Intermediation in Five Inclusive States:
Estimates of Financial Performance and Growth Strategies
Jonathan Wareham
Rich Klein
Dept. of Computer Information Systems
J. Mack Robinson College of Business
Georgia State University
Atlanta, GA 30302-4015
Tel: 404 463-9293
Fax:404 651-3842
wareham@acm.org
rklein@gsu.edu
September 29, 2001
Intermediation in Five Inclusive States:
Estimates of Financial Performance and Growth Strategies
ABSTRACT
The theoretical literature concerning commercial intermediation is
surveyed to develop a framework of five, inclusive intermediary states.
Data collected on 58 electronic healthcare portals are used to test
hypotheses concerning intermediation states, growth strategies and firm
profitability, in a sample representing both traditional and electronic
intermediaries. The study finds a positive relationship between higher
states of intermediation and profitability, as well as finding a higher
frequency of acquisitions as growth strategies in higher intermediary
states.
1.0 Introduction
The topic of electronic intermediation has received a significant amount of attention since the
widespread acceptance of the Internet. The seminal work of Malone, Yates and Benjamin (1987)
proposed that reductions in coordination costs would permit a greater use of market
mechanisms as predicted by transaction costs economics. This paper also suggested that the
movement toward electronic markets would eliminate traditional middlemen in the supply
chain. Hence, the concept of disintermediation has attracted considerable interest from the
academic community (Bakos, 1991; Benjamin and Wigand, 1995; Tapscot, 1996), in addition to
its wide-spread popularity as an underlying business philosophy for the wave of dot-com
business start-ups that were actively funded until the April 2000 fall in the NASDAQ stock
market. However, this research seeks to gain greater understanding of the fundamental role of
electronic intermediaries in contributing toward firm performance, i.e., profitability, as well as
for how firms should expect their role as an intermediary to develop over time.
Brousseau (1999), following Spulber (1996, 1999), delineates five distinct functions, or
states of intermediation, specifically: information management, transaction securitization,
liquidity, logistics management, and insurance. Using this typology, this research first seeks to
understand how these intermediation strategies relate to profitability. Given the assumption
Page 1
that these five stages are both hierarchical and inclusive, that is, firms at higher or later
intermediation stages possess the capabilities available at the lower or earlier levels, the second
focus of this research is an examination of how intermediaries evolve through time. In an effort
to address these two issues, this paper is organized as follows:
After a brief review of prior intermediation-focused research, Section 2 introduces the
healthcare industry and electronic healthcare portals as a focal point for studying electronic
intermediation. In Section 3, the concepts of commercial intermediation and corporate
diversification are discussed, the research hypotheses are formulated, and the choice of
dependent and independent variables motivated. Section 4 presents the research method
employed in data collection and analysis, reviews the estimation methods, and presents the
relevant findings. Section 5 contrasts these findings with the industry level observations and
evaluates the generalizability the results for other industry sectors, as well as exploring
implications for theory and practice.
1.1 Previous Research
The concept of intermediation has generated interest in both academic and business
communities. This idea was, paradoxically, raised in the Malone, Yates and Benjamin (op. cit.)
paper, where the movement to electronic markets was seen as facilitating new opportunities for
firms to intermediate as market providers. The expansion of novel business models such as
Amazon and eBay ushered in terms such as “cybermediaries” (Sakar, Butler and Steinfeld, 1995)
or “infomediaries” (Hagel and Singer, 1999). Underlying these terms is the idea that third-party
entities can serve to aggregate, filter, and profile information in a novel manner and use this
information to coordinate sales and supply chains to extract economic rents. Other scholars,
such as Baily and Bakos (1997), argued that new roles for electronic intermediaries included
roles such as trust provision, information aggregation, information exchange and brokering,
where Bakos (1998) argues that electronic markets require intermediaries for matching,
transaction facilitation and mitigation of moral hazard. Chircu and Kaufman (2000) provide an
in-depth review of this literature.
Page 2
However, the electronic markets hypothesis has found limited or ambiguous support in
empirical studies (Hess and Kemer, 1994; Hitt 1999), and because of the fact that few Internetdriven businesses based on pure info-mediation have demonstrated long term viability, there is
now a general consensus that the electronic versus traditional markets dichotomy has less
theoretical or empirical utility than originally believed. While we do not discount the
significance of electronic channels, the low nominal frequency with which such purely
electronic businesses exist does not warrant their continued segregation in theoretical and
empirical studies. In fact, as our study will indicate, many of the enterprises construed as
eCommerce infomediaries have, either in origin or through evolution, assumed the roles that
have traditionally been found within all forms of commercial intermediation. Hence, this study
is posited on the idea that intermediaries exist, the majority of which assume a combination of
electronic and traditional channels. Many purely electronic intermediaries have assumed
traditional intermediary functions through organic growth, joint ventures or acquisition,
whereas many traditional intermediaries have developed electronic channel capabilities in
parallel. In fact, it is the co-evolutionary paths of both intermediary types that are a focus of this
study.
Accordingly, we turn to an industry that is very information intensive and has
experienced a great deal of growth in both purely electronic as well as traditional intermediary
forms – healthcare intermediaries. After a brief survey of recent developments in the North
American healthcare market, we thereafter present a review of commercial intermediation
theory and formulate hypotheses concerning intermediary growth strategies, financial
performance and firm age. These hypotheses center around questions of how intermediation
strategies relate to profitability, and, secondly, how do intermediaries evolve through time. The
hypotheses are tested with data collected from 58 individual firms.
2.0 The State of Healthcare in the Year 2001
The U.S. healthcare industry has recently been marked by ever increasing numbers of capitated
plans, decreases in fees for service plans, declines in the staff-model, reductions in provider
Page 3
reimbursements and growing emphasis on reducing the overall costs of delivering care. More
recently, managed care organizations have been faced with dissatisfied consumers who are
demanding more open access to provider networks, driven in part by the explosion of Internet
usage and growing availability of medical content from portals like WebMD and DrKoop.com.
Consider that in 1999, 41 percent of U.S. adults, or some 74 million people, logged on to the
Internet, with 54 percent, or some 40 million people, doing so for health or medical reasons
(Weber, 1999).
In response to this consumer demand, many managed care organizations are raising
premiums in order to finance these open networks that allow for less control and contracting,
which have been the traditional sources of cost reductions for these organizations. Not
surprisingly, these increases have been met with backlash from businesses and the Federal
government. Healthcare costs are expected to rise from 14 percent of gross national product to
17 percent, or $2.1 trillion, by 2007 (Duncan, 1999). Given increasing globalization, businesses
are finding it harder to compete with European firms that are already enjoying significantly
lower healthcare costs.
U.S. managed care organizations, and healthcare providers in general, are being forced
to: (1) increase the level and quality of service afforded to consumers, and (2) increase their
operating efficiencies to achieve cost reductions. Virtually all major healthcare organizations are
investing heavily in reducing the wait-time for patients, improving billing processes and
accuracy, improving scheduling efficiency and trying to demonstrate clinical quality. All of
these initiatives are directly aimed at improving the level of service afforded to members or
patients, and have the added benefit of contributing to cost reductions. Managed care
organizations grew over the 1990s, achieving dramatic cost efficiencies through tight controls
and competitive contracting with care providers. Since these organizations are unable to raise
prices when faced with ever increasing pressure from both the private and public sectors, the
challenge is now to capitalize on the cost-saving potential of industry-wide standardization and
increased adoption of electronic data interchange opportunities through e-commerce.
As the IT focus in the healthcare industry, like many others, shifts away from Year 2000
compliance, the Health Insurance Portability and Accountability Act (HIPAA) sets forth new
Page 4
mandates for IT professionals in this field. HIPAA provides mandates for the standardization of
administrative and financial transactions and code sets within the healthcare industry. In many
ways these mandates will be the impetuous for change in an industry that has been traditionally
slow to adopt technological innovations (Raghupathi, 1997). Opportunities for information
systems providers in the healthcare industry are tremendous as organizations struggle to meet
HIPAA requirements by 2003, with the electronic commerce initiative and the Internet at the
forefront.
In general, the healthcare industry offers an optimal industry segment for a study in
intermediation in that it is: 1) highly information intensive; 2) has experienced significant
growth in both electronic and traditional intermediaries; 3) offers a broad, yet cohesive,
portfolio of intermediary functions that can be identified, mapped and transcribed to
commercial intermediation theory independent of the sector. Hence, after first discussing
representative cases of electronic healthcare portals, we will turn to commercial intermediation
theory and develop hypotheses concerning intermediation state, growth strategies and
profitability.
