market facilitator’s MCS in a largely unregulated, online
market, with a large, global clientele of small business and individual traders. One
organization, i.e. eBay, almost single-handedly created this market, and invented a now
widely copied MCS for it, which redefined the nature and limits of MCS monitoring.
While journalistic and academic accounts discuss aspects of how eBay “changed the
world”, existing accounts do not describe the process of creating an MCS to support
eBay’s business strategy. This is the purpose of the present paper. This case adapts the
Simons’ LOCaIT framework to investigate the role of the eBay MCS in its successful
“changing of the world”.
At inception, the eBay MCS was unique and “capitalized” on emerging technologies,
changing social relations and a reconceptualization of the meaning and extent of, and
need for, privacy. Over time, eBay evolved into an odd electronic “flea market” within
which professional and amateur sellers offered an unprecedented variety of goods and
buyers and sellers never met but could exchange unlimited numbers of costless text
 QRAM messages with one another. The eBay MCS provided buyers and sellers with
information about the trustworthiness and reliability of buyers and sellers but beyond
11,4 the policing of the most egregious offenders of its rules, conducted the market consistent
with its founder’s laissez faire philosophy. Within Ouchi’s (1980) characterization of
three means of achieving organizational control, i.e. markets, bureaucracies and clans,
eBay, initially was a clever combination of market and clan control. While the extent of
374 eBay’s rules, monitoring and enforcement substantially increased over the period of this
case, the data imply that bureaucratic controls were seemingly less important to the
eBay MCS than its market and clan controls. The major exception to eBay’s limited use
of bureaucratic control was its aggressive actions against a small number of
high-profile, highly visible frauds (Walton, 2006).
Downloaded by UNIVERSITAET OSNABRUCK At 17:11 30 January 2016 (PT)

6.2 Implications
Evidence suggests that the eBay MCS was rooted in its core strategies but largely
advanced by improvisation (Weick, 1998), instead of careful and deliberate strategic
planning (Bunnell and Luecke, 2000; Cohen, 2002). Accordingly, we chose a theory
(LOCaIT) and methods (qualitative) that were adapted to an eclectic, opportunistic
strategy and dataset.
We apply Simons’ (1995a, 1995b) LOC as organizing constructs for considering the
structure and evolution of the eBay control system. While these constructs helped
conceptualize eBay’s MCS innovations, they required reconceptualization to capture the
unprecedented nature of the eBay MCS. Specifically, Simons’s model assumes that
management seeks to structure controls around a stable, small number of vendors and
customers. In contrast, eBay’s business created a global consumer market of perhaps the
largest cohort of vendors and customers ever assembled. Hence, the construct of a
“manager” in Simons’ model has a different meaning in a MCS with over 100 million
users (eBay 10K, 2011) most of whom give and receive performance feedback about one
another, and many of whom have multiple user identities, meaning that even identifying
the number of actual system users (managers?) is problematic (Steiner, 2009). Hence, the
nature of eBay’s control, in some cases, tugs at the limits, and demands extension of the
constructs of the Simons’ taxonomy.
The LOC and LOCaIT models fail to capture the irony of some aspects of the relations
between eBay’s strategy and its belief system, for example, “don’t be corporate but
grow, grow, grow!”). Perhaps this failure is because LOC and LOCaIT were based on
Simons’ interviews with C-suite executives who may be unaware of, or unwilling to
disclose, the ironies (i.e. inherent contradictions) of their business strategies. In addition,
the LOCaIT model serves as a useful guide in many predicting aspects of the eBay MCS.
For example, the model accurately depicts the levels of codification in the eBay MCS
regarding its belief, DCS and ICS, and a high level of diffusion of the belief system.
LOCaIT is less successful in predicting eBay’s boundary system, which was an eclectic,
ambiguous mix of threats, disclaimers and enforcement, about which eBay provided
little insight to those who were potentially subject to it. In addition, LOCaIT does not
anticipate that eBay’s business strategy was impossible without strong IT capabilities,
and the extent of necessary diffusion of its systems to all eBay users. An alternative
theoretical frame for exploring the intersection of MCS and IT, successfully adopted by
Jeacle and Carter (2011) is based in Gidden’s (1990, 1991) analysis of abstract systems
and Miller’s (2001) exploration of calculative practices.
 The eBay MCS included several ironical processes, i.e. when the espoused eBay MCS Strategy, IT, and
was at variance with the stated eBay MCS boundary rules. For example, eBay’s belief
and boundary system espoused community honesty and a commitment to discipline
control @ eBay,
deceptive traders. However, eBay’s founding included and promoted a falsified myth, 1995-2005
failed to act against “mundane” deceptive traders and seemingly increased and
implemented controls primarily in response to bad publicity (Walton, 2006). Such
ironical processes are also found in other research investigating control systems in large 375
organizations. For example, Alvesson and Kärreman, (2004) investigated management
controls at a large management consulting firm. Investigation revealed widespread
under-reporting of time worked on engagements (called “ghosting”), meaning that the
actual hours worked on an engagement were never known or formally recorded in the
system. Hence, while the control system of the consulting firm purported to accurately
Downloaded by UNIVERSITAET OSNABRUCK At 17:11 30 January 2016 (PT)

