<<

. 7
( 15)



>>

• Audio recordings of music or spoken words (for example, lectures) presented on
Web sites (Manjoo, 2003)
• Financial market reports converted to financial information Web sites or Internet-
delivered information feeds using, for example, technologies such as Really Simple
Syndication (Gillmor, 2003)


Characteristics of Digital Products

Most digital products have common characteristics that identify them as digital prod-
ucts. These characteristics distinguish them from physical products or intangible
products without a digital existence, such as financial instruments. These characteristics
include:
• High fixed cost to produce the first unit, but low marginal costs to produce
subsequent units,
• Quality is difficult to judge without actually experiencing the product,



Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
156 Schneider


• Capacity constraints do not limit production output in any significant way, and
• Storage, retrieval, and forwarding the product is easy and inexpensive to do.


One consequence of these characteristics is that unauthorized use by purchasers of the
product (including the forwarding of the digital product to other parties who have not
paid for the product) can be difficult or impossible to control. A second consequence is
that sellers of such products must adopt different strategies from those used by sellers
of physical products to ensure that a revenue stream flows from the continuing use of
the digital product.
Many digital products are, in their essence, things that are experienced by customers.
They often have no meaningful physical existence separate from the experience.
Krishnamurthy (2003) discusses this characteristic of digital products as being either
experience or credence goods. Unlike an item of clothing, which can be examined before
purchase in a physical store, an experience good requires that the customer be exposed
to the product before determining its quality. An artwork or performance is an experience
good because the customer cannot judge its quality without experiencing it. A credence
product is even more complex. A customer often cannot judge the quality of a credence
good even after experiencing it. For example, the quality of a physician™s services could
be difficult for a person without medical training to judge even after the services have
been rendered. Purchasers rely on third-party reviews of experience and credence goods
for pre-purchase information.


Changing Technologies and Digital Products

Providers of digital products must maintain a current knowledge of underlying technolo-
gies that are used or could be used in the future for delivery of their products. One serious
case of vendors™ failure to keep up with delivery and transmission technologies is the
music recording industry, which grossly underestimated the impact of file-size compres-
sion technologies and increasingly inexpensive Internet bandwidth (Christman, 2002;
Dahl, 2003; Lee and Capell, 2003). The ability of customers to adapt and reformat digital
products is also an essential characteristic of digital products”a characteristic that can
be affected by changes in technologies, as well.




Pricing and Distribution of Digital
Products
Issues regarding pricing and distribution control of digital products arose before the
Internet and the World Wide Web (Web) became prevalent. However, the availability of
an inexpensive and near-immediate electronic transmission medium has added new
issues and complicated existing issues. These pricing and distribution issues affect the
nature, quantity, and quality of competition in markets for these products.


Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
Digital Products on the Web 157


Digital products require an approach to pricing that differs from that used for physical
products. Some digital products are made available at no charge, thus, an alternative
revenue stream that is somehow related to the product must be devised. Some digital
products are bundled with other products (digital or physical) to avoid some of the
problems inherent in the pricing of digital products alone. Another pricing strategy is to
create an artificial distinction within a subset of digital products and use differential
pricing to extract the highest revenue possible from each set of customers for the product.
Perhaps the most common pricing method is to use a licensing approach of one kind or
another.
On the Web, combinations of all these pricing methods are often seen. For example, Web
sites often make free content available and charge for other, related content. This is a
combination of the “no charge” approach with the differential pricing approach.
Although a number of studies (Bailey, 1998; Bakos, 1997; Brynjolfsson and Smith, 2000;
Lee, 1998) have been published regarding pricing on the Internet, these studies consider
the Internet as a part of the marketing and pricing mechanism in a general way for all types
of goods sold on the Internet. They do not address the specific issues that arise when
selling digital products on the Internet. Although these general analyses of the impact
of the Internet on marketing channels do conclude that in many cases the transaction
costs in the channel are reduced, the studies are less conclusive on the question of which
parties in the channel are able to reap the benefits of those transaction cost reductions
(Schneider, 2004). For most digital products, however, the real effect on pricing and
distribution strategy does not derive from the introduction of the Internet into the
marketing channel, but from the products™ very nature as digital products. In this section,
specific pricing and distribution strategies for digital products are outlined.


Low Price or Barter Strategies

Many digital products are offered on the Internet in a way that makes them appear to be
free. In almost all cases, however, there is a low price exacted. In general, this price is non-
monetary. In some cases, the site visitor obtains the digital product in exchange for
personal information, which the site is then able to use for marketing or other purposes.
In other cases, the site visitor agrees to have advertising appear in his or her Web browser
window. In effect, the customer is bartering personal information or time spent viewing
advertisements in exchange for the digital good.
Examples of these types of digital products and services include sites such as The New
York Times. The Times provides news stories in exchange for a site visitor™s registration,
which discloses a small amount of personal information, and for the visitor™s willingness
to view advertisements that are included on the Web pages that display the digital
product (the news stories). Many other newspapers, magazines, news services, and
other online information portals use this pricing and distribution strategy to sell their
digital products.
Another example of a low-price or barter strategy is the offering of “free” e-mail accounts.
Companies such as Yahoo! and Microsoft (through its Hotmail business) offer a limited
personal e-mail service. The companies do not charge for the service, but do collect



Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
158 Schneider


personal information that they use for marketing purposes. The e-mail service is provided
on Web pages that include targeted advertising. The advertising is targeted based on
the personal information collected.
In most cases, the barter transaction is fully disclosed. In some cases, however,
companies offer “free” information or software services but include in the software a
feature that collects information from the customer™s computer and reports it back to the
seller for as long as the software is installed. This type of software is called stealthware
or spyware because it often installs the reporting feature in a way that makes it difficult
for the customer to detect its operation (Hagerty and Berman, 2003; Metz, 2003).
Sometimes, a digital product is given away so that customers can try it before purchasing
it. Sites that sell information by subscription often offer a 30-day free trial period.
Software is often sold this way. Called “shareware,” the software is provided for
download and can be used for a limited time. After the trail period expires, the software
will either disable itself or launch periodic reminders (called nag boxes) to register and
pay for the software. Some shareware does not include any type of disabling code or
programmed reminders. The vendors simply ask users who like the product to contribute
money. This has not been a terribly effective way to market digital products, but it was
widely used in the early days of the Internet and it does still continue today as a
distribution model for some software products (Liao-Troth and Griffith, 2002).


Subscription Strategies

In a subscription arrangement, the customer agrees to pay for access to content or the
use of a digital service over time. Companies that offer free e-mail services, such as
Yahoo!, often also offer additional services such as increased disk storage space or the
right to send and receive larger sized e-mail messages or attachments on a paid
subscription basis.
Some newspapers and magazines also offer paid subscriptions to Web site content. This
Web content is distinguished from the content that is offered at no charge by either being
additional content not available to non-subscriber site visitors, or it is content that is
offered free of the advertising messages that accompany free content on the site.
Subscriptions are attractive to sellers because they reduce the administrative costs
associated with tracking and billing individual consumption of digital products. Sub-
scriptions can also appeal to customers because they provide a simple pricing arrange-
ment that has a known and certain price (Fishburn and Odlyzko, 1999). In some cases,
such as AOL™s adoption of a fixed subscription price in 1996, the known and certain price
can lead customers to consume greater quantities of what they perceive to be free digital
goods. Once they have paid the subscription fee, their marginal cost of consuming
additional units of the goods is near zero (Roth, 1998).