2.1 Electronic Healthcare Portals
In early December of 1999, healthcare giant McKesson and its software subsidiary,
McKesson/HBOC, announced two initiatives toward establishing the firm as a market leader in
the evolving Internet healthcare race, currently dominated by a recent Healtheon Corporation
and WebMD merger (Egger, 1999). In their first move, McKesson acquired Abaton.com, a
privately held developer of Internet-based clinical applications for use by physicians, providing
among other things the ability to prescribe electronically. Additionally, McKesson signed a
three-year licensing agreement with Claimsnet.com, which processes medical claims for
physicians via the Internet (Miller, 1999; King Jr., 1999). McKesson and Healtheon/WebMD are
positioning themselves to provide market participants with the ability to connect to doctors,
medical institutions, consumers, and payers with comprehensive products to manage
information, communications and transactions – all via the Internet – contending that the
medium is the “platform common to everyone” (Egger, 1999). Moreover, the information
Page 5
needed by the various participants, i.e., doctors, hospitals, insurers, pharmacies, and patients,
can easily and efficiently be moved via the Internet (Egger, 1999; Cole-Gomolski, 1999), saving
various market participants what has been speculated to be unnecessary inefficiencies and
unproductive overhead costs (Downsend, 1999). In early 1999, Healtheon estimated that some
30 billion transactions are being completed each year utilizing paper, phone and/or facsimile
(Egger, 1999).
3.0 Commercial Intermediation
Despite the fact that commercial intermediation constitutes over 25% of the U.S. GDP, with
roughly two million firms (Spulber, 1999), the subject has not commanded a great deal of
attention from mainstream managerial and economic theorists. Most contributions in this area
are predominantly from finance (Lewis, 1995; Berlin, Mester and Pennacchi, 1998; Rousseau and
Wachtel, 1998). However, there are several publications which have focused on the
intermediation of goods and services (Hackett, 1992; Bentacourt and Gautschi, 1993; Michael,
1994; Spulber, 1996, 1999; Brousseau, 1999).
It is generally accepted that intermediaries play two roles (Brousseau, 1999): (1) A purely
informational role whereby intermediaries are perceived as entities gathering, sorting and
arranging information about both parties’ plans in order to match them, or (2) an economic
matching role where the assumption is made that intermediaries do not have the capability to
perfectly match producers’ and consumers’ plans.
In the first case, information matching is sufficient, and the business model is often
based upon the intermediary seizing a margin of the transaction amount. In the second case, the
inability to perfectly match producers’ and consumers’ plans requires the intermediary to hold
inventories and assume a risk-bearing partnership in the subsequent exchanges, thereby
facilitating economic matching.
The popular press has promoted an extreme argument that the provision of complete
information will enable consumers and producers to match plans and deem intermediaries as
superfluous third parties. Hence, the growth of the Internet has been viewed as determinant of,
if not synonymous with, disintermediation. However, even if one assumes that the astronomical
Page 6
task of matching all transaction parties’ optimization plans is tractable, additional coordination
challenges may persist.
First, information asymmetries constitute more than just asymmetries in knowledge of
plans (Ackerlof, 1970; Milgrom and Roberts, 1987; Stiglitz, 2000). Agents in a decentralized
economy know different things, ranging from skill sets to the Hayekian “knowledge of the
particular circumstances of time and place” (Hayek, 1945, p. 524). These asymmetries in
knowledge enable the realization of economic rents. In addition, a great deal of consumption is
not simply a function of endogenous needs assessment and explicit plans, but is largely dictated
by exogenous factors such as the weather. Governments, for example, stockpile food reserves
for many reasons, including the acknowledgement that circumstances outside the control of the
farmer can result in a bad crop yield and consequent food shortages.
Second, even in the instance that all plans could be made available to all parties, there is
no reason to assume that they would match. Production plans are often dictated by economies
of scale. Production and logistic cycles take time, and are often much longer than the typical
needs assessment and fulfillment cycles of consumers. Business models based upon the creation
and fulfillment of wants in the present moment in a specific geographic location, augmented by
the instantaneous or compulsive behavior of consumers, would have little validity without the
assistance of intermediaries to coordinate needs within local markets.
Finally, asymmetries in skill, experience, general and specific knowledge of time and
place will cause difficulties in assessing the amount of return obtainable in any given exchange.
This leads to the classical problems of moral hazard and adverse selection (Milgrom and
Roberts, 1987) that are very often mitigated to some degree by the participation of a credible
third party intermediary.
While this argument is not exhaustive, it does serve to suggest that intermediaries do
more than merely coordinate information. Moreover, it implies that changes in the cost and
character of information that we have witnessed in the last decade are not only reducing the
aggregate level of intermediation, but are spawning new forms of intermediaries which may be
fulfilling very traditional processes of commercial intermediation with unique methods or in
novel combinations.
Page 7
Hence, a review of the classical roles of commercial intermediation may aid
understanding of the new class of intermediation currently witnessed in the healthcare sector.
While unanimous definitions of commercial intermediaries are difficult to extract, Brousseau
(1999), following Spulber (1996, 1999), delineates the following typology in which
intermediaries ensure adjustments in terms of availability, volume and quality, as well as
securing transactions and liquidity in five inclusive states:
1. Information Management: as an aggregator and filter of information, they support the
producer’s need to determine demand as well as the consumer’s assessment of supply
levels and capabilities.
2. Transaction Securitization: controlling and guaranteeing the products delivered and
assuring payment to producers, third party intermediaries often mitigate adverse
selection and moral hazard problems via the endorsement of the transaction with their
own reputation or legally binding guarantee. As a subset, some other form of value
added service as well as a reconfiguration of the bundle of goods and services is often
performed.
3. Liquidity: through the extension of credit to both sides of the transaction, intermediaries
facilitate the systematic clearing of markets despite possible liquidity constraints in both
ends of the channel.
4. Logistics Management: performing the basal tasks of sorting, packaging, storing,
stocking and transportation of goods from production locations to intermediate
warehouses to consumption sites.
5. Insurance: by purchasing production before consumer demands are expressed, they
provide producers with some stability and security of demand for their production.
While not a guarantee, the intermediary is able to aggregate demand curves and offer
some insurance that production can be disposed of. Likewise, some insurance is offered
to the consumer that the goods will be available, insulating them from fluctuations in
levels of supplies.
These states are the result of what has been observed in the economy. Theoretically, they can be
viewed as a function of the classic endogenous and exogenous problems caused by bounded
rationality, adverse selection and moral hazard, information asymmetry and uncertainty, as
determinants of the decision to enter an intermediated transaction by an economic agent, i.e.,
either consumer or firm level decision making (Rose, 1999; Spulber, 1996, 1999). We categorize
the antecedent determinants of intermediation choice according to uncertainty,  , and
Page 8
preferences,  . Define  as the set of factors governing uncertainty that is perceived market
and environmental uncertainty. These include, but are not limited to: information asymmetry,
opportunism, moral hazard, adverse selection and perceived and real volatility of supply
(Milgrom and Roberts, 1987). Define  as the factors related to consumer preferences and
sensitivities, which include, but are not limited to: temporal sensitivity, immediacy and spatial
needs, aversion to search and coordination costs (Williamson, 1975, 1985), as well as individual
risk aversion (Rose, 1999; Spulber, 1999). The choice to conduct a transaction through an
intermediary is independent for each transaction and will be a function of the product and
service being transacted; moreover, it is independent from the customer’s preferences and
determinants of uncertainty leading to an infinite number of outcomes.
Figure 1: Determinants of Choice in Intermediated Exchanges

Determinants
of Uncertainty
Hypothetical Decision Path B
Insurance
Logistics Management

B
Liquidity Management

A
Hypothetical Decision Path A
Transaction Securitization
Information
Management


B
A

Determinants
of Preferences
Consider that some customers have a greater need for local inventories (temporal or
spatial sensitivities), have higher aversion towards risk and fear of transacting with unknown
entities, or have decreased access to market search mechanisms, giving them a high sensitivity
Page 9
to search and coordination costs. In this case, the aggregate set of preferences could influence
consumption decisions that follow Hypothetical Decision Path A in Figure 3. Here, individual
preferences can yield a higher propensity to engage in intermediated exchanges. Similarly,
higher levels of perceived uncertainty for any market can also influence the propensity to
engage in intermediated transactions, as demonstrated in Decision Path B. For example, the
perception that markets are characterized by an unusually opportunistic culture, or
alternatively, that information transparency is low, would yield agents more likely to transact
through intermediaries in an effort to mitigate the problems of moral hazard and information
asymmetry. Greater levels of environmental instability would also increase the propensity for
intermediated exchanges.