record engagement hours worked, all who worked for the firm recognized these hours as
under-reported, i.e. a necessary fiction. We find an analogous complex strategy, or
“fiction”, in the eBay MCS. Hence, certain MCSs, such as eBay’s, may necessarily imbed
elements of irony and hypocrisy to achieve and balance conflicting control objectives (cf.
Mundy, 2010).

6.3 Conclusions
In 2005, eBay for the first time, exceeded $1 billion in revenue for each of its quarterly
reporting periods (eBay 10K, 1999/2007). As of December 2005, eBay had over 11,000
employees only about one-half of whom were in the USA, and, anticipated substantial
additional growth both in the USA and overseas, through additional acquisitions and
expansion and by aggressively expanding its presence as a facilitator in the
business-to-business ecommerce market. Because of strong loyalty in its user base,
eBay’s substantial fee increases to sellers in 2003 and 2004 increased, rather than
decreased, its reported revenue. Hence, the state of eBay’s business, as of 2005, remained
promising, despite a net stock price decline in calendar year 2005.
The online auction markets make it possible for Americans to collect rare, antique
Russian nesting dolls, for Africans to buy newly released Italian designer clothing and
for Brazilians to savor top-grade Iranian caviar. eBay’s MCS, created to support this
market, was a clever mix of public relations, controls, disclaimers and denials. The
present analysis suggests that the eBay MCS was based on a core business strategy that
sought (and achieved) extraordinary growth, required strong IT capabilities and fueled
considerable improvisation to achieve the seemingly contradictory goals of high growth
and an aversion to restricting user behavior. That the result was a success is evident in
the extent to which eBay’s MCS is now the standard online market DCS (e.g. Yelp,
TripAdvisor and Amazon).

Note
1. The original list included six strategies. We combine similar elements to collapse the list to
four strategies.

References
Albert, M.R. (2002), “E-buyer beware: why online auction fraud should be regulated”, American
Business Law Journal, Vol. 39 No. 4, pp. 575-643.
 QRAM Alvesson, M. and Kärreman, D. (2004), “Interfaces of control. technocratic and socio-ideological
control in a global management consultancy firm”, Accounting, Organizations and Society,
11,4 Vol. 29 Nos 3/4, pp. 423-444.
Baker, C.R. (2002), “Crime, fraud and deceit on the internet: is there hyperreality in cyberspace?”,
Critical Perspectives on Accounting, Vol. 13 No. 1, pp. 1-15.
Baron, D.P. (2002), “Private ordering on the internet: the eBay community of traders”, Business
376 and Politics, Vol. 4 No. 3, pp. 245-274.
Boyd, J. (2002), “In community we trust: online security communication at eBay”, Journal of
Computer Mediated Communication, Vol. 7 No. 3, p. 23.
Brivot, M. and Gendron, Y. (2011), “Beyond panopticism: on the ramifications of surveillance in a
contemporary professional setting”, Accounting, Organizations and Society, Vol. 36 No. 3,
pp. 135-155.
Downloaded by UNIVERSITAET OSNABRUCK At 17:11 30 January 2016 (PT)