Differential Price Strategies

Each person wanting to buy a product has a maximum price that he or she is willing to
pay for it. This price, often called a reservation price, is known to the buyer but not to

Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
Digital Products on the Web 159


the seller. Much of the seller™s pricing task is to estimate potential purchasers™ reserva-
tion prices and set the price of the product low enough to attract a large number of
purchases, but no lower than necessary (Edwards, 1942).
One way for a seller to extract the highest possible reservation prices from a diverse group
of potential purchasers is to segregate them into multiple categories and then charge a
different price to purchasers in each category (Bichler and Loebbecke, 2000; Forsyth,
Lavoie and McGuire, 2000). For example, some manufacturers sell their products primarily
to wholesalers and retail stores, but also sell direct to end-user customers from time to
time. These manufacturers typically charge a lower price to their wholesale customers
than to their end-user customers. If sellers can identify any reasonable basis for
discriminating in their pricing, they will optimize their profits if they can charge different
prices to different customers.
The ultimate in price discrimination is charging a different price to every single customer
(Peppers and Rogers, 1997). The Web, with its ability to identify site visitors and
customize the shopping experience, makes this not only feasible, but relatively easy to
accomplish (Godin and Peppers, 1999).
Some countries, such as the United States, have laws that prohibit certain types of price
discrimination (Purchasing Law Report, 2001). However, most price discrimination is
legal throughout the world. The ethics of price discrimination are open to debate
(Campbell, 1999) and companies such as Amazon.com have faced highly critical customer
reactions and press coverage when they have been identified as engaging in price
discrimination (Adamy, 2000; Cox, 2001).
One form of price discrimination that works very well with digital products is versioning
(Krishnamurthy, 2003). Information content on the Web can be versioned by:
• Offering a version of the product without advertising or with a smaller amount of
advertising or advertising in forms that are less obtrusive. The low-advertisement
content version is sold for a higher price or made available only to subscribers.
• More current information offered to customers who pay higher prices or who
subscribe. Those paying lower prices or not subscribing to the site receive
information after a time delay.
• A more complete feature set is offered at a higher price. For example, a news site
might include streaming video for subscribers.
• Varying the quality of the offering. Many sites that sell or lease graphics for use
on Web pages or in print applications offer a small, low-resolution picture on the
Web site at no cost or at a very low price. This graphic might also include a visible
mark that reduces the value of the image. For a higher price, customers can buy a
higher resolution version of the graphic.


These versioning strategies do not work well if the content can be converted readily. That
is, if a low-price version of the digital product can be converted to the high-price version
easily, customers will not see any reason for the price differential and will object to it
(Shapiro and Varian, 1998).



Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
160 Schneider


Product Bundling Strategies

Some products can be offered in combinations, or bundles. For example, a music CD is
a bundle of individual song recordings. A newspaper is a bundle of stories and
advertisements. A software suite is a bundle of individual programs. Research (Venkatesh
and Mahajan, 1993) has shown that if customer preferences are not homogeneous, the
prices of the individual items in the bundle are not close to each other, and a significant
proportion of customers are indifferent to a large number of the products in the bundle,
a mandatory bundling of products can yield greater profits than offering the products
separately (Estalami, 1999).
If the individual products function better as a bundle, as might occur with software
programs offered as a suite, the seller can even charge a premium for the bundle (Estelami,
1999). Companies offering digital products on the Web can easily bundle products and
services (Sieber and Sabatier, 2003) and can charge premiums for complementary
products in different amounts to different customers, thus combining the product-
bundling strategy with the price discrimination strategy described above (Venkatesh and
Kamakura, 2003).




Revenue Models for Digital Products
Companies have combined the basic pricing and distribution strategies described in the
preceding section into a number of different revenue models that they are currently using
to sell digital products on the Internet. These include subscription-based models,
advertising-supported models, per-item sales models, and a variety of mixed models
(Schneider, 2003). This section discusses specific industries that sell digital products
and provides a brief description of how each industry is using various revenue models.
According to Cohan (2001), Internet business models must be based on selling one or
more of four digital product elements to customers. These four elements include:
charging for access to content, charging for copies of content, charging to transmit
messages to current or potential customers, and charging for transactions that result
from exposure to or consumption of content.


Newspaper and Magazine Sites

Many newspapers and an increasing number of magazines publish all or part of their print
content on the Web. It is unclear whether a newspaper™s presence on the Web helps or
hurts the newspaper™s business as a whole. Choi (2003) argues that the reputation of an
old product (the print newspaper, in this case) can carry over to a new product (the news
Web site, in this case) if the products are bundled. However, other researchers, such as
Cripps and Schmidt (1996) have argued that the reputation transfer is unpredictable.




Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
Digital Products on the Web 161


Advertising Revenue Model

Although news Web sites can provide greater exposure for newspapers™ names and can
provide larger audiences for the advertising that the newspapers carry, they can also
cannibalize sales from print editions. Like retailers or distributors whose online sales lead
to cannibalization of their brick-and-mortar sales, publishers also can experience sales
losses as a result of online distribution. Newspapers and other publishers worry about
these sales losses, because they are very difficult to measure. Some publishers have
conducted surveys in which they ask people whether they have stopped buying the
newspaper because the content they want to see is available online. In addition to the
concern about lost sales of print editions, most newspaper publishers have found that
the cost of operating their Web sites cannot be covered by the revenue they can generate
from selling advertising on the site. Thus, many newspaper publishers have experi-
mented with various other ways of generating revenue from their Web sites.
The advertising revenue model is used by network television in the United States.
Broadcasters provide free programming to an audience along with advertising messages.
The advertising revenue is sufficient to support the operations of the network and the
creation or purchase of the programs. Many observers of the Web in its early growth
period believed that the potential for Internet advertising was tremendous. Web adver-
tising had grown from essentially zero in 1994 to $2 billion in 1998 (Sharples, 1999).
However, Web advertising has been flat or declining from 2000 until quite recently. In
2002, Web advertising revenues began to increase again, but at a fraction of their former
growth rate (Tynan and Gilbert, 2003). After some years of experience in trying to develop
profitable advertising revenue models without the growth rates of the early years of the
Web, many companies are less optimistic about the potential for advertising as the sole
basis for revenue generation.
The success of Web advertising has been hampered by two major problems. First, no
consensus has emerged on how to measure and charge for site-visitor views. Since the
Web allows multiple measurements, such as number of visitors, number of unique
visitors, number of click-throughs, and other attributes of visitor behavior, it has been
difficult for Web advertisers to develop a standard for advertising charges. In addition
to the number of visitors or page views, stickiness is a critical element to creating a
presence that will attract advertisers. The stickiness of a Web site is its ability to keep
visitors at the site and to attract repeat visitors. People spend more time at a sticky Web
site and are thus exposed to more advertising.
The second problem is that very few Web sites have sufficient numbers of visitors to
interest large advertisers. Most successful advertising on the Web is targeted to very
specific groups. The characteristics that marketers use to group visitors is called
demographic information, and includes such things as address, age, gender, income
level, type of job held, hobbies, and religion. It can be difficult to determine whether a
given Web site is attracting a specific market segment unless that site collects demo-
graphic information from its visitors, which is information that visitors are increasingly
reluctant to provide because of privacy concerns.




Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
162 Schneider


Subscription Revenue Model

One alternative to a pure advertising revenue model is to add the sale of subscriptions
to the information content of the Web site. Few newspaper or magazine publishers use
a pure subscription model. Most of these businesses that sell subscription access to their
sites use a combination of advertising and subscription revenues, much as print
newspapers and magazines do. Consumers Union, the publisher of the monthly product
evaluations and ratings magazine Consumer Reports, is an exception to this rule. It does
operate a Web site that relies heavily on subscriptions. Consumers Union is a not-for-
profit organization that does not accept advertising as a matter of policy (because it might
appear to influence its testing and research results for the products of its advertisers or
their advertisers™ competitors). Therefore, the site is supported by a combination of
subscription revenue and some donations. The Web site does offer some free information
as a way to attract subscribers and to fulfill its organizational mission of encouraging
improvements in product safety, but this is not a revenue generator for the organization.