The products and/or services being transacted also determine intermediary type, where
purely information-based products would have a higher frequency in earlier, or lower, states of
intermediation, although not exclusively. Implicitly, products with greater economic value
impose a greater risk of moral hazard and can therefore also increase the propensity to engage
in intermediated exchanges. As an example, consider the need for regulated exchanges to
govern the conduct of financial transactions.
This framework applies to both the firm and consumer level decision-making. The
depiction is clearly stylized, and decision-making differences as related to agent type (consumer
versus managerial) might be a topic well suited to future research.
3.1 Hypothesis Formulation
Intermediary State and Profitability
This framework has several unique implications. First, the structure is hierarchical and
inclusive, such that organizations at the higher levels of intermediation possess all of the
abilities detailed at the lower levels, although direct employment in business activities is not
requisite. For example, the highest state, insurance, implies the legal ownership of goods and
services, which hence implies corresponding roles of logistics management, liquidity,
transaction securitization, as well as information management. Thus, the higher the level of
intermediation, the greater the proportion of the value chain that is under the intermediary’s
Page 10
control, and the greater the level of resulting economic rents. Accordingly, we formulate
Hypothesis 1.
H1: The profitability of intermediaries is positively related to the state of intermediation.
While the assumption of inclusion may not hold in some instances, we did not find any firms in
our sample for which this assumption was inconsistent. We discuss the consequences of
relaxing this assumption later in the analysis.
Corporate Diversification
The concept of inclusive intermediary states implies that firms assume greater aggregate levels
of potential and realized functions in the economy. Greater intermediation is thus somehow
related to traditional concepts of diversification, although not synonymous therewith. The
literature concerning corporate diversification is both extensive and inconclusive. The large
diversified concerns that grew out of the 1970s were often justified on those arguments from
financial economics that emphasized the reduction of systematic risk (Jensen, 1986; Lubatkin,
1987; Chatterjee and Lubatkin, 1990), stabilizing cash flows (Amit and Livnat, 1988), or as a
mechanism in the market for corporate control that replaces inefficient managers with more
competent management (Jensen and Ruback, 1983). However, the resulting wave of down- or
right-sizing that ensued in the 1990s was often justified on arguments of decreasing returns to
management (Gurbaxani and Whang, 1991) that claimed the optimal level of managerial
bureaucracy had been exceeded for many organizations. Other studies pointed to the negative
cross-subsidization of low-performing, unrelated segments (Jensen, 1986; Berger and Ofek,
1995), or to the general consensus that investments in capital markets could diversify corporate
risk more effectively than managerial governance.
Financial economics literature has been accused of being “rearward-looking” in its
emphasis on risk management and its use of diversification to improve the efficiencies of
existing operations (Karim and Mitchell, 2000). In contrast, strategic management literature has
tended to be more forward-looking, arguing that organizations must engage in ongoing
Page 11
changes of their operations and strategic positions due to changing competitive environments.
However, even in a so-called “forward-looking” view, the empirical literature on diversification
has been mixed. Some studies have found gains from related diversification (Singh and
Montgomery, 1987; Wernerfelt and Montgomery, 1988), where others have found gains from
unrelated diversification (Chatterjee, 1986). Unsurprisingly, other studies have found little
conclusive evidence that either related or unrelated diversification increases firm performance
(Bradley, Desai and Kim, 1988; Lubatkin, 1987; Lubatkin and O’Neill, 1988).
We argue that the level of diversification is positively related to the level of intermediary
profitability. This position is based upon the idea that firms diversify through time as financial
resources grow, independent of the intermediary state. Increased profitability permits the
managerial freedom to expand operations in both related and unrelated areas. As even related
functional diversification would be considered a greater financial and managerial risk, such
positions are easier to take when profitability in normal operations is unconstrained.
H2: The profitability of intermediaries is positively related to the level of diversification.
Firm Age, Pure-Play Strategies and Public Listing
Organizational studies have shown ambiguous results concerning the effects of age on business
survival and profitability. While some studies have demonstrated support for the “liability of
newness argument” by finding declining rates of bankruptcy or dissolution with greater firm
age, particularly in manufacturing (Stinchcombe, 1965; Carroll and Swaminathan, 1982;
Freeman, Carroll and Hannan, 1983; Evans, 1987), other studies have found the opposite effects,
especially in the service sector (Barnett, 1990; Baum and Mezias, 1992; Amburgey, Dacin and
Kelly, 1994). Likewise, studies have been inconclusive concerning the relationship between firm
age and aggregate survival rates.
Other studies have examined the effects of firm age on modes of exit. Mitchell (1994)
found that older businesses enabled prospective buyers to better evaluate the companies’ track
records, where Chang and Singh (1999) found that greater age was correlated with the
development of idiosyncratic assets that prevented them from obtaining high market prices in
Page 12
the case of a sell-off. Hence, even in the instance of an exit decision, empirical evidence remains
inconclusive concerning the relationship between firm age and exit mode.
Intermediary types that have emerged from the Internet comprise a major portion of our
sample. As many of these younger firms are struggling to obtain positive profitability, we
hypothesize that the older, established businesses are more profitable than the younger startups.
H3a: The profitability of intermediaries is positively related to firm age.
Likewise, we believe that established industrial enterprises will have greater
profitability levels than firms which are recently launched with the express intent of capitalizing
on the capabilities offered by the Internet. While we have limited theoretical justification for the
hypothesis, we refer to the high frequency of bankruptcy and dissolution that has characterized
Internet start-ups.
H3b: Profitability is negatively related to intermediaries with an Internet pure-play strategy.
The relationship between size and organizational profitability has been a major focus of
managerial studies as well (Evans, 1987; Baum and Mezias, 1992). A variety of measures have
been used to assess firm size, including market share, absolute sales volume, market
capitalization, employees, and asset evaluations. While the factors attributed to firm size and
profitability are complex and interdependent, general arguments suggesting that larger firms
are more profitable point to natural barriers to entry, scale economies, greater financial
resources, as well as advantageous access to managerial talent and capital markets (Chang,
1996; Chang and Singh, 1999). We assume that firms that are publicly listed have higher
capitalization rates than otherwise, which enables increases in marketing expenditures, strategic
and operational investments, financial longevity, as well as incentive structures implicit in a
shareholder wealth-maximization paradigm that emphasizes earnings growth and return on
Page 13
equity. Hence, we hypothesize that the companies in our sample that are publicly traded will
have greater profitability than those that are not.
H3c: Profitability is positively related to intermediaries that are publicly traded.
Intermediary States and Growth Strategy
Our second research question focuses on how electronic intermediaries evolve through time. A
survey of the popular press will indicate that many of the businesses with electronic channels
have a particularly high frequency of announcements concerning acquisitions, joint ventures
and alliances. While we have no indication of whether these announcements are more frequent
than in the normal population of firms, or whether this is merely symptomatic of a greater level
of press savvy requisite for pre- and post-IPO companies, the observation does provoke a
question of the nature of intermediary evolution, and how these strategies are related to
different intermediary states.
While much of financial economics literature addresses why firms diversify (Jensen and
Ruback, 1983; Jensen, 1986; Lubatkin, 1987; Amit and Livnat, 1988; Chatterjee and Lubatkin,
1990), strategic management literature has focused on the question of why a firm would choose
one growth strategy as opposed to another (Yip, 1982; Chatterjee, 1990; Chang and Singh, 1999).
Economic and strategic literature has generically categorized growth strategies in broad terms,
often as acquisition, organic or internal development, or some type of joint venture or alliance
(Yip, 1982; Chatterjee, 1990; Chaterjee and Wernerfelt, 1991).