Brown, J. and Morgan, J. (2006), “Reputation in online auctions: the market for trust”, California
Management Review, Vol. 49 No. 1, pp. 61-81.
Brunker, M. (2002), “Evidence undercuts eBay’s tough talk on fraud: wrongdoing goes
unpunished, MSNBC investigation finds”, available at: www.msnbc.msn.com/id/3078736/
(accessed 25 October 2008).
Bunnell, D. and Luecke, R. (2000), The eBay Phenomenon: Business Secrets Behind The World’s
Hottest Internet Company, Wiley, New York, NY.
Burnett, J. and Schill, M.J. (1999), eBay Inc. (A), University of Virginia, Darden Business
Publishing, Charlottesville, VA, available at: https://store.darden.virginia.edu/
Caglio, A. and Ditillo, A. (2008), “A review and discussion of management control in inter-firm
relationships: achievements and future directions”, Accounting Organizations and Society,
Vol. 33 Nos 7/8, pp. 865-898.
Cohen, A. (2002), The Perfect Store: Inside eBay, 1st ed., Little, Brown and Co, Boston, MA.
Committee of Sponsoring Organizations (COSO) (2009), Guidance on Monitoring Internal Control
Systems, AICPA, New York, NY.
Dellarocas, C. (2003a), “The digitization of word of mouth: promise and challenges of online
feedback mechanisms”, Management Science, Vol. 49, No. 10, pp. 1407-1424.
Dellarocas, C. (2003b), “Building trust online: the design of robust reputation reporting
mechanisms in online trading communties” Doukidis, G., Mylonopoulos, N. and Pouloudi,
N, Information Society or Information Economy? A combined perspective on the digital era,
Idea Book Publishing, Russia.
Duh, R.-R., Jamal, K. and Sunder, S. (2002), “Control and assurance in e-commerce: privacy,
integrity, and security at eBay”, Taiwan Accounting Review, Vol. 3 No. 1, pp. 1-27.
eBay (1998), “Form S-1 SEC registration statement, July 15”, available at: http://
investor.ebayinc.com/sec.cfm?DocType⫽&Year⫽1998 (accessed 22 October 2013).
eBay (1999/2007), “10-K annual reports”, available at: www.sec.gov/edgar.shtml (accessed 22
October 2013).
eBay (2008), “Discussion boards: index”, available at: http://pages.ebay.com/community/boards/
index.html (accessed 11 October 2008).
eBay (2010a), “Code of business conduct and ethics”, available at: http://investor.ebayinc.com/
documentdisplay.cfm?DocumentID⫽649 (accessed 22 January 2010).
eBay (2010b), “All listings (including item counts)”, available at: http://listings.ebay.com
(accessed 11 May 2010).
eBay (2011), “What we believe”, available at: www.ebayinc.com/values (accessed 1 March 2011).
 Eisenhardt, K.M. and Graebner, M.E. (2007), “Theory building from cases: opportunities and Strategy, IT, and
challenges”, Academy of Management Journal, Vol. 50 No. 1, pp. 25-32.
Escrow.com (2013), “About us”, available at: https://escrow.com/why-escrowcom/about-us.aspx
control @ eBay,
(accessed 22 October 2013). 1995-2005
Farkas, M.T. and. Hill, L.A. (2001), Meg Whitman and eBay Germany, Harvard Business Press,
Boston, MA.
Frei, F.X. and Rodriguez-Farrar, H. (2005), eBay (A): The Customer Marketplace, Harvard 377
Business Press, Boston, MA.
Gatewood, R.D. and Carroll, A.B. (1991), “Assessment of ethical performance of organization
members - a conceptual framework”, Academy of Management Review, Vol. 16 No. 4,
pp. 667-690.
Downloaded by UNIVERSITAET OSNABRUCK At 17:11 30 January 2016 (PT)