Advertising-Subscription Combination Model

In the advertising-subscription combined revenue model used by most online newspa-
pers and magazines, subscribers pay a fee and accept some level of advertising. On Web
sites that use the advertising-subscription revenue model, subscribers are typically
subjected to much less advertising than they are on advertising-supported sites. Firms
have had varying levels of success in applying this model and a number of companies
have moved to or from this model over their lifetimes.
Two leading newspapers, The New York Times and The Wall Street Journal, use a
combined advertising-subscription model. The New York Times version is mostly
advertising-supported with a small subscription fee for visitors who want full access to
enhanced online versions of the newspaper™s crossword puzzles. The New York Times
also provides a searchable archive of past articles and charges a small fee for access to
non-current articles. The Wall Street Journal™s combination model is weighted more
heavily to subscription revenue. The site allows non-subscriber visitors to view the
classified ads and certain stories from the newspaper, but most of the content is reserved
for subscribers who pay an annual fee for access to the site. Visitors who already
subscribe to the print edition are offered a reduced rate on subscriptions to the online
edition.
Note that both of these newspapers use one version of this revenue model for their print
edition and another version for their online editions. Increasingly, newspapers and
magazines are finding that they need to use different revenue models for their print and
online editions. Other newspapers, including The Washington Post and the Los Angeles
Times, use another variation of the combination revenue model. These newspapers do
not charge any subscription fees for access to their Web sites. Instead, they offer current
stories free of charge on their Web sites, but require payment for older articles.
Business Week offers yet another combination revenue variation. It offers some free
content at its Business Week online site, but requires the purchase of a subscription to


Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
Digital Products on the Web 163


the Business Week print magazine if visitors want to gain access the entire site.
Subscribers who want to read archived articles that are more than five years old are levied
an additional charge per article. Business Week does place content in the subscriber
section of its Web site before the magazine appears on the newsstands or is delivered
to subscribers.
Sports fans visit the ESPN site for all types of sports-related information. Leveraging its
brand name from its cable television businesses, ESPN is one of the most visited sites
on the Web. It sells advertising and offers a vast amount of free information, but serious
sports fans can subscribe to its Insider service to obtain access to even more sports
information. Thus, ESPN uses a revenue model that includes advertising and subscrip-
tion revenue, but it only collects the subscription revenue from Insider subscribers, who
are a small percentage of site visitors.


Classified Advertising Sites: An Advertising Revenue
Model

Although attempts to create general-interest Web sites that generate sufficient adver-
tising revenue to be profitable have met with mixed results, sites that target niche markets
have been more successful. For newspapers, classified advertising is very profitable.
Thus, Web sites that specialize in providing only classified advertising do have profit
potential. This is especially true if they can reach a narrow target market and charge higher
rates because the advertising reaches the right audience.
One implementation of the advertising-supported revenue model that does appear to be
successful is Web employment advertising. As the number of people using the Web
increases, these businesses will be able to move out of their current focus on technology
and higher-level jobs and include advertising for all kinds of positions. These sites can
use the same approach that search-engine sites use to offer advertisers target markets.
When a visitor specifies an interest in, for example, engineering jobs in Dallas, the results
page can include a targeted banner ad for which an advertiser will pay more, because it
is directed at a specific segment of the audience.
Employment ad sites can also target specific categories of job seekers by including short
articles on topics of interest. These articles increase the site™s stickiness and it helps draw
people to the site who are not necessarily looking for a job. This is a good tactic because
people who are not looking for a job are often the candidates most highly sought by
employers. Classified employment advertising site Monster.com includes links to ar-
ticles, reports, a message board, and chat sessions that might interest employees at
various levels. It also offers a variety of newsletters tailored to employees at various
levels in their current positions.
Another type of classified advertising Web site that can generate sufficient revenue to
be profitable is the “used vehicle” site. Trader Publishing has printed advertising
newspapers for many years and now operates a number of vehicle classified advertising
sites under names such as AutoTrader.com, CycleTrader.com, BoatTrader.com, and
AeroTrader.com. These sites accept paid advertising from individuals and companies



Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
164 Schneider


that want to sell cars, motorcycles, boats, and airplanes. Trader Publishing charges a fee
for each listing and gives the seller the option of running the ad on the Web site only
or on the Web and in the print version of the advertising newspaper. If the product has
a dedicated following, this type of site can be successful by catering to small audiences.
For example, the VetteFinders site sells classified ads for Corvette automobiles only.
In 2002, another classified advertising category emerged, the dating services and
personal ads sector. This category yielded more than $87 million in sales revenue in 2002
and is expected to continue growing rapidly (Smith, 2003).


Information Providers: Subscription Revenue Models

Some companies that own intellectual property or the rights to such property have begun
using the Internet as an important distribution channel for selling intellectual property
rights as digital products. For example, LexisNexis began as a legal research tool that has
been available as an online product for years. Today, LexisNexis offers a variety of
information services, including legal information, corporate information, government
information, news, and resources for academic libraries (Electronic Information Report,
2001). The original legal information product exists on the Web today as Lexis.com and
provides full-text search of court cases, laws, patent databases, and tax regulations. In
the past, law firms had to subscribe and install expensive dedicated computer systems
to obtain access to this information.
The Web has given LexisNexis customers much more flexibility in how they purchase
information. Through the Lexis.com Web site, law firms can subscribe to several versions
of the service that are customized for different firm sizes and usage patterns. The Web
site even offers a credit card charge option for infrequent users who do not want a
subscription. LexisNexis has used the Web to improve the delivery and variety of its
existing product line and has been able to devise new products that take advantage of
the Web™s features. Chung and Rao (2003) outline a general model for using product
bundling in market segmentation applications such as the LexisNexis example.
ProQuest, a Web site that sells digital copies of published documents, has its roots in
two businesses: the former Bell and Howell learning materials business and University
Microfilms International (UMI). These firms had acquired reproduction rights to a variety
of published and unpublished materials (Hane, 1999). For example, UMI had contracts
with most North American universities to publish all doctoral dissertations and masters
theses on demand. ProQuest offers digital versions of these documents for sale, along
with a number of newspapers, journals, and other specialized academic publications.
Many schools and libraries have subscriptions to ProQuest. Other companies, such as
SilverPlatter Information and EBSCO Information Services sell subscriptions to digital
versions of journals and books to corporate and university libraries. These companies
also sell access to bibliographic databases and electronic journals to individuals,
schools, companies, and libraries.
Dow Jones is a major business publisher that has sold subscriptions to digitized
newspaper, magazine, and journal content for a number of years. The company offered




Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
Digital Products on the Web 165


a customized digital clipping service that provided subscribers with a daily e-mail
message of news on topics of interest to them (O™Leary, 1998). Dow Jones and Reuters,
a British company, joined in 2002 to create an online content management and integration
service called Factiva. Factiva now provides the same types of services that Dow Jones
had provided, but also gives companies the ability to integrate their existing content
(such as a corporate library) with Dow Jones and Reuters news sources (Conhaim, 2002).
One of the first academic organizations to use Internet distribution was the Association
for Computer Machinery (ACM) with its ACM Digital Library. This Web site offers
subscriptions to electronic versions of ACM journals to its members and to library and
institutional subscribers (White, 2001). Academic publishing has always been a busi-
ness with thin margins because the number of potential subscribers is much smaller than
more mainstream publications. Even the most highly regarded academic journals often
have fewer than 2,000 subscribers. However, the fixed costs of publishing remained very
much the same for all publishers, both large and small. To break even, academic journals
often must charge hundreds or even thousands of dollars for a one-year subscription
(Parks, 2002). Electronic publishing eliminates the high costs of paper, printing, and
delivery, and makes dissemination of research results less expensive and more timely.
As was the case for other technologies, such as VCRs and subscription cable television,
many of the early commercial users of Web technology were dealers in adult-themed
entertainment material. Many of the first profitable sites on the Web were sellers of adult
digital content (Simons, 1996). These sites pioneered the processing of credit card
payment transactions on the Internet and many different digital video technologies that
are now used by all types of businesses on the Web.


