Transcript Network Economics 101 - Progress and Freedom Foundation

Network Economics 101
Phil Weiser
Aspen, Colorado
Sunday, April 29, 2007
What is a “Network Industry?”
 Broad definition: An industry that connects users to each other or to
producers through a connected system (or systems) of physical
infrastructure.
–
–
–
–
Communications networks
Power grids
Transportation systems
Software platforms
 Economic definition: An industry whose good or service increases in
value to each individual user as the number of other users of the same
good or service increases.
– Communications networks
– Software platforms
– Transportation systems (?)
What is a
“Network Externality”
 The consumer’s perspective: A “network externality” is the added
benefit a consumer gets as additional consumers join the same
network or use the same product. Phone service may benefit
consumers if they can call 10 other people, but it becomes increasingly
valuable as they can call 100, 1000, 1 million+ other people. The same
holds for a word processing program, for example. It is the benefit that
comes from being able to communicate or interact with a greater
number of people, which thereby makes the given network more
valuable to each user.
 The competitive firm’s perspective: A rival’s “network externality” is a
barrier to entry. Once a firm has the market lead and provides the
greatest network benefit to subscribers, rivals must not only beat the
price and technology of the leading firm, but must do so by enough to
compensate consumers for the network benefit they lose in switching
from the leader to the smaller rival.
Network Externalities Distinguished
from Economies of Scale
 Scale economies arise when a producer’s per-unit
costs fall as output increases.
 In contrast, network externalities arise when a
consumer’s benefit grows as total consumption
increases.
 Scale economies and network externalities need
not go together; it is possible to have one without
the other.
Network Externalities Distinguished
from Lock-In Effects

Lock-in effects arise when consumers find it more economically rational to stick
with an existing product/service rather than to switch to a competing one.

Network externalities can create lock-in by making the alternative provider less
attractive in terms of the benefit it will provide to the consumer. Even if offered
payment to switch, a consumer might decline because of the network benefit
she would lose.
– Regulation can require firms to “share” their network externality (through an
interconnection or interoperability mandate).

But not all lock-in arises from network externalities; lock-in can occur even
when the competing product/service is more attractive if the “switching costs”
of moving to the better service are too high. Service termination penalties,
incompatibility with already-purchased complementary goods, and sunk costs
are factors that might keep consumers from switching even to better choices.
– Regulation can limit switching costs by mandating cooperation between rivals (e.g.,
number portability).
The Vertical Dimension: Network
Effects and Complementary Goods

The benefit of a network or software platform may derive from the availability
of complementary products. My operating system is more valuable to me the
more applications programs I can run on it. My DVD player is more valuable
to me the more movies there are on DVD. As more other consumers by my
operating system or DVD players, the more such complementary products
get made, making the operating system or DVD player more valuable to me.
This is sometimes called an “indirect” network effect or “feedback effect.”

Whether or not there is a network monopoly will depend on rivals’ access not
to the underlying platform or network and its customers, but to the
complementary products.

Different network (or platform) providers will adopt different strategies toward
complementors (or application providers).
– Modularity—Palm, Microsoft Xbox, Linux, Comcast DVR
– Vertical Integration—Microsoft (OS and browser), Comcast broadband and VoIP

ICE Helps Us Understand Choice of Strategies and Whether they pose
competition policy concerns.
Consequences for Market
Performance and Firm Strategy

Effects on market structure: When the externality cannot be shared among firms, a
network industry can “tip” toward monopoly. When AT&T refused in the early 20th century
to interconnect with rivals, it quickly gained dominance because of its larger subscriber
base, which in turn attracted more new subscribers because of the larger network
benefit, thus creating a cycle that reinforced itself and led to monopoly.

Implications for competitive strategy: Early acquisition of customers is critical in the
race to capture a network market. The first firm to gain a decisive lead will become
dominant, at least for some period of time. Low prices, giveaways and other promotions
are likely to be common.

Dynamic effects on innovation and competition: Network effects will drive firms to
innovate and compete over time for the market. Just because one firm becomes
dominant does not mean all competitors melt away. Some rivals will try to innovate
ahead of the incumbent, initiating a new round of competition that may lead to a new
firm’s becoming dominant. Competition in R&D will likely be an important strategy in
network markets. At a point in time one might observe several firms competing in R&D
even if at that same time there is little competition in the product market.

Non-market strategies: Legal and political means of obtaining access to the
incumbent’s networks are likely to be attempted.
Network Externalities Pose
Regulatory Challenges

Network market dominance is not necessarily a result of anticompetitive
strategies, but can be consumer-driven. Monopoly is not, as in conventional
markets, clearly bad for consumers. They receive a large benefit from the
network externality. If that externality cannot be had absent the monopoly, rules
designed to end the monopoly may harm consumers.