Giddens, A. (1990), The Consequences of Modernity, Polity Press in association with Basil
Blackwell, Cambridge, Oxford.
Giddens, A. (1991), Modernity and Self-Identity: Self and Society in the Late Modern Age, Stanford
University Press, Stanford, CA.
Gilbert, S. (2012), The Story of eBay, Creative Education, Mankato, MN.
Gu, J. (2007), Consumer Rights Protection on the Online Auction Website — Situations and
Solutions: A Case Study of eBay, British and Irish Law, Education, and Technology
Association, Hertfordshire.
Guadamuz González, A. (2003), “Ebay law: the legal implications of the C2C electronic commerce
model”, Computer Law & Security Review, Vol. 19 No. 6, pp. 468-473.
Hakansson, H. and Lind, J. (2004), “Accounting and network coordination”, Accounting
Organizations and Society, Vol. 29 No. 1, pp. 51-72.
Hasker, K. and Sickles, R. (2010), “EBay in the economic literature: analysis of an auction
marketplace”, Review of Industrial Organization, Vol. 37 No. 1, pp. 3-42.
Hawkins, D. and Cohen, J. (2003), “eBay Inc.: internet success or fairy tale?”, Harvard Business
School Case, November.
Helft (2001), “What makes eBay unstoppable?” Industry Standard, 6 August, pp. 32-37.
Hill, L.A. and Farkas, M.T. (2000), Meg Whitman at eBay, Inc. (A), Harvard Business Press,
Boston, MA.
Hoyt, D. and Baron, D. (2001), eBay: Private Ordering for an Online Community, Stanford
Business School Press, Palo Alto, CA.
Internet Crime Complaint Center (ICCC) (2014), available at: www.ic3.gov/default.aspx (accessed
31 October 2014).
Jayanti, R.K. (2010), “A netnographic exploration listening to online consumer conversations”,
Journal of Advertising Research, Vol. 50 No. 2, pp. 181-196.
Jeacle, I. and Carter, C. (2011), “In TripAdvisor we trust: rankings, calculative regimes and
abstract systems”, Accounting, Organizations and Society, Vol. 36 Nos 4/5, pp. 293-309.
Kenny, M. (2008), “Dogpuss: everything eBay reviewed!”, available at: http://dogpuss.com/
(accessed 11 November 2008).
Ketokivi, M. and Mantere, S. (2010), “Two strategies for inductive reasoning in organizational
research”, Academy of Management Review, Vol. 35 No. 2, pp. 315-333.
Klink, E.E. and Klink, S.G. (2005), Dawn of the eBay Deadbeats: True Tales of Treachery, Lies, and
Fraud from the Dark Recesses of the World’s Largest Online Auction Marketplace,
Mooncusser Media LLC, Allendale, NJ.
 QRAM Lacy, S. (2010), “Pierre Omidyar on eBay and Pez dispensers, leaving the valley and the most
important thing he’s ever done”, available at: http://techcrunch.com/2010/03/20/
11,4 pierre-omidyar-on-ebay-and-pez-dispensers-leaving-the-valley-and-the-most-important-
thing-he%E2%80%99s-ever-done/ (accessed 22 October 2013).
Lewis, E. (2008), The eBay Phenomenon, Marshall Cavendish Business, Chatham, Kent.
Lincoln, A. (1861), “First inaugural address”, available at: www.bartleby.com/124/pres31.html
378 (accessed 22 October 2013).
Malmi, T. and Brown, D.A. (2008), Management control systems as a package-opportunities,
challenges and research directions, Management Accounting Research, Vol. 19 No. 4,
pp. 287-300.
Malsch, B. and Gendron, Y. (2013), “Re-theorizing change: institutional experimentation and the
Downloaded by UNIVERSITAET OSNABRUCK At 17:11 30 January 2016 (PT)