Online Games: Subscription and Combination Revenue
Models

Computer and video games have become a huge industry, with revenue now exceeding
that of the film industry. An increasing portion of the industry™s revenue is generated
online. In the past, many sites that offered games relied on advertising revenue. A
growing number of these sites now include subscription and fee-for-play games. Some
sites sell software that gamers must buy and download to play the games, others require
payment of a subscription fee to enter the fee-for-play games area on the site. Microsoft™s
MSN Games, Sony™s Station.com, Electronic Arts™ EA.com, and RealNetworks™ RealOne
Arcade are among the leading game sites that include subscription game services. For
example, Sony™s EverQuest adventure game has drawn more than 400,000 players who
have purchased a $40 software package and pay $10 per month to continue playing the
game (Kushner, 2002). Most of the game sites charge a monthly subscription of between
$5 and $20 for access to all their fee-based, games offerings. The Interactive Digital
Software Association estimates that about 170 million people in the United States alone
are regular game players, a number that is growing about 17 percent each year (McLaughlin,
2001).




Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
166 Schneider


Concerts and Films: Subscription Revenue Models

As more households obtain broadband access to the Internet, an increasing number of
companies will provide streaming video of concerts and films to paying subscribers. With
a revenue model patterned after cable television companies, firms such as Intertainer and
RealNetworks are selling subscriptions for delivery of video content to computers and
other devices through cable modem and DSL connections.
The main technological limitation these companies face is that each additional customer
who downloads a video stream requires that the provider purchase additional bandwidth
from its ISP. Television broadcasters, on the other hand, need only pay the fixed cost of
a transmitter because the airwaves are free and carry the transmission to an unlimited
number of viewers at no additional cost. In contrast, as the number of an Internet-based
provider™s subscribers increases, the cost of the provider™s Internet connection in-
creases. However, if these Web entertainment companies can charge a high enough
monthly fee, they should be able to cover the additional costs of technology upgrades
and still make a profit.




Revenue Models in Transition
Many companies have gone through transitions in their revenue models for digital
products. As more people use the Web and as their behavior changes, some companies
have altered their revenue models to meet their customers™ needs. This section describes
the revenue model transitions undertaken by five different companies as they gained
experience in the online world and as they faced the changes that occurred in that world.
These and other companies might well face the need to make further adjustments to their
revenue models in the future.


Transition: Subscription to Advertising

Microsoft created an online magazine, Slate, to provide news and current events
information. Slate included experienced writers and editors on its staff and industry
observers expected the magazine to be a success. Microsoft felt that the magazine had
a high value, too. At a time when most online magazines were using an advertising-
supported revenue model, Slate charged an annual subscription fee after a limited free
introductory period.
Although Slate drew a wide readership and received acclaim for its quality reporting and
writing, it was unable to draw a sufficient number of paid subscribers. At its peak, Slate
had about 27,000 subscribers generating annual revenue of $500,000, which was far less
than the cost of creating the content and maintaining the Web site (Sanderfoot and
Jenkins, 2001).




Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
Digital Products on the Web 167


Slate is now operated as an advertising-supported site. Because it is a part of Microsoft,
Slate does not report detailed profit numbers, but the magazine announced that it was
profitable as of April 2003 on revenues of $7 million (Carr, 2003). Microsoft maintains the
Slate site as part of its MSN portal, so it is likely that Slate increases the stickiness of the
portal, which is an additional benefit to Microsoft not reflected in the accounting
determination of unit profits for Slate.


Transition: Pure Advertising to Combination Revenue

Another online magazine, Salon.com, that has also received acclaim for its content, has
moved its revenue model in a direction opposite to that of Slate. After operating for
several years as an advertising-supported site, Salon.com now offers an optional
subscription version of its site. The subscription offering was motivated by the
company™s inability to raise the additional money from investors that it needed to
continue operations (Sanderfoot and Jenkins, 2001).
Subscribers pay $30 per year for access rights to a version of the magazine called Salon
Premium that is free of advertising and that can be downloaded for storage and later
offline reading on the subscriber™s computer. Premium subscribers also gain access to
additional content such as downloadable music, e-books, and audio books.


Transition: Advertising to Fee-for-Services

Xdrive Technologies opened its original advertising-supported Web site in 1999. Xdrive
offered free disk storage space online to users. The users would see advertising on each
page and had to provide personal information that would allow Xdrive to send targeted
e-mail advertising to them. Its offering was very attractive to Web users who had begun
to accumulate large files, such as MP3 music files, and who wanted to access those files
from several computers in different locations (Schwarz, 2001).
After two years of offering free disk storage space, Xdrive found that it was unable to
pay the costs of providing the service with the advertising revenue it had been able to
generate. It switched to a subscription-supported model and began selling the service
to business users as well as individuals. The amount of the monthly subscription is based
on the amount of disk space reserved for the user and on the number of people who have
access to the disk space (Schuchart, 2003).


Transition: Advertising to Subscription

Northern Light was founded in 1995 as a search engine with a twist. In addition to
searching the Web, it searched its own database of journal articles and other publications
to which it had acquired reproduction rights. When a user ran a search, Northern Light
would return a results page that included links to Web sites and to abstracts of the items




Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
168 Schneider


in its own database. Users could then follow the links to Web sites, which were free, or
could purchase access to the database items (Kalin, 1997).
Thus, Northern Light™s revenue model was a combination of the advertising-supported
model used by most other Web search engines plus a fee-based information access
service similar to the subscription services offered by ProQuest and EBSCO mentioned
earlier in this chapter. The difference in the Northern Light model was that users could
pay for just one or two articles (the cost was typically between one and five dollars per
article) instead of paying a large amount of money for unlimited access to its database
on an annual subscription basis. Northern Light did also offer subscription access to
most of its database to companies, schools, and libraries, however.
In January 2002, Northern Light decided that the advertising revenue it was earning from
the ads it sold on search results pages was insufficient to justify continuing to offer that
service. It stopped offering public access to its search engine and converted to a new
revenue model that is primarily subscription-supported (Kontzer, 2002).


Multiple Transitions

Encyclopedia Britannica is a company that transferred its existing reputation for high
quality to the Web (Schneider, 2004). The Encyclopedia Britannica has developed a very
respected brand name in research and education over its many years in print publishing.
Britannica began in the late 1700s when a group of academics collected notes they had
made while conducting research and decided to publish them as a series of articles. After
more than 200 years of successful print publications, Britannica moved to electronic
distribution with a CD offering in the mid-1990s (Ferris, 1996).
About the same time, Britannica launched its online presence with two Web offerings.
The Britannica Internet Guide was a free Web navigation aid that classified and rated
information-laden Web sites. It featured reviews written by Britannica editors who also
selected and indexed the sites. The company™s other Web site, Encyclopedia Britannica
Online, was available for a subscription fee or as part of its Encyclopedia Britannica CD
package. Britannica used the free site to attract users to the paid subscription site
(Conhaim, 2001).
In 1999, disappointed by low subscription sales, Britannica converted to a free, adver-
tiser-supported site (Schneider, 2004). The first day the new site, Britannica.com, became
available at no cost to the public, it had over 15 million visitors, forcing Britannica to shut
down for two weeks to upgrade its servers (Book Publishing Report, 1999). The
Britannica.com site offered the full content of the print edition in searchable form, plus
access to the Merriam-Webster™s Collegiate Dictionary and the Britannica Book of the
Year. One of the most successful aspects of the site was the way it integrated the
Britannica Internet Guide Web-rating service with its print content. The Britannica Store
sold the CD version of the encyclopedia along with other educational and scientific
products to help generate revenue.
After two years of trying to generate a profit using advertising to generate revenue,
Britannica succumbed to its insufficient advertising revenues (Schneider, 2004). In 2001,
Britannica returned to a combination model in which it offered free summaries of


Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
Digital Products on the Web 169


encyclopedia articles and access to the Merriam-Webster™s Collegiate Dictionary on the
Web, but placed the full text of the encyclopedia in a restricted part of their Web site.
Full text access was available for a subscription fee of $50 per year or $5 per month
(DiSabatino, 2001).
Schneider (2004) notes that Britannica has undergone four major revenue model transi-
tions, from being a print publisher to a seller of information on the Web to an advertising
revenue-supported Web site to a combination advertising subscription model, in just a
few short years. The main thing that Britannica sells is its reputation and the expertise
of its editors, contributors, and advisors. Britannica has decided that the best way to
capitalize on that reputation and expertise is through a combined format of subscriptions
and advertising support.