Several authors have argued that acquisitions serve to minimize the issues of bounded
rationality and time compression diseconomies that constrain the content and speed with which
people learn (Simon, 1945; Dierickx and Cool, 1989). This makes acquisitions preferred to
internal development as a mode of growth. Hence, if the goal of acquisitions is to secure the
tacit knowledge and organizational memory that reside in an organization’s structure via the
routines that the organization maintains (Nelson and Winter, 1982; Karim and Mitchell, 2000),
the question remains as to under what circumstances such acquisitions are most appropriate.
Yip (1982) argued that market relatedness does not reduce the cost of entry via acquisition, as
Page 14
the market for corporate control sets the price of the acquisition. Therefore, firms tend to enter
related businesses via internal development, where unrelated areas are entered via acquisition.
Chatterjee (1990) furthered this argument by claiming that the more closely related the two
markets are, the fewer the requisite complements to the firm’s own physical and knowledge
base assets. Hence, internal development may reduce entry costs in a related market. In
contrast, growth into a related market via acquisition is more likely to entail the purchase of
redundant resources. Accordingly, growth via acquisition is more attractive in unrelated
markets. However, attempts to verify the hypothesis that firms are more likely to enter
unrelated businesses via acquisitions have thus far been unsuccessful, and the aggregate
empirical literature concerning growth strategies remains inconclusive (Karim and Mitchell,
2000).
We argue that, as the intermediation state becomes more complex and demanding in
terms of physical capital, tacit and process-based knowledge and organizational memory, the
barriers to removing constraints to individual and organizational learning are most easily
overcome via acquisition. Accordingly, it is assumed that intermediaries that can perform the
functions of logistics management, liquidity management and legal inventory ownership will
often choose to develop these capabilities through acquisition, as these capabilities frequently
require significant capital investment and process capabilities. Alternatively, it is assumed that
intermediaries that perform functions which require less capital investment, such as
information management, would attempt to extend their breadth through a higher frequency of
joint ventures, licensing agreements and alliances. Accordingly, we formulate hypotheses H4a
and H4b:
H4a: A higher relative frequency of acquisitions is positively related to higher states of
intermediation.
H4b: A higher relative frequency of joint ventures is positively related to lower states of
intermediation.
Page 15
While we do posit a higher frequency of acquisitions for higher intermediary states, intuition
does not necessitate that the inverse holds as well. Intermediaries with higher states of
intermediation could just as easily opt for organic growth in instances that the management felt
were opportune, given the ambiguous results currently in the literature concerning growth
strategies. Hence, we posit that a relationship exists between the state of intermediation and the
relative frequency of internal growth. However, we refrain from defining the direction of the
relationship.
H4c: A relationship exists between the relative frequency of organic growth and the state of
intermediation.
4.0 Data Collection and Analysis Method
This study has been conducted in two phases. The first round of data collection was completed
in the Fall of 1999 through the Spring of 2000. In the second phase, we returned to the data in
the Summer of 2001 in order to expand the size of the sample, reassess the evolutionary
patterns, and award specific attention to developments within the sector after the stock market
fall in April of 2000. Because this study is exploratory, our definition of portal was broadened to
include any type of Internet presence that aggregated the products or services of multiple
organizations within the healthcare sector. Thus, individual medical supply companies with
Internet sites that exclusively represent their own products and/or services were not included.
A many-to-many relationship between suppliers and customers was required to fulfill the
definition of an intermediary (Spulber, 1999). By some estimates, there are over 17,000 web sites
with some form of medical content (Fox, 2000). Consequently, a sample of 58 organizations
cannot be considered comprehensive. However, a thorough scanning of the media was made to
identify the most significant members of the sector in order to form a representative, theoretical
sample (Eisenhardt, 1989).
Our preferred form of data collection was personal interviews with senior and middle
management (Creswell, 1997; Mason, 1996; Miles and Huberman, 1994). The interviews were
semi-structured and guided by the assumption that there were two types of companies entering
Page 16
this venue - either new startups/joint ventures that are seeking to build a completely new client
base as a first mover, or existing companies in more traditional areas of pharmaceutical
distribution or insurance claims processing that are seeking to expand into an Internet lead
business by leveraging an existing client base. With the former organizations, we sought to
illuminate the characteristics of the medical industry and the technological environment that
motivated the move to a web-intermediated business model. Why did the management believe
that such a proposition was feasible? What antecedents presented the opportunity of
considerable gains in this sector (e.g., fragmentation, high search costs, etc.)?
With the latter group of organizations, we focused on the rationale behind their
movement into an intermediary-based business model. How did they intend to leverage their
existing business in this different media? Could they identify specific exogenous forces that
motivated their expansion beyond their traditional venue?
In the instance that a personal interview was not possible, secondary data was
employed, including information collected from web sites, annual reports, newspaper articles,
and third party analyses such as stock analysts and venture capital media (Miles and
Huberman, 1994). Although the secondary data does not permit an exploration of the
managements’ assessments of risks and opportunities at the same level as personal interviews,
it does offer historical documentation concerning companies’ development, economic
environment, diagnostics of the business model, as well as cross-validation of managerial
opinion.
Finally, in an effort to ensure that our results are consistent with the assessment of
experienced specialists within the healthcare management domain, we validated the results of
the analysis with members of the School of Hospital Management within our own institution.
4.1 Analysis
In an analysis of the type of services offered by the healthcare portals, a number of broad,
generic functions were identified. Traditional industry classification schemes such as SIC codes
were consulted, but did not offer sufficient granularity or relevance for our analysis. Our
classifications included functions such as content aggregation, patient and records
Page 17
management, insurance claims processing and supply chain management and procurement
portals. Our classification typology was developed in the first phase of the data collection
process. In the second phase, we attempted to reevaluate the typology and concluded that the
classifications retained explanatory utility in the period after the stock market fall in April 2000.
However, we did add one additional class in the second phase – consumer retailing. Definitions
of each type are presented in Table 1. These functions are not exclusive, meaning that most
subjects in the sample offered multiple functions.
Page 18
Table 1. Functional Categories of Medical Healthcare Portals
Function
Description
Consumer content aggregation
Aggregation of medical content aimed towards
consumer education, prevention and empowerment
Professional content aggregation
Aggregation of professional content to educate and
improve the competencies of medical professionals,
including nurses and administrative management
Consumer retailing
Sale of products to consumers, such as health supplies,
vitamins and pharmaceuticals
Patient management
Tools designed to facilitate the effective management of
patient booking, case data and diagnosis
Records management
Digitization and management of patient records
Practice management
Administration and analysis tools for financial
administration of small, medium and large practices
Physician career management
Professional recruiting, credential renewal and
verification
Insurance claims processing
Clearing and processing of insurance claims, verification
of codes and correct payment, revenue stream analysis
Supply chain management
Integration with chosen vendors of medical supplies
and pharmaceuticals for purchasing and inventory
management
Procurement portals
Domain-specific markets for the bid/offer of medical
supplies and fixed assets
Application service provision
Asp aggregators that serve to consolidate a portfolio of
applications and services on a single platform
4.2 Profile of Sampled Healthcare Portals
Out of the 58 companies that constitute our sample, 44.2% are publicly listed, predominantly on
the NASDAQ stock exchange. 55.2% were categorized as Internet pure-plays, while the
remaining 44.8% are companies that existed prior to the growth in Internet-based business
models. The average age of the companies is 5.8 years, where 27.6% of the companies were
Page 19
above this average. The companies that existed prior to the Internet were mainly involved in
insurance claims processing, practice management, and records management, whereas the
younger companies are characterized by higher concentration within consumer and
professional content aggregation, asset procurement markets, and physician career
management.
4.3 Dependent Variable – Profitability
The dependent variable for Hypotheses 1 through 3 is the level of average profitability of the
healthcare portal through the 3 years of 1998 to 2000. The variable was registered as an ordinal
measure of net profit: less than negative 10%; greater or equal to negative 10%, but less than
negative 5%; greater or equal to negative 5%, but less than 0%; greater or equal to 0%, but less
than 5%; greater or equal to 5%, but less than 10%; greater or equal to 10%. The brackets were
used for several reasons. First, given the relative youth of this type of business, a traditional
binomial logistic estimation of survival would not adequately capture many of the companies
that are currently struggling for their existence. A large portion of our sample has previously
had stock traded well above 20 dollars per share, but are now traded at only a fraction of these
prices. Second, while it was possible to obtain exact profitability information on 53% of our
sample, estimates for the remainder of our sample were obtained through interviews (34%) or
estimation from publicly available material (13%). In the instance of an interview, the
respondents were offered brackets so they were not compelled to disclose exact profitability
information. In addition, these brackets offered sufficient granularity for the study while
permitting the freedom requisite when estimating profitability.