struggle for domination in the field of public accounting”, Journal of Management Studies,
Vol. 50 No. 5, pp. 870-899.
Maney, K. (2005), “10 years ago, eBay changed the world”, USAToday, 22 March, available at:
http://usatoday30.usatoday.com/tech/news/2005-03-21-ebay-cover_x.htm (accessed 22
October 2013).
Mayer, R.C., Davis, J.H. and Schoorman, F.D. (1995), “An integrative model of organizational
trust”, Academy of Management Review, Vol. 20 No. 3, pp. 709-734.
Miles, M.B. and Huberman, A.M. (1994), Qualitative Data Analysis, Sage, Thousand Oaks, CA.
Miller, P. (2001), “Governing by numbers: why calculative practices matter”, Social Research,
Vol. 68 No. 2, pp. 379-396.
Mintzberg, H., Raisinghani, D. and Theoret, A. (1976), “Structure of unstructured
decision-processes”, Administrative Science Quarterly, Vol. 21 No. 2, pp. 246-275.
Mouritsen, J. (1999), “The flexible firm: strategies for a subcontractor’s management control”,
Accounting, Organizations and Society, Vol. 24 No. 1, pp. 31-55.
Mouritsen, J. and Thrane, S. (2006), “Accounting, network complementarities and the
development of inter-organisational relations”, Accounting, Organizations and Society,
Vol. 31 No. 3, pp. 241-275.
Mundy, J. (2010), “Creating dynamic tensions through a balanced use of management control
systems”, Accounting Organizations and Society, Vol. 35 No. 5, pp. 499-523.
National Consumers League (2000), “National consumers league warns consumers millions are
lost to internet fraud”, available at: www.fraud.org/internet/99final.htm
Omidyar, P. (1996), “Founders letter”, available at: http://pages.ebay.com/services/forum/
feedback-foundersnote.html (accessed 24 November 2009).
Ouchi, W. (1980), “Markets, bureaucracies, and clans”, Administrative Science Quarterly, Vol. 25
No. 1, pp. 129-141.
Rokka, J. (2010), “Netnographic inquiry and new translocal sites of the social”, International
Journal of Consumer Studies, Vol. 34 No. 4, pp. 381-387.
Schoorman, F.D., Mayer, R.C. and Davis, J.H. (1996), “Organizational trust: philosophical
perspectives and conceptual definitions”, Academy of Management Review, Vol. 21 No. 2,
pp. 337-340.
Shaughnessy, B. (2004), “Security Q&A with Rob Chesnut, eBay’s Trust & Safety VP”, available
at: www.auctionbytes.com/cab/abn/y04/m06/i26/s03 (accessed 25 May 2010).
Simons, R. (1987), “Accounting control systems and business strategy: an empirical analysis”,
Accounting, Organizations and Society, Vol. 12 No. 4, pp. 357-374.
 Simons, R. (1990), “The role of management control-systems in creating competitive advantage - Strategy, IT, and
new perspectives” Accounting Organizations and Society, Vol. 15 Nos 1/2, pp. 127-143.
Simons, R. (1991), “Strategic orientation and top management attention to control-systems”,
control @ eBay,
Strategic Management Journal, Vol. 12 No. 1, pp. 49-62. 1995-2005
Simons, R. (1992), “The strategy of control: how accounting information helps to formulate and
implement business strategy”, CA Magazine, pp. 44-50.
Simons, R. (1994), “How new top managers use control systems as levers of strategic renewal”, 379
Strategic Management Journal, Vol. 15 No. 3, pp. 169-189.
Simons, R. (1995a), “Control in the age of empowerment”, Harvard Business Review, March-April,
pp. 80-88.
Simons, R. (1995b), Levers of Control : How Managers Use Innovative Control Systems to Drive
Downloaded by UNIVERSITAET OSNABRUCK At 17:11 30 January 2016 (PT)

Strategic Renewal, Harvard Business School Press, Boston, MA.

Skogøy, K.O. (2010), “Online auction sites gives the smartest marketing and the highest profits
possible”, available at: www.anesite.com/onlineauction.html (accessed 22 October 2013).
Steiglitz, K. (2007), Snipers, Shills, & Sharks: EBay and Human Behavior, Princeton University
Press, Princeton, NJ.
Steiner, I. (2009), “eBay number of registered users inaccurate?”, available at: www.
ecommercebytes.com/C/letters/blog.pl?/pl/2009/10/1255904186.html (accessed 22 October
2013).
Vendio Services Inc (2008), “Discussion groups”, available at: www.vendio.com/mesg/ (accessed
11 October 2008).
Walton, K. (2006), Fake: Forgery, Lies & eBay, Simon Spotlight Entertainment, New York, NY.
Warner, M. (2003), “eBay’s worst nightmare”, Fortune, available at: http://money.cnn.
com/magazines/fortune/fortune_archive/2003/05/26/343106/ (accessed 22 October 2013).
Wegner, D.M. (2009), “How to think, say, or do precisely the worst thing for any occasion”, Science,
Vol. 325 No. 5936, pp. 48-50.
Weick, K.E. (1998), “Improvisation as a mindset for organizational analysis”, Organization
Science, Vol. 9 No. 5, pp. 543-555.
Wired (1998), “EBay suspends fraud suspect”, Wired, available at: www.wired.com/
techbiz/media/news/1998/12/16797 (accessed 22 October 2013).

Further reading
Dellarocas, C. (2005), “Reputation mechanism design in online trading environments with pure
moral hazard”, Information Systems Research, Vol. 16 No. 2, pp. 209-230.
eBay (2004), “eBay workshop: eBay history 101”, available at: http://workshops.ebay.
com/thread.jspa?threadID⫽410122850
eBay (2012), “Everyday source seller profile”, available at: http://members.ebay.com/
ws/eBayISAPI.dll?ViewUserPage&userid⫽everydaysource (accessed 22 October 2013).
E-commerce Land (2004), “History of E-commerce”, available at: www.ecommerce-land.com/
history_ecommerce.html (accessed 22 October 2013).

Corresponding author
Dan N. Stone can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected]

Or visit our web site for further details: www.emeraldinsight.com/reprints