The Future of Digital Product Sales:
Web Services
A key element in the delivery of digital products is the ability to communicate and send
products across organizational boundaries: from one company to another, or from a
company to an individual consumer. Web services hold the hope of providing this
capability in the near future (Homan, 2002).
Web services are combination of software tools that let application software in one
organization communicate with other applications over a network by using a specific set
of standard protocols. Web services can be described as self-contained, modular units
of application logic that provide some business functionality to other applications
through an Internet connection (Ismail, Patil, and Saigal, 2002). A growing number of
companies are using Web services today to improve customer service and reduce costs
of operations within their enterprises.
Web services allow programs written in different languages on different platforms to
communicate with each other and accomplish transaction processing and other business
tasks. The common format of this machine-to-machine communication was originally
HTML, however, most newer Web services implementations use XML.


Development of Web Services

The first Web services were information sources that programmers incorporated into
software applications. For example, a company that wanted to collect all of its financial
management information into one spreadsheet might use Web services to obtain bank
account and loan balances, stock portfolio holdings, and current interest rates on
financial instruments. If this information is available through Web services, the spread-
sheet program can use those services to update itself automatically. Some of the
information might be available as a Web service at no cost. Other information access




Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
170 Schneider


might require a subscription. But Web services can make automated access of the
information much easier (Fingar, 2002).
A more advanced example is a purchasing software application that could use Web
services to obtain price information from a variety of vendors. After a purchasing agent
reviews the price and delivery information and authorizes the purchase, the software can
submit the order and track it until the shipment is received. On the other side of this
transaction, the vendor™s software can use Web services (in addition to providing price
and delivery information) to check the buyer™s credit and contract with a freight company
to handle the shipment (Dyck, 2002).


Current Applications of Web Services

J.P. Morgan Chase & Co., a major investment bank, uses Web services in its investment
information portal to pull information such as economic forecasts, analyses of specific
companies, industry forecasts, and current securities trading markets results into online
reports that customers can access on the company™s customer portal site. The bank™s
customers could each obtain all of this information independently, but the bank™s
aggregation provides a service to its customers. Another example is CUNA Mutual
Group, which sells services such as check clearing and construction management to
credit union customers throughout the United States. CUNA provides many of these
services by running programs on its legacy systems. Instead of reprogramming every-
thing so it could be accessible on the Web, CUNA uses Web services to take information
from the legacy systems and generate Web pages that it makes accessible to its
customers (Pallatto, 2002).
Dollar Rent-a-Car developed a Web services implementation, built with XML, that
connects its reservation system to Southwest Airline™ system. This allows current
information about availability and pricing to be automatically transferred to the South-
west system that runs its reservations Web site. Customers who buy a Southwest airline
ticket can book a rental car on the Southwest site. The transaction is automatically
transmitted back through a Web services connection to Dollar, where the inventory
control and accounting systems are updated in real time (Choi and Whinston, 2002).


The Promise of Web Services

For years, the IT industry has been driven by vendors who promoted proprietary
programming languages and systems that could not communicate with each other easily.
Large companies have been forced to hire substantial programming staffs or consulting
firms to create middleware that could integrate their hodgepodge of programs for order
entry, financial management, inventory control, marketing, and other functions. Since
the idea of successfully and easily connecting software within an organization is still an
unachieved goal in many organizations, connecting software across organizational
boundaries is revolutionary.




Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
Digital Products on the Web 171


Software has been traditionally sold as a product. Web services allow software and
related knowledge-based products to be sold as a service. This gives sellers more
flexibility to use bundling, versioning, and price-differentiation strategies to increase
profits.
The promise of easy integration (Choi and Whinston, 2002) provides a significant benefit
to customers. Web services built with XML can give organizations the ability to adapt
to changing software environments, as well (Dyck, 2002). However, many companies are
concerned about managing Web services given that the standards are evolving, no
quality-of-service monitoring function has been included in the standards, and the
security of Web services has yet to be tested by a concerted attack (Hall, 2003).




Conclusions
This chapter defined digital products, described their characteristics, and outlined the
environment in which they are offered for sale on the Web. The chapter analyzed pricing
and distribution options for digital products in an online environment and provided
examples of how companies are combining those options into viable revenue models. The
chapter concluded with an analysis of the promises and risks of a combined“delivery,
digital-product medium, Web services.




References
Adamy, J. (2000, September 26). E-tailer price tailoring may be wave of future. Chicago
Tribune, 4.
Bailey, J. (1998). Intermediation and electronic markets: aggregation and pricing in
internet commerce. Ph.D. dissertation. Cambridge, MA: Massachusetts Institute
of Technology.
Bakos, J. (1997). Reducing buyer search costs: implications for electronic marketplaces.
Management Science, 43(12), 1676-1692.
Barsh, J., Kramer, E., Maue, D. & Zuckerman, N. (2001). Magazines™ home companion.
McKinsey Quarterly, 2, 83-91.
Bichler, M. & Loebbecke, C. (2000). Pricing strategies and technologies for on-line
delivered content. Journal of End User Computing, 12(2), 4-10.
Book Publishing Report. (1999). Britannica™s online move signals critical point for
reference market, 24(44), 4-5.
Brynjolfsson, E. & Smith, M. (2000). Frictionless commerce? a comparison of Internet and
conventional retailers. Management Science, 46(4), 563-585.
Buyer and seller liability for price discrimination. (2001, September). Purchasing Law
Report, 2-5.


Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
172 Schneider


Campbell, M. (1999). Perceptions of price unfairness: antecedents and consequences.
Journal of Marketing Research, 36(2), 187-199.
Choi, J. (2003). Bundling new products with old to signal quality, with application to the
sequencing of new products. International Journal of Industrial Organization,
21(8), 1179-1200.
Choi, S.Y. & Whinston, A. (2000). The Internet economy: technology and practice.
Austin: SmartEcon.
Choi, S.Y. & Whinston, A. (2002). Web services: laying down the infrastructure for the
digital economy. SmartEcon.com. Retrieved from the World Wide Web: http://
www.smartecon.com/articles/web_services.asp.
Chung, J. & Rao, V. (2003). A general choice model for bundles with multiple-category
products: application to market segmentation and optimal pricing for bundles.
Journal of Marketing Research, 40(2), 115-130.
Cohan, P. (2001). Considerations for profitable business models. eContent, 60-61.
Cox, J. (2001). Can differential prices be fair? Journal of Product & Brand Management,
10(5), 264-275.
Cripps, M. & Schmidt, K. (1996). Reputation in perturbed repeated games. Journal of
Economic Theory, 69(2), 387-410.
Christman, E. (2002, June). Wishful Thinking. Billboard, 114(23), C 1-2.
Conhaim, W. (2001). Encyclopedia Britannica makes some changes. Link-Up, 18(4), 1-
2.
Conhaim, W. (2002). Upgrading from Dow Jones Interactive to Factiva.com. Link-Up,
19(4), 2-3.
Dahl, E. (2003). Online music: new hits and misses. PC World, 21(9), 119-123.
DiSabatino, J. (2001). From fee to free to fee again: Britannica.com restructures.
Computerworld, 35(12), 16.
Dyck, T. (2002). Web services impact. eWeek, September 16, 39-51.
Edwards, C. (1942). Types of differential pricing. Journal of Marketing, 6(4), 156-167.
Electronic Information Report. (2001). LexisNexis counterpunches competitors with
product launch, 22(22), 1-2.
Estalami, A. (1999). Consumer savings in complementary product bundles. Journal of
Marketing Theory and Practice, 7(3), 107-115.
Fingar, P. (2002, January 21). Web services among peers. Internet World.
Fishburn, P. and Odlyzko, A. (1999). Competitive pricing of information goods: subscrip-
tion pricing versus pay-per-use. Economic Theory, 13(2), 447-470.
Forsyth, J., Lavoie, J. & McGuire, T. (2000). Segmenting the e-market. The McKinsey
Quarterly, 4. Retrieved from the World Wide Web: http://mckinseyquarterly.com/
article_page.asp?tk=302016:939:24&ar=939&L2=24&L3=44.
Gillmor, D. (2003). RSS starting to catch on. Computerworld, 37(30), 19.




Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
Digital Products on the Web 173


Godin, S. & Peppers, D. (1999). Permission Marketing: Turning Strangers into Friends,
and Friends into Customers. New York: Simon & Schuster.
Hagerty, J. & Berman, D. (2003, August 27). New battleground over Web privacy: ads
that snoop. The Wall Street Journal, A1-A2.
Hall, M. (2003). Web services™ sharp edge. Computerworld, 37(20), 34.
Hane, P. (1999). UMI™s new identity. Information Today, 16(5), 1-4.
Homan, D. (2002, August 26). Look past products to Web services™ true promise.
InformationWeek, 44“45.
Ismail, A., Patil, S. & Saigal, S. (2002). When computers learn to talk: a web services primer.
McKinsey Quarterly, 70“78.
Jones, R. (2001). Factiva.com launched at SLA. Information Today, 18(7), 1-2.
Kalin, S. (1997). Northern Light search engine finds more than web pages. InfoWorld,
19(33), 56.
Kontzer, T. (2002, January 28). In Search Of Business Data. InformationWeek, 63-64.
Krishnamurthy, S. (2003). E-Commerce Management: Text and Cases. Mason, OH:
South-Western.
Krumenaker, L. (2003). Tribunes and tribulation. Searcher, 11(7), 28-30.
Kushner, D. (2002, July 11). Multiplayer game servers provide worlds of fun. The New
York Times. Retrieved from the World Wide Web: http://www.nytimes.com/2002/
07/11/technology/circuits/11HOWW.html.
Lee, H. (1998). Do electronic marketplaces lower the price of goods? Communications
of the ACM, 41(1), 73-80.
Lee, L. & Capell, K. (2003, June 23). Taps form music retailers? Business Week, 40.
Liao-Troth, M. & Griffith, T. (2002). Software, shareware and freeware: multiplex commit-
ment to an electronic social exchange system. Journal of Organizational Behav-
ior, 33(5), 635-653.
Manjoo, F. (2003, April 29). I have seen the future of music and its name is i-tunes.
Salon.com. Retrieved from the World Wide Web: http://www.salon.com/tech/
feature/2003/04/29/itunes/index.html.
McLaughlin, K. (2001, March 21). Real makes gaming play: hopes its customer base will
pay for online diversions. Business 2.0. Retrieved from the World Wide Web: http:/
/www.business2.com/ebusiness/2001/03/28783.htm.
Metz, C. (2003, April 22). Just how rampant is spyware? PC Magazine, M7-M8.
Moon, I. (2001, July 23). The champs in online games. Business Week, 51.
O™Leary, M. (1998). Dow Jones interactive rounds out Dow Jones web migration.
Information Today, 15(1).
Pallatto, J. (2002). Web services deliver, Internet World, 8(10).
Parks, R. (2002). The Faustian grip of academic publishing. Journal of Economic
Methodology, 9(3), 317-335.




Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
174 Schneider


Peppers, D. & Rogers, M. (1997). The One to One Future: Building Relationships One
Customer at a Time. New York: Currency Books.
Roth, D. (1998, April 6). Well connected. Forbes, 161(7), 44.
Sanderfoot, A. & Jenkins, C. (2001). Content sites pursue fee-based model. Folio: The
Magazine for Magazine Management, 30(6), 15-16.
Schneider, G. (2004). Electronic Commerce. (5th ed.). Boston, MA: Course Technology.
Schuchart, S. (2003, February 20). Xdrive revitalizes file sharing, Network Computing,
14(3), 32-33.
Schwartz, E. (2001). Next frontier: storage. InfoWorld, 23(22), 91.
Shapiro, C. & Varian, H. (1998). Information rules: A strategic guide to the network
economy. Cambridge, MA: Harvard Business School Press.
Sharples, H. (1999). Internet advertising spending is projected to rise sharply. Graphic
Arts Monthly, 71(11), 86.
Sieber, S. & Sabatier, J. (2003). Market bundling strategies in the horizontal portal
industry. International Journal of Electronic Commerce, 7(4), 37-54.
Simons, J. (1996). The Web™s dirty secret. U.S. News & World Report, 121(7), 51-52.
Smith, S. (2003). Content plus: what™s selling online. eContent, 26(4), 40.
Tynan, D. & Gilbert, J. (2003). Web ads on the rebound? Sales & Marketing Manage-
ment, 155(2), 17-18.
Venkatesh, R. & Kamakura, W. (2003). Optimal bundling and pricing under a monopoly:
contrasting complements and substitutes from independently valued products.
Journal of Business, 76(2), 211-231.
Venkatesh, R. & Mahajan, V., (1993). A probabilistic approach to pricing a bundle of
products. Journal of Marketing Research, 30(4), 494-508.
White, J. (2001). ACM opens portal to computing literature. Communications of the
ACM, 44(7), 14-17.




Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
On Software Piracy 175




Chapter IX



On Software Piracy*
Sougata Poddar
National University of Singapore (NUS), Singapore




Abstract
The pervasiveness of the illegal copying of software is indeed a worldwide phenomenon.
Economists argue that when the piracy takes place at the end-users™ level, the original
software developer finds it profitable to allow limited piracy when the effect of network
externality is reasonably strong in the users market. We argue when the piracy is of
retail in nature, the same logic cannot be extended as the reason for piracy and show
that it is always optimal for the original software developer to protect its software even
when the effect of network externality is strong in the end-users™ market. We suggest that
piracy depends on more fundamental issues like demand environment, market structure,
nature of piracy and nature of competition. The other issue we cover here is the
economic impact of piracy on the welfare of a society. We discuss various policy
implications on regulating piracy in developing as well as developed markets.