Given the lack of maturity in the medical portal sector, other dependent variables were
considered such as market shares or relative growth rates. Intuitively, the high level of
investments required for establishment of brand recognition and infrastructure may make
short-term profitability a biased indicator of successful, long-term growth strategies. However,
as many electronic intermediaries have not survived the initial rounds of venture funding, the
issue of short-term strategy and short-term survival is appropriately pertinent.
Page 20
4.3 Dependent Variable – Intermediation State
The dependent variable for Hypothesis 4 is intermediation state. This construct serves as an
independent variable in Hypotheses 1 through 3, and the dependent variable in Hypothesis 4.
As our second main research question is how intermediaries evolve over time, a variable that
captures the development of relative intermediary growth is appropriate for testing hypotheses
concerning growth patterns.
4.4 Independent Variables
Our focal independent variable is an intermediation index, based upon the Spulber (1996)
framework previously discussed. For each company, it was necessary to assess what level of
intermediation it had assumed, and give it an ordinal score from one to five. Unfortunately a
one to one mapping between our functional typology and intermediation level was not
possible. This problem can be illustrated in the following manner. Take, for example, a
procurement portal. In their simplest form, procurement portals perform some type of market
aggregation and economic matching. In the instance that the portal does nothing other than
match buyers and sellers, it would receive a score of 1 for information management. However,
should the portal take a more active role in facilitating the transaction, accepting payment, and
coordinating the delivery of the goods, a score of 2 for transaction securitization would be
applied. Likewise, should the portal extend credit to either side of the transaction, a score of 3
would be awarded. If the portal assumed legal ownership of the inventory, provided
warehousing and logistics, a score of 4 or 5 was awarded accordingly. The index was measured
either at the time of exit of the company or at the final date of measurement.
Another problematic example is consumer retailing. If the portal simply provided some
type of link or inclusion into the activities of another entity via a joint venture or marketing
alliance, then no function other than information management exists. Should the portal actually
hold the inventory, in legal and accounting terms, then higher levels of intermediation were
registered.
Insurance claims processing is another function we frequently encountered in this
market sector. In some cases, insurance claims are processed based upon a variable rate per line
Page 21
item or claim, whereas in other instances, the claims can be bought at a discount, offering the
equivalent of a forward credit. The first case would be awarded a score of 2 for transaction
securitization, and the latter instance would receive a score of 3 for liquidity management.
Other independent variables include publicly traded, a binary variable equal to 1 if the
company is listed on any major stock exchange, 0 if not. Likewise, Internet pure-play is also a
binary variable that assumed a value of 1 if the company was launched with the express intent
of capitalizing on the capabilities of the Internet, and 0 if the company had an existing industry
function independent of the Internet. Age is a continuous variable that measures the period
between establishment of the company through the point of data collection.
Diversification index is an index we developed based upon our functional typology. We
utilized the Berry-Herfindal index defined as one minus

i  1,  I
Pi2, where Pi is the proportion
of sales in the ith business (Neter et al., 1996). However, as SIC codes, which are normally used,
did not offer sufficient granularity for this analysis, we substituted our own functional typology
of 11 functions of medical healthcare portals, measured and time t at either the point of exit or
at the date of measurement. The event of diversification was registered when the firm entered a
different functional area.
Growth strategies were registered in the data collection process. Each time a company
expanded its functional capabilities, we registered the instance as either organic growth, growth
by acquisition, or growth by joint venture/alliance. Typically, growth was defined as the
entrance of the company into a new functional area as defined in our typology. However, in
many instances, companies made acquisitions of companies in areas already included in the
company’s current portfolio of functions in an effort to expand market share. Hence, all cases
where significant growth was achieved via acquisition, joint venture, or internally were
registered. Many of the larger, mature entities had over 30 instances of some kind of growth,
where the younger, smaller companies had less than five registered instances. In order to
normalize the data, we transformed the nominal counts to relative frequencies. Hence, a
company could have 30% of growth via internal/organic means, 40% via acquisitions, and 30%
via joint venture or alliance (Seddighi, Lawler and Katos, 2000).
Page 22
4.5 Estimation and Findings
Hypotheses 1, 2 and 3 were tested using the following ordinal logistic estimation where the
dependent variable has categorical values based upon the values of ordinal properties. In this
instance, ordered response logistic regression is considered an appropriate technique for
analysis (Chu and Anderson, 1992; Levin, 1998; Neter et al., 1996) and is based on the methods
developed by McCullagh (1980, 1998). ORL gives estimates that are unbiased, and at the same
time allows estimation of a parsimonious model that is readily interpretable (Pan and Chi,
1999). Scale variables are not appropriate due to their inconsistency in the intervals between the
extreme levels (less than negative 10% and greater or equal to 10%) and the remainder. Hence,
the simplifying assumptions on which basic linear regression relies are not satisfied, as it is
highly sensitive to the manner in which categories are defined.
We use a negative log-log function due to the fact that the categories in the dependent
variable are not normally distributed. The negative log-log is appropriate when the lower
categories of the dependent variable are more probable (SPSS, 1999) as was the case with our
sample, where 46.6% of the companies were categorized in the lowest profitability category. We
test the significance of the entire OLR model with the model log-likelihood chi-square, which is
analogous to the multivariate F-test in linear regression testing the null hypothesis that the
coefficients are zero (Neter et al., 1996). Further, for each estimate, we conduct a two-tailed
significance test of the Wald statistic, which is the ratio of the estimated coefficient to its
estimated standard error and follows a chi-squared distribution. Accordingly, we write the
likelihood function as
 log  log     5    1inindex   2 log age   3divindex   4 pubtrade   5 pureplay 
As in binary logistic regression, this function can be estimated using log-likelihood
techniques, and estimates of  can be used to test the probability that various state
characteristics significantly affect the probability of higher levels of profitability. Unlike OLS,
estimates of  cannot be interpreted as marginal effects, but rather the marginal impact these
Page 23
variables have on the probability of higher levels of the dependent variable, in this case
profitability (Neter et al., 1996).
The ordinal logistic regression analysis of profitability determinants is presented in
Table 2. The model is significant (chi-square for covariates significant for P<0.001). The Cox and
Snell, Nagelkerke and McFadden pseudo R-square statistics are acceptable for the exploratory
nature of this research at .437, .469 and .214, respectively. A discussion of the results of
Hypotheses 1 through 3 follows.
Table 2. Determinants of Profitability from 1997 to 2000: Ordinal Logistic Regression
Model
Intercept Only
Final
-2
Log
Likelihood
155.319
122.007
ChiSquare
Df
33.312
5
Pseudo R-Square
Cox and Snell
Nagelkerke
McFadden
Significance
.000
.437
.469
.214
Parameter Estimates
Threshold
Profit=1
Profit=2
Profit=3
Profit=4
Location
Intermediation index
Log(Age)
Diversification index
Non-publicly traded
Non-Internet pure-play
Estimate
3.591
4.693
5.775
7.472
Std. Error
.806
.907
1.015
1.234
Wald
19.857
26.750
32.403
36.676
Significance
.000
.000
.000
.000
0.651
1.306
(0.163)
0.427
0.658
.268
.433
.133
.456
.475
5.905
9.075
1.496
0.876
1.196
.015
.003
.221
.349
.166
Hypothesis 1 posited that the profitability of healthcare intermediaries is positively
related to the state of intermediation and to the level of diversification. The coefficient for the
Page 24
intermediation index is positive, although the nominal level is moderate at 0.651. Hence, while
the hypothesis can be supported with a high statistical significance at .015, the economic
significance of the coefficient is less dramatic. The findings suggest that there is a moderately
positive relationship between intermediation state and profitability, establishing support for
Hypothesis 1.
Hypothesis 2 found no statistical support, with significance of .221. As previously
mentioned, we observed that many of the companies we studied had an abnormally high
degree of growth activity. This could be due to the fact that the sample had a significant number
of such start-ups, which, after the endowment of either private or public funding, had a limited
period of time in which to establish a viable business model and defendable competitive
position.
We explicitly attempted to limit the amount of correlation between the intermediation
level construct and the diversification index. Hence, the results suggest that higher profitability
is attainable through a focused position at higher intermediary states, not merely the acquisition
of functional areas for the sake of diversification alone.
