Memorandum by Science and Technology Policy
Research (SPRU), University of Sussex, on Alternative Models for
Creating Confidence and Stimulating e-Commerce
New institutions are emerging to support electronic
commerce. Institutions in this case refer to both the rules and
norms that assist in the co-ordination and control of human behaviour
as well as to the organisations that emerge to create, promote,
and enforce such rules and norms. The evolution of these institutions
is influencing new market opportunities for e-commerce providers.
The following outlines two alternative models of the most likely
pathways for their development.
Electronic Environment Scenarios
Companies participating in e-commerce confront
the phenomenon of "Internet time", a process of extremely
rapid change as new users and services flood into the electronic
environment. The directions of these changes have important implications
for e-commerce offerings. Recent developments are leading toward
a scenario that we label the competing services model. In
this model, e-commerce companies such as financial service companies
or retailers will have to make choices between the alternative
institutions that could support their offerings and it is unlikely
that market developments will lead to a single set of technical
The competing services model suggests a more
uncertain and complex world than the outcome of the second scenario
that we label the common infrastructure model. In this
second, less likely, model, there would be widespread acceptance
of common standards and institutions. e-Commerce companies would
have more certain and straightforward platforms for their offerings.
A Framework for Analysis
Two important axes of change are affecting the
development on Internet institutions. The first axis is the design
principle employed by service providers to create new Internet
institutions, ie, whether a new service is meant to be highly
differentiated and specialised (custom) or broadly employed and
accepted (standard). The second axis is the state of competition
among these service providers. This axis indicates the extent
of competition among new Internet institutions. A service may
be broadly endorsed by market success and essentially be the dominant
service in that market, or it may be competing with other services.
Figure 1 illustrates these possibilities.
With respect to the design principle axis, service
providers may seek to develop a new institution that will become
a universal and commonly endorsed standard for the Internet. If
they are successful, that institution will be located in the lower
left or quadrant (3) of Figure 1. The design of this kind of institution
has to be endorsed broadly by the user community and other service
providers must either fail to offer an alternative, or fail to
win acceptance for their alternative from the user community.
An endorsed institution is one that is generally accepted as appropriate.
This may be augmented by formal endorsements such as those granted
to suppliers promoting the institution, by major user adoption,
or by formal voluntary public standards. The usual role of formal
mechanisms is to reinforce social and economic acceptance.
If service providers offer alternatives and
they are accepted, the result will be a proliferation of competing
standards. In this case, the new institutions would be located
in the lower right or quadrant (4) of Figure 1. This occurs because
standards that could have been universally endorsed fail to be
endorsed and become competing standards. This is likely to be
an unstable outcome. Under conditions that often apply in markets
for information goods and services, it is possible to achieve
widespread acceptance of a common standard when that standard
provides network externalities to its users. Thus, a slight advantage
in favour of one of the competing standards may be amplified into
widespread endorsement and produce an endorsed standard institution
(in the lower left quadrant (3)).
However, service providers may recognise that
differences in user needs and interests dictate a different design
principle, that of customisation. In this case they may offer
a service in quadrants 1 or 2. The service is located in these
quadrants by a mixture of the supplier's choice (how specialised
to make the service) and market conditions (the extent of rivalry).
The service provider may fully expect that other service providers
will offer alternative institutions addressing other real or imagined
user needs and interests. It is possible that this kind of "custom"
institution will be endorsed broadly by users and become a universal
solution (an institution in the lower left quadrant).
A more likely outcome is that "custom"
institutions will retain distinct characteristics. They achieve
some degree of mutual accommodation through the construction of
bridges or gateways that allow them to be interconnected. In this
case, in the upper left hand quadrant of Figure 1, there is an
endorsement of a cluster of interconnected, but partially customised,
institutions (rather than a universal standard). Such mutual accommodation
is unstable. It is likely to be resolved in one of two ways. The
institution may become a standard despite the fact that it was
designed according to the custom design principle. The most likely
possibility is that competing service providers will further differentiate
the institution that they are sponsoring. In this case, fully
competitive, alternative custom institutions will be offered (the
upper right hand quadrant (2) of Figure 1).
To summarise, a position in either the upper
left (1) or lower right (4) quadrant of Figure 1 is unstable.
Thus, the lower left (3) or upper right (2) quadrants are the
regions in which institutions are most likely to be located as
a result of the interaction between the design principles (standardisation
and customisation) and the competitive process. The directions
for the maturation of these institutions are illustrated in Figure
The Current State of Play
Figure 3 illustrates the current position of
some of the key institutions for e-commerce.
The technical institutions concerned with basic
connectivity and some services such as e-mail have achieved widespread
acceptance or "endorsement". They are located to the
left of Figure 3. Examples are the technical standards used for
connecting to the Internet and World Wide Web (respectively, the
Internet Protocol (IP) and Hyper Text Mark-up Language (HTML)).
Simple text e-mails organised around the Simple
Mail Transfer Protocol (SMTP) technical standards have been widely
accepted. However, e-mail communication is best represented as
an interconnected set of services. This is because SMTP allows
users to attach messages, some of which other users will find
indecipherable because they do not have the appropriate application
to "decode" the contents of the message. Thus, the institution
of e-mail (which includes both simple text messages and attachments)
is an example of an endorsed (in widespread use) custom institution.
This is an unstable position. The most likely direction of change
is the further differentiation of file formats attached to e-mails
which implies movement toward the upper right quadrant of Figure
3. Somewhat less likely is the development of a common application
for accessing attachments (independent of the application used
to create the attachment) in which case e-mail would move toward
a location in the lower left portion of Figure 3.