Introduction
In this age of digital technology, the heavy use of computer-related jobs using various
software packages in our day-to-day activity has become a rule rather than exception.
With the advent of digital technology and the popular usage of software packages, one



Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
176 Poddar


thing that is also making headlines at the same time is software piracy. The pervasiveness
of the illegal copying of software is indeed a worldwide phenomenon. It is not only having
a profound effect on the users of the software, but also on the software industry as a
whole. It is also having a tremendous effect on the development of digital intellectual
properties and technologies. Software piracy is rampant because of the very nature of
the product. Software production incurs large development costs, but once developed,
the manufacturing costs of fabricating a copy of the software program are almost
negligible. In other words, replicated copies of the original software incur zero costs and
this is precisely why software piracy presents such a lucrative and effective option for
those who are out to make a quick profit.1 This implies a huge loss of potential customers
of original software buyers, which directly translates into revenue losses for the software
industry. Software manufacturers, through their trade organizations, have been assert-
ing the huge damage inflicted on their businesses by the illegal use of software. In 1995,
the Business Software Alliance (BSA) claimed that the industry lost “$13 billion per
year,” “$35 million per day,” and “$407 per second” from software piracy. The 1998
Global Software Piracy Report released (in May 1999) by the BSA and the Software &
Information Industry Association (SIIA), the two leading trade associations for the
software industry, estimates that of the 615 million new business software applications
installed worldwide during 1998, 231 million “ or 38% “ were pirated. In other words, one
out of every three software applications installed worldwide was pirated! In 2001, the
corresponding figure remains at 40%.2 Revenue losses to the global software industry
due to piracy were estimated at $13.08 billion in 2002. Asia, North America, and Western
Europe accounted for the majority of world revenue losses. In 2002, the combined total
losses for these regions stood over $10.5 billion, and within that Asia alone accounted
for a loss over $5 billion. These losses not only pose a serious constraint on the growth
of the software industry but also adversely affect investment decisions and limit the
development of new software products. At the same time, rampant piracy inhibits job
creation and government revenue contributions. As a matter of fact,
PricewaterhouseCoopers (1998) estimated that if world governments had reduced
software piracy rates to benchmark levels3, direct and indirect employment would have
increased by 521,663 jobs and tax revenues by as much as $13.7 billion in 1996/97 alone
for the non-U.S. economy. For the U.S. economy, reducing piracy would have generated
an additional 130,000 jobs and nearly $1.0 billion in tax revenues in 1996. And this problem
of software piracy only gets bigger with the revolution and intensification of the Internet.
“What Do You Want To Pirate Today?” reads a banner at one of the many sites that can
be found by any user doing a basic Internet search for the word “warez” “ the online term
for unlicensed programs. The emergence of the “Web” has added a new dimension to
software piracy by permitting electronic sales and transmissions of illegal software on
a grand scale.
Given this, conventional wisdom suggests the need for the legal software firms and
governments to take a harsh approach on piracy of software. Interestingly, a group of
economists would ask the question, in reality, is the original software developer or the
government or the controlling authority seriously interested to stop piracy? In their
recent work they actually show that the answer is not necessarily positive. This strand
of literature (Conner and Rumelt (1991), Takeyama (1994), Slive and Bernhardt (1998), Shy
and Thisse (1999)) provides us with the unconventional wisdom on the issue of software



Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
On Software Piracy 177


piracy. It shows that in some situations even the original software developer may not
necessarily want to clamp down on piracy “ even when it has the means to do so. In other
words, it is actually profitable for the original software developer to allow limited piracy.
The arguments to establish this basically stand on the feature of network externality that
is observed in the software user market.4 It is true that the occurrence of network
externality is a very prevalent feature in the software market and the existence of it plays
a central role for this (piracy) phenomenon. But, in this chapter, we argue that all these
unconventional results cannot be accepted as a general explanation for the existence of
software piracy in the real world. To prove the point, the chapter comes up with a model
where its shows that even in the presence of strong effect of network externality,
protection as opposed to allowing piracy is always optimal for the original software
developer. It also shows that the incentive to protect is even higher with the presence
of network externality as opposed to the case without any network externality. Therefore,
to understand the real reason for existence of software piracy, one needs to have a closer
look to the more fundamental economic issues that lie behind this phenomenon. The main
message is whether piracy is profitable or not to the original developer depends on the
market structure, demand environment and the nature of the competition.
The whole chapter will be done in three parts. In the first part (part I), we begin following
the existing literature, why network externality could be a reason for the existence of
software piracy in certain situations. After that a game-theoretic model of price compe-
tition between an original software developer and a pirate will be studied and analyzed
in detail to see the effect of network externality on piracy. In this part, we will assume that
both of the competitors move simultaneously, i.e., choose their strategies for operation
simultaneously. In the second part (part II), a game of sequential moves will be focused,
where the original firm acts as a leader and the pirate as the follower. The mode of piracy
that will be considered in both of these analyses is retail piracy. A retail piracy takes place
when a pirate (just like another firm) competes with the original software developer by
producing a copy which may not be as reliable as the original product, but comes with
a cheaper price. In the third part (part III) of the chapter, a complete welfare analysis will
be done by assuming (i) when a pirate is present in the market, and (ii) when the pirate
is absent. The question we ask, if there is a social planner (say, the government) whose
objective is to maximize society™s welfare, then what would be the policy recommendation
with respect to piracy? In other words, the question is whether allowing piracy is welfare
improving or welfare reducing from the society™s point of view. In this part, a rigorous
analysis will be done on this issue. This will be followed by a general discussion on the
welfare aspects of software piracy in a wider perspective. Towards the end, we highlight
the contribution of this chapter to the academic as well as business practitioners, like
managers and entrepreneurs. Finally, we conclude the chapter with a brief summary of
the main findings and by pointing out some future directions of research.
The whole study that will be done here is a theoretical analysis. Tools of economic theory
and game theory are used to model various economic situations. It will develop analytical
models in order to explain and understand the important economic issues of software
piracy. The study will also provide testable hypotheses for empirical and applied work.
Before concluding the introduction, we would like to say a few words on game theory
and the aspect of network externality that we consider here. Firstly, for those readers who



Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
178 Poddar


are not that familiar with game theory (some reading material is listed in bibliography),
game theory is a formalization of strategic thinking. It is useful in a situation where the
number of players (e.g., firms, economic agents) is not very large. In a situation like this
every player thinks strategically to maximize his/her payoff knowing about the fact that
every other player is doing the same. In such a situation, the payoff to one player not
only depends on his/her action (strategy), but it also depends on other players™ actions
(strategies) as well. In real life, in industries where there are only a few firms operating,
game theory could be a useful tool to analyze the strategic behaviour of firms. Examples
of such industries are: airline, pharmaceutical, telecommunication, banking, computer
(hardware and software) and so on.
The second issue is pertaining to the feature of network externality that we have already
talked about. We would like to emphasize that when we talk about network externality,
we are not limiting ourselves to the physical internal or external network among users,
but a network in much broader sense. All possible users (whether or not they are
physically connected by some network) of particular software or generally an operating
system (e.g., Windows, Linux, etc.) potentially form a group/network, where they can
gain much by sharing information/files among themselves. Naturally, the greater the
number of users in the network, the higher the gain from sharing, hence, the higher the
utility to the individual user. 5 As we will see, this aspect of network externality is going
to play a major role in our forthcoming analysis.


Part I

Network Externality “ A Reason for Piracy

We start with the assumption that consumers™ willingness to pay for software increases
with the total number of consumers who use (legally or illegally) the same software. That
is, the presence of network externalities reflects the fact that software users place a high
value on compatibility and file sharing. Suppose now that software is protected, and let
us assume that installed protective devices make it prohibitively costly for any consumer
to pirate the software. In this case, some consumers buy the software, whereas all others
simply do not use any software. Notice that if firms keep prices fixed, legal users have
all the incentives in the world to give this software for free to non-users, thereby
increasing the compatibility value of the software. However, legal users are prevented
from sharing their software by the protective devices installed into the software. If,
however, firms would remove protection from the software, then some consumers may
be willing to pay a higher price since the value of the software increases with the total
number of legal and illegal users. This is the core of the argument developed in the
section.
In particular, Takeyama (1994) considers a model of unauthorized reproduction of
intellectual property in the presence of demand network externalities and shows that
unauthorized copying can induce greater profits to the original firm. She considered a




Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
On Software Piracy 179


discreet demand model with two groups of consumers who have different valuations for
a good. She analyzed the situations when copying is easily possible by the users and
when cost of copying is prohibitively high, hence copying is not possible. In these two
contrasting situations, she showed that if the network effect is sufficiently large, profits
for the firm are higher with copying than without copying. The result follows from the
fact that unauthorized copying can be relatively efficient means of increasing network
size. In effect, unauthorized copying allows the firm to price-discriminate among different
classes of consumers. With copying, large network size can be achieved by the existence
of marginal consumers who make reproductions (at zero cost to the firm), while infra-
marginal consumers purchase originals at a price that may largely appropriate the
externality of increased network size created by copiers. Without copying, the same
network size may only be obtained at a possibly lower price on all existing units. Increase
in network size increases the value of the product unambiguously to all consumers
therefore enabling firms to raise price to those who buy it from the firm. In other papers,
Conner and Rumelt (1991) and Slive and Berhardt (1998) come up with a similar finding
explaining why a software manufacturer may permit limited piracy of its product. These
studies concentrated in a situation where there is only one original manufacturer, in other
words, a monopolistic industry. On the other hand a strategic approach to software
piracy is found in Shy and Thisse (1999). They show that there is a strategic reason why
software firms have followed consumers™ desire to drop software protection. They
analyze software protection policies in a price-setting duopoly software industry selling
differentiated software packages, where also consumers™ preference for particular
software is affected by the number of other consumers who (legally or illegally) use the
same software. Their results show when firms protect their software a low-price equilib-
rium emerges if network effects are strong, whereas a high-price equilibrium arises under
weak network effects. Therefore, all firms are better off with software protection when
network effects are weak. In contrasts, firms prefer not to protect their software when
network effects are strong. In another set of results which deals with a market situation
where firms choose to protect or not, prior to price competition they found that for very
weak network effects, both firms choose to protect their software because the impact of
piracy on sales is insignificant. For the intermediate value of the network effects, one firm
chooses to protect whereas the other does not. This is because the network effects are
now strong enough to induce one firm not to protect, thereby benefiting from the larger
network size, whereas these effects are still too low for the other firm to be able to afford
to do it. Furthermore, the non-protecting firm earns a higher profit than the protecting
firm. Finally, when network effects are sufficiently strong, both firms choose non-
protection.
Hence, all these studies unambiguously try to make a point that the existence of strong
demand network externalities is the central reason for the existence of piracy. In the
following section, we lay out a model of software piracy with the feature of network
externality and show that above arguments are not generally valid.




Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
180 Poddar


Network Externality “ A Reason for Protection

The Model

Consider an original software firm and a pirate software firm. The pirate has the
technology to copy the original software. In reality, we know that the cost of copying
software is negligible, hence we assume the cost of copying is zero for the pirate. The
probability that a pirated software works is q, q∈(0,1) and this probability is common
knowledge. Therefore q serves as a proxy for the quality of the pirated software. Usually
pirated copies does not come with the supporting services, so one can think even if the
pirated software is exactly same as the original one (because of digital coping), but the
lack of supporting service does not allow the user to get the full value of the pirated
software, hence quality of the pirated software q can also be interpreted like this. For
simplicity, we also assume that the marginal costs of software production (i.e., making
copies) are nil for both firms.6
There is a continuum of consumers indexed by X, X∈[0,1]. A consumer™s willingness to
pay for the software depends on how much he/she values it “ measured by X. A high
value of X means higher valuation for the software and low value of X means lower
valuation for the software. Therefore, one consumer differs from another on the basis
of his valuation for the particular software. Valuations are uniformly distributed over the
interval [0,1] and the size of the market is normalized to 1.
A consumer™s utility function is given as:


X “ PO if buys original software

7
U= q X“ PP if buys pirated software

0 if buys none




There is no way a consumer can get pirated software which has a defect replaced, since
there is no warranty for it.8 Hence, the consumer enjoys the benefit of the pirated software
only with probability q. In the event that the pirated software purchased does not work
at all, the loss to the consumer is the price paid for it. The original software is fully
guaranteed to work. PO and P P are the prices of the original and pirated software
respectively. It must be true that PO > PP . PO - PP can be viewed as the premium a consumer
pays for buying “guaranteed-to-work” software.
In our model, just as before, network externality implies that the value of a particular piece
of software for a consumer increases as more and more consumers use it. With the
presence of a pirate, and due to lower price of the pirated software, more people tend to
buy the software, which in turn increases the number of software users in the society.




Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
On Software Piracy 181


This intensifies the network effect, and as a result, this increases the value of that
software for any potential buyers (both legal and illegal). Under this, we will consider
two cases in turn. First, where the original developer protects its software, and secondly,
where the original developer does not protect.9


Software Protection (No Piracy)

Without piracy, consumers would choose only between either buying the original one
or not buying, depending on their valuation of the software.
Thus a consumer™s utility in the presence of network externality is given by:


X + θ D NP “ P NP
U= if buys original software

0 if buys none




DNP denotes the total demand of the software under protection (i.e., no piracy) 10 and PNP
denotes the price. Now θ ∈[0, ½]11 is a coefficient which measures the importance of
network size to the software user. It can be viewed as the degree of network externalities.
For example, if θ is close to ½, it implies the stronger effect of network externality and
when θ is close to zero, there is almost no effect of any network externality at all.


Figure 1. Distribution of buyers (case of protection)

None Original


0 X 1




X is the marginal consumer who is indifferent between buying the original software and
not buying any software at all:


X+ θ D NP - P NP = 0

X = P NP - θ D NP




Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
182 Poddar


Demand for the original software is:



1 ’ X = 1- P NP + θ D NP
D NP =

1 ’ PNP
’ D NP =
1’θ


The monopolist™s profit is:


πNP = PNP . DNP

1 ’ PNP
= PNP .
1’θ


Solving for the profit-maximizing monopolist price, we get:


P*NP = ½ (1)


And demand is:


1
D*NP = 2(1 ’ θ ) (2)


Note that when θ = ½, D *NP = 1, i.e., the full market is served.
Hence, the profit of the monopolist software firm in the case is:


1
π*NP = (3)
4(1 ’ θ )



No Software Protection (Piracy)

This time, a consumer™s utility is given by:


X + θ D O + qθ D P “ P O 12 if buys original software

q X +qθ D O + q 2 θ D P “ P P 13
U= if buys pirated software

0 if buys none


Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
On Software Piracy 183


DO, PO and D P, PP are the demand and prices for the original and pirated software
respectively.
As mentioned earlier, q is the probability that the pirated software works. This time,
though, the loss to the consumer if the pirated software does not work is comprised of
the price paid for the illegal software and the intangible cost which arises from not being
able to enjoy the positive network externality.



Figure 2. Distribution of buyers (case of non-protection)


None Pirate Original



ˆ ˆ
0 Y X 1




ˆ
Like before, the marginal consumer, X , who is indifferent between buying the original
software and the pirated version is given by:


X + θDO + qθ DP “ PO = q X + qθ DO + q2θ DP “ PP
ˆ ˆ

PO ’ PP
’ θ (DO + qDP )
ˆ
X= 1’ q


ˆ
The marginal consumer, Y , who is indifferent between buying the pirated software and
not buying any software at all is given by:


q Y + qθ DO + q 2θ DP “ P P = 0
ˆ

’θ (DO + qDP )
PP
ˆ
Y=
q


The demand for original software is given by:


ˆ
DO = 1- X




Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written
permission of Idea Group Inc. is prohibited.
184 Poddar


(1 ’ q ) + θ (qPO ’ PP )’ (PO ’ PP )
’ DO =
(1 ’ q )(1 ’ θ )

and the demand for pirated software is given by:


ˆˆ
DP = X - Y

qPO ’ PP
’ DP =
q (1 ’ q )


The original firm and the pirate compete by choosing price strategically. The respective
reaction functions are given by:


1 ’ q + PP (1 ’ θ )
2(1 ’ θq )
PO(PP) =

qPO
PP( P O ) =
2

<<

. 7
( 15)



>>