Hypothesis 3 explores the relationship between intermediary profitability, firm age,
Internet pure-play strategies, and public listing. As expected, the coefficient for age is positive
and very significant at .003. However, the value of the coefficient is low, even for logarithmic
form, suggesting that while there is a positive association between firm age and profitability,
the effects are moderate. Hence, we find weak support for Hypothesis 3a.
Hypothesis 3b suggests that a negative relationship exists between intermediaries that
were founded on an Internet pure-play strategy, and profitability. This hypothesis finds no
support. While the direction of the coefficient is as expected, the statistical significance is .166.
Hypothesis 3c suggested that a positive relationships exists between intermediaries that
are publicly listed and profitability. With a significance of .349, this hypothesis finds no support.
Thus, we were not able to find any statistically or economically significant relationship
between profitability, Internet pure-play strategies and public listing. However, we did find a
weak relationship between profitability and firm age. Intuitively, the selection forces that
permeate the sample are sufficient to explain this relationship.
Page 25
Our second research question relates to how intermediaries evolve over time, that is,
what is their primary method of growth? Hypothesis 3 was tested by using intermediation
index as the dependent variable, in an attempt to explore how the frequency of different growth
strategies varies with different states of intermediation. As there is no scalar relationship
between the intermediation states, that is, the differences cannot be expressed numerically, only
ordinally, we chose to apply ordinal logistic regression in the test of Hypothesis 4 as well.
 log  log     4    1organic   2acquisitio ns   3 joinventures   4divest 
The coefficients are estimated by maximizing the likelihood function.
The ordinal logistic regression analyses of intermediation state and growth strategy are
presented in Table 3. The model is significant (chi-square for covariates significant for P<0.001).
The Cox and Snell, Nagelkerke and McFadden pseudo R-square statistics are relatively high at
.639, .703 and .425, respectively. A discussion of the results of Hypothesis 4 follows.
Table 3. Intermediation State and Growth Strategy: Ordinal Logistic Regression
Model
Intercept Only
Final
-2
Log
Likelihood
116.594
57.451
ChiSquare
Df
59.143
4
Pseudo R-Square
Cox and Snell
Nagelkerke
McFadden
.639
.703
.425
Parameter Estimates
Page 26
Significance
.000
Threshold
IntIndex=1
IntIndex=2
IntIndex=3
Location
Freq. organic growth
Freq. growth by acquisition
Freq. growth by joint venture
Freq. divestitures
Estimate
.357
2.399
6.559
Std. Error
1.505
1.561
2.201
Wald
0.056
2.362
8.884
Significance
.812
.124
.003
.412
3.142
(1.575)
27.432
1.484
1.537
1.675
10.316
0.077
4.177
0.884
7.072
.781
.041
.347
.008
Hypothesis 4a suggested that a higher relative frequency of acquisitions is positively related to
higher states of intermediation. The coefficient for frequency of acquisitions is positive at 3.142
and statistically significant at .041. The test supports our expectation that higher intermediation
states require more sophisticated capital investment, process knowledge and organizational
routines that are easier to obtain through acquisitions. However, our complementary
Hypothesis 4b, that a higher relative frequency of joint ventures is positively related to lower
states of intermediation, did not find support. Where the direction and size of the coefficient are
in agreement with expectations, the statistical significance was not even close to the 10% level at
.347. One possible interpretation is that where acquisitions are the preferred growth method for
higher intermediary states, joint ventures and alliances are equally employed by all types of
intermediaries resulting from the lower access costs to this growth form.
Our final hypothesis, that a relationship exists between the relative frequency of organic
growth and the state of intermediation, found no support. Not only is the coefficient limited at
.412, there is no statistical significance at .781. An explanation similar to that for joint ventures,
that the growth strategy may be viable for all intermediary states, may be applicable.
Interestingly, we included the registration of divestitures in the regression, although we
did not hypothesize about its role in growth strategies. This coefficient was extremely high at
27.432 and highly significant at .008. This suggests that, not only are companies with higher
intermediary states more likely to employ acquisitions in growth strategies, they are even more
likely to use divestitures in the instance that refocusing of the companies’ strategies becomes
necessary. This result is not counterintuitive to our logic. We expect higher intermediary states
to be more highly capitalized organizations due to the sheer level of investment needed to
Page 27
provide the infrastructures. Those companies adept at employing acquisitions for growth
would feel equally comfortable utilizing divestitures, as both methods require similar access to
venture capital or investment banking networks, legal and accounting support.
5.0 Discussion
The results of the analysis are summarized in Tables 4a and 4b. Our hypothesis that profitability
is positively related to higher intermediary states found strong statistical support, although the
coefficient was moderate. This suggests that intermediary types that perform the more capitalintensive functions, such as supply chain management, transaction management and inventory
management, are more profitable than intermediary types that perform the roles of information
management and economic matching. Empirically, this is consistent with the general lack of
success witnessed within content aggregators and B2B exchanges that emerged from the
electronic commerce boom of 1998-2001. While we did find some support for a positive
relationship between firm age and profitability, we found no relationship between
diversification levels, Internet pure-play strategy or public listing.
Table 4a.Variables and Relationship to Profitability
Variable
Expected
Identified
Intermediation state
+
Positive
Diversification level
+
None
Age
+
Weak
Internet pure-play strategy
-
None
Publicly listed
+
None
Table 4b. Variables and Relationship to Intermediation State
Variable
Expected
Identified
Freq. of acquisitions
+
Positive
Freq. of joint venture
-
None
Freq. of organic growth
?
None
Page 28
Consider the paths of two similar content aggregators, DrKoop and WebMD. DrKoop
has made significant and costly attempts to institutionalize their brand through the use of the
former Surgeon General’s reputation for integrity and through expensive marketing endeavors,
including restrictive agreements with AOL and Walt Disney-Go.com that call for DrKoop to
make huge payments in return for being the exclusive health content provider to the two
companies' web properties (Yates, 2000). Yet, despite the arduous attempts at first mover
branding and advantages, the position of content aggregator in isolation has proven very
difficult to maintain, such that the company is projected to run out of cash in late 2001, despite
several attempts for continued financial support (Yates, 2000). Whereas DrKoop has withheld
its strategy of information management, WebMD, by its merger with Healtheon, has broadened
its activities into higher intermediary states such as insurance claims processing and supply
chain management, and has a much sounder revenue profile as a consequence. While content
aggregation can be viewed as provision of an asset in its own right, the general challenge faced
by pure content providers is that the assets provided are often not requisite to the operations of
the business. Thus, the most central platform for marketing supplementary content may in fact
be the applications facilitating the revenue stream, such as the insurance claims processors.
Through management of the revenue stream, intermediaries can logically extend liquidity
management through the purchasing of accounts receivables forward. Moreover, logistics
functions, central to the metabolism of the organization, also command a central role in the
needs for intermediation. In this instance, companies like McKesson/HBOC, which control large
networks of pharmaceutical and medical supply provision, also realize advantages in
marketing complementary products of their own and associated companies.
Likewise, we have seen similar difficulties in the many emergent B2B portals that offer
little other than economic matching. Generating sufficient interest in these markets can be
prohibitively difficult in the case that the markets do not choose to perform some functions of
transaction guaranteeing and clearing, liquidity management, and market making/insurance by
physically or legally managing inventories. While information aggregation and the governance
of opportunism, bounded rationality and moral hazard are requisite in any economy and,
Page 29
hence, secure a role for intermediaries, the ability to perform such tasks normally increases as a
function of intermediary state.
Our analysis of intermediation states and growth strategy found a positive relationship
between the frequency of acquisitions and intermediation state. However we could not identify
a negative association between the use of joint ventures and intermediation states, nor any
association with organic growth levels. Surprisingly, we did see a highly significant coefficient
with the use of divestitures. This evidence suggests that higher intermediation states are
characterized by a greater need to acquire existing capital assets, tested organizational routines
and business processes. Moreover, such resources are best acquired under a single governing
body, rather than running the risks implicit in a bilateral governance agreement such as a joint
venture. Finally, we do see that the higher intermediary states are more likely to use
divestitures as a method of exit strategy. This suggests that the propensity to use acquisitions
may also be a function of other characteristics that yield the organization adept at accessing
investment banking networks the legal and accounting support requisite for such transactions.