Regulation may therefore have to make trade-offs between conventional
price/output objectives and network externalities. One policy approach is to
promote and preserve network externalities for consumers, but to do so at the
lowest possible price to consumers by simultaneously introducing competition
where possible.

Interconnection and interoperability become key policy instruments in
managing this policy objective; more important than retail price regulation. Rate
regulation my deter innovation and dynamic competition by reducing the payoff
for competing R&D efforts. Interconnection and interoperability, on the other
hand, try to preserve competition among firms that collectively provide
consumers with a network benefit.
Entry In Network Industries
 Combination of IP, network externalities, and
rapid growth make measurement of
monopoly difficult.
 Lessons From Standard Fashion—
• What are barriers to piecemeal entry?
• Does monopolist exclusionary dealing extend
duration of monopoly? (Note Schumpeterian
critique.)
Considerations when Intervening in
an Evolving Network Market

As rival firms compete to provide a network service, the ideal is to have rapid
deployment, continued competition, and the ability of customers of all providers to share
a common network externality. Interconnection is the way to do this. Because
subscribers to one mobile phone carrier can call subscribers to all other mobile phone
carriers, no carrier has a monopoly on the network externality regardless of market
share. Despite some costs, there is little debate over basic interconnection.

But interconnection can be taken much further than basic termination of calls originating
on a rival network. For example, one carrier might introduce proprietary features or
calling plans that could lead the market to tip. Should a firm have to give rivals access to
novel technology? Should carriers be barred from implementing plans that benefit
uniquely their own subscribers? Should carriers have to grant wholesale network access
to rivals that lack necessary facilities? Over these questions there is far more debate,
because the tradeoffs for technological innovation and even for conventional price
competition will be much greater.

Key lesson: Interconnection is not unambiguously good, and the fact that it is beneficial
in its basic form does not mean it should be taken to greater lengths in the name of
access and competitive neutrality.
Intervention when the Market has
Tipped to Network Monopoly
 If the market does tip and only one firm delivers the network externality,
the question is whether monopoly effects can be weakened without
diminishing the network benefit for consumers, and without weakening
dynamic competition.
 Easy case is when basic interconnection will do the trick, as it might
have in 1910 when AT&T began to regain its monopoly. But this often
will not be enough.
 3 More Aggressive Remedies: In many if not most cases more
intense intervention will be needed to stimulate competition in the short
run. Possibilities, depending on context, are (1) unbundling or
wholesale access to incumbent’s facilities, (2) divestiture of the
monopoly into separate firms, and (3) licensing of proprietary interfaces
to potentially competing platforms. Each of these involves potentially
important economic tradeoffs.
(continued)

Wholesale access/unbundling: Can have negative effects on investment and
innovation. To avoid such harm, network access pricing must be correct, which
is very difficult.

Divestiture: Could waste economies of scale and, if divested entities evolve in
ways not fully compatible with each other, will diminish network externalities
(potential Microsoft divestiture problem).

Licensing of interfaces: Could diminish innovation incentives on a forwardlooking basis because benefits would have to be shared. Such licensing would
have to be carefully calibrated both as to timing and price. But often a useful
solution, increasingly adopted to remedy merger concerns.

Lesson: Introducing competition into network markets can be costly and the
tradeoffs need to be considered carefully.
A Central Role for Antitrust
 It is particularly important where network dominance has occurred to
make sure that dominance is not maintained by anticompetitive
strategies. Antitrust must play a big role in ensuring that a network
monopoly lasts no longer than justified on the competitive merits.
 Antitrust will be particularly important where a regulated network
operator has an unregulated line of business that is complementary to
its regulated service. Example is a cable network that owns and
produces content and services that it sells to its internet access
subscribers. DOJ and FTC will play the primary role in evaluating
potential tying arrangements and vertical mergers that could give rise
to monopoly leveraging or monopoly maintenance allegations.
 Hiqh Stakes institutional question—can the FCC perform antitrust-like
function (ex post evaluation of dominant firm practices) and can
antitrust courts act effectively (both knowledgeable and quickly) in new
economy markets
Three Hard Questions
 Intervention in evolving network market:
Should regulators intervene to ensure sustained
competition in an evolving network market—either
vertical access (i.e., network neutrality) or
horizontal access (i.e., interoperability).
 Remedies for a network monopoly: If a network
monopoly arises, should regulators try to limit the
duration of the monopoly or otherwise constrain its
behavior? And if so, when and how?
 How to judge impact of price discrimination via
vertical integration?
Questions?