In the case of virtual community institutions,
although there are some common standards, many alternatives already
exist. It is likely that these alternatives will continue to compete
due to several factors including the diversity of user needs.
Virtual communities are good examples of the `competing services
Intelligent agents such as search engines have
the potential to become a universal standard with broad endorsement
although this has not happened as yet. Instead, several approaches
to intelligent agents are vying for a share of the e-commerce
market. This is an unstable position as suggested by the disproportionate
market share of Yahoo! Yahoo!'s success indicates the possibility
of the emergence of a common standard in the search engine type
of intelligent agent. Movement in this area is almost entirely
based upon patterns of user endorsement. There is little doubt
that specialised search engine applications will continue to exist.
The business model supporting the operation of specialised service
providers will need to change if trends toward growing endorsement
of the Yahoo! common standard continue.
It is less clear where trust service provision
should be located. There is some optimism that an endorsed standard
will emerge based upon government or private sector initiatives.
However, if these alternatives fail, there is likely to be movement
toward further customisation of this institution. This could lead
to the location of such institutions in the upper right quadrant
of Figure 3.
This framework provides a basis for examining
the two emerging scenarios for the evolution of e-commerce. The
first scenario, the common infrastructure model, is shown in Figure
In this scenario, there is a broad endorsement
of common institutions as standards for the conduct ofe-commerce.
Trust services are standardised around a single model. Broad endorsement
of a common standard for intelligent agents leads to a strong
position of a few (or a single) approaches to intelligent agent-based
services such as portals. The advantages of interconnectivity
lead to a broad acceptance of standards for message communication
and the development of universal standards beyond SMTP. Even in
the area of virtual communities, common standards offer advantages.
For example, one possibility is the creation of virtual identities
that would allow an individual to be recognised across different
virtual communities. Such a standard would allow individuals to
accumulate reputation, credit worthiness and other attributes
that could be recognised outside their local virtual community,
while preserving their privacy.
This scenario has numerous advantages for companies
that are attempting to build e-commerce businesses. The most important
is a reduction in the number of platforms or configurations of
institutions that these companies have to evaluate and maintain
to provide services. The simplifications offered by this model
are also important in reducing user confusion and uncertainty
about how to interact with service providers. It is not surprising
that there are frequent statements in the trade press and in government
policy forums about the desirability of this type of model.
The alternative scenario, the competing services
model, is shown in Figure 5. It is the more likely outcome of
the interaction between the design principles and competition
in the electronic environment.
The driving force favouring this scenario is
the quest for variety. Variety is not an end in itself but a means
of catering for diverse user needs. The enormous flexibility allowed
by information and communication technologies to deliver customised
services is both a benefit and a cost in the development of new
e-commerce institutions. As a benefit, it provides the opportunity
to tailor services according to individual preferences and capabilities.
For example, novice users can be treated differently than "power"
users. The possibilities for differentiation are endless. At the
centre of these developments is the process of creating virtual
communities in which people are able to conduct their business
or pursue their interests in different ways.
The competing services model scenario is consistent
with the view that the Internet represents an anarchistic community
where the history of respected institutions offers little advantage.
This view has little validity however. This is because there are
clear advantages for many services providers and users in a more
systematic integration of services under common standards. Although
early users of the Internet were over-represented by those with
anti-establishment sympathies, the floor of new users is making
the Internet user base somewhat more representative of the general
population. Government policy, the interests of large users, and
the interests of many service providers, tend to support the growth
of common standards. For this reason, the common infrastructure
model is a credible alternative to the competing services model.
The social process of constructing e-commerce
institutions provides junctures at which it is difficult to achieve
closure in establishing a common framework for technical and institutional
standards. Public authorities are seeking to promote an institutional
framework that will support the rapid development ofe-commerce.
They often seek to do so based on the premise that a prevailing
institutional standard fore-commerce will emerge from market competition
or that such a standard can be promoted through selective policy
intervention. In the case of virtual communities and intelligent
agents, no such institutional standard appears likely to emerge
in the short and medium term. In the case of trust services, Government
efforts to provide an institutional framework for delivery of
trust services are often regarded by providers as incidental to
the issues they are facing in building trust service institutions.
Each of these scenarios represents a possible
path of evolution for e-commerce institutions but the competing
services model is the path of most likely development. In the
industry-favoured competing services model, the main objective
is for trust service providers to establish ways of building trust
with users and of resolving potential conflicts as and when they
arise. Because many institutions already perform this role, this
approach is regarded as favouring the entry of incumbents and
new players into this market. Establishing a new regime for governing
e-commerce is not a simple matter of replicating the ground rules
for trading in physical markets. Government interests in national
security, legal enforcement of contracts, individual privacy,
and the competitiveness of national markets within a global trading
environment, point to the likelihood that some form of public
common infrastructure model will emerge. In the short and medium
term, however, the predominance of the competing services model
will encourage diversity and variety in service offerings but
it may also create sufficient uncertainty in the potential user
population for business-to-consumer e-commerce to slow the growth
of new markets.
22 May 2000
91 This text is based on R Mansell, I Schenk
and WE Steinmueller (2000 forthcoming) Net Compatible: The
Economic and Social Dynamics of Electronic Commerce, Communications
& Strategies, presented at the EURO Communication Policy Research
Conference, Venice, 26-28 March 2000. Back