5.1 Contributions for Theory and Practice
This analysis offers the following contributions. First, it is one of the few empirical
investigations of intermediation theory outside the discipline of finance. By mapping
intermediation states directly onto a specific, cohesive sector of the economy, it offers a unique
engagement and validation of the theory’s explanatory utility. Moreover, the variance in the
sample of intermediary states offered by the inclusion of Internet pure-play content aggregators
compared to established brokers and wholesalers offers a unique opportunity to statistically test
the robustness of the theoretical constructs of all types of intermediaries. This contribution is
particularly poignant given the recent demise of many electronic commerce businesses and the
current disillusionment that the business community has expressed with the proselytizing that
has characterized the sector.
5.2 Limitations
Page 30
While we could not find any evidence to the contrary in our sample, the assumption that
intermediation states are inclusive may not hold in some sectors of the economy. Feasibly,
intermediation states may be discrete alternatives rather than inclusive, ordinal outcomes.
Future studies could focus on the interdependence of such relationships and explore the
degrees of inclusion within intermediaries in different sectors of the economy.
Whereas the majority of companies in this sample displayed a high propensity to
announce growth activity through press releases and the Internet, a great deal of the internal
growth that occurred in the sample may not have been registered. This is due to the fact that,
where groundbreaking developments that have some marketing value may be publicly
announced, much of the significant growth that is routine and less dramatic may not be applied
in a marketing or public relations venue. While we attempted to control for this by defining
growth as any significant advancement into or within a functional area, the potential for
delineation and identification variance must be acknowledged.
6.0 Conclusion
This study has developed a framework of five, inclusive, commercial intermediary
states, and used it to formulate hypotheses concerning intermediary states, profitability and
growth strategy. Using data collected from nearly five-dozen electronic healthcare portals, the
study identified a positive relationship between intermediary state and profitability. Increased
profitability at higher stages of intermediation suggests that these firms perform the more
capital-intensive functions such as supply chain management, transaction management and
inventory management, command greater economic rents as a function of greater proportions
of the supply chain under their domain, as well as increased yields due to greater capitalization
levels. The lack of success found for pure content aggregators and B2B exchanges within the
health care industry supports the notion that it is difficult to maintain profitability performing
solely economic matching and information management, as has been depicted in the case of
WebMD versus Dr. Koop.
The increased use of acquisitions as growth strategies at higher intermediation states
may demonstrate a greater need to acquire processes realized through greater capital
Page 31
investment, tested organizational routines, and business processes. These resources could be
viewed as best suited to acquisition under a single governing body, as opposed to a bilateral
governance scenario such as joint ventures.
Page 32
References
Ackerlof, G. A. (1970) “The Market for "Lemons": Quality Uncertainty and the Market
Mechanism,” Quarterly Journal of Economics, 84, pp. 488-500.
Amburgey, T., T. Dacin and D. Kelly (1994) “Disruptive Selection and Population
Segmentation” in J. Baum and J. Singh (eds.) Evolutionary Dynamics of Organizations.
Oxford University Press, New York, pp. 240-254.
Amit, R. and L. Livnat (1988) “Diversification Strategies, Business Cycles and Economic
Performance,” Strategic Management Journal 9(2), pp. 99-110.
Amit R. and P.J.H. Schoemaker (1993) “Strategic Assets and Organizational Rents,” Strategic
Management Journal, 14(1) pp. 99-110.
Baily, J.P. and J.Y Bakos (1997) “An Exploratory Study of the Emerging Role of Electronic
Intermediaries,” International Journal of Electronic Commerce, 1, 3, pp. 7-20.
Bakos, J.Y. (1998) “The Emerging Role of Electronic Marketplaces on the Internet,”
Communications of the ACM, (August) 41, 8, pp. 35-42.
Bakos, J.Y.(1991) “A Strategic Analysis of Electronic Marketplaces,” MIS Quarterly, 15, 4, pp.
295-310.
Barnett, W. (1990) “The Organizational Ecology of a Technical System,” Administrative Science
Quarterly, 35, pp. 31-60.
Baum, J. and S. Mezias (1992) “Localized Competition and Organizational Failure in the
Manhattan Hotel Industry,” Administrative Science Quarterly, 37, pp. 580-604.
Benjamin, R. and R. Wigand (1995) “Electronic Markets and Virtual Value Chains on the
Information Super Highway,” Sloan Management Review, 36, 2, pp. 62-72.
Bentacourt R.R. and D.A. Gautschi (1993) “The Outputs of Retail Activities: Concepts,
Measurement, and Evidence” University of Maryland Department of Economics
Working Paper Series 90-17, September ,39.
Berger P.G. and E. Ofek (1995) “Diversification’s Effect on Firm Value,” Journal of Financial
Economics, 37, pp. 39-65.
Berlin, M., L. Mester and G. Pennacchi (1998) “Intermediation and Vertical Integration,” Journal
of Money, Credit and Banking, 30(3), pp. 500-523.
Bradley, M., A. Desai and E.H. Kim (1988) “Synergistic Gains from Corporate Acquisitions and
their Division between the Stockholders and Target Acquiring Firms,” Journal of
Financial Economics, 21, pp. 3-40.
Brousseau, E. (1999) “The Governance of Transaction by Commercial Intermediaries: An
Analysis of the Re-engineering of Intermediation by Electronic Commerce,” Third
Conference of the International Society for New Institutional Economics, Washington DC,
1999.
Page 33
Carroll, G. and A. Swaninathan (1982) “The Organizational Ecology of Strategic Groups in the
American Brewing Industries,” Industrial and Corporate Change, 1, pp. 65-98.
Chang S.J. and H. Singh (1999) “The Impact of Modes of Entry and Resource Fit on Modes of
Exit by Multibusiness Firms,” Strategic Management Journal, 20, pp. 1019-1035.
Chang, S.J. (1996) “An Evolutionary Perspective on Diversification and Corporate
Restructuring,” Strategic Management Journal, 17(8), pp. 587-611.
Chatterjee, S. (1986) “Types of Synergy and Economic Value: The Impact of Acquisitions on
Merging and Rival Firms,” Strategic Management Journal, 7(2), pp.119-140.
Chatterjee, S. (1990) “Excess Resources, Utilization Costs, and Mode of Entry,” Academy of
Management Journal, 33, pp. 780-800.
Chatterjee, S. and B. Wernerfelt (1991) “The Link between Resources and Type of
Diversification: Theory and Evidence,” Strategic Management Journal, 12(1), pp. 33-48.
Chatterjee, S. and M.H. Lubatkin (1990) “Corporate Mergers, Stockholder Diversification, and
Changes in Systematic Risk,” Strategic Management Journal, 11(4), pp. 119-140.
Chircu, A. and R. Kaufman (2000) “Reintermediation Strategies in Business-to-Business
Electronic Commerce,” International Journal of Electronic Commerce, 4, 4, pp. 7-43.
Chu, W. and E.M. Anderson (1992) “Capturing Ordinal Properties of Categorical Dependent
Variables: A Review with Application to Modes of Foreign Entry,” International Journal of
Research In Marketing, 9, pp. 149-160.
Cole-Gomolski, B. (1999) “Extranet Services Target Health Care,” Computerworld, 32(37), p. 10.
Creswell, J. (1994) Research Design: Qualitative and Quantitative Approaches. Sage Publications,
Thousand Oaks, California, U.S.A.
Dierickx, I. and K. Cool (1989) “Asset Stock Accumulation and Sustainability of Competitive
Advantage,” Management Science, 35, pp. 1504-1510.
Downsend, P. (1999) “Technology Energizes Health Insurance Distribution Process,” National
Underwriter, 103(16), pp. 35-36.
Duncan, M. and D. Garets (1999) “ The State of Healthcare IT: Business, Management and
Strategy Drivers, and Trends,” Gartner Group, Strategic Analysis Report (R-09-3928).
Egger, E. (1999) “Fleet of Innovative, New Firms Racing to Make Health Care Information
Systems Obsolete,” Health Care Strategic Management, 17(8), pp. 16-17.
Eisenhardt, K.M. (1989) “Building Theories from Case Study Research,” Academy of Management
Review, 14:4 (October) pp. 532-550.
Evans, D. (1987) “The Relationship between Firm Growth, Size and Age,” Journal of Industrial
Economics, 35, pp. 567-581.
Fox Broadcasting Network (2000) “Special Report on Internet and Healthcare,” April 26, 2000.
Page 34
Freeman, J., G. Caroll and M. Hannan (1983) “The Liabilities of Newness: Age Dependence in
Organizational Death Rates,” American Sociological Review, 48, pp. 692-710.
Gurbaxani V. and S. Whang (1991) “The Impact of Information Systems on Organizations and
Markets,” Communications of the ACM, 34(1), pp. 59-73.
Hagel, J and M. Singer (1999) Net Worth: Shaping Markets When Consumers Make the Rules.
Harvard Business School Press, Boston.
Hayek, F. A. (1945) “The Use of Knowledge in Society,” American Economic Review (35:4) pp.
519-530.
Hess, C.M. and C.F. Kemerer (1994) “Computerized Loan Origination Systems: An Industry
Case Study of the Electronic Markets Hypothesis,” MIS Quarterly, 18, 3, pp. 251-275.
Hitt, L.M. (1999) “Information Technology and Firm Boundaries: Evidence from Panel Data,”
Information Systems Research, 10(2) pp. 134-149.
Hitt, L.M. and E. Brynjolfsson (forthcoming) “Beyond Computation: Information Technology,
Organizational Transformation and Business Performance.”
Jensen, M.C. (1986) “Agency Costs of the Free Cash Flow, Corporate Finance, and Takeovers,”
American Economic Review, 76, pp. 323-329.
Jensen, M.C. and R. Ruback (1983) “The Market for Corporate Control: The Scientific Evidence,”
Journal of Financial Economics, 11, pp. 5-50.
Kaplan, S. and M. Sawhney (2000) “E-Hubs: the New B2B Marketplaces,” Harvard Business
Review, (May-June) pp. 97-103.
Karim, S. and W. Mitchell (2000) “Path-Dependent and Path Breaking Change: Reconfiguring
Business Resources Following Acquisitions in the U.S. Medical Sector, 1978-1995,”
Strategic Management Journal, 21, pp. 1061-1081.
Levin, I. P. (1998) Relating Statistics and Experimental Design. Sage Publications, Inc.,Thousand
Oaks, California, U.S.A.
Lewis, M.K. (1995) Financial Intermediaries, Elgar Reference Collection. International Library of
Critical Writings in Economics, vol 43. Alsershot, U.K: Elgar.
Lubatkin, M.H. (1987) “Merger Strategies and Stockholder Value,” Strategic Management Journal,
8(1), pp. 39-53.
Lubatkin, M.H. and H.M. O’Neill (1988) “Merger Strategies, Economic Cycles, and Stockholder
Value,” Interfaces, 18(6), pp. 65-71.
Malone, T.W., J. Yates and R.I. Benjamin (1987) “Electronic Markets and Electronic Hierarchies,”
Communications of the ACM, 30(6), pp.484-497.
Mason, J. (1996) Qualitative Researching. Sage Publications, Thousand Oaks, California, U.S.A.
Mason, R. and I. Mitroff (1973) “A Program for Research in Management Information Systems,”
Management Science, 19(5), pp. 475-487.
Page 35
McCullagh, P. (1980) “Regression Models for Ordinal Data (with discussion),” Journal of the
Royal Statistical Society, Series B, Methodological, 42, pp. 109-142.
McCullagh, P. (1998) “The Proportional Odds Model” In: Encyclopedia of Biostatisitics. John
Wiley and Sons, New York.
Michael S.C. (1994) “Competition in Organizational Form: Mail Order versus Retail Stores,
1910-1940” Journal of Economic Behavior and Organization 23(3), pp. 269-286.
Miles, M. and A. Huberman (1994) Qualitative Data Analysis. Sage Publications, Thousand Oaks,
California, U.S.A.
Milgrom, P. and J. Roberts (1987) “Informational Asymmetries, Strategic Behavior and
Industrial Organization,” The American Economic Review, 77, 2, pp. 184-193.
Mitchell, W. “The Dynamics of Evolving Markets: The Effects of Business Sales and Age on
Dissolutions and Divestitures,” Administrative Science Quarterly, 39, pp. 575-602.
Nelson, R.R. and S. Winter (1982) An Evolutionary Theory of Economic Change. Belknap Press,
Cambridge, MA.
Neter, J., M.H. Kutner et al. (1996) Applied Linear Statistical Models. WCB McGraw-Hill, Boston,
Massachusetts, USA.
Pan, Y. and P. Chi (1999) “Financial Performance and Survival of Multinational Corporations in
China,” Strategic Management Journal, 20, pp. 359-374.
Raghupathi, W. “Health Care Information Systems,” Communications of the ACM, 40(8), pp. 8082.
Rose, F. (1999) Information Age Economy: The Economics, Concept, and Design of Information
Intermediaries. Physica-Verlag, Heidelberg.
Rousseau, P. and P. Wachtel (1998) “Financial Intermediation and Economic Performance:
Historical Evidence from Five Industrialized Countries,” Journal of Money, Credit and
Banking, 30(4), pp. 657-678.
Sakar, M.B., B. Butler and C. Steinfeld (1995) “Intermediaries and Cybermediaries: A
Continuing Role for Mediating Players in the Electronic Marketplace,” Journal of
Computer-Mediated Communication, 1, 3, (jcmc.huji.ac.il/vol1/issue3/sakar.html)
Schubert, P. (2000) “Pivotal Role of Community Building in Electronic Commerce,” The 33rd
Annual Hawaii International Conference on System Sciences (HICSS-33), Maui, Hawaii,
U.S.A., IEEE.
Seddighi, H.R. , K.A. Lawler and A.V. Katos (2000) Econometrics, A Practical Approach.
Routledge, New York.
Shapiro, C. and H. R. Varian (1999) Information Rules: A Strategic Guide to the Network Economy.
Harvard Business School Press, Boston, Massachusetts, U.S.A.
Simon, H.A. (1945) Administrative Behavior: A Study of Decision Making Processes in Administrative
Organization. Free Press, New York.
Page 36
Singh, H. and C. Montgomery (1987) “Corporate Acquisition Strategies and Economic
Performance,” Strategic Management Journal, 8(4), pp. 377-386.
SPSS (1999) SPSS Advanced Regression Models 10.0. SPPS Inc., Chicago.
Spulbler, D.F. (1996) “Market Microstructure and Intermediation,” Journal of Economic
Perspectives, 10(3), pp. 135-152.
Spulbler, D.F. (1999) Market Microstructure: Intermediaries and the Theory of the Firm.Cambridge
University Press, Cambridge.
Stiglitz, J.E. (2000) “The Contributions of the Economics of Information to Twentieth Century
Economics,” The Quarterly Journal of Economics, 115, 4, pp. 1441-1478.
Stinchcombe, A. (1965) “Social Structure and Organizations” in J. March (ed.) Handbook of
Organizations. Rand McNally, Chicago, IL., pp. 153-259.
Tapscott, D. (1996) The Digital Economy: Promise and Peril in the Age of Networked Intelligence.
McGraw-Hill, New York.
Tapscott, D., D. Ticoll et al. (2000) Digital Capital: Harnessing the Power of Business Webs. Harvard
Business School Press, Boston, Massachusetts, U.S.A.
Timmers, P. (1999) Electronic Commerce Strategies and Models for Business-To-Business Trading.
John Wiley & Sons Ltd., Chichester, West Sussex, England.
Webb, J. and C. Gile (2001) “Reversing the Value Chain,” The Journal of Business Strategy, 22, 2,
pp. 13-17.
Weber, D. (1999) “Web Sites of Tomorrow: How the Internet will Transform Healthcare,” Health
Forum Journal, 42(3), pp. 40-45.
Werbach, K. (2000) “Syndication: The Emerging Model for Business in the Internet Era,”
Harvard Business Review, pp. 85-91.
Wernerfelt, B. and C.A. Montgomery (1988) “Tobin’s q and the Importance of Focus in Firm
Performance,” American Economic Review, 78(1), pp. 246-250.
Williamson, O. (1975) Markets & Hierarchies. Free Press, New York.
Williamson, O. (1985) The Economic Institutions of Capitalism. Free Press, New York.
Yates, E. (2000) “DrKoop on Life Support as AOL Takes 10% Equity Stake” The Street.com, April
25.
Yip, G. (1982) “Diversification Entry: Internal Development versus Acquisition,” Strategic
Management Journal, 3(4), pp. 331-345.
Page 37